UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Neutrophil collagenase (matrix metalloproteinase-8) : regulatory roles in inflammation and autoimmunity Cox, Jennifer H. 2008

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

Item Metadata

Download

Media
24-ubc_2008_fall_cox_jennifer_h.pdf [ 4.1MB ]
Metadata
JSON: 24-1.0066986.json
JSON-LD: 24-1.0066986-ld.json
RDF/XML (Pretty): 24-1.0066986-rdf.xml
RDF/JSON: 24-1.0066986-rdf.json
Turtle: 24-1.0066986-turtle.txt
N-Triples: 24-1.0066986-rdf-ntriples.txt
Original Record: 24-1.0066986-source.json
Full Text
24-1.0066986-fulltext.txt
Citation
24-1.0066986.ris

Full Text

NEUTROPHIL COLLAGENASE (MATRIX METALLOPROTEINASE-8): REGULATORY ROLES IN INFLAMMATION AND AUTOIMMUNITY by  Jennifer H. Cox B.Sc., University of Victoria, 2001  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Biochemistry and Molecular Biology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) October 2008  © Jennifer H. Cox, 2008  ABSTRACT Inflammation is an essential process in wound healing and for the elimination of invading pathogens. However, unregulated inflammation can lead to numerous pathologies including autoimmu nities, tu morigenesis, and atherosclerosis.  Matrix metalloproteinases (MMPs),  once thought to be only extracellular matrix degrading enzymes, are now known to be key regulators of inflammatory and immune responses through proteolysis of bioactive molecules. MMP-8, a neutrophil-specific MMP, is protective in skin cancer models where MMP-8 knockout mice have an initial delay in neutrophil infiltration followed by a massive accumulation at the site of treatment.  We investigated this delay in a murine air pouch  model of acute inflammation, where MMP-8 deficiency caused decreased neutrophil migration in response to LPS. This was attributed to MMP-8 processing and activation of LPS-inducible CXC chemokine (LIX), a murine neutrophil chemoattractant. Indeed, MMP-8 knockout mice had normal neutrophil infiltration in response to synthetic analogs of cleaved LIX. Furthermore, homologous pathways with human chemokines CXCL5 and CXCL8 were described. In vivo, an indirect interaction between MMP-8 and LIX also occurs, whereby MMP-8 processes and inactivates cLl-proteinase inhibitor causing increased neutrophil elastase activity, which then efficiently cleaves and activates LIX. MMP-8 was protective in a model of rheumatoid arthritis where synovial tissues from MMP-8 deficient mice had an abundance of neutrophils. This prolonged neutrophil accumulation correlated with a loss of caspase-1 1 expression, consequent decreased caspase-3 activity and reduced apoptosis. MMP-8 shedding of TNF-a was also decreased in MMP-8 deficient leukocytes, potentially dampening a key apoptotic pathway in neutrophils. The role of MMPs in processing the Thi cell CXCR3-binding chemokines CXCL9, CXCL1O, and CXCL11 was investigated.  The  leukocytic MMPs -7, -8, -9, and -12 cleaved CXCL11 at both the amino and carboxy terminus. N-terminal cleavage resulted in the conversion of a receptor agonist to antagonist whereas C-terminal cleavage by MMP-8 caused a significant loss in glycosaminoglycan binding, demonstrating for the first time that direct chemokine proteolysis can regulate the formation of haptotactic gradients. Therefore, MMP-8 is a pivotal regulator in the onset and termination of inflammation, and has multifaceted roles in innate and acquired immunity as well as the autoimmune disorder rheumatoid arthritis.  TABLE OF CONTENTS ABSTRACT  ii  TABLE OF CONTENTS  iii  LIST OF TABLES  vi  LIST OF FIGURES  vii  LIST OF ABBREVIATIONS  ix  ACKNOWLEDGEMENTS  x  CO-AUTHORSHIP STATEMENT  xi  CHAPTER 1. INTRODUCTION  1  The Basics of Matrix Metalloproteinases  1  The Neutrophil Collagenase: MMP-8  3  Cytokine Substrates: MMP Regulation of Inflammatory Signalling Molecules Perspective  5 5  MMPs and Inflammation  6  Genetic Models of Inflammation  7  MMP Regulation of Cytokines  10  MMP Regulation of Chemokines  17  Future Perspectives  22  Themes and Hypotheses  23  References  25  CHAPTER 2. LPS RESPONSIVENESS AND NEUTROPHIL CHEMOTAXIS IN VIVO REQUIRE PMN MMP-8 ACTIVITY Perspective  3 35  Introduction  36  Materials and Methods  39  Results  41  III  Discussion References CHAPTER 3. MATRIX METALLOPROTEINASE PROCESSING OF CXCL1 1/I-TAG RESULTS IN LOSS OF CHEMOATTRACTANT ACTIVITY AND ALTERED GLYCOSAMINOGLYCAN BINDING Perspective Introduction Materials and Methods Results Discussion References  54 58 62 62 63 65 70 87 94  CHAPTER 4. MMP-8 DEFICIENCY DELAYS NEUTROPHIL APOPTOSIS AND EXACERBATES RHEUMATOID ARTHRITIS Perspective Introduction Materials and Methods Results Discussion References CHAPTER 5. CONCLUSIONS AND PERSPECTIVES References APPENDIX A. PROCESSING OF ELR+ CHEMOKINES BY NEUTROPHIL SERINE PROTEASES Perspective Results Summary and Future Experiments References  gg 100 102 106 126 131 136 143 144 144 145 154 156  iv  APPENDIX B. PROTEOMIC ANALYSIS OF MMP-8 CLEAVAGE SPECIFICITY AND SUBSTRATE IDENTIFICATION Perspective Results Summary and Future Experiments References APPENDIX C. UBC RESEARCH ETHICS BOARD CERTIFICATES  157 157 158 162 164 165  V  LIST OF TABLES Table 1.1.  Extracellular matrix and bioactive substrates of MMP-8  4  Table 1.2.  Inflammatory and immune phenotypes of MMP-deficient mice  8  Table 1.3.  Bioactive MMP substrates: The pro-inflammatory and anti-inflammatory proteolysis events by MMPs  12  vi  LIST OF FIGURES Figure 1.1.  Schematic structure of MMPs  Fl g ure 1 2  MMP proteolysis of cytokines and chemokines to precisely control the induction and termination of inflammation  15  Figure 2.1.  Impaired PMN responsiveness to LPS in MMP-8 deficient mice  42  Figure 2.2.  MMP-8 cleavage of LIX  44  Figure 2.3.  LIX is selectively cleaved by multiple MMPs  46  Figure 2.4.  In vitro cellular responses to MMP-8 cleaved LIX  47  Fi g ure 2 5  MMP-8 is required for PMN chemotaxis torard LIX in vivo, but is not required for PMN cell migration  Figure 2.6.  MMP proccessing of CXCL8 and CXCL5  50  Figure 2.7.  Enhanced bioactivity of MM P-truncated CXCLS in vitro and in vivo  53  Figure 3.1.  MMP processing of CXCR3 ligands  71  Figure 3.2.  MMP processing of murine CXCL1 I  74  Fi g ure 3 3  MMP cleavage sites in CXCL9, CXCL1 0, CXCL1 1 and murine CXCLI1  75  Figure 3.4.  Titration of MMP processing of CXCLI I  77  Figure 3.5.  Exosite interactions with CXCLI 1  79  Fi g ure 3 6 Fl g ure 3 7  MMP-mediated cleavages of CXCL1 1 result in loss of agonism and the generation of antagonism Chemotactic migration is decreased and antagonized in response to CXCL1 1 (5-73) and (5-58)  2  80 82  Figure 3.8.  Structural analysis of full length and truncated CXCL1 1  83  Figure 3.9.  Heparin binding is altered upon CXCL1I cleavage  85  Fi g ure 3 10  Model representing the potential role of leukocytic MMPs 7, 8, 9, and 12 in disrupting the haptotactic gradient of CXCL1 1  88  Figure 4.1.  Generation and characterization of MRLJIprx Mmp8-I- mice  107  Figure 4.2.  Increased joint swelling in male MMP-8 deficient MRL/lpr mice  109  vii  Fi g ure 4 3 Fi g ure 44 .  Fi g ure 4 5 Figure 4.6. Figure 4.7. Figure 4 8 Figure 4.9. Fi g ure 5 1 .  .  Histological analysis of joints reveals increase neutrophil infiltration in male MMP-8 deficient mice CLIP-CHIP analysis of bone marrow neutrophils from MMP-8 knockout and wild-type MRL/ipr mice Neutrophils and mononuclear cells from MMP-8 deficient mice have abolished constitutive and inducible levels of caspase-1 1 MMP-8 deficient neutrophils have unaltered caspase-1 activity and IL-I I secretion MMP-8 mediated cleavage of proTNF-a and shedding of TN F-cL from neutroph ils Reduced caspase-3 activity and TUNEL staining in MMP-8 knockout neutrophils Proposed mechanisms for MMP-8 promotion of programmed cell death through induction of caspase-Il expression and shedding of TNF-a Solving the MMP-8 puzzle: diverse functions of MMP-8 in inflammation and acquired immunity .  112 115 117 119 122 124 125 140  Figure A.1.  Neutrophil processing of LIX is independent of MMP-8  146  Figure A.2.  Neutrophil-derived serine proteases cleave LIX  147  Figure A.3.  Neutrophil elastase processing of murine ELR chemokines  149  Figure A.4.  Neutrophil elastase processing of human ELR chemokines  151  Figure A.5.  MMP-8 cleavage of cLl-proteinase inhibitor  152  Figure A.6. Fi g ure A 7  Reduced LIX processing and increased full length al-PI in BALF from MMP-8 deficient mice Proposed mechanism for MMP-8 regulation of neutrophil elastasemediated LIX activation .  .  153 155  Figure B.1.  Cleavage site specificities of murine and human MMP-8  159  Fi g ure B 2  CLIP-TAILS analysis of Mmp8-I- MEF secretome in the presence and absence of murine MMP-8  163  Viii  LIST OF ABBREVIATIONS al-PI AEBSF ADAM APMA BALF BSA CATG CD DMEM DMSO DTT GRO HBSS ICE IL IP-lO l-TAC LCD LIX LPS MALDI-TOF MCP MEF MIG MMP MS MS/MS MT NE PAGE PBS PMA PMN RPMI SDF SDS SLPI TACE TGF TIMP TNF  al-proteinase inhibitor 2-aminoethyl benzenesulfonyl fluoride hydrochloride a disintigrin and metalloproteinase 4-aminophenylmercuric acetate bronchoalveolar lavage fluid bovine serum albumin cathepsin G circular dichroism dulbecco’s modified essential medium dimethyl sulfoxide dithiothreitol growth related oncogene Hank’s balanced salt solution interleu kin-I converting enzyme interleukin inducible protein 10 interferon-gamma inducible T cell alpha chemoattractant linker hemopexin C domain LPS-inducible CXC chemokine lipopolysaccharide matrix-assisted laser desorption ionization-time of flight monocyte chemoattractant protein murine embryonic fibroblast monokine induced by gamma interferon matrix metalloproteinase mass spectrometry tandem mass spectrometry membrane type neutrophil elastase polyacrylamide gel electrophoresis phosphate-buffered saline phorbol 12-myristate 13-acetate polymorphonuclear cell Roswell Park Memorial Institute medium stromal derived factor sodium dodecyl sulfate secreted leukocyte protease inhibitor TNF alpha converting enzyme transforming growth factor tissue inhibitor of metalloproteinases tumor necrosis factor  ix  ACKNOWLEDGEMENTS First and foremost, I would like to thank Dr. Chris Overall for his optimi sm, encouragement, and genuine enthusiasm for science. I could not have imagin ed a more supportive supervisor, who always had confidence in me and truly gave me every opportunity to succeed. Also, I would like to acknowledge my superv isory committee for their involvement in this research: Dr. Francois Jean, Dr. Dougl as Waterfield, and Dr. Leonard Foster. Furthermore, I would like to give a special thanks to collaborators Dr. Clive Roberts and Rendi Yan for their valuab le contributions to this research. The many members of the Overall lab, past and present, have made for an exceptional and memorable experience. Thanks especially to Amanda and Rich for so many insighiful “chemokine” discussions; George, Charlotte, and Reini for great leadership and keeping the lab running smoothly; Yili for her purification skills and early morning cheer; Patrick and Oded for never turning down a coffee break; Alain and Caroline for mouse help; Verena for her contagious laugh; Anna, Ulrich, Oliver , and Olivier for sharing their proteomics expertise; and David and Magda for their latin energy. I have sincerely enjoyed my time as a graduate student, in large part because I had the opportunity to work with a friendly and talented group of scienti sts. Most of all, I would like to thank my parents for their unrelenting support throug hout all of my endeavors. They are not only excellent parents, but also great role models and friends. Thanks to Curtis for standing by my side during all the successes and failures that accompany a PhD. Finally, thanks to my big brother Jason for teachin g me to be tough enough to endure all the challenges of life, whether it be a major earthquake or just a comprehensive exam.  x  CO-AUTHORSHIP STATEMENT The chapters presented herein are all first-authored manuscripts prepared by the candidate with guidance from the graduate supervisor, Dr. Christopher Overall. Chapter 2 represents a manuscript that was a shared first authorship with Dr. Angus Tester, and the candidate would like to acknowledge Dr. Tester for his contribution of biochemical cleavage, chemotaxis, and air pouch experiments as noted in figure legends. Amanda Starr, Richard Dean, and Andrea Connor assisted with animal experiments for Chapter 2. All of the data in Chapter 3 was acquired by the candidate and Dr. Richard Dean was a co-author for his contribution of recombinant MMP-12. Dr. Clive Roberts was acknowledged for the previous development of unpublished heparin binding techniques using fluorescence polarization.  Dr. Alain Doucet contributed the  representative model of CXCLI 1, as described in the manuscript. The data for the manuscript in Chapter 4 was acquired by the candidate except for the TNF-a proteolysis experiment, which was performed by Dr. Christopher Overall, as noted in the figure legend. Amanda Starr is co-author for assistance with mouse experimentation, Dr. Reinhild Kappelhoff designed and developed the CLIPCHlPTM microarray, Rendi Yan assisted with mouse experiments and histological analysis, and Dr. Clive Roberts performed pathological grading of histological specimens. All data in the Appendices was acquired by the candidate. Dr. Alain Doucet is acknowledged for obtaining the BALF fluid in Appendix A. Furthermore, Dr. Ulrich auf dem Keller, Dr. Anna Prudova, and Dr. Oliver Schilling are acknowledged for assistance in experimental design and data analysis in Appendix B.  xi  CHAPTER 1. INTRODUCTION 1 The Basics of Matrix Metalloproteinases The first matrix metalloproteinase (MMP) was discovered in 1962 by Jerome Gross and  Charles  Lapiere  upon the  observation that tadpole tails  undergoing  metamorphosis had the unique ability to degrade fibrillar collagen (1).  Similar  activity was then demonstrated in human gingival tissues (2). It was not until the following decade that the first human collagenase, MMP-1, was isolated (3). Since then, the number of MMPs has grown to 23 in human and 24 in mice. This family of zinc-dependent endopeptidases is proposed to degrade all components of the extracellular matrix. As such, MMPs are thought to be involved in many normal physiological processes and have traditionally been associated with pathologies characterized by increased extracellular matrix degradation, most notably cancer and arthritis. However, recent substrate discovery efforts have revealed that MMPs specifically and efficiently cleave and modulate the activities of bioactive substrates including growth factors and their receptors, cytokines and chemokines, and integrins [reviewed in (4)]. Therefore, MMPs are key regulators of many processes such  as angiogenesis,  apoptosis,  tumor metastasis and  progression, and  inflammation. The simplistic view of MMPs as “matrix bull-dozers” is outdated and drug-targeting strategies must account for the emerging pleiotropic roles of MMP5 in disease and normal homeostatic processes. MMPs are produced by a variety of cell types and are either secreted or membranebound. The general structure of mammalian MMPs consists of an auto-inhibitory pro-domain and a catalytic domain (Figure 1.1).  The pro-domain acts through a  cysteine switch mechanism where a conserved cysteine coordinates the active site zinc, thereby preventing catalysis. All MMPs except MMP-7, MMP-23, and MMP-26 1  A version of the third section of Chapter 1 has been accepted for publication as a book chapter: Cox JH and Overall CM. Cytokine Substrates: MMP Regulation of Inflammatory Signalling Molecules. The Cancer Degradome Proteases and Cancer Biology. Springer Science and Business Media. 2008 (In Press). -  1  Basic MMP-1 MMP-3 MMP-8 MMP-1O MMP-12  MMP-13 MMP-19 MMP-20 MMP-21 MMP-27  Minimal MMP-7 MMP-26  Gelatin-binding MMP-2 MMP-9  F u ri n-activated MMP-11 MMP-28  Furin-activated (Membrane-anchored)  E  ‘‘I  Fu ATALYii  I  MMP-14 (TM) MMP-15(TM) MMP-16(TM) MMP-24 (TM) MMP-17 (GPI) MMP-25 (GPI) MMP-23 (Type II, not shown)  Figure 1.1. Schematic structure of MMPs. The domain compositions of the 23 human MMPs are shown. (A) The basic structure consists of a pro-domain (Pro), catalytic domain, and hemopexin C domain (Hem), attached by a linker domain. The pro-domain cysteine (Cys) associates with the catalytic Zn 2 to inhibit activity. (B) Minimal MMPs lack the hemopexin domain. (C) Gelatin-binding MMPs have 3 fibronectin-domain inserts (Fn) in the catalytic domain. (D) Furin-activated MMP5 have a furin recognition domain (Eu) upstream of the catalytic domain. (E) Membrane-anchored MMPs are furin-activated and have C-terminal transmembrane (TM) domains with a cytoplasmic tail or alternatively are GPI-anchored. MMP-23 is a Type II transmembrane protein (not shown). 2  also contain a hemopexin C domain, which is a C-terminal four-bladed propeller structure important for protein-protein interactions and substrate recognition (5). The gelatinases, MMP-2 and MMP-9, have fibronectin type II repeats inserted into the catalytic domain to facilitate collagen binding (6). Of the six membrane-type (MT) MMPs, MMP-14, -15, -16, and -24 have transmembrane domains and short cytoplasmic  tails.  In  contrast,  MMP-17  and  MMP-25  have  glycosylphosphatidylinositol (GPI) linkages and MMP-23 is a type II transmembrane protein.  MMPs have been attractive drug targets for the treatment of cancer and arthritis since their association with these diseases several decades ago.  In the case of  cancer, strong correlations between MMP overexpression and tumor progression prompted the development of many small molecule and peptidomimetic broadspectrum MMP inhibitors, some of which advanced to phase Ill clinical trials. Universally, these trials were unsuccessful and has prompted re-evaluation of MMP5 as therapeutic targets (7). In recent years with MMP-knockout mice, the individual functions of several MMPs have been examined in cancer, as well as other diseases, uncovering some unexpected host-protective roles.  As such, several  MMPs are now defined as drug anti-targets because inhibition would cause increased disease severity (8), giving some insight into the failure of broad-spectrum antagonists.  The first such anti-target was MMP-8, the neutrophil-specific  collagenase.  The Neutrophil Collagenase: MMP-8 Neutrophils, also called polymorphonuclear cells (PMNs), are often considered the  front line of defense in innate immunity. Neutrophils express MMP-8, MMP-9, MMP 14, and MMP-25.  MMP-8, which was the second MMP identified, is produced  primarily by neutrophils and is released from specific granules at sites of inflammation (9, 10). Other cell types including macrophages (11), T lymphocytes (12), and synovial fibroblasts (13) have been reported to express low levels of MMP  3  8.  Along with MMP-1 and MMP-13, MMP-8 has traditionally been considered a  collagenase for its potent ability to degrade collagen (14), in particular type I collagen which is a major component of bone and skin. In addition to proteolysis of structural components of the extracellular matrix, several bioactive substrates of MMP-8 are also known, including chemokines, adhesion molecules, and protease inhibitors, however physiological evidence was lacking (Table 1.1). Table 1.1. Extracellular matrix and bioactive substrates of MMP-8 Known MMP-8 substrates Reference Extracellular matrix substrates Aggrecan (15) Collagen type l>type lll>type II (14) Collagen VII (16) Collagen X (16) Fibrinogen (17) Fibronectin (18) Laminin-5 (19) Nidogen (18) Proteoglycan core protein (18) Bioactive substrates Alpha-i antitrypsin (20) Alpha-2 antiplasmin (21) Angiotensin (22) Bradykinin (22) Cl inhibitor (23) CCL2/MCP-1 (24) mCXCL5/LIX (25, 26) CXCL6/GCP-2 (25) CXCL1O/lP-i0 (27) IGFBP-5 (28) L-selectin (18) Tachykinin substrate P (22) Like other MMPs, MMP-8 is secreted as an inactive pro-enzyme and is converted to an active protease through the cysteine switch mechanism.  This can occur by  reactive oxidative species released from activated neutrophils (29), or proteolytically by cathepsin G and chymotrypsin (30), as well as a number of MMPs [reviewed in (31)]. However, it is unclear how MMP-8 becomes activated in vivo or how MMP-8 retains activity in the presence of endogenous inhibitors such as the tissue inhibitors  4  of metalloproteinases (TIMPs).  Although MMP-8 is secreted through traditional  pathways, it is now thought to function in a membrane-associated state which confers increased resistance to inhibition (32). MMP-8 is often associated with the pathogenesis of diseases characterized by excessive neutrophil infiltration such as rheumatoid arthritis (33), cystic fibrosis (34), and periodontal disease (35), but the biological role of MMP-8 remains uncertain. The MMP-8 knockout mouse, generated by Balbin et al., exhibits normal development and fertility but surprisingly has increased susceptibility to skin cancer (26), thus defining MMP-8 as the first host-protective MMP in cancer.  In MMP-8  deficient mice, there was an initial delay in neutrophil influx to the site of chemical carcinogen treatment followed by a massive neutrophil accumulation that persisted. Other phenotypes with the MMP-8 knockout mouse (summarized in Table 1.2, Page 8) include increased PMNs in LPS and allergen-induced airway inflammation (32, 36), resistance to TNF-induced lethal hepatitis (37), delayed wound healing (38), and less severe experimental autoimmune encephalomyelitis (EAE) (39).  It is  becoming evident that MMP-8 has pleiotropic roles in various diseases and inflammatory situations yet the specific mechanisms underlying these phenotypes remain poorly understood. Cytokine Substrates: MMP Regulation of Inflammatory Signaling Molecules Perspective Inflammation is a highly regulated process involving tissue parenchyma, vascular and connective tissues, as well as host defense and immune cells, and is an essential component of both innate and acquired immunity. MMPs were traditionally considered to assist in the inflammatory response through the breakdown of extracellular matrix molecules, providing an environment condusive to leukocyte infiltration, edema, wound healing and tissue repair with continued upregulation contributing to the onset of chronic inflammation. It is now evident, however, that the repertoire of MMP substrates, the substrate degradome, extends far beyond the extracellular matrix and includes many key bioactive molecules, the processing of  5  which results in altered actions. In the context of inflammation, several MMPs have been shown biochemically to efficiently and selectively process cytokines and chemokines, as well as associated molecules, thereby changing their bioactive properties. Many of these substrates have been validated physiologically in genetic models of inflammatory disease, where a remarkable array of inflammatory phenotypes has been observed.  As a whole, research involving MMPs in  inflammation has demonstrated their integral role in orchestrating both the initiation and termination of inflammatory processes, rather than only tissue degradation. MMPs and Inflammation Inflammation is an essential process for the elimination of invading pathogens and in wound healing.  Inflamed tissue is a dynamic interactive multi-cellular and multi-  tissue system embedded in normal stroma. A wide variety of signaling networks operate in multi-directional communication between the component cells and tissues—the connective tissue cells and epithelium, vascular endothelium, reactive peripheral stroma, and host defense cells—that orchestrate innate immunity, inflammation, resolution and healing.  Cytokine pathways that are perturbed in  unresolved acute inflammation drive tissue destruction and inflammatory/immune cell responses in chronic inflammatory disease.  Dysregulated inflammation has  been linked to numerous pathologies including autoimmune diseases (40), tumorigenesis  (41),  and atherosclerosis (42).  Stringent regulation  of the  inflammatory process is critical for both resolution and to buffer against pathological destruction. Proteolytic enzymes are now regarded as key effectors and regulators in inflammation and immunity, through not only protein degradation and turnover but also precise and efficient cleavage events leading to altered bioactivity of inflammatory mediators (43, 44).  Matrix metalloproteinases (MMP5) were historically thought to be exclusively involved in turnover of extracellular matrix components.  Accordingly, attractive  hypotheses evolved that involved matrix degradation in normal development and  6  physiological turnover as well as in many of the steps in inflammation that appeared to be dependent on extracellular matrix turnover.  However, with the exception of  MMP-14, none of the MMP-deficient mice generated thus far exhibit severe defects in matrix turnover for normal development or homeostasis. In addition, drug trials with broad-spectrum MMP inhibitors have not demonstrated significant abnormalities in extracellular matrix although a small number of patients did develop joint strictures (45).  In contrast, MMP-deficient mice display a remarkable array of inflammatory  and immune phenotypes (Table 1.2).  Although MMPs have generally been  considered pro-inflammatory molecules, promoting leukocyte migration through the breakdown of basement membrane and extracellular matrix, the observed phenotypes clearly show that MMPs can also function in an anti-inflammatory capacity and the mechanisms underlying these phenotypes are now beginning to emerge. As such, it is now evident that matrix metalloproteinases have a multitude of pleotropic functions in regulating inflammation.  In recent years, substrate discovery efforts have uncovered a plethora of novel bioactive substrates that link MMPs with both the onset and the resolution of inflammation (4, 46). Notably, MMPs are now known to have pro- and anti-target effects in cancer (8), and analogous opposing roles appear to exist in inflammation as well. As such, several authors have proposed that MMPs, by playing critical roles in regulating the activity of numerous inflammatory mediators, are homeostatic regulators and controllers of inflammation. Here we describe the conclusions from genetic models of inflammation as well as biochemical and proteomic evidence of the direct proteolytic actions of MMP5 on cytokines and chemokines, and also indirect effects of MMPs on the inflammatory response. Genetic Models of Inflammation Murine models of inflammation and innate immunity have demonstrated diverse and unexpected functions of MMPs in the inflammatory context (Table 1.2). Currently, knockout mice have been generated for at least fifteen MMP5: MMP-2, 3, 7, 8, 9, 10,  7  11, 12, 13, 14, 19, 20, 23, 24, and 28 (47). Of these, MMP-2, 3,7,8,9, 10, 12, and 28 have been reported to have altered inflammatory phenotypes.  Due to the  significant skeletal defects and short life-span of the MMP-14-null mice, inflammatory studies using these mice are not feasible, however organ transplantation allows for tissue specific evaluation of MM P-i 4 deficiency. Table 1.2. Inflammatory and immune phenotypes of MMP-deficient mice Mouse Phenotypes relating to inflammatory processes References Mmp2’  Mmp3’  Mmp7’  MmpB  Mmp’  More severe antibody induced arthritis Diminished egress of lung inflammatory cells Earlier and more severe EAE due to increased MMP-9 Prolonged cardiac allograft survival and lower cellular infiltration More severe myocardial inflammation and dysfunction Delay in inflammation associated corneal neovascularization Exacerbated experimental colitis Markedly impaired contact hypersensitivity Reduced disc resorption and generation of macrophage chemoattractant Reduced PMN recruitment in acute lung injury Reduced macrophages in atherosclerotic plaques Impaired clearance of intestinal bacteria Delayed macrophage and lymphocyte recruitment in cardiac remodelling Marked resistance to S. typhimurium infection Increased bacterial survival, lack of intestinal a-defensiris Decreased epithelial cell apoptosis due to reduced FasL shedding Reduced macrophages in herniated discs, loss of TNFa shedding Reduced PMNs in lung due to altered chemokine gradients Less severe septic arthritis Delayed PMN infiltration then accumulation in skin carcinogenesis model Increased PMN infiltration in intratracheal LPS model Increased PMN infiltration in allergen induced airway inflammation Resistant to TN F-induced lethal hepatitis Decreased PMN infiltration in LPS-treated air pouch Increased inflammation delays wound healing Decreased susceptibility in experimental autoimmune encephalomyelitis Decreased PMN infiltration in LPS-treated corneas More persistant inflammation in contact hypersensitivity Less susceptible to experimental autoimmune encephalomyelitis Resistant to subepidermal blisters with reduced PMN recruitment Reduced macrophages in experimental cardiac infarction Impaired hematopoietic stem cell motility due to reduced sKitL Less severe antibody-induced arthritis Impaired cellular infiltration in airway inflammation Impaired defense in bacterial meningitis Reduced hematopoietic stem cell mobilization Impaired dendritic cell migration through tracheal epithelium Enhanced allergen-induced airway inflammation Inflammatory cell accumulation in allergic lung model Attenuated dextran sulphate-induced colitis Increased severity of septic arthritis Reduced resistance against E. coli peritonitis  (48) (49) (50) (51) (52) (53) (54) (55) (56) (57) (58) (59) (60) (61) (62) (63) (64) (65) (66) (26) (32) (36) (37) (67) (38) (39) (68) (55) (69) (70) (71) (72) (48) (73) (74) (75) (76) (77) (78) (79) (80) (81)  8  Table 1.2. (continued) Mouse Phenotypes relating to inflammatory processes MmplU’ Increased inflammation in models of infection and wound healing Mmpl2’ Reduced macrophages in smoke-induced emphysema  Mmp26’  References (82)  Less TNFa shedding in smoke-induced inflammation Improved recovery after spinal cord injury, reduced inflammation Reduction in OVA-induced airway eosinophilia Reduction in antigen-induced airway inflammation Reduced infiltrating macrophages in aortic aneurysms Less experimental autoimmune encephalomyelitis, higher Thl/Th2 ratio Reduced macrophages in atherosclerotic plaques Decreased macrophages in ligament healing No change in bleomycin-induced lung fibrosis BALF lacks chemotactic elastin fragments Increased inflammatory response  (83) (84) (85) (86) (87) (88) (89) (90) (91) (92) (93) (W.C. Parks, Unpublished)  Individual MMPs show remarkably divergent physiological functions in inflammatory disease models.  For instance, Mmp2-null mice have increased inflammation and  consequent disease severity in models of arthritis (48), experimental autoimmune encephalomyelitis (50) and experimental colitis (54), suggesting a general antiinflammatory role for the enzyme. This protective role of MMP-2 is consistent with its near-constitutive expression, where a pro-inflammatory role would be very destructive.  In contrast, MMP-7 deficient mice exhibit reduced inflammation in  models of septic arthritis (66), herniated discs (64) and acute lung injury (65), demonstrating a pro-inflammatory function for MMP-7.  MMP-3 and MMP-12 can  also be generalized as pro-inflammatory effector molecules as these knockouts are characterized by decreased infiltrating leukocytes in several disease models (summarized in Table 1.2).  The functions of certain MMPs cannot be classified so easily.  For example, the  neutrophil protease MMP-8 seems to have opposing effects in acute and chronic inflammatory situations. In acute models such as the LPS-treated air pouch (67) and TNF-induced hepatitis (37), MMP-8 promotes neutrophil infiltration, at least in part due to ELW chemokine processing and activation. In more chronic models of skin carcinogenesis (26) and wound healing (38), MMP-8 is involved in terminating and  9  resolving inflammation.  MMP-9 also demonstrates complex regulatory properties.  For instance, Mmp9-null mice have reduced PMN recruitment in subepidermal blisters (70) and decreased macrophages in response to cardiac infarction (71), but conversely show enhanced allergen-induced airway inflammation (78) and increased severity in a model of septic arthritis (80). Therefore both MMP-8 and MMP-9 can exert differential regulatory effects depending on the particular stimulus and the temporal status. However, in the neutrophil MMP-8 dominates as its absence leads to decreased neutrophil infiltration despite normal or elevated MMP-9 levels. MMP Regulation of Cytokines Cytokines have a central role straddling both innate and adaptive immunity, functioning in both pro-inflammatory and anti-inflammatory manners. Cytokines are produced by a variety of cell types and signal either in cis or in trans to nearby cells or throughout the organism. Through the binding of specific cell-surface receptors, cytokines trigger signalling cascades resulting in altered gene and protein expression during an inflammatory reaction. Therefore, precise control of cytokine activity is essential for the accurate control of inflammation. Selective proteolysis of cytokines provides a rapid and effective means to modulate the initiation and termination of an inflammatory response. Interleukin- 1/3 lnterleukin-1 I is an early and potent pro-inflammatory cytokine that has been linked to fever, leukocytosis, anemia, and elevated acute phase proteins (94). IL-i  ,  along  with TNF-a, are classical inducers of MMPs in numerous cell types. The production of IL-i 1 is primarily in macrophages, monocytes and PMNs, where secretion occurs through non-classical pathways as IL-i 3 lacks a secretory signal peptide (95). It is translated as an inactive 33-kDa precursor, some of which seems to be secreted through specialized lysosomes.  Caspase-i, also known as IL-i 3-converting  enzyme/ICE, is an intracellular cysteine protease that cleaves latent IL-i 13 at Asp 116 7 releasing the N-terminus to generate the fully active i7-kDa molecule (96). Ala 10  Although caspase-1 is thought to play a dominant role in IL-i 3 activation, studies of local inflammation with caspase-1 deficient mice suggest that alternate mechanisms exist (97).  IL-i 1 processing by MMPs has been described in vitro, generating biologically active 1 7-kDa forms (98). Activation occurs with the following MMPs in order of increasing efficiency: MMP-2, MMP-3, and MMP-9, whereas MMP-1 lacks the ability to activate latent IL-i  (Table 1.3). With prolonged exposure and high enzyme concentrations,  MMP-i, -2, -3, and -9 degrade IL-i f yielding a loss in activity (98, 99). Hence, the physiological role of MMPs in the regulation of IL-i 3 activity remains to be determined, but biochemical studies appear to show the potential for both proinflammatory and anti-inflammatory effects in vivo.  The important caveat that  biochemical cleavage does not equate to a physiologically relevant substrate must be recognized, and in the case of prolL-i(3, which is normally processed intracellularly, substrate-enzyme colocalization must be considered. Similarly, high enzyme-substrate ratios reflect extreme parameters that may not be met in vivo. Regardless, IL-i 1 induced expression of MMPs possibly involves feed-forward or feed-back loops by which MMPs might further activate the cytokine, enhancing inflammation, or degrade lL-i3 in an inhibitory loop to dampen or terminate the response.  Tumor necrosis factor-a Tumor necrosis factor-a (TNF-a) is a multifunctional cytokine with pro-inflammatory and immunomodulatory roles (100). Overproduction of TNF-a has been implicated in inflammatory conditions such as rheumatoid arthritis, ankylosing spondylitis and psoriasis (101). membrane-associated  multiple sclerosis,  TNF-a is produced as a 26-kDa  inactive precursor (proTNF-a) that is activated upon  proteolytic shedding, thus releasing the soluble form.  This cleavage was first  attributed to metalloproteinases based on broad-spectrum metalloprotease inhibitors  11  Table 1.3. Bioactive MMP substrates: The pro-inflammatory and antiinflammatory proteolysis events by MMPs Substrate MMPs Cleavage site References Pro-inflammatory cleavages lnterIeukin-1 proTNF-a  CXCL5 CXCL8 CL1 3 CX LIX (mCXCL5)  2, 3, 9 1,2,3,7, 9 12 14, 15 17 1 8, 9 1,8, 9, 13, 14 2 1,2,8,9, 12, 13  ND ND SLISPLA,IQA,VRSS 68 SLlSP.j,LAQAVR 68 SLISPLAJ,QAVR 68 AAAVLR,J,ELRC 4 AA,Aj,VLRELRC 4 AVLPR,j,SAKE 1 QAAA,j,LTKN 68 APSSVlAA 1  (98) (102) (103) (104, 105) (106) (25, 67) (67, 105, 107) (108) (25, 26, 67, 109)  Anti-inflammatory cleavages InterIeukin-1  1,2,3,9  Degradation  proTNF-c LatentTGF-13 CCL2 CCL7 CCL8 CCL13 CXCL1  14, 15 2,3,9, 13, 14 1,3, 8, 12 1,2,3, 12, 13, 14 3, 12 1,3, 12 9 12 12 12 9 12 12 9 12 1,2,3,9,13,14 2 12 12 9, 12  Degradation ND QPDAlNAP 1 QPVG,IINTS 1 QPDSVSIP 1 QPDALNVP 1 Degradation VATELRCQ 3 LATELRCQ 3 VVTE,LRCQ 3 Degradation VLREILRCV 7 RCTCI,LRVT 11 Degradation AKELRCQC 7 KPVSLSYR 1 QHLGJ,MRKC 1 IANE,j,LRCQ 3 VASELRCQ 3 VASEILRCQ 3  CXCL2 CXCL3 CXCL4 CXCL5 CXCL6 CXCL7 CXCL8 CXCL12 CL1 3 CX mCXCL1 mCXCL2 mCXCL3  (99) (98) (104) (110-113) (24, 109) (24, 109, 114) (24, 109) (24, 109) (107) (109) (109) (109) (107) (109) (109) (107) (109) (115) (108) (109) (109) (109)  Unaltered bloactivity cleavages CXCL6 CXCL9 CXCL1O LIX (mCXCL5)  8,9 9 8,9 8 1,9, 12  ‘GPVSI,A,j,VJ,LTELR VLKJ,VRK,,SJ,QRSR 89 KAVJ,SKE,j,MSI,KRSP 66 KKAKJ,RNAL 76 NALA,j,VERT 81  (25) (27) (27) (67) (109)  12  preventing the shedding of TNF-a in cultured leukocytes and endotoxemia models (116-118). However, cloning and purification studies revealed that the major enzyme responsible is the metalloproteinase disintegrin ADAM-17, also called TNF-a converting enzyme (TACE) (119, 120).  Although ADAM-17 is considered to be the major generator of soluble TNF-a, studies with MMP-7 and MMP-12 knockout mice in models of herniated disc resorption (64) and cigarette smoke-induced inflammation (84), respectively, support a role for MMPs in the physiological release of TNF-a.  In the case of MMP-7,  intervertebral discs cocultured with peritoneal macrophages from Mmp7-null mice had a defect in macrophage influx. This was attributed to a decrease in soluble TNF a, which was found to be necessary for MMP-3 induction and the consequent generation  of  a  macrophage  chemoaffractant.  In  cigarette  smoke-induced  inflammation, MMP-12 knockout mice have decreased levels of soluble TNF-a and diminished ability to shed TNF-a from cultured alveolar macrophages in response to smoke exposure. Further, MMP-12 mediated shedding of TNF-a was found to promote endothelial activation, PMN infiltration and the proteolytic tissue damage associated with emphysema.  Biochemically, several MMPs can cleave proTNF-a to generate the active form in vitro (Figure 1.2).  For example, purified MMP-1, MMP-2, MMP-3, MMP-7, and  MMP-9 can cleave recombinant GST-TNF-a fusion protein (102, 118). In addition, MMP-12 (103), MMP-14 and MMP-15 (104), and MMP-17 (106) are capable of cleaving proTNF-a.  In a proteomic screen evaluating breast cancer cells  overexpressing MTI-MMP, TNF-a was increased in the conditioned medium in the MMP-14 transfected cells compared to vector control and biochemically MMP-14 was shown to cleave and generate the active form (105). The MMP cleavage sites in proTNF-a are variable, but lie either at or slightly upstream of the ADAM-17 cleavage site of AIa -VaV (Table 1.3). 76  Notably, ADAM-17 is stimulated by  Iipopolysaccharide or experimentally by phorbol myristyl acetate, where its role in 13  Figure 1.2. MMP proteolysis of cytokines and chemokines to precisely control the induction (left) and termination (right) of inflammation Initiation of an inflammatory response (left) is triggered by injury or an infectious agent resulting in the rapid local release of neutrophil chemoattractants and proteases, including MMP5. Circulating neutrophils roll on the endothelial surface through interactions with selectins (not shown). Engagement with chemokines, immobilized by glycosaminoglycan binding, causes firm adhesion to the endothelium through integrins and subsequent extravasation into the underlying tissue. Once in the tissue, neutrophils are activated and degranulate releasing MMP-8 and MMP-9. MMP-8 processes and activates CXCL5 and CXCL8, resulting in additional PMN recruitment. Monocyte chemoattractant proteins (CCL2, 7, 8, 13) and CXCLI2 promote the influx of monocytes and lymphocytes, respectively. The membranebound chemokine CX CL1 is shed by fibroblast-derived MMP-2 to generate a 3 soluble chemoattractant. Macrophage-derived MMP-7 and MMP-12 are involved in shedding TNF-a, in cooperation with ADAM-17, yielding a soluble pro-inflammatory cytokine. TNF-a is also shed by MMP-14, depicted here on fibroblast cells. Once the initiating stimulus is removed the pro-inflammatory signals are no longer generated (right). Local chemokine production is decreased and neutrophils no longer enter the tissue. Furthermore, some neutrophil chemokines (CXCLI, CXCL4, CXCL7) are degraded by MMP-9 resulting in inactivation. MMP-2, as well as other MMPs, is involved in dampening the response through the generation of chemokine antagonists. Shown here are CCL2, CCL7, CCL8, CCLI3, and CX CL1 as 3 substrates of MMPs, forming potent receptor antagonists and inhibiting chemotactic migration. Also, CXCLI2 is processed by MMP-2 resulting in a switch in receptor specificity. MMP-14 is potentially involved in down-regulating TNF-a through degradation. Furthermore, MMP-2 and MMP-9 (not shown) cleave the latency associated protein from TGF-13 to liberate the active TGF-13 homodimer with immunosuppressive properties. Active TGF-3 induces MMP-2 and MMP-9 expression and extracellular matrix formation in a feedback mechanism.  14  IJFjT1E iEJI_zi Neutrophil  Neutrophil  12  i F 1 *  •CXCR1 /2  ::‘  GAGs  -  Z1JrJLJ  [a iciwi  Endothelial layer  Endothelial layer  I Macrophage  •  •  Active CXCL5 CXCL8  Macrophage  4 MMP-8 MMP-9  .4—  MMP-7 MMP-i 2  ‘S UI.  -J  Active TNFa cCR4.  MMP-8 MMP-9  CC  Degraded’ CXCL1 CXCL4 CXCL7 MMP-7 MMP-i2  Degraded TNFa  C  —— —  _,JproTNFa CCL2 CCL7 CCL8 CCL13  ADAM-i 7 S.  Antagonist CCL2 CCL7 CCL13  ADAMi7 CCR  MMP-2  zv:3acist MMP-2 MMP-14  . CXCL12  . .  Fibroblasts Latent  Lymphocyte  TGFI3  Initiation  Lymphocyte  + Fibroblasts  Loss of agonist CXCL12  c  MMP-2  Active TGF  Termination  15  TNF-cL shedding is well established. Conversely, in the absence of LPS stimulation such as non-infective situations, MMP-mediated shedding of TNF-a may be important and as such a variety of experimental approaches at different levels of complexity are needed to determine the relative contibutions of various proteases (121). It is likely that ADAM-17 and several MMPs have complementary roles in the rapid shedding and activation of proTNF-a in response to varying stimuli.  