Matrix Metalloproteinases in the Central Nervous System

Matrix Metalloproteinases in the Central Nervous System May 26, 2005 16:16 WSPC/Book-SPI-B282 (9in x 6in) Matrix Me...

Author: Katherine Conant | Paul E Gottschall

58 downloads 1123 Views 15MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Matrix Metalloproteinases

in the Central Nervous System

May 26, 2005

16:16 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System fm

This page intentionally left blank

Matrix Metalloproteinases

in the Central Nervous System editors

Katherine Conant Johns Hopkins University, USA

Paul E Gottschall University of South Florida, USA

Imperial College Press

Published by Imperial College Press 57 Shelton Street Covent Garden London WC2H 9HE Distributed by World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, Singapore 596224 USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601 UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE

British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library.

Cover design: Image courtesy of Professor Dr. Thomas Arendt Paul Flechsig Institute of Brain Research Department of Neuroanatomy Leipzig University, Germany

MATRIX METALLOPROTEINASES IN THE CENTRAL NERVOUS SYSTEMS Copyright © 2005 by Imperial College Press All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the Publisher.

For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy is not required from the publisher.

ISBN 1-86094-559-7

Typeset by Stallion Press Email: [email protected]

Printed in Singapore.

May 26, 2005

16:16 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System fm

CONTENTS

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vii

I. A Brief Overview Chapter 1.

The Matrix Metalloproteinases and Their Inhibitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Madeleine M. Handsley, Janet Cross, Jelena Gavrilovic, and Dylan R. Edwards

3

II. Regulation of MMP Expression Chapter 2.

Chapter 3.

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins . . . . . . . . Tammie Roy and Elizabeth A. Milward Post-Translational Modiﬁcation . . . . . . . . . . . . . . . . . . Zezong Gu, Marcus Kaul, and Stuart A. Lipton

19 67

III. MMP Function Chapter 4.

Chapter 5.

Substrates for Metalloendopeptidases in the Central Nervous System . . . . . . . . . . . . . . . . . . . Paul E. Gottschall, John D. Sandy, and Dieter R. Zimmermann Examples of Signalling by MMPs . . . . . . . . . . . . . . . Katherine Conant

87

119

IV. MMPs and TIMPs in Development Chapter 6.

Metalloproteinases in Development — Breaking Things Down to Build a Nervous System . . . . . . 153 Sarah McFarlane v

May 26, 2005

16:16 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System fm

vi

Chapter 7.

Chapter 8.

Contents

Matrix Metalloproteinases in Myelin Formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Peter H. Larsen and V. Wee Yong TIMPs in CNS Development . . . . . . . . . . . . . . . . . . . . Diane Jaworski

189 207

V. MMPs and TIMPs in Disease Chapter 9.

Matrix Metalloproteinases in Cerebral Ischemia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monica Wetzel, Lee Anna Cunningham, and Gary A. Rosenberg

227

Chapter 10. Are Matrix Metalloproteinases Detrimental or Beneﬁcial to Artherosclerosis? . . . . . . . . . . . . . . . Vincent Lemaˆıtre and Jeanine D’Armiento

249

Chapter 11. Role and Regulation of Matrix Metalloproteases in Brain Tumours . . . . . . . . . . . . . S. Sajani Lakka and Jasti S. Rao

263

Chapter 12. Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease, Parkinson’s Disease and Other Neurodegenerative Disorders . . . . . . . 279 Carl H. Parsons, Irene Koolwijk, Tammie Roy, Cherie Roy, Daniel Johnstone, Catharine M.P. Vos, Katherine Conant, Elizabeth A. Milward Chapter 13. Tissue Inhibitors of Matrix Metalloproteinases, Inﬂammation, and The Central Nervous System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 Stephen J. Crocker and Iain L. Campbell Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

333

May 26, 2005

16:16 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System fm

FOREWORD

Matrix metalloproteinases (MMPs) are proteolytic enzymes initially named for their ability to act on components of the extracellular matrix. It is now appreciated that MMPs also target numerous soluble molecules and cell surface receptors. Included among these are select cell-cell adhesion molecules, many of which play a role in blood brain barrier function, cell survival, and/or synaptic stability. MMPs also process varied proforms of trophins that act on cells of the central nervous system (CNS). Thus, it can easily be imagined that these enzymes play critical roles in both brain development and disease. In the chapters to follow, topics related to MMPs in the CNS are presented. Included are chapters that focus on the expression and activation of these enzymes within the brain, MMP substrates within the nervous system, potential roles for MMPs in the development of the CNS, and potential roles for MMPs in brain disease. This collection is not meant to be all-inclusive. In addition, chapters can generally be read as separate entities and thus a small amount of overlap between each may be noted. Nonetheless, it is hoped that the chapters to follow will give the reader a true appreciation for an exciting and ever growing ﬁeld of neurobiology. Katherine Conant and Paul Gottschall, November 2004

vii

May 26, 2005

16:16 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System fm

This page intentionally left blank

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

PART I A BRIEF OVERVIEW

1

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

This page intentionally left blank

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

CHAPTER 1 THE MATRIX METALLOPROTEINASES AND THEIR INHIBITORS

M. M. Handsley, J. Cross, J. Gavrilovic∗ and D. R. Edwards† School of Biological Sciences, University of East Anglia, Norwich, NR4 7TJ, UK E-mails: ∗ [email protected] † [email protected]

1. Introduction It has been over 40 years since the identiﬁcation by Gross and Lapiere (1962) (1) of a collagenolytic activity involved in tadpole tail resorption, which was subsequently revealed as collagenase-1, the prototype of the protease family called the Matrix Metalloproteinases (MMPs) or matrixins (2). With the completion of the human genome sequence, the human MMP family is now known to contain 24 members (3). The MMPs have been and remain the focus of much research eﬀort, which was fuelled originally by the recognition that these enzymes played critical roles in tissue remodelling processes, including the involvement in amphibian metamorphosis that set the ball rolling. Over the last twenty years it has been recognised that diseases involving pathological tissue destruction are associated with aberrant production or activation of MMPs, or a lack of their natural tissue inhibitors, the TIMPs. These perceptions led the pharmaceutical industry to develop synthetic MMP inhibitors (MPIs) which entered clinical trials in the late 1990s as potential treatments for cancer and arthritis. That the MPIs did not prove to be wonder drugs was a disappointment, but the lesson learned was that the biological roles of MMPs are much more complex than originally envisaged. That in addition to extracellular matrix (ECM) degradation during tissue remodelling, MMPs regulate the pericellular milieu by

Correspondence to: J. Gavrilovic and D. R. Edwards 3

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

4

M.M. Handsley et al

speciﬁc cleavage of ECM and cell-associated molecules to control the bioactivity of key cell signalling eﬀectors. This review sets out to summarise current knowledge of the composition of the MMP family, as well as their regulation and molecular activities. In particular we draw attention to new insights into the functions of the MMPs, many of which have originated from studies outside the CNS, but which may have relevance to an understanding of brain development and disease pathogenesis. 2. The MMP Family The MMPs are a family of zinc- and calcium-dependent endopeptidases that form a subgroup of the metzincins, characterised by a conserved motif of three histidine residues binding a Zn2+ ion at the catalytic site followed by a methionine that introduces a turn into the molecule. The MMPs are the primary matrix degrading proteases, collectively able to degrade all protein components of the ECM (for a comprehensive list of MMPs and their substrates, see Egeblad and Werb (4); see also the excellent website of the Overall group at: http://www.clip.ubc.ca/mmp timp folder/ mmp substrates.shtm). All MMPs share a basic structural organisation comprising a signal peptide that targets them for secretion, a pro-peptide domain and an N -terminal catalytic domain. Most MMPs, with the exception of MMP-7, -23 and -26, also have a hinge region and C-terminal hemopexin-like domain. The MMPs can be arranged into diﬀerent subgroups based on their structural features, as shown in Fig. 1. MMPs are produced as inactive zymogens, with a cysteine residue within a PRCGV/NPD motif in the propeptide (with the possible exception of MMP-23) that binds to the Zn2+ in the active site cleft, thereby preventing activity (5). Activation requires disruption of the interaction between the cysteine and the Zn2+ and is referred to as the “cysteine switch”. This involves cleavage of the prodomain or disruption of its structure by thiol reactive reagents such as organomercurial compounds, reactive oxygen or denaturants. This can allow further cleavages to occur via interor intra-molecular proteolysis, thus generating the fully active enzyme (6). The C-terminal hemopexin-like domain has a β-propeller structure with pseudo-fourfold symmetry and is involved in both substrate binding and interaction with the tissue inhibitors of metalloproteinases (TIMPs) (7). The hinge region is a ﬂexible linker peptide of variable length that links the N -terminal catalytic and C-terminal hemopexin domains and is important

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

The Matrix Metalloproteinases and their Inhibitors

5

Fig. 1. Structural sub-grouping of the matrix metalloproteinases. The MMPs can be divided into several subgroups depending on their structure. All MMPs possess a signal peptide that targets the MMPs for secretion, a pro-peptide domain (containing a conserved Cys residue), and a catalytic domain. Most MMPs (with the exception of MMP-7 and MMP-26) contain a C-terminal haemopexin domain and a hinge region. Other MMP subgroups contain unique features such as a transmembrane domain, cytoplasmic tail and an MT-loop (MT1-, MT2, MT3, and MT5-MMP), a GPI anchor (MT4-MMP and MT6-MMP), a furin recognition site (MT-MMPs, MMP-11, -21, -23, -28), fibronectin type II repeats (MMP-2 and -9), and an N -terminal signal anchor, a cysteine array and an Ig-like domain (MMP-23).

May 26, 2005

6

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

M.M. Handsley et al

for MMP function. Both gelatinases (MMP-2 and -9) have a ﬁbronectinlike domain consisting of three tandem copies of a ﬁbronectin type II-like module. This domain has been shown to bind denatured type IV and V collagens, elastin, and type-I collagen thereby contributing to substrate speciﬁcity (8). A group of six cell membrane-associated MMPs has also been identiﬁed, known as the membrane-type MMPs (MT-MMPs). The MT-MMPs can be divided into two subgroups, the transmembrane type (MT-MMP-1, -2, -3, -5) and the glycosylphosphatidyl inositol (GPI)-anchored forms (MTMMP-4 and -6) (9–13). Like most MMPs, the GPI-anchored type are sensitive to all members of the TIMP-family, but the transmembrane MT-MMPs are inhibited by TIMP-2, -3, -4, but are relatively insensitive to TIMP-1 inhibition (10, 14, 15). The cytoplasmic tail of the transmembrane MT-MMPs interacts with intracellular proteins that regulate the subcellular traﬃcking of the enzymes from the Golgi to the cell surface, and to speciﬁc membrane domains, for instance to protruding structures called ‘invadopodia’ in invasive cancer cells (4, 16). 3. Regulation of MMP Activity The activity of MMPs can be regulated by various mechanisms — gene transcription, mRNA stability, translational control, cell compartmentalisation, zymogen activation via proteolysis, and inhibition by endogenous inhibitors. MMP activation is often the result of a complex proteinase cascade. Certain MMPs, including MMP-1, MMP-3, MMP-7, MMP-8, MMP-9 and MMP-10, can be cleaved in their propeptide domain, at least in vitro, by serine proteinases such as the uPA–plasmin system and trypsin (17, 18). In turn, some of these activated MMPs can then go on to activate other proMMPs, e.g. MMP-3, which can activate proMMP-1 and proMMP-9 (19, 20). MT1-MMP can activate proMMP-13 (21) and this activated MMP-13 can then go on to activate MMP-9 (22). A subset of MMPs is activated primarily intracellularly by serine proteases of the pro-protein convertase class such as furin (23). This includes MMP-11, MMP-21, MMP-23 and MMP-28 (24, 25) as well as the MT-MMPs (13, 26–29). ProMMP-2 has a unique cell-mediated activation mechanism, which is illustrated in Fig. 2. Activation of proMMP-2 by MT1-MMP has been the most extensively studied mechanism and is used as the basis for the

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

The Matrix Metalloproteinases and their Inhibitors

7

Fig. 2. Cell surface model of proMMP-2 activation. MT1-MMP on the cell surface acts as a receptor for TIMP-2, which binds via its N -terminal domain to the active site of the MT-MMP. This binary complex acts as a receptor for pro-MMP-2, the TIMP-2 C-terminal domain binding to the C-terminal domain of pro-MMP-2. A free MT1-MMP molecule in close proximity can then cleave the propeptide of pro-MMP-2 generating an intermediate species. Further proteolysis of the propeptide through an autocatalytic mechanism results in the generation of the fully active enzyme.

current model, in which MT1-MMP acts as a receptor for TIMP-2, which binds via its N -terminal domain to the active site of the MT-MMP. This binary complex can then act as a receptor for proMMP-2, the TIMP-2 C-terminal domain binding to the C-terminal domain of proMMP-2 (30). A free MT1-MMP molecule, positioned in close proximity via the interaction of the hemopexin domains of the two MT1-MMPs (31), can then cleave the propeptide of proMMP-2 at the N37-L38 bond, generating an intermediate MMP-2 species. Further autocatalytic proteolysis of the intermediate MMP-2 generates the fully active enzyme (32). Activation of proMMP-2 in this model requires TIMP-2 to be at a level that is adequate for the generation of the tri-molecular complex, but not high enough to saturate all the MT1-MMPs. High levels of TIMP-2 (as well as TIMP-3 and -4) will inhibit all MT1-MMP activity, preventing proMMP-2 activation (33). However, if no TIMP-2 is present, MT1-MMP undergoes autocatalytic processing to a 45 kDa inactive form (34). Certain synthetic MMP inhibitors have been

May 26, 2005

8

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

M.M. Handsley et al

Fig. 3. The Biology of the CNS Degradome MMPs and other metalloproteases play a number of roles (which will interact and/or overlap with each other) in the CNS both in normal and pathological conditions.

shown to stabilize ‘active’ MT1-MMP on the cell surface, thereby potentiating proMMP-2 activation (35). MT2-MMP-mediated proMMP-2 activation does not require TIMP-2 and so appears to function via a diﬀerent mechanism than that described for MT1-MMP (36). The transmembrane MT-MMPs (MT1-, MT2-, MT3- and MT5-) are all able to activate proMMP-2 in vitro (37–40). However, MT4-MMP is without activity in this respect (26) and MT6-MMP is at best poorly active (10). The formation of homophilic complexes of MT1-MMP molecules at the plasma membrane is necessary for eﬃcient activation of proMMP-2, as well as for MT1-MMP’s invasion promoting properties. Complex formation involves the interaction between the hemopexin domains and also between the cytoplasmic domains of the MT1-MMP molecules (31, 41). 4. Endogenous Inhibitors of Metalloproteinases There are four vertebrate TIMPs which can inﬂuence the degradome (Fig. 3) (42). TIMP-1, -2, and -4 are all diﬀusible, secreted proteins while TIMP-3 is matrix-associated (15, 43). The TIMPs share a basic structural arrangement consisting of 12 conserved cysteine residues paired into 6 disulphide bonds forming 6 peptide loops and two knots. This common structure can be divided into 2 discrete domains, the N -terminal domain, primarily

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

The Matrix Metalloproteinases and their Inhibitors

9

responsible for its MMP inhibitory activity, and the C-terminal domain. The C-terminal domain is more variable in sequence than the N -terminal domain between the 4 TIMPs and may in part be responsible for the distinct properties of the 4 TIMPs (44). While α2 -macroglobulin is the principal endogenous MMP-inhibitor in the circulation, the TIMPs are the primary tissue inhibitors of MMPs and as such their primary role is to limit proteolysis during ECM remodelling (15). The TIMPs can inhibit most MMPs without major selectivity between them (with the exception that TIMP-1 is a very poor inhibitor of MT1-, MT2-, MT3-, MT5-MMP and MMP-19) (32). However, TIMPs diﬀer in other properties such as tissue distribution, transcriptional regulation, and speciﬁc association with latent MMPs (i.e. TIMP-1/proMMP-9 and TIMP-2/proMMP-2). These diﬀerences suggest that they each have separate and speciﬁc physiological roles. TIMPs bind noncovalently the activated MMPs in a 1 : 1 stoichiometry, with inhibition constants in the subnanomolar range (45), forming an extended wedge structure with contacts in both the catalytic and hemopexin domains of the protease. The N terminal cysteine of the mature TIMP molecule is essential for coordination with the active site Zn2+ in the MMP, resulting in inhibition. TIMPs can also be inactivated by a variety of proteinases such as neutrophil elastase and trypsin but are mainly regulated at the level of gene expression (46). “Reversion-inducing cysteine-rich protein with kazal motifs” (RECK) is a recently discovered endogenous cell surface glycoprotein that inhibits the catalytic activities of MMP-9 (47), MMP-2 and MT1-MMP (48). RECK does not bear any structural resemblance to the TIMPs and apparently acts diﬀerently to the TIMPs, leading to the suppression of MMP-9 expression as well as inhibition of pro-MMP-2 activation. RECK expression has been detected in a wide range of tissues and loss of RECK expression has been linked to cancer progression (47). Other molecules that have been suggested as possible MMP inhibitory molecules include thrombospondins-1 (49) and -2, and proteins that resemble the N -terminal structure of the TIMPs, including netrin, secreted frizzled-related proteins, type I collagen C-proteinase enhancer protein (PCPE), and the serine proteinase tissue factor pathway inhibitor-2 (TFPI-2) [reviewed in Baker et al, 2002] (42). TIMPs also possess properties independent of their MMP inhibitory activity, including the ability of both TIMP-1 and TIMP-2 to promote growth in a variety of cell types (50). TIMP-1 proteins that have been engineered to lack anti-proteolytic activity retain growth-stimulating activity, indicating that the anti-proteolytic and growth factor activities of TIMP-1

May 26, 2005

10

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

M.M. Handsley et al

are separable (51). Overexpression of TIMP-3 (but not TIMP-1 or TIMP-2) by adenoviral delivery induces apoptosis in smooth muscle cells and cancer cell lines (52–54). However, it is probable that these pro-apoptotic eﬀects of TIMP-3 relate to the selective ability of this TIMP to inhibit another family of cell surface metalloproteinases called the adamalyins, or ADAMs (for a disintegrin and metalloproteinase). The ADAM metalloproteinases are involved both in cell adhesion and proteolysis, in particular, ADAM-17 acts as an “ectodomain sheddase” of various receptors and adhesion molecules as well as processing membrane-bound pro-TNF-α and pro-TGF-α liberating the active ligand (55). Several pro-apoptotic ligands and receptors are substrates of ADAM-17, thus its inhibition by TIMP-3 can potentiate the actions of apoptosis-inducing stimuli (56). The ADAMs and their close relatives, the ADAMTS metalloproteinases (a disintegrin and metalloproteinase with thrombospondin motifs) (57), are outside the scope of this current review, but they are likely to be as important as the MMPs in the development and disease of the CNS. 5. Novel Roles for MMPs Whilst many ECM substrates have been identiﬁed for MMPs, interest has shifted to potential roles in the generation of matricryptic sites within ECM substrates, sometimes generating for example gradients of motogenic factors as well as liberating growth factors from the matrix or in some cases from the cell surface. Here we will also consider data obtained in other systems which may reveal ways in which novel roles for MMPs in the CNS could be explored (some of which are discussed in later chapters). 6. ECM Targets in the CNS ECM components in the CNS include proteoglycans, laminins, tenascins and to a certain extent collagens, although these proteins are less abundant than in the PNS under normal conditions. Following injury to the CNS and subsequent formation of the glial scar, the production of several ECM molecules, including collagens, is up-regulated (58). The major recent focus for ECM study in the glial scar has, however been the proteoglycans, including brevican, versican and neurocan (59). These chondroitin sulphate PGs (CS-PGs) have been shown to be inhibitory for axon outgrowth in vitro and for the regeneration of severed axons in lesioned CNS in vivo. Degradation of CS side chains with bacterial chondrotinase ABC

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

The Matrix Metalloproteinases and their Inhibitors

11

relieves this inhibition (60). In the CNS, this type of role for MMPs has not been extensively explored but in the PNS degradation of CSPGs by MMP-2 enhances axon outgrowth (61). However MMP-9 has been strongly implicated in oligodendrocyte outgrowth (62) although the mechanism involved remains to be elucidated. Nidogen has been shown to be critical to normal CNS function (63) and deletion of the nidogen-binding domain of the laminin gamma 1 chain results in disrupted cortical histogenesis (64). It is unknown however whether nidogen degradation plays a role in normal or pathological CNS development/repair although it has long been established that nidogen-1 is a substrate for many MMPs (65). 7. Generation of Matricryptic Sites in ECM Components Although degradation of the ECM by MMPs has been extensively studied, the cell biological consequences of this process have not often been investigated in detail. However, some information has been obtained in systems other than the CNS which will be explored brieﬂy here. One of the most abundant ECM components in the body, type I collagen, when degraded by collagenolytic MMPs into classic 3/4 and 1/4 fragments exposes multiple RGD sites thus enabling cells to adhere through for example alpha v beta 3 integrins. This can alter radically the ability of, for instance vascular smooth muscle cells, to migrate in response to motogens such as PDGF-BB (66). It would be of interest in the future to establish whether collagen degradation occurs in the permissive environment of the PNS but not in the glial scar of the CNS. A less abundant but extremely important ECM component, laminin-5, also contains matricryptic sites, which are revealed following cleavage with MT1-MMP resulting in enhanced tumour cell migration (67). Laminins are important ECM components for neurite outgrowth in vitro and axon outgrowth in vivo but as yet roles for MMPs in generation of bioactive matricryptic sites has not been demonstrated. However, Costa et al (68) have shown that in neuron-astrocyte co-cultures laminin plays a key role in neuronal migration and MMP-2 is concomitantly synthesised. 8. Growth Factor Liberation from ECM “Stores” It is well-established in a number of systems that growth factors bound to the ECM can be liberated from these “stores” following degradation by MMPs. For example, MMP-9 degrades ECM to release VEGF and thus

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

12

M.M. Handsley et al

inﬂuences ossiﬁcation through modulation of angiogenesis in developing bone (69). In the CNS, however, whilst there are many reports of growth factor ECM association as yet roles for MMPs in the liberation of growth factors remains little investigated. 9. Chemokine Gradient Generation Whilst chemokines have been implicated in the recruitment of leukocytes across the blood brain barrier (BBB) in inﬂammatory conditions within the CNS, the role for MMPs in the generation of a chemokine gradient has not been explored. MMP-7 has been shown to play a key role in the generation of soluble syndecan-1 from lung alveolar epithelial cells (70). The chemokine IL8 binds to syndecan-1 and MMP-7 was shown to be critical in the generation of an IL8 gradient important in recruiting leukocytes to the alveolar space. As yet it is not established whether MMPs could play such a role in the CNS but it is established that syndecans play a role in transendothelial migration of monocytes across the brain endothelium (71). Another important way that MMPs may act to blunt the immune response is through the direct cleavage and inactivation of the CC class of chemokines. McQuibban et al (72) demonstrated that certain MMPs could cleave MCPs 1-4, showing speciﬁcity depending on the substrate and enzyme. Of great interest is the fact that once cleaved by MMPs, MCP2, -3 and -4 were able to block monocyte migration in vitro and inﬂammation in vivo. 10. Blood-Brain Barrier Substrates for MMPs Whilst MMPs have been implicated in penetration of the BBB for some years, it is only recently that the nature of relevant substrates for MMPs has been explored. Wachtel et al (73) and Lohmann et al (74) have provided evidence that when loss of BBB integrity is induced by tyrosine phosphatase inhibition, the tight junction protein, occludin, is degraded by MMP(s) in an in vitro model of the BBB. The identity of the MMP(s) involved remains to be determined. 11. Conclusion Metalloprotease research in the CNS is still in its infancy although MMPs have been a growing focus of research in this area over the past 10 years,

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

The Matrix Metalloproteinases and their Inhibitors

13

revealing roles in many aspects of CNS biology. Future studies will witness discovery of roles for ADAM and ADAMTS proteases whose expression and function in the nervous system is only beginning to emerge.

References 1. Gross, J., and Lapiere, C. M. (1962) Proc Natl Acad Sci 48, 1014–1022 2. Brinckerhoﬀ, C. E., and Matrisian, L. M. (2002) Nat Rev Mol Cell Biol 3, 207–214 3. Puente, X. S., Sanchez, L. M., Overall, C. M., and Lopez-Otin, C. (2003) Nat Rev Genet 4, 544–558 4. Egeblad, M., and Werb, Z. (2002) Nat Rev Cancer 2, 161–174 5. Becker, J. W., Marcy, A. I., Rokosz, L. L., Axel, M. G., Burbaum, J. J., Fitzgerald, P. M., Cameron, P. M., Esser, C. K., Hagmann, W. K., and Hermes, J. D. (1995) Protein Sci 4, 1966–1976 6. Nagase, H., Ogata, Y., Suzuki, K., Enghild, J. J., and Salvesen, G. (1991) Biochem Soc Trans 19, 715–718 7. Gohlke, U., Gomis-Ruth, F. X., Crabbe, T., Murphy, G., Docherty, A. J., and Bode, W. (1996) FEBS Lett 378, 126–130 8. Steﬀensen, B., Wallon, U. M., and Overall, C. M. (1995) J Biol Chem 270, 11555–11566 9. Itoh, Y., Kajita, M., Kinoh, H., Mori, H., Okada, A., and Seiki, M. J. (1999) Biol Chem 274, 34260–34266 10. Kojima, S., Itoh, Y., Matsumoto, S., Masuho, Y., and Seiki, M. (2000) FEBS Lett 480, 142–146 11. Will, H., and Hinzmann, B. (1995) Eur J Biochem 231, 602–608 12. Takino, T., Sato, H., Shinagawa, A., and Seiki, M. (1995) J Biol Chem 270, 23013–23020 13. Pei, D. (1999) J Biol Chem 274, 8925–8932 14. Bigg, H. F., Morrison, C. J., Butler, G. S., Bogoyevitch, M. A., Wang, Z., Soloway, P. D., and Overall, C. M. (2001) Cancer Res 61, 3610–3618 15. Brew, K., Dinakarpandian, D., and Nagase, H. (2000) Biochim Biophys Acta 1477, 267–283 16. Nakahara, H., Howard, L., Thompson, E. W., Sato, H., Seiki, M., Yeh, Y., and Chen, W. T. (1997) Proc Natl Acad Sci 94, 7959–7964 17. Okada, Y., Gonoji, Y., Naka, K., Tomita, K., Nakanishi, I., Iwata, K., Yamashita, K., and Hayakawa, T. (1992) J Biol Chem 267, 21712–21719 18. Nagase, H., Suzuki, K., Enghild, J. J., and Salvesen, G. (1991) Biomed Biochim Acta 50, 749–754 19. Nagase, H., Suzuki, K., Morodomi, T., Enghild, J. J., and Salvesen, G. (1992) Matrix Suppl 1, 237–244 20. Ogata, Y., Enghild, J. J., and Nagase, H. (1992) J Biol Chem 267, 3581–3584 21. Knauper, V., Bailey, L., Worley, J. R., Soloway, P., Patterson, M. L., and Murphy, G. (2002) FEBS Lett 532, 127–130

May 26, 2005

14

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

M.M. Handsley et al

22. Cowell, S., Knauper, V., Stewart, M. L., d’Ortho, M. P., Stanton, H., Hembry, R. M., Lopez-Otin, C., Reynolds, J. J., and Murphy, G. (1998) Biochem J 331, 453–458 23. Pei, D., and Weiss, S. J. (1995) Nature 375, 244–247 24. Ahokas, K., Lohi, J., Lohi, H., Elomaa, O., Karjalainen-Lindsberg, M. L., Kere, J., and Saarialho-Kere, U. Gene 301, 31–41 25. Illman, S. A., Keski-Oja, J., Parks, W. C., and Lohi, J. (2003) Biochem J 375, 191–197 26. English, W. R., Puente, X. S., Freije, J. M., Knauper, V., Amour, A., Merryweather, A., Lopez-Otin, C., and Murphy, G. (2000) J Biol Chem 275, 14046–14055 27. Yana, I., and Weiss, S. J. (2000) Mol Biol Cell 11, 2387–2401 28. Kang, T., Nagase, H., and Pei, D. (2002) Cancer Res 62, 675–681 29. Pei, D. (1999) Cell Res 9, 291–303 30. Butler, G. S., Butler, M. J., Atkinson, S. J., Will, H., Tamura, T., van Westrum, S. S., Crabbe, T., Clements, J., d’Ortho, M. P., and Murphy, G. (1998) J Biol Chem 273, 871–880 31. Itoh, Y., Takamura, A., Ito, N., Maru, Y., Sato, H., Suenaga, N., Aoki, T., and Seiki, M. (2001) EMBO J 20, 4782–4793 32. Will, H., Atkinson, S. J., Butler, G. S., Smith, B., and Murphy, G. (1996) J Biol Chem 271, 17119–17123 33. Hernandez-Barrantes, S., Shimura, Y., Soloway, P. D., Sang, Q. A., and Fridman, R. (2001) Biochem Biophys Res Commun 281, 126–130 34. Hernandez-Barrantes, S., Toth, M., Bernardo, M. M., Yurkova, M., Gervasi, D. C., Raz, Y., Sang, Q. A., and Fridman, R. (2000) J Biol Chem 275, 12080–12089 35. Toth, M., Bernardo, M. M., Gervasi, D. C., Soloway, P. D., Wang, Z., Bigg, H. F., Overall, C. M., DeClerck, Y. A., Tschesche, H., Cher, M. L., Brown, S., Mobashery, S., and Fridman, R. (2000) J Biol Chem 275, 41415–41423 36. Morrison, C. J., Butler, G. S., Bigg, H. F., Roberts, C. R., Soloway, P. D., and Overall, CM. (2001) J Biol Chem 276, 47402–47410 37. Imai, K., Ohuchi, E., Aoki, T., Nomura, H., Fujii, Y., Sato, H., Seiki, M., and Okada, Y. (1996) Cancer Res 56, 2707–2710 38. Kinoshita, T., Sato, H., Takino, T., Itoh, M., Akizawa, T., and Seiki, M. (1996) Cancer Res 56, 2535–2538 39. Butler, G. S., Will, H., Atkinson, S. J., and Murphy, G. (1997) Eur J Biochem 244, 653–657 40. Shofuda, K., Yasumitsu, H., Nishihashi, A., Miki, K., and Miyazaki, K. (1997) J Biol Chem 272, 9749–9754 41. Lehti, K., Valtanen, H., Wickstrom, S., Lohi, J., and Keski-Oja, J. (2000) J Biol Chem 275, 15006–15013 42. Baker, A. H., Edwards, D. R., and Murphy, G. (2002) J Cell Sci 115, 3719–3727 43. Leco, K. J., Khokha, R., Pavloﬀ, N., Hawkes, S. P., and Edwards, D. R. (1994) J Biol Chem 269, 9352–9360

May 26, 2005

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

The Matrix Metalloproteinases and their Inhibitors

15

44. Gomis-Ruth, F. X., Maskos, K., Betz, M., Bergner, A., Huber, R., Suzuki, K., Yoshida, N., Nagase, H., Brew, K., Bourenkov, G. P., Bartunik, H., and Bode, W. (1997) Nature 389, 77–81 45. Bode, W., Fernandez-Catalan, C., Grams, F., Gomis-Ruth, F. X., Nagase, H., Tschesche, H., and Maskos, K. (1999) Acad Sci 878, 73–91 46. Okada, Y., Watanabe, S., Nakanishi, I., Kishi, J., Hayakawa, T., Watorek, W., Travis, J., and Nagase, H. (1988) FEBS Lett 229, 157–160 47. Takahashi, C., Sheng, Z., Horan, T. P., Kitayama, H., Maki, M., Hitomi, K., Kitaura, Y., Takai, S., Sasahara, R. M., Horimoto, A., Ikawa, Y., Ratzkin, B. J., Arakawa, T., and Noda, M. (1998) Proc Natl Acad Sci 95, 13221–13226 48. Oh, J., Takahashi, R., Kondo, S., Mizoguchi, A., Adachi, E., Sasahara, R. M., Nishimura, S., Imamura, Y., Kitayama, H., Alexander, D. B., Ide, C., Horan, T. P., Arakawa, T., Yoshida, H., Nishikawa, S., Itoh, Y., Seiki, M., Itohara, S., Takahashi, C., and Noda, M. (2001) Cell 107, 789–800 49. Rodriguez-Manzaneque, J. C., Lane, T. F., Ortega, M. A., Hynes, R. O., Lawler, J., and Iruela-Arispe, M. L. (2001) Proc Natl Acad Sci 98, 12485–12490 50. Jiang, Y., Goldberg, I. D., and Shi, Y. E. (2002) Oncogene 21, 2245–2252 51. Chesler, L., Golde, D. W., Bersch, N., and Johnson, M. D. (1995) Blood 86, 4506–4515 52. Baker, A. H., Zaltsman, A. B., George, S. J., and Newby, A. C. (1998) J Clin Invest 101, 1478–1487 53. Smith, M. R., Kung, H., Durum, S. K., Colburn, N. H., and Sun, Y. (1997) Cytokine 9, 770–780 54. Ahonen, M., Poukkula, M., Baker, A. H., Kashiwagi, M., Nagase, H., Eriksson, J. E., and Kahari, V. M. (2003) Oncogene 22, 2121–2134 55. Itai, T., Tanaka, M., and Nagata, S. (2001) Eur J Biochem 268, 2074–2082 56. Edwards, D. R. (2004) J Pathol 202, 391–394 57. Apte, S. S. (2004) Int J Biochem Cell Biol 36, 981–985 58. Hermanns, S., Reiprich, P., and Muller, H. W. (2001) J Neurosci Methods 110, 141–146 59. Rhodes, K. E., and Fawcett, J. W. (2004) J Anat 204, 33–48 60. Bradbury, E. J., Moon, L. D., Popat, R. J., King, V. R., Bennett, G. S., Patel, P. N., Fawcett, J. W., and McMahon, S. B. (2002) Nature 416, 636–640 61. Ferguson, T. A., and Muir, D. (2000) Mol Cell Neurosci 16, 157–167 62. Ohu, L. Y., Larsen, P. H., Krekoski, C. A., Edwards, D. R., Donovan, F., Werb, Z., and Yong, V. W. (1999) J Neurosci 19, 8464–8475 63. Dong, L., Chen, Y., Lewis, M., Hsieh, J. C., Reing, J., Chaillet, J. R., Howell, C. Y., Melhem, M., Inoue, S., Kuszak, J. R., DeGeest, K., and Chung, A. E. (2002) Lab Invest 82, 1617–1630 64. Halfter, W., Dong, S., Yip, Y. P., Willem, M., and Mayer, U. (2002) J Neurosci 22, 6029–6040 65. Mayer, U., Mann, K., Timpl, R., and Murphy, G. (1993) Eur J Biochem 217, 877–884

May 26, 2005

16

15:59WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch01

M.M. Handsley et al

66. Stringa, E., Knauper, V., Murphy, G., and Gavrilovic, J. (2000) J Cell Sci 113, 2055–2064 67. Gilles, C., Polette, M., Coraux, C., Tournier, J. M., Meneguzzi, G., Munaut, C., Volders, L., Rousselle, P., Birembaut, P., and Foidart, J. M. (2001) J Cell Sci 114, 2967–2976 68. Costa, S., Planchenault, T., Charriere-Bertrand, C., Mouchel, Y., Fages, C., Juliano, S., Lefrancois, T., Barlovatz-Meimon, G., and Tardy, M. (2002) Glia 37, 105–113 69. Vu, T. H., Shipley, J. M., Bergers, G., Berger, J. E., Helms, J. A., Hanahan, D., Shapiro, S. D., Senior, R. M., and Werb, Z. (1998) Cell 93, 411–422 70. Li, Q., Park, P. W., Wilson, C. L., and Parks, W. C. (2002) Cell 111, 635–646 71. Floris, S., van den, B. J., van der Pol, S. M., Dijkstra, C. D., and De Vries, H. E. (2003) J Neuropathol Exp Neurol 62, 780–790 72. McQuibban, G. A., Gong, J. H., Wong, J. P., Wallace, J. L., Clark-Lewis, I., and Overall, C. M. (2002) Blood 100, 1160–1167 73. Wachtel, M., Frei, K., Ehler, E., Fontana, A., Winterhalter, K., and Gloor, S. M. (1999) J Cell Sci 112, 4347–4356 74. Lohmann, C., Krischke, M., Wegener, J., and Galla, H. J. (2004) Brain Res 995, 184–196

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

PART II REGULATION OF MMP EXPRESSION

17

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

This page intentionally left blank

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

CHAPTER 2 GENETIC REGULATION OF THE MATRIX METALLOPROTEINASES AND RELATED PROTEINS

T. Roy and E.A. Milward∗ School of Biomedical Sciences, University of Newcastle, Callaghan, NSW 2308, Australia E-mail: ∗[email protected]

1. Introduction The prominent role of post-translational protein processing in controlling MMP activity has sometimes overshadowed the regulation of MMP expression that is exerted at the earlier stages in MMP generation. This can occur through the control of MMP gene transcription, changes in mRNA stability or other mechanisms. These mechanisms make important contributions to the tissue and cell speciﬁcities of particular MMPs, to MMP responses during injury or disease and to the coordinated regulation of the diﬀerent MMPs and the other proteins interacting with them. This chapter outlines some of the general mechanisms by which MMP gene expression is regulated throughout the body before going on to examine particular mechanisms important in the healthy nervous system and in nervous system disease. It also reviews some of the main genetic variants proposed to inﬂuence expression of MMPs and disease susceptibility. Post-translational processing will be discussed in a following chapter. There are various common features in the transcriptional regulation of many of the better-studied MMPs (e.g. MMP-1, 3, 7 and 9) and these will generally be discussed collectively. Homologous domains in the promoter regions of these MMPs facilitate coordinated regulation of expression. In contrast, the promoter region of the MMP-2, MMP-11 and MMP-14 genes show less correspondence with other MMP promoters and their regulation will be discussed separately. Correspondence to: E.A. Milward 19