Transforming growth factor-/31  Transforming  growth  factor-31  (TGF-1)  is  generally  considered  an  immunosuppressive cytokine but encompasses a multitude of functions that cannot be defined so simply.  TGF-131 is known for its ability to instigate and maintain  immune tolerance (122, 123). Under normal circumstances, TGF-f31 is secreted in a latent form where its furin-cleaved 80-kDa N-terminus, also called the latencyassociated peptide, remains non-covalently associated and requires proteolytic, conformational, or acidic conditions for its removal (124). MMP proteolysis, amongst other mechanisms, has been proposed to release the active 25-kDa cytokine homodimer from the latency-associated peptide.  In chondrocyle cultures, MMP-3  (111) and MMP-13 (113) have been found to activate TGF-31 and likewise MMP-14 expression in osteoblasts causes activation of the latent cytokine and thereby promotes osteoblast survival (110). In the context of tumor biology, the gelatinases MMP-2 and MMP-9 were found to activate TGF-13, potentially promoting tumor invasion and angiogenesis (112). Notably, TGF-31 induces the expression of MMP 2 (125, 126) and MMP-9 (127), perhaps contributing to the potentiation of TGF-1 activity in a feed-forward manner. Furthermore, TGF-131 stimulates production of the extracellular matrix, a critical event in wound healing, and through increased collagen deposition leads to induction (128) and activation of MMP-2 (129) and consequent processing of monocyte chemoattractants (CCL2, 7, 8, and 13) to generate potent anti-inflammatory molecules, thereby terminating macrophage influx and promoting healing (24). Although the link between MMPs and TGF-fE has yet to be clearly determined in an inflammatory setting, MMP activation of TGF-3 could 16  provide a novel anti-inflammatory mechanism by which MMPs function to dampen an immune response. MMP Regulation of Chemokines Chemokines are a super-family of low molecular weight chemotactic cytokines that function in directing the migration of cells in normal leukocyte trafficking and the inflammatory response (130).  In vivo, chemokines form gradients through  interactions  glycosaminoglycans  with  proteoglycan  and  signal  through  7-  transmembrane G protein-coupled receptors to induce a chemotactic response. Several chemokines are known to be regulated by proteolysis, many of which are MMP-mediated cleavages, resulting in a multitude of functional consequences including either enhanced or decreased receptor binding, conversion to a receptor antagonist, changing receptor specificity, and shedding of membrane-anchored chemokines.  Chemokines are characterized by the spacing of conserved cysteine residues that function in the formation of disuiphide bonds.  Although the sequence similarity  amongst chemokines is quite low, ranging from 20-50% homology, the tertiary structure of these proteins is remarkably conserved (131). The general structure consists of an undefined N-terminus, three central anti-parallel beta-sheets, and a Cterminal alpha-helix.  The N-terminal portion of chemokines appears to be most  susceptible to proteolysis and cleavage often results in significant changes in activity. In contrast, some C-terminal processing events have been described but no functional changes have been ascribed to these truncations (27, 67, 132).  CC Chemokines The CC family of chemokines is defined by adjacent cysteine residues in the N terminus and its members typically induce the migration of monocytes, dendritic cells and natural killer cells. The monocyte chemoattractant proteins (MCPs), also known as CCL2, 7, 8, 13, were amongst the first chemokines to be identified as MMP  17  substrates. In a yeast genetic screen with the hemopexin domain of MMP-2, CCL7 was found to have a strong interaction with the exosite and was processed efficiently at G1y 5 4 11e with recombinant enzyme and by monocytes in culture and in vivo (114, 133). The truncated chemokine, CCL7 (5-76) still binds its cognate CC receptors but no longer promotes chemotaxis and instead functions as a receptor antagonist with potent anti-inflammatory properties in vivo, as determined in mouse models of subcutaneous and peritoneal inflammation. CCL7 is also processed by MMP-1, 3, 13, and 14 (24) and the cleaved form has been detected in rheumatoid synovial fluid (114).  All of the MCPs have been identified as MMP substrates (Table 1.3) with characteristic cleavage between residues 4-5 in the N-terminus, causing diminished activity and the generation of potent receptor antagonists (24). In addition to MMP 2, MMP-1, -3, -8, -13, and -14 cleave MCPs with different profiles of substrate preference, revealing specificity in this interaction (Table 1.3).  Notably, this work  was the first to demonstrate an unexpected dampening of inflammation by MMP5, traditionally considered to be pro-inflammatory proteases. CXC Chemokines The CXC chemokines predominantly influence the migration of polymorphonuclear neutrophils and T-lymphocytes. A subset of CXC chemokines are characterized by an ELR (glutamic acid-leucine-arginine) sequence proximal to the conserved CXC motif and act exclusively on CXCR1 and CXCR2 receptors, thereby promoting neutrophil chemotaxis.  CXCL8, the most potent of the ELW chemokines, is  processed by MMP-1, -8, -9, -13, and -14 at 6 -Ser causing a 10-fold increase in 5 Arg , receptor binding affinity and chemotactic activity (67, 105, 107).  In the cellular  context, MMP-1 4-overexpressing cells were discovered to have decreased levels of CXCL8-derived peptides in the conditioned medium compared to vector controls, confirmed biochemically to be due to proteolysis (105).  18  Activation of PMN chemoattractants also applies to the human ELW chemokine CXCL5 as well as the murine chemokine LIX (25, 26, 67). LIX, also called mCXCL5, is thought to be the mouse orthologue to human CXCL8 in that it is the most potent and abundant of the ELR chemokines. LIX is efficiently cleaved at Ser 5 4 Val by the neutrophil enzyme  MMP-9  chemoattractant activity (25).  in  vitro,  resulting  in  a  significant  increase  in  The generation of the MMP-8 deficient mouse  revealed decreased processing of LIX in LPS-stimulated bronchoalveolar lavage fluid, suggesting a specific role for MMP-8 in the activation of LIX (26).  This  observation is coupled with delayed neutrophil infiltration to the site of challenge in a skin carcinogenesis model in Mmp8-nuIl mice, followed by a massive accumulation of neutrophils.  Furthermore, in an LPS-treated subcutaneous air pouch model of  acute inflammation, Mmp8-null mice have a significantly decreased neutrophil influx compared to wild type controls (67). This difference is lost when mice are treated with the truncated chemokines LIX (5-92) or CXCL8 (6-77), demonstrating that LIX and CXCL8 are physiologically relevant substrates of MMP-8 and that the normal cell migration machinery and the ability of neutrophils to migrate through basement membrane and connective tissue is intact in the absence of MMP-8. Interestingly, several MMPs in addition to MMP-8 cleave LIX in vitro, yet animal studies indicate a lack of physiological redundancy in this pathway (67). This result also questions the importance of MMP-8 in collagen degradation in these matrices, a role traditionally assumed to be essential for neutrophil chemotaxis and extravasation.  There is mounting evidence to support the notion that MMP5 are involved in promoting  PMN  infiltration  via  the  proteolytic  activation  of  neutrophil  chemoattractants and recent data suggests an opposing role for MMPs in the inactivation of ELR chemokines. The activation mechanism for ELR chemokines differs from the majority of chemokines, where N-terminal truncation usually results in the loss of agonism coupled with the generation of potent receptor antagonists. MMP-1 2 cleaves multiple murine and human ELW chemokines within the ELR motif (Table 1.3), thereby decreasing activity (109). This cleavage was reflected  19  physiologically by increased neutrophil accumulation in LPS-treated lungs of MMP 12 knockout mice suggesting that macrophage-derived MMP-12 is involved in terminating neutrophil infiltration.  Likewise, MMP-9 has been reported to slowly  degrade the ELW chemokines CXCL1, CXCL4,  and CXCL7 in vitro, but  physiological evidence of this interaction is lacking (107).  CXCL1 2, also known as SDF-1, is a multifunctional chemokine with a critical role in development as well as a potent chemoattractant of T lymphocytes, monocytes, and CD34 stem cells. Proteolysis of CXCL12 occurs at Ser 5 4 Leu by MMP-1, -2, -3, -9,  -  13, and -14, resulting in a loss of binding to the cognate receptors CXCR4 (115) and CXCR7 (134) and therefore loss of HIV protection and decreased chemotactic properties.  Interestingly, the cleavage product, CXCL12 (5-67), is a potent and  specific neurotoxin (135), an effect shown to be mediated through CXCR3 binding and signalling (136), demonstrating for the first time that proteolysis of a chemokine can cause a shift in receptor specificity. Membrane-Bound Chemokines CX C 3 L1, also known as fractalkine, is unique amongst the chemokine superfamily in that it can exist in both membrane-anchored and soluble forms, giving it the dual functionality of an adhesion molecule and a chemoattractant (137). When shed from the membrane, 3 CX C L1 is a potent chemoattractant for T cells and monocytes. Disintegrin-like metalloproteinases ADAM-b  and ADAM-17 have been reported to  shed 3 CX C L1 by cleaving adjacent to the transmembrane region, hence regulating CX C 3 L1 -mediated cell-cell adhesion and chemoattractant activity (138, 139).  A  proteomic approach recently identified 3 CX C L1 as an MMP-2 substrate, where endogenous 3 CX C L1 was found to be increased in the conditioned medium of MMP 2 expressing fibroblasts compared to Mmp2-null control cells (108). Tandem mass spectrometry coupled with peptide mapping enabled the identification of two cleavage sites: A1a 72 in the C-terminus of the chemokine ectodomain, yielding -Leu 71 a soluble chemoattractant, and 5 -Met a truncation known to result in potent 4 Gly ,  20  antagonism of the receptor 3 CX C R (140). Hence, it appears that MMP-2 can both activate and inactivate a single chemokine, allowing for precise biphasic regulation of CX CL1 activity in the conversion of a cell surface adhesion molecule/agonist to a 3 soluble agonist and antagonist depending on the site of cleavage.  The only other  membrane-anchored chemokine, CXCL16, is also shed in a metalloproteinase dependent manner. Solubilization of CXCL1 6 is inducible by TNF-a and significantly decreased upon treatment with GM6001, a broad-spectrum MMP and ADAM inhibitor, but the specific proteases involved are yet to be identified (141). Notably, in genetic and inhibitor studies, ADAM-i 0 and gamma-secretases were found to be major components of proteolytic cascades leading to CX CL1 and CXCL1 6 shedding 3 (142).  Proteoglycan-mediated chemokine regulation Direct processing of chemokines themselves is not the only way by which MMPs regulate chemokine activity. Many chemokines are known to require binding to the glycosaminoglycan side chains of proteoglycans for the formation of gradients and chemotactic activity in vivo (143, 144).  As such, altering the proteolglycans  themselves can affect chemokine gradients and physiological functions.  The  clearest example of such an interaction came from a study showing that MMP-7 deficient mice have altered neutrophil infiltration in lung inflammation, where neutrophils remain confined to the interstitium without advancing to the alveolar space (65).  This defect was attributed to MMP-7 dependent shedding of the  syndecan-i ectodomain complexed with mCXCL1/KC, a murine ELR chemokine, a process required to direct and confine neutrophils to the site of injury. Hence, in the absence of MMP-7, neutrophils are unable to enter the lumen of the lung because the appropriate KC-syndecan-1 gradient has not been generated. Notably, sydecan 1 is also shed by MMP-14 and MMP-16 resulting in enhanced migration in vitro (145), potentially representing analogous mechanisms involving other MMPs.  21  Additional MMPs have also been suggested to affect chemokine activity through altering interactions with the extracellular matrix. In a model of TNF-induced acute hepatitis, MMP-8 deficient mice show improved survival coupled with defective neutrophil recruitment at 6 hours post-treatment, compared to wildtype controls (37). This phenotype was proposed to be due to the indirect effect of MMP-8 dependent release of LIX from the extracellular matrix, although direct processing of LIX and modulation of activity has also been observed (25, 26, 67).  Furthermore, MMP-8  processing of LIX does not alter glycosaminoglycan binding, as determined with heparin sulphate affinity assays (67). Regardless, potential regulation of chemokine interactions with proteoglycans in the formation of gradients should be appreciated when considering chemokine bioactivity in vivo. Future perspectives Deciphering the individual functions of MMPs in the regulation of inflammatory processes is an overwhelming task but the importance of the field is highlighted by failed clinical trials with MMP inhibitors and the coinciding unpredicted side effects (7, 146-149).  It is now evident that the physiological roles of MMPs and their  respective underlying mechanisms need to be further defined before attempting to inhibit individual MMPs for clinical applications. It is critical that only well established targets are inhibited and anti-targets are completely avoided (8). With this approach the detrimental actions of MMPs are blocked while still preserving the beneficial antitarget MMP activities.  As such, animal model and cell-based approaches  complemented by biochemical and proteomic studies will be instrumental in achieving this objective.  Current methods of MMP substrate discovery are predominantly hypothesis-driven, which is a very time-consuming approach. However, more systematic approaches to uncover novel MMP substrates and downstream effects of MMP activity are beginning to emerge.  For instance, two-dimensional gel electrophoresis was  recently used in the analysis of bronchoalveolar lavage fluid of allergen-challenged  22  Mmp2 ‘/Mmp9 mice, leading to the identification of Ymi, Si 00A8, and Si 00A9 as novel MMP substrates (150).  Furthermore, in the cellular context, MMP-14 and  MMP-2 transfected cells have been used to identify a variety of new bioactive substrates utilizing mass spectrometry-based proteomic analysis (105, 108, 151). This work has been extended with the use of MMP inhibitors, thus uncovering additional substrates with the potential to identify adverse inhibitor effects (152). The remarkable progress in proteomic approaches, such as those listed above, will undoubtedly have far-reaching implications in the field of MMP research, as well as that concerning all proteases (153). Applying such techniques to in vivo models of inflammation will enable rapid and unbiased identification of physiologically relevant substrates and hence a deeper understanding of the numerous inflammation-related phenotypes described here. Regardless of the specific mechanisms, it is undeniable that  MMPs  are  key  pleiotropic  regulators  of inflammation  and  immunity,  demonstrating a wide array of pro-inflammatory and anti-inflammatory functions, orchestrating both the onset and termination of inflammation.  Hence, the former  view that the major role of MMPs is to cleave extracellular matrix is no longer tenable. Rather, MMPs are key signalling proteases involved in normal homeostasis and initiating and terminating inflammation through efficient and specific processing of inflammatory mediators.  Theme and Hypotheses The broad focus of the research presented herein is the role of MMPs, predominantly the neutrophil collagenase MMP-8, in regulating inflammation and immunity. At the time this thesis research began, the MMP-8 knockout mouse had been generated and was more susceptible to skin carcinogenesis, making MMP-8 the first MMP anti-target in cancer (26). MMP-8 deficient mice had an initial delay in neutrophil infiltration followed by a massive accumulation over time, presumably providing a microenvironment conducive to tumorigenesis (41).  The following  themes and hypotheses are addressed in this thesis.  23  i) In Chapter 2 we hypothesized that the mechanism of delayed neutrophil infiltration in MMP-8 deficient mice involved LIXImCXCL5 chemokine proteolysis and activation. We tested this hypothesis using acute inflammatory models with MMP-8 deficient mice and these findings were published in PLoS ONE. ii) Next, in Chapter 3 we hypothesized that MMPs would also process interferon-y inducible CXCR3-binding chemokines and provide proteolytic regulation of T cell chemoattraction.  This was evaluated with recombinant MMPs and synthetic  chemokines, where we discovered that MMPs cleave both the N- and C-terminus of CXCLI 1, resulting in a switch from a receptor agonist to antagonist and altered glycosaminoglycan binding, respectively. This study was published in the Journal of Biological Chemistry. iii)  In Chapter 4, we hypothesized that MMP-8 would be protective in a murine  model of arthritis. To accomplish this objective, an MMP-8 deficient mouse strain naturally susceptible to arthritis and other autoimmunities was created by backcross breeding.  We also examined the underlying mechanisms by gene microarray  analysis.  Here we found that neutrophils from MMP-8 deficient mice undergo  delayed apoptosis, which we propose is due to abolished caspase-il levels and decreased TNF-a shedding.  The resulting manuscript has been submitted for  publication. Appendix A describes unpublished results relating to indirect MMP-8 regulation of LIXImCXCL5 and Appendix B reports cleavage specificity of mouse and human MMP8 and identification of MMP-8 substrates. Taken together, this research has advanced our knowledge of the function and mechanism by which MMP-8 regulates both the onset and resolution of inflammation.  24  References 1.  Gross, J., and C. M. Lapiere. 1962. Collagenolytic activity in amphibian tissues: a tissue culture assay. Proc NatlAcad Sd U SA 48:1014-1022.  2.  Bennick, A., and A. M. Hunt. 1967. Collagenolytic activity in oral tissues. Arch Oral Biol 12:110.  3.  Bauer, E. A., A. Z. Eisen, and J. J. Jeffrey. 1970. Immunologic relationship of a purified human skin collagenase to other human and animal collagenases. Biochim Biophys Acta 206:152-160.  4.  McCawley, L. J., and L. M. Matrisian. 2001. Matrix metalloproteinases: they’re not just for matrix anymore! Curr Opin Cell Biol 13:534-540.  5.  Overall, C. M. 2002. Molecular determinants of metalloproteinase substrate specificity: matrix metalloproteinase substrate binding domains, modules, and exosites. Mol Biotechnol 22:5186.  6.  Steffensen, B., U. M. Wallon, and C. M. Overall. 1995. Extracellular matrix binding properties of recombinant fibronectin type Il-like modules of human 72-kDa gelatinase/type IV collagenase. High affinity binding to native type I collagen but not native type IV collagen. J Biol Chem 270:11555-11566.  7.  Coussens, L. M., B. Fingleton, and L. M. Matrisian. 2002. Matrix metalloproteinase inhibitors and cancer: trials and tribulations. Science 295:2387-2392.  8.  Overall, C. M., and 0. Kleifeld. 2006. Tumour microenvironment opinion: validating matrix metalloproteinases as drug targets and anti-targets for cancer therapy. Nat Rev Cancer 6:227-239.  9.  Lazarus, G. S., R. S. Brown, J. R. Daniels, and H. M. Fullmer. 1968. Human granulocyte collagenase. Science 159:1483-1485.  10.  Murphy, G., J. J. Reynolds, U. Bretz, and M. Baggiolini. 1977. Collagenase is a component of the specific granules of human neutrophil leucocytes. Biochem J 162:195-197.  11.  Prikk, K., P. Maisi, E. Pirila, R. Sepper, T. Salo, J. Wahlgren, and T. Sorsa. 2001. In vivo collagenase-2 (MMP-8) expression bronchial epithelial cells and by human monocytes/macrophages in bronchiectasis. J Pathol 194:232-238.  12.  Toft-Hansen, H., R. K. Nuttall, D. R. Edwards, and T. Owens. 2004. Key metalloproteinases are expressed by specific cell types in experimental autoimmune encephalomyelitis. J Immunol 173:5209-5218.  13.  Hanemaaijer, R., T. Sorsa, Y. T. Konttinen, Y. Ding, M. Sutinen, H. Visser, V. W. van Hinsbergh, T. Helaakoski, T. Kainulainen, H. Ronka, H. Tschesche, and T. Salo. 1997. Matrix metalloproteinase-8 is expressed in rheumatoid synovial fibroblasts and endothelial cells. Regulation by tumor necrosis factor-alpha and doxycycline. J Biol Chem 272:31504-31509.  14.  Hasty, K. A., J. J. Jeffrey, M. S. Hibbs, and H. G. Welgus. 1987. The collagen substrate specificity of human neutrophil collagenase. J Biol Chem 262:10048-10052.  15.  Arner, E. C., C. P. Decicco, R. Cherney, and M. D. Tortorella. 1997. Cleavage of native cartilage aggrecan by neutrophil collagenase (MMP-8) is distinct from endogenous cleavage by aggrecanase. J 8101 Chem 272:9294-9299.  16.  Schmid, T. M., R. Mayne, J. J. Jeffrey, and T. F. Linsenmayer. 1986. Type X collagen contains two cleavage sites for a vertebrate collagenase. J Biol Chem 261 :4184-4 189.  17.  HilIer, 0., A. Lichte, A. Oberpichler, A. Kocourek, and H. Tschesche. 2000. Matrix metalloproteinases collagenase-2, macrophage elastase, collagenase-3, and membrane type  -  25  1-matrix metalloproteinase impair clotting by degradation of fibrinogen and factor XII. J Biol Chem 275:33008-33013. 18.  Shapiro, S. D. 1998. Matrix metalloproteinase degradation of extracellular matrix: biological consequences. Curr OpIn Cell 810110:602-608.  19.  Pirila, E., A. Sharabi, T. Salo, V. Quaranta, H. Tu, R. Heljasvaara, N. Koshikawa, T. Sorsa, and P. Maisi. 2003. Matrix metalloproteinases process the Iaminin-5 gamma 2-chain and regulate epithelial cell migration. Biochem Biophys Res Commun 303:101 2-101 7.  20.  Michaelis, J., M. C. Vissers, and C. C. Winterbourn. 1990. Human neutrophil collagenase cleaves alpha 1-antitrypsin. Biochem J 270:809-814.  21.  Knauper, V., A. Osthues, Y. A. DeClerck, K. E. Langley, J. Blaser, and H. Tschesche. 1993. Fragmentation of human polymorphonuclear-leucocyte collagenase. Biochem J 291 ( Pt 3):847-854.  22.  Diekmann, 0., and H. Tschesche. 1994. Degradation of kinins, angiotensins and substance P by polymorphonuclear matrix metalloproteinases MMP 8 and MMP 9. Braz J Med Biol Res 27:1865-1876.  23.  Knauper, V., S. Triebel, H. Reinke, and H. Tschesche. 1991. Inactivation of human plasma Cl-inhibitor by human PMN leucocyte matrix metalloproteinases. FEBS Lett 290:99-1 02.  24.  McQuibban, G. A., J. H. Gong, J. P. Wong, J. L. Wallace, I. Clark-Lewis, and C. M. Overall. 2002. Matrix metalloproteinase processing of monocyte chemoattractant proteins generates CC chemokine receptor antagonists with anti-inflammatory properties in vivo. Blood 100:1160-1167.  25.  Van Den Steen, P. E., A. Wuyts, S. J. Husson, P. Proost, J. Van Damme, and G. Opdenakker. 2003. Gelatinase B/MMP-9 and neutrophil collagenase/MMP-8 process the chemokines human GCP-2/CXCL6, ENA-78/CXCL5 and mouse GCP-2ILIX and modulate their physiological activities. Eur J Biochem 270:3739-3749.  26.  Balbin, M., A. Fueyo, A. M. Tester, A. M. Pendas, A. S. Pitiot, A. Astudillo, C. M. Overall, S. D. Shapiro, and C. Lopez-Otin. 2003. Loss of collagenase-2 confers increased skin tumor susceptibility to male mice. Nat Genet 35:252-257.  27.  Van den Steen, P. E., S. J. Husson, P. Proost, J. Van Damme, and G. Opdenakker. 2003. Carboxyterminal cleavage of the chemokines MIG and IP-lO by gelatinase B and neutrophil collagenase. Biochem Biophys Res Commun 310:889-896.  28.  Hemers, E., C. Duval, C. McCaig, M. Handley, G. J. Dockray, and A. Varro. 2005. Insulin-like growth factor binding protein-5 is a target of matrix metalloproteinase-7: implications for epithelial-mesenchymal signaling. Cancer Res 65:7363-7369.  29.  Okamoto, T., T. Akaike, T. Nagano, S. Miyajima, M. Suga, M. Ando, K. Ichimori, and H. Maeda. 1997. Activation of human neutrophil procollagenase by nitrogen dioxide and peroxynitrite: a novel mechanism for procollagenase activation involving nitric oxide. Arch Biochem Biophys 342:261-274.  30.  Knauper, V., S. Kramer, H. Reinke, and H. Tschesche. 1990. Characterization and activation of procollagenase from human polymorphonuclear leucocytes. N-terminal sequence determination of the proenzyme and various proteolytically activated forms. Eur J Biochem 189:295-300.  31.  Van Lint, P., and C. Libert. 2006. Matrix metalloproteinase-8: cleavage can be decisive. Cytokine Growth Factor Rev 17:21 7-223.  32.  Owen, C. A., Z. Hu, C. Lopez-Otin, and S. D. Shapiro. 2004. Membrane-bound matrix metalloproteinase-8 on activated polymorphonuclear cells is a potent, tissue inhibitor of metalloproteinase-resistant collagenase and serpinase. J Immunol 172:7791-7803.  26  33.  Matsuki, H., N. Fujimoto, K. Iwata, V. Knauper, Y. Okada, and T. Hayakawa. 1996. A onestep sandwich enzyme immunoassay for human matrix metalloproteinase 8 (neutrophil collagenase) using monoclonal antibodies. Clin Chim Acta 244:129-143.  34.  Power, C., C. M. O’Connor, D. MacFarlane, S. O’Mahoney, K. Gaffney, J. Hayes, and M. X. FitzGerald. 1994. Neutrophil collagenase in sputum from patients with cystic fibrosis. Am J Respir Crit Care Med 150:818-822.  35.  Lee, W., S. Aitken, J. Sodek, and C. A. McCulloch. 1995. Evidence of a direct relationship between neutrophil collagenase activity and periodontal tissue destruction in vivo: role of active enzyme in human periodontitis. J Periodontal Res 30:23-33.  36.  Gueders, M. M., M. Balbin, N. Rocks, J. M. Foidart, P. Gosset, R. Louis, S. Shapiro, C. Lopez-Otin, A. Noel, and D. D. Cataldo. 2005. Matrix metalloproteinase-8 deficiency promotes granulocytic allergen-induced airway inflammation. J Immunol 175:2589-2597.  37.  Van Lint, P., B. Wielockx, L. Puimege, A. Noel, C. Lopez-Otin, and C. Libert. 2005. Resistance of collagenase-2 (matrix metalloproteinase-8)-deficient mice to TN F-induced lethal hepatitis. J Immunol 175:7642-7649.  38.  Gutierrez-Fernandez, A., M. Inada, M. Balbin, A. Fueyo, A. S. Pitiot, A. Astudillo, K. Hirose, M. Hirata, S. D. Shapiro, A. Noel, Z. Werb, S. M. Krane, C. Lopez-Otin, and X. S. Puente. 2007. Increased inflammation delays wound healing in mice deficient in collagenase-2 (MMP 8). Faseb J.  39.  Folgueras, A. R., A. Fueyo, 0. Garcia-Suarez, J. Cox, A. Astudillo, P. Tortorella, C. Campestre, A. Gutierrez-Fernandez, M. Fanjul-Fernandez, C. J. Pennington, D. R. Edwards, C. M. Overall, and C. Lopez-Otin. 2008. Collagenase-2 deficiency or inhibition impairs experimental autoimmune encephalomyelitis in mice. J Biol Chem 283:9465-9474.  40.  Wucherpfennig, K. W. 2001. Mechanisms for the induction of autoimmunity by infectious agents. J Clin Invest 108:1097-1104.  41.  Dvorak, H. F. 1986. Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing. N Engl J Med 315:1650-1659.  42.  Hansson, G. K., and P. Libby. 2006. The immune response in atherosclerosis: a doubleedged sword. Nat Rev Immunol 6:508-519.  43.  Overall, C. M. 2004. Dilating the degradome: matrix metalloproteinase 2 (MMP-2) cuts to the heart of the matter. Biochem J 383:e5-7.  44.  Turk, B. 2006. Targeting proteases: successes, failures and future prospects. Nat Rev Drug Discov 5:785-799.  45.  Bramhall, S. R., M. T. Hallissey, J. Whiting, J. Scholefield, G. Tierney, R. C. Stuart, R. E. Hawkins, P. McCulloch, T. Maughan, P. D. Brown, M. Baillet, and J. W. Fielding. 2002. Marimastat as maintenance therapy for patients with advanced gastric cancer: a randomised trial. BrJ Cancer 86:1864-1 870.  46.  Parks, W. C., C. L. Wilson, and Y. S. Lopez-Boado. 2004. Matrix metalloproteinases as modulators of inflammation and innate immunity. Nat Rev lmmunol4:617-629.  47.  Page-McCaw, A., A. J. Ewald, and Z. Werb. 2007. Matrix metalloproteinases and the regulation of tissue remodelling. Nat Rev Mol Cell Biol 8:221-233.  48.  Itoh, T., H. Matsuda, M. Tanioka, K. Kuwabara, S. Itohara, and R. Suzuki. 2002. The role of matrix metalloproteinase-2 and matrix metalloproteinase-9 in antibody-induced arthritis. J Immunol 169:2643-2647.  49.  Corry, D. B., K. Rishi, J. Kanellis, A. Kiss, L. Z. Song Lz, J. Xu, L. Feng, Z. Werb, and F. Kheradmand. 2002. Decreased allergic lung inflammatory cell egression and increased susceptibility to asphyxiation in MMP2-deficiency. Nat Immunol 3:347-353.  27  50.  Esparza, J., M. Kruse, J. Lee, M. Michaud, and J. A. Madri. 2004. MMP-2 null mice exhibit an early onset and severe experimental autoimmune encephalomyelitis due to an increase in MMP-9 expression and activity. Faseb J 18:1682-1691.  51.  Campbell, L. G., S. Ramachandran, W. Liu, J. M. Shipley, S. Itohara, J. C. Rogers, N. Moazami, R. M. Senior, and A. Jaramillo. 2005. Different roles for matrix metalloproteinase-2 and matrix metalloproteinase-9 in the pathogenesis of cardiac allograft rejection. Am J Transplant 5:517-528.  52.  Matsusaka, H., M. Ikeuchi, S. Matsushima, T. Ide, T. Kubota, A. M. Feldman, A. Takeshita, K. Sunagawa, and H. Tsutsui. 2005. Selective disruption of MMP-2 gene exacerbates myocardial inflammation and dysfunction in mice with cytokine-induced cardiomyopathy. Am J Physiol Heart Circ Physiol 289:H 1858-1864.  53.  Samolov, B., B. Steen, S. Seregard, I. van der Ploeg, P. Montan, and A. Kvanta. 2005. Delayed inflammation-associated corneal neovascularization in MMP-2-deficient mice. Exp Eye Res 80:159-1 66.  54.  Garg, P., M. Rojas, A. Ravi, K. Bockbrader, S. Epstein, M. Vijay-Kumar, A. T. Gewirtz, D. Merlin, and S. V. Sitaraman. 2006. Selective ablation of matrix metalloproteinase-2 exacerbates experimental colitis: contrasting role of gelatinases in the pathogenesis of colitis. J Immunol 177:4103-4112.  55.  Wang, M., X. Qin, J. S. Mudgett, T. A. Ferguson, R. M. Senior, and H. G. Welgus. 1999. Matrix metalloproteinase deficiencies affect contact hypersensitivity: stromelysin-1 deficiency prevents the response and gelatinase B deficiency prolongs the response. Proc NatI Acad Sd U SA 96:6885-6889.  56.  Haro, H., H. C. Crawford, B. Fingleton, J. R. MacDougall, K. Shinomiya, D. M. Spengler, and L. M. Matrisian. 2000. Matrix metalloproteinase-3-dependent generation of a macrophage chemoattractant in a model of herniated disc resorption. J Clin Invest 105:133-141.  57.  Warner, R. L., L. Beltran, E. M. Younkin, C. S. Lewis, S. J. Weiss, J. Varani, and K. J. Johnson. 2001. Role of stromelysin 1 and gelatinase B in experimental acute lung injury. Am J Respir Cell Mol Biol 24:537-544.  58.  Silence, J., F. Lupu, D. Collen, and H. R. Lijnen. 2001. Persistence of atherosclerotic plaque but reduced aneurysm formation in mice with stromelysin-1 (MMP-3) gene inactivation. Arterioscler Thromb Vasc Biol 21:1440-1445.  59.  Li, C. K., S. L. Pender, K. M. Pickard, V. Chance, J. A. Holloway, A. Huett, N. S. Goncalves, J. S. Mudgett, G. Dougan, G. Frankel, and T. T. MacDonald. 2004. Impaired immunity to intestinal bacterial infection in stromelysin- 1 (matrix metalloproteinase-3)-deficient mice. J Immunol 173:5171-51 79.  60.  Mukherjee, R., J. A. Bruce, D. M. McClister, Jr., C. M. Allen, S. E. Sweterlitsch, and J. P. Saul. 2005. Time-dependent changes in myocardial structure following discrete injury in mice deficient of matrix metalloproteinase-3. J Mol Cell Cardiol 39:259-268.  61.  Handley, S. A., and V. L. Miller. 2007. General and specific host responses to bacterial infection in Peyer’s patches: a role for stromelysin-1 (matrix metalloproteinase-3) during Salmonella enterica infection. Mol Microbiol 64:94-110.  62.  Wilson, C. L., A. J. Ouellette, D. P. Satchell, T. Ayabe, Y. S. Lopez-Boado, J. L. Stratman, S. J. Hultgren, L. M. Matrisian, and W. C. Parks. 1999. Regulation of intestinal alpha-defensin activation by the metalloproteinase matrilysin in innate host defense. Science 286:113-117.  63.  Powell, W. C., B. Fingleton, C. L. Wilson, M. Boothby, and L. M. Matrisian. 1999. The metalloproteinase matrilysin proteolytically generates active soluble Fas ligand and potentiates epithelial cell apoptosis. Curr Biol 9:1441-1447.  28  64.  Haro, H., H. C. Crawford, B. Fingleton, K. Shinomiya, D. M. Spengler, and L. M. Matrisian. 2000. Matrix metalloproteinase-7-dependent release of tumor necrosis factor-alpha in a model of herniated disc resorption. J Clin Invest 105:143-1 50.  65.  Li, Q., P. W. Park, C. L. Wilson, and W. C. Parks. 2002. Matrilysin shedding of syndecan-1 regulates chemokine mobilization and transepithelial efflux of neutrophils in acute lung injury. Cell 111:635-646.  66.  Gjertsson, I., M. Innocenti, L. M. Matrisian, and A. Tarkowski. 2005. Metalloproteinase-7 contributes to joint destruction in Staphylococcus aureus induced arthritis. Microb Pathog 38:97-105.  67.  Tester, A. M., J. H. Cox, A. R. Connor, A. E. Starr, R. A. Dean, X. S. Puente, C. Lopez-Otin, and C. M. Overall. 2007. LPS Responsiveness and Neutrophil Chemotaxis In Vivo Require PMN MMP-8 Activity. PLoS ONE 2:e312.  68.  Lin, M., P. Jackson, A. M. Tester, E. Diaconu, C. M. Overall, J. E. Blalock, and E. Peariman. 2008. Matrix Metalloproteinase-8 Facilitates Neutrophil Migration through the Corneal Stromal Matrix by Collagen Degradation and Production of the Chemotactic Peptide Pro-Gly Pro. Am J Pathol.  69.  Dubois, B., S. Masure, U. Hurtenbach, L. Paemen, H. Heremans, J. van den Oord, R. Sciot, T. Meinhardt, G. Hammerling, G. Opdenakker, and B. Arnold. 1999. Resistance of young gelatinase B-deficient mice to experimental autoi m mune encephalomyelitis and necrotizing tail lesions. J Clin Invest 104:1 507-1515.  70.  Liu, Z., X. Zhou, S. D. Shapiro, J. M. Shipley, S. S. Twining, L. A. Diaz, R. M. Senior, and Z. Werb. 2000. The serpin aiphal-proteinase inhibitor is a critical substrate for gelatinase B/MMP-9 in vivo. Cell 102:647-655.  71.  Ducharme, A., S. Frantz, M. Aikawa, E. Rabkin, M. Lindsey, L. E. Rohde, F. J. Schoen, R. A. Kelly, Z. Werb, P. Libby, and R. T. Lee. 2000. Targeted deletion of matrix metalloproteinase-9 attenuates left ventricular enlargement and collagen accumulation after experimental myocardial infarction. J Clin Invest 106:55-62.  72.  Heissig, B., K. Hattori, S. Dias, M. Friedrich, B. Ferris, N. R. Hackett, R. G. Crystal, P. Besmer, D. Lyden, M. A. Moore, Z. Werb, and S. Rafii. 2002. Recruitment of stem and progenitor cells from the bone marrow niche requires MMP-9 mediated release of kit-ligand. Cell 109:625-637.  73.  Cataldo, D. D., K. G. Tournoy, K. Vermaelen, C. Munaut, J. M. Foidart, R. Louis, A. Noel, and R. A. Pauwels. 2002. Matrix metalloproteinase-9 deficiency impairs cellular infiltration and bronchial hyperresponsiveness during allergen-induced airway inflammation. Am J Pathol 161:491-498.  74.  Bottcher, T., A. Spreer, I. Azeh, R. Nau, and J. Gerber. 2003. Matrix metalloproteinase-9 deficiency impairs host defense mechanisms against Streptococcus pneumoniae in a mouse model of bacterial meningitis. NeurosciLett338:201-204.  75.  Pelus, L. M., H. Bian, A. G. King, and S. Fukuda. 2004. Neutrophil-derived MMP-9 mediates synergistic mobilization of hematopoietic stem and progenitor cells by the combination of G CSF and the chemokines GRObeta/CXCL2 and GRObetaT/CXCL2delta4. Blood 103:110119.  76.  lchiyasu, H., J. M. McCormack, K. M. McCarthy, D. Dombkowski, F. I. Preffer, and E. E. Schneeberger. 2004. Matrix metalloproteinase-9-deficient dendritic cells have impaired migration through tracheal epithelial tight junctions. Am J Respir Cell Mo! Biol 30:761-770.  77.  McMillan, S. J., J. Kearley, J. D. Campbell, X. W. Zhu, K. Y. Larbi, J. M. Shipley, R. M. Senior, S. Nourshargh, and C. M. Lloyd. 2004. Matrix metalloproteinase-9 deficiency results in enhanced allergen-induced airway inflammation. J Immunol 172:2586-2594.  29  78.  Corry, D. B., A. Kiss, L. Z. Song, L. Song, J. Xu, S. H. Lee, Z. Werb, and F. Kheradmand. 2004. Overlapping and independent contributions of MMP2 and MMP9 to lung allergic inflammatory cell egression through decreased CC chemokines. Faseb J 18:995-997.  79.  Santana, A., C. Medina, M. C. Paz-Cabrera, F. Diaz-Gonzalez, E. Farre, A. Salas, M. W. Radomski, and E. Quintero. 2006. Attenuation of dextran sodium sulphate induced colitis in matrix metalloproteinase-9 deficient mice. World J Gastroenterol 12:6464-6472.  80.  Calander, A. M., S. Starckx, G. Opdenakker, P. Bergin, M. Quiding-Jarbrink, and A. Tarkowski. 2006. Matrix metalloproteinase-9 (gelatinase B) deficiency leads to increased severity of Staphylococcus aureus-triggered septic arthritis. Microbes Infect 8:1434-1 439.  81.  Renckens, R., J. J. Roelofs, S. Florquin, A. F. de Vos, H. R. Lijnen, C. van’t Veer, and T. van der Poll. 2006. Matrix metalloproteinase-9 deficiency impairs host defense against abdominal sepsis. J Immunol 176:3735-3741.  82.  Kassim, S. Y., S. A. Gharib, B. H. Mecham, T. P. Birkland, W. C. Parks, and J. K. McGuire. 2007. Individual matrix metalloproteinases control distinct transcriptional responses in airway epithelial cells infected with Pseudomonas aeruginosa. Infect Immun 75:5640-5650.  83.  Hautamaki, R. D., D. K. Kobayashi, R. M. Senior, and S. D. Shapiro. 1997. Requirement for macrophage elastase for cigarette smoke-induced emphysema in mice. Science 277:20022004.  84.  Churg, A., R. D. Wang, H. Tai, X. Wang, C. Xie, J. Dai, S. D. Shapiro, and J. L. Wright. 2003. Macrophage metalloelastase mediates acute cigarette smoke-induced inflammation via tumor necrosis factor-alpha release. Am J Respir Crlf Care Med 167:1083-1089.  85.  Wells, J. E., T. K. Rice, R. K. Nuttall, D. R. Edwards, H. Zekki, S. Rivest, and V. W. Yong. 2003. An adverse role for matrix metalloproteinase 12 after spinal cord injury in mice. J Neurosci23:10107-101 15.  86.  Pouladi, M. A., C. S. Robbins, F. K. Swirski, M. Cundall, A. N. McKenzie, M. Jordana, S. D. Shapiro, and M. R. Stampfli. 2004. lnterleukin-13-dependent expression of matrix metalloproteinase-12 is required for the development of airway eosinophilia in mice. Am J RespIr Cell Mo) Biol 30:84-90.  87.  Warner, R. L, N. W. Lukacs, S. D. Shapiro, N. Bhagarvathula, K. C. Nerusu, J. Varani, and K. J. Johnson. 2004. Role of metalloelastase in a model of allergic lung responses induced by cockroach allergen. Am J Patho) 165:1921-1 930.  88.  Longo, G. M., S. J. Buda, N. Fiotta, W. Xiong, T. Griener, S. Shapiro, and B. T. Baxter. 2005. MMP-12 has a role in abdominal aortic aneurysms in mice. Surgery 137:457-462.  89.  Weaver, A., A. Goncalves da Silva, R. K. Nuttall, D. R. Edwards, S. D. Shapiro, S. Rivest, and V. W. Yong. 2005. An elevated matrix metalloproteinase (MMP) in an animal model of multiple sclerosis is protective by affecting ThllTh2 polarization. Faseb J 19:1668-1670.  90.  Johnson, J. L., S. J. George, A. C. Newby, and C. L. Jackson. 2005. Divergent effects of matrix metalloproteinases 3, 7, 9, and 12 on atherosclerotic plaque stability in mouse brachiocephalic arteries. Proc Nat) Acad Sd USA 102:15575-15580.  91.  Wright, R. W., T. Allen, H. B. El-Zawawy, M. D. Brodt, M. J. Silva, C. S. Gill, and L. J. Sandell. 2006. Medial collateral ligament healing in macrophage metalloelastase (MMP-12)deficient mice. J Orthop Res 24:2106-2113.  92.  Manoury, B., S. Nenan, I. Guenon, E. Boichot, J. M. Planquois, C. P. Bertrand, and V. Lagente. 2006. Macrophage metalloelastase (MMP-12) deficiency does not alter bleomycin induced pulmonary fibrosis in mice. J lnflamm (Lond) 3:2.  93.  Houghton, A. M., P. A. Quintero, D. L. Perkins, D. K. Kobayashi, D. G. Kelley, L. A. Marconcini, R. P. Mecham, R. M. Senior, and S. D. Shapiro. 2006. Elastin fragments drive disease progression in a murine model of emphysema. J ClIn Invest 116:753-759.  30  94.  Dinarello, C. A. 2005. Blocking IL-i in systemic inflammation. J Exp Med 201:1355-1359.  95.  Andrei, C., C. Dazzi, L. Lotti, M. R. Torrisi, G. Chimini, and A. Rubartelli. 1999. The secretory route of the leaderless protein interleukin ibeta involves exocytosis of endolysosome-related vesicles. Mo! Biol Cell 10:1463-1475.  96.  Kostura, M. J., M. J. Tocci, G. Limjuco, J. Chin, P. Cameron, A. G. Hillman, N. A. Chartrain, and J. A. Schmidt. 1989. Identification of a monocyte specific pre-interleukin 1 beta convertase activity. Proc Nail Acad Sd U S A 86:5227-5231.  97.  Fantuzzi, G., G. Ku, M. W. Harding, D. J. Livingston, J. D. Sipe, K. Kuida, R. A. Flavell, and C. A. Dinarello. 1997. Response to local inflammation of IL-i beta-converting enzymedeficient mice. J lmmunol 158:1818-1824.  98.  Schonbeck, U., F. Mach, and P. Libby. 1998. Generation of biologically active IL-i beta by matrix metalloproteinases: a novel caspase-1 -independent pathway of IL-i beta processing. J Immunol 161:3340-3346.  99.  Ito, A., A. Mukaiyama, Y. Itoh, H. Nagase, I. B. Thogersen, J. J. Enghild, Y. Sasaguri, and Y. Mon. 1996. Degradation of interleukin ibeta by matrix metalloproteinases. J Biol Chem 271:14657-14660.  100.  Kodama, S., M. Davis, and D. L. Faustman. 2005. The therapeutic potential of tumor necrosis factor for autoimmune disease: a mechanistically based hypothesis. Cell Mol Life Sd 62:1850-1862.  101.  Taylor, P. C., R. 0. Williams, and M. Feldmann. 2004. Tumour necrosis factor alpha as a therapeutic target for immune-mediated inflammatory diseases. Curr Opin Biotechnol 15:557563.  102.  Gearing, A. J., P. Beckett, M. Christodoulou, M. Churchill, J. M. Clements, M. Cnimmin, A. H. Davidson, A. H. Drummond, W. A. Galloway, R. Gilbert, and et al. 1995. Matrix metalloproteinases and processing of pro-TNF-alpha. J Leukoc Biol 57:774-777.  103.  Chandler, S., J. Cossins, J. Lury, and G. Wells. 1996. Macrophage metalloelastase degrades matrix and myelin proteins and processes a tumour necrosis factor-alpha fusion protein. Biochem Biophys Res Commun 228:421-429.  104.  d’Ortho, M. P., H. Will, S. Atkinson, G. Butler, A. Messent, J. Gavrilovic, B. Smith, R. Timpi, L. Zardi, and G. Murphy. 1997. Membrane-type matrix metalloproteinases 1 and 2 exhibit broad-spectrum proteolytic capacities comparable to many matrix metalloproteinases. Eur J Biochem 250:751-757.  105.  Tam, E. M., C. J. Morrison, Y. I. Wu, M. S. Stack, and C. M. Overall. 2004. Membrane protease proteomics: Isotope-coded affinity tag MS identification of undescribed MT1-matrix metalloproteinase substrates. Proc Nat! Acad Sd USA 101:6917-6922.  106.  English, W. R., X. S. Puente, J. M. Freije, V. Knauper, A. Amour, A. Merryweather, C. Lopez Otin, and G. Murphy. 