May 26, 2005

20

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

It should be kept in mind that much of the information on MMP gene expression relating to the nervous system comes from transformed models and does not always provide an accurate picture of nontransformed systems. The levels and regulation of both basal and inducible MMP gene expression can diﬀer substantially in transformed as opposed to nontransformed systems and between in vitro and in vivo systems (1–3). Basal (‘constitutive’) levels of MMP transcription in a tumor-derived cell line such as an astrocytoma can be far greater than basal levels in comparable nontransformed cell types (e.g. astrocytes or glial precursor cells). Furthermore, the basal patterns of MMP expression observed in tumor-derived or other cell lines in vitro are not even necessarily representative of patterns displayed by the parent cells in vivo (3). Similarly, species diﬀerences can be substantial — e.g. regulation of the endogenous MMP-1 gene is very diﬀerent in mice compared to humans (4) — so caution also needs to be exercised when extrapolating across species. We will ﬁrst give an overview of the wide range of diﬀerent stimuli that inﬂuence MMP gene expression. We will then consider the main downstream pathways through which these stimuli act and the cis- and trans-acting genetic regulatory elements inﬂuenced by these pathways. We will also touch on the nervous system regulation of the closely related adamalysin metalloproteinase family (ADAMs) and the naturally occurring tissue inhibitors of metalloproteinases (TIMPs). 2. Stimuli Triggering Alterations in MMP Gene Expression Alterations in MMP expression can occur in a wide range of normal physiological conditions, as well as in injury or disease, and can be triggered by many diﬀerent stimuli. Some of the better characterised stimuli pertinent to the nervous system that have been the subject of recent reports include growth factors such as nerve growth factor NGF (5, 6), epidermal growth factor EGF (7) and transforming growth factor beta TGFβ, (3, 7, 8), integrins (9), peroxisome proliferator-activated receptor PPAR gamma (10), retinoids (11), cytokines such as the pro-inﬂammatory cytokines tumor necrosis factor alpha TNFα (12–16) and interleukin 1α (16) and 1β (12, 17, 18), chemokines (19), oxidants such as iron (20), reactive oxygen species (21–24), reactive nitrogen species (25, 26), prostaglandins and other mediators of the cyclooxygenase and lipoxygenase pathways of arachidonic acid metabolism (27–29), adenosine 3 ,5 -cyclic monophosphate cAMP (2), protein kinase C PKC (15, 30), p53 (31), tumour promoting agents such as

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

21

phorbol esters (15, 32, 33) and oncogenes such as ras and jun, as reviewed elsewhere (e.g. 34). Other important stimuli include nuclear receptor superfamily members such as estrogen (28), progesterone, androgens and thyroid hormone (reviewed (35)) and glucocorticoids (36). The MMPs are also up-regulated in response to infectious and inﬂammatory agents such as lipopolysaccharide (37), viruses such as human T cell lymphotropic virus type 1 (HTLV-I) (38) and viral proteins such as the human immunodeﬁciency virus HIV-1 glycoprotein 120 (39) and the transmembrane component of the HIV envelope gp41 (40). Conversely, antibiotics such as tetracycline can reduce expression of MMPs and their transcripts (41). Infections and inﬂammatory diseases will be examined in depth in later chapters. In addition to these relatively well-known MMP stimuli, some other more novel stimuli are given below. 3. Alterations in Other MMPs Besides well-recognised interactions at the level of post-translational activation (see following chapters), changes in some MMPs may also trigger alterations in the transcription of other MMPs. For example, compensatory up-regulation of substitute MMPs occurs in some MMP deletion mutant models (42, 43). It is not always realised that elevation of MMP enzymatic activity sometimes precede elevation of mRNA levels in response to particular stimuli. For example, this can occur following excitotoxic stimuli and seizures (44). In such instances, it has been proposed that transcriptional changes might be better viewed as a response to reductions in cellular reserves of pro-enzyme or transcripts rather than the initial cause of active enzyme release (45). 4. Neurotransmitters Recently, the list of known stimuli has also been expanded to include neurotransmitters. The neurotransmitter serotonin can up-regulate MMP expression in muscle and might also have a similar function in the nervous system (46). By binding to the 5-hydroxytryptamine 5-HT2A receptor, serotonin can inﬂuence muscle MMP-13 expression by signalling through phospholipase C, inositol phosphates and diacylglycerol, PKC and the Shc/mitogen activated protein kinase/extracellular response kinase cascade

May 26, 2005

22

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

(MAPK/ERK) pathway (see below). As MMP-13 is expressed in the central nervous system (47) and serotonin can activate a similar signalling pathway in rat cerebrum, serotonin is likely to also regulate cerebral MMP expression (46). This could provide a means of coupling neural activity to neural plasticity in learning, memory and a range of other nervous system functions. 5. Cell-Cell and Cell-Matrix Interactions Given the importance of MMPs in shaping the extracellular environment, it is not surprising that expression of MMPs can also be inﬂuenced by cellcell (9, 48–50) and cell-matrix (9, 48, 49, 51, 52) interactions, as reviewed previously (e.g. 81, 118). Few studies have examined this in the nervous system. However, one relevant example involves transient contact between human astrocytes and inﬁltrating CD4+ T cells activated by human T lymphotropic virus HTLV-1. The transient cell-cell contact causes changes in expression of astrocytic MMP-2, 3 and 9 and the tissue inhibitors of metalloproteinases TIMP-1 and 3. These alterations are mediated in part by cytokines and in part by eﬀects involving integrin-mediated cell adhesion (9). The relationships between MMPs and integrins are discussed at length in a later chapter. Another situation where cell-matrix interactions alter MMP expression involves the disruption of the hippocampal extracellular matrix that occurs in response to transient ischemia. This triggers post-transcriptional up-regulation of hippocampal MMP-2 and MMP-9 expression, apparently by a mechanism involving G-protein coupled receptor alterations and upregulation of tyrosine-phosphorylated focal adhesion kinase pp125FAK (52). This kinase transduces signals to intracellular compartments from the matrix in general, and from speciﬁc integrins in particular. This transduction is achieved through modiﬁcations of cytoskeletal actin interactions with the cell membrane, triggering downstream cascades (52). Such mechanisms are of additional interest because they also provide potential routes for feedback control of matrix modelling by the MMPs; changes in the composition of the matrix lead to changes in MMP expression, which in turn further alter the composition of the matrix. 6. Physical and Biochemical Stress Expression of MMPs can also be aﬀected by either biochemical stress involving activation of the hypothalamic-pituitary-adrenal and

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

23

sympathetic-adrenal medullary axes (53) or physical stress in the form of unidirectional or oscillatory shear stress on blood vessels or cells. For example, physical stress increases MMP-14 (MT1-MMP) gene expression through the early growth response transcription factor (EGR-1), which is induced within an hour of exposing microvascular endothelial cells to physical strain (54). Physical stresses arising from blood ﬂow turbulence in the vicinity of atherosclerotic plaques can also lead to up-regulation of endothelial MMP-9 gene expression in vasculopathy (24). Similar eﬀects may occur in cerebrovasculopathy and stroke, which are discussed in later chapters.

7. Suppressive Factors As reviewed elsewhere (55), both basal and induced MMP gene expression in nervous system cells can be suppressed by glucocorticoids (12, 56), interferons (IFN) and other agents (57, 58). Notably, suppression through mechanisms involving glucocorticoid receptors is of immediate clinical relevance (56). This is discussed further below. In addition, nonsteroidal anti-inﬂammatory drugs (NSAID) inhibit MMP-2 gene transcription by reducing signalling through the ERK/SP1 route in human lung cancer cells and may exert similar eﬀects in the nervous system (59). Several of the positive regulatory factors listed above can also inhibit MMP expression in some systems, notably cAMP (60), p53 (61), TGFβ (62, 63), TNFα (57) and retinoids (64), as reviewed elsewhere (34, 80, 202). A stimulus can sometimes simultaneously up-regulate one MMP and down-regulate another in the same system. In some systems, including human ﬁbroblasts, TGFβ down-regulates MMP-1 but induces other MMPs, such as MMP-2 (65) and MMP-13 (66, 67).

8. Feedback Control Expression of MMP genes can also be subject to feedback control. For example, in guinea pig retinal M¨ uller glial cells, MMPs are proposed to increase activation of EGF receptors by cleaving pro-heparin-binding EGF, resulting in the release of active ligand from extracellular matrix reservoirs (68). In turn, EGF receptor activation can up-regulate both MMPs (69) and TIMPs (70). Other feedback mechanisms may involve MMPs (e.g. MMP-1, 2, 3, 9) that can degrade proteins such as IL-1β which regulate MMP gene transcriptional activity (71, 72).

May 26, 2005

24

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

Feedback regulation of MMP-1 expression can also occur through discoidin domain-containing receptor-like tyrosine kinases. These are activated by binding of intact collagen, a prominent MMP-1 substrate. This leads to the induction of MMP-1 expression. This inductive mechanism is suppressed by cleaved collagen, which inactivates these kinases (73, 74). Conversely some MMPs, including MMP-1, 2, 3, 7 and 9 (75) and MMP-12 (macrophage metalloelastase) (76) can process pro-TNFα to its mature, active form. Since active TNFα is a potent regulator of MMP expression in many systems, as described elsewhere in this chapter, this may constitute a form of positive, amplifying, feed-forward regulation that could be important in some inﬂammatory scenarios. Another possible route of feedback control involves the Akt pathway, discussed further below, which can up-regulate some MMPs but may also in turn be regulated by one or more MMP (77).

9. Signalling Cascades Involved in Regulating MMP Gene Expression 9.1. The MAPK pathway Many of the above stimuli, including certain potentially suppressive stimuli such as TGFβ (67, 80), signal through three important cascades within the mitogen activated protein kinase (MAPK) signalling pathway. These cascades have all been demonstrated in nervous system models and utilise (i) the extracellular response kinases (notably ERK 1, 2) (6, 18, 78, 79), (ii) the p38 kinases and heat shock protein 27 (33, 34, 78, 79) and (iii) the c-Jun N -terminal kinases (JNK), also known as stress-activated protein kinases (78). The MAPK system has been reviewed extensively by Widmann and colleagues (82) and is also discussed in various articles on the MMPs and TIMPs (34, 45, 80, 81). It can be activated by most of the factors listed in the previous section including cytokines, growth factors, tumour promoters, cell-matrix interactions, endotoxins and environmental stress. Conversely, it can be suppressed by agents such as glucocorticoids or retinoids. The ERK1/2 kinases are proposed to be preferentially activated by growth factors and phorbol esters, with cytokines or environmental stress preferentially activating the JNK and p38 MAP kinases (82). However, this can vary depending on the stimuli, the cellular context and the MMP in question. Besides the MAPK cascades, there are several other well-established pathways now known to be important in MMP gene regulation. Here, we

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

25

will brieﬂy describe four of the better characterised ones. There is also extensive interaction (‘cross-talk’) between the diﬀerent pathways. This has important implications both for data interpretation and for potential treatment strategies that aim to control MMP expression, and will be discussed further below. 9.2. The NF-κB pathway One important pathway is the nuclear factor kappa B (NF-κB) pathway. Complexes formed between receptors for the cytokines IL-1β and TNFα and various accessory proteins activate the NF-κB inducing kinases, setting oﬀ a cascade that leads to degradation of IκB, the cytosolic inhibitor of NF-κB. This releases the p50 and p65 subunits of NF-κB, which then translocate to the nucleus and transactivate MMPs and other genes, as reviewed elsewhere (34). Induction of MMP-9 by all-trans retinoic acid in neuroblastoma cells (11) or by oxidative stress in human brain capillary endothelial cells (23) involves similar mechanisms. In contrast, NF-κB may repress cytokine-induced MMP-3 up-regulation by binding to the IL-1 responsive element in the promoter region (however note that NF-κB can also up-regulate MMP-3 in other circumstances) (83). The –1612del/ins MMP-3 promoter polymorphism (discussed in the ﬁnal section of this chapter) occurs within the IL-1 responsive element of the MMP-3 promoter. This might contribute to variability in how diﬀerent individuals regulate MMP gene expression. 9.3. The ‘JAK/STAT’ pathway Another important pathway in MMP gene regulation involves the ‘JAK/STAT’ system, triggered by factors such as the interferons. In this pathway, downstream signalling through Janus kinases (JAKs) activates the protein “signal transducer and activator of transcription” STAT-1α. Activated STAT forms homodimers or heterocomplex with other proteins and translocates to the nucleus where it regulates transcription in a variety of ways. These can include binding to sites in target gene promoters (e.g. IFNstimulated response elements or γ-activated sequences), interacting with AP-1 binding proteins, co-activators and other components of the general transcriptional machinery and chromatin remodelling (58), as reviewed elsewhere (34, 84, 85). There are complex interactions between this system and other pathways regulating MMP expression. For example, in addition

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

26

T. Roy & E.A. Milward

to collaborating with focal adhesion kinase to transduce MMP responses to alterations in cell adhesion by activating the MAPK ERK cascades, the src family of nonreceptor tyrosine kinases can also act through the STAT system by enhancing DNA binding of STAT complexes (reviewed 34). The STAT system is particularly important since it is one of the few pathways known to be capable of exerting suppressive eﬀects on MMP gene expression in some circumstances (34, 58), as will be discussed further below. 9.4. The PI3K/Akt system The phosphatidylinositol 3 kinase/protein kinase B (PI3K/Akt) system utilises the serine/threonine protein kinase B, also known as Akt (18, 86–88). Activation of Akt by phosphorylation can be triggered by various stimuli, including growth factors and cytokines, and can occur through various routes including an integrin-linked kinase-PI3K pathway (89–91) and a pathway involving phospholipase Cγ-Ca2+ -calmodulin which is independent of PI3K (92). Numerous downstream eﬀectors of the Akt cascade have been identiﬁed including Bad, caspase 9, glycogen synthase kinase (GSK)-3β (89, 93, 94), the forkhead transcription factors AFX and Foxo1 (Fkhr) (95, 96), endothelial nitric oxide synthase (eNOS) (e.g. 92) and AMP-activated protein kinases such as adhesion related kinase 5 (ARK5). This kinase induces expression of MMP-14 (MT1-MMP), which in turn activates MMP-2 and MMP-9 (88). The PI3K/Akt system is of particular interest in the nervous system as Akt is proposed to be a downstream mediator of the neurotoxic eﬀects of MMP-1 (77). 9.5. The Smad pathway The last pathway we will cover here involves the Smad family. Smads were initially identiﬁed as products of the C.elegans Sma and Drosophila Mad genes and are important in pathways involving TGFβ. Type 1 TGFβ receptors phosphorylate C-terminal serine residues of receptor-activated (R-Smads; Smad2, 3), which can then form heteromeric complexes with a common mediator Smad (co-Smad; Smad 4) and translocate to the nucleus to regulate gene expression directly or through various mechanisms involving co-activators and co-repressors. These include eﬀects on turnover of transcription factors that involve increases in the degradation of transcriptional repressors through interactions with ubiquitin ligase. This area has recently been reviewed elsewhere (97). As is described in more detail below,

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

27

Smad proteins appear likely to be important in enabling TGFβ to induce MMP-inhibiting TIMPs while simultaneously attenuating MMP induction by other stimuli such as cytokines. 9.6. Cross-talk and competition between pathways There is a complex network of interactions between these pathways that is only now beginning to be unravelled. Retinoids can signal through both NF-κB (11) and AP-1 binding sites (99) and can synergise with cAMP to up-regulate MMP-2 basal activity in some systems (100). Similarly, as reviewed elsewhere (e.g. 34), synergistic interactions and cross-talk can also occur between cascades triggered by the various growth factors and cytokines (18, 101–103). This has consequences for designing experimental or treatment strategies since genes that can be transactivated through two or more pathways will require more complex suppressive regimens. One important example of competition among the pathways that has implications for tumour metastasis occurs between the PI3K/Akt cascade and the MAPK/ERK cascade and involves insulin-like growth factor 1 (IGF-1). At a concentration of 10 ng/ml, which promotes mitogenesis, cell proliferation and motility in lung carcinoma cell lines, IGF-1 acts through the PI3K/Akt pathway, with concomitant inactivation of Raf kinase and attenuated ERK response, to increase MMP-2 gene expression. This is associated with increased invasiveness. Conversely, with higher IGF-1 (100 ng/ml), ERK activation dominates and MMP-2 expression is reduced (18). For reasons that are not fully understood, the reverse is seen in human breast cancer cells with signalling through MAPK/ERK at lower concentrations and Akt at higher concentrations (98). Which pathway predominates is determined at least in part by ligand load and probably also by diﬀerences in availabilities of surface receptors and downstream substrates (18, 98). There is also the potential for cross-talk between these pathways and other systems including neuronal death pathways since, as mentioned above, Akt is also proposed to be a downstream mediator of neurotoxic eﬀects of MMP-1, which can rapidly dephosphorylate Akt in human neuronal cultures (77). 9.7. Pathways operating in the nervous system Relatively little is known about cascade utilisation in diﬀerent nervous system cell species and regions. However, the few studies available, mainly

May 26, 2005

28

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

on MMP-9, show that while all the above cascades can regulate expression of MMP-9 in nervous system cells, diﬀerent pathways operate in different circumstances. For example, in astrocytes TNFα can act through all three cascades whereas the tumour promoter and PKC agonist PMA appears to induce MMP-9 speciﬁcally through the ERK pathway (15). In glioblastoma cells, on the other hand, PMA induction of MMP-9 and other MMPs and TIMPs occurs via the p38 MAPK cascade (33). In contrast, upregulation of nervous system MMP-9 in response to oxidative stress appears to involve activation of NF-κB rather than MAPK cascades (104). Induction of MMP-9 in primary human or mouse astrocytes and astroglioma lines can be inhibited by IFN-β and IFN-γ through STAT-1α (58).

10. Genetic Elements Regulating MMP Gene Expression 10.1. Constitutive and inducible transcription factors All the above pathways ultimately regulate MMP gene activity through overlapping sets of genetic elements, with signalling by distinct stimuli through the diﬀerent cascades being integrated at the level of the gene. This integration is exerted in part through interacting constitutive or inducible transcription factors. The general mechanisms underlying activation of these diﬀerent types of transcription factors will be covered only brieﬂy here, having been reviewed at length elsewhere, for example by Hager and colleagues (105) and by Herdegen and Leah (106), the two main sources from which the following description is adapted. In one type of response mechanism, the end kinases of the MAPK and other cascades translocate to the nucleus where they phosphorylate (and thereby activate) constitutive transcription factors that are already bound to DNA. One well-known example of a constitutive transcription factor is the cAMP response element binding protein (CREB), which can bind to MMP promoter sites and has been proposed to mediate transcriptional repression of MMP-1 expression following TGFβ treatment (reviewed 80). In a second type of response (exempliﬁed by certain STATs), phosphorylated, constitutive transcription factors residing in the cytoplasm move into the nucleus and associate with other transcription factors and/or activate the general transcription machinery. This can lead to increased transcription of genes encoding inducible transcription factors such as Jun, Fos and Krox family members. The message from these genes is translated in the cytoplasm. Whether or not a particular factor ultimately has an inductive

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

29

or suppressive eﬀect on gene expression is inﬂuenced by post-translational modiﬁcations as well as by prior alternative splicing. The resulting proteins translocate back to the nucleus where they form homodimers or heterodimers with other factors. These dimers then bind to speciﬁc DNA sequences to modulate the transcription of target genes. This may activate the general transcription machinery or instead may act to repress gene transcription. This can occur when binding either does not activate the general transcription factors or prevents other transcription factors from accessing the DNA at nearby sites. Conversely, anything overcoming repressive eﬀects of this nature will transactivate gene expression. The relative concentration of active transcription factors (reﬂecting message and protein turnover as well as dephosphorylation or dimer dissociation rates) and their relative aﬃnities for speciﬁc DNA sequences determine the strength and duration of their interactions with DNA. Most factors are extremely mobile and engage in multiple, rapid, transient interactions with numerous gene targets. Site occupancies or interactions between factors, co-factors and accessory molecules are often in the order of seconds or minutes rather than hours. Hager and colleagues (105) comprehensively review the dynamics of transcription factor interactions. A very thorough review of transcription factor interactions in the nervous system is provided by Herdegen and Leah (106). 10.2. Orchestrating multigene responses Often MMP expression is up-regulated or down-regulated as part of an orchestrated response that involves several MMPs together with other proteins that interact directly or indirectly with MMPs, including ADAMs, TIMPs and plasminogen activators, as described in many previous reports and reviews (34, 55, 80, 81, 107–112). This synchronised response is facilitated in part by common cis- and trans-acting genetic regulatory elements that enable diﬀerent genes to respond in concert. Complex, coordinated alterations in the expression proﬁles of these genes are seen in nervous system-derived cells in response to a range of factors, including cytokines (109) and growth factors (109, 113–115), as well as in many nervous system disorders or injury scenarios (below and other chapters). Similar mechanisms may also be important in compensatory responses in MMP-null mice in which ablation of one MMP leads to up-regulation of others (42, 43). Despite the wide range of neurological conditions involving coordinated expression of the MMPs and other interacting proteins, there is virtually no

May 26, 2005

30

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

information on the mechanisms underlying such orchestrated responses in the nervous system. In other tissues, the coordinated regulation of MMPs and other MMP-interacting proteins in response to stimuli such as growth factors or cytokines involves interelement cooperativity of composite activator protein-1 (AP-1) and polyoma virus enhancer A-binding protein-3/E26 virus (PEA-3/ETS) motifs, NF-κB-like sites, silencer/anti-silencer sequences and other elements (reviews include (34, 45, 80, 81, 116, 117). The few studies to date suggest these elements are also important in regulating MMP gene expression in the nervous system but, not surprisingly, sometimes in very diﬀerent ways to those seen in other tissues. 10.3. AP-1 binding sites and related elements Some of the main 5 elements implicated in regulation of human MMP gene expression are shown in Fig. 1. Regulation through AP-1 binding sites is a major contributor to basal and inducible MMP gene transcription, both peripherally and in the CNS, and has been covered in many previous reviews (34, 45, 80, 81, 116–118). Pathways such as the MAPK ERK and JNK cascades exert many of their eﬀects through diﬀerent dimeric combinations of the Fos and Jun proteins which regulate transcription by binding to AP-1 sites. A proximal AP-1 site is located in the region −65 to −79 bp in most MMP promoters with the notable exceptions of MMP-2, 11 and 14 (see below). Mutations at this site can substantially reduce promoter activity, as reviewed elsewhere (34, 45, 80). In addition to eﬀects exerted at AP-1 sites as a result of activation of MAPK and other systems described above, factors such as retinoids and glucocorticoids which bind nuclear hormone receptors can inﬂuence MMP gene expression through direct or indirect eﬀects on AP-1 sites, usually by means of interactions with Jun or Fos proteins (reviewed 34, 80, 202). Many MMP promoters also contain a second distal site highly homologous to the AP-1 site consensus sequence ((119, 120), reviewed (80, 116) and see also Fig. 1). These distal sites appear to have relatively small eﬀects on basal transcription but contribute to inducible expression in response to stimuli such as growth factors or phorbol esters and other PKC agonists (120, 121) and reviewed (45, 80). In addition, AP-2 sites are found in the promoter regions of some MMP genes and appear essential for transcription of both MMP-2 and MMP-9 in some cell types (122, 123), as reviewed previously (80).

May 26, 2005

31

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

Fig. 1. Promoter regions of the human MMP genes. Elements identified as regulating each promoter are shown as follows: activator protein-1 (AP-1), activator protein-2 (AP-2), polyoma enhancer A binding protein-3 (PEA-3), TGF-β inhibitory element (TIE), SIAT binding element (SBE), CCATT/enhancer binding protein-β (C/EBP), stromelysin-1 PDGF responsive element (SPRE), octamer binding protein (TRE), silencer sequence-1 and -2 (S1 and S2), nuclear factor-κB (NF-κB), SP-1 binding site (Sp1), p53 binding site (p53), nerve growth factor response element (NGFRE), DR1-type retinoic acid responsive element (DR1-RARE), DR2-type retinoic acid responsive element (DR2-RARE), nuclear factor-1 (NF-1).

May 26, 2005

32

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

The AP-1 sites do not operate in isolation. Synergistic interactions occur between these elements and other cis-acting domains such as PEA-3/ETS sites (124 and reviewed 34, 80, 81) and NF-κB sites (119 and reviewed 34, 80). With the exception of MMP-2, PEA-3/ETS binding sites are also present in most MMP promoters examined to date (reviewed (34, 80, 81)). Both basal and inducible MMP gene activity appear to involve cooperative interactions between the proximal AP-1 site and one or more other sites, notably PEA-3/ETS sites (124), as reviewed elsewhere (34, 80). The nature of these interactions depends on the particular transcription factor involved. For example, depending on their structure, diﬀerent ETS factors can either up-regulate or down-regulate induction of MMP-1 through AP-1 (125). In addition to exerting eﬀects more directly through AP-1 sites, kinases such as the ERKs also phosphorylate ETS transcription factors, resulting in either activation or inactivation, again depending on the particular factor involved (reviewed 34). Down-regulation of MMP expression is also exerted through other motifs. For example, several MMP genes contain TGFβ inhibitory elements (TIEs), which can mediate suppressive eﬀects of TGFβ on MMP gene expression through mechanisms involving binding of c-Fos or other factors ((126) and reviewed elsewhere (34, 80); see also Fig. 1). 10.4. Other promoter elements Other important MMP promoter sites include SP1 binding sites (reviewed brieﬂy elsewhere (80, 127)) and inverted CCAAT boxes that bind NF-Y transcription factors (109). The zinc-ﬁnger SP1 transcription factor can produce bends in DNA to bring together distant regulatory sites (128, 129). It can interact with AP-1 (119) and directly modify the basal transcription complex, as occurs for MMP-2 (130), and can also modulate inducible activity, often in a tissue-speciﬁc fashion, as brieﬂy reviewed elsewhere (80, 127). The SP1 and NF-Y transcription factors can also interact — for example, signalling triggered by cAMP leads to a delayed up-regulation of TIMP-2 expression via cooperation between SP1 and NF-Y (109). A number of other human MMP promoter elements are also shown in Fig. 1. Some of these are likely to be important in the nervous system, for example the nerve growth factor response element (NGFRE) and the MMP-3 PDGF responsive element. However, as yet there is still little if any information available on the functioning of these elements in nervous system derived cells. [Note that conversely some sites known to be important

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

33

in MMP gene expression in other species have not yet been investigated in humans.]

10.5. Chromatin remodelling Another recently proposed mechanism for modulating MMP gene expression is chromatin remodelling. The metastases-associated gene MTA1 encodes a protein that binds histone deacetylases 1 and 2 and is one of the subunits of the human nucleosome remodelling and deacetylation complex. Binding of MTA1 protein to the MMP-9 promoter can suppress MMP-9 expression without apparently aﬀecting induction through AP-1, SP1 and NFκB sites, possibly through alterations in chromatin conformation that limit access to the basal transcription machinery (131). As noted in the previous section, interferon suppression of MMP-9 gene transcription might also involve chromatin remodelling (58). These modes of MMP gene regulation are still not well understood but probably also operate in nervous system cells, with a recent study reporting that an indirect, methylationmediated mechanism contributes to dysregulation of the TIMP-2/MMP equilibrium in neuroblastoma cells (132).

11. Regulation of the MMP-2, MMP-11 and MMP-14 (MT1-MMP) Genes 11.1. The MMP-2 gene The MMP-2 gene diﬀers from most other MMP family genes in several important features, including the lack of both a proximal AP-1 element and a TATA box. For this reason and because it does not always respond to stimuli that regulate other MMPs (1), MMP-2 has sometimes been considered to be unregulated. However, recent sequence analysis has identiﬁed potential cis-acting elements in the MMP-2 promoter. These include binding sites for AP-1 (133), p53, AP-2 and Y-box transcription factor YB-1, all reviewed brieﬂy by Mertens (117), as well as binding sites for SP1 and SP3 (117, 130, 133), GATA-2 (134), the CCAAT/enhancer-binding protein (C/EBP) and a cAMP response related element (100). Some of these, including the AP-1 (133), AP2, p53 and YB-1 (117, 130) and GATA-2 (134) binding sites have also been conﬁrmed to have functional relevance, although this depends on the system being studied. For example, the human MMP-2 gene contains a region referred to as the r2 enhancer element in

May 26, 2005

34

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

its 5 region that binds AP-2, YB-1 and p53. This element is highly conserved across species and is functionally important in many cells but is non-functional in some human astroglioma cells (130).

11.2. The MMP-11 gene The MMP-11 or stromelysin 3 (ST3) gene also has unusually regulated gene expression as it has two promoters. The second one is a human-speciﬁc, inducible promoter accessible to nuclear factors such as C/EBP and retinoic acid receptors. (In general, only a very restricted set of factors appears able to regulate MMP-11 gene expression compared to other MMPs.) This second promoter controls expression of an alternatively spliced transcript that encodes a highly unusual MMP protein, termed ST3β, which lacks a signal peptide and is produced as an already active MMP that remains within the cell (135). Relatively little is known about MMP-11 in the nervous system. It is expressed in neuroepithelial cells during development (136), is constitutively expressed in both mouse (107) and human nervous system (137) and by astrocytes in vitro (138) and, unlike most other MMPs examined to date, does not appear to show altered expression in models of infection such as meningitis (137) or lipopolysaccharide stimulation (138). As far as we are aware, the alternatively spliced isoform has not been investigated in the nervous system.

11.3. The MMP-14 gene Finally, the gene promoter for the membrane-associated MMP-14 (or ‘membrane type 1-MMP’; MT1-MMP) also diﬀers from most other MMP promoters so far characterised. Apart from one motif with high homology to the SP1 site consensus sequence, three motifs highly homologous to the TGF-β1 inhibitory element and an NF-κB site, it has very few elements resembling consensus transcription factor binding sites. However, it does have some resemblance to the MMP-2 gene promoter in that it lacks a TATA box and proximal AP-1 site and contains multiple (four) transcription start sites (139). As will be discussed in other chapters, MMP-14 is important in producing active MMP-2 from the inactive, pro-MMP-2 prozyme. Consistent with this, the MMP-2 and MMP-14 genes are co-regulated in some circumstances, often in conjunction with TIMP-2, as reviewed elsewhere (81),

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

35

although only a small amount of information on this is available in the nervous system (140). On the other hand, there are also considerable diﬀerences in how the MMP-2 and MMP-14 genes are regulated. For example, transcription of the MMP-14 gene is strongly up-regulated in endothelial cells by the transcription factor EGR-1 (mentioned above), which regulates a set of genes involved in vascular remodelling and can compete with SP1 for promoter binding. In contrast, there do not appear to be EGR-1 binding sites in the rat MMP-2 promoter (141). 12. Regulation of ADAMs Family Gene Expression Another family of metalloproteinases that may share certain regulatory control mechanisms with the MMPs is the ADAMs family (a disintegrin and metalloproteinases). As reviewed elsewhere (142–144), the ADAMs belongs to the adamalysin subgroup of the MB metzincin superfamily, to which the MMPs also belong. Structural resemblances between the ADAMs and the MMPs include the N -terminal signal peptides as well as the existence of similar propeptide forms containing a conserved cysteine ‘switch’ residue that binds the metalloproteinase catalytic zinc atom at the active site (145). There are over 30 diﬀerent ADAMs family gene products and related proteins such as the ‘ADAMs with thrombospondin motifs’ ADAM-TS (146), reviewed elsewhere (147) and decysin ADAMDEC 1 (148). Of these, at least 17 are normally expressed in the adult CNS, as reviewed previously (149, 150) and see also more recent research reports (151–153). The ADAMs have numerous functions, including many that overlap with MMP activities, such as roles in cell-matrix or cell-cell adhesion and migration (154). As reviewed previously (149), the family members examined to date appear to be expressed in most brain areas, predominantly in astrocytes and endothelial cells (155). Increased levels of ADAMs gene expression occur after excitotoxic insults (156) and other nerve injuries (157, 158). One important role of ADAMs in the nervous system may involve their ability to release soluble protein species from membrane-bound precursors. This is also referred to as ‘convertase’ or ‘sheddase’ activity, as the membrane-bound proteins are converted to soluble forms that are shed into the extracellular matrix. This is reviewed elsewhere (143, 159, 160). The archetypal convertase is ADAM-17 (also known as TNFα converting enzyme because of its ability to release soluble TNFα from its membranebound precursor). This area has been the focus of attention in recent

May 26, 2005

36

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

years because several ADAMs are able to generate soluble forms of the Alzheimer’s disease amyloid precursor protein (APP) in this way. This process can involve cleavage at sites within the amyloidogenic Aβ peptide domain, preventing the formation of pathogenic amyloid, with potential implications for Alzheimer’s disease therapies (161). This is reviewed elsewhere (162, 163), including in a later chapter. Some ADAMs might also have roles in other disease such as multiple sclerosis (MS), which is also discussed in a later chapter, since increases in ADAMs have been observed in the cerebrospinal ﬂuid (CSF) of MS patients as well as in acute and chronic active MS plaques (152). However it has been suggested that these increases primarily reﬂect ADAMs from inﬁltrating peripheral T -lymphocytes rather than up-regulation of constitutive ADAM expression in astrocytes and other endogenous nervous system cells (152). The signalling pathways regulating the ADAMs are still relatively undeﬁned but share at least some common mechanistic features with MMP signalling pathways. The main MAPK cascades (i.e. ERK, JNK, p38), and transcription factors such as the AP-1 family and NF-κB all appear able to transduce modulation of transcription of at least some ADAMs family genes (164). Various ADAMs are also involved in transactivation of the EGF receptor and related downstream signalling through MAPK pathways by G-protein coupled receptor ligands (165). In addition, epigenetic silencing of ADAM-23 expression through hypermethylation of the promoter region occurs in some breast malignancies and may be related to reduced cellcell or cell-matrix adhesiveness and increased metastatic potential (166). Yet, despite their probable importance in various serious diseases, almost nothing is known about how the expression of the ADAM’s family genes is regulated in general, let alone in the nervous system.

13. Regulation of TIMP Gene Expression The tissue inhibitors of metalloproteinases (TIMPs; described in more detail elsewhere in this book) are also regulated by many of the same factors that regulate MMPs. For example, growth factors and cytokines such as TGFβ, TNFα and interleukins (36, 167), retinoids and glucocorticoids (36), reviewed (34, 35, 202), have all been reported to increase TIMP-1 and TIMP-3. Conversely the tumour promoter and protein kinase C activator phorbol 12-myristate 13-acetate (PMA) has been reported to down-regulate TIMP-1 and TIMP-2 in glioblastoma cells (33). The TIMPs have been

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

NF- B

AP-1

PEA3

Sp1

GATA

37

GC Box

+1 TATA

CCAAT

Fig. 2. Representation of common promoter region motifs of the human TIMP-1-4 genes. Elements that have been identiﬁed as regulating a TIMP member are shown as follows: activator protein-1 (AP-1), polyoma enhancer A binding protein-3 (PEA-3), nuclear factor-κB (NF-κB), and SP-1 binding site (Sp-1).

reviewed at length in many recent articles (34, 45, 110–112, 168–170) and are also discussed in depth in other chapters. As mentioned above, the TIMP genes can sometimes be coordinately regulated in concert with MMP genes. Some of the regulatory elements identiﬁed within the TIMP gene promoter regions are summarised in Fig. 2. None of TIMPs 1–3 has a classical TATA box. However, all contain a 22-bp ‘serum response element’ 75 base pairs upstream of the major transcription start site that contains binding sites for AP-1, STAT and PEA-3/ETS. This element is essential for basal and inducible transcription (1, 113, 170–173) and has been discussed in many other reviews (34, 45, 118, 170, 174). The regional and cellular expression patterns of the TIMPs during embryonic and subsequent development of rodent CNS have been recently reviewed by Crocker and colleagues (170) and suggest possible roles in neurogenesis and migration (64, 175–177). Much of the information presently available on TIMP gene expression relates to TIMP-1. At least 6 AP-1 and 12 AP-2 binding sites, 6 PEA-3/ ETS sites, 10 SP1 sites, 5 CCAAT boxes and 6 transcription start points are contained within the TIMP-1 promoter (171–173, 178), along with other less well-characterised regulatory motifs (174, 179, 180) and repressive elements binding SP1, SP3, ETS-related and other factors (181). Elements downstream (3 ) from the transcription start sites can also modify TIMP gene transcription, including an element spanning the exon1/intron1 boundary (173, 182). While these motifs may not all be functional, the presence of so many potential regulatory elements within the gene are consistent with the possibility that the mechanisms controlling expression of this gene are extremely complex, as might be expected from the intricate behaviours of the TIMP proteins. The TIMP-2 promoter also contains AP-1, AP-2, PEA-3/ETS and SP1 binding sites (197). However, it diﬀers from TIMP-1 and TIMP-3 in being

May 26, 2005

38

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

refractory to regulation through cytokines, PKC-dependent tumour promoters and AP-1 or PEA-3/ETS binding sites in some cell systems. For this reason its regulatory features have been proposed to be closer to those of housekeeping genes than to other TIMPs (197). Nonetheless, TIMP-2 gene expression can be modulated in nervous system-derived and other cells. For example, the DNA methyltransferase inhibitor 5-azacytidine can act through a mechanism involving an NF-Y site in the TIMP-2 promoter to resurrect TIMP-2 activity in SH-SY5Y neuroblastoma cells, leading to a decrease in invasiveness (132). Regulation of TIMPs in nervous system disease and injury is discussed further below.

14. Inverse Regulation of MMPs and TIMPs Insight into how the MMPs and their naturally occurring inhibitors the TIMPs can be inversely regulated through common sites and transcription factors may come from considering the actions of TGFβ, which is able to exert reciprocal regulatory eﬀects on MMP and TIMP genes (70). In human ﬁbroblasts, TGFβ regulates both the MMP-1 and TIMP-1 genes through c-Fos, c-Jun and JunD and the proximal AP-1 site. However, while TGFβ acts to induce TIMP-1 through this pathway, in contrast TGFβ is proposed to attenuate the induction of MMP-1 that occurs through the AP-1 pathway in response to agents such as cytokines or the tumour promoter PMA. This attenuation is thought to involve Smad proteins (described earlier in this article). The Smad proteins interact with the proximal AP-1 site of the MMP-1 gene to suppress MMP-1 expression induced through this AP-1 site in response to cytokines and other agents (187). Similarly Smads have also been implicated in attenuation by TGFβ of IL-1 induction of MMP-1 through competition with NF-κB (188).

15. Regulation of TIMPs Independent of MMP Inhibitory Functions Although the presence of common regulatory motifs enables coordinated regulation of TIMP and MMP genes, the TIMPs can also be independently regulated and have various functions distinct from their roles in MMP inhibition. These are reviewed elsewhere (55, 81, 110–112, 169, 170, 198, 199) and include cell-type speciﬁc growth-promoting or growth-suppressing

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

39

actions. Growth-related eﬀects of TIMPs can involve attenuating the downstream signalling of tyrosine kinase-type growth factor receptor by reducing receptor phosphorylation in response to growth factor ligands (e.g. EGF, bFGF, PDGF and vascular endothelial growth factor VEGF). These TIMP eﬀects can also involve activation of both the GTP-binding protein Gαs and adenylate cyclase, leading to increases in cytosolic cAMP second messenger proximal to the ERK pathway (200). Additional activities include anti-apoptotic or anti-excitotoxic activities (TIMP-1,2) and pro-apoptotic (TIMP-3,4) activities that might involve stabilisation of cell surface TNFα receptors through actions on a receptor shedding ADAMs family member or other non-matrix metalloproteinase (170, 193). The TIMPs can also interact with G protein and cAMP signalling pathways by means of MMPdependent links (200). There is relatively little information about TIMP actions within the nervous system that are independent of TIMP inhibition of MMPs. This is partly because few studies have been performed in this area but also because other functions of TIMPs often still appear to require interactions with MMPs, although not necessarily of an inhibitory nature. This can make it diﬃcult to establish whether particular TIMP functions are indeed independent of contemporaneous inhibition of metalloproteinase activity. For example, as discussed further below, hippocampal neurons are protected against glutamate-induced excitotoxic injury by TIMP-1. This occurs through a mechanism involving reduction of calcium inﬂux that is not blocked by a broad spectrum MMP inhibitor yet still seems to require the interaction of TIMP-1 with one or more MMPs (193). Irrespective of the extent to which particular TIMP actions are independent of eﬀects on MMPs, as will be discussed further below, the ability to confer neuroprotection against excitotoxic injury or neural damage in other settings has potential clinical relevance, making TIMP regulation in the nervous system of even greater interest.

16. Post-Transcriptional Regulation of MMP Expression This chapter focuses mainly on alterations in gene expression at the transcriptional level. However, changes in MMP protein levels can also reﬂect post-translational eﬀects such as changes in protein turnover or alterations in MMP mRNA stability occurring in response to cytokines or other factors (201–204), as reviewed previously (e.g. 34, 81). Another mode of regulation

May 26, 2005

40

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

may involve translational eﬃciency, where an increase in the proportion of mRNA that is associated with the polysomes is proposed to increase the rate of MMP synthesis without any alteration in mRNA transcription (205). Again, virtually nothing is known about the contributions of mRNA stability and other post-transcriptional mechanisms to MMP expression in the nervous system. The post-translational regulation of MMPs will be covered in a following chapter. 17. Regulation of MMP Gene Expression in the Healthy Nervous System The mechanisms controlling tissue-speciﬁc and cell type-speciﬁc expression of MMPs and related proteins are not well understood in general (reviewed 34, 80), let alone in the nervous system, where few relevant studies have been performed to date. Basal expression of MMPs within the healthy nervous system is generally low or non-detectable. However, improved antibodies and increasingly sensitive techniques are revealing that there is more constitutive expression of MMPs in the nervous system than was previously thought, although this is sometimes of uncertain physiological signiﬁcance. For example, while some earlier studies did not detect MMP-9 protein in normal adult rat hippocampal neurons or other brain regions, it has now been detected by Western blotting and immunohistochemistry, with mRNA demonstrated by in situ hybridisation (44, 175). Yet zymography suggests that while low levels of MMP9 activity are present in adult cerebellum (175), much of the hippocampal MMP-9 is not normally active (44, 206), although it could act as a reservoir for rapid activation in response to stressors. It is outside the scope of this chapter to review in depth the proﬁle of constitutive or induced MMP expression in the nervous system, especially as this varies enormously between regions and cell types as well as between species and has been addressed at length in various previous reports (17, 107, 175) and reviews (45, 150, 207). In brief, constitutive expression of a subset of MMPs and TIMPs, including MMPs 1–3, 7, 9, 11–14 and 24 and TIMPs 2–4, has been demonstrated in rodent or human nervous system cells by real time PCR and other techniques (107, 175, 208–213). Among the more recently discovered MMPs, the membrane-associated MMP-24 (MT5-MMP) is of interest as it is preferentially expressed in brain, with strongest expression in hippocampus and cerebellum (64, 177, 214). As noted above, relatively few studies have examined ADAMs in the nervous system. However, many ADAMs are constitutively expressed in

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

41

the adult CNS (149, 150), with expression reported in various cell types including neurons and oligodendrocytes (e.g. ADAM-8 (215)) and astrocytes and endothelial cells (e.g. ADAM-10 and ADAM-17 (152, 155)). All major nervous system cell types, namely neurons (44, 64, 175, 216), astrocytes (44, 138, 175, 216–218), microglia (19, 47, 209, 219, 220), oligodendrocytes (221, 222), Schwann cells (2, 223–225), choroid plexus epithelial cells (107, 176, 226), vessel endothelial cells (12, 14, 23, 209) and nervous system progenitor cells (175, 216, 227, 228) have been reported to express various MMP or TIMP genes. However, both mRNA and protein studies point to large diﬀerences in MMP expression not only between diﬀerent cell species but in diﬀerent brain regions and even between diﬀerent locations within a cell. In addition to the orchestrated expression of MMPs and TIMPs by a given cell or cell species described above, it is also probable that the diﬀerent nervous system cell species exhibit interdependent co-regulation in some circumstances. Again more research is required in this area. 18. Neuronal Expression As already noted, neuronal expression of particular MMPs diﬀers between diﬀerent brain regions. This is illustrated by consideration of MMP-9, one of the better characterised, neuronally expressed MMPs. Expression of MMP-9 message has been detected in both limbic system (e.g. piriform cortex, amygdala) and nonlimbic system (e.g. neocortex) structures and generally appears to parallel protein levels (44, 229). One preferential site of MMP-9 expression is the hippocampus, where message for MMP-9 is predominantly localised to restricted subsets of neurons (e.g. pyramidal neurons but not dentate gyrus granule neurons), with MMP-9 mRNA present both in the perinuclear region and in dendrites. In contrast, in the rat cerebellum, MMP-9 mRNA expression appears to be conﬁned predominantly to the cell bodies of granule and Purkinje neurons, along with MMP-2 and MMP-3 transcripts (44, 175). This delimited intracellular distribution appears to be speciﬁc to a restricted subset of MMPs since MMP-24 mRNA is present at high levels in both the soma and the dendrites of these cells (44). Both MMP-24 message and protein are also present in hippocampal dentate gyrus neurons and CA1, CA2 and CA3 subﬁeld neurons in both the soma and dendrites and, to a lesser extent, in thalamus and olfactory bulb neurons (64, 214, 230). In general, MMP-2 message and protein is not as strongly expressed as MMP-9 and is not preferentially expressed by the hippocampus but more

May 26, 2005

42

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

uniformly distributed within brain gray matter, primarily in glia although also present at low levels in dentate gyrus granule neurons (44). Surprisingly few studies have examined the genetic regulation of MMP expression in neurons. Speciﬁc regulatory mechanisms operating in neurons have the potential to aﬀect nervous system function and dysfunction in diverse settings and may well diﬀer from mechanisms operating in other cell types. An in vitro comparison of various cytokines and growth factors, including bFGF, TGFα and TGFβ, ciliary neurotrophic factor (CNTF), TNFα and IFNβ, found that only IL-1β up-regulated MMP-9 expression in neurons, with slight down-regulation by TGFβ (17). There are also likely to be diﬀerences between diﬀerent types of neurons. 19. Astrocyte Expression In contrast to neurons, astrocytes appear to primarily express MMP-2 constitutively (17, 44, 138, 219). While low levels of apparently constitutive astrocytic MMP-9 or MMP-3 expression have been observed in some in vitro studies (17, 224), this may reﬂect activation of astrocytes during the culture process. Astrocytes also constitutively express TIMPs 1–3 at high levels in vitro (224, 231). Astrocyte expression of MMPs and TIMPs can be up-regulated by inﬂammatory mediators such as lipopolysaccharide and cytokines. Interleukins (in particular IL-1β) have been reported to up-regulate MMP-1, 2, 3, 9, TIMP-1 and other related species and TNFα can increase MMP-9 expression (218, 224, 232–234). These eﬀects can be suppressed by glucocorticoids or IFNγ (218, 232). The up-regulation of TIMPs by reactive astrocytes has been proposed to be part of attempted protective or reparative responses to nervous system diseases (111). Responses to chronic inﬂammation may diﬀer from responses to acute inﬂammatory stimuli since prolonged astrocyte treatment with IL-1β eventually leads to reductions in MMP-2 and TIMP-1, although MMP-1 remained elevated (234). It is unclear whether such longer term changes are ultimately beneﬁcial or instead epitomise the failure of attempted compensatory responses. 20. Microglial Expression Microglia in vitro have been reported to express MMP-2, 9, 13 and 14 (MT1-MMP) constitutively (19, 47, 219, 232). Cytokines such as IL-1β, TNFα and IFN-γ can up-regulate MMP-2, 3 and 9 in adult rat microglia

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

43

in vitro (19). These cytokines do not seem to regulate microglial MMP-13, although microglial MMP-13 can be down-regulated by TGFβ (47), in contrast to the up-regulation that occurs in human ﬁbroblast cells which was mentioned above (67). Induction of MMP-9 by TNFα and IFN-γ can also be suppressed by TGFβ (47). Up-regulation of MMP-9 gene promoter and mRNA accumulation in primary rat microglia cultures in response to the inﬂammatory mediator lipopolysaccharide can be inhibited by estrogen (28). However, while the genetic factors regulating microglial MMP gene expression are likely to be important in diverse neuroinﬂammatory conditions, as well as many other neuropathological circumstances, again there is little information available in this area. 21. Endothelial and Other Cells While both MMP-2 and MMP-9 are also constitutively expressed by rat nervous system endothelial cells, IL-1β and TNFα selectively up-regulate only MMP-9 and not MMP-2 or TIMP-2 in rat CNS endothelial cells in vitro (12). This up-regulation can be partially inhibited with the glucocorticoid steroid dexamethasone (12). While there is some variation between human studies, probably relating to methodologies, healthy human cerebral endothelial cells appear to express MMP-3 and MMP-9 but not MMP-1, with MMP-2 mRNA but without any protein detected (14, 209). Consistent with the ﬁndings in rat, transcripts for MMP-3, 9 and 12 but not MMP-2 have been reported to be up-regulated in human endothelial cells in response to TNFα, although zymography suggests only the MMP-3 enzyme is active in either control or TNFα-treated cells and not MMP-2 or 9 (14). These responses may be relevant to vascular remodelling in nervous system lesions. Secretion by choroid plexus epithelia of active MMP-2 and MMP-9 into the cerebrospinal ﬂuid (CSF) is also increased by pro-inﬂammatory cytokines, and probably contributes to increased levels of CSF MMPs in neuroinﬂammatory diseases such as HIV dementia (233). However, it has been suggested that this increase may involve primarily post-translational mechanisms (226). 22. Oligodendrocyte and Schwann Cell Expression While perhaps more commonly perceived to be associated with demyelination in conditions such as multiple sclerosis and other neuroinﬂammatory

May 26, 2005

44

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

diseases, in some settings, including following injury, the MMPs (and also the TIMPs) are also proposed to facilitate myelination of nerve axons in both the peripheral and the central nervous system. In the peripheral nervous system, myelin is produced by Schwann cells. These cells produce high amounts of MMP-3, with Schwannomas producing MMP-2, 3 and 9 (224). In the CNS, myelin is produced by oligodendrocytes. While there is again almost no information on MMP production by oligodendroglial cells in the literature, activation of PKC has been shown to induce elevation of MMP-9 expression in oligodendrocytes and this is proposed to be important in the extension of oligodendroglial cell processes during both developmental myelination and remyelination associated with disease or injury (221). As discussed elsewhere in this chapter, the TIMPs appear likely to have various neuroprotective roles and might also be involved in myelinationpromoting eﬀects. Schwann cells constitutively express TIMPs 1–3 at high levels in vitro (224). 23. Expression of TIMPs in the Nervous System In general, basal expression of TIMP-1 is very low in the normal, unstimulated adult CNS (175, 183, 189) and is restricted to particular cell types such as cerebellar Bergmann glia, some cerebellar Purkinje neurons (175) and adult hippocampal cells (183). In contrast, TIMPs 2–4 are constitutively and relatively abundantly expressed in many regions of the adult CNS, as reviewed previously (170). The most recently discovered TIMP, TIMP-4, is of interest since the CNS, in particular cerebellum, is one of the few tissues in which it is expressed (other major sites are heart and ovaries) and levels of its major transcripts increase through embryonic and postnatal development to adulthood (183). However, as yet, little is known about the mechanisms by which it is regulated. Based on RNA levels estimated from Northern hybridisation, the most abundantly expressed TIMP in the CNS at all times before and after birth appears to be TIMP-2, which is strongly expressed by neurons (175, 183). Message for TIMP-3 is more diﬀusely distributed than TIMP-2 message and has been reported in the choroid plexus, meninges, thalamic neurons and cortical astrocytes and endothelial cells (177, 183). It has also been reported in cerebellar granular neurons, molecular layer interneurons and Purkinjes cells (175) although there may be some disparities between studies depending on the detection method and probes (183). Expression of TIMP-4 message appears to be most prominent in cerebellum as well as in dorsal root ganglia (183).

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

45

Expression of the TIMP-1 gene can be up-regulated in various rat and human cell systems by cytokines such as TNFα, interleukins IL-1β and IL-6 and oncostatin M as well as by PKC-dependent tumour promoters such as 12-O-tetradecanoylphorbol-13-acetate (TPA), through mechanisms that can involve increases in TIMP-1 mRNA stability (184–186) as well as actions on the proximal AP-1 and PEA-3/ETS sites. These sites are also important in basal expression, as reviewed elsewhere (170). Regulation of TIMP-1 following nervous system injury is discussed further below. 24. Expression in Nervous System Development and Plasticity Changes in expression of MMP and TIMP genes occur during the development of the nervous system in association with cortical neuritogenesis and CNS vascularisation (176, 183, 216, 227, 228). This has been reviewed previously (150, 170, 194). Conversely, some MMPs may also have roles in programmed cell death during neurogenesis or in neuronal cell death at later stages in life. Withdrawing nerve growth factor (NGF) from sympathetic neurons in vitro causes apoptosis, involving induction of a wave of immediate early genes, such as c-fos and c-jun, followed by transient upregulation of at least two MMPs, including MMP-1 (235), which has now been shown to be neurotoxic in vivo (236). Diﬀerentiation of pheochromocytoma-derived PC12 cells and neuritic process extension in vitro in response to NGF or FGF is accompanied by increases in MMP-3 expression (237, 238) through a mechanism involving the zinc ﬁnger transcription factor EGR1/NGFI-A and p21 (WAFI) that is independent of MAPK activation (5). Similarly, induction of MMP-2 by NGF has been reported to be required for dorsal root ganglia neurite outgrowth (242). Furthermore, diﬀerentiation to mature CNS cell species in vitro is accompanied by selective alterations in speciﬁc MMPs and TIMPs, notably reduction of MMP-2 and TIMP-4 (227). Following from such ﬁndings, it has been proposed that treatment with MMPs may help overcome the restricted regenerative response of the adult mammalian nervous system (243, 244). Increased expression of select MMPs and TIMPs is observed in association with plasticity during normal functions such as long-term potentiation (LTP), learning and memory, as the adult nervous system ages and in the course of regeneration during injury repair (see also below), consistent with roles in various important nervous system activities (44, 175, 176, 190, 239–241). This has also been reviewed previously (111, 150, 170, 194).

May 26, 2005

46

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

The extension of neuronal cell processes is a common feature in most of the neurodevelopmental and neuroplastic phenomena described above. The growth cone at the tip of an extending neuronal cell process secretes MMPs, plasminogen activator and other proteases that carve a tunnel through the matrix ahead of the cone, as reviewed elsewhere (299). Detection of MMP mRNA in dendrites as well as in cell bodies might indicate local translation of MMPs in actively growing processes or in the vicinity of select, activated synapses, as has been observed for some other mRNAs (45). This is also reviewed elsewhere (245, 246). Glial cells such as oligodendrocytes also secrete active MMPs at the growing tips of their cell processes and this may inﬂuence myelination and other glial functions (222). Almost nothing is known about how diﬀerential expression of genes encoding MMPs, TIMPs and related molecules in diﬀerent CNS regions and diﬀerent cell types is regulated in the healthy nervous system. The mechanisms involved are likely to incorporate many of the same control elements that regulate diﬀerential expression in other systems. However, very little has been published on the transcriptional regulation of MMP expression in neurons or other nervous system cell species. 25. MMP Gene Regulation in Nervous System Injury and Disease Similarly, surprisingly few studies have examined the pathways and elements involved in the regulation of MMP gene expression in nervous system injury and disease. In other parts of the body, cytokines and other inﬂammation- and disease-associated stimuli frequently act through the same downstream signalling pathways that are activated in physiological MMP regulation. This probably also occurs in the nervous system. Much of the research in this area has used nervous system tumour-derived cells but more information is now also starting to appear on MMP gene regulation in nervous system injury and in neuroinﬂammatory and other diseases. As the latter are covered in depth in the following chapters, here we will only focus on gene regulation in nervous system tumours and injury. 26. Regulation in Nervous System Malignancies Expression of MMP genes is up-regulated in various nervous system tumours including neuroblastomas, gliomas, schwannomas and meningiomas, as reviewed elsewhere (2, 118, 130, 194). For example, the MMP-2 gene

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

47

is highly expressed in gliomas and schwannomas and its levels often correlate particularly well with tumour invasiveness, as has also been reviewed previously (2, 57, 118, 130, 194). Constitutive MMP-2 gene expression in human astrogliomas requires functional SP1A and AP-2 binding sites and is transactivated by the transcription factors SP1, SP3 and AP-2 (130). Factors such as TNFα and IFNγ (57, 130), cAMP (reviewed (2)) and TGFβ (8) that can control MMP-2 gene transcription through these or other elements could therefore potentially modify tumour metastatic potential. Two silencer elements (S1, S2) adjacent to the p53 site can negatively regulate MMP-2 promoter activity in human astrogliomas and may have potential therapeutic relevance (130). The ability of the growth factor CNTF to up-regulate TIMP-2 mRNA expression by human SK-N-BE neuroblastoma cells without substantially altering MMP-2 expression may provide another possible approach to reducing neuroblastoma invasiveness (114). Other groups have proposed similar ideas (109, 132). However, experimental evidence for this has been mixed — while overexpression of TIMP-1 reduces invasiveness of astrocytoma cells in some studies (e.g. 247), TIMP-1 was not particularly eﬀective in reducing invasiveness in a rodent glioma model (248). Moreover, the ability of TIMPs to promote growth appears to counteract their tumour suppressive capabilities in some circumstances (249). Activity of the MMP-9 promoter is also heightened in many CNS tumours or tumour-derived cell lines, again often in association with increased metastatic potential as well as resistance to diﬀerentiation or to apoptosis (250, 251). The NF-κB cascade has been implicated in elevated MMP-9 expression both in human SK-N-BE 9N neuroblastoma cells treated with alltrans-retinoic acid (11) and in rat C6 glioma cells following over-expression of the atypical (calcium and diacyl glycerol insensitive) protein kinase C isoform PKC-ζ (30). Up-regulation can also occur by MAPK routes. Dominant negative expression of JNK or ERK-1 or synthetic inhibitors reduce MMP-9 promoter activity in the human gliobastoma cell line SNB19, decreasing invasiveness (252, 253) and also suppressing angiogenic eﬀects, as gauged by reduced formation of capillary-like structures in co-cultures of the glioblastoma cells with endothelial cells (254). Interferons can suppress induction of MMP-9 expression in human astrogliomas as well as in primary astrocytes whereas TNFα induces MMP-9 in these systems (58). Regulation of MMP gene expression through MAPK pathways is also believed to contribute to the seemingly paradoxical propensity of irradiation at sublethal, therapeutic doses to increase the invasiveness and possibly

May 26, 2005

48

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

also the growth of human glioblastomas. This appears to reﬂect changes in BAX and other BCL-2 family proteins (255), as well as eﬀects transduced by p53 and the Y -box binding transcription factor YB-1 in response to ionising radiation (256–260), as reviewed elsewhere (115, 117). These responses lead to increased MMP through generation of reactive oxygen species and reactive nitrogen species and activation of AP-1 through a cross-talk mechanism involving activation of EGF receptor, cytokine receptors and associated kinases including ERK, c-Jun kinase and the serine/threonine protein kinase CK2 (103, 260, 261). This has been reviewed previously (25). Interactions between MMPs and the αv β3 integrin and perhaps other integrins are also important in irradiation-enhanced motility as well as in constitutive motility (255). Controlling these various eﬀects might reduce the high (90%) rate of relapses following radiotherapy, most of which originate from the immediate vicinity of the irradiated region (255).

27. Regulation in Nervous System Injury Like many molecules that participate in inﬂammatory responses, increases in MMP activity have often been considered undesirable and an unambiguous target for therapeutic inhibition but it is now clear that inhibiting MMP responses in injury or disease can have both advantages and disadvantages that need to be carefully considered. This issue is of immediate clinical relevance. The synthetic glucocorticoid methylprednisolone is already approved by the Food and Drug Administration for treating patients with acute traumatic spinal cord injury. It has been reported to induce TGFβ and TIMP expression in response to trauma and to reduce AP-1 and NF-κB activation and expression of MMP-1 and MMP-9 through a mechanism involving glucocorticoid receptor activation (56, 262). These actions could conceivably also be beneﬁcial in other settings. [However it should be noted that as no persistent eﬀects on MMP-9 activity in CSF were observed a week after completing methylprednisolone treatment in patients with multiple sclerosis or optic neuritis, at least some of these eﬀects might be of only limited duration (263, 264).] Down-regulation of nervous system expression of MMPs in human patients is therefore immediately feasible but it is not yet clear to what extent this may be detrimental rather than advantageous. This is likely to depend on various factors, including the MMP in question, the type of injury or disease and temporal considerations. For example, downregulation of certain MMPs might be beneﬁcial immediately following an

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

49

injury but harmful at later times, with timing depending at least in part on the type and severity of injury. This is another area where large amounts of work still remain to be done. Both primate (265) and rodent models of nerve injury, whether physical or caused by excitotoxins or other neurotoxic agents, exhibit altered levels and activities of numerous MMPs (e.g. MMPs 1–3, 7, 9–14, 19, 20, 23, 24), ADAMs and TIMPs in the injured nerves (16, 17, 56, 225, 239, 240, 266–275), probably in conjunction with changes in cytokines such as TNFα and other factors (270). In normal mice, TIMPs and MMP-2 and 9 are expressed at the neuromuscular junction both in Schwann cells and in the perineurium of the intramuscular nerves and are modulated in diﬀerent ways following denervation or injury to the nerve depending on the type of injury and the particular MMP (167, 225, 239). Increases in TNFα correlate with increases in gelatinase activity in injured sciatic nerve (167, 274) although MMP increases can also precede TNFα changes in some scenarios (37). The particular patterns and temporal proﬁles of the various changes depend on the model being studied. A comparative study of transcripts of 22 mammalian MMPs found multiple MMP changes in a mouse spinal injury model, with by far the greatest changes seen for MMP-12, although it remains to be established whether this reﬂected primarily microglial or macrophage expression or both (275). Neuronal regeneration and injury repair in the adult probably involves MMP activities similar to those likely to be important in neurogenesis during development. Other MMP functions that might also contribute to regeneration include clearing cellular debris and remodelling damaged matrix structure as well as facilitating both axonal regrowth (64, 243, 244, 271) and remyelination (276) by degrading chondroitin sulfate proteoglycans and other growth inhibiting molecules (224, 243, 244). In one recent study of entorrhinal cortex lesions, treatment with the MMP inhibitor FN-439 led to failure to clear neuronal debris and reduced collateral neuronal sprouting together with the impairment of long-term potentiation and other electrophysiological measures compared to vehicle treated controls (277). On the other hand, MMPs can also be deleterious. Evidence for this has come from nerve injury experiments where inhibiting MMPs has been found to foster myelin preservation and to delay Wallerian degeneration (272). Further evidence comes from studies of MMP-12 null mice, which show improved recovery of hind limb functions after spinal injury compared to wildtype mice (275). Whether or not inhibiting expression of a particular MMP is harmful or beneﬁcial after nervous system injury is again likely

May 26, 2005

50

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

to depend on the type of injury, the MMP in question and temporal considerations. This is illustrated by various studies which suggest that early blockade of MMPs might reduce neural tissue damage following injury and promote healing but that inhibiting MMPs later in the healing process might be deleterious. For example, MMP-9 null mice exhibit improved locomotor recovery after a moderate contusion injury to the spinal cord, prompting the suggestion that early blockade of MMPs expressed by glia, vascular elements and peripheral immune cells within a critical 3 day period after injury might be protective and improve motor outcomes (278). [Interestingly, no change in TIMPs was observed over this period in these mice although TIMP changes have been reported in comparable studies in unmodiﬁed mice (274).] The proposal that early blockade of MMP-9 may assist healing is further supported by ﬁndings that MMP-9 null mice are also protected against the early eﬀects of transient focal ischemia, with less blood-brain barrier disruption, reduced degradation of myelin basic protein and smaller 24 hour lesion volumes (279). Manipulating the MMP regulating cascades discussed above may provide a means of achieving similar eﬀects in non-engineered organisms including human patients. For example, a decrease in MMP-9 production following brain trauma in mice can be achieved by inhibiting the ERK pathway (280). On the other hand, there is also evidence that MMPs (initially MMP-9 then MMP-2) are important in axonal regeneration (274) and revascularisation in later stages of injury repair (273) as well as contributing to glial scar formation (273), suggesting that at least at later times, MMP inhibition may be disadvantageous. Concerted regulation of MMPs and TIMPs following nerve injury may be part of a protective response that serves to remove debris and clear the way for tissue remodelling yet at the same time protects tissue against uncontrolled degradation. The low basal expression of TIMP-1 in the adult nervous system, mentioned previously, may occur because the transcription factor c-Fos, important in maintaining basal transcription through AP-1 sites, is not constitutively expressed in the CNS (170). Instead c-Fos may be involved in inducing TIMP-1 as part of the brain’s defense mechanism against injury. Excitotoxic stimulation and seizures result in up-regulation of TIMP-1 in astrocytes in various brain regions as well as hippocampal dentate gyrus neurons and other resistant cells (44, 189–192). This up-regulation appears to involve c-Fos protein binding to an AP-1 responsive promoter element (192). The response is probably part of a cellular defense against injury as

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

51

TIMP-1 can protect hippocampal neurons against excitotoxic injury (see also below) (193). The neuroprotective eﬀects of TIMPs probably include curbing the proteolytic activities of MMPs up-regulated in response to injury. The plasmintissue plasminogen activator (tPA) system and MMPs such as MMP-3 and 9 are implicated in excitotoxicity (reviewed (45, 194)) and over-expression of MMP-3 in mammary epithelial cells of transgenic mice causes unscheduled apoptosis (195) through mechanisms likely to involve integrins and caspases (196). This apoptosis can be reduced by crossing with mice overexpressing TIMP-1 (195). The neuroprotective eﬀects of TIMPs might also involve MMP-independent eﬀects such as have been described earlier in this chapter. Again, almost nothing is known about regulation of MMP and TIMP gene expression in these contexts. Delineating the factors regulating the temporal sequence of MMP expression after diﬀerent types of nervous system injury or disease has potential clinical relevance for a large number of pathological situations.

28. Genetic Polymorphisms and MMP Expression Several single nucleotide polymorphisms (SNPs) are important in regulating MMP gene expression in diseases outside the nervous system. Various studies (reviewed (81, 281) and see also (83, 282)) have reported associations between cardiovascular disease and MMP polymorphisms, in particular (i) the −1612 del/ins MMP-3 promoter polymorphism (ii) the C-1562T MMP-9 promoter SNP (iii) the (CA)n microsatellite repeat polymorphism at −90 in the MMP-9 promoter, (iv) the A-82G MMP-12 promoter SNP and (v) the A-77G MMP-13 promoter SNP. However, the few studies to date in the nervous system have generally failed to reveal disease associations. The C-1562T polymorphism increases MMP-9 expression in vitro, which could in principle increase blood-brain barrier breakdown in stroke. However, it did not aﬀect plasma MMP-9 levels or hemorrhage risk in 59 stroke patients (283) although the sample was small and contained no individuals homozygous for the polymorphism. Similarly, neither the C-1562T SNP nor microsatellite polymorphisms at −90 of the MMP-9 promoter were associated with increased susceptibility to multiple sclerosis (284). Various polymorphisms in TIMPs 1–3 have also failed to show associations with Alzheimer’s disease (285) or intracranial aneurysms (286).

May 26, 2005

52

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

Mutations that alter MMP-2 promoter activity might alter risk, susceptibility or progression of gliomas and other conditions featuring MMP-2 (127). Two MMP-2 promoter SNPs have been found to appreciably aﬀect MMP-2 expression (127). One is a G → A substitution at −1575 that decreases estrogen receptor-α binding to the promoter, reducing estrogen regulation of MMP-2 gene transcription (287). The second SNP is a −1306C → T substitution that destroys an SP1 site and decreases promoter activity (127). This SNP is relatively frequent (∼25% in a North American Caucasian population) and important in lung, breast and other cancers, where it can sometimes reduce cancer risks (288–290). Polymorphisms in molecules regulating MMP gene expression are also likely to be important. Some cytokine gene polymorphisms modify an individual’s expression of cytokines in response to inﬂammation (291, 292). Different combinations of these polymorphisms modify inﬂammatory responses in various diseases (291, 292) and can inﬂuence risk or severity of disease or treatment responsiveness (292). Cytokine genotypes relevant to MMP expression include genotypes associated with high or low production of TNF-α (promoter −308 alleles) or IL-6 (promoter −174), IL-1 cluster haplotypes (e.g. IL-1α promoter −899, IL-1β promoter −511, exon 5 +3953), and high, intermediate or low production of IFN-γ (intron 1 +874) or transforming growth factor beta TGF-β (codons 10, 25) (291). It is probably na¨ıve to expect to detect strong disease associations for single polymorphisms regulating expression of downstream eﬀectors such as the MMPs if the potential confounding eﬀects of polymorphisms inﬂuencing upstream regulatory molecules are not taken into account. For example, a person who is a ‘high responder’, producing high levels of TNFα or IL-1β in disease or injury, may be substantially more aﬀected by a SNP modifying an MMP promoter site than a ‘low responder’. The future in this ﬁeld may lie in the use of risk analysis and modulated intervention based on arrays of relevant polymorphisms across a spectrum of interacting factors. Encouragingly, the eﬀects of a particular genotype can sometimes be overcome. One way of achieving this is by utilising alternative modes of regulating MMP expression through manipulating diﬀerent combinations of signal transduction pathways and cis-acting sequences (293). This is illustrated by a −1607 del/ins MMP-1 promoter polymorphism associated with melanomas and other cancers (281, 294–297). The presence of an additional G at this position (2G allele) creates an ETS binding site that is absent in the 1G allele form. The phenotype of the 2G isoform is generally

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

53

characterised by higher levels of MMP-1 expression and is more invasive (294, 297). But the eﬀect of the 2G isoform can be overcome by inhibiting the Fos-like region antigen (Fra-1), an AP-1 family member that is necessary for mitogen regulation of MMP-1. Inhibition of Fra-1 down-regulates MMP-1 expression from the 2G allelic form of the promoter relative to the 1G form (298). This is a valuable reminder that genotype, while important, can nonetheless often be overridden, improving the prospects for manipulating MMPs, TIMPs and related molecules in both experimental and clinical settings.

29. Conclusion The foregoing discussion illustrates how little is presently known about the regulation of MMP gene expression in the nervous system. While displaying many features observed in systemic MMP gene regulation, there are nonetheless important diﬀerences not only between nervous system and non-nervous system cells but also between diﬀerent nervous system regions and cell types. This is likely to prove a particularly fertile area for future research. The rapid advancement of array technology should help provide the necessary tools to unravel the complex network of interdependent regulation operating between the many MMPs, ADAMs and TIMPs and the other molecules that interact with them. Since these enzymes are important in so many neurophysiological and neuropathological events, extending the knowledge base in this area will be of substantial beneﬁt to our understanding of the nervous system in health and disease.

References 1. Mackay, A. R., Ballin, M., Pelina, M. D., Farina, A. R., Nason, A. M., Hartzler, J. L., and Thorgeirsson, U. P. (1992) Invasion Metastasis 12(3–4), 168–84 2. Muir, D. (1995) Clinical & Experimental Metastasis 13(4), 303–14 3. Nakano, A., Tani, E., Miyazaki, K., Yamamoto, Y., and Furuyama, J. (1995) J Neurosurg 83(2), 298–307 4. Balbin, M., Fueyo, A., Knauper, V., Lopez, J. M., Alvarez, J., Sanchez, L. M., Quesada, V., Bordallo, J., Murphy, G., and Lopez-Otin, C. (2001) J Biol Chem 276(13), 10253–62 5. Qu, Z., Wolfraim, L. A., Svaren, J., Ehrengruber, M. U., Davidson, N., and Milbrandt, J. (1998) J Cell Biol 142(4), 1075–82

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

54

T. Roy & E.A. Milward

6. Khan, K. M., Falcone, D. J., and Kraemer, R. (2002) J Biol Chem 277(3), 2353–9 7. Rooprai, H. K., Rucklidge, G. J., Panou, C., and Pilkington, G. J. (2000) Br J Cancer 82(1), 52–5 8. Wick, W., Platten, M., and Weller, M. (2001) J Neurooncol 53(2), 177–85 9. Giraudon, P., Szymocha, R., Buart, S., Bernard, A., Cartier, L., Belin, M. F., and Akaoka, H. (2000) J Immunol 164(5), 2718–27 10. Worley, J. R., Baugh, M. D., Hughes, D. A., Edwards, D. R., Hogan, A., Sampson, M. J., and Gavriloc, J. (2003) J Biol Chem 278(51), 51340–6 11. Farina, A. R., Masciulli, M. P., Tacconelli, A., Cappabianca, L., De Santis, G., Gulino, A., and Mackay, A. R. (2002) Cell Growth Diﬀer 13(8), 343–54 12. Harkness, K. A., Adamson, P., Sussman, J. D., Davies-Jones, G. A., Greenwood, J., and Woodroofe, M. N. (2000) Brain 123 (Pt 4), 698–709 13. Rathke-Hartlieb, S., Budde, P., Ewert, S., Schlomann, U., Staege, M. S., Jockusch, H., Bartsch, J. W., and Frey, J. (2000) FEBS Lett 481(3), 227–34 14. Hummel, V., Kallmann, B. A., Wagner, S., Fuller, T., Bayas, A., Tonn, J. C., Benveniste, E. N., Toyka, K. V., and Rieckmann, P. (2001) J Neuropathol Exp Neurol 60(4), 320–7 15. Arai, K., Lee, S. R., and Lo, E. H. (2003) Glia 43(3), 254–64 16. Hebert, G., Arsaut, J., Dantzer, R., and Demotes-Mainard, J. (2003) Neurosci Lett 349(3), 191–5 17. Vecil, G. G., Larsen, P. H., Corley, S. M., Herx, L. M., Besson, A., Goodyer, C. G., and Yong, V. W. (2000) J Neurosci Res 61(2), 212–24 18. Zhang, X., and Chintala, S. K. (2004) Exp Eye Res 78(4), 849–60 19. Cross, A. K., and Woodroofe, M. N. (1999) Glia 28(3), 183–9 20. Brenneisen, P., Wenk, J., Klotz, L. O., Wlaschek, M., Briviba, K., Krieg, T., Sies, H., and Scharﬀetter-Kochanek, K. (1998) J Biol Chem 273(9), 5279–87 21. Yoon, S. O., Park, S. J., Yoon, S. Y., Yun, C. H., and Chung, A. S. (2002) J Biol Chem 277(33), 30271–82 22. Kim, G. W., Gasche, Y., Grzeschik, S., Copin, J. C., Maier, C. M., and Chan, P. H. (2003) J Neurosci 23(25), 8733–42 23. Kolev, K., Skopal, J., Simon, L., Csonka, E., Machovich, R., and Nagy, Z. (2003) Thromb Haemost 90(3), 528–37 24. Magid, R., Murphy, T. J., and Galis, Z. S. (2003) J Biol Chem 278(35), 32994–9 25. Brenneisen, P., Sies, H., and Scharﬀetter-Kochanek, K. (2002) Annals NY Acad Sci 973, 31–43 26. Gu, Z., Kaul, M., Yan, B., Kridel, S. J., Cui, J., Strongin, A., Smith, J. W., Liddington, R. C., and Lipton, S. A. (2002) Science 297(5584), 1186–90 27. Reich, R., and Martin, G. R. (1996) Prostaglandins 51(1), 1–17 28. Vegeto, E., Bonincontro, C., Pollio, G., Sala, A., Viappiani, S., Nardi, F., Brusadelli, A., Viviani, B., Ciana, P., and Maggi, A. (2001) J Neurosci 21(6), 1809–18 29. Kyrkanides, S., Moore, A. H., Olschowka, J. A., Daeschner, J. C., Williams, J. P., Hansen, J. T., and Kerry O’Banion, M. (2002) Brain Res Mol Brain Res 104(2), 159–69