2000. Membrane type 4 matrix metalloproteinase (MMP17) has tumor necrosis factor-alpha convertase activity but does not activate pro-MMP2. J Biol Chem 275:14046-14055.  107.  Van den Steen, P. E., P. Proost, A. Wuyts, J. Van Damme, and G. Opdenakker. 2000. Neutrophil gelatinase B potentiates interleukin-8 tenfold by aminoterminal processing, whereas it degrades CTAP-lll, PF-4, and GRO-alpha and leaves RANTES and MCP-2 intact. Blood 96:2673-2681.  108.  Dean, R. A., and C. M. Overall. 2007. Proteomics discovery of metalloproteinase substrates in the cellular context by iTRAQ labeling reveals a diverse MMP-2 substrate degradome. Mo! Cell Proteomics 6:611-623.  109.  Dean, R. A., J. H. Cox, C. L. Bellac, A. Doucet, A. E. Starr, and C. M. Overall. In Press. Macrophage-specific metalloelastase (MM P-i 2) truncates and inactivates ELR+ CXC  31  chemokines and generates CCL2, 7, 8, and 13 antagonists: Potential role of the macrophage in terminating PMN influx. Blood:ln Press. 110.  Karsdal, M. A., L. Larsen, M. T. Engsig, H. Lou, M. Ferreras, A. Lochter, J. M. Delaisse, and N. T. Foged. 2002. Matrix metalloproteinase-dependent activation of latent transforming growth factor-beta controls the conversion of osteoblasts into osteocytes by blocking osteoblast apoptosis. J Biol Chem 277:44061-44067.  111.  Maeda, S., D. D. Dean, I. Gay, Z. Schwartz, and B. D. Boyan. 2001. Activation of latent transforming growth factor betal by stromelysin 1 in extracts of growth plate chondrocyte derived matrix vesicles. J Bone Miner Res 16:1281-1290.  112.  Yu, Q., and I. Stamenkovic. 2000. Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-beta and promotes tumor invasion and angiogenesis. Genes Dev 14:163-176.  113.  D’Angelo, M., P. C. Billings, M. Pacifici, P. S. Leboy, and T. Kirsch. 2001. Authentic matrix vesicles contain active metalloproteases (MMP). a role for matrix vesicle-associated MMP-13 in activation of transforming growth factor-beta. J Biol Chem 276:11347-11353.  114.  McQuibban, G. A., J. H. Gong, E. M. Tam, C. A. McCulloch, I. Clark-Lewis, and C. M. Overall. 2000. Inflammation dampened by gelatinase A cleavage of monocyte chemoattractant protein-3. Science 289:1202-1206.  115.  McQuibban, G. A., G. S. Butler, J. H. Gong, L. Bendall, C. Power, I. Clark-Lewis, and C. M. Overall. 2001. Matrix metalloproteinase activity inactivates the CXC chemokine stromal cellderived factor-I. J Biol Chem 276:43503-43508.  116.  Mohler, K. M., P. R. Sleath, J. N. Fitzner, D. P. Cerretti, M. Alderson, S. S. Kerwar, D. S. Torrance, C. Otten-Evans, T. Greenstreet, K. Weerawarna, and et al. 1994. Protection against a lethal dose of endotoxin by an inhibitor of tumour necrosis factor processing. Nature 370:218-220.  117.  McGeehan, G. M., J. D. Becherer, R. C. Bast, Jr., C. M. Boyer, B. Champion, K. M. Connolly, J. G. Conway, P. Furdon, S. Karp, S. Kidao, and et al. 1994. Regulation of tumour necrosis factor-alpha processing by a metalloproteinase inhibitor. Nature 370:558-561.  118.  Gearing, A. J., P. Beckett, M. Christodoulou, M. Churchill, J. Clements, A. H. Davidson, A. H. Drummond, W. A. Galloway, R. Gilbert, J. L. Gordon, and et al. 1994. Processing of tumour necrosis factor-alpha precursor by metalloproteinases. Nature 370:555-557.  119.  Black, R. A., C. T. Rauch, C. J. Kozlosky, J. J. Peschon, J. L. Slack, M. F. Wolfson, B. J. Castner, K. L. Stocking, P. Reddy, S. Srinivasan, N. Nelson, N. Boiani, K. A. Schooley, M. Gerhart, R. Davis, J. N. Fitzner, R. S. Johnson, R. J. Paxton, C. J. March, and D. P. Cerretti. 1997. A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells. Nature 385:729-733.  120.  Moss, M. L., S. L. Jin, M. E. Milla, D. M. Bickett, W. Burkhart, H. L. Carter, W. J. Chen, W. C. Clay, J. R. Didsbury, D. Hassler, C. R. Hoffman, T. A. Kost, M. H. Lambert, M. A. Leesnitzer, P. McCauley, G. McGeehan, J. Mitchell, M. Moyer, G. Pahel, W. Rocque, L. K. Overton, F. Schoenen, T. Seaton, J. L. Su, J. D. Becherer, and et al. 1997. Cloning of a disintegrin metalloproteinase that processes precursor tumour-necrosis factor-alpha. Nature 385:733736.  121.  Overall, C. M., and C. P. Blobel. 2007. In search of partners: linking extracellular proteases to substrates. Nat Rev Mo! Cell Biol 8:245-257.  122.  WahI, S. M., J. Wen, and N. Moutsopoulos. 2006. TGF-beta: a mobile purveyor of immune privilege. Immunol Rev 213:213-227.  123.  Shull, M. M., I. Ormsby, A. B. Kier, S. Pawlowski, R. J. Diebold, M. Yin, R. Allen, C. Sidman, G. Proetzel, D. Calvin, and et al. 1992. Targeted disruption of the mouse transforming growth factor-beta 1 gene results in multifocal inflammatory disease. Nature 359:693-699.  32  124.  Khalil, N. 1999. TGF-beta: from latent to active. Microbes Infect 1:1255-1263.  125.  Overall, C. M., J. L. Wrana, and J. Sodek. 1989. Transforming growth factor-beta regulation of collagenase, 72 kDa-progelatinase, TIMP and PAl-i expression in rat bone cell populations and human fibroblasts. Connect Tissue Ros 20:289-294.  126.  Overall, C. M., J. L. Wrana, and J. Sodek. 1991. 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 266:14064-14071.  127.  Zhou, H., E. J. Bernhard, F. E. Fox, and P. C. Billings. 1993. Induction of metalloproteinase activity in human T-lymphocytes. Biochim Biophys Acta 1177:174-178.  128.  Ellerbroek, S. M., Y. I. Wu, C. M. Overall, and M. S. Stack. 2001. Functional interplay between type I collagen and cell surface matrix metalloproteinase activity. J Biol Chem 276:24833-24842.  129.  Tam, E. M., T. R. Moore, G. S. Butler, and C. M. Overall. 2004. Characterization of the distinct collagen binding, helicase and cleavage mechanisms of matrix metalloproteinase 2 and 14 (gelatinase A and MT1-MMP): the differential roles of the MMP hemopexin c domains and the MMP-2 fibronectin type II modules in collagen triple helicase activities. J Biol Chem 279:43336-43344.  130.  Moser, B., M. Wolf, A. WaIz, and P. Loetscher. 2004. Chemokines: multiple levels of leukocyte migration control. Trends Immunol 25:75-84.  131.  Fernandez, E. J., and E. Lolis. 2002. Structure, function, and inhibition of chemokines. Annu Rev Pharmacol Toxicol 42:469-499.  132.  Hensbergen, P. J., D. Verzijl, C. I. Balog, R. Dijkman, R. C. van der Schors, E. Raaij-Helmer, M. J. van der Plas, R. Leurs, A. M. Deelder, M. J. Smit, and C. 2004. Furin is a chemokine-modifying enzyme: in vitro and in vivo processing generates a C-terminally truncated chemokine retaining full activity. J Biol Chem 13411.  133.  Overall, C. M., G. A. McQuibban, and I. Clark-Lewis. 2002. Discovery of chemokine substrates for matrix metalloproteinases by exosite scanning: a new tool for degradomics. Biol Chem 383:1059-1066.  134.  Balabanian, K., B. Lagane, S. Infantino, K. Y. Chow, J. Harriague, B. Moepps, F. Arenzana Seisdedos, M. Thelen, and F. Bachelerie. 2005. The chemokine SDF-1/CXCL12 binds to and signals through the orphan receptor RDC1 in T lymphocytes. J Biol Chem 280:35760-35766.  135.  Zhang, K., G. A. McQuibban, C. Silva, G. S. Butler, J. B. Johnston, J. Holden, I. Clark-Lewis, C. M. Overall, and C. Power. 2003. HIV-induced metalloproteinase processing of the chemokine stromal cell derived factor-i causes neurodegeneration. Nat Neurosci 6:10641071.  136.  Vergote, D., G. S. Butler, M. Ooms, J. H. Cox, C. Silva, M. D. Hollenberg, J. H. Jhamandas, C. M. Overall, and C. Power. 2006. Proteolytic processing of SDF-ialpha reveals a change in receptor specificity mediating HIV-associated neurodegeneration. Proc NatI Acad Sd U S A 103:19182-19187.  137.  Bazan, J. F., K. B. Bacon, G. Hardiman, W. Wang, K. Soo, D. Rossi, D. R. Greaves, A. Zlotnik, and T. J. Schall. 1997. A new class of membrane-bound chemokine with a CX3C motif. Nature 385:640-644.  138.  Garton, K. J., P. J. Gough, C. P. Blobel, G. Murphy, D. R. Greaves, P. J. Dempsey, and L. W. Raines. 2001. Tumor necrosis factor-alpha-converting enzyme (ADAM17) mediates the cleavage and shedding of fractalkine (CX3CL1). J Biol Chem 276:37993-38001.  M. van der P. Tensen. of CXCL1O 279:13402-  33  139.  Hundhausen, C., D. Misztela, T. A. Berkhout, N. Broadway, P. Saftig, K. Reiss, D. Hartmann, F. Fahrenholz, R. Postina, V. Matthews, K. J. Kallen, S. Rose-John, and A. Ludwig. 2003. The disintegrin-like metalloproteinase ADAM10 is involved in constitutive cleavage of CX3CL1 (fractalkine) and regulates CX3CL1-mediated cell-cell adhesion. Blood 102:11861195.  140.  lnoue, A., H. Hasegawa, M. Kohno, M. R. Ito, M. Terada, T. lmai, 0. Yoshie, M. Nose, and S. Fujita. 2005. Antagonist of fractalkine (CX3CL1) delays the initiation and ameliorates the progression of lupus nephritis in MRLJIpr mice. Arthritis Rheum 52:1522-1533.  141.  Hara, T., T. Katakai, J. H. Lee, Y. Nambu, N. Nakajima-Nagata, H. Gonda, M. Sugai, and A. Shimizu. 2006. A transmembrane chemokine, CXC chemokine ligand 16, expressed by lymph node fibroblastic reticular cells has the potential to regulate T cell migration and adhesion. mt Immunol 18:301-311.  142.  Schulte, A., B. Schulz, M. G. Andrzejewski, C. Hundhausen, S. Mletzko, J. Achilles, K. Reiss, K. Paliga, C. Weber, S. R. John, and A. Ludwig. 2007. Sequential processing of the transmembrane chemokines CX3CL1 and CXCL16 by alpha- and gamma-secretases. Biochem Biophys Res Commun 358:233-240.  143.  Handel, T. M., Z. Johnson, S. E. Crown, E. K. Lau, and A. E. Proudfoot. 2005. Regulation of protein function by glycosaminoglycans--as exemplified by chemokines. Annu Rev Biochem 74:385-410.  144.  Proudfoot, A. E., T. M. Handel, Z. Johnson, E. K. Lau, P. LiWang, I. Clark-Lewis, F. Borlat, T. N. Wells, and M. H. Kosco-Vilbois. 2003. Glycosaminoglycan binding and oligomerization are essential for the in vivo activity of certain chemokines. Proc Nati Acad Sci U S A 100:18851890.  145.  Endo, K., T. Takino, H. Miyamori, H. Kinsen, T. Yoshizaki, M. Furukawa, and H. Sato. 2003. Cleavage of syndecan-1 by membrane type matrix metalloproteinase-1 stimulates cell migration. J Biol Chem 278:40764-40770.  146.  Puente, X. S., L. M. Sanchez, C. M. Overall, and C. Lopez-Otin. 2003. Human and mouse proteases: a comparative genomic approach. Nat Rev Genet 4:544-558.  147.  Pavlaki, M., and S. Zucker. 2003. Matrix metalloproteinase inhibitors (MMPIs): the beginning of phase I or the termination of phase Ill clinical trials. Cancer Metastasis Rev 22:177-203.  148.  Overall, C. M., and C. Lopez-Otin. 2002. Strategies for MMP inhibition in cancer: innovations for the post-trial era. Nat Rev Cancer 2:657-672.  149.  Zucker, S., J. Cao, and W. T. Chen. 2000. Critical appraisal of the use of matrix metalloproteinase inhibitors in cancer treatment. Oncogene 19:6642-6650.  150.  Greenlee, K. J., D. B. Corry, D. A. Engler, R. K. Matsunami, P. Tessier, R. G. Cook, Z. Werb, and F. Kheradmand. 2006. Proteomic identification of in vivo substrates for matrix metalloproteinases 2 and 9 reveals a mechanism for resolution of inflammation. J Immunol 177:7312-7321.  151.  Dean, R. A., G. S. Butler, Y. Hamma-Kourbali, J. Delbe, D. R. Brigstock, J. Courty, and C. M. Overall. 2007. Identification of candidate angiogenic inhibitors processed by matrix metalloproteinase 2 (MMP-2) in cell-based proteomic screens: disruption of vascular endothelial growth factor (VEGF)/heparin affin regulatory peptide (pleiotrophin) and VEGF/Connective tissue growth factor angiogenic inhibitory complexes by MMP-2 proteolysis. Mo! Cell Biol 27:8454-8465.  152.  Butler, G. S., R. A. Dean, E. M. Tam, and C. M. Overall. 2008. Pharmacoproteomics of a metalloproteinase hydroxamate inhibitor in breast cancer cells: Dynamics of matrix metalloproteinase-14 (MTI-MMP) mediated membrane protein shedding. Mo! Cell Biol.  153.  Schilling, 0., and C. M. Overall. 2007. Proteomic discovery of protease substrates. Curr Opin Chem 810111:36-45.  34  CHAPTER 2. LPS RESPONSIVENESS AND NEUTROPHIL CHEMOTAXIS IN VIVO REQUIRE PMN MMP-8 ACTIVITY 2 Perspective We identify matrix metalloproteinase (MMP)-8, the polymorphonuclear (PMN) leukocyte collagenase, as a critical mediator initiating lipopolysaccharide (LPS) responsiveness in vivo. PMN infiltration towards LPS is abrogated in Mmp8-null mice. MMP-8 cleaves LPS-induced CXC chemokine (LIX) at 5 -Va1 and Lys 4 Ser T9 . LIX bioactivity is increased upon N-terminal cleavage, enhancing intracellular 80 Arg calcium mobilization and chemotaxis upon binding its cognate receptor, CXCR2. As there is no difference in PMN chemotaxis in Mmp8-null mice compared with wildtype mice towards synthetic analogues of MMP-8-cleaved LIX, MMP-8 is not essential for extravasation or cell migration in collagenous matrices in vivo. However, with biochemical redundancy between MMP5 1, 2, 9, and 13, which also cleave LIX at position 4-5, it was surprising to observe such a markedly reduced PMN infiltration towards LPS and LIX in Mmp8 -I- mice. This lack of physiological redundancy in vivo identifies MMP-8 as a key mediator in the regulation of innate immunity. Comparable results were found with CXCL8/lL-8 and CXCL5IENA-78, the human orthologues of LIX. MMP-8 cleaves CXCL8 at Arg 6 5 Ser and at Val 8 7 Leu in CXCL5 to activate respective chemokines. Hence, rather than collagen, these PMN chemoattractants are important MMP-8 substrates in vivo; PMN-derived MMP-8 cleaves and activates LIX to execute an in cis PMN-controlled feed-forward mechanism to orchestrate the initial inflammatory response and promote LPS responsiveness in tissue.  2  A version of this chapter has been published. Tester AM*, Cox JH*, Connor AR, Starr AE, Dean RA, Puente XS, Lopez-Otin C, and Overall CM. (2007) LPS responsiveness and neutrophil chemotaxis in vivo require PMN MMP-8 activity. PL0S ONE. 2(3):e312. (* equal contribution). This topic has been examined further and is presented in Appendix A (unpublished data).  35  Introduction Polymorphonuclear neutrophils (PMNs) are crucial inflammatory leukocytes in host protection from infection, where their primary role is in phagocytosis and killing of bacteria, fungi and protozoa, and in wound debridement and healing (1, 2). Given these critical roles of PMNs, it has long been recognised that neutropenic patients are at greater risk of infection (3), and that is often observed after intensive cancer chemotherapy (4, 5). Proteolysis of phagosome contents and damaged extracellular matrix are key PMN actions in inflammation.  Cell migration, crossing basement  membrane and connective tissue matrix barriers are other aspects of PMN function traditionally thought to require proteolytic activity (6).  Additionally, PMN5 are a  source of chemotactic factors that guide the recruitment of specific and non-specific immune effector cells (7) and so these first line defence cells play key roles in innate and acquired immunity. Of the two major chemokine subfamilies that provide directional cues for leukocyte migration and activation (8), the CXC chemokines predominantly influence PMN5 and T-lymphocytes whereas the CC chemokines are active on monocytes, basophils and eosinophils (9). The expression of CXC chemokines is rapidly upregulated during acute inflammatory responses, such as that initiated by the endotoxin lipopolysaccaride (LPS) (10-13). A subset of the CXC chemokines are characterised by an ELR (glutamic acid-leucine-arginine) sequence proximal to the conserved CXC motif. ELR is essential for binding CXC-receptors (CXCR) I and 2 (14) leading to PMN activation, degranulation and release of proteases (15). The murine ELR CXC chemokines act through a single receptor that is homologous to human CXCR2 (16). In humans there are seven ELR CXC chemokines; CXCL8/interleukin-8 (IL 8); CXCL7/neutrophil-activating peptide-2 (NAP-2); CXCL6/granulocyte chemotactic protein-2 (GCP-2); CXCL5/epithelial cell-derived neutrophil activating peptide-78 (ENA-78); and CXCLI, -2 and -3 (also known as growth-related oncogenes (GRO) a,  -,  and -y).  Only CXCL8/lL-8, the most potent of these chemokines, and  CXCL6/GCP-2 bind CXCR1, whereas all members signal through the closely related receptor CXCR2 (14). Mice lack a homologue of CXCL8/IL-8 having only four ELR  36  CXC chemokines: LPS-induced CXC chemokine (LIX), the most abundant and potent of the murine chemokines and regarded as the orthologue of CXCL8 (17); keratinocyte-derived chemokine (KC); macrophage inflammatory protein-2 (MIP-2); and dendritic cell  inflammatory protein-I  (DCIP-1). Physiological  N-terminal  cleavage of chemokines modifies their bioactivity—either enhancing activity of the ELR CXC chemokines (15) or generating potent receptor antagonists from the CC chemokines CCL2, -7, -8 and -13 (also known as macrophage chemotactic proteins I to 4) (18, 19). Although several candidate proteases are proposed for ELR+ CXCL proteolytic activation, none have been validated in vivo. Matrix metalloproteinases (MMP5) are traditionally associated with extracellular matrix protein degradation in many physiological and pathological processes, including  inflammation,  bacterial  infection, wound  healing, and  cancer cell  invasiveness (reviewed in ref (20)). However, it is now clear that MMPs mediate homeostasis of the extracellular environment (21) by modulating the biological activity of many bioactive molecules involved in cell function (22, 23), innate immunity (24) including chemokines (18, 19, 23, 25-27), TNF-a (22, 28), ct-defensin (29), and mannose binding lectin (30), and in tumour initiation and progression. Inflammation in cancer, particularly macrophage infiltration and MMP-9 release, generates a microenvironment advantageous to neoplastic progression (31), with recent evidence indicating that a PMN source of MMP-9 can also promote tumorigenesis (32).  MMP-8 (human: P22894, mouse: CAA73786, rat: AJ007288), the neutrophil collagenase, is produced primarily by PMNs and is released from the specific granules at sites of inflammation (33). Mmp8-nuII mice have no overt phenotype, with normal embryonic development, fertility, and long-term survival (34). In contrast to other MMP deficient mice (35), Mmp8-nulI mice challenged with carcinogens showed a markedly increased susceptibility to tumorigenesis (34), but this only occurred in male mice. This was the first report of a MMP having a protective role in tumorigenesis, so validating MMP-8 as an anti-target in cancer therapy (35).  37  As a potent type I collagenase (36, 37) it had been predicted that mice lacking the Mmp8 gene would show reduced PMN migration through collagenous matrices (34). Indeed, at the tumor stromal interface an abnormal inflammatory response is observed, characterised by an initially delayed and then a more diffuse PMN influx in the Mmp8-null mice (34). However, once established there was a prolonged chronic accumulation of PMNs that did not dissipate. Overall, this phenotype is reversed following transplantation of wild type bone marrow, confirming that the absence of MMP-8 produced by PMNs and not a tissue or tumour source resulted in the higher incidence of tumors.  In a model of TNF-induced acute hepatitis, MMP-8 deficient  mice showed dampened levels of PMN infiltration into the liver that was postulated to result from reduced LIX mobilization from an unidentified binding protein in the matrix (38).  Together, these studies suggest a coordinating role for MMP-8 in  physiological leukocyte trafficking both in acute and chronic inflammation, either through cleavage of collagen or chemokine binding proteins (38), or by processing of bioactive molecules, such as LIX (34) to control PMN migration or longevity. Here we have investigated the role of MMP-8 in PMN recruitment during acute inflammation using LPS responsiveness as a trigger; the PMN influx was abrogated in the Mmp8 -I- mouse. LIX is identified as a key inflammatory substrate of MMP-8 where N-terminal processing by MMP-8 activates the chemokine and so increases PMN chemotaxis and LPS responsiveness in vitro and in vivo. Although these activities are reduced in the Mmp8-nulI mouse, the loss of neutrophil collagenolytic activity did not alter in vitro PMN chemokinesis or in vivo chemotaxis when challenged with truncated LIX or truncated CXCL8/IL-8 chemokines. Hence, these data reveal a new auto-regulatory mechanism of PMN chemotaxis that is initiated by MMP-8 release from PMN5 and executed, directly or indirectly, by the proteolytic activation of LIX in mice and CXCL8 and CXCL5/ENA-78 in man. This drives further PMN migration in a novel feed-forward mechanism that, remarkably, is a major determinant of LPS responsiveness.  38  Materials and Methods Animals. Mice deficient in Mmp8 on a C57BL6/J x 129 S background were provided by Dr. S. Shapiro (Boston, USA). Wild type C57BL6/J x 129 S mice were purchased from the Jackson Laboratory. Animal breeding and experimental procedures were approved by the Animal Care Committee of the University of British Columbia (Appendix C). 6-8 week old mice, segregated according to sex, were used for all experiments. In vivo PMN chemotaxis. The air pouch model of PMN chemotaxis was used as  described previously (59). Sterile air (3 mL) was injected under the dorsal skin of mice, two days later the air pouch was reinflated with 2 mL of sterile air. On day five, I pg LPS (Sigma) in phosphate buffered saline or 5 pg of chemokine in 1 mL 0.5% carboxymethylcellulose was injected into the air pouch. After 0, 4, 8, or 12 h the mice were sacrificed and air pouches lavaged with 2 mL of sterile PBS. The resulting cell suspensions were lysed with 0.1% Triton-X100 and freeze-thawed. The PMN content was determined by myeloperoxidase activity (60) using isolated PMN cells as a standard. Cell lysate aliquots were separated by 7.5% SDS-PAGE and MMP-8 was detected by western blot with 1:10,000 rabbit lgG against recombinant mouse MMP-8 (34). Chemokines and proteinases.  LIX (1-92), LIX (5-92), LIX (5-79), MIP-2, KG,  DCIP-1, CXCL8, CXCL8 (6-77), CXCL5 and CXCL5 (8-78) were chemically synthesized and purified (61).  Recombinant human MMP-1, -2, -8, -9, -13, -14,  rodent MMP-8, -13, and recombinant human MMP-8 hemopexin C-domain were expressed and purified (18, 19).  The synthetic hydroxamate MMP inhibitor  Batimastat (BB94) was from British Biotech (Oxford, UK). Chemokine cleavage assays.  Analysis of chemokine cleavage by MMPs was  performed at enzyme/chemokine ratios from 1:1000 up to 1:10 (w/w), at 37 °C in assay buffer (150 mM NaCI, 20 mM Tris, 5 mM CaCI , pH 7.5) in the presence of I 2 mM APMA to activate proMMPs for 16-22 h. Digests were terminated by adding 5  39  pM EDTA. Recombinant MMP-8 hemopexin C-domain, EDTA or BB94 were added to cleavage assays as indicated. Reaction products were analyzed by 15% Tris Tricine SDS-PAGE and stained with Coomassie Brilliant Blue R250. The [M+H]/z of each cleavage product was determined by matrix assisted laser desorption ionization time of flight (MALDI-TOF) mass spectrometry on a Voyager-DEw STR Biospectrometry Workstation (ABI). Mass spectrometry data was deconvoluted to identify the substrate cleavage sites and confirmed by Edman sequencing. Heparin binding. To assess the effect of MMP truncations of LIX on heparin binding, 0.5 mL of 4 M chemokine in 10 mM potassium phosphate, pH 7.5 was loaded onto a I mL Hitrap TM heparin-Sepharose column (GE Healthcare). Bound  LIX and synthetic analogues of MMP-cleaved LIX were eluted using a linear gradient of 0 to 1.0 M NaCI over 20 mm at a flow rate of 1.0 mLimin and monitored by in-line absorbance at 215 nm. Cells. Murine PMNs were isolated from bone marrow as previously described (46) except PMN5 were recovered from a density gradient comprised of Histopaque 1077 layered on top of Histopaque 1119 according to manufacture’s instructions (Sigma, St Louis, USA). The murine pre-B 300-19 cell line stably expressing human CXCR2 was supplied by Dr. B. Moser (Bern, Switzerland). Cells were cultured in RPM 1-1640 medium containing 10% fetal bovine serum, 1% glutamine and 5 x i0 M f3mercaptoethanol (Sigma) under puromycin (1.5 pg/mL) selection. Measurement of intracellular calcium mobilization. Murine PMNs or CXCR2transfected B300-19 cells (I x 10 /mL in RPMI-1640 media supplemented with 1% 7 serum) were incubated with 2 1 iM Fluo-4-acetoxymethyl ester (Molecular Probes), for 30 mm, 37 °C. Cells were washed to remove unincorporated agent and resuspended at I x 106 cells/mL in Hanks Balanced Salt Solution (Gibco), 20 mM HEPES, 2.5 mM probenecid (Sigma). The cells were allowed to equilibrate at 37 °C for 5 mm  prior to addition of ligand as indicated.  Calcium concentration was  monitored by excitation at 485 nm and emission of 520 nm with a PerkinElmer  40  LS5OB spectrophotometer.  Calibration was performed by addition of 5 1 iM  ionomycin (Sigma) and 1 mM MnCI 2 (Fisher Biotech). In vitro chemotaxis assays.  Murine PMNs or CXCR2-transfected murine pre-B  cells were plated (1 x 6 /10 well in ctMEM, 0.1% BSA) in transwell inserts (pre-coated with 10% fetal bovine serum) containing 3 tm or 8 m pores respectively (Costar). Chemokines were added to the lower chamber and the plates incubated at 37 °C. PMN chemotaxis assays were performed for I h. The migrated cells were then fixed with 4% paraformaldehyde prior to counting. The murine pre-B cell transfectants were assayed for 4 h after which the migrated cells were counted.  Results LPS induced PMN response in mice To ascertain the role of MMP-8 in PMN cell migration and LPS responsiveness in vivo we compared Mmp8 -I- with Mmp8 +1+ mice. The PMN influx was significantly reduced (P  0.005) when LPS was injected in air pouches formed under the dorsal  skin of the MMP-8 knockout mice compared with wild type mice (Figure 2.1A). This was observed in both male and female mice, although the PMN infiltrate was generally greater in the females towards both LPS and the PBS control. Hence, the reduced PMN migration and accumulation in Mmp8 -I- mice reveals a critical role for this PMN-specific protease in neutrophil function in acute inflammation.  Notably,  MMP-8 was only detected in cell lysates from LPS-treated air pouches of wild type mice, revealing both the pro and active forms of the enzyme at 85 and 65 kDa, respectively (Figure 2.IB). A 30-kDa inactive degradation product of MMP-8 was also detected, as observed previously (39). MMP-8 processing of murine ELR CXC chemokines We investigated the potential role of MMP-8 to modulate the activity of the cognate ELR ligands of CXCR2 that may underlie the defect in LPS-induced PMN migration. The four murine ELR CXC chemokines LIX, KC, MIP-2, and DCIP-1 were incubated  41  A  B 10.0  a  7.5  K0 WT  kDa KO WT 188—  Cl)  Z  4proMMP-8 4 active MMP-8  62—  5.0  49— 2.5 .  0.0  430-kDa MMP-8  ——  PBS  LPS  Figure 2.1. Impaired PMN responsiveness to LPS in MMP-8 deficient mice. (A) Infiltration of PMNs in vivo in response I tg of LPS (n = 8) or phosphate buffered saline control (n = 4) injected into the air pouch of male Mmp8 -I- (black bar) and wild type mice (white bar) was assessed 8 h post-injection. The PMN influx was quantitated by myeloperoxidase activity. Error bars, standard error. (B) Western blot analysis of murine MMP-8 in LPS-treated air pouch PMN lysates corresponding to 50,000 cells per lane.  42  with recombinant MMP-8.  Of these, LIX was the only chemokine susceptible to  proteolytic processing (Figure 2.2A)  —  KC, MIP-2, and DCIP-1 were MMP-8-  resistant even at high enzyme:substrate ratios (1:10) and after prolonged incubation times revealing protease substrate specificity. MALDI-TOF mass spectrometry analysis of the LIX cleavage products showed that MMP-8 processed the chemokine at two sites. The major product of 9,511 Da represents a deletion of the first four NH 2 terminal amino acid residues. Edman sequencing confirmed the deconvoluted MALDI-TOF data (Figure 2.2B). Hence, MMP-8 cleaves the 92-amino acid residue LIX between Ser-Val at position 4-5 to generate a new NH -terminus at Va1 2 5 that we designate LIX (5-92). With a measured mass of 8,113 Da, the second cleavage product was processed near the COOH-terminus after Lys , resulting in the removal 79 of 13 amino acid residues and generating the truncated form of LIX designated LIX (5-79) (Figure 2.2B). LIX (1-79) was never detected. Edman sequencing confirmed the deconvoluted mass spectrometry analysis of the 5 -Va1 amino-terminal 4 Ser cleavage site of LIX (5-79). As determined by MALDI-TOF mass spectrometry, the 2 NH and COOH-terminal processing of LIX was efficient; the cleavage products were first detected by mass spectrometry after 1 h with no full-length chemokine remaining at 18 h (Figure 2.2C). The COOH-terrninal cleavage product was not detected in the absence of NH 2 terminal processing at any time point indicating it occurs subsequent to 2 -NH terminal cleavage. Although the C-terminal cleavage was at an unusual highly cationic site 84 it was confirmed to be MMP dependent as two MMP inhibitors, KKKAK-RNALA 75 EDTA and the synthetic hydroxamate small molecule chemical inhibitor BB94, blocked all LIX processing with no cleavage products detected by MALDI-TOF mass spectrometry or Tris-tricine SDS-PAGE (Figure 2.2D). Chemokine cleavage is often enhanced by MMP exosite interactions (40). For instance, the hemopexin C domain of MMP-2 greatly increases the catalytic rate constants of CCL2, -7, -8 and -13 cleavage (19). When molar excess of recombinant MMP-8 hemopexin C-domain was added in the cleavage assays, we likewise found that all LIX processing by MMP-8 was abrogated revealing that a binding site for LIX was present on the  43  A  mlz [M+H]  B KC  DCIP-1 MMP-8+  -  +  -  LIX +  -  MIP-2  +  +  -  —  —  \%c,’\  Measured  Predicted  N-Terminal Sequence  LIX (1-92)  9,850  9,850  A-P-S-S-V-I  LIX (5-92)  9,511  9,507  V-I-A-A-T-E  LIX (5-79)  8,113  8,111  V-I-A-A-T-E  DCIP (1-72)  7,912  7,911  n.d.  KC (1-72)  7,811  7,812  A-P-I-A-N-E  MIP-2 (1-73)  7,845  7,847  n.d.  _\  C  D  Time 0 .51  -  2  — —  34  -  5  618  — —  MMP-8-  = c1 ?%_  1-92, 5-92, 5-79  +  +  —  —  b ;)‘  +  +  ccv t.  çv 0 1.0  Figure 2.2. MMP-8 Cleavage E E of LIX. (A) Tris-tricine 15% 0.8 IC) SDS-PAGE gel analysis of MMP-8 cleavage of murine 0.6 z w C, C.) CXCR2-binding chemokines C 0.4 DCIP-1, KC, LIX and MIP-2. .0 0 The m/z [M+H] of reaction 0.2 .0 products are as shown. (B) Identification of cleavage 0.0 products by MALDI-TOF mass 10 spectrometry and NH -terminal 2 Volume (mL) Edman sequencing. n.d., not determined. (C) Tris-tricine 15% SDS-PAGE and MALDI-TOF analysis of LIX cleavage products generated over time (h) by MMP-8. (D) Tris-tricine 15% SDS PAGE and MALDI-TOF analysis of LIX cleavage by MMP-8 in the presence of 10fold molar excess of recombinant hemopexin C-domain (MMP-8 CD), 10 .tM EDTA, or 10 tM BB94. (E) Heparin Sepharose AKTA chromatograms of synthetic analogues of MMP-8 cleavage products of LIX eluted with a linear gradient of NaCI as indicated. (Panels C&D contributed by Dr. Angus Tester) 44  hemopexin C domain of MMP-8 (Figure 2.2D). The C-terminal cleavage at 79 Lys ° only slightly reduced heparin affinity (Figure 2.2E) consistent with the removal 8 Arg of just two basic residues in the cleaved peptide 80 92 RNALAVER TASVQ while the Nterminal truncation had no effect on heparin binding. Processing of LIX by other MMPs Proteolytic screening of LIX cleavage by several MMPs showed that MMP-9, a prominent PMN MMP, processed LIX at 5 -Va1 (Figure 2.3). In addition to MMP 4 Ser 9, other important stromal, endothelial and leukocytic MMPs 1, -2, and -13 could also process LIX at 5 -Va1 (Figure 2.3), but like MMP-8, did not cleave KC, MIP-2, 4 Ser or DCIP-1 (not shown). Similar redundancy has been shown with MMPs for CXCLI2 (also know as SDF-1) (25) and for CCL2, -7, -8 and -13 (19). Protease selectivity was also shown with MMP-14, the RNA of which is also expressed by PMNs (data not shown), but was incapable of processing LIX at any position (Figure 2.3A). Further, only MMP-8 could process the C-terminus of LIX at Lys . 80 -Arg T9  Effect of LIX processing on in vitro biological activity Upon binding to the receptor CXCR2, LIX mobilizes intracellular Ca ion stores. As measured in recombinant CXCR2-expressing murine pre-B 300-19 cells, synthetic analogues of the MMP-truncated forms of LIX (5-92) and LIX (5-79) both induced an —2-fold greater intracellular Ca ion release compared to full-length LIX (1-92) (Figure 2.4A). This was confirmed using murine Mmp8 -I- PMN5 where an even greater stimulation in Ca ion release was observed upon binding LIX (5-92) compared with full-length LIX (Figure 2.4B).  Functionally, this translated into  enhanced chemoattraction for the CXCR2-expressing pre-B 300-19 cells by both of the truncated forms of LIX versus the full-length chemokine (Figure 2.4C), and in purified PMNs from Mmp8 +1+ mice compared to the unprocessed LIX (1-92) (Figure 2.4D). Notably, PMNs isolated from Mmp8 -I- mice migrated towards LIX (5-92) in a comparable manner to the Mmp8 +1+ PMNs (Figure 2.4D and data not shown) indicating that the locomotor functions of PMN5 isolated from both wild type and  45  A MMP  8  -  LIX(1-92) LlX(5-92) L IX(5-79)”  8  2  9 131314  —  —  c’\-  1  —  ,\Oj  4)’  4)’ 4)  4)’  (  \  4)’  B MMP-1, 2,8,9,13 10 20 I APSS+VIAATELRCVCLTVTPKINPKLIAN I  30  40  50  LEVI PAGPQCPTVEVIAKLKNQKEVCLDPE 60  70  80  90  APVI KKI IQKI LGSDKKKAKtRNALAVERTASVQ MMP-8  Figure 2.3. LIX is selectively cleaved by multiple MMPs. (A) Tris-tricine SDS PAGE and MALDI-TOF mass spectrometry analysis of MMP processing of LIX (enzyme:substrate ratio of 1:100 (w:w)) showing that MMPs 1, 2, 8, 9 and 13 cleave LIX at position 4-5, whereas MMP-14 does not, and that only MMP-8 also cleaves at position 79—80. Cleavage assays of rodent MMP-8 and MMP-13 are shown in the second 8 and 13 lanes. (B) Cleavage data are summarized using the full-length sequence of LIX. Experiment was performed by Dr. Angus Tester.  46  A  B 12  2  D Ci  C) 03  1o. C C  0 (‘3  N  8-  LIX (1-92) lJLX (5-92) LX (5-79)  D .  T  (‘3’— L.0  C’3 —(‘3  6-  2N  a) o  4.  C.)  =  C)  0-  10 nM  25 nM  C•’1 C)  0)  0) F.—  I  I  I  U)  x  ><  x  _J  _J  _I  D  C 100-  9-  0  a)  80-  -a.  _J  E  2 60-  a)  a).  40(‘3  200  c’.J  C) 1’;.  0) U)  (‘4  (‘.1  0)  I ‘-  I  I  C)  U)  -J  Mmp8  C) U)  F•—  CJ  0) I  U)  U) ‘--  %_  —  —  ><  ><  ><  ><  _J  _J  _I  J  +1+  —I—  Figure 2.4. In vitro cellular responses to MMP-8 cleaved LIX. (A) Enhanced intracellular calcium mobilization was induced by LIX (5-92) and LIX (5-79) compared to full-length LIX (1-92) in recombinant CXCR2-expressing B300-19 cells and (B) PMNs isolated from Mmp8 -I- mice (100 nM chemokine). (C) By transwell cell migration assay, both LIX (5-92) and LIX (5-79) truncated forms are more potent chemoatiractants compared with the full-length LIX (1-92) for both CXCR2-expressing B300-19 cell transfectants and (D) murine PMNs isolated from either Mmp8 +1+ or Mmp8 -I- mice, all at 10 nM chemokine concentration. (Panels C&D contributed by Dr. Angus Tester) 47  MMP-8 knock out mice were equivalent in vitro and unaffected by the presence or absence of MMP-8. In vivo PMN chemotaxis towards LIX Infiltration of PMNs towards LIX (1-92) injected in a dorsal skin air pouch of Mmp8-Imice was impaired at all time points compared to PMN infiltration in wild type mice, with an —2-fold lower number of PMNs seen at 8 and 12 h in knock out compared with wild type mice (Figure 2.5). In contrast, when LIX (5-92) or LIX (5-79) were used as chemoattractants there was no significant difference in PMN infiltration into the air pouches of wild type and mice lacking MMP-8 (Figure 2.5). This indicates that MMP-8 activity is not essential for blood vessel extravasation and PMN cell migration in vivo and that there is little physiological redundancy by PMN MMP-9, or from tissue MMPs that we found competent in cleaving and activating LIX in the biochemical context in vitro.  MMP-8 processes and activates IL-8 and ENA-78 Our experimental data suggest that upon LPS-induced release of LIX and resultant PMN chemoattraction a feed-forward PMN activation mechanism operates in vivo. MMP-8 released from degranulating PMNs at the site of challenge fully activates LIX in the tissue to further enhance PMN migration towards the LPS stimulus. To ascertain whether a similar autologous CXCR2 ligand activation mechanism occurs in man we assessed every human CXCR2 ligand for MMP-8 cleavage. Of the seven ELR CXC chemokines (CXCL8II L-8, CXCL7/NAP-2, CXCL6IGCP-2, CXCL5/ENA78, CXCL3/GROy, CXCL2IGRO3 and CXCLI/GROa) only CXCL8 and CXCL5 were processed by MMP-8. By MALDI-TOF MS and confirmation by Edman sequencing, CXCL8 was NH -terminally processed by MMP-8 at 6 2 -Ser to generate CXCL8 5 Arg (6-77) (Figure 2.6A, C) whereas CXCL5 was cleaved at 8 -Leu to generate 7 Va1 CXCL5 (8-78) (Figure 2.6C). There were no COOH-terminal cleavages detected. Another difference from the murine system was that CXCL8 cleavage was not hemopexin C domain dependent  —  proteolysis was not inhibited in the presence of  a molar excess of hemopexin C domain (Figure 2.6B). 48  22.0  1.5  LIX (1-92)  — KO  I  EEJ WT  Cl,  Z 0.5 0  a  4  8  12  -J  E  U,  C  x C,,  z 0  Time (h)  Figure 2.5. MMP-8 is required for PMN chemotaxis towards LIX in vivo, but is not required for PMN cell migration. PMN infiltration was greatly reduced in response to full-length LIX (1-92) injected into the dorsal skin air pouch of Mmp8I- mice (black bars) compared to wild type mice (white bars). PMN numbers were calculated from myeloperoxidase assay after sacrifice at 0, 4, 8 and 12 h following injection of chemokine (n = 4). Unaltered PMN cell migration into air pouches of Mmp8 -I- mice (black bars) compared to wild type mice (white bars) injected with MMP-cleaved analogues of LIX (5-92 or 5-79) reveals no intrinsic cell kinesis defects or chemotactic ability in the PMNs of the knock out mice. (Experiment performed by Dr. Angus Tester)  49  Figure 2.6. MMP processing of CXCL8 and CXCL5. (A) Tris-tricine 15% SDS PAGE gel and MALDI-TOF mass spectrometry analysis of CXCL8 cleavage products after assay with the indicated human MMP5 at an enzyme to substrate ratio of 1:100. Cleavage assays of rodent MMP-8 are shown in the second 8 lane. The NH -terminus of the CXCL8 truncated forms were deconvoluted from 2 the mass spectrometry data and confirmed by Edman sequencing (as shown in C) with the CXCL8 forms identified shown below the corresponding gel lanes. 177, full-length CXCL8. (B) The effect of recombinant MMP-8 hemopexin C domain (MMP-8 CD), 10 M EDTA, or 10 M BB94 on CXCL8 cleavage by human MMP 8 as determined by Tris-tricine 15% SDS-PAGE and MALDI-TOF mass spectrometry. The m/z [M+H] of reaction products is as shown. (C) Identification of the NH -termini of MMP cleavage products of CXCL8 and CXCL5 by MALDI 2 TOF mass spectrometry and NH -terminal Edman sequencing. 2 n.d., not determined. (D) Location of the various MMP cleavage sites on the CXCL8 and CXCL5 sequences. (Experiment performed by Dr. Angus Tester)  50  A  B  MMP  -  8  2  8  9 13 14  1  00  MMP-8  CXCL8  -  +  +  +  +  CXCL8  _\\  ,.\-  #  -(\,(\ j•’  is’  I.’.-  C  m/z[M+H] Measured Predicted N-terminus 8,919 8,919 A-V-L-P-R-S 8,387 8,383 S-A-K-E-L-R 8,295 8,296 n.d.  CXCL8 (1-77) CXCL8 (6-77) CXCL8 (7-77) CXCL5 (1-78) CXCL5(8-78) CXCL5 (10-78)  D  8,402 7,860 7,589  8,399 7,861 7,592  n.d. n.d. n.d.  MMP-8, 13, 14, r8 1  10  I  20  CXCL8 AVLPR$StAKELRCQCIKTYSK... MMP-1, 9 MMP-8 1  I  10  20  CXCL5 AGPAAAV’LRELRCVCLQTTQ... MMP-1, r8  51  To determine protease specificity, other MMPs were screened for cleavage of CXCL5 and CXCL8. Only MMP-1 and MMP-8 cleaved CXCL5, with MMPs 1, 8, 9, 13 and 14 cleaving CXCL8. Some differences in cleavage site specificity were identified, but all cleavages were NH -terminal to the ELR motif (Figure 2.6D) and 2 none cut in the C-terminal a-helix, as occurs for LIX. Consistent with previous studies (15, 26, 27) and our results with MMP-8 cleavage of LIX, MMP-8 processing of CXCL8 markedly activated the chemokine, with CXCL8 (6-77) leading to increased intracellular Ca mobilization (Figure 2.7A) and commensurate cell migration in transwells in vitro (Figure 2.7B). Despite several MMPs biochemically characterised to cleave and activate CXCL8, the critical importance of MMP-8 in CXCL8 activation was shown by injecting full-length CXCL8 in the air pouches of  Mmp8 -I- mice (Figure 2.7C). Here, the early PMN migration was decreased by >50% at 4 h compared with wild type mice. At later time points, there was less of a difference, but still it was always depressed in the knock out mice. Mmp8 -I- PMN responsiveness and cell migratory behaviour in vivo was also shown to be unaffected by the absence of MMP-8 when challenged with the synthetic analogue of MMP-8-cleaved CXCL8 (6-77) (Figure 2.7C). This reconfirms the critical role of MMP-8 in directing chemotaxis by chemokine processing rather than cleavage of other molecules such as those in the blood vessel wall or extracellular matrix. Similar results were obtained using synthetic analogues of MMP-8-cleaved CXCL5 (8-78) compared with full-length CXCL5 (1-78) in the air pouch model (data not shown). In using human chemokines in a murine setting, it was important to show that rodent MMP-8 did cleave CXCL8 at the same site as human MMP-8 (Figure 2.6A, D).  However, rodent MMP-8 cleaved human CXCL5 at 1 -Glu with no 9 Arg °  cleavage detected at 8 -Leu (Figure 2.6D). 7 Va1  Hence, these in vitro and in vivo  studies indicate that similar to PMN migration mechanisms towards LIX in mice, human PMN chemoattraction in response to CXCL8 and CXCL5 also exhibits a unique MM P-8 dependent feed-forward activation mechanism.  52  A  B 100  15  oEL_  -  10  80  60  —  0  20  ——  —  lOnM  25nM 0•  C -J  I  0  4  8  12  0  4  8  12  Time (h)  Figure 2.7. Enhanced bioactivity of MMP-truncated CXCL8 in vitro and in vivo. (A) Ca ion mobilization with 10 nM CXCL8 and (B) chemotaxis of CXCR2transfected B300-19 cells stimulated with 10 nM full-length CXCL8 (1-77) and CXCL8 (6-77) synthetic analogue of MMP-8-cleaved CXCL8. (C) Time course of air pouch PMN influx in response to CXCL8 (1-77) and CXCL8 (6-77) in Mmp8-Imice (black) and wild type (white) mice as quantitated by myeloperoxidase assay. (Panels B & C contributed by Dr. Angus Tester)  53  Discussion We have addressed the role of MMPs in the activation of ELR CXC chemokines in vivo.  Using Mmp8 -I- mice, the essential role of PMN MMP-8 was shown in the  activation pathway of the murine CXC chemokine LIX and human CXCL5 and CXCL8. Reflecting this, the absence of MMP-8 led to a profoundly defective PMN infiltration response in vivo to LPS or to full-length LIX, CXCL8, and CXCL5. This occurred despite the biochemical redundancy in chemokine activation by several commonly expressed MMPs including 1, 2, 9, 13, and 14 in a chemokine specific manner. PMN MMP-8 proteolysis leads to the activation of selected ELR CXC chemokines responsible for directing PMN cell migration and activation in vivo. With MMP-8 being primarily expressed by PMNs, our study identifies MMP-8 as an essential mediator of an interesting and unique activation mechanism of PMNs in innate immunity. This highlights an unexpectedly important role of the PMN itself in the integration of stimuli for the appropriate release of MMP-8 for LIX, CXCL8, and CXCL5 activation and so reveals an autologous cellular activation mechanism that acts in a feed-forward  manner to orchestrate the  PMN  influx and  LPS  responsiveness. Precise control of innate immunity is key for the development of successful inflammatory responses, resolution of infection, tissue healing, and ultimately host survival. As first line defence cells, the rapid recruitment and efficient activation of PMNs is of critical importance. It has long been known that the ELR CXCR2 cognate Iigands that direct PMN chemotaxis and activation are expressed as relatively inactive ligands requiring proteolytic truncation of up to eight amino acid residues from the amino terminus for full activity, providing that the essential ELR motif remains intact (15). However, relatively little attention has been focused on identifying the responsible enzymes in vivo and the location of chemokine activation —  key questions for understanding the control of PMN function and innate immunity  mechanisms. Naturally occurring NH -terminal truncated forms of ELR CXC 2 chemokines including LIX (17, 41), CXCL8 (42), CXCLI, 3 and 5 (43), have been reported but the proteases involved have not been identified in vivo. Mice deficient in  54  MMP-9 (44), neutrophil elastase (45), both MMP-9 and neutrophil elastase (46), or cathepsin G (47) exhibit a normal PMN chemotactic response, indicating that PMNs neither require these enzymes for CXC chemokine activation, nor for migration and efficient chemotaxis in vivo.  Hence, the deficient LPS and PMN migratory  responses in the Mmp8 -I- mouse are unique and reveal the singular importance of MMP-8 in these processes. However, we do not discount the possibility that in vivo other proteases downstream of MMP-8 activity may also activate LIX. Consistent with previous observations (15, 34, 41) was our demonstration through the use of in vitro calcium flux and chemotaxis assays that the MMP-8-processed chemokine products LIX (5-92), LIX (5-79), CXCL8 (6-77), and CXCL5 (8-78) were more potent than the full-length chemokines. Notably, LIX (5-92) and (5-79) were equally potent chemokines indicating that the C-terminal cleavage at position Lys 79 ° did not influence chemotactic activity. 8 Arg  Chemokines interact with the highly  negatively charged glycosaminoglycan chains of proteoglycans resulting  in  immobilization and the generation of a haptotactic gradient within the extracellular matrix that is responsible for directing leukocyte migration (48).  This has been  confirmed physiologically for CC type chemokines (49) and murine ELR CXC chemokines (50).  Binding is typically through a heparin binding consensus  sequence BBXB where B is a basic residue (51). The MMP-8 C-terminal cleavage of LIX only removes two basic amino acids, both arginines, in the truncated product and so only slightly reduces heparin affinity. The biological significance of this cleavage is therefore unknown. None of the MMP5 analysed could cleave any of the other murine ELR chemokines (data not shown) revealing the specificity of MMP chemokine cleavage. This susceptibility of LIX to MMP processing is likely due to an extended N-terminus before the ELR motif, as compared with KC, MIP-2, and DCIP-1.  Interestingly,  rodent MMP-8 cleaved CXCL5 at 8 -Leu and not at the same site as human 7 Val MMP-8, 10 -G1u The active site of murine and rat MMP-8 differs from that of 9 Arg . human MMP-8 by the presence of Lys 187 at S 187 as found in human ’ instead of Ala 3 MMP-8. Charge repulsion between Lys’ 87 in rodent MMP-8 at S P2 ’ Arg ’ with the 3 3  55  in CXCL5 may preclude CXCL5 cleavage at 1 -Glu and instead favour cleavage 9 Arg ° at . ° of CXCL5 would be the P 1 8 7 Va1 Leu Here G1u ’ residue and so likely forms a 3 salt bridge with Lys . 187  Although LIX shows structural homology to CXCL5 and  CXCL6 (52), MMP-8 exclusively cleaved CXCL5 and CXCL8 amongst all seven human ELR CXC chemokines. The activation of CXCL5 and CXCL8 were shown by  in vitro calcium flux and chemotaxis assays, with the singular importance of MMP-8 in activating these human chemokines in vivo shown from studies comparing the  Mmp8 -I- mice with the wild type controls. Our results reveal that chemokine processing is one of the most important functions of MMP-8 in vivo and casts into doubt the importance of MMP-8 in collagen degradation, a role that has long been assumed to be of particular importance for PMN cell migration and chemotaxis. Indeed, using air pouch models, no difference in PMN cell migration and infiltration was observed in response to synthetic analogues of the MMP-8 cleaved LIX, CXCL8, and CXCL5. This shows that the PMN cell migration machinery to chemotactic agents does not require MMP-8 activity for responsiveness and that MMP-8 proteolysis of blood vessel basement membrane and interstitial extracellular matrix components is not essential for effective cell migration in vivo. Although MMPs have traditionally been thought to cleave extracellular matrix components and so disrupt extracellular matrix contacts with the tumour and potentiate tumour cell spread and metastasis, other biological roles for MMPs in cancer are now known (20, 53). Potentially related to this, sustained inflammatory responses maintain a microenvironment advantageous to tumour growth (54). Indeed, MMP-8 modulates the innate immune response induced by carcinogens leading to a protective role in preventing tumour progression (34). Mice lacking MMP-8 exhibited an abnormal inflammatory response upon application of carcinogen, with a delayed and more diffuse PMN influx to the site of the host challenge. Once established though, the inflammatory response was sustained and the mechanism for this is under investigation in our laboratory. With MMP-8 being the first MMP reported to have a protective role in tumorigenesis (34), the recognition of further MMP anti-targets in cancer continues (35). MMP-3 has a protective role in squamous cell carcinoma (55) and macrophage MMP-12 is an anti 56  target in lung carcinoma (56). In these cases the proteases were reported to alter leukocyte infiltration, although the mechanism and substrates were not elucidated. In view of the considerable number of chemokines now known to be processed by MMPs, chemokines are strong candidate substrates to phenotypically explain cancer anti-target activity in these MMP genetic knockout mice. The role of MMPs in LPS responsiveness and PMN migration differs from that found for macrophages and CXCR4-displaying leukocytes. Instead of promoting cell migration, MMP cleavage of CC chemokines CCL2, -7, -8 and -13 results in the loss of agonist activity and the generation of potent in vitro and in vivo CCR antagonists (18, 19). Interestingly, the MMP-2 cleavage and inactivation of CXCLI2 (25) in the brain generates a potent and selective neurotoxin implicated in HIV dementia (57, 58). We also recently found that MMP-2 induces the shedding of the integral plasma membrane chemokine 3 CX C L1 (fractalkine) by release of the chemokine domain from the stalk at A1a 72 (23). -Leu 71  Further, the cell surface agonist activity of  CX3CLI was converted to a soluble antagonist due to processing at . 5 4 Gly Met Hence, MMPs dynamically regulate the biological activity of chemokines and inflammatory and immune cell function in pleiotropic ways.  Our present studies  suggest that an in cis feed-forward activation mechanism occurs in which the PMN integrates the tissue signalling milieu leading to controlled release of MMP-8 that either directly or indirectly activates LIX in the mouse, and CXCL8 and CXCL5 in man for further PMN migration.  57  References 1. 2. 3.  4. 5.  6. 7.  8. 9.  10.  11.  12.  13.  14.  15.  16. 17.  Faurschou, M, and N. Borregaard. 2003. Neutrophil granules and secretory vesicles in inflammation. Microbes Infect 5:1317-1327. Park, J. E., and A. Barbul. 2004. Understanding the role of immune regulation in wound healing. Am J Surg 187:11S-16S. Bodey, G. P., M. Buckley, Y. S. Sathe, and E. J. Freireich. 1966. Quantitative relationships between circulating leukocytes and infection in patients with acute leukemia. Ann Intern Med 64:328-340. Pizzo, P. A. 1993. Management of fever in patients with cancer and treatment-induced neutropenia. N Engi J Med 328:1323-1332. Hughes, W. T., D. Armstrong, G. P. Bodey, E. J. Bow, A. E. Brown, T. Calandra, R. Feld, P. A. Pizzo, K. V. Rolston, J. L. Shenep, and L. S. Young. 2002. 2002 guidelines for the use of antimicrobial agents in neutropenic patients with cancer. CIin Infect Dis 34:730-751. Madri, J. A., and D. Graesser. 2000. Cell migration in the immune system: the evolving inter related roles of adhesion molecules and proteinases. Dev Immunol 7:103-116. Lillard, J. W., Jr., P. N. Boyaka, 0. Chertov, J. J. Oppenheim, and J. R. McGhee. 1999. Mechanisms for induction of acquired host immunity by neutrophil peptide defensins. Proc NatlAcad Sd U S A 96:651-656. Foxman, E. F., J. J. Campbell, and E. C. Butcher. 1997. Multistep navigation and the combinatorial control of leukocyte chemotaxis. J Cell Biol 139:1349-1360. Clark-Lewis, I., K. S. Kim, K. Rajarathnam, J. H. Gong, B. Dewald, B. Moser, M. Baggiolini, and B. D. Sykes. 1995. Structure-activity relationships of chemokines. J Leukoc Biol 57:703711. Utgaard, J. 0., F. L. Jahnsen, A. Bakka, P. Brandtzaeg, and G. Haraldsen. 1998. Rapid secretion of prestored interleukin 8 from Weibel-Palade bodies of microvascular endothelial cells. J Exp Med 188:1751-1 756. Olszyna, D. P., E. De Jonge, P. E. Dekkers, S. J. van Deventer, and T. van der Poll. 2001. Induction of cell-associated chemokines after endotoxin administration to healthy humans. Infect Immun 69:2736-2738. Arcaroli, J., J. Kupfner, H. K. Yum, R. Shenkar, J. S. Park, and E. Abraham. 2002. Lipopolysaccharide-induced neutrophil gene expression under in vivo or in vitro conditions. Chest 121 :76S. Malcolm, K. C., P. G. Arndt, E. J. Manos, D. A. Jones, and G. S. Worthen. 2003. Microarray analysis of lipopolysaccharide-treated human neutrophils. Am J Physiol Lung Cell Mol Physiol 284:L663-670. Wuyts, A., P. Proost, J. P. Lenaerts, A. Ben-Baruch, J. Van Damme, and J. M. Wang. 1998. Differential usage of the CXC chemokine receptors 1 and 2 by interleukin-8, granulocyte chemotactic protein-2 and epithelial-cell-derived neutrophil attractant-78. Eur J Biochem 255:67-73. Clark-Lewis, I., C. Schumacher, M. Baggiolini, and B. Moser. 1991. Structure-activity relationships of interleukin-8 determined using chemically synthesized analogs. Critical role of NH2-terminal residues and evidence for uncoupling of neutrophil chemotaxis, exocytosis, and receptor binding activities. J 8101 Chem 266:23128-231 34. Lee, J., G. Cacalano, T. Camerato, K. Toy, M. W. Moore, and W. I. Wood. 1995. Chemokine binding and activities mediated by the mouse IL-8 receptor. J Immunol 155:2158-2164. Wuyts, A., A. Haelens, P. Proost, J. P. Lenaerts, R. Conings, G. Opdenakker, and J. Van Damme. 1996. Identification of mouse granulocyte chemotactic protein-2 from fibroblasts and epithelial cells. Functional comparison with natural KC and macrophage inflammatory protein 2. J Immunol 157:1736-1 743.  58  18.  19.  20. 21. 22.  23.  McQuibban, G. A., J. H. Gong, E. M. Tam, C. A. McCulloch, I. Clark-Lewis, and C. M. Overall. 2000. Inflammation dampened by gelatinase A cleavage of monocyte chemoattractant protein-3. Science 289:1202-1206. McQuibban, G. A., J. H. Gong, J. P. Wong, J. L. Wallace, I. Clark-Lewis, and C. M. Overall. 2002. Matrix metalloproteinase processing of monocyte chemoattractant proteins generates CC chemokine receptor antagonists with anti-inflammatory properties in vivo. Blood 100:1160-1167. Egeblad, M., and Z. Werb. 2002. New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2:161-174. Overall, C. M. 2004. Dilating the degradome: matrix metalloproteinase 2 (MMP-2) cuts to the heart of the matter. Biochem J 383:e5-7. Tam, E. M., C. J. Morrison, Y. I. Wu, M. S. Stack, and C. M. Overall. 2004. Membrane protease proteomics: Isotope-coded affinity tag MS identification of undescribed MT1-matnx metalloproteinase substrates. Proc Nat! Acad Sd U S A 101:6917-6922. Dean, R. A., and C. M. Overall. 2007. Proteomic discovery of metalloproteinase substrates in the cellular context by iTRAQO labeling reveals a diverse MMP-2 substrate degradome. Mo!  Cell Proteomics. 24. 25.  26.  27.  28.  29.  30.  31. 32.  33. 34.  35.  Parks, W. C., C. L. Wilson, and Y. S. Lopez-Boado. 2004. Matrix metalloproteinases as modulators of inflammation and innate immunity. Nat Rev !mmunol 4:617-629. McQuibban, G. A., G. S. Butler, J.-H. Gong, L. Bendall, C. Power, I. Clark-Lewis, and C. M. Overall. 2001. Matrix Metal loproteinase Activity Inactivates the CXC Chemokine Stromal Cellderived Factor-i. J. Biol. Chem. 276:43503-43508. Van den Steen, P. E., P. Proost, A. Wuyts, J. Van Damme, and G. Opdenakker. 2000. Neutrophil gelatinase B potentiates interleukin-8 tenfold by aminoterminal processing, whereas it degrades CTAP-lll, PF-4, and GRO-alpha and leaves RANTES and MCP-2 intact. Blood 96:2673-2681. Van Den Steen, P. E., A. Wuyts, S. J. Husson, P. Proost, J. Van Damme, and G. Opdenakker. 2003. Gelatinase B/MMP-9 and neutrophil collagenase/MMP-8 process the chemokines human GCP-2/CXCL6, ENA-78/CXCL5 and mouse GCP-2/LIX and modulate their physiological activities. Eur J Biochem 270:3739-3749. Gearing, A. J., P. Beckett, M. Christodoulou, M. Churchill, J. Clements, A. H. Davidson, A. H. Drummond, W. A. Galloway, R. Gilbert, and J. L. Gordon. 1994. Processing of tumour necrosis factor-alpha precursor by metalloproteinases. Nature 370:555-557. Wilson, C. L., A. J. Ouellette, D. P. Satchell, T. Ayabe, Y. S. Lopez-Boado, J. L. Stratman, S. J. Hultgren, L. M. Matrisian, and W. C. Parks. 1999. Regulation of intestinal alpha-defensin activation by the metalloproteinase matrilysin in innate host defense. Science 286:113-i 17. Butler, G. S., D. Sim, E. Tam, D. Devine, and C. M. Overall. 2002. Mannose-binding lectin (MBL) mutants are susceptible to matrix metalloproteinase proteolysis: potential role in human MBL deficiency. J Biol Chem 277:17511-17519. Coussens, L. M., and Z. Werb. 2002. Inflammation and cancer. Nature 420:860-867. Acuff, H. B., K. J. Carter, B. Fingleton, D. L. Gorden, and L. M. Matrisian. 2006. Matrix metalloproteinase-9 from bone marrow-derived cells contributes to survival but not growth of tumor cells in the lung microenvironment. Cancer Res 66:259-266. Weiss, S. J., G. Peppin, X. Ortiz, C. Ragsdale, and S. T. Test. 1985. Oxidative autoactivation of latent collagenase by human neutrophils. Science 227:747-749. Balbin, M., A. Fueyo, A. M. Tester, A. M. Pendas, A. S. Pitiot, A. Astudillo, C. M. Overall, S. D. Shapiro, and C. Lopez-Otin. 2003. Loss of collagenase-2 confers increased skin tumor susceptibility to male mice. Nat Genet 35:252-257. Overall, C. M., and 0. Kleifeld. 2006. Validating matrix metalloproteinases as drug targets and anti-targets for cancer therapy. Nat Rev Cancer 6:227-239.  59  36. 37.  38.  39.  40.  41.  42.  43.  44.  45.  46.  47. 48. 49.  50.  51.  Hasty, K. A., J. J. Jeffrey, M. S. Hibbs, and H. G. Welgus. 1987. The collagen substrate specificity of human neutrophil collagenase. J Biol Chem 262:10048-10052. Pelman, G. R., C. J. Morrison, and C. M. Overall. 2005. Pivotal molecular determinants of peptidic and collagen triple helicase activities reside in the S3’ subsite of matrix metalloproteinase 8 (MMP-8): the role of hydrogen bonding potential of ASN188 and TYR189 and the connecting cis bond. J Biol Chem 280:2370-2377. Van Lint, P., B. Wielockx, L. Puimege, A. Noel, C. Lopez-Otin, and C. Libert. 2005. Resistance of collagenase-2 (matrix metalloproteinase-8)-deficient mice to TN F-induced lethal hepatitis. J Immuno! 175:7642-7649. Owen, C. A., Z. Hu, C. Lopez-Otin, and S. D. Shapiro. 2004. Membrane-bound matrix metalloproteinase-8 on activated polymorphonuclear cells is a potent, tissue inhibitor of metalloproteinase-resistant collagenase and serpinase. J Immunol 172:7791-7803. Overall, C. M., G. A. McQuibban, and I. Clark-Lewis. 2002. Discovery of Chemokine Substrates for Matrix Metalloproteinases by Exosite Scanning: A New Tool for Degradomics. Biol Chem 383:105-1 066. Wuyts, A., A. D’Haese, V. Cremers, P. Menten, J. P. Lenaerts, A. De Loof, H. Heremans, P. Proost, and J. Van Damme. 1999. NH2- and COOH-terminal truncations of murine granulocyte chemotactic protein-2 augment the in vitro and in vivo neutrophil chemotactic potency. J immunol 163:6155-6163. Van Damme, J., M. Rampart, R. Conings, B. Decock, N. Van Osselaer, J. Willems, and A. Billiau. 1990. The neutrophil-activating proteins interleukin 8 and beta-thromboglobulin: in vitro and in vivo comparison of NH2-terminally processed forms. Eur J Immunol 20:21132118. Wuyts, A., C. Govaerts, S. Struyf, J. P. Lenaerts, W. Put, R. Conings, P. Proost, and J. Van Damme. 1999. Isolation of the CXC chemokines ENA-78, GRO alpha and GRO gamma from tumor cells and leukocytes reveals NH2-terminal heterogeneity. Functional comparison of different natural isoforms. EurJ BIochem 260:421-429. Betsuyaku, T., J. M. Shipley, Z. Liu, and R. M. Senior. 1999. Neutrophil emigration in the lungs, peritoneum, and skin does not require gelatinase B. Am J Respir Cell Mo! Biol 20:1303-1309. Belaaouaj, A., R. McCarthy, M. Baumann, Z. Gao, T. J. Ley, S. N. Abraham, and S. D. Shapiro. 1998. Mice lacking neutrophil elastase reveal impaired host defense against gram negative bacterial sepsis. Nat Med 4:615-618. Allport, J. R., Y. C. Lim, J. M. Shipley, R. M. Senior, S. D. Shapiro, N. Matsuyoshi, D. Vestweber, and F. W. Luscinskas. 2002. Neutrophils from MMP-9- or neutrophil elastase deficient mice show no defect in transendothelial migration under flow in vitro. J Leukoc 810! 71:821-828. Maclvor, D. M., S. D. Shapiro, C. T. Pham, A. Belaaouaj, S. N. Abraham, and T. J. Ley. 1999. Normal neutrophil function in cathepsin G-deficient mice. Blood 94:4282-4293. Rot, A. 1993. Neutrophil attractantlactivation protein-i (interleukin-8) induces in vitro neutrophil migration by haptotactic mechanism. Eur J Immuno! 23:303-306. Proudfoot, A. E., T. M. Handel, Z. Johnson, E. K. Lau, P. LiWang, I. Clark-Lewis, F. Borlat, T. N. Wells, and M. H. Kosco-Vilbois. 2003. Glycosaminoglycan binding and oligomerization are essential for the in vivo activity of certain chemokines. Proc Nat! Acad Sd U S A 100:18851890. Li, Q., P. W. Park, C. L. Wilson, and W. C. Parks. 2002. Matrilysin shedding of syndecan-1 regulates chemokine mobilization and transepithelial efflux of neutrophils in acute lung injury. Cell 111:635-646. Perrimon, N., and M. Bernfield. 2000. Specificities of heparan sulphate proteoglycans in developmental processes. Nature 404:725-728.  60  52.  53. 54. 55.  56.  57.  58.  59.  60.  61.  Smith, J. B., L. E. Rovai, and H. R. Herschman. 1997. Sequence similarities of a subgroup of CXC chemokines related to murine LIX: implications for the interpretation of evolutionary relationships among chemokines. J Leukoc Bio!62:598-603. Coussens, L. M., and Z. Werb. 1996. Matrix metalloproteinases and the development of cancer. Chem Biol 3:895-904. Dvorak, H. F. 1986. Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing. NEngIJMed3l5:1650-1659. McCawley, L. J., H. C. Crawford, L. E. King, Jr., J. Mudgett, and L. M. Matrisian. 2004. A protective role for matrix metalloproteinase-3 in squamous cell carcinoma. Cancer Res 64:6965-6972. Houghton, A. M., J. L. Grisolano, M. L. Baumann, D. K. Kobayashi, R. D. Hautamaki, L. C. Nehring, L. A. Cornelius, and S. D. Shapiro. 2006. Macrophage elastase (matrix rnetalloproteinase-12) suppresses growth of lung metastases. Cancer Res 66:6149-6155. Zhang, K., G. S. McQuibban, C. Silva, G. S. Butler, J. B. Johnston, I. Clark-Lewis, J. Holden, C. M. Overall, and C. Power. 2003. Metalloproteinase Cleavage of the Chemokine SDF-1x Induces Neuronal Apoptosis in I-f IV Encephalitis. Nature Neuroscience 6:1064-1071. Vergote, D., G. S. Butler, M. Corns, J. H. Cox, C. Silva, M. D. Hollenberg, J. H. Jhamandas, C. M. Overall, and C. Power. 2006. Proteolytic processing of SDF-lalpha reveals a change in receptor specificity mediating HIV-associated neurodegeneration. Proc Nat! Acad Sd U S A 103:19182-19187. Edwards, J. C., A. D. Sedgwick, and D. A. Willoughby. 1981. The formation of a structure with the features of synovial lining by subcutaneous injection of air: an in vivo tissue culture system. J Pathol 134:147-156. Tateda, K., T. A. Moore, M. W. Newstead, W. C. Tsai, X. Zeng, J. C. Deng, G. Chen, R. Reddy, K. Yamaguchi, and T. J. Standiford. 2001. Chernokine-dependent neutrophil recruitment in a murine model of Legionella pneumonia: potential role of neutrophils as immunoregulatory cells. Infect Immun 69:2017-2024. Clark-Lewis, I., L. Vo, P. Owen, and J. Anderson. 1997. Chemical synthesis, purification, and folding of C-X-C and C-C chemokines. Methods Enzymol 287:233-250.  61  CHAPTER 3. MATRIX METALLOPROTEINASE PROCESSING OF CXCLI Ill-TAG RESULTS IN LOSS OF CHEMOATTRACTANT ACTIVITY AND ALTERED GLYCOSAMINOGLYCAN BINDING 3 Perspective The CXCR3 chemokine receptor regulates the migration of Thi lymphocytes and responds to three ligands: CXCL9/MIG, CXCLIO/IP-1O, and CXCLII/I-TAC. We screened for potential regulation of T cell responses by matrix metalloproteinase (MMP) processing of these important chemokines. The most potent of the CXCR3 ligands, CXCLI 1, was identified by MALD I-TOE MS as a substrate of the PMN specific MMP-8, macrophage-specific MMP-12, and the general leukocyte MMP-9. The 73-amino acid residue CXCLI I is processed at both the amino and carboxy termini to generate CXCLI 1 (5-73), (5-63), and (5-58) forms. N-terminal truncation results in loss of agonistic properties, as shown in calcium mobilization and chemotaxis experiments using CXCR3-transfectants and human T lymphocytes. Moreover, CXCLI 1 (5-73) is a CXCR3 antagonist and interestingly shows enhanced affinity to heparin. However, upon C-terminal truncation to position 58 there is loss of antagonist activity and heparin binding. Together this highlights an unexpected site for  receptor  interaction  and  that  the  carboxy  terminus  is  critical  for  glycosaminoglycan binding, an essential function for the formation of chemokine gradients in vivo. Hence, MMP activity might regulate CXCL1 I tissue gradients in two ways. First, the potential of CXCLI 1 (5-73) to compete active CXCLI 1 from glycosaminoglycans might lead to the formation of an antagonistic haptotactic chemokine gradient. Second, upon further truncation, MMPs disperse the CXCLI I gradients by proteolytic loss of a C-terminal GAG binding site. Hence, these results reveal potential new roles in downregulating Thi lymphocyte chemoafiraction through MMP processing of CXCL1 1.  A version of this chapter has been published. Cox JH, Dean RA, Roberts CR, and Overall CM. (2008) Matrix metalloproteinase processing of CXCLI 1/1-TAC results in loss of chemoattractant activity and altered glycosaminoglycan binding. J Biol Chem. 283:19389-99.  62  Introduction Chemokines are a super-family of low molecular weight chemotactic cytokines that function in directing the migration of leukocytes and other cell types in a multitude of processes including development, lymphocyte homing, inflammation and wound repair (1). Chemokines form haptotactic gradients in vivo through associations with proteoglycan glycosaminoglycans (2).  Upon interaction with 7-transmembrane G  protein-coupled receptors, chemokines induce a chemotactic response.  The  expression and secretion of inducible chemokines is stimulated during infection or injury to promote rapid and efficient inflammatory and immune responses. Conversely, dampening of inflammation, a critical event in allowing tissue repair to continue unimpeded and in preventing excessive tissue damage and autoimmunity, is known to involve coordinated down-regulation of chemokine expression (3), receptor internalization (4), scavenger receptors (5), and proteolytic mechanisms of inactivation and conversion to antagonists (6). Specific and limited proteolysis, termed processing (7), of chemokines results in altered bioactivity with functional consequences such as increased or decreased receptor binding (8), conversion of an agonist to an antagonist (6, 9), shedding of membrane-anchored chemokines (10, 11), and changing receptor specificity (12). Cleavage of proteoglycan core protein also disrupts chemokine gradients formed by chemokine binding to the glycosaminoglycan side chains (13).  A variety of  proteases are thought to be involved in chemokine cleavage including CDI 3, also known as aminopeptidase N (14), CD26, also known as dipeptidylpeptidase-IV (15, 16), dipeptidylpeptidase-8 (16),  neutrophil elastase (17),  cathepsin G (18),  cathepsins B and D (19), and the matrix metalloproteinases (MMPs) [Reviewed in (20-22)1.  Several chemokines are important physiological substrates of MMPs, a  family of proteases once thought to be only associated with extracellular matrix turnover (23), including the monocyte chemoattractant proteins CCL2, 7, 8, and 13 (6, 9), the ELR chemokines CXCL5/ENA-78, CXCL8/IL-8 and murine LIXImCXCL5 (24-27),  CXCL9 and  CXCL1O (28),  CXCLI2/SDF-Ia and  (8,  29), and  CL1/Fractalkine (10). 3 CX  63  MMPs are zinc-dependent extracellular endopeptidases, with most having a carboxy-terminal hemopexin domain containing exosites for substrate binding (30). MMPs are upregulated in many pathologies including rheumatoid arthritis (31), multiple sclerosis (32, 33), and tumongenesis (7, 34), all of which have a chronic inflammatory component. The expression of numerous MMPs by immune cells is commonly thought to promote cellular migration through extracellular matrix degradation and hence removal of physical barriers (23). However, this concept is outmoded following T lymphocyte migration studies in 3D extracellular matrices in the presence of potent protease inhibitors (35),  MMP-8 knockout mice, where  deficiency in this neutrophil collagenase did not reduce PMN cell migration in response to activated chemokines (25), and recognition of a multitude of bioactive substrates of MMP5 through genetic (23) and proteomic analyses (10, 36, 37). Interestingly, MMPs are produced by cells of both the innate and adaptive arms of immunity, suggesting pleiotropic roles coordinating rapid host defense and more specific or specialized acquired immunity. Examples include the neutrophil-specific MMP-8, the macrophage-specific MMP-12 (also known as metalloelastase), and MMP-9 (also known as gelatinase B), which is expressed by a variety of leukocytes. Interstitial collagenase (MMP-1) and MMP-7 (also known as matrilysin), an endothelial/epithelial-derived protease, are also expressed by macrophages (20). CD4 T helper lymphocytes (Thi) are involved in managing humoral and cellmediated responses through cytokine release and subsequent activation of effector cells. Notably, the gelatinases MMP-2 (also known as gelatinase A) and MMP-9 are preferentially expressed by the Thi subset and CD4 migration is reduced in the presence of MMP inhibitors (38). Furthermore, membrane type-i MMP (MMP-14) is the major physiological activator of MMP-2 (39) and is expressed by monocytes and dendritic cells (40, 41). The CXCR3 chemokine receptor is preferentially expressed on Thi cells and has three known ligands: CXCL9/MIG, CXCLIO/IP-10, and CXCLIIII-TAC.  These  chemokines are induced by the pro-inflammatory cytokine interferon-gamma in  64  several cell types including neutrophils, monocytes, macrophages, T cells, astrocytes, fibroblasts, and endothelial cells (1).  The CXCR3 receptor and its  cognate chemokines have been implicated in inflammatory disorders such as multiple sclerosis, rheumatoid arthritis, and allograft rejection, as they are upregulated in these pathological states, presumably causing enhanced infiltration of CXCR3-expressing Thi cells (42-44). As such, precise regulation of the CXCR3 ligands is key in preventing excessive Thi recruitment and the resulting pathology. Previous reports have shown that CXCLI 0 is cleaved at the carboxy-terminus by furin in the cellular context (45), and that CXCL9 and CXCLIO can be truncated at the carboxy-terminus by MMP-8 and MMP-9 (28), however no functional changes were found for any of these cleavages. Like many chemokines, CXCLI I is cleaved at the amino-terminus after proline-2 by dipeptidylpeptidase IV (46, 47) and aminopeptidase N (14), resulting in loss of agonism, but the propensity of MMP5 in processing this chemokine are unknown. Here, we have biochemically evaluated CXCL9, CXCL1 0, and CXCLI I as leukocytic-MMP substrates and report that CXCLII is processed by several MMPs resulting in altered receptor binding and glycosaminoglycan affinity.  In particular, CXCLI I is first converted to a CXCR3  antagonist by N-terminal processing at position 4-5, but this is lost upon C-terminal truncation revealing an unexpected site for receptor binding. Therefore, with proteolytic loss of a C-terminal glycosaminoglycan-binding site this might result in dispersal of chemotactic gradients in vivo and so contribute to regulation of Thi cell recruitment and cell accumulation.  Materials and Methods Chemokine cleavage assays.  All chemokines and analogues of MMP-cleaved  chemokines were synthesized using tBoc (tertiary butyloxycarbonyl) solid phase chemistry as described previously (48).  Recombinant human MMP-1, 2, 8, 9, 12,  and soluble MMP-14 (lacking the transmembrane domain) were expressed and purified using standard techniques (49, 50).  Human MMP-7 was purchased from  65  USB Corporation.  Enzyme activity was followed by quenched fluorescence  synthetic peptide cleavage assay with Mca-Pro-Leu-GIy-Leu-Dpa-Ala-Arg-NH 2 (51). Chemokine cleavage assays were performed in 50 mM Tris, 200 mM NaCI, 5 mM , pH 7.5 with initial screening at an enzyme to substrate ratio of 1:10 (w:w) for 2 CaCI 16 h at 37 °C, then assessed at lower enzyme concentrations to establish kinetic parameters, biological efficacy and relevance.  Unlike the other MMPs assayed,  MMP-12 showed significant loss of activity through autodegradation in 16 h incubations, therefore kinetic analysis was performed after 2 h.  Inhibition  experiments were done in the presence of 10 1 M EDTA, 10 1 iM Batimastat, or 10fold molar excess MMP-8 hemopexin C domain. Assay results were analyzed by matrix-assisted  laser desorption  ionization  time-of-flight  (MALDI-TOF)  mass  spectrometry on a Voyager-DE STR (Applied Biosystems) in sinapinic acid matrix and confirmed by 15% Tris-Tricine SDS-polyacrylamide gel electrophoresis (PAGE).  Hemopexin C domain expression and binding. MMP-8 linker and hemopexin C domain with an NH -terminal His-tag was cloned into the pGYMX vector and 2 expressed in E. coil BL21 gold cells.  Cell lysates were solubilized and protein  purified by Ni -chelate chromatography followed by Superdex 2 TM 75 gel filtration chromatography (GE Healthcare) yielding a pure 26-kDa product, confirmed to be MMP-8 linker and hemopexin C domain by western blotting with antibody against the MMP-8 C domain (Chemicon).  Chemokines were immobilized in 96-well plates  (Falcon) at 500 ng/well in phosphate buffered saline (PBS) (138 mM NaCI, 2.7 mM KCI, 20 mM 4 HPO 1.5 mM , 2 Na , 4 P 2 KH O pH 7.4) overnight at 4 °C.  Wells were  blocked for I h with PBS containing 2.5% bovine serum albumin (BSA). Binding of recombinant N-terminal His-tagged MMP-8 hemopexin C domain to immobilized chemokines was performed for I h in PBS at room temperature and was assessed by direct ELISA with alkaline phosphatase detection (Sigma). Washes were with PBS containing 0.05% Tween-20.  66  Transfected and isolated cells.  Human CXCR3-transfected B300-19 cells (52)  were kindly provided by B. Moser (Bern, Switzerland) and were cultured in RPMI 1640 supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, 50 iM  -  mercaptoethanol, and 1.0 mglmL G-418. Human T cells were isolated from healthy volunteers as approved by the University of British Columbia Clinical Ethics Review Board (Appendix C). Peripheral blood was drawn into EDTA-treated vaccutainers and layered on Ficoll-Paque Plus (Amersham Biosciences), as per manufacturers protocol.  Monocytes were removed by adhesion to flask surfaces following 3 h  incubation at 37 °C, 5% CO . Isolated lymphocytes were cultured in RPMI-1640, 2 2 mM L-glutamine, 100 U/mL penicillin, and 0.1 mg/mL streptomycin.  Cells were  activated by the addition of 5 igImL phytohemagglutinin-P (Sigma) for three days followed by six to twelve days in the presence of 100 U/mL lnterleukin-2 (Peprotech).  Media was changed every three days.  Expression of CXCR3 was  confirmed by flow cytometry (Beckman Coulter EPICS XL-MCL) using mouse anti hCXCR3 (R&D Systems, clone 49801). Calcium mobilization. CXCR3-transfected B300-1 9 cells were resuspended at I x 10 cells/mL in RPMI-1640 media supplemented with 1% FBS and incubated with 2 iM Fluo-4-acetoxymethyl ester (Molecular Probes) for 30 mm  at 37 °C. Cells were  washed to remove unincorporated agent and resuspended at I x 106 cells/mL in Hanks Balanced Salt Solution (without calcium or magnesium, Sigma), 20 mM HEPES, 2.5 mM probenecid (Sigma). 3.5 x i0 5 cells were plated per well in 96-well clear bottom fluorescent plates (Nunc) and centrifuged at 1,000 x g for 5 mm without braking. Cells were allowed to equilibrate at 37 °C for 15 mm  prior to addition of  ligand. Calcium concentration was monitored by excitation of 485 nm and emission of 520 nm with a Molecular Devices Flexstation II as described previously (53). Calibration was performed by addition of 5 1 iM ionomycin (Sigma) followed by 1 mM 2 (Fisher Biotech) to determine Fmax and Fmin, resepctively. MnCI  Absolute Ca 2  concentrations were calculated as Kd x [(FFmin)/(FmaF)j (54) where the Kd of Fluo-4 was 345 nM, as reported by Molecular Probes. Antagonist experiments were performed by preincubating cells in the presence of CXCLI 1 (5-73) or CXCL1 1 (5-  67  58), synthetic analogues of MM P-processed CXCLI 1 at position 4-5 (CXCL1 1 (573)) and 58-59 (CXCLI 1 (5-58)), for 120 s prior to the addition of 3 nM full-length CXCLI 1, designated CXCLI 1 (1-73). CXCR3 Internalization. Internalization of CXCR3 on CXCR3-transfected B300-19 cells was measured following the incubation of 5 x i0 5 cells with 10 nM and 100 nM full-length or truncated CXCL1 1 for 30 mm at 37 °C. Cells were washed three times with ice cold PBS containing 0.5% BSA, incubated for 45 mm at 4 °C with 2 Ig/mL mouse anti-CXCR3 (R&D Systems, Clone 49801) in PBS, 0.5% BSA. Cells were washed three times then incubated for 45 mm at 4 °C with 1:200 diluted sheep antimouse IgG-FITC (Sigma) followed by 3 final washes and resuspension in 0.5 mL PBS. Staining was analyzed by flow cytometry (Beckman Coulter EPICS XL-MCL) and internalization was calculated from the mean fluorescence intensity values relative to untreated cells. Chemotaxis.  Chemotactic migration of CXCR3-transfectants and activated T  lymphocytes was performed in a 48-well Boyden chamber across a 5 Lm pore size polycarbonate membrane (Neuroprobe). Cells and chemokine were diluted in RPMI containing 20 mM HEPES and 1% BSA. To the upper chamber, 2.5 x added.  cells were  The chamber was incubated for 3 h at 37 °C, 5% C0 , then the upper 2  chamber was aspirated and washed twice with distilled water.  Inhibition  experiments were performed as above except the lower chamber contained 10 nM CXCLI 1 (1-73) in combination with up to 10 1 iM of the truncated CXCLI I forms. Following incubation, the contents of the lower chamber were transferred to a MaxiSorp 96-well plate (Nunc) and frozen at -80 °C for at least 2 hours. Cell content was determined by CyQUANT analysis, a fluorescent nucleic acid-based cellular quantitation  assay,  by  comparison  with  a  standard  curve,  according  to  manufacturers protocol (Molecular Probes). Chemotactic index was calculated as a ratio of cells migrating in response to chemokine compared to the buffer control.  68  Circular dichroism.  CD spectroscopy measurements were made in the far-  ultraviolet range from 190-260 nm using a JASCO J-810 spectropolarimeter. Fulllength and truncated CXCLI I forms, and full-length CXCL8, were resuspended at 10 M in nanopure water and measured in a quartz cuvette of 0.1 cm path length at 25 °C. Data are reported as the average of 4 scans with a I nm band width. When possible, secondary structure composition was predicted with the SOMCD algorithm (55). Glycosaminoglycan binding.  CXCLI I binding to fluorescein-heparin (Molecular  Probes) was determined using fluorescence polarization analysis on the Polarstar Optima 96-well fluorimeter (BMG) essentially as described previously for other binding proteins (56). Incubations contained 0.1 1 iM heparin-FITC and chemokines at concentrations of 0 to 2 1 iM, performed in 100 mM Tris, 150 mM NaCI, pH 7.4 for I h at 37 °C. To further assess the effect of MMP truncations of CXCL1 1 on heparin binding, 0.5 mL of 4 M chemokine in 10 mM potassium phosphate, pH 7.5 was loaded onto I mL Hitrap TM heparin-Sepharose and cation exchange Sepharose columns (GE Healthcare) as described previously (57).  Bound CXCLII and  synthetic analogues of MMP-cleaved CXCLI I were eluted using a linear gradient of 0 to 1.5 M NaCI over 30 mm  at a flow rate of 1.0 mL/min and monitored by in-line  absorbance at 215 nm on an AKTA Purifier (Amersham Pharmacia Biotech). The effect of soluble glycosaminoglycans on CXCLI I cleavage by MMP-8 was assessed in 16 h reactions at 37 °C containing a chemokine to GAG ratio of 1:5 (w:w) and an enzyme to substrate ration of 1:10 (w:w). Heparan sulfate and chondroitin sulfate A, B, and C were obtained from Seikagaku and Healon hyaluronate was from Pharmacia.  69  Results Selective processing of CXCL9 and CXCLIO by MMPs. To screen for potential regulation of CXCR3 ligands by MMPs, recombinant MMP-1, 2, 7, 8, 9, 12, and 14 were incubated in vitro with CXCL9 and CXCL1O at an enzyme-substrate ratio of 1:10 (w:w).  Of these MMP5, MMP-7 and 12 had the  broadest activity, either cleaving or degrading both chemokines (Figure 3.IA, B). In the case of CXCL9, MMP-7 proteolysis generated a single product with a m/z of 10136 Da ([M+Hfl, determined by MALDI-TOF mass spectrometry.  Upon  deconvolution, this corresponds to a truncation of CXCL9 at position 90-91 with the cleavage site 88 93 (Figure 3.1 D). The same C-terminal truncation of CXCL9 VLK-VRK was observed with MMP-12 incubation, but this cleavage was not as efficient as with MMP-7, as some CXCL9 remained intact. CXCL1O was partially processed by MMP-12, resulting in truncated forms of 8150 and 7931 Da ([M+H]), corresponding to CXCLIO (1-73) and CXCL1O (1-71), respectively (Figure 3.IB). The cleavage sites in CXCL1O were 69 74 and SKE-MSK . 76 EMS-KRS 71  Although MMP-7 did not generate stable cleavage products of  CXCLIO, reproducibly reduced chemokine band intensity indicates that degradation occurred due to MMP-7 activity. Interestingly, proteolysis of CXCL9 and CXCLIO was exclusively in the carboxy-terminal region, as determined by MALDI-TOF mass spectrometry (Figure 3.ID). The cleavage products CXCL9 (1-90), CXCLIO (1-71), and CXCL1O (1-73) have been identified previously with MMP-8 and MMP-9 (28) and the latter was found to retain full activity when compared to full-length CXCL1O (45). Hence, the present results expand the repertoire of MMPs that process these CXCR3 chemokines.  However, under our assay conditions, and even with high  enzyme-substrate ratios, we did not detect cleavage of CXCL9 and CXCLI 0 by MMP-8 or MMP-9 as previously reported (28).  70  Figure 3.1. MMP processing of CXCR3 ligands. Recombinant MMP-1, 2, 7, 8, 9, 12, and 14 were incubated with CXCL9 (A), CXCL1O (B), and CXCLII (C) at a 1:10 ratio (w:w) for 16 h at 37 °C and proteolysis was visualized on 15% Tris tricine gels, silver-stained. (D) MALDI-TOF mass spectrometry and deconvolution enabled the assignment of cleavage fragments with mass to charge ratios indicated (mlz) with +1 charge ionization ([M+H]). Chemokine m/z values are corrected for disulfide bond formation (-4) and +1 charge ionization. Lanes with * show substrate degradation as observed by loss of band intensity. Note that CXCLI 1 incubation with MMP-8 resulted in complete conversion represented by slight shift in mass of band.  71  A  MMP  1  2  7  8  9  — —  “S  Cl  14 CXCL9  —.  4’  q>  12  14  [M÷H]  “S  “S  BMMP  12  1  2  7  8  9  -  CXCL1O [M+H]  4’ C  MMP  1  2  7  8  9  12  14 CXCL11  (5  —___d _  D Chemokine CXCL9 (1-103) CXCL9 (1-90) CXCL1O (1-77) CXCL1O (1-73) CXCL1O (1-71) CXCL11 (1-73) CXCL1 1 (5-73) CXCL1 1 (5-63) CXCL11 (1-58) CXCL1 1 (5-58)  [M+H]  m/z [M÷H] Measured Predicted 11721 11722 10136 10137 8617 8618 8150 8150 7931 7931 8303 8304 7779 7781 6524 6525 6451 6451 5928 5928 72  Selective processing of CXCLII by MMPs Of the CXCR3-binding chemokines, CXCLI I was the most susceptible to MMP cleavage, being significantly processed by MMP-7, 8, 9, and 12 (Figure 3.IC). Although MMP-7 appeared to almost entirely degrade CXCL1 1, one peptide ion peak at 6451 Da ([M+H1) was detected by MALDI-TOF MS following cleavage at 61 reprsenting a transient cleavage product, CXCLI 1 (1-58) (Figure PKS-KQA 56 3.ID). MMP-12 also degraded CXCL1I, but three transient truncations were found generating peptides with m/z of 7779, 6451, and 5928 Da ([M÷H]) corresponding to CXCLI 1 (5-73), (1-58), and (5-58) with the N-terminal cleavage at 7 PMF-KRG The 2 . neutrophil-specific rotease MMP-8 specifically cleaved CXCLI I generating stable 7779, 6524, and 5ë28 Da ([M÷H]) truncations corresponding to CXCLI 1 (5-73), (563), and (5-58), respectively.  MMP-9 processed the chemokine, generating two  cleavage products of 6451 and 5928 Da ([M+H]), which represent CXCL1 1 (1-58) and (5-58). Hence, it is evident that several MMPs cleave CXCLI I at residues 4-5 and 58-59, while MMP-8 is unique in its cleavage of the 63-64 bond at 61 66 ARL-11K (Figure 3.3). Processing of mCXCL 11 Murine CXCLII, which is 68% identical to the human form but lacks a proline at position 2, was alsO examined as a MMP substrate. MMP-1, 7, 8, and 12 cleaved the chemokine (Figure 3.2A). Several products were detected by MALDI-TOF mass spectrometry and upon deconvolution were determined to represent N-terminal cleavage at 6 FLM-FKQ and C-terminal cleavages at 61 1 , 66 and 71 ARL-1MQ KNF76 (Figure 3.2B). MMP-8 cleaved mCXCLI I at all three sites, while MMP-7 and LRR MMP-1 cleaved exclusively at the N-terminus and C-terminus, respectively (Figure 3.3).  