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

55

30. Esteve, P. O., Chicoine, E., Robledo, O., Aoudjit, F., Descoteaux, A., Potworowski, E. F., and St Pierre, Y. (2002) J Biol Chem 277(38), 35150–5 31. Bian, J., and Sun, Y. (1997) Mol Cell Biol 17(11), 6330–8 32. Chandrasekar, N., Jasti, S., Alfred-Yung, W. K., Ali-Osman, F., Dinh, D. H., Olivero, W. C., Gujrati, M., Kyritsis, A. P., Nicolson, G. L., Rao, J. S., and Mohanam, S. (2000) Clin Exp Metastasis 18(4), 337–42 33. Park, M. J., Park, I. C., Hur, J. H., Kim, M. S., Lee, H. C., Woo, S. H., Lee, K. H., Rhee, C. H., Hong, S. I., and Lee, S. H. (2002) J Neurosurg 97(1), 112–8 34. Vincenti, M. P. (2001) Methods Mol Biol 151, 121–48 35. Schroen, D. J., and Brinckerhoﬀ, C. E. (1996) Gene Expr 6(4), 197–207 36. Leco, K. J., Khokha, R., Pavloﬀ, N., Hawkes, S. P., and Edwards, D. R. (1994) J Biol Chem 269(12), 9352–60 37. Mun-Bryce, S., Lukes, A., Wallace, J., Lukes-Marx, M., and Rosenberg, G. A. (2002) Brain Research 933(1), 42–9 38. Giraudon, P., Buart, S., Bernard, A., and Belin, M. F. (1997) Mol Psychiatr 2(2), 107–10, 84 39. Misse, D., Esteve, P. O., Renneboog, B., Vidal, M., Cerutti, M., St Pierre, Y., Yssel, H., Parmentier, M., and Veas, F. (2001) Blood 98(3), 541–7 40. Chong, Y. H., Seoh, J. Y., and Park, H. K. (1998) J Mol Neurosci 10(2), 129–41 41. Sadowski, T., and Steinmeyer, J. (2001) Inﬂamm Res 50(3), 175–82 42. Robinson, S. N., Pisarev, V. M., Chavez, J. M., Singh, R. K., and Talmadge, J. E. (2003) Stem Cells 21(4), 417–27 43. Oh, J., Takahashi, R., Adachi, E., Kondo, S., Kuratomi, S., Noma, A., Alexander, D. B., Motoda, H., Okada, A., Seiki, M., Itoh, T., Itohara, S., Takahashi, C., and Noda, M. (2004) Oncogene 44. Szklarczyk, A., Lapinska, J., Rylski, M., McKay, R. D., and Kaczmarek, L. (2002) J Neurosci 22(3), 920–30 45. Kaczmarek, L., Lapinska-Dzwonek, J., and Szymczak, S. (2002) EMBO J 21(24), 6643–8 46. Shum, J. K., Melendez, J. A., and Jeﬀrey, J. J. (2002) J Biol Chem 277(45), 42830–40 47. Paglinawan, R., Malipiero, U., Schlapbach, R., Frei, K., Reith, W., and Fontana, A. (2003) Glia 44(3), 219–31 48. Bair, E. L., Massey, C. P., Tran, N. L., Borchers, A. H., Heimark, R. L., Cress, A. E., and Bowden, G. T. (2001) Exp Cell Res 270(2), 259–67 49. Galt, S. W., Lindemann, S., Medd, D., Allen, L. L., Kraiss, L. W., Harris, E. S., Prescott, S. M., McIntyre, T. M., Weyrich, A. S., and Zimmerman, G. A. (2001) Circ Res 89(6), 509–16 50. Sato, T., Sakai, T., Noguchi, Y., Takita, M., Hirakawa, S., and Ito, A. (2004) Gynecol Oncol 92(1), 47–56 51. Wernicke, D., Schulze-Westhoﬀ, C., Brauer, R., Petrow, P., Zacher, J., Gay, S., and Gromnica-Ihle, E. (2002) Arthritis Rheum 46(1), 64–74 52. Zalewska, T., Ziemka-Nalecz, M., Sarnowska, A., and Domanska-Janik, K. (2003) Brain Res 977(1), 62–9

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

56

T. Roy & E.A. Milward

53. Yang, E. V., Bane, C. M., MacCallum, R. C., Kiecolt-Glaser, J. K., Malarkey, W. B., and Glaser, R. (2002) J Neuroimmunol 133(1–2), 144–50 54. Yamaguchi, S., Yamaguchi, M., Yatsuyanagi, E., Yun, S. S., Nakajima, N., Madri, J. A., and Sumpio, B. E. (2002) Lab Invest 82(7), 949–56 55. Brinckerhoﬀ, C. E., and Matrisian, L. M. (2002) Nat Rev Mol Cell Biol 3(3), 207–14 56. Xu, J., Kim, G. M., Ahmed, S. H., Yan, P., Xu, X. M., and Hsu, C. Y. (2001) J Neurosci 21(1), 92–7 57. Qin, H., Moellinger, J. D., Wells, A., Windsor, L. J., Sun, Y., and Benveniste, E. N. (1998) J Immunol 161(12), 6664–73 58. Ma, Z., Qin, H., and Benveniste, E. N. (2001) J Immunol 167(9), 5150–9 59. Pan, M. R., and Hung, W. C. (2002) J Biol Chem 277(36), 32775–80 60. Mahloogi, H., Gonzalez-Guerrico, A. M., Lopez De Cicco, R., Bassi, D. E., Goodrow, T., Braunewell, K. H., and Klein-Szanto, A. J. (2003) Cancer Res 63(16), 4997–5004 61. Sun, Y., Wenger, L., Rutter, J. L., Brinckerhoﬀ, C. E., and Cheung, H. S. (1999) Ann N Y Acad Sci 878, 638–41 62. Feinberg, M. W., Jain, M. K., Werner, F., Sibinga, N. E., Wiesel, P., Wang, H., Topper, J. N., Perrella, M. A., and Lee, M. E. (2000) J Biol Chem 275(33), 25766–73 63. Chen, H., Li, D., Saldeen, T., and Mehta, J. L. (2003) Am J Physiol Heart Circ Physiol 284(5), H1612–7 64. Hayashita-Kinoh, H., Kinoh, H., Okada, A., Komori, K., Itoh, Y., Chiba, T., Kajita, M., Yana, I., and Seiki, M. (2001) Cell Growth Diﬀer 12(11), 573–80 65. Overall, C. M., Wrana, J. L., and Sodek, J. (1989) J Biol Chem 264(3), 1860–9 66. Pendas, A. M., Balbin, M., Llano, E., Jimenez, M. G., and Lopez-Otin, C. (1997) Genomics 40(2), 222–33 67. Uria, J. A., Jimenez, M. G., Balbin, M., Freije, J. M., and Lopez-Otin, C. (1998) J Biol Chem 273(16), 9769–77 68. Milenkovic, I., Weick, M., Wiedemann, P., Reichenbach, A., and Bringmann, A. (2003) Invest Ophthalmol Vis Sci 44(3), 1211–20 69. Delany, A. M., and Brinckerhoﬀ, C. E. (1992) J Cell Biochem 50(4), 400–10 70. Edwards, D. R., Murphy, G., Reynolds, J. J., Whitham, S. E., Docherty, A. J., Angel, P., and Heath, J. K. (1987) EMBO J 6(7), 1899–904 71. Ito, A., Mukaiyama, A., Itoh, Y., Nagase, H., Thogersen, I. B., Enghild, J. J., Sasaguri, Y., and Mori, Y. (1996) J Biol Chem 271(25), 14657–60 72. Schonbeck, U., Mach, F., and Libby, P. (1998) J Immunol 161(7), 3340–6 73. Vogel, W., Gish, G. D., Alves, F., and Pawson, T. (1997) Mol Cell 1(1), 13–23 74. Shrivastava, A., Radziejewski, C., Campbell, E., Kovac, L., McGlynn, M., Ryan, T. E., Davis, S., Goldfarb, M. P., Glass, D. J., Lemke, G., and Yancopoulos, G. D. (1997) Mol Cell 1(1), 25–34 75. Gearing, A. J., Beckett, P., Christodoulou, M., Churchill, M., Clements, J., Davidson, A. H., Drummond, A. H., Galloway, W. A., Gilbert, R., Gordon, J. L., and et al (1994) Nature 370(6490), 555–7

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

57

76. Chandler, S., Cossins, J., Lury, J., and Wells, G. (1996) Biochem Biophys Res Commun 228(2), 421–9 77. Conant, K., St Hillaire, C., Nagase, H., Visse, R., Gary, D., Haughey, N., Anderson, C., Turchan, J., and Nath, A. (2004) J Biol Chem 279(9), 8056–62 78. Reunanen, N., Westermarck, J., Hakkinen, L., Holmstrom, T. H., Elo, I., Eriksson, J. E., and Kahari, V. M. (1998) J Biol Chem 273(9), 5137–45 79. Lai, W. C., Zhou, M., Shankavaram, U., Peng, G., and Wahl, L. M. (2003) J Immunol 170(12), 6244–9 80. Benbow, U., and Brinckerhoﬀ, C. E. (1997) Matrix Biology 15(8–9), 519–26 81. Sternlicht, M. D., and Werb, Z. (2001) Annu Rev Cell Dev Biol 17, 463–516 82. Widmann, C., Gibson, S., Jarpe, M. B., and Johnson, G. L. (1999) Physiol Rev 79(1), 143–80 83. Borghaei, R. C., Rawlings, P. L., Jr., Javadi, M., and Woloshin, J. (2004) Biochem Biophys Res Commun 316(1), 182–8 84. Darnell, J. E., Jr. (1997) Science 277(5332), 1630–5 85. Ramana, C. V., Chatterjee-Kishore, M., Nguyen, H., and Stark, G. R. (2000) Oncogene 19(21), 2619–27 86. Kubiatowski, T., Jang, T., Lachyankar, M. B., Salmonsen, R., Nabi, R. R., Quesenberry, P. J., Litofsky, N. S., Ross, A. H., and Recht, L. D. (2001) J Neurosurg 95(3), 480–8 87. Fromigue, O., Louis, K., Wu, E., Belhacene, N., Loubat, A., Shipp, M., Auberger, P., and Mari, B. (2003) Int J Cancer 106(3), 355–63 88. Suzuki, A., Lu, J., Kusakai, G., Kishimoto, A., Ogura, T., and Esumi, H. (2004) Mol Cell Biol 24(8), 3526–35 89. Delcommenne, M., Tan, C., Gray, V., Rue, L., Woodgett, J., and Dedhar, S. (1998) Proc Natl Acad Sci USA 95(19), 11211–6 90. Kumar, C. C. (1998) Oncogene 17(11 Reviews), 1365–73 91. Persad, S., Attwell, S., Gray, V., Delcommenne, M., Troussard, A., Sanghera, J., and Dedhar, S. (2000) Proc Natl Acad Sci USA 97(7), 3207–12 92. Gelinas, D. S., Bernatchez, P. N., Rollin, S., Bazan, N. G., and Sirois, M. G. (2002) Br J Pharmacol 137(7), 1021–30 93. Brazil, D. P., and Hemmings, B. A. (2001) Trends Biochem Sci 26(11), 657–64 94. Hashimoto, T., Wen, G., Lawton, M. T., Boudreau, N. J., Bollen, A. W., Yang, G. Y., Barbaro, N. M., Higashida, R. T., Dowd, C. F., Halbach, V. V., and Young, W. L. (2003) Stroke 34(4), 925–31 95. Brownawell, A. M., Kops, G. J., Macara, I. G., and Burgering, B. M. (2001) Mol Cell Biol 21(10), 3534–46 96. Nakae, J., Kitamura, T., Ogawa, W., Kasuga, M., and Accili, D. (2001) Biochemistry 40(39), 11768–76 97. Moustakas, A., Souchelnytskyi, S., and Heldin, C. H. (2001) J Cell Sci 114 (Pt 24), 4359–69 98. Moelling, K., Schad, K., Bosse, M., Zimmermann, S., and Schweneker, M. (2002) J Biol Chem 277(34), 31099–106 99. Schroen, D. J., and Brinckerhoﬀ, C. E. (1996) J Cell Physiol 169(2), 320–32

May 26, 2005

58

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

100. Hasan, S., and Nakajima, M. (1999) Biochem Biophys Res Commun 258(3), 663–7 101. Korzus, E., Nagase, H., Rydell, R., and Travis, J. (1997) J Biol Chem 272(2), 1188–96 102. Bond, M., Fabunmi, R. P., Baker, A. H., and Newby, A. C. (1998) FEBS Lett 435(1), 29–34 103. Wan, Y., Belt, A., Wang, Z., Voorhees, J., and Fisher, G. (2001) Int J Mol Med 7(3), 329–34 104. Gasche, Y., Copin, J. C., Sugawara, T., Fujimura, M., and Chan, P. H. (2001) J Cereb Blood Flow Metab 21(12), 1393–400 105. Hager, G. L., Nagaich, A. K., Johnson, T. A., Walker, D. A., and John, S. (2004) Biochim Biophys Acta 1677, 46–51 106. Herdegen, T., and Leah, J. D. (1998) Brain Res Rev 28, 370–490 107. Pagenstecher, A., Stalder, A. K., Kincaid, C. L., Shapiro, S. D., and Campbell, I. L. (1998) Am J Pathol 152(3), 729–41 108. Cuzner, M. L., and Opdenakker, G. (1999) J Neuroimmunol 94(1–2), 1–14 109. Zhong, Z. D., Hammani, K., Bae, W. S., and DeClerck, Y. A. (2000) J Biol Chem 275(24), 18602–10 110. Baker, A. H., Edwards, D. R., and Murphy, G. (2002) J Cell Sci 115(Pt 19), 3719–27 111. Gardner, J., and Ghorpade, A. (2003) J Neurosci Res 74(6), 801–6 112. Visse, R., and Nagase, H. (2003) Circ Res 92(8), 827–39 113. Campbell, C. E., Flenniken, A. M., Skup, D., and Williams, B. R. (1991) J Biol Chem 266(11), 7199–206 114. Negro, A., Onisto, M., Pellati, D., and Garbisa, S. (1997) Brain Res Mol Brain Res 48(1), 30–6 115. Brenneisen, P., Wlaschek, M., Schwamborn, E., Schneider, L. A., Ma, W., Sies, H., and Scharﬀetter-Kochanek, K. (2002) Biochem J 365(Pt 1), 31–40 116. Crawford, H. C., and Matrisian, L. M. (1996) Enzyme Protein 49(1–3), 20–37 117. Mertens, P. R., Steinmann, K., Alfonso-Jaume, M. A., En-Nia, A., Sun, Y., and Lovett, D. H. (2002) J Biol Chem 277(28), 24875–82 118. Yong, V. W., Krekoski, C. A., Forsyth, P. A., Bell, R., and Edwards, D. R. (1998) Trends Neurosci 21(2), 75–80 119. Sato, H., and Seiki, M. (1993) Oncogene 8(2), 395–405 120. White, L. A., and Brinckerhoﬀ, C. E. (1995) Matrix Biol 14(9), 715–25 121. Chamberlain, S. H., Hemmer, R. M., and Brinckerhoﬀ, C. E. (1993) J Cell Biochem 52(3), 337–51 122. Frisch, S. M., and Morisaki, J. H. (1990) Mol Cell Biol 10(12), 6524–32 123. Fini, M. E., Bartlett, J. D., Matsubara, M., Rinehart, W. B., Mody, M. K., Girard, M. T., and Rainville, M. (1994) J Biol Chem 269(46), 28620–8 124. White, L. A., Maute, C., and Brinckerhoﬀ, C. E. (1997) Connect Tissue Res 36(4), 321–35 125. Westermarck, J., Seth, A., and Kahari, V. M. (1997) Oncogene 14(22), 2651–60 126. Kerr, L. D., Miller, D. B., and Matrisian, L. M. (1990) Cell 61, 267–78

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

59

127. Price, S. J., Greaves, D. R., and Watkins, H. (2001) J Biol Chem 276(10), 7549–58 128. Su, W., Jackson, S., Tjian, R., and Echols, H. (1991) Genes Dev 5(5), 820–6 129. Sjottem, E., Andersen, C., and Johansen, T. (1997) J Mol Biol 267(3), 490–504 130. Qin, H., Sun, Y., and Benveniste, E. N. (1999) J Biol Chem 274(41), 29130–7 131. Yan, C., Wang, H., Toh, Y., and Boyd, D. D. (2003) J Biol Chem 278(4), 2309–16 132. Cappabianca, L., Farina, A. R., Tacconelli, A., Mantovani, R., Gulino, A., and Mackay, A. R. (2003) Exp Cell Res 286(2), 209–18 133. Bergman, M. R., Cheng, S., Honbo, N., Piacentini, L., Karliner, J. S., and Lovett, D. H. (2003) Biochem J 369 (Pt 3), 485–96 134. Han, X., Boyd, P. J., Colgan, S., Madri, J. A., and Haas, T. L. (2003) J Biol Chem 278(48), 47785–91 135. Luo, D., Mari, B., Stoll, I., and Anglard, P. (2002) J Biol Chem 277(28), 25527–36 136. Lefebvre, O., Regnier, C., Chenard, M. P., Wendling, C., Chambon, P., Basset, P., and Rio, M. C. (1995) Development 121(4), 947–55 137. Kieseier, B. C., Paul, R., Koedel, U., Seifert, T., Clements, J. M., Gearing, A. J., Pﬁster, H. W., and Hartung, H. P. (1999) Brain 122 (Pt 8), 1579–87 138. Wells, G. M., Catlin, G., Cossins, J. A., Mangan, M., Ward, G. A., Miller, K. M., and Clements, J. M. (1996) Glia 18(4), 332–40 139. Lohi, J., Lehti, K., Valtanen, H., Parks, W. C., and Keski-Oja, J. (2000) Gene 242(1–2), 75–86 140. Yuan, L., and Neufeld, A. H. (2001) J Neurosci Res 64(5), 523–32 141. Haas, T. L., Stitelman, D., Davis, S. J., Apte, S. S., and Madri, J. A. (1999) J Biol Chem 274(32), 22679–85 142. Rawlings, N. D., and Barrett, A. J. (1995) Methods Enzymol 248, 183–228 143. Killar, L., White, J., Black, R., and Peschon, J. (1999) Ann N Y Acad Sci 878, 442–52 144. Stone, A. L., Kroeger, M., and Sang, Q. X. (1999) J Protein Chem 18(4), 447–65 145. Loechel, F., Overgaard, M. T., Oxvig, C., Albrechtsen, R., and Wewer, U. M. (1999) J Biol Chem 274(19), 13427–33 146. Kuno, K., Kanada, N., Nakashima, E., Fujiki, F., Ichimura, F., and Matsushima, K. (1997) J Biol Chem 272(1), 556–62 147. Apte, S. S. (2004) Int J Biochem Cell Biol 36(6), 981–5 148. Bates, E. E., Fridman, W. H., and Mueller, C. G. (2002) Immunogenetics 54(2), 96–105 149. Karkkainen, I., Rybnikova, E., Pelto-Huikko, M., and Huovila, A. P. (2000) Mol Cell Neurosci 15(6), 547–60 150. Yong, V. W., Power, C., Forsyth, P., and Edwards, D. R. (2001) Nat Rev Neurosci 2(7), 502–11 151. Bernstein, H. G., Bukowska, A., Krell, D., Bogerts, B., Ansorge, S., and Lendeckel, U. (2003) J Neurocytol 32(2), 153–60

May 26, 2005

60

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

152. Kieseier, B. C., Pischel, H., Neuen-Jacob, E., Tourtellotte, W. W., and Hartung, H. P. (2003) Glia 42(4), 398–405 153. Sun, Y. P., Deng, K. J., Wang, F., Zhang, J., Huang, X., Qiao, S., and Zhao, S. (2004) Gene 325, 171–8 154. Herren, B., Garton, K. J., Coats, S., Bowen-Pope, D. F., Ross, R., and Raines, E. W. (2001) Exp Cell Res 271(1), 152–60 155. Goddard, D. R., Bunning, R. A., and Woodroofe, M. N. (2001) Glia 34(4), 267–71 156. Yuan, W., Matthews, R. T., Sandy, J. D., and Gottschall, P. E. (2002) Neuroscience 114, 1091–1101 157. Lemons, M. L., Sandy, J. D., Anderson, D. K., and Howland, D. R. (2001) J Neurosci 21(13), 4772–81 158. Sasaki, M., Seo-Kiryu, S., Kato, R., Kita, S., and Kiyama, H. (2001) Brain Res Mol Brain Res 89(1–2), 158–63 159. Black, R. A. (2002) Int J Biochem Cell Biol 34, 1–5 160. Moss, M. L., and Lambert, M. H. (2002) Essays Biochem 38, 141–53 161. Arribas, J., Lopez-Casillas, F., and Massague, J. (1997) J Biol Chem 272(27), 17160–5 162. Allinson, T. M., Parkin, E. T., Turner, A. J., and Hooper, N. M. (2003) J Neurosci Res 74(3), 342–52 163. Asai, M., Hattori, C., Szabo, B., Sasagawa, N., Maruyama, K., Tanuma, S., and Ishiura, S. (2003) Biochem Biophys Res Commun 301(1), 231–5 164. Sylvester, J., Liacini, A., Li, W. Q., and Zafarullah, M. (2004) Cell Signal 16(4), 469–76 165. Schafer, B., Gschwind, A., and Ullrich, A. (2004) Oncogene 23(4), 991–9 166. Costa, F. F., Verbisck, N. V., Salim, A. C., Ierardi, D. F., Pires, L. C., Sasahara, R. M., Sogayar, M. C., Zanata, S. M., Mackay, A., O’Hare, M., Soares, F., Simpson, A. J., and Camargo, A. A. (2004) Oncogene 23(7), 1481–8 167. La Fleur, M., Underwood, J. L., Rappolee, D. A., and Werb, Z. (1996) J Exp Med 184(6), 2311–26 168. Campbell, I. L., and Pagenstecher, A. (1999) Trends Neurosci 22(7), 285–7 169. Brew, K., Dinakarpandian, D., and Nagase, H. (2000) Biochim Biophys Acta 1477(1–2), 267–83 170. Crocker, S. J., Pagenstecher, A., and Campbell, I. L. (2004) J Neurosci Res 75(1), 1–11 171. Edwards, D. R., Rocheleau, H., Sharma, R. R., Wills, A. J., Cowie, A., Hassell, J. A., and Heath, J. K. (1992) Biochim Biophys Acta 1171(1), 41–55 172. Logan, S. K., Garabedian, M. J., Campbell, C. E., and Werb, Z. (1996) J Biol Chem 271(2), 774–82 173. Clark, I. M., Rowan, A. D., Edwards, D. R., Bech-Hansen, T., Mann, D. A., Bahr, M. J., and Cawston, T. E. (1997) Biochem J 324 (Pt 2), 611–7 174. Trim, J. E., Samra, S. K., Arthur, M. J., Wright, M. C., McAulay, M., Beri, R., and Mann, D. A. (2000) J Biol Chem 275(9), 6657–63 175. Vaillant, C., Didier-Bazes, M., Hutter, A., Belin, M. F., and Thomasset, N. (1999) J Neurosci 19(12), 4994–5004

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

61

176. Canete Soler, R., Gui, Y.-H., Linask, K. K., and Muschel, R. J. (1995) Dev Brain Res 88, 37–52 177. Jaworski, D. M., and Fager, N. (2000) J Neurosci Res 61(4), 396–408 178. Lee, M., Song, S. U., Ryu, J. K., and Suh, J. K. (2004) J Cell Biochem 91(6), 1260–8 179. Wang, M., Hu, Y., Shima, I., and Stearns, M. E. (1998) Oncol Res 10(4), 219–33 180. Kanazawa, M., Ogata, Y., Aoyagi, K., Stearns, M. E., and Shirouzu, K. (2001) J Exp Clin Cancer Res 20(1), 145–51 181. Dean, G., Young, D. A., Edwards, D. R., and Clark, I. M. (2000) J Biol Chem 275(42), 32664–71 182. Ponton, A., Coulombe, B., Steyaert, A., Williams, B. R., and Skup, D. (1992) Gene 116(2), 187–94 183. Fager, N., and Jaworski, D. M. (2000) Mech Dev 98(1–2), 105–9 184. Bugno, M., Witek, B., Bereta, J., Bereta, M., Edwards, D. R., and Kordula, T. (1999) FEBS Lett 448, 9–14 185. Botelho, F. M., Edwards, D. R., and Richards, C. D. (1998) J Biol Chem 273(9), 5211–8 186. Sohara, N., Trojanowska, M., and Reuben, A. (2002) J Hepatol 36(2), 191–9 187. Hall, M. C., Young, D. A., Waters, J. G., Rowan, A. D., Chantry, A., Edwards, D. R., and Clark, I. M. (2003) J Biol Chem 278(12), 10304–13 188. Yuan, W., and Varga, J. (2001) J Biol Chem 276, 38502–10 189. Rivera, S., Tremblay, E., Timsit, S., Canals, O., Ben-Ari, Y., and Khrestchatisky, M. (1997) J Neurosci 17(11), 4223–35 190. Nedivi, E., Hevroni, D., Naot, D., Israeli, D., and Citri, Y. (1993) Nature 363(6431), 718–22 191. Zhang, J. W., Deb, S., and Gottschall, P. E. (1998) Eur J Neurosci 10, 3358–68 192. Jaworski, J., Biedermann, I. W., Lapinska, J., Szklarczyk, A., Figiel, I., Konopka, D., Nowicka, D., Filipkowski, R. K., Hetman, M., Kowalczyk, A., and Kaczmarek, L. (1999) J Biol Chem 274(40), 28106–12 193. Tan, H. K., Heywood, D., Ralph, G. S., Bienemann, A., Baker, A. H., and Uney, J. B. (2003) Mol Cell Neurosci 22(1), 98–106 194. Rivera, S., and Khrestchatisky, M. (2000) in Advances in Synaptic Plasticity (Baudry, M., Davis, J. L., and Thompson, R. F., eds), pp. 53–86, MIT Press, Cambridge, MA 195. Alexander, C. M., Howard, E. W., Bissell, M. J., and Werb, Z. (1996) J Cell Biol 135 (6 Pt 1), 1669–77 196. Boudreau, N., Sympson, C. J., Werb, Z., and Bissell, M. J. (1995) Science 267(5199), 891–3 197. Hammani, K., Blakis, A., Morsette, D., Bowcock, A. M., Schmutte, C., Henriet, P., and DeClerck, Y. A. (1996) J Biol Chem 271(41), 25498–505 198. Hayakawa, T., Yamashita, K., Ohuchi, E., and Shinagawa, A. (1994) J Cell Sci 107 (Pt 9), 2373–9 199. Mannello, F., and Gazzanelli, G. (2001) Apoptosis 6(6), 479–82

May 26, 2005

62

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

200. Hoegy, S. E., Oh, H. R., Corcoran, M. L., and Stetler-Stevenson, W. G. (2001) J Biol Chem 276(5), 3203–14 201. Whitelock, J. M., Paine, M. L., Gibbins, J. R., Keﬀord, R. F., and O’Grady, R. L. (1993) Immunol Cell Biol 71 (Pt 1), 39–47 202. Vincenti, M. P., White, L. A., Schroen, D. J., Benbow, U., and Brinckerhoﬀ, C. E. (1996) Crit Rev Euk Gene Exp 6(4), 391–411 203. Rutter, J. L., Benbow, U., Coon, C. I., and Brinckerhoﬀ, C. E. (1997) J Cell Biochem 66(3), 322–36 204. Sehgal, I., and Thompson, T. C. (1999) Mol Biol Cell 10(2), 407–16 205. Jiang, Y., and Muschel, R. J. (2002) Cancer Res 62(6), 1910–4 206. Lorenzl, S., Albers, D. S., Narr, S., Chirichigno, J., and Beal, M. F. (2002) Exp Neurol 178(1), 13–20 207. Rosenberg, G. A. (2002) Neuroscientist 8(6), 586–95 208. Cuzner, M. L., Gveric, D., Strand, C., Loughlin, A. J., Paemen, L., Opdenakker, G., and Newcombe, J. (1996) J Neuropathol Exp Neurol 55(12), 1194–204 209. Maeda, A., and Sobel, R. A. (1996) J Neuropathol Exp Neurol 55(3), 300–9 210. Anthony, D. C., Ferguson, B., Matyzak, M. K., Miller, K. M., Esiri, M. M., and Perry, V. H. (1997) Neuropathol Appl Neurobiol 23(5), 406–15 211. Cossins, J. A., Clements, J. M., Ford, J., Miller, K. M., Pigott, R., Vos, W., Van der Valk, P., and De Groot, C. J. (1997) Acta Neuropathologica 94(6), 590–8 212. Lindberg, R. L., De Groot, C. J., Montagne, L., Freitag, P., van der Valk, P., Kappos, L., and Leppert, D. (2001) Brain 124 (Pt 9), 1743–53 213. Lorenzl, S., Albers, D. S., Chirichigno, J. W., Augood, S. J., and Beal, M. F. (2004) J Neurol Sci 218(1–2), 39–45 214. Sekine-Aizawa, Y., Hama, E., Watanabe, K., Tsubuki, S., Kanai-Azuma, M., Kanai, Y., Arai, H., Aizawa, H., Iwata, N., and Saido, T. C. (2001) Eur J Neurosci 13(5), 935–48 215. Schlomann, U., Rathke-Hartlieb, S., Yamamoto, S., Jockusch, H., and Bartsch, J. W. (2000) J Neurosci 20(21), 7964–71 216. Del Bigio, M. R., and Jacque, C. M. (1995) Brain Res Dev Brain Res 86(1–2), 345–7 217. Gottschall, P. E., and Yu, X. (1995) J Neurochem 64(4), 1513–20 218. Deb, S., and Gottschall, P. E. (1996) J Neurochem 66(4), 1641–7 219. Colton, C. A., Keri, J. E., Chen, W. T., and Monsky, W. L. (1993) J Neurosci Res 35(3), 297–304 220. Gottschall, P. E., Yu, X., and Bing, B. (1995) J Neurosci Res 42(3), 335–42 221. Uhm, J. H., Dooley, N. P., Oh, L. Y., and Yong, V. W. (1998) Glia 22(1), 53–63 222. Oh, L. Y., Larsen, P. H., Krekoski, C. A., Edwards, D. R., Donovan, F., Werb, Z., and Yong, V. W. (1999) J Neurosci 19(19), 8464–75 223. Hughes, P. M., Wells, G. M., Clements, J. M., Gearing, A. J., Redford, E. J., Davies, M., Smith, K. J., Hughes, R. A., Brown, M. C., and Miller, K. M. (1998) Brain 121 (Pt 3), 481–94

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

63

224. Muir, E. M., Adcock, K. H., Morgenstern, D. A., Clayton, R., von Stillfried, N., Rhodes, K., Ellis, C., Fawcett, J. W., and Rogers, J. H. (2002) Brain Res Mol Brain Res 100(1–2), 103–17 225. Demestre, M., Wells, G. M., Miller, K. M., Smith, K. J., Hughes, R. A., Gearing, A. J., and Gregson, N. A. (2004) Neuroscience 124(4), 767–79 226. Strazielle, N., Khuth, S. T., Murat, A., Chalon, A., Giraudon, P., Belin, M. F., and Ghersi-Egea, J. F. (2003) J Neuropathol Exp Neurol 62(12), 1254–64 227. Frolichsthal-Schoeller, P., Vescovi, A. L., Krekoski, C. A., Murphy, G., Edwards, D. R., and Forsyth, P. (1999) Neuroreport 10(2), 345–51 228. Tsukatani, T., Fillmore, H. L., Hamilton, H. R., Holbrook, E. H., Costanzo, R. M., Oba, T., Ukeda, H., Matsumoto, K., Miwa, T., Furukawa, M., Ishimaru, T., Hatanaka, S., Yata, T., Horikawa, I., Nishimura, T., Kanehira, E., Ge, Y., DiNardo, L. J., Reiter, E. R., Yamamoto, M., Katayama, Y., Moriya, M., Tanaka, S., Okoyama, S., Moriizumi, T., Sakashita, H., Awaji, T., and Shigemitsu, K. (2003) Neuroreport 14(8), 1135–40 229. Backstrom, J. R., Lim, G. P., Cullen, M. J., and Tokes, Z. A. (1996) J Neurosci 16(24), 7910–9 230. Jaworski, D. M. (2000) Brain Res 860(1–2), 174–7 231. Apodaca, G., Rutka, J. T., Bouhana, K., Berens, M. E., Giblin, J. R., Rosenblum, M. L., McKerrow, J. H., and Banda, M. J. (1990) Cancer Res 50(8), 2322–9 232. Gottschall, P. E., and Deb, S. (1996) Neuroimmunomodulation 3(2–3), 69–75 233. Conant, K., McArthur, J. C., Griﬃn, D. E., Sjulson, L., Wahl, L. M., and Irani, D. N. (1999) Ann Neurol 46(3), 391–8 234. Suryadevara, R., Holter, S., Borgmann, K., Persidsky, R., Labenz-Zink, C., Persidsky, Y., Gendelman, H. E., Wu, L., and Ghorpade, A. (2003) Glia 44(1), 47–56 235. Estus, S., Zaks, W. J., Freeman, R. S., Gruda, M., Bravo, R., and Johnson, E. M., Jr. (1994) J Cell Biol 127 (6 Pt 1), 1717–27 236. Vos, C. M., Gartner, S., Ransohoﬀ, R. M., McArthur, J. C., Wahl, L., Sjulson, L., Hunter, E., and Conant, K. (2000) J Neuroimmunol 109(2), 221–7 237. Machida, C. M., Rodland, K. D., Matrisian, L., Magun, B. E., and Ciment, G. (1989) Neuron 2(6), 1587–96 238. Fillmore, H. L., Mainardi, C. L., and Hasty, K. A. (1992) Matrix Suppl 1, 408 239. Kherif, S., Dehaupas, M., Lafuma, C., Fardeau, M., and Alameddine, H. S. (1998) Neuropathol Appl Neurobiol 24(4), 309–19 240. Jaworski, D. M. (2000) Glia 30(2), 199–208 241. Misko, A., Ferguson, T., and Notterpek, L. (2002) J Neurochem 83(4), 885–94 242. Muir, D. (1994) Exp Cell Res 210(2), 243–52 243. Zuo, J., Ferguson, T. A., Hernandez, Y. J., Stetler-Stevenson, W. G., and Muir, D. (1998) J Neurosci 18(14), 5203–11

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

64

244. 245. 246. 247. 248. 249. 250. 251. 252. 253.

254. 255. 256. 257. 258. 259. 260. 261.

262. 263. 264. 265. 266. 267. 268.

T. Roy & E.A. Milward

Ferguson, T. A., and Muir, D. (2000) Mol Cell Neurosci 16(2), 157–67 Job, C., and Eberwine, J. (2001) Nat Rev Neurosci 2(12), 889–98 Steward, O., and Schuman, E. M. (2001) Annu Rev Neurosci 24, 299–325 Matsuzawa, K., Fukuyama, K., Hubbard, S. L., Dirks, P. B., and Rutka, J. T. (1996) J Neuropathol Exp Neurol 55(1), 88–96 Takahashi, M., Fukami, S., Iwata, N., Inoue, K., Itohara, S., Itoh, H., Haraoka, J., and Saido, T. (2002) Pharmacol Res 46(2), 155–63 Soloway, P. D., Alexander, C. M., Werb, Z., and Jaenisch, R. (1996) Oncogene 13(11), 2307–14 Wild-Bode, C., Weller, M., and Wick, W. (2001) J Neurosurg 94(6), 978–84 Morelli, C., Campioni, K., Parolin, C., Palu, G., and Tognon, M. (2004) J Cell Physiol 199, 126–133 Lakka, S. S., Jasti, S. L., Kyritsis, A. P., Yung, W. K., Ali-Osman, F., Nicolson, G. L., and Rao, J. S. (2000) Clin Exp Metastasis 18(3), 245–52 Lakka, S. S., Jasti, S. L., Gondi, C., Boyd, D., Chandrasekar, N., Dinh, D. H., Olivero, W. C., Gujrati, M., and Rao, J. S. (2002) Oncogene 21(36), 5601–8 Chandrasekar, N., Mohanam, S., Lakka, S. S., Dinh, D. H., Olivero, W. C., Gujrati, M., and Rao, J. S. (2003) Clin Cancer Res 9(6), 2342–9 Wild-Bode, C., Weller, M., Rimner, A., Dichgans, J., and Wick, W. (2001) Cancer Res 61(6), 2744–50 Ko, L. J., and Prives, C. (1996) Genes Dev 10(9), 1054–72 Boulikas, T. (1996) Anticancer Res 16(1), 225–42 Ohga, T., Koike, K., Ono, M., Makino, Y., Itagaki, Y., Tanimoto, M., Kuwano, M., and Kohno, K. (1996) Cancer Res 56(18), 4224–8 Ohga, T., Uchiumi, T., Makino, Y., Koike, K., Wada, M., Kuwano, M., and Kohno, K. (1998) J Biol Chem 273(11), 5997–6000 Wan, Y. S., Wang, Z. Q., Voorhees, J., and Fisher, G. (2001) Cell Signal 13(2), 139–44 Nirmala, C., Jasti, S. L., Sawaya, R., Kyritsis, A. P., Konduri, S. D., Ali-Osman, F., Rao, J. S., and Mohanam, S. (2000) Int J Cancer 88(5), 766–71 Rosenberg, G. A., Dencoﬀ, J. E., Correa, N., Jr., Reiners, M., and Ford, C. C. (1996) Neurology 46(6), 1626–32 Sellebjerg, F., Christiansen, M., Jensen, J., and Frederiksen, J. L. (2000) Eur J Neurol 7(3), 281–9 Sorensen, T. L., Sellebjerg, F., Jensen, C. V., Strieter, R. M., and Ransohoﬀ, R. M. (2001) Eur J Neurol 8(6), 665–72 Agapova, O. A., Kaufman, P. L., Lucarelli, M. J., Gabelt, B. T., and Hernandez, M. R. (2003) Brain Res 967(1–2), 132–43 Rosenberg, G. A. (1995) J Neurotrauma 12(5), 833–42 Shibayama, M., Kuchiwaki, H., Inao, S., Ichimi, K., Yoshida, J., and Hamaguchi, M. (1997) Acta Neurochir Suppl (Wien) 70, 220–1 Bosse, F., Petzold, G., Greiner-Petter, R., Pippirs, U., Gillen, C., and Muller, H. W. (2000) Glia 32(3), 313–27

May 26, 2005

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

Genetic Regulation of the Matrix Metalloproteinases and Related Proteins

65

269. de Castro, R. C., Jr., Burns, C. L., McAdoo, D. J., and Romanic, A. M. (2000) Neuroreport 11(16), 3551–4 270. Shubayev, V. I., and Myers, R. R. (2000) Brain Res 855(1), 83–9 271. Duchossoy, Y., Horvat, J. C., and Stettler, O. (2001) Mol Cell Neurosci 17(6), 945–56 272. Siebert, H., Dippel, N., Mader, M., Weber, F., and Bruck, W. (2001) J Neuropathol Exp Neurol 60(1), 85–93 273. Goussev, S., Hsu, J. Y., Lin, Y., Tjoa, T., Maida, N., Werb, Z., and Noble-Haeusslein, L. J. (2003) J Neurosurg 99 (2 Suppl), 188–97 274. Platt, C. I., Krekoski, C. A., Ward, R. V., Edwards, D. R., and Gavrilovic, J. (2003) J Neurosci Res 74(3), 417–29 275. Wells, J. E., Rice, T. K., Nuttall, R. K., Edwards, D. R., Zekki, H., Rivest, S., and Yong, V. W. (2003) J Neurosci 23(31), 10107–15 276. Larsen, P. H., Wells, J. E., Stallcup, W. B., Opdenakker, G., and Yong, V. W. (2003) J Neurosci 23(35), 11127–35 277. Reeves, T. M., Prins, M. L., Zhu, J., Povlishock, J. T., and Phillips, L. L. (2003) J Neurosci 23(32), 10182–9 278. Noble, L. J., Donovan, F., Igarashi, T., Goussev, S., and Werb, Z. (2002) J Neurosci 22(17), 7526–35 279. Asahi, M., Wang, X., Mori, T., Sumii, T., Jung, J. C., Moskowitz, M. A., Fini, M. E., and Lo, E. H. (2001) J Neurosci 21(19), 7724–32 280. Mori, T., Wang, X., Aoki, T., and E.H., L. (2002) J Neurotrauma 19, 1411–9 281. Ye, S. (2000) Matrix Biology 19(7), 623–9 282. Yoon, S., Kuivaniemi, H., Gatalica, Z., Olson, J. M., Buttice, G., Ye, S., Norris, B. A., Malcom, G. T., Strong, J. P., and Tromp, G. (2002) Matrix Biol 21(6), 487–98 283. Montaner, J., Fernandez-Cadenas, I., Molina, C. A., Monasterio, J., Arenillas, J. F., Ribo, M., Quintana, M., Chacon, P., Andreu, A. L., and Alvarez-Sabin, J. (2003) Stroke 34(12), 2851–5 284. Nelissen, I., Vandenbroeck, K., Fiten, P., Hillert, J., Olsson, T., Marrosu, M. G., and Opdenakker, G. (2000) J Neuroimmunol 105(1), 58–63 285. Wollmer, M. A., Papassotiropoulos, A., Streﬀer, J. R., Grimaldi, L. M., Kapaki, E., Salani, G., Paraskevas, G. P., Maddalena, A., de Quervain, D., Bieber, C., Umbricht, D., Lemke, U., Bosshardt, S., Degonda, N., Henke, K., Hegi, T., Jung, H. H., Pasch, T., Hock, C., and Nitsch, R. M. (2002) Psychiatr Genet 12(3), 155–60 286. Krex, D., Rohl, H., Konig, I. R., Ziegler, A., Schackert, H. K., and Schackert, G. (2003) Stroke 34(12), 2817–21 287. Harendza, S., Lovett, D. H., Panzer, U., Lukacs, Z., Kuhnl, P., and Stahl, R. A. (2003) J Biol Chem 278(23), 20490–9 288. Yu, C., Pan, K., Xing, D., Liang, G., Tan, W., Zhang, L., and Lin, D. (2002) Cancer Res 62(22), 6430–3 289. Miao, X., Yu, C., Tan, W., Xiong, P., Liang, G., Lu, W., and Lin, D. (2003) Cancer Res 63(14), 3987–90 290. Zhou, Y., Yu, C., Miao, X., Tan, W., Liang, G., Xiong, P., Sun, T., and Lin, D. (2004) Carcinogenesis 25(3), 399–404

May 26, 2005

66

16:0 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch02

T. Roy & E.A. Milward

291. Barrett, S., Collins, M., Kenny, C., Ryan, E., Keane, C. O., and Crowe, J. (2003) J Med Virol 71(2), 212–8 292. Cooper, A. C., Mikhail, A., Lethbridge, M. W., Kemeny, D. M., and Macdougall, I. C. (2003) J Am Soc Nephrol 14(7), 1776–84 293. Benbow, U., Tower, G. B., Wyatt, C. A., Buttice, G., and Brinckerhoﬀ, C. E. (2002) J Cell Biochem 86(2), 307–19 294. Rutter, J. L., Mitchell, T. I., Buttice, G., Meyers, J., Gusella, J. F., Ozelius, L. J., and Brinckerhoﬀ, C. E. (1998) Cancer Res 58(23), 5321–5 295. Kanamori, Y., Matsushima, M., Minaguchi, T., Kobayashi, K., Sagae, S., Kudo, R., Terakawa, N., and Nakamura, Y. (1999) Cancer Res 59(17), 4225–7 296. Nishioka, Y., Kobayashi, K., Sagae, S., Ishioka, S., Nishikawa, A., Matsushima, M., Kanamori, Y., Minaguchi, T., Nakamura, Y., Tokino, T., and Kudo, R. (2000) Jpn J Cancer Res 91(6), 612–5 297. Noll, W. W., Belloni, D. R., Rutter, J. L., Storm, C. A., Schned, A. R., Titus-Ernstoﬀ, L., Ernstoﬀ, M. S., and Brinckerhoﬀ, C. E. (2001) Am J Pathol 158(2), 691–7 298. Tower, G. B., Coon, C. I., Belguise, K., Chalbos, D., and Brinckerhoﬀ, C. E. (2003) Eur J Biochem 270(20), 4216–25 299. Gordon-Weeks, P. R. (2000) in Neuronal Growth Cones (Bard, J. B. L., Barlow, P. W., and Kirk, D. L., eds), pp. 75–136, Cambridge University Press, Cambridge, UK

May 26, 2005

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

CHAPTER 3 POST-TRANSLATIONAL MODIFICATION

Z. Gu, M. Kaul and S.A. Lipton∗ Center for Neuroscience & Aging, The Burnham Institute, 10901 North Torrey Pines Road, La Jolla, CA 92037, USA E-mail: ∗[email protected]

Recent studies have implicated matrix metalloproteinases (MMPs) in the pathogenesis of stroke and neurodegenerative disorders, including Alzheimer’s disease, HIV-associated dementia, multiple sclerosis, as well as those aﬀecting the retina. The mechanism for regulating MMP activity in these disorders, however, remains largely unknown. Nitric oxide (NO), as a signalling molecule, can regulate the function of many proteins by S-nitrosylation/denitrosylation (transfer of NO to/from a critical cysteine thiol group within an acid-base or hydrophobic structural motif). Such dynamic regulation is also prone to malfunction in many diseases. Here, we summarise a newly discovered post-translational modiﬁcation of MMPs involving S-nitrosylation that is responsible for their activation. We found evidence that S-nitrosylation can activate the proform of MMP-9 in vitro as well as in vivo. When MMP-9 was activated by NO, it induced neuronal apoptosis in cerebrocortical cultures. During cerebral ischemia/reperfusion injury, MMP-9 co-localised with neuronal nitric oxide synthase (nNOS), and MMP-9 activity increased in the ischemic brains. Activation of MMP was abrogated after stroke in nNOS knockout mice or in wild-type animals treated with a relatively speciﬁc nNOS inhibitor. Mass spectrometry identiﬁed the active derivative of MMP-9 both in vitro and in vivo as a stable sulﬁnic or sulfonic acid, whose formation was triggered by S-nitrosylation following nitrosative and oxidative stress. These ﬁndings indicate a novel extracellular proteolysis pathway to neuronal cell death in which S-nitrosylation leads to activation of MMPs, and further oxidation results in a stable post-translational modiﬁcation with pathological activity.

1. Introduction Matrix metalloproteinases (MMPs) represent a family of extracellular soluble or membrane-bound endopeptidases that are prominently involved in Correspondence to: S.A. Lipton 67

May 26, 2005

68

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Z. Gu, M. Kaul & S.A. Lipton

maintaining extracellular matrix (ECM) integrity and modulating interactions between cells during development and tissue remodelling (1, 2). Under normal physiological conditions, the activities of MMPs are precisely regulated at the level of transcription, activation of the proform of MMPs, and inhibition by endogenous inhibitors. Imbalance of MMP activity levels is thought to underlie many neurological disorders as well as other inﬂammatory and malignant diseases (1, 2). The post-translational modiﬁcations of MMPs, such as glycosylation, covalent dimerisation and noncovalent interactions, have been described before (3). In the present review, we focus on post-translational modiﬁcation of MMPs by S-nitrosylation and their functions related to the pathogenesis of neurodegenerative disorders. 2. Cysteine Switch Mechanism The MMP family currently has at least 23 members in human and is characterised by the presence of the signal peptide, propeptide domain with the conserved PRCGXPD motif, and catalytic domain with the zinc (Zn2+ )-binding motif HEXGHXXGXXH. All MMPs, except MMP-7 and MMP-26, contain an additional hinge region and carboxy terminal hemopexin domain (4). One cysteine residue in the conserved autoinhibitory region of the propeptide domain coordinates a Zn2+ in the catalytic site and thus inhibits the proform of the enzyme (5, 6). In addition to the well-characterised activation by proteinases and organomercurial compounds, the enzyme can be activated in vitro to various extents by surfactants such as sodium dodecyl sulfate, by chaotropic ions such as SCN− , by disulﬁde compounds such as oxidised glutathione, by sulfhydryl alkylating agents such as N -ethylmaleimide, and by oxidants such as NaOCl (7). The underlying basis for these activations is the modiﬁcation, exposure, or proteolytic release of the cysteine residue from its habitat in the latent enzyme where it is thought to be complexed to the activesite zinc atom. All modes of activation of latent MMPs are believed to involve the dissociation of cysteine from the active-site zinc atom and its replacement by water, with the concomitant exposure of the active site. This is thought to be the primary event that precedes the well-known autolytic cleavages. The dissociation of cysteine from the zinc atom in the latent enzyme ‘switches’ the role of the zinc from a non-catalytic to a catalytic one (5).

May 26, 2005

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Post-Translational Modiﬁcation

69

3. Modiﬁcation of MMPs by S-Nitrosylation Nitric oxide (NO) is a signalling molecule implicated in regulation of many biological processes in the nervous system, including neurotransmitter release, plasticity, and apoptosis (8–10). The chemical reactions of NO are largely dictated by its redox state (11). NO can modulate the biological activity of many proteins by reacting with cysteine thiol to form an S-nitrosylated derivative. Such reactions regulate the activity of circulating, membrane-bound, cytosolic, and nuclear proteins, including hemoglobin, NMDA receptors, caspases, and NF-κB (12–15). Cerebral ischemia/reperfusion results in nitrosative and oxidative stress, and hence the production of NO and reactive oxygen species (ROS) (16, 17). The regulation of protein function by S-nitrosylation has led to the proposal that nitrosothiols function as post-translational modiﬁcations analogous to phosphorylation or acetylation. Although the factors governing cysteine reactivity towards nitrosylating agents are not completely understood, critical features include basic and acidic residues ﬂanking the reactive cysteine, either in linear sequence or as a consequence of the three-dimensional organisation of the protein; these acidic and basic residues mechanistically participate in the nitrosylation and denitrosylation steps (18). In our work, we initially examined whether MMP-9 can be S-nitrosylated in vitro. To eliminate eﬀects of TIMP-1 binding to the hemopexin domain, which might interfere with catalysis and activation of MMP-9, we used recombinant proMMP-9 encoding the propeptide and catalytic domains of MMP-9 but lacking the hemopexin domain (R-proMMP-9). R-proMMP-9, puriﬁed from conditioned medium of stably transfected human embryonic kidney 293 (HEK293) cells (19), was incubated with the physiological NO donor S-nitrosocysteine (SNOC). The generation of S-nitrosothiol was detected by the measurement of the ﬂuorescent compound 2,3-naphthyltriazole (NAT). NAT is stoichiometrically converted from 2,3-diaminonaphthalene (DAN) by NO released from S-nitrosylated proteins and thus provides a quantitative measure of S-nitrosothiol formation (20). SNOC-treated R-proMMP-9 resulted in signiﬁcant S-nitrosothiol formation (21). To insure that the S-nitrosothiol generated under these conditions represented S-nitroso-MMP-9 rather than residual SNOC, we examined the stability of these S-nitrosothiols at different incubation times. We found that the S-nitrosylation product of SNOC-treated R-proMMP-9 was much more stable than SNOC alone; within 15 min of incubation, over 95% of the SNOC had decayed while

May 26, 2005

70

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Z. Gu, M. Kaul & S.A. Lipton

over 80% of the S-nitroso-MMP-9 remained. This temporal separation allowed us to distinguish SNOC from S-nitroso-MMP-9 in the ﬂuorescent S-nitrosothiol assay. 4. S-Nitrosylation Leads to MMP-9 Activation In Vitro To determine if S-nitrosylation of R-pro-MMP-9 resulted in its activation, we compared the eﬀects of the known exogenous MMP-9 activator, p-aminophenylmercuric acetate (APMA) with those of SNOC and another nitrosylating agent, acidiﬁed sodium nitrite. Gelatin zymography revealed that incubation with APMA, SNOC, or acidiﬁed sodium nitrite led to a partial conversion of the 53.5 kD R-proMMP-9 into the 41.2 kD activated form of MMP-9; the respective masses were conﬁrmed by mass spectrometry (21). The activation was inhibited in the presence of the MMP-speciﬁc inhibitors GM6001 or 1,10-phenathroline. We then compared the activity of R-proMMP-9 incubated with APMA or SNOC by assaying the ability to cleave a synthetic peptide substrate. The initial velocity of R-proMMP-9 activation was 4.80 µmoles/hr by APMA compared to 0.88 µmoles/hr by SNOC (21). Taken together, these ﬁndings demonstrate that MMP-9 can undergo S-nitrosylation, and furthermore show for the ﬁrst time that NO can directly promote activation of MMP-9. 5. NO-Induced Subsequent Oxidation From the experiments using DAN to NAT conversion to show nitrosothiol generation, we knew that S-nitroso-MMP-9 formation was associated with MMP-9 activation. However, nitrosothiols can be relatively short-lived and their reaction can be reversed by chemical reducing agents (22). Alternatively, S-nitrosothiol formation could potentially lead to irreversible oxidative reactions that would result in the permanent activation of MMPs. It has been suggested that S-nitrosylation of proteins via NO may generally represent a signal transduction cascade, whereas subsequent oxidation via ROS can lead to irreversible modiﬁcations (22). In fact, our observation that S-nitroso-MMP-9 was not stable over time and decayed within 1.5 hours while MMP-9 activation continued over the ensuing day, suggested that a more long-lasting derivative of activated MMP-9 might be produced, especially in the presence of an oxidative insult (21). Possible reactions of S-nitrosylated enzymes (E) with ROS include the

May 26, 2005

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Post-Translational Modiﬁcation

71

following (22): E-SH + NO ↔ E-SNO + H+ (S-nitrosylated, often labile, reversible with reducing agents) E-SNO + H2 O ↔ E-SOH + HNO (sulfenic acid; reversible with reducing agents) • E-SOH + O− 2 → E-SO2 H + OH

(sulﬁnic acid formation; irreversible reaction) E-SO2 H + 2OH• → E-SO3 H + H2 O (sulfonic acid formation; irreversible reaction) 6. Peptide Mass Fingerprinting Analysis of Cysteine Residue Post-Translational Modiﬁcations To assess the possibility of these additional oxidative products and further identify the chemical nature of the NO-triggered-modiﬁcation of MMP-9 responsible for activation, we conducted peptide mass ﬁngerprinting (23). Mass spectra were obtained after in-gel digestion of human R-proMMP-9 by trypsin using matrix-assisted laser desorption/ionisation time-of-ﬂight (MALDI-TOF) mass spectrometry. To perform chemical reduction and in-gel digests without disrupting the peptide fragments of interest in the MALDI-TOF analysis, free cysteines had to be ﬁrst protected to avoid cleavage followed by uncontrolled disulﬁde formation. Therefore, we initially protected cysteine by iodoacetamide alkylation in the absence of SNOC exposure. We observed four signature masses of human MMP-9 fragments that were virtually identical (< 0.1% variation) to those predicted from theoretical tryptic fragments of MMP-9 deduced from the published amino acid sequences (Fig. 1(A), left). One of these peaks represented the region responsible for the cysteine switch in the propeptide domain fragment CGVPDLGR (816 Da) of proMMP-9. We then observed a 48 Da shift in the mass spectrum of the 816 Da fragment after SNOC exposure, yielding a peak at 864.8 Da, consistent with further oxidation to the sulfonic acid derivative (SO3 H-CGVPDLGR; Fig. 1(A), right). Furthermore, we examined mass spectra of tryptic fragments from aﬃnity-precipitated MMP-9 obtained from rat brain after a 2-hr MCA occlusion/15-min reperfusion injury or from the contralateral (control) side

May 26, 2005

72

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Z. Gu, M. Kaul & S.A. Lipton

Fig. 1. Peptide mass ﬁngerprinting analysis of the modiﬁed thiol group of the cysteine residue within the highly conserved auto-inhibitory prodomain of human and rodent MMP-9. (A) Left: MALDI-TOF spectra of in-solution tryptic digest of R-proMMP-9 revealed four signature masses from six tryptic fragments (arrows). Right: The tryptic fragment CGVPDLGR at 816.7 Da shifted by 48 Da to 864.8 Da (arrow) after exposure to SNOC, representing SO3 H-CGVPDLGR. (B) Detection of tryptic fragments by MALDI-TOF mass spectrometry of gel-puriﬁed MMP-9 from rat brains following 2-hour MCA occlusion and 15-min reperfusion. MMP-9 was extracted in Tris buﬀer with 1% Triton X-100, aﬃnity precipitated with Gelatin-Sepharose 4B, subjected to SDS-PAGE gel under non-reducing conditions, and visualised by silver staining. Left: Gel-puriﬁed MMP-9 was reduced and alkylated prior to digestion. Detergent was removed as previously described (23) MALDI-TOF mass spectrometry revealed a mass peak at 830.3 Da (arrow), representing the iodoacetamide (57 Da)-alkylated rat peptide acet-CGVPDVGK (57 + 774 Da) from the propeptide domain isolated from control brains. Right: A mass of 821.8 Da (arrow), representing the 774 Da propeptide domain fragment plus a 48 Da modiﬁcation (SO3 H-CGVPDVGK) was observed in the ischemic side of the brain. MALDI-TOF spectra did not detect modiﬁcation of other cysteine residues within MMP9 tryptic fragments. (C) Treatment with 3br7NI prior to ischemia blocked the formation of the sulﬁnic or sulfonic acid modiﬁcations of MMP-9. Left: In soybean oil vehicletreated rats, MALDI-TOF mass spectrometry revealed three signature mass peaks of

May 26, 2005

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Post-Translational Modification

Fig. 1.

73

(Continued )

of the brain. For these experiments, we performed in-gel digestion with trypsin because gel separation oﬀered better protein resolution. MALDITOF analysis of specimens obtained from the control side of the brain revealed that after reduction and alkylation by iodoacetamide (57 Da), the rat propeptide domain fragment (CGVPDVGK, mass 774 Da) yielded a peak at 830.3 Da, representing the alkylated fragment (acet-CGVPDVGK) (Fig. 1(B), left). In contrast, on the side of the brain with the stroke, the Fig. 1. (Continued ) the MMP-9 tryptic fragments (at 831, 866, and 1070 Da), plus a mass peak of 821 Da representing the propeptide domain fragment containing a 48 Da modiﬁcation (SO3 H-CGVPDVGK). Right: In rats treated with 3br7NI (30 mg/kg body weight, intraperitoneal), the mass peak at 821 Da was not detected in the ischemic side of the brain.

May 26, 2005

74

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Z. Gu, M. Kaul & S.A. Lipton

Fig. 1.

(Continued )

propeptide domain was not as susceptible to reduction and alkylation as evidenced by the appearance of an additional peak indicating a propeptide tryptic fragment at 821.8 Da; this peak represented the addition of a 48 Da adduct in accord with sulfonic acid derivatisation of the thiol group (SO3 HCGVPDVGK) (Fig. 1(B), right), and was similar to that found in vitro after NO activation of human MMP-9 (Fig. 1(A)). Additionally, MALDITOF mass ﬁngerprinting analysis revealed that of the 19 cysteine residues present in MMP-9, only the cysteine in the propeptide domain that coordinates Zn2+ in the active site was irreversibly modiﬁed to a sulﬁnic (-SO2 H) or sulfonic (-SO3 H) acid in these experiments. Our ﬁndings indicate that S-nitrosylation of this cysteine residue in the prodomain followed by further oxidation to a sulﬁnic or sulfonic acid derivative leads to activation of MMP-9. Unlike S-nitrosylation, these latter oxidative reactions are irreversible and therefore contribute to the pathophysiological activation of MMP-9, as found during cerebral ischemia/reperfusion. To conﬁrm the pathophysiological relevance of these ﬁndings, we performed

May 26, 2005

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Post-Translational Modiﬁcation

75

the same ischemia and reperfusion experiments after nNOS inhibition with 3-bromo-7-nitroindazole (3br7NI), which is known to be neuroprotective and decrease stroke size. Under these conditions with NO formation blocked, the sulﬁnic and sulfonic acid oxidation products of activated MMP-9 were not observed in our MALDI-TOF analysis (Fig. 1(C), right). Taken together, it is likely that NO activation of MMPs participates in neuronal injury in vivo. 7. Structural Model of S-Nitrosylation of MMPs As discussed above, we noted an S-nitrosylation motif in MMP-9 when we made an atomic model of its structure using the related MMP-2 crystal structure (PDB code 1CK7) (5). In proMMP-9, a glutamate (E402) is located ∼2.8 ˚ A from the cysteine sulfur, and may act as a general base to remove the sulfhydryl proton (in the activated enzyme, this glutamate acts as a base to activate the Zn2+ -bound water in a similar fashion). The reactivity of the cysteine sulfur is likely to be further enhanced by its binding to the Zn2+ ion, which increases its nucleophilicity. Nitrosylation of this cysteine would be expected to reduce the nucleophilicity of the cysteine sulfur, weakening the bond to the Zn2+ ion, and thus activating the enzyme. However, NO is also a good leaving group in the context of MMPs, thus allowing other oxidative reactions to occur on this cysteine. One of the pathways proposed for oxidation of the nitrosylated cysteine is via hydrolysis to form a sulfenic acid: E-S-N=O + H2 O → E-S-OH + HNO (22). The MMP is set up to carry out hydrolysis of a peptide bond using an activated water molecule, and it is likely that the same machinery can be used to hydrolyze nitrosocysteine (Fig. 2). Additionally, the sulfenic acid is labile and thus susceptible to further oxidation to the stable sulﬁnic or sulfonic acid derivatives that we observed during MALDI-TOF peptide ﬁngerprinting (Fig. 1). Interestingly, activation of the enzyme can occur prior to cleavage, but after sulﬁnic or sulfonic acid modiﬁcation, since we were able to observe these derivatives in our peptide analysis of pro-MMP-9. 8. MMPs in Neurological Diseases Substantial evidence shows that MMPs play a crucial role in the pathogenesis of both acute and chronic neurological disorders, including stroke, Alzheimer’s disease, HIV-associated dementia, multiple sclerosis, as well as retinal degeneration (2, 24–26). MMPs appear to contribute both

May 26, 2005

76

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Z. Gu, M. Kaul & S.A. Lipton

Fig. 2. Model of MMP-9 activation by S-nitrosylation and subsequent oxidation. (A) Molecular surface of a partial sequence of human MMP-9 (from 97 Pro to 411 His without the ﬁbronectin repeats found between 216 Val and 391 Gln) (5). Colour coded by charge with positive charge in violet, negative charge in red; propeptide domain (97 Pro to 106 Arg) designated by a yellow ribbon; catalytic domain (401 His to 411 His) in green. In proMMP-9, Zn2+ is coordinated by a cysteine and three histidine residues. R98, C99, and E402 ﬁt the proposed consensus motif for S-nitrosylation (18). R, Arg; C, Cys; E, Glu; H, His. (B, C) Proposed structure-based chemistry of NO-induced MMP-9 activation. Reactivity of the catalytic cysteine sulfur of MMP-9 appears to be enhanced by increased nucleophilicity of 402 Glu (shown in red) to S-nitrosylating agents (SNOC = Cys-NO, for example). The sulfur bound at the zinc site appears to be highly nucleophilic, which may give high initial reactivity to NO from its endogenous donors. The S-nitroso-MMP-9 propeptide domain appears to be more easily broken up in this highly polar environment and replaced by a nucleophilic water molecule. Reaction with H2 O of the S-nitrosothiol group forms sulfenic acid (-SOH), as observed in glutathione reductase (22, 46). The reversible sulfenic acid can serve as an intermediate leading to subsequent irreversible oxidation steps via ROS to sulﬁnic (-SO2 H) and sulfonic (-SO3 H) acids.

directly and indirectly to these neuropathologic processes. One example for an indirect neurotoxic mechanism mediated by MMP-2 involves the α-chemokine SDF-1 and its possible role in the development of HIVassociated dementia (27). SDF-1 is the natural ligand of the HIV-1 coreceptor CXCR4. Importantly, increased expression and activation of

May 26, 2005

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Post-Translational Modification

Fig. 2.

77

(Continued )

MMPs, including MMP-2 and MMP-9, were detected in HIV-infected macrophages and also in post mortem brain specimens from AIDS patients compared with uninfected controls (28). Furthermore, the expression of SDF-1 also appears elevated in the brains of HIV patients (29). We had reported previously that SDF-1 per se can indeed be toxic to mature neurons (30). As elegantly shown by Power and colleagues, MMP-2 released from HIV-infected macrophages is able to proteolytically remove four amino acid from the N -terminus of SDF-1. This truncated form of SDF-1 no longer

May 26, 2005

78

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Z. Gu, M. Kaul & S.A. Lipton

binds CXCR4 and is an even more powerful neurotoxin than full length SDF-1 (27). In contrast to chronic disorders like HIV-associated dementia, MMPs appear to directly contribute to neuronal cell death and blood-brain barrier damage in the case of stroke. MMP-9 in particular is signiﬁcantly elevated in humans after stroke (31), and MMP-2 levels are acutely increased in the brains of baboons after stroke (35). Mice deﬁcient in MMP-9 manifest a reduction in cerebral infarct size; in addition, treatment with MMP inhibitors or antibodies also reduces infarct size (32–34). 9. Neuronal NOS-Associated Activation of MMP During Stroke We have examined the association of MMP-9 and NO after focal cerebral ischemia and reperfusion in rodents. Accumulating evidence has shown that brain damage after middle cerebral artery (MCA) occlusion involves ischemia/reperfusion-induced NO production (17, 36). We monitored immunoreactivity of nNOS and MMP-9 in the ischemic cortex by double immunoﬂuorescence staining and observed substantial co-localisation of MMP-9 and nNOS (21), indicating the coincident production of NO and MMP-9 activity following ischemia and reperfusion. Gelatin zymography revealed an increase in both the level of proMMP-9 and in MMP-9 activity in the ischemic hemisphere compared to the contralateral control hemisphere (21). Immunoblotting with an anti-MMP-9 antibody also showed increased MMP-9 levels in the ischemic hemisphere. The slight decrease in actin in the damaged hemisphere may reﬂect cell loss. Under our conditions, MMP-2 was not activated (Gu and Lipton, unpublished observations). Similar changes in MMP-9 have recently been reported after human embolic stroke (31). In situ zymography and immunocytochemistry were used to examine the cellular localisation of MMP-9 enzymatic activity. We found that MMP activity was signiﬁcantly elevated in ischemic brain parenchyma after ischemia and reperfusion (21). Moreover, to further demonstrate the pathophysiological relevance of these ﬁndings and their relationship to NO, activation of MMP was abrogated after stroke in nNOS knockout mice or in wild-type animals that were treated with the relatively speciﬁc nNOS inhibitor 3-bromo-7-nitroindazole (3br7NI). Previously, neuroprotection had been demonstrated under either of these conditions of NOS inhibition (Gu and Lipton, unpublished observations and Ref. 36). In wild-type animals not treated with NOS inhibitors, immunocytochemistry revealed

May 26, 2005

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Post-Translational Modiﬁcation

79

that many cells positive for the neuronal marker NeuN also manifested MMP activity (21). 10. Exogenous NO-Activated MMP-9 Induces Neuronal Apoptosis in Cerebrocortical Cultures We also evaluated the eﬀects of NO-activated MMP-9 on neuronal cell death in cerebrocortical cultures. MMP activity was assessed by in situ zymography, neurons were identiﬁed by immunoreactivity for MAP-2, and nuclear morphology was monitored with Hoechst 33342. We found that the percentage of neurons exhibiting MMP activity signiﬁcantly increased after exposure to R-proMMP-9 pre-activated with SNOC compared to R-proMMP-9 alone. SNOC from which NO was dissipated did not activate R-proMMP-9 and did not increase the percentage of neurons exhibiting MMP activity. Additionally, 18 hours after exposure to SNOC-activated R-proMMP-9, we assessed apoptotic neurons by staining with anti-MAP-2 and terminaldeoxynucleotidyl-transferase-mediated deoxyuridine triphosphate nick-end labelling (TUNEL) in conjunction with condensed nuclear morphology assessed with Hoechst 33342. For these experiments, R-proMMP-9 was preexposed and hence preactivated by SNOC; NO had already been released from SNOC by the time the cultures were incubated with the activated MMP, as evidenced by measurement with an NO-sensitive electrode (WPI, Sarasota, FL) (9). Hence, direct release of NO from SNOC or the formation of peroxynitrite (ONOO− ) due to the release of NO from SNOC and subsequent reaction with superoxide anion (O− 2 ) could not have accounted for the observed neuronal apoptosis (9). NO-activated MMP-9 resulted in signiﬁcantly increased neuronal apoptosis, whereas treatment with the MMP inhibitor GM6001 blocked the neuronal cell death. We also observed many neurons coming up oﬀ the dish after exposure to NO-activated MMP-9. These results strongly suggest that even high levels of inactivated proMMP-9 protein do not have a deleterious eﬀect on neurons. However, activation of MMP-9 by NO has toxic eﬀects. One caveat with these ﬁndings is that nNOS deletion or NOS inhibition diminishes stroke damage, and hence one could argue that other stroke-related processes responsible for MMP activation would be reduced. Nonetheless, taken together with the data demonstrating S-nitrosylation of MMPs and the fact that we found MMPs activated in this fashion cause neuronal apoptosis in vitro makes it likely that NO activation of MMPs participates in neuronal injury in vivo.

May 26, 2005

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

80

Z. Gu, M. Kaul & S.A. Lipton

11. MMPs Degrade Extracellular Matrix Components Maintenance of normal functions of neurons depends on cell-matrix interactions that underlie phosphoinositol 3-kinase (PI3-K)/Akt (protein kinase B) cascades, and interruption of these survival-signalling pathways can trigger anoikis (apoptosis resulting from loss of cell-matrix homeostasis) (37, 38). In addition to this indirect inﬂuence on apoptosis, MMPs may also possibly directly trigger apoptosis via proteolysis of extracellular matrix. MMP-mediated degradation of parenchymal basal lamina may aﬀect cell survival (39, 40; Gu and Lipton, unpublished observations). Another possible pathway for the deleterious eﬀects of MMPs after stroke involves the ability of MMPs to digest the components of vascular matrix, including basal lamina and tight junction proteins. Chan and colleagues (41) suggested that superoxide anion might participate in MMP activation and alteration of capillary permeability. Damage to vascular integrity would then lead to disruption of the blood-brain barrier with secondary brain edema and cell death (33, 35, 42). Moreover, degradation of the tight junction protein zonae occludens-1 (ZO-1), a substrate for MMP-9, was observed after cerebral ischemia, and a targeted knockout of the MMP-9 gene ameliorated this degradation and subsequent blood-brain barrier degradation (43). However, one caveat here is that the authors could not make a ﬁrm link between the two diﬀerent cellular compartments for extracellular MMP-9 activation and intracellular ZO-1 degradation. Another substrate that may be targeted by deleterious MMP activity after ischemia and trauma is represented by the white matter component, myelin basic protein (MBP) (44). White matter damage is a major pathological outcome and cause of functional impairment after brain injury. After cerebral ischemia, MBP is degraded, and deletion of the MMP-9 gene reduced the severity of damage to this critical white matter protein (43). Additional data, initially from the pioneering work of the Rosenberg laboratory, suggest that abnormal MMP activation participates in blood-brain barrier damage after ischemia (42, 44).

12. Summary S-nitrosylation and subsequent oxidation of protein thiol in the prodomain of MMP-9 can lead to enzyme activation. It is likely that other, homologous MMPs are activated in a similar manner. This series of reactions confers responsiveness of the extracellular matrix to nitrosative and

May 26, 2005

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Post-Translational Modiﬁcation

81

oxidative stress. Such insults are relevant to a number of pathophysiological conditions, including cerebral ischemia and neurodegenerative diseases. Extracellular proteolytic cascades triggered by MMPs can disrupt the extracellular matrix, contribute to cell detachment, and lead to a form of apoptotic cell death known as anoikis, similar to that observed in our neuronal cultures (45). The elucidation of this novel extracellular signalling pathway to neuronal apoptosis involving NO-activated MMPs may contribute to the development of new therapies for stroke, multiple sclerosis, HIVassociated dementia, Alzheimer’s disease, and other disorders associated with nitrosative and oxidative stress. The reactions described here are believed to represent the ﬁrst case of combined nitrosative/oxidative activation of an enzyme that leads to cell apoptosis, and as such may represent a more general paradigm in molecular toxicology.

Acknowledgments We thank Chung Ju for cerebrocortical cultures, Jiankun Cui for surgical and histological works, Weizhong Li and Kosi Gramatikoﬀ for modelling and illustration of the MMP-9 crystal structure, and Robert Liddington, Steve Kridel, Jeﬀ Smith, Alex Strongin and Grey del Zoppo for helpful discussions. Z.G. was supported by NIH NRSA fellowship; M.K. was supported by the American Foundation for AIDS Research and the NIH; and S.A.L. was supported in part by grants from NIH and the American Heart Association.

References 1. Sternlicht, M. D., and Werb, Z. (2001) Annual Review of Cell & Developmental Biology 17, 463–516 2. Yong, V. W., Power, C., Forsyth, P., and Edwards, D. R. (2001) Nature Reviews Neuroscience 2(7), 502–11 3. Van den Steen, P. E., Dubois, B., Nelissen, I., Rudd, P. M., Dwek, R. A., and Opdenakker, G. (2002) Critical Reviews in Biochemistry & Molecular Biology 37(6), 375–536 4. Visse, R., and Nagase, H. (2003) Circulation Research 92(8), 827–39 5. Morgunova, E., Tuuttila, A., Bergmann, U., Isupov, M., Lindqvist, Y., Schneider, G., and Tryggvason, K. (1999) Science 284(5420), 1667–70 6. Egeblad, M., and Werb, Z. (2002) Nature Reviews Cancer 2(3), 161–74 7. Van Wart, H. E., and Birkedal-Hansen, H. (1990) Proceedings of the National Academy of Sciences of the United States of America 87(14), 5578–82

May 26, 2005

82

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Z. Gu, M. Kaul & S.A. Lipton

8. Dawson, T. M., and Snyder, S. H. (1994) Journal of Neuroscience 14(9), 5147–59 9. Lipton, S. A., Choi, Y. B., Pan, Z. H., Lei, S. Z., Chen, H. S., Sucher, N. J., Loscalzo, J., Singel, D. J., and Stamler, J. S. (1993) Nature 364(6438), 626–32 10. Melino, G., Bernassola, F., Knight, R. A., Corasaniti, M. T., Nistico, G., and Finazzi-Agro, A. (1997) Nature 388(6641), 432–3 11. Stamler, J. S. (1994) Cell 78(6), 931–6 12. Choi, Y. B., Tenneti, L., Le, D. A., Ortiz, J., Bai, G., Chen, H. S., and Lipton, S. A. (2000) Nature Neuroscience 3(1), 15–21 13. Jaﬀrey, S. R., Erdjument-Bromage, H., Ferris, C. D., Tempst, P., and Snyder, S. H. (2001) Nature Cell Biology 3(2), 193–7 14. Jia, L., Bonaventura, C., Bonaventura, J., and Stamler, J. S. (1996) Nature 380(6571), 221–6 15. Matthews, J. R., Botting, C. H., Panico, M., Morris, H. R., and Hay, R. T. (1996) Nucleic Acids Research 24(12), 2236–42 16. Kumura, E., Kosaka, H., Shiga, T., Yoshimine, T., and Hayakawa, T. (1994) Journal of Cerebral Blood Flow & Metabolism 14(3), 487–91 17. Sato, S., Tominaga, T., Ohnishi, T., and Ohnishi, S. T. (1994) Brain Research 647(1), 91–6 18. Stamler, J. S., Toone, E. J., Lipton, S. A., and Sucher, N. J. (1997) Neuron 18(5), 691–6 19. Kridel, S. J., Chen, E., Kotra, L. P., Howard, E. W., Mobashery, S., and Smith, J. W. (2001) Journal of Biological Chemistry 276(23), 20572–8 20. Wink, D. A., Kim, S., Coﬃn, D., Cook, J. C., Vodovotz, Y., Chistodoulou, D., Jourd’heuil, D., and Grisham, M. B. (1999) Methods in Enzymology 301, 201–11 21. Gu, Z., Kaul, M., Yan, B., Kridel, S. J., Cui, J., Strong, A., Smith, J. W., Liddington, R. C., and Lipton, S. A. (2002) Science 297(5584), 1186–90 22. Stamler, J. S., and Hausladen, A. (1998) Nature Structural Biology 5(4), 247–9 23. Yan, B., and Smith, J. W. (2000) Journal of Biological Chemistry 275(51), 39964–72 24. Campbell, I. L., and Pagenstecher, A. (1999) Trends in Neurosciences 22(7), 285–7 25. Lukes, A., Mun-Bryce, S., Lukes, M., and Rosenberg, G. A. (1999) Molecular Neurobiology 19(3), 267–84 26. Sivak, J. M., and Fini, M. E. (2002) Progress in Retinal & Eye Research 21(1), 1–14 27. Zhang, K., McQuibban, G. A., Silva, C., Butler, G. S., Johnston, J. B., Holden, J., Clark-Lewis, I., Overall, C. M., and Power, C. (2003) Nature Neuroscience 6(10), 1064–71 28. Johnston, J. B., Jiang, Y., van Marle, G., Mayne, M. B., Ni, W., Holden, J., McArthur, J. C., and Power, C. (2000) Journal of Virology 74(16), 7211–20 29. Langford, D., Sanders, V. J., Mallory, M., Kaul, M., and Masliah, E. (2002) Journal of Neuroimmunology 127(1–2), 115–26

May 26, 2005

16:1 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch03

Post-Translational Modiﬁcation

83

30. Kaul, M., and Lipton, S. A. (1999) Proceedings of the National Academy of Sciences of the United States of America 96(14), 8212–6 31. Montaner, J., Alvarez-Sabin, J., Molina, C., Angles, A., Abilleira, S., Arenillas, J., Gonzalez, M. A., and Monasterio, J. (2001) Stroke 32(8), 1759–66 32. Romanic, A. M., White, R. F., Arleth, A. J., Ohlstein, E. H., and Barone, F. C. (1998) Stroke 29(5), 1020–30 33. Gasche, Y., Fujimura, M., Morita-Fujimura, Y., Copin, J. C., Kawase, M., Massengale, J., and Chan, P. H. (1999) Journal of Cerebral Blood Flow & Metabolism 19(9), 1020–8 34. Asahi, M., Asahi, K., Jung, J. C., del Zoppo, G. J., Fini, M. E., and Lo, E. H. (2000) Journal of Cerebral Blood Flow & Metabolism 20(12), 1681–9 35. Heo, J. H., Lucero, J., Abumiya, T., Koziol, J. A., Copeland, B. R., and del Zoppo, G. J. (1999) Journal of Cerebral Blood Flow & Metabolism 19(6), 624–33 36. Huang, Z., Huang, P. L., Panahian, N., Dalkara, T., Fishman, M. C., and Moskowitz, M. A. (1994) Science 265(5180), 1883–5 37. Bachelder, R. E., Wendt, M. A., Fujita, N., Tsuruo, T., and Mercurio, A. M. (2001) Journal of Biological Chemistry 276(37), 34702–7 38. Gary, D. S., and Mattson, M. P. (2001) Journal of Neurochemistry 76(5), 1485–96 39. Chen, Z. L., and Strickland, S. (1997) Cell 91(7), 917–25 40. Lee, S. R., Tsuji, K., and Lo, E. H. (2004) Journal of Neuroscience 24(3), 671–678 41. Gasche, Y., Copin, J. C., Sugawara, T., Fujimura, M., and Chan, P. H. (2001) Journal of Cerebral Blood Flow & Metabolism 21(12), 1393–400 42. Rosenberg, G. A., Dencoﬀ, J. E., McGuire, P. G., Liotta, L. A., and Stetler-Stevenson, W. G. (1994) Laboratory Investigation 71(3), 417–22 43. Asahi, M., Wang, X., Mori, T., Sumii, T., Jung, J. C., Moskowitz, M. A., Fini, M. E., and Lo, E. H. (2001) Journal of Neuroscience 21(19), 7724–32 44. Rosenberg, G. A., Sullivan, N., and Esiri, M. M. (2001) Stroke 32(5), 1162–8 45. Cardone, M. H., Salvesen, G. S., Widmann, C., Johnson, G., and Frisch, S. M. (1997) Cell 90(2), 315–23 46. Becker, K., Savvides, S. N., Keese, M., Schirmer, R. H., and Karplus, P. A. (1998) Nature Structural Biology 5, 267–71

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

This page intentionally left blank

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

PART III MMP FUNCTION

85

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

This page intentionally left blank

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

CHAPTER 4 SUBSTRATES FOR METALLOENDOPEPTIDASES IN THE CENTRAL NERVOUS SYSTEM

P.E. Gottschall1,∗ , J.D. Sandy1,2 and D.R. Zimmermann3 1

University of South Florida College of Medicine, Department of Pharmacology and Therapeutics, Tampa, FL 33612-4799 2

Shriners Hospital for Children, Tampa, FL 33612-9499

3

University Hospital of Zurich, Department of Pathology, CH-8091 Zurich, Switzerland ∗ E-mail: [email protected]

1. Introduction The matrix metalloproteinases (MMPs) and related matrix-degrading metalloendopeptidases (ADAMTSs and ADAMs) most often appear in physiology and pathology during events related to cell and tissue remodelling, and function per their name, in the cleavage and/or degradation of extracellular matrix (ECM) proteins as well as non-matrix protein substrates. Other sources including chapters in this volume have delineated the fact that MMPs are important modulators of key functions in the nervous system from progenitor migration during development, to diﬀerentiation of glial cells, to enhancing axonal sprouting (1, 2). Yet the crucial role for metalloendopeptidases in developmental physiology somewhat belies their potential for tissue destruction in various neuropathologies in the adult ranging from multiple sclerosis to cerebrovascular stroke to Alzheimer’s disease. In this regard, however, it should be noted that even in the adult nervous system, the actions of metalloendopeptidases are clearly not conﬁned to pathological eﬀects. Yet, how do proteolytic enzymes that primarily act outside cells, most often by binding to and cleaving matrix substrates, inﬂuence cell behaviour? Although metalloendopeptidases may directly activate intracellular signalling pathways via ‘receptors’, their Correspondence to: P.E. Gottschall 87

May 26, 2005

88

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

classical and primary action is exerted by a ‘simple’ cleavage of a matrix or cell-associated substrate. Thus, cleavage and/or further proteolytic degradation of a metalloproteinase substrate has the potential to: (1) alter the composition, ﬂuidity and rigidity of the extracellular milieu providing an environment that changes the likelihood of diﬀerential cell growth and behaviour within it (3, 4), (2) mobilise and activate (or de-activate) matrixbound growth factors (5) that use ECM as storage depots, (3) change matrix-to-cell-directed signal transduction that involves a matrix protein directly or indirectly linked to the cell surface, or (4) result in the formation of a molecule with novel biological activity compared to its (latent) parent protein by proteolytically exposing a cryptic site (6, 7) (Fig. 1). Thus, the ultimate function of these metalloendopeptidases involves not only the protease itself, but one or more substrate partners that specify the molecular change resulting from proteolysis. Most evidence for identifying a substrate for a metalloendopeptidase has come from in vitro experiments that incubate a puriﬁed proteinase with crude or puriﬁed substrate, and then chemically or immunochemically determining whether the amount of intact substrate was reduced or whether a novel cleavage fragment was generated. This tactic alone cannot establish whether a particular protein substrate is proteolytically cleaved by a speciﬁc proteinase in vivo. Since somewhat selective inhibitors for metalloendopeptidases are available, one approach to this question is to selectively block proteolytic activity with one of these small molecule inhibitors. Newer in vitro methods to identify novel and more selective substrates for these proteases are being developed at a rapid pace (8–12). The purpose of this chapter is to review some structural-functional aspects of proteins that bind to and are proteolytically cleaved by metalloendopeptidase members of the metzincin family, the MMPs, the ADAMs (a disintegrin and metalloproteinase) and the ADAMTSs (a disintegrin and metalloproteinase with thrombospondin motifs). Because this is such a broad topic, the review will be limited to what we believe to be the major metalloendopeptidase substrates in the basal lamina of the cerebrovasculature, substrates present in the neuropil and perineuronal (pericellular) nets, and other identiﬁed substrates present in the central nervous system (CNS). It is well beyond the scope of this review to cover the structure and function of each of these matrix proteins in great detail. In discussing each protein substrate, focus will be placed on the proteolytic role of the metalloendopeptidases in cleaving that protein.