Notably, no stable truncation products were detected with MMP-12, as  substrate degradation was evident.  73  A MMP  -  2  1  —  \<S c\  7  — *..  8 ;a____  9  12  14  —  mCXCL11  *  c  -  qbo  cz,  [M÷H]  ‘\  B Chemokine mCXCL11 (1-79) mCXCL1 1 (4-79) mCXCL1 1 (1-73) mCXCL1 1 (4-73) mCXCL1 1 (1-63) mCXCL1 1 (4-63)  m/z [M+H] Measured Predicted 9112 9110 8719 8718 8312 8311 7921 7920 7108 7108 6717 6716  Figure 3.2. MMP processing of murine CXCLI 1. (A) Recombinant MMP-1, 2, 7, 8,9, 12, and 14 were incubated with murine CXCL11 for 16 h ata 1:10 ratio (w:w) at 37 °C, visualized on 15% Tris-tricine gels, silver-stained. (B) Cleavage products were assigned by MALDI-TOF mass spectrometry by comparing measured and predicted mass to charge ratios (m/z) with +1 charge ionization ([M÷H]). Lane with * shows substrate degradation as observed by loss of band intensity.  74  CXCL9 1  10  80  100  90  TPVVRKGRCS...KQKNGKKHQKKKVLKtVRKSQRSRQKKTT MMP-7, 12  CXCL1O 1 70 10 60 VPLSRTVRCT...CLNPESKAIKNLLKAVSKEtMStKRSP  MMP-12  CXCL11 MMP-8 ,9, 12 I 10 60 70 FPMF*KRGRCL...GQRCLNPKStKQARLtIIKKVERKNF MMP-7, 8, 9, 12  MMP-8  mCXCL1 1 I  MMP-7, 8 10  60  70  FLM+FKQGRCL...DPRSKQARLtIMQAIEKKNFtLRRQNM MMP-1,8  MMP-1,8  Figure 3.3. MMP cleavage sites in CXCL9, CXCLIO, CXCLII, and murine CXCLII. Carboxy-terminal processing of CXCL9 by MMP-7 and MMP-12 at 91 yields CXCL9 (1-90). CXCLIO cleavage by MMP-12 generates C -Va1 90 Lys terminal truncations CXCL1O (1-71) and (1-73) at G1u 72 and Ser -Met 71 . 74 -Lys 73 MMP-8, 9, and 12 cleave CXCLI I at two sites to generate CXCLI 1 (5-73) and (558) whereas MMP-7 only cleaves at Ser 59 to give CXCLI 1 (1-58). MMP-8 -Lys 58 alone cleaves CXCL1 I at Leu . Murine CXCL1 1 is cleaved by MMP-7 and 64 -lle 63 8 at 4 -Phe and by MMP-1 and 8 at Leu 3 Met 64 and Phe -1le 63 . 74 -Leu 73  75  Kinetic analysis of CXCL 11 processing We first screened for cleavage at high enzyme-substrate ratios. Since cleavage alone does not indicate a biologically relevant substrate (7), we assessed the enzyme kinetics of MMP cleavage of CXCLI I (Figure 3.4). Based on densitometry analysis, where ‘(cat/KM = (In2)I(E)(ti, ), the kt/KM of CXCL1 I processing was 2 calculated to be in the range from 96-190 M s for MMP-7, 8, 9, and 12 (Figure 1 3.4E).  Notably,  MMP-7 partially cleaved CXCLI I  at the lowest enzyme  concentration tested, 16 nM, which corresponds to an enzyme-substrate ratio of 1:800 (mol:mol) (Figure 3.4A). However, degradation is apparent at concentrations of 62.5 nM MMP-7 and greater, as was also observed for MMP-12, at concentrations of 1000 nM (Figure 3.4D). In control incubations MMP activity against fluorogenic substrate was maintained at 16 h except for MMP-12, which exhibited loss of activity due to autodegradation. Hence, MMP-12 was analyzed at higher concentrations for only 2 hours due to instability observed in longer incubations.  Both MMP-8 and  MMP-9, which are expressed and secreted by neutrophils, efficiently and precisely cleaved CXCL1 1 at enzyme concentrations of 125 nM and greater, resulting in both amino and carboxy-terminal truncations (Figure 3.4B, C; See also Figure 3.8A), so indicating the potential for Thi  lymphocyte regulation by neutrophil-derived  proteases. Interaction of the MMP-8 hemopexin C domain with CXCLII It is well established that MMPs contain exosites, which are substrate-binding sites outside of the catalytic domain important in dictating substrate specificity and efficiency in cleavage (30, 58). In particular, the hemopexin C domains of several MMPs interact with chemokines to promote proteolysis (6, 25, 58). In the case of MMP-8 processing of CXCLII, the addition of 10-fold molar excess MMP-8 hemopexin C domain inhibited the formation of truncation products indicating that CXCL1 I cleavage is hemopexin C domain-dependent (Figure 3.5A).  As positive  controls, the hydroxamate MMP inhibitor Batimastat (BB94) and the general metalloprotease inhibitor EDTA also blocked proteolysis. Interestingly MMP-7 and  -  12 can process several chemokines yet MMP-7 does not have a hemopexin C 76  A MMP-7 (nM)  0  16  32  62.5 125  250  500  CXCL11—— (1-73)  4  CXCL11 (1-62) CXCL11 (1-58)  B MMP-8 (nM)  16  0  32  62.5 125 250  CXCL1 1-+ (1-73)  500  4  CXCL1 1 (5-73) CXCL1 1 (5-63) CXCL1 1 (5-58)  4  CXCL1 1 (5-73)  4  CXCL1 1 (5-58)  4  CXCL1 1 (5-73)  I  CXCL1 1 (5-58)  -  —-  —  C MMP-9 (nM)  0  16  32  CXCL1 (1-73)  62.5 125  250  500  .  D MMP-12(nM) 0  125  250 500 1000 20004000  CXCL1 1-+ (1-73)  E  Enzyme MMP-7 MMP-8 MMP-9 MMP-12  kca/KM (Ms) 2 1.9x10  1.3x10 2 9.6x 101 9.6x 101  Figure 3.4. Titration of MMP processing of CXCLI 1. Recombinant MMP-7 (A), MMP-8 (B), MMP-9 (C) were incubated with 10 ji.M CXCLI I at concentrations ranging from 16 500 nM for 16 h at 37 °C. (D) Due to decreased enzyme stability, 125-4000 nM MMP-12 was incubated with 10 1 tM CXCLII for 2 h at 37 °C. Cleavage products were visualized on 15% Tris-tricine gels and were assigned by MALDI-TOF mass spectrometry. (E) Specificity constant (kcat/KM) values were determined by densitometry from gels. -  77  domain and MMP-12 autolytically removes this during synthesis.  Interestingly the  hemopexin C domain of MMP-12 has significant antibacterial actions (59), which also likely contributes to the beneficial activities of MMP-12 in innate immunity. To investigate the molecular basis for the selective interaction of MMP-8 with CXCL1 1 compared to CXCL9 and CXCL1O, where there was no cleavage, the affinity of the CXCR3 ligands for the MMP-8 hemopexin C domain was assessed. ELISA experiments demonstrated that CXCLI I has a significantly higher affinity for MMP-8 hemopexin C domain compared with CXCL9 and CXCL1O, which only bound at or below background controls (Figure 3.5B). The CXCL1 1 MMP-truncated forms CXCL1 1 (5-73) and CXCLI 1 (5-58) bound with similar affinity to the MMP-8 hemopexin C domain as full-length CXCLI I (data not shown), suggesting that the interaction occurs with the central region of CXCL1 1, distant to the cleavage sites. Calcium mobilization of CXCR3-transfected 8300-1 9 cells Synthetic analogues of MMP-truncated CXCLI I were tested for activity with human CXCR3-transfected B300-19 cells. At a concentration of 800 nM chemokine, there was a clear loss of activity in the MMP-truncated forms of CXCL1 1, where calcium mobilization is measured as a burst in relative fluorescence (Figure 3.6A).  Full-  length CXCLII (1-73) caused a dose-dependent increase in intracellular calcium mobilization with an EC 50 of I nM, consistent with previous reports (60) (Figure 3.6B). In contrast, the truncated forms of CXCLI I had no detectable agonist activity at any concentration tested.  Since MMP-cleaved CCL2, 7, 8, and 13 (9) and  CL1 (10) are receptor antagonists, we assessed the CXCL1 I cleaved analogues 3 CX as receptor antagonists by calcium mobilization.  A two-minute pre-incubation,  shown previously to be sufficient time for receptor interaction (61, 62), with CXCLI I (5-73) resulted in significant inhibition of full-length CXCLI I with a moderate 50 IC of 500 nM (Figure 3.6C), as observed previously (61). However, the CXCLII (5-58) truncation had reduced inhibition of calcium mobilization, suggesting that the C terminus is involved in binding the CXCR3 receptor.  Neither CXCLI 1 (5-73) nor  CXCL1 1 (5-58) caused CXCR3 receptor internalization as measured by flow  78  A Inhibitor MMP-8  -  -  -  +  4  Q,  +  +  +  14$ — i—CXCL1 1 (1-73)  CXCL1 1 (5-73) 6  B 1 CXCL11 C LC) —  0. CXCL1O BLAN K  0,  C.) C  CU  CXCL9  0  Cl)  0. 0  500  1000  1500  Hemopexin L-C Domain (M x 10w)  Figure 3.5. Exosite interactions with CXCLI 1. (A) MMP-8 cleavage of CXCLI I (16 h at 37 °C at 1:10 enzyme to substrate ratio (w:w)) was inhibited in the presence of EDTA, Batimastat (BB94), and 10-fold molar excess of the MMP-8 linker and hemopexin C domain (LCD). (B) ELISA analysis of MMP-8 hemopexin linker-C (L-C) domain binding to the CXCR3 chemokines, immobilized in 96-well plate format. The order of hemopexin C domain binding affinity was CXCL1 1 >CXCLI 0>CXCL9.  79  A  B 600  200 —CXCL11 (1-73)  • CXCLII (1-73)  I  Time (s)  Chemokine (log M)  D  C 125  100  CXCLI1 (1-73)  1  1100  Chemokine (log M)  Chemokine (M x 10)  Figure 3.6. MMP-mediated cleavages of CXCLII result in loss of agonism and the generation of antagonist activity in calcium mobilization assays. (A) Representative trace of calcium flux, represented by a burst in relative fluorescence units (RFU) following addition of 800 nM CXCL1I in CXCR3transfected B300-1 9 cells loaded with Fluo-4 calcium indicator reagent. Arrow indicates time of CXCLI I addition. (B) Dose response of calcium mobilization in CXCR3-transfected B300-1 9 cells in response to CXCLI 1 (1-73), (5-73), and (558). Calcium concentrations were calculated from relative fluorescence based on calibration with ionomycin and MnCI . (C) Antagonism of calcium mobilization in 2 CXCR3-transfected B300-19 cells by CXCL1 1 (5-73) and (5-58). Cells were preincubated with antagonists for two minutes prior to the addition 3 nM fulllength CXCLI 1. Activity values are reported as a percentage relative to that of uninhibited control cells. (D) CXCR3 internalization of CXCR3-transfected B30019 cells following 30 minute incubation of 10 and 100 nM CXCLII at 37 °C. Detection of cell surface levels of CXCR3 was performed by flow cytometry with anti-CXCR3 monoclonal antibody and mean fluorescence intensities were compared to untreated controls and reported as percentage CXCR3 levels. 80  cytometry (Figure 3.6D), confirming the loss of agonistic activity, whereas 100 nM CXCLI 1 (1-73) caused approximately 60% internalization, consistent with previous reports (63).  Chemo tactic migration of CXCR3-transfected B300-19 cells and human T lymphocytes To confirm the calcium mobilization results, CXCR3-transfected cells and activated T lymphocytes were evaluated for chemotaxis toward full-length and truncated forms of CXCLII.  In both cell types CXCLII  (1-73) promoted dose-dependent  chemotactic migration as expected (Figure 3.7A, C). However, both CXCL1 1 (5-73) and CXCLII (5-58) had minimal activity, even at concentrations as high as 100 nM. Again, the N-terminal truncated CXCLI 1 (5-73) had antagonistic activity with IC 50 values of 100 nM and 1000 nM for CXCR3-transfected B300-19 cells and T lymphocytes, respectively (Figure 3.7B, D).  Also, the CXCL1 1 (5-58) truncated  chemokine was only a very weak antagonist, with an 1C 50 of greater that 10 !IM in both cell types. However, it is important to note that this represents a concentration that is biologically unattainable to achieve antagonism. Structural analysis of truncated CXCLII MMP processing of CXCLI 1 results in the removal of 4 N-terminal amino acids and either 10 or 15 C-terminal amino acids (Figure 3.8A). The carboxy terminal a-helix removed by MMP cleavage is markedly cationic, supporting a role in GAG interactions, as shown in this structural representation based on the published NMR structure (Protein Databank ID I RJT) (64). Notably, none of the MMP truncations affect the cysteine residues involved in disulfide formation, hence tertiary protein structure is not expected to be adversely affected. Circular dichroism comparing full length and truncated CXCLI I suggests that the structural integrity remains intact (Figure 3.8B). Interestingly, the overall CD profile of CXCLI I resembles that of CC chemokines with a high percentage of random structure, as indicated by strong negativity at 200 nm (65). CXCL8 was analyzed for comparison with another CXC chemokine and this spectrum of a highly ordered protein correlates with published 81  A  B 4 • CXCLII (1-73)  ioo  _____________ Chemokine (log M)  C  :: I  Chemokine (log M)  D  !1  ..m.=:==f::  Chemokine (log M)  Chemokine (log M)  Figure 3.7. Chemotactic migration is decreased and antagonized in response to CXCLII (5-73) and (5-58). Chemotaxis of CXCR3-transfected B300-19 cells (A) and isolated T lymphocytes (C) is abolished in response to CXCLI 1 (5-73) and (5-58). Chemotaxis was measured across a 5 tm pore size for 3 hours in response to 1-100 nM chemokine. Migrated cells were quantitated by CyQUANT assay and a chemotactix index was calculated, defined as the ratio of cells migrating in response to stimulus compared to the buffer control. Migration is inhibited moderately by CXCLI 1 (5-73) in CXCR3-transfected cells (B) and T cells (D) whereas CXCL1 1 (5-58) is a weak inhibitor. Chemotaxis was performed for 3 hours in response to 10 nM to 10 tM twncated CXCLI1 co incubated with 10 nM full length CXCLI1. Percentage of maximum was calculated relative to uninhibited samples. 82  A g Phe - Lys 4 5  - Lys 58 Ser 59  r  64 -lle 63 Leu  B a—. 10 a)  0  0  E  D10. c.J  E ci 0) G)  SI  —  II  —CXCL11 (1-73) CXCL1 1 (5-73) CXCL11 (5-58) CXCL8 (1-77)  -30  —  -40  ———  ‘  -50 180  200  220  240  260  Wavelength (nm)  Figure 3.8. Structural analysis of full length and truncated CXCLII. (A) Structural representation of full--length CXCLI I demonstrating MMP cleavage sites (arrows) at position 4-5 in the N-terminus and positions 58-59 and 63-64 in the C-terminus. The blue coloring highlights the C-terminal cationic GAG-binding region removed by MMP proteolysis. The amino terminus (NH , 1) and caroxy 2 terminus (COOH, 73) are shown. I acknowledge Dr. Alain Doucet for his contribution of this figure. (B) Circular dichroism of CXCLI 1 (1-73), (5-73), (5-58) and CXCL8 (1-77) in the far-ultraviolet wavelengths ranging from 190-260 nm. Strong negativity at 200 nm illustrates a high degree of disorder for all three CXCL1 1 peptides. Ellipticity at 220 nm represents a-helical content and this signal is lost upon C-terminal truncation.  83  data (65). The higher degree of disorder in CXCLI 1, in comparison with other CXC chemokines, has been reported previously by NMR analysis (64).  Here the CD  spectra for CXCL11 (1-73) and (5-73) are super-imposable and SOMCD analysis predicts 17.9-19.4% cL-helix, 27.3-29.9% n-sheets, 11.8-15.6% turns, and 39-39.3% random structure (55).  In contrast, the CXCLI 1 (5-58) spectrum reveals reduced  ellipticity at 220 nm, suggesting a loss in a-helical structure as expected, but otherwise is comparable to the spectrum for full length CXCL1 1. The potential role of MMP  disruption  of the  haptotactic CXCLI I  proteoglycan  gradient on  glycosaminoglycans is depicted (Figure 3.10). Heparin binding affinity of CXCLII The abundance of positively charged amino acids in the carboxy-terminal a-helix of CXCLI 1 (Figure 3.8A) prompted the prediction that C-terminal truncation would result in reduced glycosaminoglycan binding. In a heparin-binding assay KDS were measured by fluorescence polarization spectroscopy. CXCLI 1 (1-73) and CXCLI I (5-73) bound to heparin with similar affinity, with binding constants ranging from 0.4 to 0.6 1 tM (Figure 3.9A). In contrast, CXCLI 1 (5-58) had weakened heparin binding, and as important, showed significantly reduced levels of bound chemokine upon saturable  heparin  binding.  Together this  indicates  loss  of  a  C-terminal  glycosaminoglycan-binding site that reduced the overall affinity and amounts of chemokine binding. Consistent with this, CXCLI 1 (5-58) was also shown to have impaired binding to chondroitin sulfate A (data not shown). Heparin-Sepharose chromatography was used to refine the characterization of the heparin binding properties of the cleaved forms of CXCLI I in relation to the full length chemokine. Here again the C-terminal truncation had significantly reduced affinity (Figure 3.9B), confirming the assessment by fluorescence polarization. CXCLII (1-73) eluted at 0.74 M NaCI whereas CXCLI1 (5-58) eluted at 0.52 M NaCI. Surprisingly, the N-terminal truncation caused enhanced affinity as CXCLI 1 (5-73) eluted at 0.81 M NaCl, significantly higher than the full-length counterpart.  84  Figure 39. Heparin binding is altered upon CXCLII cleavage. (A) Fluorescence polarization with heparin-FITC reveals a loss in affinity for CXCL1 1 (5-58). Chemokines were incubated for 1 hour with 0.1 1 M heparin-FITC at 37 °C prior to analysis. (B) Elution profile of CXCLI I and cleaved analogues from heparin-Sepharose column (top) and strong cation exchange (SCX) Sepharose column (bottom). Elution was with a gradient reaching 1.5 M NaCI and absorbance was measured at 215 nm. (C) MMP-8 processing of CXCLI1 is unaffected in the presence of 5-fold excess of heparan sulfate (HS), chrondroitin sulfate A (CSA), chondroitin sulfate B (CSB), chondroitin sulfate C (CSC), and hyaluronic acid (HA). The enzyme to substrate ratio was 1:10 (w:w) and the reaction was 16 h at 37°C and products were visualized on a 15% Tris-tricine gel, silver stained. Note that MALDI-TOF analysis confirmed the presence of CXCLI I (5-73), (5-63) and (5-58) in all MMP-8 treated samples.  85  A 2  •CXCLII (1-73) • CXCLI 1 (5-73) o CXCL11 (5-58)  10  15  2O  Concentration (M x 106)  B  8O  It) 60  -g  1.5 —CXCL11 (1-73) —CXCLII (5-73)  ——  80  —  1.5  j: i: 1 Volume (mL)  C GAG MMP-8CXCL11 (1 73)  -  +  +  —  +  —  +  +  +  —  CXCL1 1 5-58  86  To assess the effect of specificity of interaction versus charge effects alone on the chemokine binding properties, an equivalent experiment was performed with a strong cation exchange column. The elutions of CXCL1 1 (1-73), (5-73), and (5-58) occurred at 0.74, 0.77, and 0.46 M NaCI, respectively.  Loss of the cationic C-  terminal a-helix in CXCLI 1 (5-58) yielded a similar reduction in binding to both the heparin and strong cation exchange columns, in comparison with CXCLII (1-73), suggesting  that  the  removal  of  charge  is  responsible  for  decreased  glycosaminoglycan binding. However, the N-terminal truncation in CXCLI 1 (5-73) appears to enhance specific heparin interactions because the elution shift relative to full-length CXCLI 1 is greater on the heparin column compared to the strong cation exchange column. The glycosaminoglycan interaction with the cationic residues of the C-terminal peptide 73 KQARLIIKKVERKNF forming an a-helix in the 59 , chemokine, might block MMP cleavage. The addition of excess heparan sulfate, chondroitin sulfate, and hyaluronic acid did not affect MMP-8 processing of CXCL1 1 at position 58-59, confirmed by MALDI-TOF MS analysis (Figure 3.9C).  Hence,  these data demonstrate that CXCLI I can be processed by MMPs in the presence of glycosaminoglycans and that the resulting cleavages alter binding affinity.  Discussion Presented herein is the first instance of chemokine cleavage directly affecting glycosaminoglycan-binding  properties.  Previously,  the  murine  chemokine  mCXCL1/KC was shown to be mobilized by MMP-7 cleavage of the core protein of syndecan-1, leading to altered neutrophil chemotaxis, but the chemokine itself was not directly processed (13). Hence, the cleavage of the cationic C-terminal peptide of CXCL1 I and resultant loss of a glycosaminoglycan-binding site reported in our biochemical study represents a novel mechanism of chemokine regulation. We suggest that the neutrophil MMPs 8 and 9, and macrophage MMP5 7, 9 and 12 mobilize CXCLI 1 from proteoglycans with disruption of the haptotactic chemokine gradient predicted (Figure 3.10). This would serve to enhance the direct effects on chemokine activity resulting from proteolysis of CXCLI I by these leukocytic MMPs.  87  ESTABLISHMENT OF A HAPTOTACTIC CHEMOKINE GRADIENT  CXCLII  CXCR3. i  I  Thi Lymphocyte  MMP CONTROL OF HAPTOTACTIC CHEMOKINE GRADIENT CXCLI 1 (5-73) or (5-58).  .  C Macrophage Neutrophi I . Loss of GAG binding  Cell dispersion  Figure 3.10. Model representing the potential role of leukocytic MMPs 7, 8, 9 and 12 in disrupting the haptotactic gradient of CXCLII. The establishment of a full-length CXCLI I gradient through glycosaminoglycan interactions is depicted in the top panel. This gradient is modified upon CXCLI I processing by neutrophil and macrophage-derived MMPs (bottom panel), resulting in the formation of CXCLI 1 (5-73) and (5-58) and leading to potential dispersion of Thi lymphocytes.  88  Amino-terminal processing at Phe 5 4 Lys resulted in loss of agonist activity in calcium flux and chemotactic migration assays with CXCR3-transfectants and isolated T lymphocytes.  However, CXCLI 1 (5-73) retained receptor affinity, as it was a  moderate antagonist against full-length CXCLI 1.  Interestingly, the C-terminal  truncation at Ser 59 caused a reduction in receptor binding as indicated by -Lys 58 decreased antagonistic properties so revealing a new CXCR3 binding site on the chemokine. Together, MMP cleavage potentially dampens T cell migration by loss of agonist activity, generation of a CXCR3 antagonist, and disruption of the chemokine gradient (Figure 3.10). The in vivo analysis of these effects are now in progress. Chemokine activity is modulated at several levels including regulated expression and secretion, proteolysis, and the formation of haptotactic gradients in tissue. High affinity interactions with the glycosaminoglycan side-chains of proteoglycans are critical for in vivo activity of many chemokines as an immobilization mechanism for cell-surface retention and possibly also for presentation to circulating leukocytes (2). MMP-mediated proteolysis has been shown previously to indirectly influence chemokine gradients, but by a different mechanism.  In the case of MMP-7,  knockout mice demonstrate altered neutrophil infiltration in lung inflammation where neutrophils remain in the interstitium without advancing to the alveolar space (13). This phenotype was attributed to MMP-7--dependent shedding of the syndecan-1 ectodomain complexed with the murine ELR chemokine mCXCLI/KC, a process required to direct neutrophils to the site of injury. However, proteolysis of KC did not occur. This mechanism contrasts the potential mobilization of CXCLI 1, where we demonstrate direct MMP cleavage of the chemokine and loss of a heparin-binding site that also has the potential to disrupt the chemokine gradient. The basic residues lysine and arginine direct glycosaminoglycan affinity through electrostatic interactions; in CCL5 a BBXB motif has been identified (66) whereas CXCL8 contains several key residues at the C-terminus (67). Notably, the heparin binding cationic cradle is rarely a simple consensus sequence, but rather a more complex combination of remote positively charged residues that are clustered in the  89  folded conformation, as illustrated with fibronectin module 111-13 (68).  We  hypothesized that the glycosaminoglycan binding-site of CXCLI 1 is located in the Cterminus due to an abundance of basic amino acids and indeed the heparin binding interaction was strongly reduced in CXCLII (5-58).  However, residual binding  suggests additional sites of heparin interaction independent of the C-terminus. Surprisingly, removal of the N-terminal tetrapeptide FPMF in CXCLI I increased heparin-Sepharose binding, perhaps due to the removal of steric hinderance providing stronger interactions with key basic residues.  The enhanced affinity of  CXCLI 1 (5-73) is not completely charge-dependent as the equivalent comparison on a strong cation exchange column showed only a minimal increase. Despite the location of a glycosaminoglycan-binding site near the site of MMP cleavage between positions 58-59, MMP cleavage of CXCLI I was unaltered when interacting with soluble heparan sulfate, chondroitin sulfate, and hyaluronic acid. Hence, these findings show for the first time for any protease that direct proteolysis of a chemokine can alter its glycosaminoglycan-binding properties. The resulting enhanced and reduced affinities, depending on the cleavage site, reveal biphasic regulation of chemokine binding to glycosaminoglycans and hence potential complex modulation of in vivo localization and chemotactic activity that is currently under investigation. Unlike many other proteases, the cleavage site specificity of MMPs remains largely unpredictable, likely due to less restricted substrate amino acid binding pockets (in particular S3-S3’) and the complex role of exosite binding in substrate recognition (58). Peptide substrate libraries have demonstrated an MMP preference for proline at P3 (69), and notably the majority of chemokines contain a proline at residue 2 in the N-terminus. Therefore this proline appears to dictate MMP proteolysis of many chemokines at the characteristic residues 4-5 in the N-terminus. However, examples at other N-terminal cleavage sites suggest that specificity is not so simple. For instance CXCL5, which has a proline at position 3, is cleaved by MMPs at Va1 8 7 Leu and ° 1 9 Arg Glu (25, 27). Interestingly, CXCL9, CXCLIO, and CXCLII share a proline  90  at position two and are approximately 40% identical (60), yet only CXCLI I is susceptible to N-terminal processing by MMPs, illustrating that additional elements are important, such as exosite interactions. Indeed, only CXCL1 I bound the hemopexin C domain of MMP-8, but neither CXCL9 or CXCL1O did and these were not cleaved. Furthermore, murine CXCLI 1, which has high sequence identity to the human homolog, contains a leucine in the place of proline at position 2 and is processed efficiently at position 4 -Phe rather than 5 3 Met -Lys by MMP-7 and -8. 4 Phe Nonetheless, the predominant C-terminal cleavage of CXCL1 I at residues 58-59 also contains a proline at P3. Hence, this data suggests that Pro 2 is not essential for CXCL1 1 processing, but rather that it directs cleavage to position 4-5 in the Nterminus. Interestingly, all three cleavage sites on CXCL1 1 lie on the same face of the protein in a line indicating potential processivity of cleavage upon chemokine tethering to the hemopexin C domain. Of the MMPs evaluated, MMP-1, -2, and -14 had no activity on any of the human CXCR3 ligands revealing specificity of the main leukocylic MMPs for chemokine cleavage and hence an immune cell-specific feedback mechanism regulating chemokine activity. In contrast to the specific and selective cleavages by MMP-8 and MMP-9, processing or degradation of the three chemokines by MMP-7 and MMP-12 represents broad specificity. Interestingly, MMP-7 and MMP-12 share structural similarities in that MMP-7 does not encode a hemopexin C domain (70) and MMP-12 loses the domain upon activation (71).  In the case of the CXCR3  ligands, it is probable that MMP-7 and MMP-12 have weak specificity, but high activity, resulting in substrate degradation in contrast to the precise processing by other MMPs. Therefore, it seems likely that exosite interactions of the hemopexin C domain contribute greatly to selective cleavage of substrates. Amino-terminal twncation of chemokines has been reported to result in conversion of receptor agonists to antagonists (6, 9). CXCLI I truncations have been previously characterized and CXCL1 1 (4-73) was a potent receptor antagonist whereas CXCLI 1 (5-73) had moderate inhibitory properties (61). In addition, CD26-mediated  91  cleavage of CXCLI I generates CXCL1 1 (3-73), another receptor antagonist (47). Antagonist studies with CXCLI 1 (5-73) are consistent with previous findings, where moderate IC s of 100-1000 nM were observed. Surprisingly, the MMP-mediated C50 terminal truncation of CXCL1 1 at residues 58-59 significantly decreased the inhibitory properties of this molecule with lC s greater than 10 tiM, concentrations 50 that are unlikely to be reached in Wvo.  The C-terminus of chemokines is not  traditionally associated with receptor binding properties.  Rather, the N-loop  following the first two cysteines is the major receptor binding site that confers receptor specificity. Electrostatic interactions in the 30s loop, the region connected the first two beta-strands, are also thought to contribute to receptor affinity (72). Therefore, we have demonstrated an unexpected role for the C-terminus of CXCLI I in receptor binding. Amino-terminal CXCLI1 processing has been described previously for CD13 (14) and CD26 (46, 47), however the physiological relevance of these interactions remains unclear, particularly given the clinical use of CD26 inhibitors for diabetes (73).  In contrast, the role of MMPs in chemokine processing in vivo is well  established with evidence from genetic mouse models and neo-epitope antibodies detecting the MMP-processed forms in human biological samples (6, 25, 29). The murine homolog of CXCLI I lacks a proline at position 2 and hence MMP processing of mCXCLI I is altered, restricting the use of genetic models to further explore this mechanism in animal models. Interestingly, MMPs are the only enzymes known to cleave the C-terminus of CXCLI 1, indicating potential synergism between multiple proteases in the precise regulation of this potent T-cell chemoattractant. Specific and limited proteolysis is emerging as a widespread mechanism of rapid and efficient modulation of both cytokine and chemokine activity [see reviews (20, 21).  To date, MMPs regulate chemokine function by activation, inactivation,  generation of antagonists, shedding from the cell membrane, and cleavage of binding proteins. Our present data demonstrates a new mechanism whereby direct cleavage of a chemokine leads to loss of proteoglycan binding predicted to result in  92  dispersion of the chemokine gradient. Here we have biochemically characterized MMP processing of the CXCR3 ligands, uncovering cleavages that alter receptor activation and binding as well as glycosaminoglycan interactions. These cleavage events have the potential to modulate Thi cell migration and hence disorders such as multiple sclerosis and rheumatoid arthritis, which are now under study in our laboratory. Neutrophils are implicated in mediating tissue damage in the context of chronic inflammation.  However, the abundance of neutrophil-derived MMP-8 and  MMP-9 may enable regulation of the adaptive immune response through the finetuning of Thi chemoattraction.  Furthermore, macrophage proteases MMP-7 and  MMP-12 have the potential to dampen the Thi response through chemokine degradation. The ensuing challenge is to assess these mechanisms in the context of homeostatic and pathological states to decipher the physiological consequences of altering these chemotactic signaling pathways in T cell chemotaxis and immune regulation.  93  References 1.  Moser, B., M. Wolf, A. Walz, and P. Loetscher. 2004. Chemokines: multiple levels of leukocyte migration control. Trends Immunol 25:75-84.  2.  Handel, T. M., Z. Johnson, S. E. Crown, E. K. Lau, and A. E. Proudfoot. 2005. Regulation of protein function by glycosaminoglycans--as exemplified by chemokines. Annu Rev Biochem 74:385-410.  3.  Hamilton, T. A., M. Novotny, S. Datta, P. Mandal, J. Hartupee, J. Tebo, and X. Li. 2007. Chemokine and chemoatiractant receptor expression: post-transcriptional regulation. J Leukoc Biol 82:213-219.  4.  Neel, N. F., E. Schutyser, J. Sai, G. H. Fan, and A. Richmond. 2005. Chemokine receptor internalization and intracellular trafficking. Cytokine Growth Factor Rev 16:637-658.  5.  Locati, M., Y. M. Torre, E. Galliera, R. Bonecchi, H. Bodduluri, G. Vago, A. Vecchi, and A. Mantovani. 2005. Silent chemoattractant receptors: D6 as a decoy and scavenger receptor for inflammatory CC chemokines. Cytokine Growth Factor Rev 16:679-686.  6.  McQuibban, G. A., J. H. Gong, E. M. Tam, C. A. McCulloch, I. Clark-Lewis, and C. M. Overall. 2000. Inflammation dampened by gelatinase A cleavage of monocyte chemoattractant protein-3. Science 289:1202-1206.  7.  Overall, C. M., and 0. Kleifeld. 2006. Tumour microenvironment opinion: validating matrix metalloproteinases as drug targets and anti-targets for cancer therapy. Nat Rev Cancer 6:227-239.  8.  McQuibban, G. A., G. S. Butler, J. H. Gong, L. Bendall, C. Power, I. Clark-Lewis, and C. M. Overall. 2001. Matrix metalloproteinase activity inactivates the CXC chemokine stromal cell-derived factor-i. J Biol Chem 276:43503-43508.  9.  McQuibban, G. A., J. H. Gong, J. P. Wong, J. L. Wallace, I. Clark-Lewis, and C. M. Overall. 2002. Matrix metalloproteinase processing of monocyte chemoattractant proteins generates CC chemokine receptor antagonists with anti-inflammatory properties in vivo. Blood 100:1160-1167.  10.  Dean, R. A., and C. M. Overall. 2007. Proteomics discovery of metalloproteinase substrates in the cellular context by iTRAQ labeling reveals a diverse MMP-2 substrate degradome. Mol Cell Proteomics 6:611-623.  ii.  Schulte, A., B. Schulz, M. G. Andrzejewski, C. Hundhausen, S. Mletzko, J. Achilles, K. Reiss, K. Paliga, C. Weber, S. R. John, and A. Ludwig. 2007. Sequential processing of the transmembrane chemokines CX3CL1 and CXCL16 by alpha- and gamma secretases. Biochem Biophys Res Commun 358:233-240.  12.  Vergote, D., G. S. Butler, M. Ooms, J. H. Cox, C. Silva, M. D. Hollenberg, J. H. Jhamandas, C. M. Overall, and C. Power. 2006. Proteolytic processing of SDF-lalpha reveals a change in receptor specificity mediating HIV-associated neurodegeneration. Proc NatlAcad Sd USA 103:19182-19187.  13.  Li, Q., P. W. Park, C. L. Wilson, and W. C. Parks. 2002. Matrilysin shedding of syndecan 1 regulates chemokine mobilization and transepithelial efflux of neutrophils in acute lung injury. Cell 111:635-646.  14.  Proost, P., A. Mortier, T. Loos, J. Vandercappellen, M. Gouwy, I. Ronsse, E. Schutyser, W. Put, M. Parmentier, S. Struyf, and J. Van Damme. 2007. Proteolytic processing of CXCL11 by CD13/aminopeptidase N impairs CXCR3 and CXCR7 binding and signaling and reduces lymphocyte and endothelial cell migration. Blood 110:37-44.  -  94  15.  Van Damme, J., S. Struyf, A. Wuyts, E. Van Coillie, P. Menten, D. Schols, S. Sozzani, I. De Meester, and P. Proost. 1999. The role of CD26/DPP IV in chemokine processing. Chem Immunol 72:42-56.  16.  Ajami, K., M. R. Pitman, C. H. Wilson, J. Park, R. I. Menz, A. E. Starr, J. H. Cox, C. A. Abbott, C. M. Overall, and M. D. Gorrell. 2008. Stromal cell-derived factors laipha and ibeta, inflammatory protein-b and interferon-inducible T cell chemo-attractant are novel substrates of dipeptidyl peptidase 8. FEBS Left 582:819-825.  17.  Petit, I., M. Szyper-Kravitz, A. Nagler, M. Lahav, A. Peled, L. Habler, T. Ponomaryov, R. S. Taichman, F. Arenzana-Seisdedos, N. Fujil, J. Sandbank, D. Zipori, and T. Lapidot. 2002. G-CSF induces stem cell mobilization by decreasing bone marrow SDF-1 and upregulating CXCR4. Nat Immunol 3:687-694.  18.  Lim, J. K., W. Lu, 0. Hartley, and A. L. DeVico. 2006. N-terminal proteolytic processing by cathepsin G converts RANTES/CCL5 and related analogs into a truncated 4-68 variant. J Leukoc Biol 80:1395-1404.  19.  Hasan, L., L. Mazzucchelli, M. Liebi, M. Lis, R. E. Hunger, A. Tester, C. M. Overall, and M. Wolf. 2006. Function of liver activation-regulated chemokine/CC chemokine ligand 20 is differently affected by cathepsin B and cathepsin D processing. J Immunol 176:65126522.  20.  Parks, W. C, C. L. Wilson, and Y. S. Lopez-Boado. 2004. Matrix metalloproteinases as modulators of inflammation and innate immunity. Nat Rev Immunol 4:617-629.  21.  Cox, J. H., and C. M. Overall. In press. Cytokine Substrates: MMP Regulation of Inflammatory Signalling Molecules. The Cancer Degradome Proteases and Cancer Biology. Springer Science and Business Media. -  22.  Cauwe, B., P. E. Van den Steen, and G. Opdenakker. 2007. The biochemical, biological, and pathological kaleidoscope of cell surface substrates processed by matrix metalloproteinases. Crit Rev Biochem Mo! Biol 42:113-185.  23.  Page-McCaw, A., A. J. Ewald, and Z. Werb. 2007. Matrix metalloproteinases and the regulation of tissue remodelling. Nat Rev Mo! Cell B1o18:221-233.  24.  Balbin, M., A. Fueyo, A. M. Tester, A. M. Pendas, A. S. Pitiot, A. Astudillo, C. M. Overall, S. D. Shapiro, and C. Lopez-Otin. 2003. Loss of collagenase-2 confers increased skin tumor susceptibility to male mice. Nat Genet 35:252-257.  25.  Tester, A. M., J. H. Cox, A. R. Connor, A. E. Starr, R. A. Dean, X. S. Puente, C. Lopez Otin, and C. M. Overall. 2007. LPS Responsiveness and Neutrophil Chemotaxis In Vivo Require PMN MMP-8 Activity. PLoS ONE 2:e312.  26.  Van den Steen, P. E., P. Proost, A. Wuyts, J. Van Damme, and G. Opdenakker. 2000. Neutroph ii gelatinase B potentiates interleukin-8 tenfold by aminoterminal processing, whereas it degrades CTAP-Ill, PF-4, and GRO-aipha and leaves RANTES and MCP-2 intact. Blood 96:2673-2681.  27.  Van Den Steen, P. E., A. Wuyts, S. J. Husson, P. Proost, J. Van Damme, and G. Opdenakker. 2003. Gelatinase B/MMP-9 and neutrophil collagenase/MMP-8 process the chemokines human GCP-2/CXCL6, ENA-78/CXCL5 and mouse GCP-2/LIX and modulate their physiological activities. Eur J Biochem 270:3739-3749.  28.  Van den Steen, P. E., S. J. Husson, P. Proost, J. Van Damme, and G. Opdenakker. 2003. Carboxyterminal cleavage of the chemokines MIG and IP-lO by gelatinase B and neutrophil collagenase. Biochem Biophys Res Commun 310:889-896.  29.  Zhang, K., G. A. McQuibban, C. Silva, G. S. Butler, J. B. Johnston, J. Holden, I. ClarkLewis, C. M. Overall, and C. Power. 2003. HIV-induced metalloproteinase processing of the chemokine stromal cell derived factor-i causes neurodegeneration. Nat Neurosci 6:1064-1071.  95  30.  Overall, C. M. 2002. Molecular determinants of metalloproteinase substrate specificity: matrix metalloproteinase substrate binding domains, modules, and exosites. Mo! Biotechnol 22:51-86.  31.  Burrage, P. S., K. S. Mix, and C. E. Brinckerhoff. 2006. Matrix metalloproteinases: role in arthritis. Front Biosci 11:529-543.  32.  Bar-Or, A., R. K. Nuttall, M. Duddy, A. Alter, H. J. Kim, I. Ifergan, C. J. Pennington, P. Bourgoin, D. R. Edwards, and V. W. Yong. 2003. Analyses of all matrix metalloproteinase members in leukocytes emphasize monocytes as major inflammatory mediators in multiple sclerosis. Brain 126:2738-2749.  33.  Kouwenhoven, M., V. Ozenci, A. Gomes, D. Yarilin, V. Giedraitis, R. Press, and H. Link. 2001. Multiple sclerosis: elevated expression of matrix metalloproteinases in blood monocytes. J Autoimmun 16:463-470.  34.  Lopez-Otin, C., and L. M. Matrisian. 2007. Emerging roles of proteases in tumour suppression. Nat Rev Cancer 7:800-808.  35.  Wolf, K., R. Muller, S. Borgmann, E. B. Brocker, and P. Friedl. 2003. Amoeboid shape change and contact guidance: T-lymphocyte crawling through fibnllar collagen is independent of matrix remodeling by MMPs and other proteases. Blood 102:3262-3269.  36.  Dean, R. A., G. S. Butler, Y. Hamma-Kourbali, J. Delbe, D. R. Brigstock, J. Courty, and C. M. Overall. 2007. Identification of candidate angiogenic inhibitors processed by matrix metalloproteinase 2 (MMP-2) in cell-based proteomic screens: disruption of vascular endothelial growth factor (VEGF)/heparin affin regulatory peptide (pleiotrophin) and VEGF/Connective tissue growth factor angiogenic inhibitory complexes by MMP-2 proteolysis. Mo! Ce!! Biol 27:8454-8465.  37.  Tam, E. M., C. J. Morrison, Y. I. Wu, M. S. Stack, and C. M. Overall. 2004. Membrane protease proteomics: Isotope-coded affinity tag MS identification of undescribed MT1matrix metalloproteinase substrates. Proc Nati Acad Sd U S A 101:6917-6922.  38.  Abraham, M., S. Shapiro, A. Karni, H. L. Weiner, and A. Miller. 2005. Gelatinases (MMP 2 and MMP-9) are preferentially expressed by Thi vs. Th2 cells. J Nouroimmuno! 163:157-164.  39.  Will, H., S. J. Atkinson, G. S. Butler, B. Smith, and G. Murphy. 1996. 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 Chem 271:17119-17123.  40.  Matias-Roman, S., B. G. Galvez, L. Genis, M. Yanez-Mo, G. de Ia Rosa, P. Sanchez Mateos, F. Sanchez-Madrid, and A. G. Arroyo. 2005. Membrane type 1-matrix metalloproteinase is involved in migration of human monocytes and is regulated through their interaction with fibronectin or endothelium. B!ood 105:3956-3964.  41.  Yang, M. X., X. Qu, B. H. Kong, Q. L. Lam, Q. Q. Shao, B. P. Deng, K. H. Ko, and L. Lu. 2006. Membrane type 1-matrix metalloproteinase is involved in the migration of human monocyte-derived dendritic cells. Immuno! Cell Biol 84:557-562.  42.  Gerard, C., and B. J. Rollins. 2001. Chemokines and disease. Nat Immunol 2:108-115.  43.  Hancock, W. W., L. Wang, Q. Ye, R. Han, and I. Lee. 2003. Chemokines and their receptors as markers of allograft rejection and targets for immunosuppression. Curr Opin Immuno! 15:479-486.  44.  Sorensen, T. L., M. Tani, J. Jensen, V. Pierce, C. Lucchinetti, V. A. Folcik, S. Qin, J. Rottman, F. Sellebjerg, R. M. Strieter, J. L. Frederiksen, and R. M. Ransohoff. 1999. Expression of specific chemokines and chemokine receptors in the central nervous system of multiple sclerosis patients. J Clin !nvest 103:807-815.  96  45.  Hensbergen, P. J., D. Verzijl, C. I. Balog, R. Dijkman, R. C. van der Schors, E. M. van der Raaij-Helmer, M. J. van der Plas, R. Leurs, A. M. Deelder, M. J. Smit, and C. P. Tensen. 2004. Furin is a chemokine-modifying enzyme: in vitro and in vivo processing of CXCL1O generates a C-terminally truncated chemokine retaining full activity. J Biol Chem 279:13402-13411.  46.  Ludwig, A., F. Schiemann, R. Mentlein, B. Lindner, and E. Brandt. 2002. Dipeptidyl peptidase IV (CD26) on T cells cleaves the CXC chemokine CXCL1 1 (1-TAC) and abolishes the stimulating but not the desensitizing potential of the chemokine. J Leukoc Biol 72:183-191.  47.  Proost, P., E. Schutyser, P. Menten, S. Struyf, A. Wuyts, G. Opdenakker, M. Detheux, M. Parmentier, C. Durinx, A. M. Lambeir, J. Neyts, S. Liekens, P. C. Maudgal, A. Billiau, and J. Van Damme. 2001. Amino-terminal truncation of CXCR3 agonists impairs receptor signaling and lymphocyte chemotaxis, while preserving antiangiogenic properties. Blood 98:3554-3561.  48.  Clark-Lewis, I., L. Vo, P. Owen, and J. Anderson. 1997. Chemical synthesis, purification, and folding of C-X-C and C-C chemokines. Methods Enzymol 287:233-250.  49.  Butler, G. S., E. M. Tam, and C. M. Overall. 2004. The canonical methionine 392 of matrix metalloproteinase 2 (gelatinase A) is not required for catalytic efficiency or structural integrity: probing the role of the methionine-turn in the metzincin metalloprotease superfamily. J Biol Chem 279:15615-15620.  50.  Tam, E. M., T. R. Moore, G. S. Butler, and C. M. Overall. 2004. Characterization of the distinct collagen binding, helicase and cleavage mechanisms of matrix metalloproteinase 2 and 14 (gelatinase A and MT1-MMP): the differential roles of the MMP hemopexin c domains and the MMP-2 fibronectin type II modules in collagen triple helicase activities. J Biol Chem 279:43336-43344.  