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

89

Fig. 1. Schematic of functional location of extracellular matrix in the central nervous system. Described are not all the possible interactions of cell-matrix interactions in the nervous system, but those for which evidence is available. (1) alter the composition, fluidity and rigidity of the extracellular milieu providing an environment that changes the likelihood of plasticity, (3, 4) or (2) mobilise and activate matrix-bound growth factors (5) (3) eliminate matrix-to-cell directed signal transduction, (4) result in the formation of a molecule with novel biological activity compared to its (latent) parent protein by proteolytically exposing a cryptic site.

2. Substrates in the Basal Lamina (Basement Membrane) Basal lamina (basement membranes) are mostly found beneath epithelia wherever they meet connective tissue (13) and consist of ECM protein aggregates weaved into a thin sheet (Fig. 1). In blood vessels, this specialised ECM separates endothelial cells and pericytes from the surrounding extracellular space. Basement membranes provide morphogenic cues to the epithelia that determine the fate of cells, the polarisation of sub cellular components and the location of receptors and transporters (14). In capillaries in brain, astroglia send ‘endfeet’ processes that form lamellae and contact the basement membrane and unlike many other tissues, astrocytes along with endothelia, contribute to the production of proteins that make up the cerebrovascular basal lamina (15). Basal lamina

May 26, 2005

90

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

proteins self assemble into heteromeric aggregates (16, 17) and the individual components include collagen type IV (13, 18), laminin (19), heparan sulfate proteoglycans, particularly perlecan (20), entactin/nidogen (21), ﬁbronectins (22) and SPARC (secreted protein acidic and rich in cysteine (23). It is well known that the presence of tight junctions and specialised transport mechanisms of brain capillary endothelial cells are the structural basis for the blood-brain-barrier, the functional specialisation of CNS capillaries that restricts the diﬀusion of hyrodrophilic solutes and protects the brain from sudden changes in plasma composition. However, the morphological and physiological integrity of the overall blood-brain barrier depends on integrin-dependent interaction of endothelial cells and astrocytes with an intact basal lamina (24, 25). Consistent with this notion is that disruption of the basement membrane by proteolysis or other means can aﬀect the leakiness of the blood-brain barrier. The ECM of the basal lamina also helps to protect the brain parenchyma from edema and hemorrhage (26). Several physiological mechanisms and disease states target the cerebral microvasculature and involve proteolytic breakdown of the basal lamina to one extent or another. These include leukocyte or tumour cell extravasation in either direction (i.e. blood to neuropil, neuropil to blood) (27), focal ischemia (28), amyloid angiopathy (29, 30), and traumatic brain injury (31) to identify just a few. In each of these conditions, there are unanswered questions about the presence, location and action of the extracellular protease(s) involved in the pathology. Nonetheless, there is good indirect evidence in most of these cases that MMPs play some role in proteolysis of the basal lamina. Details for each of these activities of the MMPs are described elsewhere in this volume. Here the purpose is a brief discussion of the individual basement membrane substrates in the cerebrovasculature. 3. Collagen Type IV The most abundant structural protein of the basal lamina is collagen IV, the component that mainly determines its biomechanical stability and macromolecular organisation. Collagen IV forms triple helical structures each containing three subunits encoded by six distinct α chain genes, although despite the many potential permutations of the six chains, collagen IV is apparently formed from only three combination sets of triple helical molecules, [α1 (IV)2α2 (IV)] the classical and most abundant molecule and that which exists in cerebral vessels, [α3 (IV)α4 (IV)α5 (IV)], and [α5 (IV)2α6 (IV)] (32). Like most collagens, collagen IV contains a variable number of Gly-X-Y repeats with several interruptions that allow for

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

91

ﬂexibility in structure and non-collagenous domains at the aminoterminal (7S) and the carboxyterminal (NC1) end. Several MMPs are capable of cleaving collagen IV including MMP-13, MMP-2, MMP-9, MMP-3, MMP-7, MMP-25 and MMP-12 (for review, see (7)), and when digested under particular conditions, the products are two fragments consisting of one-quarter and three quarters of the intact molecule (33). Collagen IV is proteolytically degraded during focal ischemia (34) via a mechanism that may or may not involve activated MMPs (28). It is interesting that in vitro a single cleavage of collagen IV did not appear to inﬂuence the binding of collagen IV to its integrin receptors (33). 4. Laminins Laminins are composed of combinations of three diﬀerent chains termed α (α1 − α5), β (β1 − β4) and γ (γ1 − γ3). These chains associate as heterotrimers (one subunit from each group) and thus far, ﬁfteen diﬀerent variants (laminin-1 to -15) have been identiﬁed that show tissue-speciﬁc and developmental-selective expression. Laminin-1 (α1 β1 γ1 ) is highly expressed in basement membrane where it aggregates to large networks and interacts with other ECM proteins and cellular receptors (17). Laminins exist as a cross shaped molecule with the N-terminal domains from the α chain (about 400 kD), β chain (about 200 kD) and γ chain (about 200 kD) making up the short arms that contain EGF-like motifs arranged into linear rods interrupted by globular domains. All three subunits make up the long arm of the cross with the α chain being the longest and the carboxy terminus consisting of ﬁve globular domains. The short and long arm of the α subunit seem to be the most biologically relevant portions of the molecule (35). Laminin-5, which is present in cerebral microvessels, is involved in hemidesmosome formation with the integrin α6β4. Interestingly, α6β4 was demonstrated to be localised at the astrocyte-cerebrovessel interface, and these integrin subunits (36), along with laminin-5 immunoreactivity, were lost early after ischemia). Laminin-1 is an in vitro substrate for several members of the MMP family including the MMP-1, MMP-2 and MMP-9, MMP-3 and MMP-11, MMP-7, MMP-12, and the membrane-type (MT)MMPs, MMP-14, MMP-15, MMP-16 (37, 38). 5. Perlecan and Other Heparan Sulfate Proteoglycans (HSPG) Perlecan is a heparan sulfate-containing proteoglycan that was shown to be present in all native basement membranes (39, 40). The core protein

May 26, 2005

92

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

of human perlecan has a predicted molecular weight of 467 kD containing attachment sites for both heparan sulfate (HS) and chondroitin sulfate (CS) chains. Based on the cDNA sequence of perlecan, there appear to be 5 distinct functional domains: a unique N-terminal domain that is essential for HS attachment, a domain comparable to the cholesterol binding region of the low density lipoprotein (LDL) receptor, a domain homologous to the short arm of the laminin α1 chain, a domain with several immunoglobulin repeats similar to those found in neural cell adhesion molecule (N-CAM), and a domain that contains EGF-like globular domains that are found in laminins. Importantly, perlecan binds both ﬁbroblast growth factor (FGF) and platelet-derived growth factor (PDGF) via its heparan sulfate chains (41, 42), yet proteolytic cleavage of endothelial-derived perlecan results in the release of bound growth factor (43). Several of the more classical MMPs in addition to MMP-13, MMP-14, and MMP-15 have been shown to cleave perlecan. Over-expression of transforming growth factor β (TGFβ) in mice results in microvascular basement membrane thickening, increased levels of perlecan and ﬁbronectin in cerebrovascular basement membranes, deposition of β-amyloid peptide in the vessels and an increased likelihood of cerebral hemorrhage in old age (44). Other basement membrane heparan sulfate proteoglycans include agrin and collagen XVIII. Both are present in the basal lamina of cerebral blood vessels and accumulate in amyloid plaques of Alzheimer patients (45) and in the case of agrin, also in the neuroﬁbrillary tangles (46). Agrin and collagen XVIII are in vitro substrates of MMP-3 and MMP-7, respectively (47, 48). 6. Entactins (Nidogens) Entactin/nidogens are monomeric molecules that arise from two separate genes. Both proteins, entactin/nidogen-1 and entactin/nidogen-2, consist of two aminoterminal (G1 and G2) and one carboxyterminal globular domains (G3), being separated by rod-like structures that include EGFrepeats. At the carboxyterminal end, several LDL receptor ‘YWTD’ motifs have been identiﬁed in the G3 domains. It has been proposed that entactins link collagen and laminin networks in basement membranes (49), because the G3 domain binds with high-aﬃnity to an EGF-like element of the laminin γ1 chain (50), whereas the G2 domain interacts tightly with collagen IV and the perlecan core protein (51, 52). Entactin-1 (nidogen-1) is cleaved by MMP-1, MMP-2, MMP-3, MMP-7, MMP-14, MMP-15, and MMP-12. Interestingly, selective disruption of the entactin-1 gene in mice

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

93

results in a phenotype that exhibited seizures and the lack of muscle control in the hind legs. Structural alterations in the basement membranes of blood vessels were found only in brain capillaries and the lens capsule (53), a more severe phenotype being most likely prevented by the compensatory action of entactin/nidogen-2. 7. Substrates in the Neuropil and Perineuronal Nets Studies in the late 19th and early 20th century, including those by Camillo Golgi and Ramon y Cajal, were seminal in describing an extracellular meshwork that surrounds neurons and other cells in the nervous system. However, with the advent of electron microscopy in the late 1950s and early 1960s, the actual existence of appreciable extracellular space within the brain was questioned by several anatomists because material they were observing showed little or no space between cellular elements in the nervous system. It turns out that early ﬁxation techniques for electron microscopy caused a marked shrinkage of tissue that left little remaining space between cells, and this and other reasons led to a lengthy period of declining interest in brain interstitial matrix (for greater detailed history, see (54)). Presently, there is general agreement that the volume of brain occupied by extracellular space is about 20% (55), and within this space lies a meshwork of PGs, glycoproteins, proteins without sugars and other less abundant, but functionally important molecules that make up the extracellular milieu of the CNS. Unlike the extracellular space of most other tissues, the matrix in the brain does not contain appreciable levels of ﬁbril-forming collagens, although astrocytes and gliomas do produce collagens when in culture). Matrix molecules that are expressed and deposited in the extracellular milieu exist as complex aggregates (3, 56, 57) and qualitative and quantitative expression of these can vary greatly, from those produced during development, to those present in normal adult brain, to those which are synthesised and secreted after injury to the nervous system. These changes in matrix expression clearly have a functional relevance as ECM molecules modulate cell migration, diﬀerentiation, pathﬁnding and topographical map formation, and even synaptogenesis during development. In the healthy adult, there is evidence to indicate that ECM can modulate neural and synaptic plasticity (3, 56–58), and after injury and in disease, glial scars that contain abundant ECM molecules are potent inhibitors of compensatory neurite outgrowth. What follows in this section is a brief description of the structural and functional elements of the

May 26, 2005

94

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

neuropil-localised, matrix molecules in the nervous system with an emphasis on their nature and function as substrates for metalloendopeptidases. 8. Proteoglycans Excellent and thorough reviews updating functional roles of PGs in the nervous system have been published very recently and the reader is referred to these for functional information (3, 56–60). The purpose here is to summarise information related to structure, expression and proteolytic processing of the PGs in the neuropil and perineuronal nets. Identiﬁcation of CS-substituted PGs in the nervous system has been conducted with several methods. A number of techniques identify the CS itself and provide an overall picture of the quantity and tissue distribution of this glycosaminoglycan. In the CNS, the predominant core proteins that bear CS chains are versican, brevican, neurocan, aggrecan (all belonging to the hyalectans), phosphacan (a splice-variant of RPTPζ/β) and NG2 (a transmembrane proteoglycan containing CS). Important reagents used to detect CSs are plant lectins, especially the lectin from Wisteria ﬂoribunda that apparently has speciﬁcity for carbohydrate structures terminating in N-acetylgalactosamine. This particular lectin intensely stains perineuronal nets and other neuropil-associated matrix in rodent brain (61). A second method for the identiﬁcation of CS is the use of antibodies that speciﬁcally recognise various epitopes of the CS chains. One example is CS56, an antibody that detects intact CS chains containing both 4-sulfated and 6-sulfated galactosamine (62). Thirdly, antibodies that recognise various epitopes on the individual core proteins of the hyalectans are important reagents to quantify and localise the individual PGs. The individual core proteins show a diverse localisation and expression in the CNS, and thus, have a diversity of function even though they are of highly similar structure. One example is the presence of aggrecan and brevican, but not versican in perineuronal nets in adult brain. 9. Hyalectans (Lecticans, Aggregating PGs) Structure: There is basic, domain structural and functional homology among the hyalectans. The N- and C-terminal globular domains are the most highly conserved among members of this family that currently includes aggrecan (63), versican (64), neurocan (65) and brevican/BEHAB (66, 67) (Fig. 2). Each of these proteoglycans consists of a core protein covalently

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

95

Fig. 2. Schematic drawings of the various hyalectans illustrating the globular G1 domain at the N-terminus, the globular G3 domain at the C-terminus, and the central domain bearing the chondroitin sulfate chains. Of note, (1) aggrecan contains an additional globular domain, G2 that may also bind hyaluronic acid, (2) versican exists in four alternatively spliced forms, (3) brevican is alternatively splice to create the GPIlinked variant. The arrow indicates the major, putative ADAMTS-cleavage site in each protein (adapted from Bandtlow and Zimmermann, 2000).

modiﬁed by glycosaminoglycan side chains bound to serine residues on a variably-sized, extended central GAG-attachment domain (see Fig. 2). The N-terminal globular domain of hyalectans, termed G1, contains an immunoglobulin-like loop and a hyaluronan-binding tandem repeat homologous to other hyaluronan-binding proteins such as link proteins, CD44 and TSG-6. The diversity in molecular size and sequence among the hyalectans is mainly due to the presence of non-homologous central domains in their core proteins. These central domains may carry several covalently bound CS chains, which are repeating disaccharide units of glucuronic acid and variably sulfated N-acetylgalactosamine. In some cases, the glucuronic acid is epimerised to iduronic acid and the glycosaminoglycan is then termed dermatan sulfate. These glycosaminoglycans are attached to the core protein central domain through a speciﬁc linkage-region oligosaccharide (Xyl-Gal-Gal-GA) which is bound to the core protein by a xylose-serine attachment. The consensus for glycosaminoglycan linkage is Gly-Ser or Ser-Gly dipeptide sequences ﬂanked by a variable number

May 26, 2005

96

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

of acidic residues. The number of potential glycosaminoglycan-attachment sites in the central region of the core protein varies markedly among the hyalectans. There are 3 glycosaminoglycan consensus attachment sites in brevican, over 7 sites in neurocan, about 20 sites in versican V0 and up to 100 in aggrecan. The individual chondroitin side chains themselves also vary in size between 10 kD and 45 kD. To add further variation, the size of the core protein of the individual hyalectans may also diﬀer as a consequence of alternative exon usage. For instance, the central GAG-attachment domain of the versican gene contains two long, alternatively spliced exons, GAGα and GAGβ that encode four core protein isoforms (68–70). Versican V3, the smallest variant, contains neither GAG-attachment and is made up entirely of the terminal globular regions, versican V2 contains the GAGα domain, versican V1 the GAGβ domain, and versican V0 includes both GAGattachment domains and is the largest hyalectan core protein described. Brevican, another alternatively spliced hyalectan, gives rise to a secreted isoform and a GPI-anchored variant with a spliced carboxy-terminus lacking the G3 domain (71). In aggrecan, the alternative splicing aﬀects the EGFlike elements and the carboxyterminal sushi-domain within G3. Human aggrecan contains in addition a unique highly repetitive sequence within the CS attachment region encoded by a variable number tandem repeat polymorphism that gives rise to individual diﬀerences among humans (72). Aggrecan also bears a second N-terminal globular domain (G2) which is not present in other hyalectans. This exclusive globular domain is separated from the G1 domain by a short, proteolytically sensitive, interglobular region (termed IGD). The G2 domain consists of an additional, tandem repeat motif highly similar to its counterpart in G1. In contrast to G1, however, it does not display any hyaluronan-binding activity. The C-terminal globular domain G3, is present on all hyalectans, (except for the GPI-linked isoform of brevican) and consists of EGF-like repeats, a C-type lectin motif, and a sushi domain. The same combination of motifs are found in the selectin family of proteins, although in a diﬀerent N -terminal to C-terminal sequence compared to the hyalectans and the tertiary structure for these domains suggests high similarity with the C-type lectin domain of mannose-binding protein (73). Each of the G3 regions of the hyalectans have been shown to bind to ﬁbronectin type III repeats of tenascin-R and tenascin-C via their c-type lectin domains (74, 75). This protein-protein interaction may act to further crosslink hyaluronanhyalectan complexes (76). Hyalectans do not always carry glycosaminoglycan chains. About 50% of brevican for instance exists in the adult CNS as chain-free core glycoprotein

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

97

and therefore belongs to the so-called ‘part-time’ proteoglycans (66). Furthermore, versican V3 lacks a glycosaminoglycan-attachment domain and therefore can only exist as glycoprotein (70).

10. Expression, Localisation and Molecular/Cellular Interactions The tissue expression of hyalectans demonstrate characteristic patterns with the highest expression of aggrecan in cartilage, versicans in a wide variety of tissues including blood vessels and neurocan and brevican restricted to the CNS. There is abundant, yet diﬀerential expression of hyalectans in the CNS. Brevican, versican V2 and aggrecan production increase steadily in rodent brain starting at about post-natal day 10, reaching a plateau in the young adult that is at a level about 15-times that observed at embryonic day 14. In contrast, neurocan and versican V1 appear transiently, expression is low in the early embryo, production increases markedly before and around the time of birth, and expression returns to low levels in the adult organism (77). There is evidence that these PGs take part in the control of cell migration processes and in axonal guidance during development (78–80). These temporal changes may be accompanied by changes of the cellular origin of a particular hyalectan during development. In the hippocampal ﬁmbria for example, oligodendrocytes and their precursors express brevican up to and during the active myelination period (post-natal day 7–21). Subsequently, brevican production and secretion is taken over by astrocytes which appear to be responsible for the high brevican expression in the adult white matter (81). In the adult, brevican is abundant and widely expressed in the CNS. It takes part in the formation of perineuronal nets (82) and is present throughout the neuropil likely being produced by both astrocytes and particular sub-populations of neurons. In general, the expression of the large secreted brevican isoforms appear to be higher than their GPIlinked counterpart (83), which is associated with neural membranes (84). It may be deposited adjacent to synapses (82), and brevican mRNA has been localised in vivo to particular populations of neuron and glia (82, 83). The secreted, but not the GPI-anchored, brevican variant is up-regulated in gliosis after a brain lesion (83), it may be deposited adjacent to synapses (82), and brevican mRNA has been localised in vivo to particular populations of neuron and glia (82, 83). The other major hyalectan in the adult CNS is versican V2 which is primarily present in white matter and produced by

May 26, 2005

98

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

oligodendrocytes and their progenitors, (85–89). Versican V1 has almost completely disappeared in the mature tissue, except for the ECM around blood vessels where it is probably expressed by astrocytes. Hence, versican V2 is by far the most abundant versican isoform and possibly also the major hyalectan of the adult CNS (85). Like versican V1, intact neurocan is less abundant in adult compared to the developing nervous system and is produced by astrocytes (89). Brevican, neurocan and aggrecan (61, 90, 91) and their proteolyticallyderived fragments (4) (our unpublished observations), are each present within perineuronal nets and with most nets, there is a high propensity for co-localisation with the ECM glycoprotein, tenascin-R (90, 92). In addition to these proteinaceous molecules, hyaluronan (hyaluronic acid) is found in the CNS (93) in perineuronal nets and the neuropil (61, 90) where it binds speciﬁcally to the G1 domain of the hyalectans. Interestingly, two novel, brain-speciﬁc ‘link proteins’ were cloned recently (94, 95). Link proteins are lower molecular weight molecules (about 50 kDa) that contain an immunoglobulin-like region and tandem hyaluronate binding repeats and are involved in stabilising binding between the hyalectans and hyaluronic acid. Most investigations into hyaluronic acid binding to hyalectans have been conducted using aggrecan and cartilage link protein. Although both tandem repeats present in G1 aggrecan are essential for binding to hyaluronan, the presence of the immunoglobulin loop enhances binding (96). Link proteins bind to both aggrecan, possibly on the N-terminal immunoglobulin loop, and to hyaluronan, and markedly stabilise the complex that forms between aggrecan and hyaluronan (97). Other hyalectans, and proteolytic fragments of hyalectans, bind to the hyaluronic acid as well (98–100). Several proteins have been shown to bind to the C-type lectin domain in the C-terminus of the hyalectans, however, the most solid evidence for physiologically-relevant binding to a protein in the CNS is with tenascin-R (90, 101). These interactions of the hyalectans with hyaluronic acid and tenascin led to the development of a model for higher-level aggregation of these molecules, suggesting that complex lattices are formed by the binding of hyalectan-tenascin-hyaluronic acid (3, 76) and these lattices are tighter when hyalectans are abundant and when shorter hyalectans are involved (3). However, upon proteolytic cleavage of the hyalectan, the lattices are opened up, the extracellular complex becomes ‘loose’ and an environment is created that is more favourable toward plasticity, i.e. axon growth, synaptic remodelling, cell migration, etc.

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

99

11. Proteolytic Processing of the Hyalectans Western analysis of CNS extracts for aggrecan, versican, brevican or neurocan generally reveals the presence of the intact PG core protein, however, in many instances, lower molecular weight immunoreactive bands are observed. These proteolytic products appear to be generated in vivo since the fragments are observed when samples are prepared in the presence of a cocktail of proteinase inhibitors and the products of versican V2 processing in rat brain extracts are not more abundant after a lengthy post-mortem interval (102). Most often, the N-terminal G1-globular domain product may be detected which terminates C-terminally at sequences which are speciﬁc for each hyalectan (NITEGE for aggrecan, EAVESE for brevican, NIVSFE for versican V2 and at an unknown site for neurocan). The presence of this type of G1 domain-containing product was ﬁrst recognised in cartilage explant experiments where aggrecan proteolysis was found to result in the separation of the G1 domain from the remainder of the molecule (103). This process was later found to be due to cleavage at a Glu-Ala bond in the interglobular domain between the G1 and G2 domains (104), a ﬁnding that provided the assay system which led to the cloning of the aggrecanase group of ADAMTS proteinases (105). Earlier a glial-derived protein was identiﬁed (106) that was shown to bind to hyaluronate and termed glial hyaluronic acid binding protein (GHAP) (100). Although originally thought to be a novel hyaluronic acid-binding protein, GHAP was eventually discovered to be the ADAMTS-generated proteolytic cleavage fragment of versican V2 (102, 107). The proteolytic processing of neurocan and brevican was recognised during isolation of the molecules (66, 98) at the same time that the genes were cloned. Speciﬁc proteolytic cleavage sites have now been identiﬁed for aggrecan (104), brevican (108), versican V0, V1 (109), and versican V2 (102) (Fig. 3) and these sites are located in a conserved region about 400 residues from the N-terminus (Fig. 3). Although some of the N-terminal G1-domain products are N- and O-glycosylated, the apparent molecular weights (as determined by SDSPAGE) of the major fragments in brain tissue (after chondroitinase treatment) are 53 kD and 300 kD (intact ∼350 kD) for aggrecan (110), 55 kD and 80 kD (intact 145 kD) for brevican (4, 66, 111), and 66 kD and 250 kD for versican V2 (intact ∼300 kD) (102, 107). Whether these proteolyticallyderived fragments have any biological activity independent of the parent protein remains to be determined, although a matrix organising function is suggested by the ﬁnding that G1 fragments remain bound to hyaluronan in

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

100

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

Fig. 3. Glutamyl-endopeptidase cleavage sites identiﬁed for brevican, aggrecan and versican. Although these sites appear in the rather unconserved central region of the proteins, the cleavage sites are highly conserved among species. ADAMTS is the likely protease that cleaves at each of these sites (adapted from Yamaguchi, 2000).

the matrix. Indeed the G1 domain of brevican has been reported to mediate glioma migration into the neuropil (112). For the major, in vivo proteolytic cleavage of aggrecan, brevican and versican V2/V0, it now seems clear that the proteases responsible are not MMPs, but glutamyl endopeptidases that belong to the ‘aggrecanase’ subgroup of the related ADAMTS family of proteases. The original ADAMTS (ADAMTS1) was identiﬁed, isolated and cloned from a cachexigenic tumour cell line (113), with subsequent cloning of a highly related gene whose protein was found to be expressed in cartilage and demonstrate potent ‘aggrecanase’ activity (114). The ADAMTS family now consists of more than 20 members with a variety of biologically relevant proteolytic actions from procollagen N -proteinase activity (ADAMTS2, 3 and 14), to the cleavage and activation of vonWillebrand factor (ADAMTS13) in a blood clotting mechanism, to eﬀective aggrecanase (hyalectanase) activity (ADAMTS1, 4, 5, 8, 9, 15) (for reviews, see (115, 116)). TIMP-3 and α2 -macroglobulin are endogenous inhibitors of ADAMTS4 activity (117–120). The ADAMTS proteins consist of an N-terminal signal peptide for entering the secretory path, followed by a proprotease domain, and metalloprotease and disintegrin domains similar to the related ADAMs family of proteases and adhesion proteins. The remainder of the ‘ancillary’ domain, consists of a thrombospondin type I repeat, a cysteine-rich region, a spacer domain, and in many of the ADAMTSs, several, additional thrombospondin type I repeats at the C-terminus. Although classical MMPs are activated upon disruption of the interaction between an

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

101

unpaired cysteine in the pro-protease region, and a Zn++ molecule bound in the catalytic domain, and although there is an unpaired cysteine residue in the pro-protease of the ADAMTSs, there is no evidence that this mechanism maintains the zymogen form in ADAMTSs. However, good evidence indicates that these proteases are partially activated in the Golgi network of the secretory pathway by proteolytic removal of the pro-domain via cleavage at the pro-proteinase convertase consensus sequence (RXX(K)R) by a furin or furin-like protease (121). ADAMTS4 is furin-cleaved from a 100 kD pro-protease to a 68 kD N-terminal product. Catalytic activity of ADAMTS4 is further regulated by proteolytic or autocatalytic sequential truncation of the C-terminus to yield products of 53 kD and 40 kD (122), and although aﬃnity for GAG-containing aggrecan may be reduced in these shorter forms (122, 123), it appears that only the shorter forms can generate the G1-domain products (123, 124). Aggrecan, brevican, versican V0, V1 and V2 are all cleaved at speciﬁc sites by human recombinant ADAMTSs (as glutamyl endopeptidases, see Fig. 3) to form G1 domain products that are 40–66 kD and that are found in vivo. Proteolysis of hyalectans by the ADAMTSs does not entirely preclude cleavage by MMPs in vivo as aggrecan (125), versican (107) and brevican (126) all are readily cleaved in vitro by MMPs. Cleavage site predictions based on sequence alignments (109) suggest that human neurocan (127) contains a typical aggrecanase site at Glu505-Ala506, consistent with the sizes of the N-terminal and C-terminal fragments that have been identiﬁed (128), however evidence for cleavage at this site in vivo is lacking. In addition, using pooled sequences of peptide library mixtures, neurocan was identiﬁed as a substrate for active MMP-2 (10). Using in situ hybridisation, ADAMTS4 mRNA was localised to pyramidal neurons of Ammon’s horn and dentate granule neurons in the rat hippocampus, in addition to several other telencephalic regions including limbic areas (4). ADAMTS1 transcript was also found in the same regions, but appeared to be signiﬁcantly less abundant. Treatment of rats with kainic acid selectively increased immunoreactivity for the ADAMTSselective, N-terminal fragment of brevican ending in ‘EAVESE’ (Fig. 3 see next section) in the outer molecular layer of the dentate gyrus as well as stimulating the abundance of ADAMTS4 and ADAMTS1 transcript in the same area (4). These results suggest that cleavage of brevican by the ADAMTSs may be involved in the neural plasticity, i.e. neurite outgrowth, targeting and synaptic plasticity, that occurs after a kainite-induced lesion (4). The ADAMTSs may play a role in plasticity in

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

102

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

the peripheral nervous system as well. After hypoglossal nerve injury, there was a rapid and marked up-regulation of mRNA for ADAMTS1 and it was suggested that this protease may be increased in reactive glial that respond to the injury (129). In addition, rat astrocytes cultured in vitro respond to the Alzheimer’s disease peptide, β-amyloid by increasing expression of ADAMTS4 mRNA (130). A preponderance of data has been generated showing that the ADAMTSs, especially ADAMTS4 cleavage of brevican, may clear a pathway in the matrix that allows for invasion of gliomas into normal brain parenchyma (111, 112, 131). This property of malignant gliomas is the most diﬃcult to manage clinically. Thus, although the evidence is not overwhelming at present, there certainly is reason to suspect that ADAMTS-cleavage of the hyalectans may be important during periods of remodelling and plasticity in the nervous system.

12. Methods of Identifying Metalloproteinase Activity In Vivo Nearly all previous studies that intended to examine the role of matrixdegrading proteinases in biological processes in the nervous system, have entailed the measurement of MMPs or plasminogen activators, or their endogenous inhibitors by: (1) mRNA transcript levels, (2) protein levels by Western blot, (3) the localisation of mRNA or protein by in situ hybridisation or immunohistochemistry, respectively, (4) measurement of proteases by substrate zymography on SDS-PAGE or in situ zymography, and/or (5) detection of the cleavage of a synthetic or other artiﬁcial substrate in the extracts of brain tissue. Data obtained with these methods often show marked changes in metalloendopeptidase levels dependent upon the model involved, and because of the bulk of data now available, suggest albeit indirectly, an important role for these proteases in the nervous system (2). Studies may make an association in time or location between the injury and the elevation of the protease or identify cell types that may (or may not) express these proteases in the nervous system in vivo. Because MMPs and ADAMTSs are latent and must undergo proteolytic cleavage themselves to be activated, and because their activity is regulated by endogenous TIMPs that are often expressed in tandem with the protease, identifying the presence and location of active protease is essential to learning about their physiological/pathophysiological function. Speciﬁc proteolytic cleavage sites (sequences) have been identiﬁed in matrix protein substrates for

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

103

MMP, plasminogen activator and/or ADAMTS cleavage (see Fig. 3 for ADAMTS-speciﬁc hyalectan cleavage sites). Thus, antisera may be raised against the novel C-terminus (or N-terminus) of the daughter fragments that are generated upon cleavage. These ‘neoepitope’ antisera are now common place to investigators, studying caspases, where proteolytic cleavage activates proteases, or those studying MMP or ADAMTS cleavage of aggrecan in cartilage. Such anti-peptide antisera exhibit a speciﬁcity for only the novel terminus of the cleavage fragment, and usually do not cross-react with the intact, parent matrix protein to any signiﬁcant extent. Using such antisera, the ‘daughter’ fragment(s) may be identiﬁed in regional brain extracts by molecular weight on Western blots (4, 102, 110). More importantly in vivo, since generation of this fragment can be assumed to result from cleavage by an active matrix-degrading protease, these fragments can also be localised to highly speciﬁc regional and cellular locations using neoepitope antibodies in immunohistochemistry. For example, when an increase in the gene or protein expression of a speciﬁc metalloproteinase is observed in a region where an elevation of a cleaved matrix fragment is co-localised, it is reasonable to conclude that there is active, and not latent, protease present. The power of localisation of these cleaved neoepitope fragments by immunohistochemistry, resides in the fact that an area of active proteolytic activity can be unequivocally identiﬁed (4). There are of course, also limitations to the interpretation of this technique. Firstly, the abundance of the fragment in an extract or on IHC does not give a direct measure of the amount of proteolysis involved since that will be determined by the half-life of the fragment in the tissue. Secondly, the absence of the fragment on Western analysis or on IHC cannot be interpreted as a complete absence of the proteolytic process, because some neo-epitope antibodies only detect certain glycovariants of the protein due to masking phenomena. Thirdly, the speciﬁc metalloproteinase or proteinase family responsible for the cleavage cannot always be identiﬁed unequivocally, e.g. many ADAMTS family members share speciﬁcity and may also cleave at previously described MMP-speciﬁc cleavage sites (132). Fourthly, the presence of a speciﬁc fragment does not imply that this is the only cleavage site in this substrate for this protease or that this is the major substrate for this protease in that location. However, given these limitations, neoepitope antisera can establish the presence of an active proteinase in situ, and therefore, they are a unique tool to investigate localised proteolytic degradation. In this regard, their application to the study of ECM proteinases in the CNS is still in its infancy.

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

104

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

13. Tenascin-R and Tenascin-C Among the four tenascin genes identiﬁed, tenascin-R, and to a lesser extent, tenascin-C, appear to play important roles in the nervous system, both during development and in the adult. Neurogenesis and neuronal migration (133, 134), axonal guidance and growth (135), and myelination (136) are each impacted by the presence of tenascin-R. Tenascin-R expression appears to be restricted to the CNS, where it is found in the neuropil as a component of perineuronal nets in a purported complex with the hyalectans and hyaluronic acid (3, 90). Mice deﬁcient in tenascin-R show reduced levels of hyalectan-components of perineuronal nets and abnormal staining of these nets with Wisteria ﬂuoribunda, although the ultrastructure of synapses were observed to be similar to that of wild-type animals (90). Tenascin-C is highly expressed in the developing nervous system and to a lesser extent in the adult, and it is expressed in both neural and non-neural tissues. Each of the tenascins interact with a host of molecules on the cell surface and in the extracellular milieu to exert their actions, for review see (137, 138).

14. Structure In rotary-shadowed, electron microscopic images, tenascin-C is seen as a highly symmetrical, hexameric structure known as a hexabrachion (139). Six polypeptide tenascin C chains are linked via their amino termini, the so-called tenascin oligomerisation domain. Individually, this oligomerisation domain is attached to linear structures consisting of EGF-like repeats followed by multiple ﬁbronectin type III domains emanating out from this central core. The number of these EGF-like and ﬁbronectin repeats vary widely among the four tenascin proteins. Each linear molecule ends with a globular domain that resembles the chains of ﬁbrinogen. Tenascin-R (and tenascin-X and tenascin-W) contains the same domains in the same arrangement as tenascin-C including conserved cysteines that form disulﬁde bonds which stabilise the multivalent structure. However, tenascin-R has been isolated as a trimeric structure and not a hexamer.

15. Expression, Localisation and Molecular/Cellular Interactions Tenascin-R is highly expressed in the CNS rather late in development, mainly by oligodendrocytes (and a small set of neurons) where its synthesis

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

105

correlates with the period of active myelination. It is found in high concentrations at nodes of Ranvier, and it is thought to bind to the β2 subunit of Na+ channels (140) and target the channel to the nodes. Mice deﬁcient in tenascin-R exhibit decreased conduction velocity in neurons of the CNS (141). In addition to nodes of Ranvier, tenascin-R is localised to contacts between unmyelinated axons, between axons and processes of actively myelinating oligodendrocytes and between myelin sheaths (136). Along with conduction defects, tenascin-R knockout mice show reduced perisomatic inhibition in hippocampus resulting in impaired NMDA-receptor-dependent, long-term potentiation (142). In the adult, tenascin-R is localised to perineuronal nets of parvalbumin-immunoreactive, inhibitory interneurons in several regions of the brain including the cerebellum, spinal cord and hippocampus (143) and expression is up-regulated after the focal brain injury (144). There are clear behavioural deﬁcits in tenascin-R deﬁcient mice, especially in open-ﬁeld and other anxiety measures (Freitag, 2003 #130). Most recently, tenascin-R was found to be involved in the activity-dependent radial migration of neuroblasts in the core of the olfactory bulb during development (134). Expression of tenascin-C is high during embryonic development and in the adult, and regions that express tenascin-C are in or adjacent to areas with active, ongoing neurogenesis, the hippocampus, the borders of the subventricular zone and the rostral migratory stream (145). Tenascin-C production is also up-regulated after injury (146). Tenascin-C has various eﬀects on neurite outgrowth and the growth of processes in other neural cell types in vitro and in vivo depending on conditions (135, 147) and is involved in hippocampal synaptic plasticity (148, 149). Signals important for enhancing neurite outgrowth induced by tenascin-C may be mediated by α7β1 integrin (150). Both tenascin-R and tenascin-C interact with a host of proteins and carbohydrates that mainly bind to the ﬁbronectin type III repeat modules and the EGF-like domain and these activities may be adhesive or counteradhesive. Cell surface ‘receptors’ for the tenascins include integrins, cell surface molecule belonging to the immunoglobulin superfamily including N-CAM, F11/contactin and axonin-1, and extracellular interacting proteins such as the hyalectans, ﬁbronectin, and heparin and heparan sulfate (for review of these activities, see (138)). Several ligands bind the ﬁbronectin type III repeats 3–5 including neurocan (74), heparin that may bind a KEDK motif in this region (151), and ﬁbronectin whose cryptic bindingsite in this region may be revealed by proteolysis (152). The carbohydrate of tenascin-R may interact with ﬁbronectin and tenascin-C (153). The

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

106

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

alternatively spliced variants of the tenascins that may or may not contain binding sites add another layer of complexity to the host of tenascin‘receptor’ interactions.

16. Proteolytic Processing of Tenascin Because of the crucial role that tenascin is thought to play in plasticity, both within and outside the nervous system, regulation of its activity by proteolytic cleavage of the molecule would not be surprising. Indeed, the tenascins are substrates for many members of the MMP family. Tenascin puriﬁed from a human melanoma cell line was cleaved by MMP-1, -3 and -7, but not MMP-2 (154) whereas there were diﬀerent susceptibilities of small and large tenascin-C to cleavage by the MMPs. MMP-2, -3, -7 and -13 degrade large tenascin-C isoforms where cleavage occurs in the ﬁbronectin repeat region, however, only MMP-7 was able to cleave the small isoform by removing the N-terminal oligomerisation region (155, 156). More work is needed to apply neo-epitope antibody approaches to this interesting area. Many studies outside of the nervous system have observed a co-regulation of the expression of tenascin-C and the MMPs, i.e. when MMP levels are induced, tenascin-C expression is increased in tandem. This suggests that certain conditions provide a modiﬁed environment favourable to cell migration and tissue plasticity (138).

17. Miscellaneous Substrates in Neuropil There certainly are additional substrates for the MMPs and ADAMs in the CNS that may not be as easily categorised as those described above, and for a complete review, see (157, 158). What follows is a selective listing of a number of MMP and other metalloendopeptidase substrates that may be important for actions in the CNS. This list is not meant to be exhaustive, but the proteins were chosen because of their potential relevance to CNS physiology and disease.

18. Extracellular Matrix Proteins 18.1. Dystroglycan This protein is usually associated with the basal lamina. However, it is found in adult mouse brain neuropil. Proteolytic cleavage of the dystroglycan precursor results in a mature protein that contains two subunits,

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

107

α-dystroglycan that binds major basal lamina components, laminin, perlecan, and β-dystroglycan which is a transmembrane protein that links the α subunit to the epithelial basement membrane and other cell types that express dystroglycan. Interestingly, CRE/loxP mediated conditional inactivation of the dystroglycan gene using Cre-recombinase under the control of the CNS-speciﬁc GFAP promoter eﬀectively depleted the brain parenchyma of dystroglycan, and recapitulated some of the structural and functional brain abnormalities seen in congenital muscular dystrophies (159). Similarly, MMP activity disrupts the β-dystroglycan link with the ECM and it has been suggested that MMPs may play a role in the pathogenesis of hereditary neuromuscular diseases. 18.2. Laminins Laminins are not usually associated with the neuropil in the adult. However, there may be expression of an α2 subunit containing isoform in the rostral migratory stream of glial cells and their ﬁne processes (160). In addition, the integrin laminin ‘receptor’ α6β1 is highly expressed during neuroprogenitor migration (161). There is an increase in laminin production after injury in the CNS (162); a series of studies have localised lamininimmunoreactive pyramidal neurons in the CA region of the hippocampus, and loss of laminin immunoreactivity accompanied excitotoxin-induced neuronal death. Blockade of injury-induced proteolysis of laminin appeared to rescue these neurons from the eﬀects of the excitotoxin (163). 19. Cytokines and Growth Factors 19.1. Chemokines One of the more interesting ﬁndings in the MMP ﬁeld in recent years is the discovery that chemokines may be cleaved by the MMPs, a process which may either inactivate the agonist ligand or even convert the ligand from an agonist to an antagonist for its respective receptor. This phenomenon is covered elsewhere in this volume (see Chapter 5). 19.2. Nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF) Several growth factors may be activated by proteolysis, converting an inactive form into a mature, biologically active factor. Interestingly, it was shown that two neurotrophins, NGF and BDNF are secreted as pro-neurotrophic factors

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

108

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

and proteolytic cleavage by MMP-3, MMP-7 or plasmin cleaves them into active molecules that support growth and survival. Even more intriguing was the report that the pro-neurotrophins may activate the p75 receptor and stimulate apoptosis rather than survival and growth (164). 19.3. Transforming growth factor (TGF) β The original observation of MMPs inﬂuencing TGFβ activity was made outside the nervous system in a keratinocytes or a mammary carcinoma cell line showing that the hyaluronic acid receptor CD44 binds to proteolytically active MMP-9 and localises it to the cell surface. Here, MMP-9 (or MMP-2) can cleave latent TGFβ, activating it where it may participate in angiogenesis, tumour growth and/or invasion (165). There is good reason to believe that a similar mechanism may be at work in gliomas, which often express high levels of CD44 (166).

20. Substrates for Sheddase Activities 20.1. β-amyloid precursor protein (APP) Amyloidogenic processing of APP is carried out by proteases that cleave the precursor at each end of the peptide, β-secretase at the N-terminus of β amyloid and γ secretase near the C-terminal end. However, cleavage by the so-called α secretase activity (cleavage at Lys16 –Leu17 ) precludes the formation of β-amyloid, and this cleavage is much more frequent than processing via the amyloidogenic pathway. Although the β and possibly the γ secretases have been identiﬁed, the identity of α secretase has been elusive. However, there is good evidence that the majority of α secretase activity occurs directly or indirectly via a metalloproteinase (167). In addition, there is a growing notion that several of the proteolytically active members of the ADAM family, ADAM9, ADAM10 and ADAM17 may be responsible for the preponderance of α secretase activity (168, 169). Although individually, their expression in the CNS, does not seem to account for the abundant APP processing that occurs, collectively the proteases may explain a signiﬁcant proportion. Other chapters in this text discuss various developmentally important substrates of the MMPs, and ADAMs such as cytokine and growth factor receptors, integrins, ephrins and Ephs, and Notch.

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

109

20.2. Syndecan Syndecans are transmembrane heparan sulfate proteoglycans that bind ECM proteins and cytoskeletal proteins with their C-terminal domain and thus may mediate transduction of signals from the ECM to the cell. It has been reported that extracellular domains of the syndecans may be released from cell surface by membrane shedding (170, 171), without any loss of binding aﬃnity for their ECM ligands. Syndecan-2 and syndecan-4 are expressed by cells of the CNS and have critical roles in synaptogenesis and other actions during development (172). Interestingly, MMP-7 can induce the shedding of syndecan-1 in lung epithelia, and together with a CXC chemokine, KC, induces transepithelial migration of neutrophils during inﬂammation. Migration to the site of injury was lost in MMP-7 null mice or in KC null mice, suggesting that both molecules are required in inﬂammatory responses (173).

20.3. NG2 NG2 is a large, membrane spanning CSPG with a short cytoplasmic region and a large extracellular domain (more than 2,200 aa residues) that is highly expressed in developing and adult CNS (174). Although it appears to be a marker for oligodendrocyte precursors in post-natal development, in adult CNS, NG2 expression is maintained in a proliferating, potentially novel cell type that has been termed synantocyte (175) or polydendrocyte (176). The cell type also becomes reactive after injury, but does not express any of the classical astrocyte markers (177). A defect in re-myelination after injury in MMP-9 deﬁcient mice was attributed to the inability of these mice to clear the accumulation of NG2 after injury (178). In addition, in vitro incubation of NG2 and MMP-9 produced NG2 immunoreactive fragments on Western blot.

20.4. Myelin basic protein This abundant protein expressed by mature oligodendrocytes and deposited in myelin is cleaved by a number of MMPs (179) and ADAMs (180). This has led to the speculation that cleavage may induce an autoimmune response against the remnant neoepitopes (181, 182).

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

110

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

21. Peptides 21.1. β-amyloid Although neprilysin and insulin-degrading enzyme may be more eﬀective, potent and widely expressed in the CNS compared to the MMPs (183), MMPs do have the potential to degrade β-amyloid peptide in vitro (184, 185). 21.2. Substance P Substance P and other peptides may be degraded by the gelatinases, MMP-2 and MMP-9, as well (186).

22. Summary Not only have the number of substrates for the MMPs increased markedly in the last few years, but the diversity of extracellular proteins upon which they may act has been increased. It is a big leap forward that the yeasttwo hybrid system has worked for proteins that act outside of cells (11). The importance of exosites as identiﬁed by Overall and colleagues may lead to exponential growth of substrates for these proteases. However, as with any method, the ﬁnal importance of cleavage remains whether it occurs in physiological or pathophysiological environments in vivo. In addition, the proteolytic degradation of standard matrix proteins that exist in the neuropil around cells on the nervous system, is in its infancy since a number of the proteases that degrade these proteins have been identiﬁed on recently. Deﬁning novel actions of proteolytic products and their role in plasticity in the nervous system should be the basis for explosive growth in the future.

References 1. Lo, E. H., Wang, X., and Cuzner, M. L. (2002) J Neurosci Res 69, 1–9 2. Yong, V. W., Power, C., Forsyth, P., and Edwards, D. R. (2001) Nat Rev Neurosci 2, 502–511 3. Yamaguchi, Y. (2000) Cell Mol Life Sci 57, 276–289 4. Yuan, W., Matthews, R. T., Sandy, J., and Gottschall, P. E. (2002) Neuroscience 114, 1091–1101 5. Fowlkes, J. L., and Winkler, M. K. (2002) Cytokine Growth Factor Rev 13, 277–287 6. McQuibban, G. A., Gong, J.-H., Tam, E. M., McCulloch, C. A. G., Clark-Lewis, I., and Overall, C. M. (2000) Science 289, 1202–1205

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

111

7. Visse, R., and Nagase, H. (2003) Circ Res 92, 827–839 8. Overall, C. M. (2001) Methods Mol Biol 151, 79–120 9. Kridel, S. J., Chen, E., Kotra, L. P., Howeard, E. W., Mobashery, S., and Smith, J. W. (2001) J Biol Chem 276, 20572–20578 10. Turk, B. E., Huang, L. L., Piro, E. T., and Cantley, L. C. (2001) Nat Biotechnol 19, 661–667 11. Overall, C. M., McQuibban, G. A., and Clark-Lewis, I. (2002) Biol Chem 383, 1059–1066 12. Tam, E. M., Morrison, C. J., Wu, Y. I., Stack, M. S., and Overall, C. M. (2004) Proc Natl Acad Sci USA 101, 6917–6922 13. Farquhar, M. G. (1991) in Cell Biology of the Extracellular Matrix (Hay, E. H., ed), 2nd Ed., pp. 365–418, Plenum Press, New York 14. Gumbiner, B. M. (1996) Cell 84, 345–357 15. Nag, S. (2003) in The Blood-Brain-Barrier: Biology and Research Protocols (Nag, S., ed) Vol. 89, pp. 3–36, Humana Press, Totowa, NJ 16. Yurchenco, P. D., Amenta, P. S., and Patton, B. L. (2004) Matrix Biol 22, 521–538 17. Sasaki, T., Fassler, R., and Hohenester, E. (2004) J Cell Biol 164, 959–963 18. Timpl, R., Wiedemann, H., van Delden, V., Furthmayr, H., and Kuhn, K. (1981) Eur J Biochem 120, 203–211 19. Timpl, R., Rohde, H., Risteli, L., Ott, U., Robey, P. G., and Martin, G. R. (1982) Methods Enzymol 82, 831–838 20. Noonan, D. M., Fulle, A., Valente, P., Cai, S., Horigan, E., Sasaki, M., Yamada, Y., and Hassell, J. R. (1991) J Biol Chem 266, 22939–22947 21. Carlin, B., Jaﬀe, R., Bender, B., and Chung, A. E. (1981) J Biol Chem 256, 5209–5214 22. Yamada, K. M. (1991) in Cell Biology of the Extracellular Matrix (Hay, E. H., ed), 2nd Ed., pp. 111–146, Plenum Press, New York 23. Lane, T. F., and Sage, E. H. (1994) FASEB J 8, 163–173 24. Risau, W. (1991) Ann NY Acad Sci 633, 405–419 25. Abbott, N. J. (2002) J Anat 200, 629–638 26. Risau, W., and Wolburg, H. (1990) Trends Neurosci 13, 174–178 27. Callahan, M. K., and Ransohoﬀ, R. M. (2004) Curr Allergy Asthma Rep 4, 65–73 28. del Zoppo, G. J., and Mabuchi, T. (2003) J Cereb Blood Flow Metab 23, 879–894 29. Fryer, J. D., Taylor, J. W., DeMattos, R. B., Bales, K. R., Paul, S. M., Parsadanian, M., and Holtzman, D. M. (2003) J Neurosci Res 23, 7889–7896 30. Lee, J.-M., Yin, K., Hsin, I., Chen, S., Fryer, J. D., Holtzman, D. M., Hsu, C. Y., and Xu, J. (2003) Ann Neurol 54, 379–382 31. Muellner, A., Benz, M., Kloss, C. U., Mautes, A., Burggraf, D., and Hamann, G. F. (2003) J Neurotrauma 20, 745–754 32. Hudson, B. G., Tryggvason, K., Sundaramoorthy, M., and Neilson, E. G. (2003) N Engl J Med 348, 2543–2456

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

112

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

33. Eble, J., Ries, A., Lichy, A., Mann, K., Stanton, H., Gavrilovic, J., Murphy, G., and Kuhn, K. (1996) J Biol Chem 271, 30964–30970 34. Hamann, G. F., Okada, Y., Fitridge, R., and del Zoppo, G. J. (1995) Stroke 26, 2120–2126 35. Aumailley, M., and Gayraud, B. (1998) J Mol Med 76, 253–265 36. Wagner, S., Tagaya, M., Koziol, J. A., Quaranta, V., and del Zoppo, G. J. (1997) Stroke 28, 858–865 37. d’Ortho, M. P., Will, H., Atkinson, S., Butler, G., Messent, A., Gavrilovic, J., Smith, B., Timpl, R., Zardi, L., and Murphy, G. (1997) Eur J Biochem 250, 751–757 38. Shimada, T., Nakamura, H., Ohuchi, E., Fujii, Y., Murakami, Y., Sato, H., Seiki, M., and Okada, Y. (1999) Eur J Biochem 262, 907–914 39. Hassell, J. R., Robey, P. G., Barrach, H. J., Wilczek, J., Rennard, S. I., and Martin, G. R. (1980) Proc Natl Acad Sci 77, 4494–4498 40. Hassell, J. R., Y., Y., and Arikawa-Hirasawa, E. (2003) Glyconjug J 19, 263–267 41. Gohring, W., Sasaki, T., Heldin, C., and Timpl, R. (1998) Eur J Biochem 255, 60–66 42. Mongiat, M., Taylor, K., Otto, J., Aho, S., Uitto, J., Whitelock, J. M., and Iozzo, R. V. (2000) J Biol Chem 275, 7095–8000 43. Whitelock, J. M., Murdoch, A. D., Iozzo, R. V., and Underwood, P. A. (1996) J Biol Chem 271, 10079–10086 44. Wyss-Coray, T., Lin, C., Sanan, D. A., Mucke, L., and Masliah, E. (2000) Am J Pathol 156, 139–150 45. van Horssen, J., Kleinnijenhuis, J., Maass, C. N., Rensink, A. A., Otte-Holler, I., David, G., van den Heuvel, L. P., Wesseling, P., de Waal, R. M., and Verbeek, M. M. (2002) Neurobiol Aging 23, 537–545 46. Verbeek, M. M., Otte-Holler, I., van den Born, J., van den Heuvel, L. P., David, G., Wesseling, P., and de Waal, R. M. (1999) Am J Pathol 155, 2115–2125 47. Lin, H. C., Chang, J. H., Jain, S., Gabison, E. E., Kure, T., Kato, T., Fukai, N., and Azar, D. T. (2001) Invest Ophthalmol Vis Sci 42, 2517–2524 48. Sole, S., Petegnief, V., Gorina, R., Chamorro, A., and Planas, A. M. (2004) J Neuropathol Exp Neurol 63, 338–349 49. Timpl, R., and Brown, J. C. (1996) Bioessays 18, 123–132 50. Poschl, E., Mayer, U., Stetefeld, J., Baumgartner, R., Holak, T. A., Huber, R., and Timpl, R. (1996) EMBO J 15, 5154–5159 51. Aumailley, M., Wiedemann, H., Mann, K., and Timpl, R. (1989) Eur J Biochem 184, 241–248 52. Battaglia, C., Mayer, U., Aumailley, M., and Timpl, R. (1992) Eur J Biochem 208, 359–366 53. Dong, L., Chen, Y., Lewis, M., Hsieh, J. C., Reing, J., Chaillet, J. R., Howell, C. Y., Melhem, M., Inoue, S., Kuszak, J. R., DeGeest, K., and Chung, A. E. (2002) Lab Invest 82, 1617–1630 54. Celio, M. R., Spreaﬁco, R., DeBiasi, S., and Vitellaro-Zuccarello, L. (1998) Trends Neurosci 21, 510–515

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75.

76. 77.

78. 79. 80.

113

Nicholson, C., and Sykova, E. (1998) Trends Neurosci 21, 207–215 Dityatev, A., and Schachner, M. (2003) Nat Rev Neurosci 4, 456–468 Kleene, R., and Schachner, M. (2004) Nat Rev Neurosci 5, 195–208 Bandtlow, C. E., and Zimmermann, D. R. (2000) Physiol Rev 80, 1267–1290 Rhodes, K. E., and Fawcett, J. (2004) J Anat 204, 33–48 Silver, J., and Miller, J. H. (2004) Nat Rev Neurosci 5, 146–156 Bruckner, G., Grosche, J., Hartlage-Rubasamen, Schmidt, S., and Schachner, M. (2003) J Chem Neuroanat 26, 37–50. Rosen, M., Edfors-Lilja, I., and Bjornsson, S. (2002) Anal Biochem 308, 210–222 Doege, K., Sasaki, M., Horigan, E., Hassell, J. R., and Yamada, Y. (1987) J Biol Chem 262, 17757–17767 Zimmermann, D. R., and Ruoslahti, E. (1989) EMBO J 8, 2975–2981 Rauch, U., Karthikeyan, L., Maurel, P., Margolis, R. U., and Margolis, R. K. (1992) J Biol Chem 267, 19536–19547 Yamada, H., Watanabe, K., Shimonaka, M., and Yamaguchi, Y. (1994) J Biol Chem 269, 10119–10126 Jaworski, D. M., Kelly, G. M., and Hockﬁeld, S. (1994) J Cell Biol 125, 495–509 Shinomura, T., Nishida, Y., Ito, K., and Kimata, K. (1993) J Biol Chem 268, 14461–14469 Dours-Zimmermann, M. T., and Zimmermann, D. R. (1994) J Biol Chem 269, 32992–32998 Zako, M., Shinomura, T., Ujita, M., Ito, K., and Kimata, K. (1995) J Biol Chem 270, 3914–3918 Seidenbecher, C., Richter, K., Rauch, U., Bassler, R., Garner, C. C., and Gundelﬁnger, E. D. (1995) J Biol Chem 270, 27206–27212 Doege, K. J., Sasaki, M., Kimura, T., and Yamada, Y. (1991) J Biol Chem 266, 894–902 Brissett, N. C., and Perkins, S. J. (1998) Biochem J 329, 415–424 Rauch, U., Clement, A., Retzler, C., Frohlich, L., Fassler, R., Gohring, W., and Faissner, A. (1997) J Biol Chem 272, 26905–26912 Aspberg, A., Miura, R., Bourdoulous, S., Shimonaka, M., Heinegard, D., Schachner, M., Ruoslahti, E., and Yamaguchi, Y. (1997) Proc Natl Acad Sci USA 94, 10116–10121 Lundell, A., Olin, A. I., Morgelin, M., Al-Karadaghi, S., Aspberg, A., and Logan, D. T. (2004) Structure (Camb) 12, 1495–1506 Milev, P., Maurel, P., Chiba, A., Mevissen, M., Popp, S., Yamaguchi, Y., Margolis, R. K., and Margolis, R. U. (1998) Biochem Biophys Res Comm 247, 207–2121 Grumet, M., Friedlander, D. R., and Sakurai, T. (1996) Perspect Dev Neurobiol 3, 319–330 Landolt, R. M., Vaughan, L., Winterhalter, K. H., and Zimmermann, D. R. (1995) Development 121, 2303–2312 Snow, D. M., Steindler, D. A., and Silver, J. (1990) Dev Biol 138, 359–376

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

114

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

81. Ogawa, T., Hagihara, K., Suzuki, M., and Yamaguchi, Y. (2001) J Comp Neurol 432, 285–295 82. Hagihara, K., Miura, R., Kasaki, R., Berglund, E., Ranscht, B., and Yamaguchi, Y. (1999) J Comp Neurol 410, 256–264 83. Jaworski, D. M., Kelly, G. M., and Hockﬁeld, S. (1999) Exp Neurol 157, 327–337 84. Seidenbecher, C. I., Smalla, K.-H., Fischer, N., Gundelﬁnger, E. D., and Kreutz, M. R. (2002) J Neurochem 83, 738–746 85. Schmalfeldt, M., Dours-Zimmermann, M. T., Winterhalter, K. H., and Zimmermann, D. R. (1998) J Biol Chem 273, 15758–15764 86. Schmalfeldt, M., Bandtlow, C. E., Dours-Zimmermann, M. T., Winterhalter, K. H., and Zimmermann, D. R. (2000) J Cell Sci 113 (Pt 5), 807–816 87. LeBaron, R. G. (1996) Perspect Dev Neurobiol 3, 261–271 88. Courel, M. N., Marret, S., Girard, N., Chauzy, Z., Olivier, A., Bertrand, P., Delpech, A., Laquerriere, A., Asou, H., and Delpech, B. (1998) J Neurocytol 27, 27–32 89. Asher, R. A., Morgenstern, D. A., Shearer, M. C., Adcock, K. H., Pesheva, P., and Fawcett, J. (2002) J Neurosci 22, 2225–2236 90. Bruckner, G., Grosche, J., Schmidt, S., Hartig, W., Margolis, R. U., Delpech, B., Seidenbecher, C. I., Czaniera, R., and Schachner, M. (2000) J Comp Neurol 428, 616–629 91. Matthews, R. T., Kelly, G. M., Zerillo, C. A., Gray, G., Tiemeyer, M., and Hockﬁeld, S. (2002) J Neurosci 22, 7536–7547 92. Perides, G., Erickson, H. P., Rahemtulla, F., and Bignami, A. (1993) Anat Embryol 188, 467–479 93. Bignami, A., and Asher, R. A. (1992) Int J Dev Neurosci 10, 45–57 94. Oohashi, T., Hirakawa, S., Bekku, Y., Rauch, U., Zimmermann, D. R., Su, W. D., Ohtsuka, A., Murakami, T., and Ninomiya, Y. (2002) Mol Cell Neurosci 19, 43–57 95. Bekku, Y., Su, W. D., Hirakawa, S., Fassler, R., Ohtsuka, A., Kang, J. S., Sanders, J., Murakami, T., Ninomiya, Y., and Oohashi, T. (2003) Mol Cell Neurosci 24, 148–159 96. Watanabe, H., Cheung, S. C., Itano, N., Kimata, K., and Yamada, Y. (1997) J Biol Chem 272, 28057–28065 97. Nieduszynski, I. A., Sheehan, J. K., Phelps, C. F., Hardingham, T. E., and Muir, H. (1980) Biochem J 185, 107–114 98. Rauch, U., Gao, P., Janetzko, A., Flaccu, A., Hilgenberg, L., Tekotte, H., Margolis, R. K., and Margolis, R. U. (1991) J Biol Chem 266, 14785–14801 99. LeBaron, R. G., Zimmermann, D. R., and Ruoslahti, E. (1992) J Biol Chem 267, 10003–10010 100. Perides, G., Lane, W. S., Andrews, D., Dahl, D., and Bignami, A. (1989) J Biol Chem 264, 5881–5887 101. Aspberg, A., Bindert, C., and Ruoslahti, E. (1995) Proc Natl Acad Sci 92, 10590–10594

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

115

102. Westling, J., Gottschall, P. E., Thompson, V. P., Cockburn, A., Perides, G., Zimmermann, D. R., and Sandy, J. D. (2004) Biochem J 377, 787–795 103. Sandy, J. D., Flannery, C. R., and Plaas, A. H. (1987) Biochem Biophys Acta 931, 255–261 104. Sandy, J. D., Flannery, C. R., Neame, P. J., and Lohmander, L. S. (1992) J Clin Invest 89, 1512–1516 105. Arner, E. C. (2002) Curr Opin Pharmacol 2, 322–329 106. Delpech, B., Delpech, A., Girard, N., Chauzy, C., and Laumonier, R. (1979) Br J Cancer 40, 123–133 107. Perides, G., Asher, R. A., Lark, M. W., Lane, W. S., Robinson, R. A., and Bignami, A. (1995) Biochemistry Journal 312, 377–384 108. Yamada, H., Watanabe, K., Shimonaka, M., Yamasaki, M., and Yamaguchi, Y. (1995) Biochem Biophys Res Comm 216, 957–963 109. Sandy, J. D., Westling, J., Kenagy, R. D., Iruela-Arispe, M. L., Verscharen, C., Rodriguez-Mazaneque, J. C., Zimmermann, D. R., Lemire, J. M., Fischer, J. W., Wight, T. N., and Clowes, A. W. (2001) J Biol Chem 276, 13372–13378 110. Lemons, M. L., Sandy, J. D., Anderson, D. K., and Howland, D. R. (2001) J Neurosci 21, 4772–4781 111. Matthews, R. T., Gary, S. C., Zerillo, C., Pratta, M., Solomon, K., Arner, E. C., and Hockﬁeld, S. (2000) J Biol Chem 275, 22695–22703 112. Zhang, H., Kelly, G. M., Zerillo, C., Jaworski, D. M., and Hockﬁeld, S. (1998) J Neurosci 18, 2370–2376 113. Kuno, K., Kanada, N., Nakashima, E., Fujiki, F., Ichimura, F., and Matsushima, K. (1997) J Biol Chem 272, 556–562 114. Tortorella, M. D., Burn, T. C., Pratta, M. A., Abbaszade, I., et al., (1999) Science 284, 1664–1666 115. Tang, B. L., and Hong, W. (1999) FEBS Lett 445, 223–225 116. Apte, S. S. (2004) Int J Biochem Cell Biol 36, 981–985 117. Kashiwagi, M., Tortorella, M., Nagase, H., and Brew, K. (2001) J Biol Chem 276, 12501–12504 118. Rodriguez-Manzaneque, J. C., Westling, J., Thai, S. N., Luque, A., Knauper, V., Murphy, G., Sandy, J. D., and Iruela-Arispe, M. L. (2002) Biochem Biophys Res Comm 293, 501–508 119. Gendron, C., Kashiwagi, M., Hughes, C., Caterson, B., and Nagase, H. (2003) FEBS Lett 555, 431–436 120. Tortorella, M. D., Arner, E. C., Hills, R., Easton, A., Korte-Sarfaty, J., Fok, K., Wittwer, A. J., Liu, R. Q., and Malfait, A. M. (2004) J Biol Chem 279, 17554–17561 121. Wang, P., Tortorella, M., England, K., Malfait, A. M., Thomas, G., Arner, E. C., and Pei, D. (2004) J Biol Chem 279, 15434–15440 122. Flannery, C. R., Zeng, W., Corcoran, C., Collins-Racie, L. A., Chockalingam, P. S., Hebert, T., Mackie, S. A., McDonagh, T., Crawford, T. K., Tomkinson, K. N., LaVallie, E. R., and Morris, E. A. (2003) J Biol Chem 277, 42775–42780

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

116

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

123. Gao, G., Westling, J., Thompson, V. P., Howell, T. D., Gottschall, P. E., and Sandy, J. D. (2002) J Biol Chem 277, 11034–11041 124. Kashiwagi, M., Enghild, J. J., Gendron, C., Hughes, C., Caterson, B., Itoh, Y., and Nagase, H. (2004) J Biol Chem 279, 10109–10119 125. Sandy, J. D., Neame, P. J., Boynton, R. E., and Flannery, C. R. (1991) J Biol Chem 266, 8683–8685 126. Nakamura, H., Fujii, Y., Inoki, I., Sugimoto, K., Tanzawa, K., Matsuki, H., Miura, R., Yamaguchi, Y., and Okada, Y. (2000) J Biol Chem 275, 38885– 38890 127. Prange, C. K., Pennacchio, L. A., Lieuallen, K., Fan, W., and Lennon, G. G. (1998) Gene 221, 199–205 128. Matsui, F., Nishizuka, M., Yasuda, Y., Aono, S., Watanabe, E., and Oohira, A. (1998) Brain Res 790, 45–51 129. Sasaki, M., Seo-Kiryu, S., Kato, R., Kita, S., and Kiyama, H. (2001) Mol Brain Res 89, 158–163 130. Satoh, K., Suzuki, N., and Yokota, H. (2000) Neurosci Lett 289, 177–180 131. Viapiano, M. S., Matthews, R. T., and Hockﬁeld, S. (2003) J Biol Chem 278, 33239–33247 132. Westling, J., Fosang, A. J., Last, K., Thompson, V. P., Tomkinson, K. N., Hebert, T., McDonagh, T., Collins-Racie, L. A., LaVallie, E. R., Morris, E. A., and Sandy, J. D. (2002) J Biol Chem 277, 16059–16066 133. Schachner, M., Taylor, J. W., Bartsch, U., and Pesheva, P. (1994) Perspet Dev Neurobiol 2, 33–41 134. Saghatelyan, A., de Chevigny, A., Schachner, M., and Lledo, P. M. (2004) Nature Neurosci 7, 347–356 135. Faissner, A. (1997) Cell Tissue Res 290, 331–341 136. Bartsch, U., Pescheva, P., Raﬀ, M., and Schachner, M. (1993) Glia 9, 57–69 137. Bartsch, U. (1996) Prog Neurobiol 49, 145–168 138. Jones, F. S., and Jones, P. L. (2000) Dev Dyn 218, 235–259 139. Erickson, H., and Iglesias, J. L. (1984) Nature 311, 267–269 140. Xiao, Z. C., Ragsdale, D. S., Malhotra, J. D., Mattei, L. N., Braun, P. E., Schachner, M., and Isom, L. L. (1999) J Biol Chem 274, 26511–26517 141. Weber, P., Bartsch, U., Rasband, M. N., Czaniera, R., Lang, Y., Bluethmann, H., Margolis, R. U., Levinson, S. R., Shrager, P., Montag, D., and Schachner, M. (1999) J Neurosci 19, 4245–4262 142. Saghatelyan, A. K., Dityatev, A., Schmidt, S., Schuster, T., Bartsch, U., and Schachner, M. (2001) Molecular and Cellular Neuroscience 17, 226–240 143. Fuss, B., Wintergerst, E. S., Bartsch, U., and Schachner, M. (1993) J Neurosci Res 29, 299–307 144. Brodkey, J. A., Laywell, E. D., O’Brien, T. F., Faissner, A., Stefansson, K., Dorries, H. U., Schachner, M., and Steindler, D. A. (1995) J Neurosurg 82, 106–112 145. Jankovski, A., and Sotelo, C. (1996) J Comp Neurol 371, 376–396 146. Crossin, K. L. (1996) J Cell Biochem 61, 592–598 147. Garcion, E., Faissner, A., and ﬀrench-Constant, C. (2001) Development 128, 2485–2496

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

Substrates for Metalloendopeptidases in the Central Nervous System

117

148. Nakic, M., Manahan-Vaughan, D., Reymann, K. G., and Schachner, M. (1998) J Neurobiol 37, 393–404 149. Evers, M. R., Salmen, B., Bukalo, O., Rollenhagen, A., Bosl, M. R., Morellini, F., Bartsch, U., Dityatev, A., and Schachner, M. (2002) J Neurosci 22, 7177–7194 150. Mercado, M. L. T., Nur-e-Kamal, A., Liu, H.-Y., Gross, S. R., Movahed, R., and Meiners, S. (2004) J Neurosci 24, 238–247 151. Jang, J. H., Hwang, J. H., Chung, C. P., and Choung, P. H. (2004) J Biol Chem (ahead of print) 152. Ingham, K. C., Brew, S. A., and Erickson, H. P. (2004) J Biol Chem (ahead of print) 153. Probmeister, R., Braunewell, K.-H., and Pesheva, P. (2000) Brain Res 863, 42–51 154. Imai, K., Kusakabe, M., Sakakura, T., Nakanishi, I., and Okada, Y. (1994) FEBS Lett 352, 216–218 155. Siri, A., Knauper, V., Veirana, N., Caocci, F., Murphy, G., and Zardi, L. (1995) J Biol Chem 270, 8650–8654 156. Knauper, V., Cowell, S., Smith, B., Lopez-Otin, C., O’Shea, M., Morris, H., Zardi, L., and Murphy, G. (1997) J Biol Chem 272, 7608–7616 157. Sternlicht, M. D., and Werb, Z. (2001) Ann Rev Cell Dev Biol 17, 463–516 158. Moss, M. L., and Lambert, M. H. (2002) Essays Biochem 2002, 141–153 159. Moore, S. A., Saito, F., Chen, J., Michele, D. E., Henry, M. D., Messing, A., Cohn, R. D., Ross-Barta, S. E., Westra, S., Williamson, R. A., Hoshi, T., and Campbell, K. P. (2002) Nature 418, 422–424 160. Hagg, T., Portera-Cailliau, C., Jucker, M., and Engvall, E. (1997) Brain Res 764, 17–27 161. Jacques, T. S., Relvas, J. B., Nishimura, S., Pytela, R., Edwards, G. M., Streuli, C. H., and ﬀrench-Constant, C. (1998) Development 125, 3167–3177 162. Hagg, T., Muir, D., Engvall, E., Varon, S., and Manthorpe, M. (1989) Neuron 3, 721–732 163. Chen, Z. L., Indyk, J. A., and Strickland, S. (2003) Molecular Biology of the Cell 14, 2665–2676 164. Lee, R., Kermani, P., Teng, K. K., and Hempstead, B. L. (2001) Science 294, 1945–1948 165. Yu, Q., and Stamenkovic, I. (2000) Genes Dev 14, 163–176 166. Ranuncolo, S. M., Ladeda, V., Specterman, S., Varela, M., Lastiri, J., Morandi, A., Matos, E., Bal de Kier Joﬀe, E., Puricelli, L., and Pallotta, M. G. (2002) J Surg Oncol 79, 30–35 167. Roberts, S. B., Ripellino, J. A., Ingalls, K. M., Robakis, N. K., and Felsenstein, K. M. (1994) J Biol Chem 269, 3111–3116 168. Asai, M., Hattori, C., Szabo, B., Sasagawa, N., Maruyama, K., Tanuma, S., and Ishiura, S. (2003) Biochem Biophys Res Comm 301, 231–235 169. Allinson, T. M. J., Parkin, E. T., Turner, A. J., and Hooper, N. M. (2003) J Neurosci Res 74, 342–352 170. Endo, K., Takino, T., Miyamori, H., Kinsen, H., Yoshizaki, T., Furukawa, M., and Sato, H. (2003) J Biol Chem 278, 40764–40770

May 26, 2005

16:2 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch04

118

P.E. Gottschall, J.D. Sandy & D.R. Zimmermann

171. Asundi, V. K., Erdman, R., Stahl, R. C., and Carey, D. J. (2003) J Neurosci Res 73, 593–602 172. Ethell, I. M., Irie, F., Kalo, M. S., Couchman, J. R., Pasquale, E. B., and Yamaguchi, Y. (2001) Neuron 31, 1001–1013 173. Li, Q., Park, P. W., Wilson, C. L., and Parks, W. C. (2002) Cell 111, 635–646 174. Nishiyama, A., Dahlin, K. J., Prince, J. T., Johnstone, S. R., and Stallcup, W. B. (1991) J Cell Biol 114, 359–371 175. Butt, A. M., Kiﬀ, J., Hubbard, P., and Berry, M. (2002) J Neurocytol 31, 551–565 176. Nishiyama, A., Watanabe, M., Yang, Z., and Bu, J. (2002) J Neurocytol 31, 437–455 177. Levine, J. M., Reynolds, R., and Fawcett, J. (2001) Trends Neurosci 24, 39–47 178. Larsen, P. H., Wells, J. E., Stallcup, W. B., Opdenakker, G., and Yong, V. W. (2003) J Neurosci 23, 11127–11135 179. Chandler, S., Coates, R., Gearing, A., Lury, J., Wells, G., and Bone, E. (1995) Neurosci Lett 201, 223–226 180. Amour, A., Knight, C. G., English, W. R., Webster, A., Slocombe, P. M., Knauper, V., Docherty, A. J., Becherer, J. D., Blobel, C. P., and Murphy, G. (2002) FEBS Lett 524, 154–158 181. Descamps, F. J., Van den Steen, P., Nelissen, I., Van Damme, J., and Opdenakker, G. (2003) Adv Exp Med Biol 535, 69–77 182. Opdenakker, G., Van den Steen, P. E., and Van Damme, J. (2001) Trends Immunol 22, 571–579 183. Carson, J. A., and Turner, A. J. (2002) J Neurochem 81, 1–8 184. Roher, A. E., Kasunic, T. C., Woods, A. S., Cotter, R. J., Ball, M. J., and Fridman, R. (1994) Biochem Biophys Res Comm 205, 1755–1761 185. Backstrom, J. R., Lim, G. P., Cullen, M. J., and Tokes, Z. A. (1996) J Neurosci 16, 7910–7919 186. Backstrom, J. R., and Tokes, Z. A. (1995) J Neurochem 64, 1312–1318

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

CHAPTER 5 EXAMPLES OF SIGNALLING BY MMPs

K. Conant Johns Hopkins University, Department of Neurology, Meyer 6-109, 600 North Wolfe Street, Baltimore, MD 21287 E-mail: [email protected]

Named for their ability to degrade proteins of the extracellular matrix (ECM), MMPs are being increasingly recognised as eﬀectors of intracellular signalling. Their ability to act as such is mediated through multiple mechanisms, including some that follow from the processing of ECM proteins, and others that are relatively independent of such. In this chapter, the broad subject of signalling by MMPs will be discussed. Topics to be covered will include signalling that follows from the cleavage of matrix proteins, soluble molecules, and cell surface receptors. Signalling that may result from enzyme activity independent eﬀects of MMPs will also be discussed.