51.  Knight, C. G., F. Willenbrock, and G. Murphy. 1992. A novel coumarin-labelled peptide for sensitive continuous assays of the matrix metalloproteinases. FEBS Left 296:263-266.  52.  Loetscher, M., B. Gerber, P. Loetscher, S. A. Jones, L. Piali, I. Clark-Lewis, M. Baggiolini, and B. Moser. 1996. Chemokine receptor specific for IP1O and mig: structure, function, and expression in activated T-lymphocytes. J Exp Med 184:963-969.  53.  Fricker, S. P., V. Anastassov, J. Cox, M. C. Darkes, 0. Grujic, S. R. ldzan, J. Labrecque, G. Lau, R. M. Mosi, K. L. Nelson, L. Qin, Z. Santucci, and R. S. Wong. 2006. Characterization of the molecular pharmacology of AMD31 00: a specific antagonist of the G-protein coupled chemokine receptor, CXCR4. Biochem Pharmacol 72:588-596.  54.  Takahashi, A., P. Camacho, J. D. Lechleiter, and B. Herman. 1999. Measurement of intracellular calcium. Physiol Rev 79:1089-1125.  55.  Unneberg, P., J. J. Merelo, P. Chacon, and F. Moran. 2001. SOMCD: method for evaluating protein secondary structure from UV circular dichroism spectra. Proteins 42:460-470.  56.  Kakehi, K., Y. Oda, and M. Kinoshita. 2001. Fluorescence polarization: analysis of carbohydrate-protein interaction. Anal Biochem 297:111-116.  57.  Lau, E. K., C. D. Paavola, Z. Johnson, J. P. Gaudry, E. Geretti, F. Borlat, A. J. Kungl, A. E. Proudfoot, and T. M. Handel. 2004. Identification of the glycosaminoglycan binding site of the CC chemokine, MCP-1: implications for structure and function in vivo. J Biol Chem 279:22294-22305.  58.  Overall, C. M., G. A. McQuibban, and I. Clark-Lewis. 2002. Discovery of chemokine substrates for matrix metalloproteinases by exosite scanning: a new tool for degradomics. Biol Chem 383:1059-1066.  59.  Shapiro, S. D. 2003. Proteolysis in the lung. EurRespirJ Suppl44:30s-32s.  97  60.  Cole, K. E., C. A. Strick, T. J. Paradis, K. T. Ogborne, M. Loetscher, R. P. Gladue, W. Lin, J. G. Boyd, B. Moser, D. E. Wood, B. G. Sahagan, and K. Neote. 1998. Interferon inducible T cell alpha chemoattractant (1-TAC): a novel non-ELR CXC chemokine with potent activity on activated T cells through selective high affinity binding to CXCR3. J Exp Med 187:2009-2021.  61.  Clark-Lewis, I., I. Mattioli, J. H. Gong, and P. Loetscher. 2003. Structure-function relationship between the human chemokine receptor CXCR3 and its ligands. J Biol Chem 278:289-295.  62.  Ogilvie, P., G. Bardi, I. Clark-Lewis, M. Baggiolini, and M. Uguccioni. 2001. Eotaxin is a natural antagonist for CCR2 and an agonist for CCR5. Blood 97:1920-1924.  63.  Sauty, A., R. A. Colvin, L. Wagner, S. Rochat, F. Spertini, and A. D. Luster. 2001. CXCR3 internalization following T cell-endothelial cell contact: preferential role of IFN inducible T cell alpha chemoattractant (CXCL11). J immunol 167:7084-7093.  64.  Booth, V., I. Clark-Lewis, and B. D. Sykes. 2004. NMR structure of CXCR3 binding chemokine CXCL1 1 (ITAC). Protein Sd 13:2022-2028.  65.  Lusti-Narasimhan, M., A. Chollet, C. A. Power, B. Allet, A. E. Proudfoot, and T. N. Wells. 1996. A molecular switch of chemokine receptor selectivity. Chemical modification of the interleukin-8 Leu25 --> Cys mutant. J Biol Chem 271:3148-3153.  66.  Proudfoot, A. E., S. Fritchley, F. Borlat, J. P. Shaw, F. Vilbois, C. Zwahlen, A. Trkola, D. Marchant, P. R. Clapham, and T. N. Wells. 2001. The BBXB motif of RANTES is the principal site for heparin binding and controls receptor selectivity. J Biol Chem 276:10620-10626.  67.  Kuschert, G. S., A. J. Hoogewerf, A. E. Proudfoot, C. W. Chung, R. M. Cooke, R. E. Hubbard, T. N. Wells, and P. N. Sanderson. 1998. Identification of a glycosaminoglycan binding surface on human interleukin-8. Biochemistry 37:11193-11201.  68.  Busby, T. F., W. S. Argraves, S. A. Brew, I. Pechik, G. L. Gilliland, and K. C. lngham. 1995. Heparin binding by fibronectin module 111-13 involves six discontinuous basic residues brought together to form a cationic cradle. J Biol Chem 270:18558-18562.  69.  Turk, B. E., L. L. Huang, E. T. Piro, and L. C. Cantley. 2001. Determination of protease cleavage site motifs using mixture-based oriented peptide libraries. Nat Biotechnol 19:661-667.  70.  Quantin, B., G. Murphy, and R. Breathnach. 1989. Pump-i cDNA codes for a protein with characteristics similar to those of classical collagenase family members. Biochemistry 28:5327-5334.  71.  Shapiro, S. D., D. K. Kobayashi, and T. J. Ley. 1993. Cloning and characterization of a unique elastolytic metalloproteinase produced by human alveolar macrophages. J Biol Chem 268:23824-23829.  72.  Fernandez, E. J., and E. Lolis. 2002. Structure, function, and inhibition of chemokines. Annu Rev Pharmacol Toxicol 42:469-499.  73.  Herman, G. A., P. P. Stein, N. A. Thornberry, and J. A. Wagner. 2007. Dipeptidyl peptidase-4 inhibitors for the treatment of type 2 diabetes: focus on sitagliptin. Clin Pharmacol Ther8l :761-767.  98  CHAPTER 4. MMP-8 DEFICIENCY DELAYS NEUTROPHIL APOPTOSIS AND EXACERBATES RHEUMATOID ARTHRITIS 4 Perspective Neutrophil accumulation in inflammation is determined by cell infiltration, apoptosis and clearance. We identify an important new mechanism regulating neutrophil apoptosis that is orchestrated by the neutrophil-specific matrix metalloproteinase-8 (MMP-8) or collagenase-2. Mmp8 -I- mice were backcrossed onto the MRL/lpr background, characterized by a defective Fas receptor that results in systemic autoimmunity including spontaneous rheumatoid arthritis. Surprisingly, male MMP-8 deficient mice had earlier and more severe joint inflammation coupled with a massive accumulation of neutrophils in synovial tissues. CLlPCHlPTM microarray analysis revealed abolished expression of caspase-1 I in neutrophils from MMP-8 knockout mice. This was confirmed at the protein level in unstimulated or LPS, IFN y and TNF-a-treated MMP-8 -I- neutrophils. Caspase-1 1 is an activator of caspase 3 and in Mmp8 -I- neutrophils caspase-3 activity was significantly reduced as was constitutive apoptosis. MMP-8 sheds TNF-a, which induces neutrophil apoptosis. In unstimulated  MMP-8 deficient  neutrophils,  TNF-o  shedding  was  reduced.  Therefore, MMP-8 is required for expression of caspase-1 I and downstream activation of caspase-3, constitutive TNF-a shedding and neutrophil apoptosis in the resolution of inflammation. In autoimmune diseases such as rheumatoid arthritis, MMP-8 deficiency leads to an exaggerated accumulation of neutrophil infiltrates with concurrent pathological changes including earlier onset, increased incidence and more severe disease. Hence MMP-8 is an anti-target in the treatment of arthritis.  A version of this chapter has been submitted for publication. Cox JH, Starr AE, Kappelhoff R, Yan R, Roberts CR, and Overall CM. Matrix metalloproteinase-8 deficiency exacerbates rheumatoid arthritis through delayed neutrophil apoptosis and reduced caspase-1 I expression. 2008. Manuscript currently under review at The Journal of Immunology.  99  Introduction Dysregulation of the cellular components of the innate and adaptive immune responses can result in autoimmune diseases such as rheumatoid arthritis. Neutrophils, typically considered acute inflammatory cells, are also present in a variety of chronic inflammatory diseases including arthritis, autoimmune colitis, and psoriasis (1, 2). In rheumatoid arthritis, neutrophils degrade cartilage components and are essential for the initiation and maintenance of inflammation in murine models of arthritis (3-5). Matrix metalloproteinases (MMPs) have long been considered promising drug targets in arthritis treatment due to their ability to degrade collagen and proteoglycans (6), major components of cartilage and bone, and MMPs are upregulated in the synovial fluid and tissues of affected joints (7). MMP-8 (also known as collagenase-2) is produced primarily by neutrophils and is released from the specific granules at sites of inflammation and some reports indicate that low levels of MMP-8 mRNA can be induced in synovial cells (8). MMPs have more recently been shown to precisely process and regulate inflammatory mediators such as chemokines (9, 10), tumor necrosis factor-a (TNF-a) (11), and cytokine binding proteins that otherwise mask bioactivity (12), identifying pleiotrophic roles for MMPs in both promoting and dampening inflammation. This has complicated MMP drugtargeting strategies that must now consider additional roles other than simply extracellu lar matrix degradation (13). The pro-inflammatory cytokines interleukin-13 (lL-13), TNF-a, and interferon-y (IFN y) are central molecules in the pathogenesis of chronic inflammation, acting in signaling cascades with chemokines, initiating the recruitment of inflammatory cells, and enabling the self-perpetuation of inflammation. In auto-inflammatory disorders such as rheumatoid arthritis, blocking IL-113 and TNF-a activity with receptor antagonists or neutralizing antibodies has proven beneficial, however animal models have shown that IL-113 is more important in joint damage (14, 15). Furthermore, deletion of caspase-1, also known as interleukin-1 converting enzyme (ICE), prevents maturation of IL-I 3 and improves inflammatory diseases (16). Caspase-I I,  100  which is an activator of both caspase-1 and the pro-apoptotic caspase-3 (17), could also represent a potential drug target to inhibit chronic inflammation, being involved in neutrophil apoptosis and hence clearance of cellular infiltrate. MMP-8 deficient mice  have  no  overt  phenotype,  with  normal  embryonic  development, fertility, and long-term survival (18) but they have diminished LPS responsiveness in a subcutaneous air pouch model (19). In skin carcinogenesis studies, an abnormal inflammatory response was observed at the tumor-stromal interface in MMP-8 knockout mice that was characterized by an initial delay in neutrophil infiltration followed by a prolonged accumulation that did not dissipate, resulting in increased tumorigenesis (18). We recently investigated the mechanism of  this  initial  lag  finding  that  MMP-8  cleaves  and  activates  murine  lipopolysaccharide-induced CXC chemokine, human CXCL5/ENA-78, and human CXCL8/lL-8, all potent neutrophil chemoattractants, in a novel in cis feed-forward mechanism that is lost upon MMP-8 gene deletion (19). Hence, MMP-8 plays a coordinating role in leukocyte trafficking with clear implications for disease pathogenesis. However, its role in chronic inflammatory disease is unknown as is the mechanism leading to chronic accumulation of neutrophils in the absence of MMP-8. Elucidating this mechanism will reveal important new aspects of the control of neutrophil function in innate immunity and pathology. The MRL/Ipr lymphoproliferative mouse is characterized by systemic autoimmune disease including inflammation of the joints, skin, salivary glands, and brain and marked lymphoid hyperplasia, glomerulonephritis, and vasculitis, resulting in significant mortality in the first 6 months of life (20). The Ipr mutation has been localized to the fas gene and causes faulty expression of Fas receptor (CD95/Apo1), leading to defective Fas-dependent lymphocyte apoptosis and impaired deletion of autoreactive lymphocytes (21, 22). The MRLIIpr strain provides a mild, age dependent, naturally susceptible model for arthritis and other autoimmune diseases. As such, it is considered a good mimic of human disease compared to acute and severe models such as collagen-induced arthritis.  Here we examine the role of  101  MMP-8 in MRL//pr autoimmune arthritis. We show a protective role for the enzyme, as MMP-8 deficient mice have an accumulation of neutrophils in the joints. Moreover, MMP-8-deficient cells were found to have no expression of caspase-1 1, reduced caspase-3 activity and hence downstream reduction in neutrophil apoptosis. TNF-a shedding from MMP-8 deficient neutrophils was also dampened. We conclude that MMP-8 has anti-inflammatory functions in the chronic inflammatory setting, in part due to regulation of caspase-1 I expression and consequent caspase 3 activity, thereby promoting neutrophil apoptosis. Hence, the neutrophil MMP-8 is a new anti-target in chronic inflammation.  Materials and Methods Generation of Mmp8 -I- mice.  MMP-8 deficient mice on a C57BL/6 x 129  background were kindly provided by Dr. Steven Shapiro (Harvard School of Medicine,  USA)  and  backcrossed  for  8  generations  onto  the  MRLJIpr  lymphoproliferative strain (Jackson Labs 006825), after which wild-type and MMP-8 deficient mice were maintained as homozygous lines. Genotyping was performed by Hot Shot DNA isolation (23) followed by three-primer PCR with wild-type primers TGT CGG GTC CTG GTT TAC ATT CTG (forward) and AAG GTC AGG GGC CAT GCT ACA (reverse) and knockout primer CGC CTT CU GAC GAG TTC TTC TGA (reverse). The PCR products were 1100 bp for wild-type and 550 bp for knockout. All animal breeding and experimental procedures were approved by the University of British Columbia Animal Care Committee (Appendix C). Western blot and immunohistochemistry. Murine bone marrow neutrophils were purified by flushing of freshly isolated femurs and tibias with Dulbeccos modified eagle’s medium (DMEM, Invitrogen) containing 10% fetal bovine serum (FBS, Hyclone) followed by separation on a discontinuous Histopaque gradient (Sigma) (24). Neutrophil content and viability was consistently confirmed to be greater than 95% by Duff-Quick staining (IMEB Inc.) and trypan blue exclusion, respectively. Cell lysates were prepared by 15 mm  incubation with 0.2% Zwittergent 3-14 detergent  102  (Calbiochem) in PBS at 4 °C. MMP-8 and MMP-9 western blots were performed on 10% SDS-PAGE gels transferred to lmmobilon-FL PVDF membrane (Millipore) and imaged on a LI-COR Odyssey using an Alexa-conjugated secondary antibody (Molecular Probes). Rabbit anti-mouse MMP-8 antibody was kindly provided by Dr. C. Lopez-Otin (University of Oviedo, Spain) (18), used at 1:10,000 and rabbit antimouse MMP-9 was used at 1:1,000 (Sigma). Rabbit anti-actin was used at 1:200 (Sigma).  Induction and measurement of arthritis.  To synchronize and enhance the  initiation of spontaneous arthritis, male and female mice at 13-14 weeks of age were injected intradermally at two inguinal sites with 0.05 mL of a 1:1 emulsion of water with complete Freund’s adjuvant (Sigma) supplemented to 10 mg/mL heat-killed Mycobacterium tuberculosis (Difco). The width of both hind ankles was measured on 3 consecutive days before arthritis induction and every 2-3 days for 30 days following injection using an electronic digital caliper (Marathon).  Baseline ankle  width was calculated as the average of pre-injection measurements. Arthritis was considered to have occurred following a 10% increase in average ankle width. Statistical analysis was done with Prism 4.0 (Graphpad Software) using two-way ANOVA with Bonferroni post-tests. Histological assessment of arthritis.  At day 30 post-injection, mice were  sacrificed by CO 2 asphyxiation, hind ankles were removed, skinned and formalin fixed for 8-12 h.  Ankles were then decalcified for 72 h in 10% formic acid,  dehydrated and paraffin-embedded.  Sections (3-4 1 km) were stained with  hematoxylin and eosin or toluidine blue. Tissue sections were coded and graded for inflammation by a blinded observer. Neutrophil immunostaining was performed with a 1:2,000 dilution of rat anti-mouse 7/4 antibody (Serotec) coupled with Vectastain ABC and 3,3’-diaminobenzidine (DAB) detection (Vector labs). Macrophages and T lymphocytes were stained with F4/80 antibody (Serotec) and anti-CD3 (Dako), respectively, both at a 1:50 dilution. Tissues were stained for collagenase cleavage sites in type II collagen using the COL2-3/4Ch t neoepitope antibody against the 01  103  peptide GPHypGPGG, kindly provided by Dr. John Mort (McGill University, Canada) (25). Aggrecan neoepitope antibodies generated against VDIPEN peptide (MMP cleavage) and NITEGE peptide (aggrecanase cleavage) were generously provided by Dr. Amanda Fosang (University of Melbourne, Australia) (26, 27). CLlPCHIPTM microarray analysis. Murine CLlPCHlPTM analysis, a focused DNA  microarray with probes for all murine proteases, their inhibitors, and inactive homologs was performed as described previously (28). Isolated bone marrow neutrophils from 2-3 mice were pooled and total RNA was purified using the Qiagen RNeasy protocol including an on-column DNasel digest. Reverse transcription and amplification was done with MessageAmp II aRNA amplification kit (Ambion). Twocolor labeling was performed with ULS aRNA fluorescent labeling kit reagents (Kreatech). All hybridizations were performed in quadruplicate including dye-swap controls.  Microarray slides were scanned on a 428 Array Scanner (MWG) and  images were analyzed with Imagene 6.1 (Biodiscovery). Data from biological replicates was analyzed with TM4 Microarray Software Suite including background subtraction and local LOWESS normalization (29). Quantitative real-time PCR. Total RNA from bone marrow neutrophils was reverse transcribed with Ambion Arrayscript and 50 ng cDNA was used for Taqman analysis with the caspase-li gene expression assay and 18S RNA as a control (Applied Biosystems). Real time PCR was performed on the BioRAD Opticon: 10 mm at 95  °C, followed by 40 cycles of 15 s at 95 °C and I mm at 60 °C. Relative quantification of transcript abundance was performed using the AACt method (30). Murine MMP-8 Expression. The murine MMP-8 pro-enzyme was cloned with a FLAG-tag in the pGWIHG vector and expressed in Chinese hamster ovary cells. Protein from conditioned medium was purified on Reactive Green Agarose (Sigma) followed by FLAG affinity chromatography (Sigma).  Before use, enzyme was  activated for 1 h with I iM 4-aminophenymercuric acetate (Sigma) at 37 °C followed by desalting on a Zeba spin column (Pierce). Concentration was determined by  104  active site titration with murine tissue inhibitor of metalloproteinases-2 (TIMP-2). Caspase-1 I immunoblotting.  Bone marrow neutrophils and mononuclear cells  were isolated by Histopaque gradient as described above and treatments included 24 h incubation at 37 °C, 5% CO 2 with 100 ng/mL LPS (E. coil 0111 :B4, Sigma), 10 Ig/mL LPS, 10 ng/mL interferon-y (Peprotech), 20 ng/mL murine TNF-a (Peprotech), or I Ig/mL active murine MMP-8. Cell lysates for caspase-1 I analysis were prepared by lysis in 50 mM Tris, 150 mM NaCI, 1 mM EDTA, and 1% Nonidet-40, pH 7.4 containing EDTA-free protease inhibitor cocktail (Roche) for 30 mm at 4 °C followed by 15 mm centrifugation at 13,000 x g at 4 °C to remove insoluble material. Protein concentration was determined by Bradford assay (Bi0RAD). Proteins were separated on a 12% SDS-PAGE gel and transferred to lmmobilon-P PVDF membrane (Millipore). Monoclonal rat anti-caspase-Il (Sigma) was used at 1:500 and secondary detection was with anti-rat-HRP (Sigma) followed by Enhanced Chemiluminescence (ECL) exposure (GE Healthcare) to X-omat blue film (Kodak). Caspase activity and interleukin-I1 ELISA. Caspase-1 and caspase-3 activity were measured following incubation of 2 x 106 cells/mL for 24 h at 37 °C, 5% CO 2 in DMEM containing streptomycin.  10% FBS and  100 units/mL penicillin and  100 tg/mL  Treatments included 100 ng/mL LPS (E. coil 0111 :B4, Sigma), 10  ig/mL LPS, and 10 ng/mL interferon-y (Peprotech).  Lysates were prepared by  incubation of 2 x 106 cells with 75 tL lysis buffer (Biovision) for 15 mm  at 4 °C  followed by centrifugation at 13,000 x g to remove insoluble material.  Protein  content was determined by Bradford assay (Bi0RAD). Activity was measured in 20 mM HEPES, 10% glycerol, pH 7.5 containing 2 mM and 10 mM DTT for caspase-3 and caspase-1, respectively. Substrate concentrations were 50 tM Ac-YVAD-AMC (Calbiochem) for caspase-1 and 10 .tM Ac-DEVD-AMC (Calbiochem) for caspase-3 and rates were measured at 390nm/460nm on a BMG Polarstar fluorescence plate reader.  Rates  were  converted  to  pmoles/h  per  g  protein  based  on  aminomethylcoumarin (Calbiochem) standard curves. lnterleukin-1 j was quantified in cell-free medium by ELISA quantikine kit (R&D Systems) as per manufacturers 105  protocol. TNF-a cleavage and ELISA. Full length proTNF-a N-terminally fused to GST was expressed in E. coil and affinity-purified as described before (11) and incubated with human and rodent MMP-8 in 200 mM NaCI, 5 mM CaCI , 0.05% Brij25, 50 mM Tris 2 HCI, pH 7.4. Incubation was at enzyme-substrate ratio of 1:100 (w:w) for 18 h at 37 °C. Synthetic hydroxymate inhibitor BB21 16 was used at 20 !M as a negative control. Reaction mixes were analyzed on 6-15% gradient SDS-PAGE gels and protein bands in the gels were stained with coomassie G250 and in parallel gels were transferred to nitrocellulose for N-terminal Edman sequencing (5 cycles). TNF a secretion from isolated neutrophils was quantified in cell-free medium by ELISA quantikine kit (R&D Systems) as per manufacturers protocol. Neutrophil apoptosis. Spontaneous apoptosis of bone marrow-derived neutrophils was determined following 24 h incubation at 37 °C, 5% Co . 2  Briefly, cytospins  (Shandon) of 30,000 cells were prepared and stained with TUNEL (Roche) following manufacturers protocol with Hoechst nuclear staining. A minimum of four random fields were analyzed per slide and percentage of TUNEL positive cells was determined relative to Hoechst control. Results  Generation of MMP-8 deficient MRLJ1pr mice To address the role of MMP-8 in inflammatory arthritis, Mmp8-nuII mice on a C57BL/6 x 129 background were backcrossed for 8 generations with the MRLIIpr lymphoproliferative strain. Genotype was determined by PCR with bands at 550 and 1100 bp for wild-type and knockout, respectively (Figure 4.IA), and was confirmed by western blofting (Figure 4.IB), showing an absence of MMP-8 in neutrophil lysates from MRL/Ipr x Mmp8 -I- mice. Mice lacking MMP-8 had normal development and fertility, comparable to wild-type controls. MMP-9, another MMP expressed in neutrophils, has been described previously to increase in the MMP-8 deficient  106  A  B Mmp8 +1+  +1+  -I-  Mmp8 1100 bp 550 bp  Mmp9 Actin  —  —-  Figure 41. Generation and characterization of MRLIIprx Mmp8 -I- mice. (A) PCR genotyping results demonstrating wild-type and knockout products at 1100 and 550 base pairs, respectively. (B) Western blot of lysates of bone marrow neutrophils from Mmp8 +1+ and Mmp8 -I- mice (both on the MRLIIpr background) showing loss of MMP-8 and an increase in MMP-9 levels in MMP-8 deficient cells. Actin was analyzed as a loading control.  107  C57BL/6 x 129 strain (31). Here we also found a similar compensation on a different genetic background, confirming this to be an MMP-8 associated effect (Figure 4.IB). More severe arthritis in male MMP-8 deficient MRL,lpr mice The MRL/Ipr model of arthritis is an age-dependent spontaneous inflammatory disease model that can be initiated and enhanced by injection of complete Freund’s adjuvant (20). Therefore, the induction of arthritis was synchronized by injection at 13-14 weeks of age and MMP-8 wild-type and knockout mice on the MRLIIpr background were monitored for ankle swelling for 30 days, at which time humane endpoints were apparent and animals were sacrificed (n= 13 and 12 wild-type male and female, n= 11 and 8 knockout males and females). In the male group, MMP-8 knockout mice developed much earlier and more severe joint swelling than wild-type controls, with significant differences as early as 11 days post-treatment (Figure 4.2A). At day 30, average changes in ankle width were 0.25 ÷1- 0.02 and 0.56 +10.06 mm for wild-type and knockout males, respectively.  In contrast, female  knockouts had no significant increase in swelling compared to wild-type females with final ankle width change of 0.32 +1- 0.03 and 0.36 -‘-I- 0.07, respectively (Figure 4.2B).  Incidence of disease, characterized as a greater than or equal to 10% increase in ankle width for two consecutive measurements, reached 100% in knockout males but only 31 % in wild-type males at 30 days, with a much earlier onset in knockout mice (Figure 4.2C).  Female knockout mice had a slight shift toward earlier  incidence, but this was due to one mouse only.  MMP-8 knockout and wild-type  females had similar penetrance (75% and 83%, respectively) (Figure 4.2D), consistent with previous reports studying arthritis in MRLIIpr mice (32). Severity of disease, as determined by maximal percent ankle width increase throughout the experiment, was greater in males deficient in MMP-8 (55% exhibiting severe swelling), and slightly more severe in MMP-8 deficient females (25% exhibiting severe swelling), whereas no wild-type mice had >20% increase in ankle width (Figure 4.2E, F). These results clearly display a protective effect of MMP-8 in the  108  Figure 4.2. Increased joint swelling in male MMP-8 deficient MRLIIpr mice. (A) Mice were monitored for changes in hind ankle width for 30 days following treatment with complete Freund’s adjuvant and male MMP-8 deficient mice developed more severe swelling than wild-type controls (N=1 3 and 11 for WT and KO males, respectively) (* p<O.05, ** p<0.01, p<0.001). (B) No significant change was detected in comparing female MMP-8 knockout mice with wild-type controls (N=12 and 8 for WT and KO females, respectively). (C, D) Disease incidence was defined as >10% increase in ankle width (relative to baseline width) for at least two consecutive measurements. (C) Male MMP-8 knockout mice reached 100% incidence at day 30 whereas only 31% of wild-type males displayed measurable disease. (D) Female wild-types and knockout mice had similar incidence with a slightly earlier onset in the knockout population. (E,F) Severity of disease was graded based on the maximal width increase (%) throughout the experiment: mild (0-10%), moderate (10-20%), and severe (>20%). (E) Male MMP-8 knockouts had much more severe disease with over 50% of mice having severe disease whereas female knockouts (F) had a modest shift toward increased severity.  109  A  B 0.8 ***  • KOMALE i1 WT MALE  *** ***  •0  E  0.6  **  • KOFEMALE 0 WTFEMALE  **  DD  C  0.2 0 0) C 0.0  .  15  20  I 1315  25  Day  0-0.2  C  20  25  30  20  25  30  Day  D 100  —;.  80’  a)  60  C.)  • KO MALE o WTMALE  • KOFEMALE o WT FEMALE a) C.) C  C  a) C.) C  a)  •0 C.) C  20  5  10  20  15  25  5  30  Day  10  15  Day  E  F 100  100  • KOMALE o WTMALE  80 60 40 20  0-10  1  10-20  Increase in Width  I  >20  (%)  • KO FEMALE 0 WT FEMALE  i 0-10  10-20  Increase in Width  >20  (%)  110  clinical manifestations of inflammatory arthritis, predominantly in male mice. Furthermore, this gender-based effect exemplifies the importance of studying both sexes in animal models of inflammatory disease. Histological assessment of arthritis Hematoxylin and eosin staining of sectioned joints revealed increased cellular infiltration in the MMP-8 knockout males, relative to wild-type males, with an abundance of inflammatory cells in the subsynovial tissue and some cellularity in the joint space (Figure 4.3A). In contrast, female wild-types and knockouts had similar levels of cellular infiltration (Figure 4.3B). Scoring of inflammatory content revealed a significant increase in knockout males (p<O.OO1) with no change in females (Figure 4.31, J). Toluidine blue staining for proteoglycans, a measure of cartilage integrity, displayed an absence of cartilage damage in all groups, a surprising finding given the extent of inflammation present in these tissues (Figure 4.3C and D).  MMP-8  cleaves aggrecan and so genetic deficiency of MMP-8 might account for this despite the large amount of cellular infiltrate (33). However, use of neo-epitope antibodies to detect MMP-8 processed murine type II collagen and murine aggrecan resulted in undetectable amounts, supporting a lack of cartilage degradation in the wild-type or MMP-8 knockout mice in this model  (data not shown).  Immunostaining  demonstrated a massive accumulation of neutrophils in the subsynovial tissue surrounding swollen joints and also in the joint space of joints from MMP-8 deficient male mice and all female synovial tissues (Figure 4.3E-H).  Macrophages and T  lymphocytes were also detected with levels correlating with overall inflammation (data not shown).  CLlPCHlPThl microarray analysis of MMP-8 deficient neutrophils The CLIP-CHIP , a DNA microarray dedicated to 823 proteases, non-proteolytic TM homologs and protease inhibitors (28), was employed to examine if compensation by other MMPs or changes in the protease web occurred following deletion of MMP 8. By the criteria of greater than 1.5-fold change in expression with p<O.O1, only 8  111  Figure 4.3. Histological analysis of joints reveals increased neutrophil infiltration in male MMP-8 deficient mice. Hematoxylin and eosin staining of sections from wild-type and MMP-8 knockout male (A) and female (B) mice. A, articular surface; B, bone marrow; S, subsynovium. Toluidine blue staining for proteoglycan content and cartilage integrity in male (C) and female (D) ankle sections. (E-H) Neutrophil immunostaining showing an abundance of neutrophils in the subsynovium and joint space of sections from male MMP-8 deficient and all female joints. Scoring of cellular infiltration (l,J) by a blinded observer demonstrating significantly increase subsynovial inflammation in male Mmp8-null mice (p<O.OO1). Original magnifications xlOO (A-F) and x400 (C-H). Scale bar is equivalent to 100 mm.  112  MALE  MmpS-f-I-’-  FEMALE Mmp8 +1+ Mmp8 -I-  Mmp8-/-  A’  B A.A  .J.  i’  .1  C  G  H  L:. a, 0 0  U,  0 0 U)  0  0  ‘4-I  E E  E E  ‘4-  ‘4-  =  WT  WT  KO  113  genes in male neutrophils and 9 in female neutrophils were differentially expressed between wild-type and knockout samples (Figure 4.4A).  Hence, protease and  inhibitor expression was essentially unaltered between wild-type and MMP-8 deficient bone marrow neutrophils, indicating minor homeostatic alterations in the protease web following loss of MMP-8 expression as demonstrated in a representative R-l (ratio versus intensity) plot (Figure 4.4B). These changes represent approximately I % of all analyzed genes. Internal controls of the transcript analysis were MMP-8 and neomycin, which is the selection marker used in generating the Mmp8-null mice. A notable common change to both male and female neutrophils was caspase-li, the mouse homolog of human caspases 4 and 5, which was unexpectedly downregulated in MMP-8 knockout samples.  Moderate changes in male MMP-8  deficient neutrophils include decreased levels of stefin Al, stefin A2, kallikrein related peptidase-l 1, and cationic trypsin. In female MMP-8 knockout neutrophils the transcripts for ubiquitin specific protease (USP)-l 3, USP-3l, tissue inhibitor of metalloproteinases (TIMP)-2, and complement factor I were downregulated. Other than neomycin, few genes were upregulated in the absence of MMP-8. Kallikrein related peptidase-9 was increased, however expression levels were very low, only slightly above background. In addition, carboxypeptidase AS was increased I .6-fold in female MMP-8 knockout neutrophils.  Caspase-Il is not expressed in MMP-8 deficient neutrophils Caspase-l 1 was explored further due to its roles in caspase-3 and caspase-l activation, with consequent implications in apoptosis and inflammation (17). Relative quantitation by real time PCR generated a DDCT value of 10.7, translating to a greater than 1600-fold change in caspase-1 1 transcript levels, showing that caspase-1 I expression is absent in MMP-8 deficient neutrophils (Figure 4.5A). Notably, this differential caspase-1 I expression was also observed in neutrophils  114  MALE (KO:WT)  A  Fold Change  Gene  13.5  Neomycin  P value  1’  2E-8  1.5 1.6 .j, 1.7 1.8 8.3 .j, 10.4  0.003 0.008 0.003 4E-5 8E-4 8E-8  1te StefinAl Stefin A2 Kallikrein-1 1 Cationic Trypsin Caspase-il MMP-8  ,  ,  .  FEMALE (KO:WT) Gene  Fold Change  P value  14.2 6.7 1.6t 1.6 4, 1.7 4, 1.9 .I 2.5 4, 6.0 4, 14.2 .  IE-7 4E-4  Neomycin Kallikrein-9 USPI3 USP3I TIMP-2 Complem. factor I Caspase-lI MMP-8  B  Not expressed 6  I—  0.005 0.006 0.004 0.007 IE-5 4E-8  Expressed  -V  43 -  -:Neomycin ;—----------  ------  -  -  1 Caspase-li  4  QMMP-8 3  4  5  6  7  8  9  10  11  12  13  I [1og (WT.KO)] 10 Figure 4.4. CLlPCHlPTM analysis of bone marrow neutrophils from MMP-8 knockout and wild-type MRLIIpr mice. (A) Differential expression of proteases and inhibitors with p<0.01 in male and female bone marrow neutrophils. Changes are relative to MMP-8 knockout samples. USP, ubiquitin specific protease; TIMP, tissue inhibitor of metalloproteinases. (B) Representative R-l (ratio versus intensity) scatter plot of data following local LOWESS normalization. An intensity sum of 7 illustrates the boundary between expression levels and low signal. Neomycin, MMP-8, and Caspase-1 1 are highlighted. 115  following ex vivo stimulation with LPS, known to induce caspase-1 I expression, and also in neutrophils isolated from the original C57BL/6 x 129 strain (Figure 4.5A). Western blots confirmed that caspase-1 I was absent in MMP-8 deficient neutrophil lysates and was not induced by treatment with interferonj’ or LPS, treatments that resulted in increased levels in wild-type neutrophils (Figure 4.5B). Surprisingly, caspase-Il was also absent and non-inducible in the mononuclear cell fraction (predominantly monocytes and T lymphocytes) from MMP-8 deficient mice, cells that do not produce MMP-8, suggesting potential paracrine activation of caspase-l I expression in development during hematopoiesis (Figure 4.5C). We attempted to induce caspase-1 I levels in MMP-8 knockout neutrophils and mononuclear cells by rescuing with exogenous active murine MMP-8 or with cell-free bone marrow fluid isolated from wild-type mice.  However, caspase-1 I levels remained unchanged  (Figure 4.5D and data not shown). Hence, MMP-8 regulation of caspase-1 I gene expression may require factors from additional cell types or could occur upstream in hematopoietic differentiation. Unaltered caspase-1 activity and interleukin-113 secretion in MMP-8 deficient neutrophils Previous reports have defined caspase-1 I as an essential activator of caspase-1, the interleukin-1 converting enzyme that is key for IL-i  maturation and release (34).  In response to increasing concentrations of LPS, bone marrow derived MRLIIpr neutrophils had elevated caspase-1 activity, as measure by the fluorometric synthetic peptide substrate Ac-YVAD-AMC (Figure 4.6A). However, comparing wild type and Mmp8-nulI neutrophils, there was no significant difference in caspase-1 activity, despite substantial differences in caspase-1 I expression.  LPS-mediated  induction of caspase-1 activity was more pronounced in mononuclear cells, reaching an activity greater than 0.4 pmoles/h/g (Figure 4.6B), compared to 0.1 pmol/h/Ig in neutrophils.  However, caspase-1 activities in wild-type and MMP-8 deficient  mononuclear cells were also unchanged (Figure 4.6B). Consistent with unaffected  116  Figure 4.5. Neutrophils and mononuclear cells from MMP-8 deficient mice have abolished constitutive and inducible levels of caspase-Il. (A) Relative quantification of caspase-1 I transcript by real-time PCR with Taqman probes revealed loss of expression in Mmp8 -I- neutrophils from either MRLI1pr or C57BLI6x129 mouse strains. This reduction was maintained following I h treatment of MRLIIpr neutrophils with 10 ig/mL LPS. Lysates (50 ig total protein) from bone marrow neutrophils (B) and mononuclear cells (C) untreated or incubated with 10 ng/mL lFN-’, 0.1 tg/mL or 10 tg/mL LPS show a loss of caspase-1 1 protein levels in cells from male MMP-8 deficient mice. (D) Addition of I tg/mL active murine MMP-8 failed to rescue the caspase-1 I expression defect. Actin intensity is used as a loading control.  117  A  T 2  AACT (WT:KO) SAMPLE MRLIIpr 10.7 +/-OA MRL/Ipr + LPS 9.2 +1- 0.2 C57BL/6x129 9.9 +1- 0.4  B  (WT:KO) 1660 590 960  Neutrophils IFNy  WT 10  -  LPS  -  -  -  0.1  10  —  —  10  -  -  -  -  -  -  -  0.1  -  10  —  .  -  -  0.1  10  ng/mL tg/mL C  Mononuclear cells WT KO 10 10  C  LPS  -  ..L.  —  IFNy  KO  -  -  -  0.1  -  10  11  Acti n  ng/mL g/mL Casp-11  —  D  —  Neutrophils WT  MMP-8LPS  —  -  Acti n  Mononuclear cells  KO  WT  +  -  -  +  -  -  +  -  +  -  -  +  -  -  -  ÷  -  -  KO ÷ -  -  +  Casp-1 I —  —  — Actin  118  Figure 4.6. MMP-8 deficient neutrophils have unaltered caspase-1 activity and IL-I 13 secretion. Caspase-1 activity, measured by Ac-YVAD-AMC cleavage, was unchanged in untreated, IFN-y, or LPS stimulated male MMP-8 deficient neutrophils (A) and mononuclear cells (B) compared to wild-type control cells (n=3). (C) IL-113 secretion (pg/mL) following 24 h culture of bone marrow neutrophils, as determined by ELISA (n=3). Histograms show the means and the SEM for samples in each group where each sample was from an independent mouse.  119  C  -  (1OgImL)  LPS  LPS (0.1 p.gImL)  Untreated  0  I  0 I  0  1-i  I  0 I  0  •  I  IL-I 3 (pg!mL)  I  -  I  0  C) c I  C,  -‘  C  I  IFN-y (lOngImL)  Untreated.  LPS (10 iglmL)  I  0  LPS (10 tg/mL)’  H  Ii  I  0  LPS (0.1 tgImL)  I  .  LI’S (0.1 ig/mL)  IFN-y (lOngImL)  Untreated•  0  0  Caspase-1 Activity Ac-YVAD-AMC Cleavage (pmoI/h/ig) o  ..  I  r.. I  H  I  a I  I  H  I  I  C,)  -  H 0  I 11  I  Caspase-1 Activity > Ac-YVAD-AMC Cleavage (pmollh4Lg)  caspase-1 activity, secreted interleukin-1 3 concentrations from untreated and LPS treated neutrophils were unaltered in MMP-8 deficient samples (Figure 4.6C).  Proteolytic cleavage and shedding of TNF-a by MMP-8 TNF-a is a central pro-inflammatory cytokine that is unique in that it can also induce apoptosis (35). TNF-cz is produced as a membrane-associated inactive precursor that is solubilized upon proteolysis, an event that is required for normal inflammatory functions via TNFRI (36). Previous work has demonstrated that a number of MMPs can process proTNF-a to generate the active soluble form, both biochemically (11) and in vivo (37, 38). In vitro incubation of a GST-proTNF-a fusion protein with human and rodent MMP-8 resulted in efficient and specific cleavage in the prodomain (Figure 4.7A). N-terminal sequencing defined the new N-terminus as VRSSS, which is the physiological terminus of soluble TNF-a. Bone marrow derived MMP-8 deficient neutrophils were examined for their ability to shed TNF-a (Figure 4.7B). In untreated MMP-8 knockout neutrophils, TNF-a shedding was absent and upon IFN-y treatment, soluble TNF-a was reduced compared with wild-type neutrophils.  However, after LPS treatment there was no difference, likely due to  induction of ADAM-17, the predominant TNF-a converting enzyme in infection (39, 40). The effect of exogenous soluble TNF-a on caspase-1 I expression was examined (Figure 4.7C). Wild-type neutrophils had significantly elevated caspase 11 expression in response to TNF-a but neutrophils from MMP-8 deficient mice had undetectable levels.  Caspase-3 activity and constitutive neutrophil apoptosis Caspase-1 I deficient mice have reduced apoptosis and cell death in pathological states due to decreased activation of caspase-3, which is known to be important in the execution phase of apoptosis (41-43). Therefore, caspase-3 activity was measured in leukocytes from wild-type and MMP-8 deficient mice using the  121  A MrX1O0’$  50  m— GST—Pr0TNF-a  ._  36  —  16  .*—  — —  TN F-a  Propeptide I  I  TN F-a  GST-HFGVIGPQREESPRDLSLISPLAQAVRSSSRT rMMP-8 hMMP-8  C  B500  TNF-a  4001  WT -  KO +  -  +  I i 100 ‘—  Casp-11  i  50  [Ii —  2  ri  -  Actin  —  2  oa) C  D  ..J 0  Figure 4.7. MMP-8 mediated cleavage of proTNF-a and shedding of TNF-a from neutrophils. (A) Full length proTNF-a N-terminally fused to GST was incubated with human (h) and rodent (r) MMP-8 resulting in cleavage in the propeptide. MMPI, assays performed in the presence of 20 pM BB2112 hydroxamate inhibitor. Proteins were analyzed on a 6-15% gradient SDS-PAGE, stained with Coomassie G250. Original gel contained other MMP reactions that were removed from figure for presentation. Cleavage products were analyzed by N-terminal sequencing to determine cleavage site. Experiment was performed by Dr. Christopher Overall. (B) TNF-a secretion from bone marrow neutrophils from wild-type and MMP-8 deficient mice (n3, histogram shows SEM). Neutrophils were incubated for 24 h in media alone (untreated), 10 ng/mL IFN-y, or 10 Ig/mL LPS and cell-free supematant was analyzed for TNF-a by ELISA. (C) Bone marrow neutrophils were stimulated for 24 h with 20 ng/mL murine TNF-a and the levels of caspase-1 I were determined by western blot. 122  fluorometric synthetic peptide substrate Ac-DEVD-AMC (Figure 4.8). Both untreated and interferon-y treated MMP-8 knockout neutrophils had significantly reduced caspase-3 activity compared to wild-type cells (p=0.005 and 0.016, respectively) (Figure 4.8A). This difference was abolished in the presence of LPS, an inhibitor of neutrophil apoptosis (44), resulting in reduced caspase-3 activity in both wild-type and knockout neutrophils.  Treatment with TNF-a caused elevated caspase-3  activity in both wild-type and knockout neutrophils. In contrast, no significant change in Ac-DEVD-AMC activity was observed  in parallel treatments of isolated  mononuclear cells from wild-type and MMP-8 deficient mice (Figure 4.8B), suggesting a neutrophil-specific effect that is driven by MMP-8. Previous phenotypes in  MMP-8 deficient mice,  including  increased airway  inflammation and delayed wound healing, were proposed to be due to an apoptotic defect (31, 45). Therefore, the accumulation of neutrophils in joints of MMP-8 deficient mice and the absence of caspase-1 1 in leukocytes from MMP-8 knockout mice prompted the examination of constitutive apoptosis in isolated neutrophils. A 34% decrease in TUNEL-positive cells in the knockout population (p=0.04), from 14.3% to 9.5%, was found (Figure 4.8C, D), and wild-type cell apoptotic staining was consistent with reported values (46). Therefore, we propose that impaired apoptosis in neutrophils from MMP-8 knockout mice involves two novel pathways: abolished caspase-Il  expression and reduced TNF-a shedding from MMP-8 deficient  neutrophils (Figure 4.9).  These mechanisms potentially contribute to the  accumulation of inflammatory cells in arthritic disease and other inflammatory disorders in MMP-8 deficient mice.  123  A  B  a  Mononuclear cells liii WT K0  r  nifli -J  4-  Cu  0  4-  0  _j  0  D  D  C  20  TUNEL *  10 U, 0  a.  Hoechst  -J Lii  5,  z D  0  WT  KO  WT  Figure 4.8. Reduced caspase-3 activity and TUNEL staining in MMP-8 knockout neutrophils. Activity assay for caspase-3 with fluorometric substrate Ac-DEVD-AMC with neutrophil (A) and mononuclear cell (B) lysates from wildtype and MMP-8 knockout male mice (n=3) (* p<O.05, p<O.O1). Constitutive apoptosis of bone marrow neutrophils from MMP-8 wild-type and knockout mice. (C, D) Cells were cultured for 24 h and cytospins were analyzed for DNA fragmentation by TUNEL staining (C) and percentage of TUNEL-positive cells was calculated relative to Hoechst control with MMP-8 knockout cells having significantly less apoptosis (p=O.04, n=3) (D). All histograms show the means and the SEM. **  124  proTN F-a  1’  Caspase-lI  I membrane  Activation  I  Caspase-3  I  Activation  I Caspase-8 I  NEUTROPHIL APOPTOSIS  Figure 4.9. Proposed mechanisms for MMP-8 promotion of programmed cell death through induction of caspase-Il expression and shedding of TNF-a. MMP-8 expression and activity in the neutrophil enables basal caspase 11 expression and responsiveness to LPS and IFN-y through an unknown pathway. MMP-8 also cleaves cell membrane proTNF-a to generate soluble TNF a which signals through TNFRI (TNF receptor 1) and TRADD (TNF receptor associated death domain) to induce caspase 11 and activate caspase-8. Caspase-8 activates caspase-3, leading to neutrophil apoptosis. Therefore, MMP-8 acts via multiple pathways that converge to execute apoptosis in the neutrophil.  125  Discussion Our experiments reveal the surprising finding that the secreted neutrophil-specific protease MMP-8 is a key regulator in the expression of caspase-1 1, basal shedding of TNF-ct, and the orchestration of neutrophil apoptosis. We reported previously that MMP-8, originally described as a collagenase, promotes neutrophil migration in acute inflammation not by collagen degradation, but by processing of the ELR chemokine LIX in mouse and CXCL5/ENA-78 and CXCL8/IL-8 in human (19). In contrast, the resolution of inflammation requires the termination of neutrophil influx and clearance of infiltrated cells. Here we report that male leukocytes from MMP-8 deficient mice lack caspase-1 I expression at the RNA transcript and protein level accompanied  by reduced activity of caspase-3 and decreased constitutive  apoptosis.  Moreover, MMP-8 deficient neutrophils exhibited reduced TNF-a  shedding.  Biologically, this was reflected by an exaggerated accumulation of  infiltrated neutrophils in a murine model of autoimmune rheumatoid arthritis, correlating with increased joint swelling, and so likely contributes to the disease pattern. Hence, MMP-8 is protective in a murine model of autoimmune arthritis by enabling clearance of neutrophils through normal levels of apoptosis, so defining MMP-8 as an anti-target for arthritis treatment. Neutrophil apoptosis is a critical process in the resolution of inflammation. Upon removal of an inflammatory stimulus, neutrophils rapidly undergo programmed cell death through TNF-a, Fas-dependent, or constitutive intrinsic mechanisms (47). Several caspases are key effector molecules in apoptosis, forming the apoptosome, with caspase-3 functioning as an execution phase caspase (48). Levels of active caspase-3 correlate with increased apoptosis (49). Caspase-3 is expressed as a pro-enzyme and can be proteolytically activated by several caspases, including caspase-I 1 (42). Here we report a novel regulatory interaction between MMP-8 and caspase-I I. Interestingly other MMPs have pro-apoptotic roles in non-immune cell types. For example, MMP-7 (matrilysin) promotes epithelial cell apoptosis through processing of FasL (50, 51).  126  By means of DNA microarray analysis, we discovered that MMP-8 deficiency abolishes expression of caspase-1 1, along with ten other alterations in the protease web.  Importantly, complete loss of caspase-1 I expression was also observed in  neutrophils from MMP-8 deficient C57BL/6 x 129 mice, demonstrating a strainindependent association between MMP-8 and caspase-1 1.  Isolation of bone  marrow neutrophils and mononuclear cells from MMP-8 knockout mice confirmed an absence of caspase-1 1, even following treatment with LPS or interferon-y, previously reported to induce caspase-1 I expression through nuclear factor-KB and signal transducer and activator of transcription (STAT)-1, respectively (52). Interestingly, the genes for Mmp8 and Caspl I are both located on mouse chromosome 9 but are separated by over 2 million base pairs and are located at 7.56 and 5.31 megabases, respectively.  Notably, the CaspII gene is adjacent to other inflammasome  caspases (53), Caspl and CaspI2, both of which had normal expression in MMP-8 deficient neutrophils. This further reduces the likelihood that genetic disruption of the capsase-1 1 gene by recombination during the knockout of Mmp8 accounted for these results. Surprisingly, high concentrations of exogenous active murine MMP-8 had no effect on the expression of caspase-1 1 in either wild-type or MMP-8 knockout cells. This suggests that the regulatory role of MMP-8 requires factors derived from surrounding cell types or that an MMP-8 dependent regulatory event occurs upstream in hematopoietic differentiation that persists independent of the presence or absence of MMP-8 activity. Caspase-1 1, the murine homolog of human caspases 4 and 5, is a cysteine protease involved in cytokine maturation and programmed cell death through the proteolytic  activation  of caspase-1  and  caspase-3,  respectively  (17,  42).  Furthermore, caspase-1 1 regulates cell migration through actin interacting protein-i, an activator of cofilin-mediated actin depolymerization (54). Nonetheless, we previously showed that the chemotactic ability of MMP-8 knockout neutrophils in vivo is no different than wild-type neutrophils (19). MMP-8 deficient bone marrow neutrophils have reduced caspase-3 activity compared to wild-type neutrophils. Reduced caspase-3 activity correlated with decreased constitutive apoptosis in  127  isolated bone marrow neutrophils from MMP-8 deficient mice. However, this difference was negated upon treatment with LPS where both wild-type and MMP-8 deficient neutrophils had reduced caspase-3 activity, consistent with the anti apoptotic effect of LPS (44). Caspase-1 activity and IL-i (3 secretion were unchanged in neutrophils and mononuclear cells from MMP-8 deficient mice. Though caspase 11 has been described as an essential activator of caspase-i (34), previous studies with caspase-1 I deficient mice have revealed tissue-dependent phenotypes.  For  instance, intraperitoneal injection of LPS causes reduced IL-i (3 levels whereas intravenous administration yields normal secretion of IL-I (3, relative to wild-type controls (55). As such, the tissue microenvironment plays a decisive role in caspase II activation of caspase-I and consequent IL-1j3 secretion, suggesting that caspase-1 1-independent pathways exist. Our results reveal the possibility of such an alternate pathway in the neutrophil. Although our data show no change in IL-I (3 levels, we cannot discount the possibility that lL-1(3 levels are reduced in MMP-8 knockouts in other cellular microenvironments or disease contexts. TNF-a plays a central role in the pathogenesis of rheumatoid arthritis through the regulation of cytokine, chemokine and MMP production, recruitment of inflammatory cells into joints, angiogenesis, and cell death (56). It was previously reported that MMP-7 and MMP-12 contribute to TNF-a shedding and activation in vivo (37, 38), although ADAM-I7 is the major TNF-cz converting enzyme in response to LPS stimulation or bacterial infection. Here we show for the first time that MMP-8 cleaves proTNF-a in vitro, specifically and efficiently generating the mature form detected physiologically. In comparison to other collagenolyic MMPs (-1, -2, -13, -14) only MMP-8 produced exclusively the correctly processed active form (data not shown). Unstimulated  and  interleron-y treated  MMP-8 deficient  neutrophils  showed  decreased shedding of TNF-a, compared to wild-type neutrophils, whereas LPS treatment resulted in equal levels presumably due to induction of ADAM-17. However, LPS stimulation and the consequent role of ADAM-i7 in TNF-a shedding is likely not relevant in the context of autoimmune disease. The reduction in soluble  128  TNF-a production in the absence of MMP-8 might reduce an autocrine TNF-a mediated apoptotic mechanism within the neutrophil, hence delaying apoptosis and resolution of inflammation. Decreased levels of soluble TNF-a in the context of arthritis could explain the lack of cartilage and bone damage observed in the MMP-8 knockout male mice, despite an abundance of infiltrated inflammatory cells. Notably, TNF-a is a potent inducer of MMP expression in arthritic chondrocytes and synovial fibroblasts (57). Therefore, less TNF-a shedding from MMP-8 deficient neutrophils in the joint should result in decreased production of other MMPs, namely MMP-1 3, now considered to be the predominant MMP drug target in treating arthritis (58). Neutrophils are essential for the initiation and maintenance of chronic joint inflammation in rheumatoid arthritis (5). However, the roles of neutrophil-derived proteases are poorly defined in the onset and progression of arthritis. The creation of the novel MMP-8 deficient MRL/Ipr mouse strain, a genetic background characterized by systemic autoimmunity including rheumatoid arthritis (59), enabled the evaluation of MMP-8 in inflammatory arthritis. Male MMP-8 knockout mice had much earlier and more severe arthritis as compared with wild-type controls, whereas female MMP-8 knockouts had unaltered clinical disease and histology, relative to female wild-type counterparts.  The gender dependence of MMP-8 in chronic  inflammation was observed previously in skin carcinogenesis where the protective role of MMP-8 only occurred in male mice and was rescued by bone marrow transplant (18). In the present study we have demonstrated the same gender effect in a different inflammatory disease and genetic background, thereby supporting the original finding. Despite the marked accumulation of neutrophils in MMP-8 deficient joints of male mice that we attribute to reduced apoptosis, there was a surprisingly low level of cartilage destruction. This is perhaps reflective of the slow progressive nature of this spontaneous model of chronic arthritis that can only be evaluated over a relatively short time span due to deteriorating health of the MRL/Ipr mice beyond 18 weeks of age. More acute and aggressive arthritis models would be useful to clarify the role of MMP-8 in cartilage destruction, but these models are less predictive of chronic human rheumatoid arthritis.  129  Our data highlight the complex effects of different MMPs in arthritic disease. The roles of individual MMPs in arthritis remain poorly defined despite progress with genetic knockout mice, recently revealing contradictory roles of MMPs in disease progression and severity. In the case of the gelatinolytic MMPs, MMP-2 (gelatinase A) deficient mice had more severe clinical and histological arthritis compared with wild-types whereas the MMP-9 (gelatinase B) knockouts displayed milder arthritis (60). In contrast, MMP-3 deficient mice had no significant change in a collageninduced arthritis model (61). The MMP-13 over-expressing transgenic mouse showed  increased  destruction  of articular cartilage  reminiscent of human  osteoarthritis (62), but the role of MMP-13 at physiological levels is unknown. Global interpretation of these studies is difficult given the diverse nature of arthritis models that range from acute inflammation to spontaneous chronic inflammation as investigated here. Due to the short life span of mice, inflammatory changes are not always associated with the cartilage destruction that is typical in chronic human disease.  In regulating neutrophil apoptosis and clearance, MMP-8 exhibits a potent and novel anti-inflammatory role in chronic inflammation in the context of rheumatoid arthritis, thereby defining MMP-8 as an anti-target in the disease. A protective role of certain MMPs in joint inflammation is not unexpected in light of musculoskeletal side effects in clinical trials with broad-spectrum MMP inhibitors (63). Undoubtedly, it will be necessary to spare MMP-8 in the development of novel selective MMP inhibitors for treating chronic diseases requiring long-term dosing. MMP-8 has also emerged as an anti-target in cancer therapy (13, 18), but surprisingly has a detrimental effect in mouse models of multiple sclerosis where MMP-8 deficient mice have improved outcome (64). Clearly the disease and organ-dependent functions of MMP-8 need to be explored further, in combination with substrate discovery efforts, to decipher the physiological mechanisms and functions of this so-called collagenase, whose roles clearly extend beyond extracellular matrix degradation.  130  References 1.  Duan, H., T. Koga, F. Kohda, H. Hara, K. Urabe, and M. Furue. 2001. lnterleukin-8-positive neutrophils in psoriasis. J. Dermatol. Sd. 26:119-124.  2.  Mitsuyama, K., A. Toyonaga, E. Sasaki, K. Watanabe, H. Tateishi, T. Nishiyama, T. Saiki, H. Ikeda, 0. Tsuruta, and K. Tanikawa. 1994. IL-8 as an important chemoattractant for neutrophils in ulcerative colitis and Crohn’s disease. C/in. Exp. 1mm. 96:432-436.  3.  Chatham, W. W., R. Swaim, H. Frohsin, Jr., L. W. Heck, E. J. Miller, and W. D. Blackburn, Jr. 1993. Degradation of human articular cartilage by neutrophils in synovial fluid. Arthritis Rheum. 36:51-58.  4.  Mohr, W., H. Westerhellweg, and D. Wessinghage. 1981. Polymorphonuclear granulocytes in rheumatic tissue destruction. Ill. an electron microscopic study of PMNs at the pannus cartilage junction in rheumatoid arthritis. Ann. Rheum. Dis. 40:396-399.  5.  Wipke, B. T., and P. M. Allen. 2001. Essential role of neutrophils in the initiation and progression of a murine model of rheumatoid arthritis. J. Immunol. 167:1601-1608.  6.  Murphy, G., and H. Nagase. 2008. Reappraising metalloproteinases in rheumatoid arthritis and osteoarthritis: destruction or repair? Nat. C/in. Pract. Rheumatol. 4:128-135.  7.  Tchetverikov, I., H. K. Ronday, B. Van El, G. H. Kiers, N. Verzijl, J. M. TeKoppele, T. W. Huizinga, J. DeGroot, and R. Hanemaaijer. 2004. MMP profile in paired serum and synovial fluid samples of patients with rheumatoid arthritis. Ann. Rheum. Dis. 63:881-883.  8.  Hanemaaijer, R., T. Sorsa, Y. T. Konttinen, Y. Ding, M. Sutinen, H. Visser, V. W. van Hinsbergh, T. Helaakoski, T. Kainulainen, H. Ronka, H. Tschesche, and T. Salo. 1997. Matrix metalloproteinase-8 is expressed in rheumatoid synovial fibroblasts and endothelial cells. Regulation by tumor necrosis factor-alpha and doxycycline. J. BIoI. Chem. 272:31504-31509.  9.  Cox, J. H., and C. M. Overall. In press. Cytokine Substrates: MMP Regulation of Inflammatory Signalling Molecules. The Cancer Degradome Proteases and Cancer Biology. Springer Science and Business Media. -  10.  Parks, W. C., C. L. Wilson, and Y. S. Lopez-Boado. 2004. Matrix metalloproteinases as modulators of inflammation and innate immunity. Nat. Rev. Immunol. 4:617-629.  11.  Gearing, A. J., P. Beckett, M. Christodoulou, M. Churchill, J. Clements, A. H. Davidson, A. H. Drummond, W. A. Galloway, R. Gilbert, J. L. Gordon, and et al. 1994. Processing of tumour necrosis factor-alpha precursor by metal loproteinases. Nature 370:555-557.  12.  Dean, R. A., G. S. Butler, Y. Hamma-Kourbali, J. Delbe, D. R. Brigstock, J. Courty, and C. M. Overall. 2007. Identification of candidate angiogenic inhibitors processed by matrix metalloproteinase 2 (MMP-2) in cell-based proteomic screens: disruption of vascular endothelial growth factor (VEGF)/heparin affin regulatory peptide (pleiotrophin) and VEGF/Connective tissue growth factor angiogenic inhibitory complexes by MMP-2 proteolysis. MoL Cell. Biol. 27:8454-8465.  13.  Overall, C. M., and 0. Klelfeld. 2006. Tumour microenvironment opinion: validating matrix metalloproteinases as drug targets and anti-targets for cancer therapy. Nat. Rev. Cancer 6:227-239.  14.  Christodoulou, C., and E. H. Choy. 2006. Joint inflammation and cytokine inhibition in rheumatoid arthritis. C/in. Exp. Med. 6:13-19.  15.  Zwerina, J., K. Redlich, K. Polzer, L. Joosten, G. Kronke, J. Distler, A. Hess, N. Pundt, T. Pap, 0. Hoffmann, J. Gasser, C. Scheinecker, J. S. Smolen, W. van den Berg, and G. Schett. 2007. TN F-induced structural joint damage is mediated by IL-i. Proc. Nati. Acad. Sci. USA 104:11742-11747.  -  131  16.  Furlan, R., G. Martino, F. Galbiati, P. L. Poliani, S. Smiroldo, A. Bergami, G. Desina, G. Comi, R. Flavell, M. S. Su, and L. Adorini. 1999. Caspase-1 regulates the inflammatory process leading to autoimmune demyelination. J. Immunol. 163:2403-2409.  17.  Kang, S. J., S. Wang, H. Hara, E. P. Peterson, S. Namura, S. Amin-Hanjani, Z. Huang, A. Srinivasan, K. J. Tomaselli, N. A. Thornberry, M. A. Moskowitz, and J. Yuan. 2000. Dual role of caspase-1 1 in mediating activation of caspase-1 and caspase-3 under pathological conditions. J. Cell Biol. 149:613-622.  18.  Balbin, M., A. Fueyo, A. M. Tester, A. M. Pendas, A. S. Pitiot, A. Astudillo, C. M. Overall, S. D. Shapiro, and C. Lopez-Otin. 2003. Loss of collagenase-2 confers increased skin tumor susceptibility to male mice. Nat. Genet. 35:252-257.  19.  Tester, A. M., J. H. Cox, A. R. Connor, A. E. Starr, R. A. Dean, X. S. Puente, C. Lopez-Otin, and C. M. Overall. 2007. LPS responsiveness and neutrophil chemotaxis in vivo require PMN MMP-8 activity. PLoS ONE 2:e312.  20.  Ratkay, L. G., L. Zhang, J. Tonzetich, and J. D. Waterfield. 1993. Complete Freund’s adjuvant induces an earlier and more severe arthritis in MRL-lpr mice. J. Immunol. 151 :50815087.  21.  Singer, G. G., and A. K. Abbas. 1994. The fas antigen is involved in peripheral but not thymic deletion of T lymphocytes in T cell receptor transgenic mice. Immunity 1:365-371.  22.  Watanabe-Fukunaga, R., C. I. Brannan, N. G. Copeland, N. A. Jenkins, and S. Nagata. 1992. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356:314-317.  23.  Truett, G. E., P. Heeger, R. L. Mynatt, A. A. Truett, J. A. Walker, and M. L. Warman. 2000. Preparation of PCR-quality mouse genomic DNA with hot sodium hydroxide and tris (HotSHOT). BioTechniques 29:52, 54.  24.  English, D., and B. R. Andersen. 1974. Single-step separation of red blood cells. Granulocytes and mononuclear leukocytes on discontinuous density gradients of Ficoll Hypaque. J. Immunol. Methods 5:249-252.  25.  Billinghurst, R. C., L. Dahlberg, M. lonescu, A. Reiner, R. Bourne, C. Rorabeck, P. Mitchell, J. Hambor, 0. Diekmann, H. Tschesche, J. Chen, H. Van Wart, and A. R. Poole. 1997. Enhanced cleavage of type II collagen by collagenases in osteoarthritic articular cartilage. J. Clin. Invest. 99:1534-1545.  26.  Mort, J. S., and D. J. Buttle. 1999. The use of cleavage site specific antibodies to delineate protein processing and breakdown pathways. Mo!. PathoL 52:11-18.  27.  Sztrolovics, R., M. Alini, P. J. Roughley, and J. S. Mort. 1997. Aggrecan degradation in human intervertebral disc and articular cartilage. Biochem. J. 326 ( Pt 1 ):235-241.  28.  Kappelhoff, R., and C. M. Overall. 2007. The CLIP-CHIP Oligonucleotide Microarray: Dedicated Array for Analysis of All Protease, Nonproteolytic Homolog, and Inhibitor Gene Transcripts in Human and Mouse. Curr. Protoc. Protein Sd. 49:21.19.21-21.19.16.  29.  Saeed, A. I., V. Sharov, J. White, J. Li, W. Liang, N. Bhagabati, J. Braisted, M. Klapa, T. Currier, M. Thiagarajan, A. Sturn, M. Snuffin, A. Rezantsev, D. Popov, A. Ryltsov, E. Kostukovich, I. Borisovsky, Z. Liu, A. Vinsavich, V. Trush, and J. Quackenbush. 2003. TM4: a free, open-source system for microarray data management and analysis. Bio Techniques 34:374-378.  30.  Pfaffl, M. W. 2001. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 29:e45.  31.  Gutierrez-Femandez, A., M. Inada, M. Balbin, A. Fueyo, A. S. Pitiot, A. Astudillo, K. Hirose, M. Hirata, S. D. Shapiro, A. Noel, Z. Werb, S. M. Krane, C. Lopez-Otin, and X. S. Puente.  132  2007. Increased inflammation delays wound healing in mice deficient in collagenase-2 (MMP 8). FASEB J. 21:2580-2591. 32.  Ratkay, L. G., B. Tait, J. Tonzetich, and J. D. Waterfield. 1994. Lpr and MRL background gene involvement in the control of adjuvant enhanced arthritis in MRL-lpr mice. J. Autoimmun. 7:561-573.  33.  Fosang, A. J., K. Last, P. J. Neame, G. Murphy, V. Knauper, H. Tschesche, C. E. Hughes, B. Caterson, and T. E. Hardingham. 1994. Neutrophil collagenase (MMP-8) cleaves at the aggrecanase site E373-A374 in the intergiobular domain of cartilage aggrecan. Biochem. J. 304 ( Pt 2):347-351.  34.  Wang, S., M. Miura, Y. K. Jung, H. Zhu, E. Li, and J. Yuan. 1998. Murine caspase-11, an ICE-interacting protease, is essential for the activation of ICE. Cell 92:501-509.  35.  Carswell, E. A., L. J. Old, R. L. Kassel, S. Green, N. Fiore, and B. Williamson. 1975. An endotoxin-induced serum factor that causes necrosis of tumors. Proc. Nat!. Acad. Sct USA 72:3666-3670.  36.  Ruuls, S. R., R. M. Hoek, V. N. Ngo, T. McNeil, L. A. Lucian, M. J. Janatpour, H. Korner, H. Scheerens, E. M. Hessel, J. G. Cyster, L. M. McEvoy, and J. D. Sedgwick. 2001. Membranebound TNF supports secondary lymphoid organ structure but is subservient to secreted TNF in driving autoimmune inflammation. Immunity 15:533-543.  37.  Churg, A., R. D. Wang, H. Tal, X. Wang, C. Xie, J. Dai, S. D. Shapiro, and J. L. Wright. 2003. Macrophage metalloelastase mediates acute cigarette smoke-induced inflammation via tumor necrosis factor-alpha release. Am. J. Respir. Crit. Care Med. 167:1083-1089.  38.  Haro, H., H. C. Crawford, B. Fingleton, K. Shinomiya, D. M. Spengler, and L. M. Matrisian. 2000. Matrix metalloproteinase-7-dependent release of tumor necrosis factor-alpha in a model of herniated disc resorption. J. Clin. Invest. 105:143-1 50.  39.  Black, R. A., C. T. Rauch, C. J. Kozlosky, J. J. Peschon, J. L. Slack, M. F. Wolfson, B. J. Castner, K. L. Stocking, P. Reddy, S. Srinivasan, N. Nelson, N. Boiani, K. A. Schooley, M. Gerhart, R. Davis, J. N. Fifzner, R. S. Johnson, R. J. Paxton, C. J. March, and D. P. Cerretti. 1997. A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells. Nature 385:729-733.  40.  Moss, M. L., S. L. Jin, M. E. Milla, D. M. Bickett, W. Burkhart, H. L. Carter, W. J. Chen, W. C. Clay, J. R. Didsbury, D. Hassler, C. R. Hoffman, T. A. Kost, M. H. Lambert, M. A. Leesnilzer, P. McCauley, G. McGeehan, J. Mitchell, M. Moyer, G. Pahel, W. Rocque, L. K. Overton, F. Schoenen, T. Seaton, J. L. Su, J. D. Becherer, and et al. 1997. Cloning of a disintegrin metalloproteinase that processes precursor tumour-necrosis factor-alpha. Nature 385:733736.  41.  Hisahara, S., J. Yuan, T. Momoi, H. Okano, and M. Miura. 2001. Caspase-li mediates oligodendrocyte cell death and pathogenesis of autoimmune-mediated demyelination. J. Exp. Med. 193:111-122.  42.  Kang, S. J., S. Wang, K. Kuida, and J. Yuan. 2002. Distinct downstream pathways of caspase-1 1 in regulating apoptosis and cytokine maturation during septic shock response. Cell Death Differ. 9:1115-1125.  43.  Lee, J., J. Hur, P. Lee, J. Y. Kim, N. Cho, S. Y. Kim, H. Kim, M. S. Lee, and K. Suk. 2001. Dual role of inflammatory stimuli in activation-induced cell death of mouse microglial cells. Initiation of two separate apoptotic pathways via induction of interferon regulatory factor-i and caspase-1 1. J. Biol. Chem. 276:32956-32965.  44.  Colotta, F., F. Re, N. Polentarutti, S. Sozzani, and A. Mantovani. 1992. Modulation of granulocyte survival and programmed cell death by cytokines and bacterial products. Blood 80:2012-2020.  133  45.  Gueders, M. M., M. Balbin, N. Rocks, J. M. Foidart, P. Gosset, R. Louis, S. Shapiro, C. Lopez-Otin, A. Noel, and D. D. Cataldo. 2005. Matrix metalloproteinase-8 deficiency promotes gran ulocytic allergen-induced airway inflammation. J. Immunol. 175:2589-2597.  46.  Giudice, E. D., A. Ciaramella, N. Balestro, D. Neumann, P. G. Romano, M. P. Cesaroni, G. Maurizi, P. Ruggiero, D. Boraschi, and P. Bossu. 2001. Neutrophil apoptosis in autoimmune Fas-defective MRL lpr/lpr mice. Eur. Cytokine. NeIw. 12:510-517.  47.  O’Neill, A., M. C. Greenan, B. Doyle, J. M. Fitzpatrick, and R. W. Watson. 2004. Gene profiling of in vitro and in vivo models of delayed neutrophil apoptosis: a common pathway? Biochem. Soc. Trans. 32:470-473.  48.  Riedl, S. J., and G. S. Salvesen. 2007. The apoptosome: signalling platform of cell death. Nat. Rev. Mol. Cell Biol. 8:405-413.  49.  Pongracz, J., P. Webb, K. Wang, E. Deacon, 0. J. Lunn, and J. M. Lord. 1999. Spontaneous neutrophil apoptosis involves caspase 3-mediated activation of protein kinase C-delta. J. Biol. Chem. 274:37329-37334.  50.  Powell, W. C., B. Fingleton, C. L. Wilson, M. Boothby, and L. M. Matrisian. 1999. The metalloproteinase matrilysin proteolytically generates active soluble Fas ligand and potentiates epithelial cell apoptosis. Curr. Biol. 9:1441-1447.  51.  Fingleton, B., T. Vargo-Gogola, H. C. Crawford, and L. M. Matrisian. 2001. Matrilysin [MMP 7] expression selects for cells with reduced sensitivity to apoptosis. Neoplasia 3:459-468.  52.  Schauvllege, R., J. Vanrobaeys, P. Schotte, and R. Beyaert. 2002. Caspase-li gene expression in response to lipopolysaccharide and interferon-gamma requires nuclear factorkappa B and signal transducer and activator of transcription (STAT) 1. J. Blot Chem. 277:41624-41630.  53.  Martinon, F., K. Burns, and J. Tschopp. 2002. The inflammasome: a molecular platform triggering activation of inflammatory caspases and processing of prolL-beta. Mol. Cell 10:417-426.  54.  Li, J., W. M. Brieher, M. L. Scimone, S. J. Kang, H. Zhu, H. Yin, U. H. von Andrian, T. Mitchison, and J. Yuan. 2007. Caspase-1 1 regulates cell migration by promoting Aipi-Cofihin mediated actin depolymerization. Nat Cell Biol. 9:276-286.  55.  Mueller, N. J., R. A. Wilkinson, and J. A. Fishman. 2002. Listeria monocytogenes infection in caspase-1 1-deficient mice. infect. Immun. 70:2657-2664.  56.  Feldmann, M., and R. N. Maini. 2001. Anti-TNF alpha therapy of rheumatoid arthritis: what have we learned? Annu. Rev. Immunol. 19:163-196.  57.  Vincenti, M. P., and C. E. Brinckerhoff. 2002. Transcriptional regulation of collagenase (MMP 1, MMP-13) genes in arthritis: integration of complex signaling pathways for the recruitment of gene-specific transcription factors. Arthritis Res. 4:157-164.  58.  Takaishi, H., T. Kimura, S. Dalal, Y. Okada, and J. D’Armiento. 2008. Joint diseases and matrix metalloproteinases: a role for MMP-13. Curr Pharm Biotechnol 9:47-54.  59.  O’Sullivan, F. X., H. G. Fassbender, S. Gay, and W. J. Koopman. 1985. Etiopathogenesis of the rheumatoid arthritis-like disease in MRL/l mice. I. The histomorphologic basis of joint destruction. Arthritis Rheum. 28:529-536.  60.  Itoh, T., H. Matsuda, M. Tanioka, K. Kuwabara, S. Itohara, and R. Suzuki. 2002. The role of matrix metalloproteinase-2 and matrix metalloproteinase-9 in antibody-induced arthritis. J. lmmunol. 169:2643-2647.  61.  Mudgett, J. S., N. I. Hutchinson, N. A. Chartrain, A. J. Forsyth, J. McDonnell, Singer, II, E. K. Bayne, J. Flanagan, D. Kawka, C. F. Shen, K. Stevens, H. Chen, M. Trumbauer, and D. M. Visco. 1998. Susceptibility of stromelysin 1-deficient mice to collagen-induced arthritis and cartilage destruction. Arthritis Rheum. 41:110-121.  134  62.  Neuhold, L. A., L. Killar, W. Zhao, M. L. Sung, L. Warner, J. Kulik, J. Turner, W. Wu, C. Billinghurst, T. Meijers, A. R. Poole, P. Babij, and L. J. DeGennaro. 2001. Postnatal expression in hyaline cartilage of constitutively active human collagenase-3 (MMP-13) induces osteoarthntis in mice. J. Clin. Invest. 107:35-44.  63.  Coussens, L. M., B. Fingleton, and L. M. Matrisian. 2002. Matrix metalloproteinase inhibitors and cancer: trials and tribulations. Science 295:2387-2392.  64.  Folgueras, A. R., A. Fueyo, 0. Garcia-Suarez, J. Cox, A. Astudillo, P. Tortorella, C. Campestre, A. Gutierrez-Fernandez, M. Fanjul-Fernandez, C. J. Pennington, D. R. Edwards, C. M. Overall, and C. Lopez-Otin. 2008. Collagenase-2 deficiency or inhibition impair experimental autoimmune encephalomyelitis in mice. J. Biol. Chem.  135  CHAPTER 5. CONCLUSIONS AND PERSPECTIVES Dynamic Regulation of Inflammation and Immunity by MMP-8 The research presented in this thesis expands on our previous knowledge of the functions of MMPs and the mechanisms by which MMPs regulate inflammation and immunity, with a particular focus on MMP-8, the neutrophil collagenase. The most significant findings of this thesis research are (i) identifying for the first time and characterizing a feed-forward neutrophil recruiting mechanism by MMP-8 regulation of mouse and human neutrophil chemoattractants, (ii) biochemically identifying a novel MMP chemokine substrate, the T cell chemoattractant CXCL1 1, and evaluating the functional consequences of truncation including a novel proteolytic effect on glycosaminoglycan binding and hence dispersal of the chemokine gradient, and (iii) defining MMP-8 as an anti-target in arthritis and identifying potential mechanisms by which MMP-8 promotes neutrophil apoptosis and the resolution of inflammation.  Taken together, this thesis highlights the diverse roles of a single  protease in the context of different immunological processes, encompassing the innate, adaptive, and autoimmune scenarios. Once considered to be simple matrix degrading enzymes, MMP5 are now known to have pleiotropic roles in normal homeostasis and pathological processes by acting through a plethora of bioactive substrates. As such, defining suitable MMP drug targets has become a challenging task.  In the case of MMP-8, the data from air  pouch models suggest that MMP-8 inhibition could be beneficial in dampening acute inflammation.  In collaboration work not presented in this thesis, we showed that  MMP-8 was detrimental in experimental autoimmune encephalomyelitis (EAE), a murine model of multiple sclerosis (1).  In contrast, MMP-8 was found to be  protective in rheumatoid arthritis, a chronic inflammatory disease.  Furthermore, I  propose that MMP-8, in conjuction with other MMPs, functions to downregulate Thi cell chemoattraction and modulate acquired immunity.  Therefore, the risks and  benefits of targeting a multi-functional protease such as MMP-8 must be carefully assessed and balanced. Protective roles in cancer and arthritis define MMP-8 as an  136  anti-target in these diseases and hence MMP-8 must be spared in the design of novel MMP inhibitors for clinical use. Despite the large number of MMP family members, which is 23 in human and 24 in mouse, there exists a remarkable lack of redundancy in vivo. Nearly every MMP deficient mouse generated to date has one or more phenotypes relating to inflammatory or immunological processes, as reviewed in Chapter 1. The neutrophil alone produces MMP-8, MMP-9, MMP-14, and MMP-25. Although several MMPs, including MMP-9 and MMP-14, biochemically process and activate the neutrophil chemoattractant LIX, MMP-8 is the dominant physiological regulator of LIX activity. This shows that while there is biochemical redundancy, there is less physiological redundancy than expected.  In additon, four MMPs cleaved the T lymphocyte  chemokine CXCLI 1/1-TAC but only MMP-8 cleaved within the C-terminal alphahelix, resulting in the loss of glycosaminoglycan affinity. In the absence of MMP-8, MMP-9 is considerably upregulated in the neutrophil, yet this does not compensate for the more severe arthritis observed in MMP-8 deficient male mice nor the delayed neutrophil apoptosis in the absence of MMP-8. Therefore, in spite of an apparently redundant system, the data presented herein supports a unique and complex role for MMP-8 as a pivotal regulator in inflammation and immunity. This research has used a wide variety of techniques to explore the biological functions of MMP-8, including biochemical analyses with mass spectrometry, cell culture, gene microarrays, and mouse models of inflammation. This multifaceted approach has proven very successful in viewing MMPs from different perspectives, at both the molecular and physiological levels, and hence gaining valuable insight into both mechanism and function. Biochemical tools, such as the MALDI-TOF MS analysis that I performed, aided in the discovery of new MMP-8 substrates, specifically LIX, CXCL5, CXCL8, CXCLI 1, and TNF-a.  Isolation and culture of  primary cells enabled the characterization of functional consequences of proteolysis of bioactive substrates.  Furthermore, gene microarrays revealed potential  mechanisms by which MMP-8 controls neutrophil accumulation and resolution.  137  Finally, the MMP-8 knockout mouse and the novel MRL/Ipr MMP-8 deficient strain that I generated by backcross breeding were instrumental in deciphering the physiological roles of MMP-8 in acute and chronic inflammation. Bridging the gap between what happens in a test tube and what occurs in an animal is far from trivial.  This is exemplified in Chapters 2 and 3 which describe  biochemical cleavages that resulted in either chemokine activation or switching a receptor agonist to an antagonist, respectively.  A direct link between MMP-8  mediated LIX activation and decreased neutrophil infiltration in MMP-8 deficient mice seemed obvious at first.  However, some key pieces to the puzzle were missing.  Most notably, the MMP processed forms of LIX were never detected in physiological samples although decreased cleavage was evident in MMP-8 deficient mice (2). More recent findings, as described in Appendix A, have prompted the hypothesis that MMP-8 is regulating LIX activation in an indirect manner.  Demonstrating  evidence of in vivo processing of CXCL1 1 has been unsuccessful to date, despite the raising of sensitive neo-epitope antibodies specific for the MMP processed forms of the chemokine. Many biochemical cleavage events are in vitro artefacts that do not occur physiologically, and must be considered when designing experiments. In hindsight, studying chemokine processing at both the biochemical and cellular level would provide a more relevant starting point for the identification of physiological substrates. Animal models of disease are without doubt extremely useful research tools. However, no model is perfect in predicting human disease. This is especially true for mouse models of arthritis. There is no consensus on the best murine model of rheumatoid arthritis, which range from acute and severe collagen-induced arthritis, serum transfer methods, and the milder MRL/Ipr systemic autoimmune model that was described in Chapter 4.  With researchers using such an array of models,  comparison across studies becomes increasingly difficult.  The MRLIIpr model of  autoimmunity provided a strong inflammatory response in the mouse joints, which was evident by clinical edema and histological analysis. However, this model did not  138  yield detectable cartilage damage, a hallmark of rheumatoid arthritis. As such, using a single model to study such a complex disease is a limitation and future arthritis studies should use multiple models to strengthen conclusions and assess different aspects of disease pathology. Over the past five years, conflicting results regarding the role of MMP-8 in inflammatory disease have emerged (Table 1.2). In acute situations, such as the LPS-treated air pouch (Chapter 2), TNF-induced lethal hepatitis (3), and LPS-treated cornea (4),  MMP-8 assumes a pro-inflammatory function with knockout mice  exhibiting decreased levels of neutrophils.  This phenotype is reversed in lung  models of acute inflammation where MMP-8 deficient mice have elevated infiltration of neutrophils (5, 6), demonstrating tissue-dependent effects.  In contrast, chronic  inflammation such as that in skin carcinogenesis (2), wound healing (7), and rheumatoid arthritis (Chapter 4), reveal an anti-inflammatory role for MMP-8, where MMP-8 knockout mice are protected from disease. Based on this research and that of others, I currently hypothesize that MMP-8 is acting through multiple pathways to promote neutrophil migration in acute inflammation and to resolve inflammation during the transition for acute to chronic inflammation (Figure 5.1). In acute inflammation, MMP-8 is released from stimulated neutrophils to cleave and inactivate ctl-proteinase inhibitor, a serine protease inhibitor. This enhances the activity of neutrophil elastase which can efficiently cleave and activate LIX, a potent murine neutrophil chemoattractant. mechanism is depicted in Figure A.7 (Appendix A).  This  In the normal resolution of  inflammation, I hypothesize that MMP-8 promotes neutrophil apoptosis and clearance through induction of caspase-1 1 expression and TNF-a shedding (Figure 4.9). Although adaptive immune processes have not been specifically studied in the MMP-8 knockout mouse, I propose that MMP-8 and other MMPs can process CXCLI 1 and convert it into a CXCR3 antagonist Thi lymphocyte chemokine, alter its  haptotactic  gradient  localization  on  glycosaminoglycans,  and  ultimately  downregulate Thi lymphocyte recruitment (Figure 3.10).  139  PRO-INFLAMMA TORY ROLE IN ACUTE SETTING  INACTIVATION OF al-PI  LIX ACTIVATION  Promotes NE activation of LIX  reased neutrophil recruitment  TNF-a SHEDDING Increases ne utrophi I apoptosis  /  fr  CXCLII INACTIVATION Decreases Thi chemoattraction  YET TO BE DISCOVERED  CASPASE-1 I EXPRESSION Enhances caspase-3 activity and promotes neutrophil apoptosis  PROMO TES THE RESOLUTION OF INFLAMMATION  Figure 5.1. Solving the MMP-8 puzzle: diverse functions of MMP-8 in inflammation and acquired immunity. MMP-8 is released from specific granules of activated neutrophils resulting in many proteolytic functions. To promote neutrophil recruitment in acute inflammation, MMP-8 either activates the neutrophil chemoattractant LIX directly or increases neutrophil elastase (NE) activation of LIX via inactivation of al-proteinase inhibitor (al-PI). Alternatively, MMP-8 promotes the normal resolution of inflammation through TNF-a shedding and the regulation of caspase-1 1 expression, pathways proposed to converge and induce neutrophil apoptosis. TNF-a shedding results in increased MMP expression (not shown). MMP-8 also processes the Thi chemoattractant CXCL1 1 yielding a receptor antagonist with predicted reduction of Thi lymphocyte recruitment in vivo. The discovery of additional physiologically relevant substrates will be essential in fully understanding the complex and multifaceted functions of MMP-8 in neutrophil physiology and inflammation.  140  Future research stemming directly from these findings should include examination of MMP processing of CXCLI I in vivo as well as further characterizing the pro apoptotic effect of MMP-8. CXCLI 1 processing could be compared in normal and diseased tissues using mass spectrometry and neo-epitope antibody-based detection.  Specifically, due to an abundance of neutrophils and T lymphocytes,  tissues from autoimmunities such as multiple sclerosis, psoriasis, and arthritis would be of particular interest. Experiments to decipher the mechanism for the regulatory interaction between MMP-8 and caspase-1 I should focus on developmental aspects. Studying leukocytes from mice of various ages could provide insight into this mechanism. Furthermore, examining caspase-1 1 levels in hematopoietic stem cells and myeloid progenitor cells from MMP-8 deficient mice could reveal the basis of this regulatory effect. Ultimately, experiments should be aimed at discovering and validating substrates through which MMP-8 promotes neutrophil apoptosis. The homeostatic and pathological functions of proteases are defined by the substrates they cleave and the biological consequences of these post-translational modifications. Until recently, substrate discovery efforts have been largely biased and limited by the classical hypothesis-driven approach.  However, novel mass  spectrometry based methods for substrate discovery are emerging with far-reaching potential to revolutionize protease research, providing systematic and unbiased routes to substrate identification (8).  Although numerous bioactive MMP-8  substrates have been identified biochemically (Table 1.1), none have been validated  in vivo, which is a challenging task due to the low abundance of many substrates such as chemokines. This highlights the necessity for multiple levels of substrate validation, which can be achieved by demonstrating biochemical, cellular, animal model, and ultimately human relevance (9).  The use of MMP-8 knockout mice,  primary neutrophils, or MMP-8 deficient cell systems in conjunction with proteomic substrate discovery techniques will validate proposed mechanisms and potentially uncover novel pathways by which MMP-8 coordinates both the initiation and resolution of inflammation. One such technique is described in Appendix B.  141  Data relating the effects of MMPs in mouse models with those in human disease are currently lacking. Fundamental differences between mouse and human MMPs are evident by the increased number of MMPs, and all proteases, in mice (10) and also from failed MMPI clinical trials, despite promising efficacy in animals (11). Such differences should be assessed at the molecular level by comparing cleavage site specificities (described in Appendix B), kinetics of substrate cleavage, and inhibition profiles for individual mouse and human proteases. This should then be expanded upon using human cell lines in conjunction with small interfering RNA and specific inhibitors, and ultimately correlation studies with disease severity should be examined in humans, as described recently for MMP-8 in cancer metastasis (12). The majority of homeostatic and pathogenic pathways appear to be conserved across species, although specific mediators may differ. However, these differences must be appreciated and clearly defined. In conclusion, this thesis has demonstrated the pivotal role of MMP-8 in both the initiation and resolution of inflammation.  Once thought to be simply a collagen-  degrading protease, MMP-8 is emerging as a multifaceted regulator of neutrophil recruitment and clearance with protective roles in arthritis and cancer. MMPs are now known to act through a wide variety of bioactive molecules and function as key regulators in numerous physiological processes, including inflammation and immunity.  As our knowledge of the mechanisms and physiological functions of  MMPs continues to expand, there is the potential to uncover novel therapeutic strategies for the many diseases that involve dysregulated inflammation.  142  References 1.  Folgueras, A. R., A. Fueyo, 0. Garcia-Suarez, J. Cox, A. Astudillo, P. Tortorella, C. Campestre, A. Gutierrez-Fernandez, M. Fanjul-Fernandez, C. J. Pennington, D. R. Edwards, C. M. Overall, and C. Lopez-Otin. 2008. Collagenase-2 deficiency or inhibition impairs experimental autoimmune encephalomyelitis in mice. J Biol Chem 283:9465-9474.  2.  Balbin, M., A. Fueyo, A. M. Tester, A. M. Pendas, A. S. Pitiot, A. Astudillo, C. M. Overall, S. D. Shapiro, and C. Lopez-Otin. 2003. Loss of collagenase-2 confers increased skin tumor susceptibility to male mice. Nat Genet 35:252-257.  3.  Van Lint, P., B. Wielockx, L. Puimege, A. Noel, C. Lopez-Otin, and C. Libert. 2005. Resistance of collagenase-2 (matrix metalloproteinase-8)-deficient mice to TN F-induced lethal hepatitis. J Immunol 175:7642-7649.  4.  Lin, M., P. Jackson, A. M. Tester, E. Diaconu, C. M. Overall, J. E. Blalock, and E. Pearlman. 2008. Matrix Metalloproteinase-8 Facilitates Neutrophil Migration through the Comeal Stromal Matrix by Collagen Degradation and Production of the Chemotactic Peptide Pro-Gly  Pro. Am J Pathol. 5.  Gueders, M. M., M. Balbin, N. Rocks, J. M. Foidart, P. Gosset, R. Louis, S. Shapiro, C. Lopez-Otin, A. Noel, and D. D. Cataldo. 2005. Matrix metalloproteinase-8 deficiency promotes gran ulocytic allergen-induced airway inflammation. J immunol 175:2589-2597.  6.  Owen, C. A., Z. Hu, C. Lopez-Otin, and S. D. Shapiro. 2004. Membrane-bound matrix metalloproteinase-8 on activated polymorphonuclear cells is a potent, tissue inhibitor of metalloproteinase-resistant collagenase and serpinase. J Immunol 172:7791-7803.  7.  Gutierrez-Fernandez, A., M. lnada, M. Balbin, A. Fueyo, A. S. Pitiot, A. Astudillo, K. Hirose, M. Hirata, S. D. Shapiro, A. Noel, Z. Werb, S. M. Krane, C. Lopez-Otin, and X. S. Puente. 2007. Increased inflammation delays wound healing in mice deficient in collagenase-2 (MMP  8). Faseb J. 8.  Schilling, 0., and C. M. Overall. 2007. Proteomic discovery of protease substrates. Curr Opin Chem Biol 11:36-45.  9.  Overall, C. M., and C. P. BIobel. 2007. In search of partners: linking extracellular proteases to substrates. Nat Rev Mo! Cell Biol 8:245-257.  10.  Puente, X. S., L. M. Sanchez, C. M. Overall, and C. Lopez-Otin. 2003. Human and mouse proteases: a comparative genomic approach. Nat Rev Genet4:544-558.  11.  Coussens, L. M., B. Fingleton, and L. M. Matrisian. 2002. Matrix metalloproteinase inhibitors and cancer: trials and tribulations. Science 295:2387-2392.  12.  Gutierrez-Fernandez, A., A. Fueyo, A. R. Folgueras, C. Garabaya, C. J. Pennington, S. Pilgrim, D. R. Edwards, D. L. Holliday, J. L. Jones, P. N. Span, F. C. Sweep, X. S. Puente, and C. Lopez-Otin. 2008. Matrix metalloproteinase-8 functions as a metastasis suppressor through modulation of tumor cell adhesion and invasion. Cancer Res 68:2755-2763.  143  APPENDIX A. PROCESSING OF ELR 4 CHEMOKINES BY NEUTROPHIL SERINE 5 PROTEASES Perspective The manuscript in Chapter 2 entitled “LPS responsiveness and neutrophil chemotaxis in vivo require PMN MMP-8 activity” describes the pro-inflammatory role of MMP-8 in neutrophil infiltration in acute immunity. In a subcutaneous air pouch model, MMP-8 deficient mice had reduced neutrophil migration in response to LPS. This was attributed to MMP-8 processing and activation of the murine ELR chemokine LIX.  Indeed, air pouch experiments with the activated form of LIX  showed equal migration of neutrophils in wild-type and MMP-8 knockout mice, validating LIX as a physiological MMP-8 dependent mechanism for promoting neutrophil infiltration.  This demonstrated that there was no defect in the PMN  chemotactic machinery and that MMP-8 cleavage of the extracellular matrix is not required for migration. Neo-epitope antibodies specific for the MMP-8 processed forms of LIX were generated, but these twncations were not detected in air pouch fluids, thereby prompting the examination of alternate MMP-8 dependent pathways for LIX activation and neutrophil recruitment. In this Appendix, efficient neutrophil elastase processing of LIX is demonstrated and MMP-8 cleavage and regulation of al-PI, a potent neutrophil elastase inhibitor, is proposed. By this indirect pathway, MMP-8 might increase neutrophil elastase activity and hence increase the proteolytic activation of LIX, thereby recruiting neutrophils in a feed-forward mechanism.  The studies presented in Appendix A are unpublished and expand on the published manuscript in Chapter 2. 144  Results Processing of LIX by isolated bone marrow neutrophils MMP-8 processing of LIX was studied in the cellular context with bone marrow neutrophils from wild-type and MMP-8 deficient mice. Cleavage was analyzed by Tris-tricine gel electrophoresis and MALDI-TOF mass spectrometry.  Processing  occurred after only a few hours with either wild-type or MMP-8 deficient neutrophils (Figure A.IA, B). Deconvolution of MALDI-TOF data identified cleavage products as LIX (9-92), LIX (1-78), and LIX (9-78) with mass to charge ratios (mlz) of 9150, 8325, and 7626 [M+H], respectively (Figure A.IC). Therefore, neutrophil-mediated proteolysis of LIX occurs at 11 IAA-TEL and 76 6 81 (Figure A.ID). These KKA-KRN cleavage products are expected to have enhanced activity based on previous findings with ELR chemokines (1, 2). Therefore, neutrophil proteases are able to process and activate LIX and this is not MMP-8 dependent in the cellular context. LIX is processed by serine proteases neutrophil elastase and cathepsin G To assess the neutrophil protease(s) responsible for LIX processing, protease inhibitors were added to the cleavage mixture.  The serine protease inhibitor 2-  aminoethyl benzenesulfonyl fluoride hydrochloride (AEBSF) was an effective inhibitor of LIX cleavage (Figure A.2A). This data indicates that LIX is processed by at least one of the three major serine proteases in neutrophils: proteinase-3, cathepsin G, and neutrophil elastase. The mechanism of serine protease mediated LIX cleavage was examined further using endogenous serine protease inhibitors (Figure A.2A). At concentrations as low as 100 nM, both al-proteinase inhibitor (al P1) and secreted leukocyte protease inhibitor (SLPI) effectively inhibited LIX cleavage by neutrophils. It is known that SLPI does not inhibit proteinase-3 (3), whereas both al-PI and SLPI inhibit neutrophil elastase and cathepsin G (4). Therefore, potential candidates for neutrophil-mediated LIX processing were neutrophil elastase and cathepsin G.  145  A Time(h) LIX PMA ( iM) 1  3 -  -  -  1  1  2  2  3  3  +  ÷  +  +  +  +  +  0.05  10  0.05  10  0.05  10  1  1  2  2  3  3  +  +  +  +  +  +  +  -  0.05  10  0.05  10  0.05  10  -  WT PMNs [M+HJ  B Time(h) LIX PMA ( iM) 1  3  -  -  KO PMNs [M÷H]  C  D  Chemokine LIX (1-92) LIX (9-92) LIX (7-78) LIX (9-78) 1  10  mlz [M÷H] Measured Predicted 9849 9850 9150 9153  8325  8325 7628  7626 80  90  APSSVIAATELR...SDKKKAtKRNALAVERTASVQ Figure A.1. Neutrophil processing of LIX is independent of MMP-8. Bone marrow neutrophils were isolated from wild-type and MMP-8 deficient mice then incubated with LIX. Reactions contained 1 x 106 cells, activated with PMA as marked, and 10 tg LIX in RPMI and were incubated at 37 °C for up to 3 hours. Cleavage with wild-type (A) and MMP-8 knockout (B) neutrophils was visualized by 15% Tris-Tricine gel electrophoresis and analyzed by MALDI-TOF MS in sinapinic acid matrix. (C) Mass to charge ratios ([M-i-H]) were deconvoluted to determine truncation products (italics). (D) N-terminal cleavage at 9 -Thr and the C-terminal 8 Ala cleavage at A1a 79 are depicted with arrows in the LIX sequence. -Lys 78 146  A  AEBSF  Inhib (tM)  WTPMNs  -  -  SLPI  100  500  0.1  0.5  0.1  0.5  +  ÷  +  +  +  +  -1-  -  al-PI  LIX (192)—  IEH* LIX  0  0 0 0 0  r. -  ‘-  0 0 0 -  0 0 0  0 0  •  0 0  L  1-  0 tf)  ‘-  -  1  9-78  LIX (1-92)  *[  MMP-8  (—LIX (1-92) 4—LIX (1-79)  CatG  LIX (1-92)  NE  —  I——  LIX (7-78)  C  D  Enzyme  ) s 1 kcat/KM (M-  MMP-8 CatG NE  6.Ox 102 6.Ox 101 4 12x10  MMP-8  NE  APSS’VI’’AATELRCVC... NE *  80  90  ...GSDKKKAKRNALAVERTASVQ NE MMP-8,CATG  Figure A.2. Neutrophil-derived serine proteases cleave LIX. Bone marrow neutrophils were isolated from wild-type mice then incubated with LIX. I x 106 cells were activated with 50 nM PMA and preincubated with AEBSF, al-PI, or SLPI before the addition of 10 g LIX. Reactions were incubated at 37 °C for 4 h. (A) Cellfree conditioned media was visualized on 15% Tris-Tricine SDS-PAGE and analyzed by MALDI-TOF MS in sinapinic acid matrix to determine truncation products. (B) Biochemical cleavage efficiencies of MMP-8, cathepsin G (CatG; Calbiochem) and neutrophil elastase (NE; Elastin Products) for LIX were assessed by titration. Enzyme-substrate ratios (E:S) are mol:mol. (C) Specificity constants (kcat/KM) were determined by densitometry. (D) MMP-8, CatG, and NE cleavage sites in LIX are depicted. Asterisks mark the sites of neutrophil-mediated cleavage.  147  Human neutrophil elastase and cathepsin G were tested in vitro for cleavage of LIX, and directly compared with MMP-8, which is known to cleave LIX biochemically (Figure A.2B). MMP-8 processed LIX at 5 -Va1 and Lys 4 Ser , as determined 80 -Arg 79 previously, with cleavage starting at an enzyme concentration of 20 nM. Cathepsin G also cleaved LIX at Lys , but only at concentrations above 100 nM. 80 -Arg 79  In  contrast, neutrophil elastase was an efficient processor of LIX at concentrations as low as 1 nM, resulting in cleavages at 7 -A1a AIa 6 11e , , and Leu 79 -Lys 78 . By 84 -AIa 83 densitometry analysis as described previously (5), specificity constants (kcat/KM) of LIX cleavage were estimated to be 600, 60 and 1200 M s for MMP-8, cathepsin G, 1 and neutrophil elastase, respectively (Figure A.2C). Notably, only the C-terminal 79 cleavage -Lys 78 A1a  by neutrophil elastase matches a  truncation of LIX (Figure A.2D).  neutrophil-mediated  The N-terminal A1a 9 8 Thr cleavage was not  reproduced by any of the enzymes tested, however these were human enzymes and murine counterparts could have slightly shifted specificity. Nonetheless, based on the kinetic efficiency of neutrophil elastase cleavage of LIX and the inhibition profile in neutrophil-mediated LIX cleavage, neutrophil elastase is the best candidate for LIX processing and activation in vivo. In vitro cleavage of mouse ELR chemokines by neutrophil elastase We tested whether neutrophil elastase cleavage of LIX was a specific interaction or if other murine ELR chemokines are processed.  Incubations containing 20 nM  enzyme and 10 1 iM chemokine (E:S of 1:500) revealed that KC and MIP-2 are resistant to cleavage and that DCIP-1 is degraded by neutrophil elastase (Figure A.3A). As expected, LIX was efficiently and specifically cleaved, generating LIX (783) and LIX (7-78), corresponding to [M+H] values of 8352 and 7769, respectively. Therefore, LIX is the only murine neutrophil chemoattractant that is precisely cleaved and activated by neutrophil elastase. In vitro cleavage of human ELR+ chemokines by neutrophil elastase Human ELR chemokines were investigated for neutrophil elastase cleavage to assess whether homologous chemokine activation mechanisms exist in human. At enzyme substrate ratios of 1:500 (mol:mol), neutrophil elastase cleaved GRO-y at 148  A  DCIP-1 NE  -  KC  +  -  LIX +  MIP-2 +  -  +  -  ÷  —  *  -  ,\c  B Chemokine DCIP-1 (1-73) KG (1-72) LIX (1-92) LIX (7-83) LIX (7-78) MIP-2 (1-73)  -  -  3cb  ,  ib4  [M+H]  m/z [M+HJ Measured Predicted 7911 7812 9852 8352 7769 7844  7911 7812 9850 8353 7770 7847  Figure A.3. Neutrophil elastase processing of murine ELR ’ chemokines. 4 Chemokines were incubated for 16 h at 37 °C in 50 mM Tris, 200 mM NaCI, 5 mM , pH 7.4. Enzyme to substrate ratio was 1:500 (mol:mol). (A) Reactions 2 CaCI were visualized by 15% Tris-Tricine gel electrophoresis (silver stained) and analyzed by MALDI-TOF MS in sinapinic acid matrix. * signifiies loss of band intensity and protein degradation. (B) Mass to charge ratios ([M+H]) were deconvoluted to determine truncation products (italics).  149  position , 5 4 Va1 Thr but left GRO-a and GRO-3 intact (Figure A.4A, B). Like DCIP-1, ENA-78 was also largely degraded by neutrophil elastase, but a peak at [M+H] of 7862 was detected, which corresponds to a cleavage at 8 -Leu (Figure A.4A, B). 7 Va1 NAP-2 was partially cleaved in the C-terminus at residues Val ° and IL-8 was 6 -G1n 59 resistant to cleavage at these conditions. The cleaved sequences reveal a strong preference for Valine in P1 (Figure A.4C), as reported previously in peptide library screens (6).  MMP-8 processing of al-proteinase inhibitor MMPs have been previously reported to cleave and inactivate al-PI (7, 8) and modulate the activity of serine proteases physiologically through this mechanism (9). Other naturally occurring neutrophil elastase inhibitors SLPI and elafin are reported to be resistant to MMP-8 processing (10). MMP-8 efficiently cleaves the 54-kDa al P1 to generate the 50-kDa inactive form (Figure A.5A), with processing evident at enzyme-substrate ratios as low as 1:5000 (w:w). A time course of MMP-8 proteolysis of al-PI (Figure A.5B) enabled the estimation of a specificity constant (kcat/KM) of 7.7 x  s from densitometry analysis. 1 M  Altered processing of LIX in BALF of MMP-8 deficient mice Several mouse tissues and fluids were examined for the presence of LIX and MMP 8 truncated LIX including air pouch fluid, corneal extracts, and arthritic joints with no success (data not shown). Previous work demonstrated decreased proteolysis in bronchoalveolar lavage fluid (BALF) from MMP-8 deficient mice and therefore processing was attributed to MMP-8 (11).  Likewise, in the present work we  observed reduced proteolysis of LIX in MMP-8 deficient lungs following 72-h treatment with LPS (Figure A.6A). However, these bands were not stained by neo epitope antibodies specific for MMP-8 cleavage products of LIX (data not shown), suggesting that the cleavage sites are distinct from those produced by MMP-8. Western blots with the same BALF samples stained for al -P1 demonstrate increased levels of full-length al-PI in knockout compared to wild-type lungs at both 48 and 72 h (Figure A.6B).  150  A  GRO-a  NE  -  +  GRO-13 GRO-y -  +  ,(‘‘ ,(\  B Chemokine GRO-a (1-72) GRO-13 (1-73) GRO-y (1-73) GRO-y (5-73)  -  ENA-78 ÷  +  -  -  -  +  IL-8 -  +  —-1 ___  m/z [M+H] Measured Predicted  7747 7889 7860 7505  NAP-2  7748 7890 7862 7506  +  [M+H]  mlz [M+H] Chemokine Measured Predicted ENA-78 (1-78) 8402 8403 ENA-78 (8-78) 7862 7861 NAP-2 (1-70) 7624 7624 NAP-2 (1-59) 6481 6482 IL-8 (1-77) 8918 8919  C GRO-cL G RO (3 GRO-y -  ENA-78 NAP-2 IL-8  ASVATELRCQCLQTLQGIHP APLATELRCQCLQTLQGIHL ASVVTE LRCQCLQTLQGI H L AGPAAAVLRELRCVCLQTTQGVHP AELRCMCIKTTSGIHP... KKIVQKKLAGDESAD AVLPRSAKELRCQCIKTYSKPFH  Figure A.4. Neutrophil elastase processing of human ELR chemokines. Chemokines were incubated for 16 h at 37 °C in 50 mM Tris, 200 mM NaCI, 5 mM , pH 7.4. Enzyme to substrate ratio was 1:500 (mol:mol). (A) Reactions 2 CaCI were visualized by 15% Tris-Tricine gel electrophoresis and analyzed by MALDI TOF MS in sinapinic acid matrix. (B) Mass to charge ratios ([M+H}) were deconvoluted to determine truncation products (italics). (C) N-terminal cleavage sites in GRO-’’ and ENA-78 and the C-terminal cleavage in NAP-2 are depicted in red. Sequences are aligned by the ELR motif, highlighted in gray.  151  o o o  A Iz •  ,  W.W)  0  o o  If)  -  _  .t_  o  0  0  It)  0 It)  -  0 i—  If)  —54 kDa al -PP 4— 50 kDa a 1-PP  B Time(h)  0  0.5  1  2  3  4  6  8  —  10 —54kDaa1-PI 4—50 kDa al-PP  Figure A.5. MMP-8 cleavage of al-proteinase inhibitor. (A) The serine protease inhibitor al-PP was incubated with MMP-8 for 16 h at 37 °C in 50 mM Tris, 200 mM NaCI, 5 mM CaCI , pH 7.4 containing 1 mM APMA. Enzyme to 2 substrate ratio ranged from 1:5 to 1:5000 (w:w). Reactions were visualized on a 10% SDS-PAGE (silver stained). (B) Time course of MMP-8 cleavage of al-PP at 1:50 (w:w) enzyme-substrate ratio.  152  48h WT KO  LIX STD  A  72h WT KO  (1-92) (5-92) —+ —i (5-79)  al-PI  B MMP-8  +  48 h WT KO  72 h KO WT  62j 49  —  38  —  —  14-  Figure A.6. Reduced LIX processing and increased full length al-PI in BALF from MMP-8 deficient mice. Female wild-type and MMP-8 deficient mice had 2 ig LPS instilled in their lungs and mice were sacrificed and lungs lavaged with PBS at 48 and 72 h (Performed by Dr. Alain Doucet). Cell-free BALF from 3 mice was pooled and concentrated by acetone precipitation. (A) Western blot for LIX in BALE samples. Standard contains 5 ng each of LIX (1-92), (5-92), and (5-79). Antibody was polyclonal anti-LIX from R&D Systems and was used at 1:100. (B) Western blot for al-PI in BALE. Standards were 50 ng al-PI (uncleaved) and 50 ng al-Pl cleaved with MMP-8 (+MMP8). Antibody was anti-human antitrypsin (Sigma) used at 1:1000. Detection was with Alexa-conjugated antibodies (Molecular Probes) on the LiCOR Odyssey.  153  Summary and Future Experiments The data presented herein supports our current hypothesis that MMP-8 processes and inactivates al-PI, leading to increased neutrophil elastase activity and LIX activation, and thereby promoting neutrophil infiltration (Figure A.7). Conversely, in the MMP-8 knockout mouse, al-PI is not inactivated and therefore neutrophil elastase remains inhibited. Notably, other neutrophil MMPs do not seem to compensate in the MMP-8 knockout mouse. As such, LIX processing and activation by neutrophil elastase is decreased in the absence of MMP-8, resulting in decreased neutrophil infiltration. However, some inconsistencies must be addressed. The Nterminal cleavage site generated by murine neutrophils 9 (Al 8 ) Thr a is different than that of human neutrophil elastase ) 7 (11 6 A1a . e This may simply reflect a slight difference in species specificity or could result from protease cooperativity in the context of the neutrophil, an effect that is absent in biochemical cleavage experiments. This could be clarified with murine neutrophil elastase, but this enzyme is not commercially available. Alternatively, LIX proteolysis and activation in lung inflammation could be studied in the presence of specific neutrophil elastase inhibitors to define the role of this serine protease in the physiological cleavage and activation of LIX.  154  ___  rWJ  iL  Neutrophil  ‘a  :d.  •CXCR2 Mature  -: / LIX GAGsY JLiLJ  Endothelial layer  *  .  ••  .  —  ccJ MMP-8  C  •d al-PI  /*  4  ••.•• Activated LIX  NE  Figure A.7. Proposed mechanism for MMP-8 regulation of neutrophil elastase-mediated LIX activation. During infection or injury, tissue and endothelial cells secrete neutrophil chemoattractants including LIXJmCXCL5. In the mature form, LIX is a weak chemoattr actant. Neutrophils that extravasate into the tissue are activated and degranul ate, releasing MMP-8 and neutrophil elastase (NE). NE is inhibited by the irreversible serpin, al proteinase inhibitor (al-PI). However, MMP-8 cleaves the C-terminus of al-Pl rendering it inactive and thereby increasing NE activ ity. Uninhibited NE cleaves the N-terminus of LIX resulting in increased chemoattraction of neutrophils, demonstrating an indirect feed-forward mech anism of neutrophil recruitment.  155  References 1.  Tester, A. M., J. H. Cox, A. R. Connor, A. E. Starr, R. A. Dean, X. S. Puente, C. Lopez-Otin, and C. M. Overall. 2007. LPS responsiveness and neutrophil chemotaxis in vivo require PMN MMP-8 activity. PLoS ONE 2:e312.  2.  Van Den Steen, P. E., A. Wuyts, S. J. Husson, P. Proost, J. Van Damme, and G. Opdenakker. 2003. Gelatinase B/MMP-9 and neutrophil collagenase/MMP-8 process the chemokines human GCP-2ICXCL6, ENA-78/CXCL5 and mouse GCP-2ILIX and modulate their physiological activities. Eur J Biochem 270:3739-3749.  3.  Rao, N. V., B. C. Marshall, B. H. Gray, and J. R. Hoidal. 1993. Interaction of secretory leukocyte protease inhibitor with proteinase-3. Am J Respir Cell Mo! Biol 8:612-616. Wright, C. D., J. A. Kennedy, R. J. Zitnik, and M. A. Kashem. 1999. Inhibition of murine neutrophil serine proteinases by human and murine secretory leukocyte protease inhibitor. BIochem Biophys Res Commun 254:614-617.  4.  5.  McQuibban, G. A., J. H. Gong, E. M. Tam, C. A. McCulloch, I. Clark-Lewis, and C. M. Overall. 2000. Inflammation dampened by gelatinase A cleavage of monocyte chemoattractant protein-3. Science 289:1202-1206.  6.  Schilling, 0., and C. M. Overall. 2008. Proteome-derived, database-searchable peptide libraries for identifying protease cleavage sites. Nat BIotechnol 26:685-694.  7.  Knauper, V., H. Reinke, and H. Tschesche. 1990. Inactivation of human plasma alpha 1proteinase inhibitor by human PMN leucocyte collagenase. FEBS Left 263:355-357.  8.  Vissers, M. C., P. M. George, I. C. Bathurst, S. 0. Brennan, and C. C. Winterbourn. 1988. Cleavage and inactivation of alpha 1-antitrypsin by metalloproteinases released from neutrophils. J Clin Invest 82:706-711.  9.  Liu, Z., X. Zhou, S. D. Shapiro, J. M. Shipley, S. S. Twining, L. A. Diaz, R. M. Senior, and Z. Werb. 2000. The serpin aiphal-proteinase inhibitor is a critical substrate for gelatinase B/MMP-9 in vivo. Cell 102:647-655.  10.  Henry, M. T., K. McMahon, C. Costello, M. X. Fitzgerald, and C. M. O’Connor. 2002. Secretory leukocyte proteinase inhibitor and elafin are resistant to degradation by MMP-8. Exp Lung Res 28:85-97.  11.  Balbin, M., A. Fueyo, A. M. Tester, A. M. Pendas, A. S. Pitiot, A. Astudillo, C. M. Overall, S. D. Shapiro, and C. Lopez-Otin. 2003. Loss of collagenase-2 confers increased skin tumor susceptibility to male mice. Nat Genet 35:252-257.  156  APPENDIX B. PROTEOMIC ANALYSIS OF MMP8 CLEAVAGE SPECIFICITY AND SUBSTRATE 6 IDENTIFICATION Perspective There are fundamental differences between mouse and human MMPs, where humans have 23 and mice have 24 MMPs. It is very important to consider such differences when interpreting results from mouse mod els. Biochemical data with the ELR chemokine CXCL5 shows a shift in cleav age specificity between human and rodent MMP-8, likely dUe to a lysine at S3’ in the rodent enzyme in place of alanine in human MMP-8 (Chapter 2). Therefore , the cleavage specificities of human and mouse MMP-8 were compared to valid ate studies with the MMP-8 knockout mouse model. Traditional peptide-centric methods for profiling the cleavage site specificity of a protease are time cons uming, expensive, and provide limited information [reviewed in (1)]. A novel techn ique that relies on proteome derived, database-searchable peptide libraries for the identification of cleavage sites, termed PICS (2), was used to compare the sequence specificities of mouse and human MMP-8. Very minor differences in consensus cleavage sites were detected, suggesting that mouse and human MMP-8 act on similar substrates and have conserved physiological functions. To identify tissu e-derived extracellular MMP-8 substrates in a systematic, non-biased manner, recombin ant murine MMP-8 was added to immortalized embryonic fibroblasts that were isolated from MMP-8 knockout mice. The protein mixture in the condition ed media was then simplified by terminal amine isotopic labeling of substrates (termed TAILS), which enriches for Nterminal peptides (KIelfeld et al., manuscript in prep aration). By comparing the intensities of isotopic labels in untreated and mMMP-8-tre ated samples, the levels of 20 peptides were significantly changed. Hence, the corresponding proteins represent potential mMMP-8 substrates. Future work will be aimed at validating potential substrates and using quantitative proteomic s to compare neutrophils isolated from wild-type and MMP-8 knockout mice.  6  The studies presented in Appendix B are unpublished. 157  Results PICS analysis of MMP-8 cleavage specificity The proteolytic cleavage specificities of murine and human MMP-8 were determined as described previously (2).  Briefly, peptide libraries were prepared from the  insoluble lysis pellet HEK 293 cells, digested with trypsin, and cysteines and amines were protected with iodoacetamide and formaldehyde treatment, respectively (library provided by Dr. Oliver Schilling).  The library was treated with murine or human  MMP-8 at a ratio of 1:100 (w:w) for 16 h at 37 °C.  Neo-amino termini were  conjugated to cleavable biotin and purified on immobilized streptavidin.  Following  LC-MS/MS on a QSTAR Pulsar (Applied Biosystems, Operated by the UBC Centre for Blood Research Mass Spectrometry Suite), peptides were identified by MASCOT in conjunction with PeptideProphet analysis. The C-terminal (non-prime) side of the original substrates was inferred from the human IPI database (Perl script by Dr. Oliver Schilling and Dr. Ulrich auf dem Keller). PICS analysis of tryptic libraries with MMP-8 yielded 187 and 266 cleaved peptides with mouse and human enzymes, respectively. The resulting heat maps, generated with Pro FIT software, reveal a strong preference for leucine at P1’, proline and alanine at P3, and glycine and alanine at P3’ (Figure B.1).  This specificity is  comparable to that reported for MMP-2 (2). Notably, there were very few differences between the mouse and human cleavage preferences. Human MMP-8 had slightly elevated preference for glycine at P5, whereas mouse MMP-8 had increased preference for serine at P1 and leucine at P4 and P5’. Overall, the specificities look very similar, suggesting conserved cleavage specificities for human and mouse  MMP-8 in vivo.  158  MOUSE  HUMAN  P6 P5 P4 P3 P2 P1 P1’ P2’ P3’ P4’ P5’ P6’  P6 P5 P4 P3 P2 P1 P1 P2’ P3’ P4’ P5’ P6’ -50  -40 >  3  0  -30 2) C)  0 0  -20 C CD  C)  -10 CD  -0  Substrate positions  Substrate positions  Figure B.1. Cleavage site specificities of murine and human MMP-8. PIGS analysis of tryptic peptide libraries was performed at an enzyme-library ratio of 1:100 (w:w) for 16 h at 37 °C. Amino acid occurences as percentages from P6 to P6’ are plotted. carboxyamidomethylated cysteine; K, dimethylated lysine. ,  159  MMP-8 substrate discovery with TAILS enrichment method Over the past few years, numerous proteomic approaches to protease substrate discovery have been reported including 2-dimensional gel electrophoresis (3), isotope-coded affinity tags (ICAT) (4-6), and isobaric tags for relative and absolute quantitation (iTRAQ) (7). A novel method with the benefit of simplifying complex proteomes is called TAILS, for terminal amine isotopic labeling of substrates (Kleifeld et al., manuscript in preparation). Like iTRAQ, TAILS relies upon isotopic labeling of protein N-termini comparing untreated and protease-treated samples. However, following tryptic digestion of proteins, internal peptides are removed by coupling of free N-termini to a large polymer containing reactive aldehyde groups. The original N-termini in the sample are blocked with isotopic labels and therefore do not bind the polymer.  As such, the sample is greatly enriched for original and  protease-derived N-terminal peptides allowing for better detection of potential protease cleavage sites. Embryonic fibroblasts from MMP-8 knockout mice were isolated and immortalized to provide a naive system for studying MMP-8 proteolysis of extracellular matrix and secreted tissue proteins.  Briefly, embryos were isolated from time-mated MMP-8  knockout females on the C57BL/6 x 129 background at 13.5 days gestation. Notably, timed mating was most successful when females were placed in a malescented cage 3 days prior to mating, thereby initiating the female estrus cycle. Embryonic heads and livers were removed and remaining tissue was homogenized  in trypsin and cultured in DMEM supplemented with 10% fetal bovine serum, 2 mM glutamine, 100 units/mL penicillin, 100 tg/mL streptomycin, and 100 tM 2mercaptoethanol. Cells were passaged every 3-5 days until immortalized and then subcloned. Recombinant murine MMP-8 was added to Mmp8-/- fibroblast cultures in serum-free conditions for 24 h at a concentration of 0.4 g/mL, as determined by active site titration with TIMP-1  and TIMP-2.  Conditioned media was collected and  160  concentrated by precipitation with trichloroacetic acid, resuspended in 50 mM NaOH, then neutralized with 250 mM HEPES, pH 8.0. 500 tg each of untreated and MMP 8 treated proteomes were reduced, denatured, isotopically labeled with heavy or light label, combined, and trypsin digested. The resultant peptide mixture was then incubated with an excess of aldehyde-containing polymer to remove internal peptides.  The remaining peptides were fractionated by strong cation exchange  HPLC and analyzed by LC-MS/MS on a QSTAR Pulsar (Applied Biosystems, Operated by the UBC Centre for Blood Research Mass Spectrometry Suite). Peptides were identified by MASCOT assignment and relative abundance quantified by Trans Proteomic Pipeline (TPP, Institute for Systems Biology). The resulting data was analyzed and annotated by Dr. Ulrich auf dem Keller. Following polymer enrichment, 432 unique peptides were identified. Of these, 76% had iTRAQ-labeled N-termini, 13% had acetylated N-termini, 10% had iTRAQ labeled lysines, and 1% were completely unlabeled (Figure B.2A). Therefore, 11% of peptides had unlabeled N-termini. In contrast, 89% of peptides had blocked N termini and hence represent either mature or proteolytically generated N-termihal peptides, demonstrating clear enrichment using the TAILS technique.  The  distribution of isotopic ratios relating MMP-8 treated to untreated controls is centered around natural log of 0, indicating a 1:1 labeling ratio (Figure B.2B). The median of this distribution was 0.96. Using a cut-off of a two fold change in isotopic intensity, 9 peptides were significantly upregulated in MMP-8 treated samples, including MMP-8 itself (Figure B.2C). The other increased peptides could represent direct MMP-8 substrates but these have yet to be validated.  In contrast, 12 peptides were  significantly downregulated in the presence of MMP-8. This could be the result of protein degradation by MMP-8, MMP-8 regulation of other proteases in the cellular context, or alternatively MMP-8 cleavage within the detected peptide. The latter of these possibilities is probable in several cases where a strong MMP consensus sequence of proline at P3 and leucine or isoleucine P1’ are evident (shown in red), such as SPARC, shown previously to be cleaved by several MMPs at this site (8). Notably, 9 of the downregulated peptides contain RL and RV on the N-terminal side  161  of cleavage (P2 and P1). Potentially, these sites are cleaved by unknown fibroblast derived  protease(s),  and  hence these  changes could  result from  MMP-8  downregulation of the activity of other proteases in the protease web.  Summary and Future Directions The data presented here has demonstrated similar cleavage site specificities for murine and human MMP-8 with a consensus of leucine at P1’, proline and alanine at P3, and glycine and alanine at P3’. However, it is known that exosites such as the hemopexin C domain are major determinants of substrate specificity, and exosite interactions are unlikely at the peptide level (9). Therefore, comparing human and mouse MMP-8 activities on protein substrates should be addressed in the future. The TAILS data shown here is really just the beginning of a project aimed at identifying MMP-8 substrates, through the use of immortalized cells as well as primary neutrophils from wild-type and MMP-8 knockout mice.  Several potential  substrates were identified and these will be confirmed with repeat experiments and validated both biochemically and physiologically. Quantitative proteomic techniques including TAILS have the potential to reveal the precise mechanisms through which MMPs such as MMP-8 are regulating fundamental cellular and physiological processes.  162  A  Lysine 10%  B:  Unlabeled i1°  Acetylated 13%  c7  Cu  In (MMP8:Control)  C PROTEIN  SEQUENCE  Neutrophil collagenase Fibronectin Beta 2 microglobulin Pfas Novel protein Tyrosine protein kinase like 7 Extracellular matrix protein Nuclear pore complex-associated TPR Thrombospondin 2 Dystroglycan  VPEH .LEEKN IKTAENYLR YLPN .VDEEVQIGHVPR TGLY.AIQKTPQIQVYSR QQLG.ASVVPLSPDHAR AGAQ.AAIVFI KEPSSQDALQG R ALAS.AASEGAFKASDQR EGGD.GTDPGTETEESMGGAESHQR RASA.GDHVKDTSFDLFSISNINR AVAQ.AHWPSEPSEAVR  >20 5.3 4.8 3.6 3.4 3.1 2.6 2.5 2.4  Periostin Transitional ER ATPase Protein disulfide isomerase A3 PDZ and LIM domain protein 1 Moesin Heat shock cognate 71 kDa protein Actin cytoplasmic 1 Transketolase Superoxide dismutase Cu Zn SPARC Fibronectin Hspg2 perlecan  KI ID.GVPVEITEKQTR RVRL.GDVISIQPCPDVKYGKR KKFI .SDKDASVVGFFR GFRL.VGGKDFEQPLAISR QKHL.KALTSELANAR TERL. lG DAAKNQVAMN PTNTVFDAKR ELRV.APEEHPVLLTEAPLNPKANR INRL.GQSDPAPLQHQVDIYQKR EDRV. ISLSGEHSIIGR EKRL. EAGDH PVELLAR GYRV.EVLPVSLPGEHGQR EARL.VSEDPINDGEWHR  0.49  MMP8:Control  0.21 0.07 0.07 0.06 0.05 0.05 <0.05 <0.05 <0.05 <0.05 <0.05  Figure B.2. CLIP-TAILS analysis of Mmp8-I- MEF secretome in the presence  and absence of murine MMP-8. (A) Distribution of 432 unique peptides identified by TAILS enrichment. (B) Relative frequency of ratios plotted as a function of In (MMP8:control) ratios based on the intensity of heavy to light isotopic labels. The median of distribution is 0.96. (C) Isotopic intensities of several peptides were significantly altered and are listed with their corresponding proteins. Sequences in red indicate potential MMP cleavage sites within the identified sequence.  163  References 1.  Schilling, 0., and C. M. Overall. 2007. Proteomic discovery of protease substrates. Curr Opin Chem Biol 11:36-45.  2.  Schilling, 0., and C. M. Overall. 2008. Proteome-derived, database-searchable peptide libraries for identifying protease cleavage sites. Nat Biotechnol 26:685-694.  3.  Greenlee, K. J., D. B. Corry, D. A. Engler, R. K. Matsunami, P. Tessier, R. G. Cook, Z. Werb, and F. Kheradmand. 2006. Proteomic identification of in vivo substrates for matrix metalloproteinases 2 and 9 reveals a mechanism for resolution of inflammation. J lmmunol 177:7312-7321.  4.  Tam, E. M., C. J. Morrison, Y. I. Wu, M. S. Stack, and C. M. Overall. 2004. Membrane protease proteomics: Isotope-coded affinity tag MS identification of undescribed MTI-matrix metalloproteinase substrates. Proc Nati Acad Sd U S A 101:6917-6922.  5.  Dean, R. A., G. S. Butler, Y. Hamma-Kourbali, J. Delbe, D. R. Brigstock, J. Courty, and C. M. Overall. 2007. Identification of candidate angiogenic inhibitors processed by matrix metalloproteinase 2 (MMP-2) in cell-based proteomic screens: disruption of vascular endothelial growth factor (VEGF)/heparin affin regulatory peptide (pleiotrophin) and VEGF/Connective tissue growth factor angiogenic inhibitory complexes by MMP-2 proteolysis. Mo! Cell Biol 27:8454-8465.  6.  Butler, G. S., R. A. Dean, E. M. Tam, and C. M. Overall. 2008. Pharmacoproteomics of a metalloproteinase hydroxamate inhibitor in breast cancer cells: Dynamics of matrix metalloproteinase-14 (MT1-MMP) mediated membrane protein shedding. Mo! Cell Biol.  7.  Dean, R. A., and C. M. Overall. 2007. Proteomics discovery of metalloproteinase substrates in the cellular context by iTRAQ labeling reveals a diverse MMP-2 substrate degradome. Mo! Cell Proteomics 6:611-623.  8.  Sasaki, T., W. Gohring, K. Mann, P. Maurer, E. Hohenester, V. Knauper, G. Murphy, and R. Timpl. 1997. Limited cleavage of extracellular matrix protein BM-40 by matrix metalloproteinases increases its affinity for collagens. J 8101 Chem 272:9237-9243.  9.  Overall, C. M. 2002. Molecular determinants of metalloproteinase substrate specificity: matrix metalloproteinase substrate binding domains, modules, and exosites. Mo! Biotechno! 22:5186.  164  APPENDIX C. UBC RESEARCH ETHICS BOARD CERTIFICATES This appendix contains the following approval certificates from the University of British Columbia.  1. Animal Care certificate of approval of mouse breeding (A06-0347) 2. Animal Care certificate of approval of air pouch model (A06-0303) 3.  Animal Care certificate of approval of MRL/lpr arthritis experimentation (A06-  0170)  4. Clinical Research Ethics Board approval of use of human blood in chemotaxis and proteolysis experiments (H6-00047)  165  https: //rise.ubc.ca/rise/Doc/0/88B6VKBPL2 KKV3LF8GF4V8jP22 /fromString.html  04/06/08 10:46 AM  THE UNIVERSITY OF BRITISH COLUMBIA  4  ANIMAL CARE CERTIFICATE Application Number: A06-0347 Investigator or Course Director: Christopher Overall  Department: Oral Biological & Medical Sciences Animals:  Mice 1500  Start Date:  June 24, 2006  °‘  August 28,2007  Funding Sources: Funding Agency: Funding Title: Funding Agency: Funding Title: Funding Agency: Funding Title:  Funding Agency: Funding Title: Funding Agency: Funding Title: Funding Agency: Funding Title:  Canadian Institutes of Health Research (CIHR) Leukocyte matrix metalloproteinases (MMPs) in arthritis National Cancer Institute of Canada Functional proteomics and activation of cancer matrix metalloproteinases Canadian Institutes of Health Research (CIHR) Neutrophil collagenase (MMP-8) in inflammation National Cancer Institute of Canada Proteomic and degradomic analysis of breat cancer metastases Canadian Institutes of Health Research (CIHR) Metalloproteinase degradomics: systems biology of the protease web in periodontitis Canadian Institutes of Health Research (CTHR) CIFIR Group in Matrix Dynamics  166  Pagelof2  https: //rise.ubc.ca/rise/Doc/0/88B6VKBPL2 KKV3I..F8GF4V8JP22 /fromString.html  Funding Agency: Funding Title: Funding Agency: Funding Title:  04/06/08 10:46 AM  Canadian Institutes of Health Research (CIHR) Molecular determinants of human MMP-2 substrate activity National Cancer Institute of Canada Functional proteomics and activation of cancer matrix metalloproteinases UBC Operating Budget  Funding Title: Fundmg Agency: Funding Title: Funding Agency: Funding Title: Funding Agency: Funding Title:  Unfunded title:  BREEDING: Production of MMP-2 and -8 deficient mice Canadian Arthritis Network (CAN)  -  Networks of Centres of Excellence (NCE)  Anti-inflammatory chemokine antagonists for treating arthritis Canadian Institutes of Health Research (CIHR) CIHR Group in Matrix Dynamics National Cancer Institute of Canada Proteomic and degradomic analysis of breast cancer metastases  N/A  The Animal Care Committee has examined and approved the use of animals for the above experimental project. This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the CCAC and some granting agencies.  A copy of this certificate must be displayed in your animal facility.  Office of Research Services and Administration 102, 6190 Agronomy Road, Vancouver, BC V6T 1Z3 Phone: 604-827-5111 Fax: 604-822-5093  167  Page2of2  https: f/rise.ubc.ca/rise/Doc/0/PCI5U 17EL64KP137AHJQNJDTC9/fromString.html  04/06/08 10:45 AM  THE UNIVERSITY OF BRITISH COLUMBIA  ANIMAL CARE CERTIFICATE Application Number: A06-0303 Investigator or Course Director: Christopher Overall Department: Oral Biological & Medical Sciences Animals:  Mice Mice Mice Mice  Start Date:  129 Sv/Ev MMP12 KO 100 B6129SF2/J MMP8 KO 100 129 Sv/Ev 100 B6129SF2/J 100  June 25, 2006  Approval  August 20, 2007  Funding Sources:  Funding Agency:  Canadian Arthritis Network (CAN)  Funding Title:  Anti-inflammatory chemokine antagonists for treating arthritis  Unfunded title:  N/A  -  Networks of Centres of Excellence (NCE)  The Animal Care Committee has examined and approved the use of animals for the above experimental project. This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the CCAC and some granting agencies.  Pagelof2 168  https://rise.ubc.ca/rise/Doc/0/7LCS2O9FOHS4TSM272QNSADR93/fromString.html  04/06/08 10:44 AM  THE UNIVERSITY OF BRITISH COLUMBIA  4  ANIMAL CARE CERTIFICATE Application Number: A06-0 170 Investigator or Course Director: Christopher Overall Department: Oral Biological & Medical Sciences Animals:  Mice MRL/MpJ-Faslpr/J KOMMP8 60 Mice MRL/MpJ-Faslpr/J 60  Start Date:  May 8, 2006  Approval  August 3, 2007  Funding Sources: Funding  Agency:  Canadian Arthritis Network (CAN)  -  Networks of Centres of Excellence (NCE)  Funding Title:  Anti-inflammatory chemokine antagonists for treating arthritis  Unfunded title:  N/A  The Animal Care Committee has examined and approved the use of animals for the above experimental project. This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the CCAC and some granting agencies.  A copy of this certificate must be displayed in your animal facility. 169  Pagelof2  https://rise.ubc.ca/rise/Doc/0/LSA5 5RL242IKSEMSMATATIJU65/fromString.html  ijBç  04/06/08 10:47 AM  The University of British Columbia Office of Research Services Clinical Research Ethics Board Room 210, 828 West 10th Avenue, Vancouver, BC V5Z 1L8 —  ETHICS CERTIFICATE OF EXPEDITED APPROVAL: RENEWAL RINClPAL INVESTIGATOR:  DEPARTMENT:  Christopher Overall  U BC/Dentistry/Oral Biological & Medical H06-00047 Sciences  UBC CREB NUMBER:  INSTITUTION(S) WHERE RESEARCH WILL BE CARRIED OUT:  I  institution  UBC Jancouver Coastal Health (VCHRINCHA)  Site  Vancouver (excludes UBC Hospital) UBC Hospital  )ther locations where the research will be conducted:  N/A  CO-INVESTIGATOR(S): manda Starr Dharlotte Morrison Richard A Dean Jain Doucet Ulrich Auf dem Keller 3eorgina S. Butler Reinhild Kappelhoff Jennifer Cox  5PONSORING AGENCIES: Danadian Arthritis Network (CAN) Networks of Centres of Excellence (NCE) “Anti-inflammatory chemokine antagonists or treating arthritis” Danadian Institutes of Health Research (CIHR) “Substrate identification for matrix metalloproteinases (MMPs) involved in 9rthritis progression” National Cancer Institute of Canada -  -  -  PROJECT TITLE: Migration and proteolytic activity of human immune cells  EXPIRY DATE OF THIS APPROVAL: April 7, 2009 APPROVAL DATE: April 7, 2008 CERTIFICATION:  In respect of clinical trials: 1. The membership of this Research Ethics Board complies with the membership requirements for Research Ethics Boards defined in )ivision 5 of the Food and Drug Regulations. The Research Ethics Board carries out its functions in a manner consistent with Good Clinical Practices. 3. This Research Ethics Board has reviewed and approved the clinical trial protocol and informed consent form for the trial which is to e conducted by the qualified investigator named above at the specified clinical trial site. This approval and the views of this Research Ethics Board have been documented in writing. .  The Chair of the UBC Clinical Research Ethics Board has reviewed the documentation for the above named project. The research study, as presented in the documentation, was found to be acceptable on ethical grounds for research involving human subjects and was approved for renewal by the UBC Clinical Research Ethics Board.  Approval of the Clinical Research Ethics Board by:  170  Pagelof2  https://rise.ubc.ca/rise/Doc/O/LSA55RL242IKSEMSMATATIJU65/fromString.html  04/06/08 10:47 AM  Dr. Bonita Sawatzky, Associate Chair  171  Page2of2  

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.24.1-0066986/manifest

Comment

Related Items