1. Signalling Related to Proteolysis of Matrix and Matrix-Like Proteins Proteolysis of matrix may aﬀect intracellular signalling through changes in the occupancy and/or conformation of integrin receptors, through the generation of matrix-derived signalling molecules known as matrikines, or through the exposure/release of previously sequestered epitopes and soluble proteins. MMP directed proteolysis of matrix-like proteins, such as the heparin sulfated proteoglycan agrin or the laminin-like netrin may also aﬀect intracellular events. Each of these possibilities will be discussed in turn.

Correspondence to: K. Conant 119

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

120

K. Conant

1.1. Anoikis, or cell death by loss of anchorage, involves changes in intracellular signalling and may be stimulated by mechanisms including proteolysis of either matrix or cadherins Anchorage of cells to the extracellular matrix is typically mediated by integrins. These are transmembrane receptors composed of an α and a β chain. The various αβ heterodimers bind diﬀerent components of the ECM including ﬁbronectin, laminin, and interstitial collagen. There is some overlap in potential interactions, in that a given matrix component can associate with more than one integrin heterodimer and vice versa. The binding of matrix components to integrins aﬀects both their conformation and signalling properties. Not surprisingly, a disruption in such binding may do the same. Changes in integrin signalling that result from disrupted interactions with matrix may be linked to changes in cell mobility, and in some circumstances, to cell death. In fact cell death that follows from disrupted cell anchorage, including that mediated by integrins, is referred to by the term anoikis. With respect to the CNS and neurons in particular, several studies are consistent with a role for integrin/matrix interactions in cell survival. For example, plasmin-catalysed laminin degradation has been linked to an increase in excitatory amino acid induced death of hippocampal neurons (1), and both laminin and the integrin receptor binding ligand, IKLLI, can protect neurons from glutamate toxicity (2). In related studies, it has been shown that collagen-1 may be neuroprotective through an integrin binding eﬀect (3), and that chondroitin sulfate proteoglycans may protect neurons from glutamate mediated cell death (4). Integrin-matrix interactions are thought to inﬂuence cell survival through eﬀects on the activity of speciﬁc intracellular signalling molecules. Among these is focal adhesion kinase (FAK) and integrin-linked kinase (ILK). FAK associates with an array of integrins while ILK has a strong preference for the cytoplasmic domains of β1 and β3 (5–7). FAK or ILK may be activated upon integrin ligation, and either may ultimately stimulate an increase in the phosphorylation and activation of Akt/protein kinase B, a kinase that can promote cell survival through a variety of mechanisms (7–9). Recent work has also shown that changes in β1 -integrin conformation may eliminate also the activity of protein phosphatase 2A (PP2A) and thus more directly aﬀect the phosphorylation of Akt (10). In this study, it was shown that Akt was localised to PP2A containing complexes which were functionally linked to β1 , and it was proposed that

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

121

changes in the conformation of β1 , which could conceivably follow from loss of matrix, might de-repress PP2A and thus lead to decreased activity of Akt and cell death. The binding to, or cleavage of, matrix proteins by an MMP might directly alter matrix conformation so that, though integrin binding is maintained, integrin conformation is altered. Similarly, MMPs might generate matrix fragments that can in turn inﬂuence matrix conformation and thus integrin signalling. An example of the latter is provided by a study which showed that a 76 amino acid fragment of ﬁbronectin could cause disassembly of preformed ﬁbronectin matrix while not aﬀecting cell adhesion (11). Adherent cells are thought to assemble ﬁbronectin into a ﬁbrillar matrix on their apical surface. Integrin binding and then ﬁbronectin selfassembly are required. The 76 amino acid fragment, which forms one of the ﬁbronectin self-assembly sites, inhibited proliferation and migration of attached ﬁbroblasts or endothelial cells, and inhibited Rho dependent processes while activating Cdc42 (11). In addition, ACK (activated Cdc42binding kinase) and p38 MAPK (mitogen-activated protein kinase), two downstream eﬀectors of Cdc42, were activated, while PAK (p21-activated kinase) and JNK/SAPK (c-Jun NH2-terminal kinase/stress-activated protein kinase) were inhibited (11). While anchorage of cells to the extracellular matrix is typically mediated by integrins, anchorage between cells may be mediated by integrins or by other proteins including immunoglobulin superfamily members, lectins, proteoglycans and cadherins. Of these, the cadherins have perhaps been the best studied. In addition, it has been shown that cadherins can be targeted by MMPs. The cadherins represent a family of transmembrane glycoprotein adhesion molecules which allow for calcium dependent homotypic and heterotypic cell-cell interactions (12). Disruption of cadherin mediated cell-cell anchorage may, like disruption of integrin mediated cell-matrix anchorage, be linked to anoikis. For example, epithelial cells can overcome anoikis stimulated by reduced matrix contact if cell-cell contact is maintained. If E-cadherin binding is disrupted, however, anoikis will ensue (13–15). In addition, ﬁbroblasts grown at high density can form cell-cell contacts which provide some resistance to the anoikis that follows a disruption in cell-matrix interactions (16). The intracellular signalling cascades that mediate the survival promoting eﬀects of cadherin engagement are less well understood than are those thought to underlie the survival promoting eﬀects of integrin engagement. Akt may, however, again be involved (17). This kinase has been localised

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

122

K. Conant

to cell-cell contacts, and E-cadherin has been shown to promote survival in a manner dependent of PI3-K (13), an upstream eﬀector of Akt. Cadherins may aﬀect intracellular signalling cascades in that these receptors may regulate the location of β-catenin. In adherent non-stimulated cells, β-catenin is localised mostly at the cell membrane adhesion complex. Thus proteolytic shedding of cadherins may release catenin from the cytoplasmic cell adhesion complex and thus aﬀect catenin mediated gene transcription (12). This pathway, however, may also be inﬂuenced by intracellular glycogen synthase kinase-3β (GSK-3β) activity, which may be increased in association with reduced levels of Akt (18). GSK-3β phosphorylates β-catenin and thus marks it for proteosome-mediated degradation (19, 20). Recent work has also implicated Fas signalling in anoikis. Loss of anchorage may lead to an up-regulation in both Fas and Fas-Ligand (Fas-L) expression, and Fas/Fas-L interactions may in turn contribute to cell death. The Fas/Fas-L system will be discussed in more detail in a section to follow, as will be the role that speciﬁc MMPs may play in its regulation.

1.2. Proteolysis may also lead to the generation of matrix-derived fragments, or matrikines, that can stimulate changes in cell signalling Partial proteolysis of matrix components or other large proteins by MMPs may be associated with the generation of bioactive protein fragments known as matrikines. Of interest, a number of these fragments can in turn stimulate an increase in MMP expression, leading to a potential ampliﬁcation of tissue remodelling in wound healing, inﬂammation or other initiating events. For example, ﬁbronectin fragments, but not intact ﬁbronectin, may signal through the α5 β1 integrin to increase MMP-1 and -3 expression (21). A 110 kDa ﬁbronectin fragment can also bind to α5 β1 to stimulate MMP-13 production from chondrocytes (22). Similarly, the central 120 kDa segment of ﬁbronectin has an RGD sequence which may increase MMP expression. In contrast, a connecting segment-1 (CS-1) region-derived peptide of ﬁbronectin may actually suppress MMP expression through an interaction with α4 β1 (23). In studies related to other matrix components, it has been shown that a peptide from type IV collagen can interact with αv β3 to stimulate FAK and PI3-kinase phosphorylation (24), and that MMP-2 can act on laminin 5 to

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

123

produce a product that supports cell migration as opposed to adhesion. This eﬀect is thought to follow from the exposure of a neoepitope on laminin 5 that can stimulate signalling via its interaction with an unknown cell surface receptor (25). Other matrix fragments which have been particularly well studied and shown to be bioactive are angiostatin and endostatin (26). The ﬁrst, or angiostatin, is a 38 kDa plasminogen fragment initially described as an inhibitor of angiogenesis. A number of proteinases can convert plasminogen to angiostatin, including MMP-3, -7, -9 and -12 (27–30). That MMPs may be important to angiostatin generation in vivo is supported by at least one study, in which reduced plasma levels of MMP-9 were associated with increased tumour vascularisation and relatively low levels of circulating angiostatin (31). Angiostatin may inhibit bFGF-stimulated migration of endothelial cells (32), and in several studies, angiostatin has been shown to stimulate apoptosis of endothelial cells (33). The latter may be related in part to the ability of angiostatin to associate with integrin receptors on this cell type. Endostatin is a 20 kDa C-terminal fragment of collagen XVIII (34). It may be generated by proteinases including MMPs (35). Like angiostatin, endostatin can inhibit endothelial cell proliferation and stimulate apoptosis (34). Endostatin also interferes with endothelial cell migration (32). The ability of endostatin to interact with α5 β1 and αv β3 is likely to play a role in its biological eﬀects. It is tempting to speculate that both angiostatin and endostatin could play a role in CNS inﬂammation. The proteinases that can generate such fragments may be elevated as part of the inﬂammatory response. Newly generated angiostatin and/or endostatin might then cause death or dysfunction of blood brain barrier endothelial cells so that the CNS ingress of leukocytes or serum derived toxins would be facilitated. Matricellular proteins such as thrombospondins (TSPs) may also be processed through mechanisms including proteolysis (36). TSPs are large proteins which function in a variety of processes including wound healing and angiogenesis (37, 38). These proteins can aﬀect signalling through cell surface receptors including CD36 and CD47. In terms of the CNS in particular, TSP may inﬂuence synapse formation (39) and/or neuronal survival (40). While this protein can act on a receptor which has been linked to neuronal death (40), overall it may actually be neuroprotective. This latter eﬀect may relate to its ability to sequester and localise growth factors. In other studies, TSP has been shown to inﬂuence MMP expression and/or availability (38, 41).

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

124

K. Conant

1.3. Sequestered molecules or their active sites may be released or exposed by targeted proteolysis MMPs may, in some circumstances, release protein-bound molecules and thereby increase their signalling potential. Such a mechanism may be involved in the case of transforming growth factor-β (TGF-β), which may be bound to the proteoglycan decorin (42). MMP-2, -3 and -7 have been shown to release TGF-β from decorin and these MMPs may thus stimulate TGF signalling (42). Insulin like growth factor (IGF) and vascular endothelial growth factor (VEGF) signalling may also be modulated by MMPs. For example, it has been shown that MMP-1 can degrade IGF binding protein-3 (43), thus potentially enhancing IGF availability, and that MMP-9 may be required for the release of VEGF during long bone development (44). Proteinases may also play a role in modulating the bioavailability of ﬁbroblast growth factor-2 (FGF-2). For example, MMP-1 and -3 can degrade endothelial cell derived perlecan to release basic FGF (45). Alternatively, MMP-2 may decrease FGF bioavailability in that it can cleave the ectodomain fragment of the FGF receptor-1 (46), which may then bind to and thus sequester tissue FGF. In addition, as discussed in the section on matrikines, MMP-2 can act on laminin 5 to expose a site which in turn aﬀects intracellular signalling (25). Stromelysin 3 or MMP-11 has been shown to release growth factors from ECM that tumour cells need to grow (47), and it has recently been shown that MMPs can cleave Tenascin-C into fragments including one with an EGF-like domain that stimulates smooth muscle cell apoptosis (48).

1.4. Proteolysis of matrix-like proteins One matrix-like protein which may be a substrate for MMP mediated proteolysis is the heparin sulphated proteoglycan agrin (49, 50). Named for its ability to aggregate acetylcholine receptors, agrin is a trophic factor which may play a critical role in the assembly of the postsynaptic apparatus. Agrin deﬁcient mice do not form functional postsynaptic structures at the neuromuscular junction (51), and may die perinatally due to breathing failure. In adult agrin deﬁcient mice, constructed so that the neuromuscular junctional defect could be rescued, it has been noted that hippocampal synaptogenesis may be impaired (52). Agrin may also play a role in the formation of the immunological synapse (53). It is thought that agrin may positively aﬀect the recruitment

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

125

of lipid rafts, a process necessary for proper localisation of proteins at this synapse (53). Agrin may also aﬀect localisation of the aquaporin water channel, and thus degradation of agrin might lead to altered localisation of this receptor (54). Dystroglycan is a heparan sulfated proteoglycan that may, like agrin, be critical to the stability of the neuromuscular junction. It is often localised to complexes that contain agrin and possibly aquaporin as well. It may also serve as a critical receptor for laminin. Evidence suggests that β-dystroglycan may be cleaved by an MT-MMP, suggesting that MT-MMP activity may impair the cell’s ability to interact with laminin (55). Another matrix like molecule that plays a role in CNS physiology is netrin, a laminin like protein that is known to play a role in axon guidance. Netrin may, in certain circumstances, also aﬀect neuronal survival (56). Potential cleavage of netrin by matrix degrading proteinases has not been well examined, but it would not be unexpected. Evidence does, however, suggest that MMPs may be involved in processing one of the netrin receptors (57). This receptor, known as deleted in colorectal cancer or DCC, plays a role in netrin associated eﬀects on axon guidance. Note: potential processing of DCC by MMPs will be discussed in more detail in a later section (Sec. 3.4).

2. Signalling Related to Changes in the Structure or Bioavailability of Select Soluble Molecules Given the vast number of MMP substrates that have thus far been identiﬁed, it is tempting to think of these enzymes as somewhat non-speciﬁc or promiscuous, and even to question the biological relevance of some of the MMP/substrate interactions that can be made to occur in vitro. Nonetheless, MMP mediated proteolysis is often extremely targeted and purposeful. One area in which this can be especially well appreciated is in the ability of MMPs to process soluble molecules that inﬂuence immune eﬀector functions of the host or invading pathogen. This will be touched upon in the sections to follow, as will be the targeted proteolysis of soluble proteins that play a role in CNS development and function. 2.1. Cleavage of host proteins by pathogen derived MPs A number of fascinating examples of MPs as eﬀectors come from studies focused on parasites. For example, it has been shown that eotaxin, a potent

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

126

K. Conant

eosinophil chemoattractant that acts via CCR3 and is believed to have evolved to control infection with helminthic parasites, may be cleaved and thus inactivated by a hookworm derived metalloproteinase (58). It therefore seems that hookworms may evade the host immune response via metalloproteinase activity. As will be discussed in sections to follow, this ability may be shared by cancer cells, which also release metalloproteinases that may in some part act to interfere with host immunity/immune surveillance.

2.2. Host cell-derived MMPs also inﬂuence immunity The immune response may also be modulated by a number of host cell derived MMPs that are up-regulated in response to pro-inﬂammatory stimuli. For example, host-derived MMPs target a number of chemotactic cytokines including monocyte chemoattractant protein-3 (MCP-3) (59), stromal derived factor-1 (SDF-1) (60) and interleukin-8 (IL-8) (61). Cleavage of IL-8 by neutrophil gelatinase B can potentiate its activity ten fold and may thus exacerbate a brisk immune response (61). Alternatively, MMP-mediated cleavage of MCP-3 and SDF may lead to their inactivation, thus dampening a brisk response. Of note, MCP-3 and SDF are both released by astrocytes (62, 63), the most numerous cell type in the brain. Their inactivation by MMPs could be considered in some part protective, in that inﬂammation within the CNS is considered particularly harmful to the host.

2.3. MMPs may process soluble molecules that inﬂuence neuronal structure, survival or migration Soluble molecules targeted by MMPs include not only those which are involved in inﬂammation, but those involved in neuronal function. Of interest, however, there is some overlap between the two. For example, while the receptor for the chemotactic cytokine SDF-1 is expressed on leukocytes, it may also be expressed on neurons. Thus, mice that are lacking CXCR4 have not only haematopoietic defects, but altered cerebellar development (64). Moreover, cytokines such as TNF-α may inﬂuence activity of the neuronal synapse (65). In addition, a number of proteins which are known to inﬂuence neuronal survival also function as, or are closely related to, cytokines that are involved in the immune response. Interleukin-1 (IL-6), for example, can act in some situations as a neurotrophin (66, 67).

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

127

Classically, the neurotrophins are soluble molecules that can promote neuronal survival, diﬀerentiation, and in some cases, death. There is obvious overlap between neurotrophins and soluble molecules known to inﬂuence neuronal or neuronal process migration in that many proteins that inﬂuence cell or cell process migration can also inﬂuence cell survival, and both processes often involve signalling through Akt. A number of classic neurotrophins have been identiﬁed and well studied for their ability to promote neuronal survival through eﬀects on the Trk family of tyrosine kinases and/or the p75 neurotrophin receptor (p75NTR). A classic example is that of nerve growth factor (NGF), a trophin which acts on both p75NTR and TrkA. NGF is synthesised as a proform which can be processed intracellularly with subsequent release of the mature form. In some cases, however, the proform is released extracellularly where it can act either as a proform or instead be processed by extracellular enzymes. Such extracellular processing is thought to be mediated by plasmin and MMPs (68). Of note, while the proform of NGF tends to activate p75(NTR) and to minimally aﬀect TrkA, the truncated form is instead a more potent stimulus for TrkA. Thus, the extracellular processing of pro NGF by MMPs could favour TrkA activation and TrkA associated eﬀects such as the promotion of diﬀerentiation. Reduced processing might instead favour p75(NGF) associated eﬀects such as apoptosis (68, 69).

2.4. MMPs may process soluble molecules involved in pain Kinins, which are often studied for their involvement in pain and inﬂammation, are produced by the proteolytic processing of kininogen precursors which are typically synthesised in the liver and released into the circulation (70). Kininogen precursors may, however, also be produced locally, within speciﬁc extra hepatic tissues (71, 72). Moreover, kininogen production may be up-regulated by inﬂammatory stimuli including prostaglandins (71). Of additional interest, a number of proteinases may process kininogens into bioactive kinins (73, 74). These proteinases include circulating kallikreins as well as serine and metalloproteinases which can be released from activated macrophages (74). Thus, neuroactive kinins might be generated within a macrophage laden inﬂamed tissue, from precursors and proteinases that are produced locally and/or introduced via leaky blood-nerve or blood-brain barriers.

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

128

K. Conant

The vasoactive peptide endothelin-1 (ET-1) may also play a role in pain (75). This is a 21 amino acid peptide, synthesised by a number of cell types, that acts on receptors including ETA and ETB . These are G protein coupled receptors which often support antagonistic functions. For example, ETA has been linked to vasoconstriction while ETB may mediate vasodilation. The ability of ET-1 to trigger pain has been linked to its activation of ETA receptors localised to nociceptive cells. In fact pain caused by metastatic cancer may be diminished by an ETA antagonist. ETB receptor activation, however, has recently been linked to the production of β-endorphins which are known to alleviate pain. With respect to proteolytic processing, it has been shown that ET is produced from a precursor, named preproendothelin. After the removal of a signal peptide, the precursor is selectively processed by a furin-like peptidase to yield an inactive intermediate known as big-endothelin (big-ET). Big-ET is further converted into active ET by an endothelin-converting enzyme. Several isoforms of ECE, which is structurally similar to neutral endopeptidase, have been identiﬁed. However, while ET-1 is typically generated by ECE, MMP-2 can also cleave big ET-1 to produce a vasoactive peptide ET-1 (1–32)(76). This peptide activates ETA receptors and up-regulates the cell surface expression of CD11b/CD18 on neutrophils, with EC50 values of 1–3 nM. Up-regulation of CD11b/CD18 expression was found to be associated with the activation of extracellular signal-regulated kinase (Erk), which followed activation of Ras, Raf-1, and MEK (MAPK kinase). ET-1[1–32] also produced slight increases in the expression of ICAM-1 and E-selectin on endothelial cells, and markedly enhanced β2 integrin-dependent adhesion of neutrophils to activated coronary artery endothelial cells (76).

2.5. MMPs may process molecules involved in neurodegenerative disease A number of soluble molecules that are targeted by MPs may play a role in neurodegenerative disease. For example, MMPs may activate IL-1β (77), a cytokine that has implicated in the pathogenesis of Alzheimer’s disease (78). IL-1β has been linked to the activation of the NF-kappaB/Rel binding site in the regulatory region of the amyloid precursor protein gene (79), and more recently, this cytokine has been shown to increase NMDA receptor mediated intracellular calcium increase through activation of tyrosine kinases and subsequent NR2A/B subunit phosphorylation, eﬀects which may contribute to glutamate-mediated neurodegeneration (80). MMPs also

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

129

activate other MMPs, and may thereby contribute to MMP associated neurodegeneration (81). It has been shown for example that MMPs including MMP-1, -2, and -9 may be neurotoxic (82–84). In one of these studies, an MMP derived SDF-1 fragment was associated with neurodegeneration (82). It was shown that MMP-2, activated by neuronal MT1-MMP, could cleave SDF-1 and thus produce a molecule lacking in the N -terminal tetrapeptide. This fragment, SDF(5−67) was then shown to be neurotoxic through a mechanism that involved signalling through Gi proteins (82). It was proposed that SDF(5−67) may play a role in the neurodegeneration associated with HIV dementia, a disease in which levels of MMP-2 are known to be relatively elevated (82, 85).

3. Signalling Related to Proteolysis of Select Cell Surface Molecules MMPs have been demonstrated to cleave a number of cell surface receptors that are expressed on CNS-derived cells and which can in turn inﬂuence intracellular signalling events. Such receptors include those which aﬀect blood brain barrier function, as well as those that can more directly inﬂuence neuronal shape and survival. MMPs may also cleave cell surface molecules that can modulate the survival of CNS derived tumour cells. Moreover, as is described in more detail in those chapters focused on arterial disease and stroke, MMPs can act on receptors that inﬂuence platelet function and aggregation. However, since the number of cell surface molecules that are targeted by MMPs is vast and ever increasing, the discussion to follow will be limited to select examples. 3.1. MMPs can act on molecules that play a role in CNS inﬂammation, such as ICAM Intercellular adhesion molecule-1 (ICAM-1) plays a critical role in leukocyte adherence and extravasation (86). This immunoglobulin supergene family member also plays a role in the homotypic aggregation of T cells, in antigen presentation, in antigen-induced lymphocyte proliferation, and in T cell mediated cytotoxicity. While it has long been known that ICAM expression can be positively regulated at the transcriptional level by proinﬂammatory stimuli, recent studies suggest that the cell surface expression of ICAM may also be negatively inﬂuenced by such. For example, TNF-α stimulated astrocytes will release membrane associated ICAM (87). This

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

130

K. Conant

release is proteinase dependent in that it can be reduced by MMP activity inhibitors (87). In a related study, it has been shown that MMP-9 can cleave ICAM-1 from the cell surface (88). The implications for MMP stimulated ICAM release are broad. ICAM engagement has been linked to intracellular signalling events. The ligation of cell surface ICAM by neutrophils, antibodies, or ﬁbrinogen in diﬀerent cell types can result in intracellular calcium increases, cytoskeletal changes, and gene transcription (89). Cross linking of ICAM in pulmonary microvascular endothelial cells, for example, has been shown to stimulate activation of p38 mitogen-activated protein kinase (p38 MAPK) and associated cytoskeletal rearrangement (89, 90). Cross linking of ICAM in brain endothelial cell lines has also been linked to the activation of pathways related to cytoskeletal changes (91). Soluble ICAM has also been implicated as an eﬀector of intracellular signalling. For example, this molecule can stimulate cytokine production from lymphocytes (92). In addition, soluble ICAM may interfere with the ability of immune eﬀector cells to target ICAM bearing tumour cells (86). Soluble ICAM may also interfere with the ability of leukocytes to adhere to membrane associated ICAM on blood brain barrier endothelial cells and thus to transmigrate into the CNS (93). A decrease in leukocyte migration may in turn, be a protective response designed to dampen inﬂammation. The brain, like other organs, is often more damaged than helped by a vigorous inﬂammatory response. In other studies that have examined the eﬀects of MMPs on cell surface proteins involved in the immune response, it has been shown that MMPs or MMP like proteins can cleave vascular cell adhesion molecule (VCAM) (94), TNF receptors (95), and membrane bound fractalkine (96–99), a chemoattractant protein for monocytes. 3.2. Eﬀects on other adhesion molecules As will be discussed, MMPs may associate with integrins. The potential consequences of such associations are many. One potential consequence that will be brieﬂy mentioned here, however, is the possibility that MMPs can cleave integrins. It has for example been shown that matrilysin (MMP-7) may cleave integrin β4 (100). In this study, α6 and β1 were instead found resistant to degradation by matrilysin. Since the loss of β4 expression is observed in association with numerous cancers, and such cancers may simultaneously be associated with increased MMP activity. Though not

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

131

tested in this study, the loss of β4 might also play a role in the progression of some cancers. In another study, it has been shown that neutrophil elastase can cleave the carboxyl terminus of the αIIb subunit of platelet αIIb β3 , and thus up-regulate its ﬁbrinogen binding activity which may in turn potentiate platelet aggregation (101). MMPs have also been implicated in the shedding of CD44 (102). Like ICAM, CD44 is thought to be involved both in cell adhesion and tumour survival. Shedding of CD44 may be linked to reduced cell contact inhibition and of relevance to signalling, a functional link has been demonstrated between CD44 cleavage and consequent signal transduction within cells. The intracellular domain of CD44 has also been shown to bind to merlin, which typically sends growth inhibitory signals when CD44 is intact and engaged by ligand (103). 3.3. Cleavage of Fas ligand Fas and Fas ligand are cell surface receptors that can interact to stimulate bi-directional signalling. Fas-Fas ligand interactions typically lead to apoptosis in the Fas bearing cell and are thought to play a role in testicular germ cell death (104) and immune system peripheral tolerance (105, 106). Loss of expression of, or functional mutations in the Fas-Fas ligand system have instead been associated with progressive lymphadenopathy and an autoimmune syndrome (106). Fas ligand expression is often observed in ‘immune privileged’ tissues such as the eye or the brain, and its expressing may be increased by pro-inﬂammatory stimuli (106). Fas ligand bearing cells may kill Fas expressing lymphocytes that intrude. Similarly, Fas ligand may be expressed on endothelial cells and play a role in their barrier function by stimulating the death of Fas expressing leukocytes. Tumour cells can also express Fas ligand and may thus trigger apoptosis of anti-tumour leukocytes. Matrilysin or MMP-7 has been shown to cleave Fas ligand (107) and thereby to generate a molecule known as soluble Fas ligand or sFasL. sFasL can interact with membrane Fas, but does not necessarily stimulate Fas signalling. It can therefore act to block membrane bound Fas ligand mediated cell death (106). Based on the discussion in the preceding paragraph, it can thus be imagined that cleavage of Fas ligand may interfere with the ability of immune privileged tissues to kill intruding leukocytes. Interestingly, however, sFasL does not necessarily act simply to block cell death by preventing Fas ligand from interacting with Fas. In some situations, sFasL

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

132

K. Conant

can itself stimulate cell death (107). Thus the consequences of an increase in MMP-7 generated sFasL may be inﬂuenced by factors such as tissue type and disease process. The generation of sFasL by MMP-7 may also aﬀect intracellular signalling cascades. Engagement of intact or membrane associated Fas ligand, which like Fas can act as a receptor, may be linked to fyn, a src family tyrosine kinase (108). As to the potential biological relevance of signalling mediated by the engagement of Fas ligand, such has been associated with CD8 cell proliferation (109). Of course the engagement of Fas is also associated with intracellular signalling events (110). Activation of this TNF/NGF family member has been linked to the recruitment of FADD and in turn, that of procaspase-8. 3.4. MMPs act on receptors important to neuronal migration, structure, and/or survival As previously discussed in the section on cleavage of matrix-like proteins, netrin is a laminin like protein that can act as a soluble axon guidance molecule during CNS development. Netrin, however, may also play a role in the mature CNS. It has been detected in oligodendroglia of the adult rat spinal cord, and following its release, tends to be associated with cell membrane or matrix components. This observation suggests that in the adult CNS, netrin may play a role in axon-oligodendroglial cell interactions (111). Netrin can interact with a number of receptors, and the function of these receptors may in turn be inﬂuenced by MMPs. A role for MMPs in regulation of the netrin system is supported by at least two studies. In one study, it was demonstrated that inhibition of MMP activity could potentiate the axon-outgrowth promoting eﬀects of netrin in dorsal spinal cord explants (57). Subsequent experiments demonstrated that staining for a netrin receptor known as deleted in colorectal cancer, or DCC, was more intense in those cultures that had been treated with the MMP inhibitor. In addition, supernatants of explants that had been treated with the inhibitor showed lower levels of a DCC cleavage product. In another study, it was shown that the migration of distal tip cells in Caenorhabditis elegans was altered in mutants for an ADAM family homologue (112). Moreover, when genes for both this ADAM homologue and a netrin homologue were disrupted, the defect in dorsal tip cell migration was enhanced. Thus, a metalloproteinase-like activity may either interfere with axonal migration or promote distal tip cell migration through eﬀects on the netrin system. That the end result of metalloproteinase activity may

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

133

be dependent on the system under study is not surprising. In some systems cleavage of DCC may be particularly important while in others, eﬀects on basement membrane proteins, processing of netrin itself, or processing of other netrin receptors could play a role. It has been shown, for example, that integrins can mediate netrin-associated migration (113) and, as will be discussed in more detail in the next section, a number of proteinases can interact with integrins. An ADAM protein has also been implicated in Notch signalling (25). An ADAM homologue can cleave Notch in a membrane proximal extracellular region (25, 114, 115). This cleavage event is thought to trigger presenilin mediated cleavage of Notch at an intramembranous site, an event which allows the cytoplasmic portion of Notch to act, with a transcription factor, as a co-stimulus for Notch dependent gene transcription (25, 116). Proteolytic processing is also important to signalling mediated by heparin binding EGF-like growth factor (HB-EGF), a protein with biological eﬀects that include the promotion of neuronal survival (117). Initial studies had shown that EGFR dependent cell migration is diminished by an MP inhibitor but rescued by exogenous EGF (118). Additional studies showed that G protein coupled receptor activation could stimulate intracellular events which could in turn activate a transmembrane ADAM to release membrane associated HB-EGF (119). Thus released, HB-EGF can act on cell surface receptors including ErbB4 and EGFR. Another transmembrane molecule which may be processed by proteolysis and inﬂuence neuronal survival is the amyloid precursor protein (APP). Cleavage of APP in its extracellular domain, mediated by a protease known as β-secretase, in combination with cleavage in its intramembranous portion, mediated by γ-secretase (a presenilin containing complex) (120), leads to the generation of the potentially neurotoxic and pro-inﬂammatory amyloid β peptide (25). An α-secretase may, however, cleave amyloid-β internally to reduce the quantity and thus negative eﬀects of full length peptide. Studies suggest that both ADAM 10 and ADAM 17 may have some ability to act as α-secretases (121, 122). In terms of other cell surface molecules that may inﬂuence neuronal structure or survival, metalloproteinase mediated cleavage of the NGF neurotrophin receptors has been shown (123). MMPs may also cleave SHPS-1, a molecule that is highly expressed in the CNS and thought to play a role in cell migration (124). For a more detailed discussion of the ability of proteinases to cleave these and other molecules important to CNS development and/or neuronal function, the reader is referred to the reference list and to the chapter by Dr. McFarlane.

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

134

K. Conant

3.5. Proteinase activated receptors and the urokinase plasminogen activator receptor Like the MMPs, serine proteinases such as thrombin and plasmin can act on matrix and soluble proteins and may in some cases eﬀect changes in intracellular signalling cascades through indirect mechanisms such as anoikis. A number of serine proteinases may, however, also act on speciﬁc cell surface receptors to more directly stimulate intracellular signalling cascades. Thrombin, for example, can activate a number of the so-called proteinase activated receptors (PARs) (125, 126). These are seven transmembrane G protein coupled receptors that are activated by cleavage of their N terminus. Such cleavage exposes a tethered peptide ligand that binds to another region on the receptor, and thus activates signalling presumably by eﬀecting a structural change. The PARs are widely expressed in the CNS, with robust expression of several subtypes on astrocytes and neurons, and their expression may be increased in association with pro-inﬂammatory stimuli. They may play a role in CNS development and/or pathology, especially in conditions such as stroke or inﬂammation that are characterised by blood brain barrier breakdown and an increase in CNS levels of thrombin. Though PARs are indeed typically and best activated by serine proteinases, there is some evidence that other proteinases, including granzymes can also target these receptors (127). With respect to MMPs, elastase can act on PAR-2 (128). Another link between MMPs and a proteinase receptor system comes from studies on the urokinase plasminogen activator receptor (uPAR). This system may play a role in the activation of MMP-2 and -9 (129), and select MMPs may also cleave this receptor (130). Cleavage of uPAR causes a loss of uPA binding, as well as a dissociation of uPAR/integrin cell surface signalling complexes. Of note, uPAR expressed on neurons and in some circumstances, this receptor may inﬂuence neuronal diﬀerentiation (131).

4. Signalling Related to the Binding of MMPs to Select Cell Surface Receptors 4.1. Like a number of snake venom MPs, MMPs may bind to integrins In several studies, MMPs have been shown to bind to, or associate with, speciﬁc cell surface receptors (132–136). Such interactions may facilitate activation or internalisation of the MMP, or serve to position the MMP so

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

135

that its activity is appropriately localised. Cell surface receptor binding by an MMP may also disrupt cell matrix interactions to inﬂuence cell motility or survival, and might at the same time alter intracellular signalling. One receptor to which MMPs and MMP-like proteinases may bind is α2 β1 . It has been shown, for example, that the snake venom proteinase jararhagin can stimulate ﬁbroblast signalling via α2 β1 (137). This eﬀect was further demonstrated to have been independent of jararhagin’s enzymatic activity and to be non-cation dependent. The region of jararhagin responsible was not identiﬁed but, based on known interactions between integrins and other snake venom metalloproteinases, possibilities include the cysteine rich domain, the RKKH motif in the metalloproteinase domain, or the ECD motif located in the disintegrin-like domain. With respect to this latter ‘disintegrin-like domain’, it should be noted that the term ‘disintegrin’ refers to snake venom proteins which can interfere with integrin function (138). A number of these proteins can bind to αIIb β3 in platelets and thereby interfere with blood clotting. Disintegrins typically, though not always, have an RGD sequence at their active site. Other integrin binding motifs present in snake venoms include MLD (139), RKKH (137) and DECD (140). A partial list of the numerous snake venom MPs, their potential integrin partners, and the biological eﬀects that may follow the listed snake venom MP-integrin interactions is given in Table 1. The motifs that underlie mammalian MMP interactions with integrins have been somewhat less well studied. Similarly, little is known regarding the biological sequelae of such interactions. With respect to MMP-1, it has been proposed that the interaction between pro-MMP-1 and keratinocyte α2 β1 can facilitate activation of pro-MMP-1 at sites of collagen contact (133). Thus, collagen could be cleaved and cell migration facilitated. Non-mutually exclusive possibilities include the potential of the pro-MMP-1/integrin interaction to alter integrin conformation and/or integrin eﬀects on intracellular signalling molecules. In neuronal cultures, pro-MMP-1 may associate with neuronal α2 β1 and stimulate neuronal death (134). This eﬀect, which is independent of MMP-1’s catalytic activity, may be related to changes in the conformation of α2 β1 that lead to the dephosphorylation of Akt (10). If parallels are to be drawn between MMPs and snake venom MPs, it is tempting to speculate that a well-exposed and conserved RGD sequence in the catalytic domain of MMP-1 plays some part in its ability to interact with α2 β1 (Fig. 1). MMP-2 also has a sequence which is known to interact with neuronal integrins, DGEA (141). And like MMP-1, MMP-9 has an

Integrin(s) with which the MP can associate

Eﬀect(s)

Reference(s)

RGD (160)

αv β3 (161)

Inhibits bFGF induced proliferation of bovine capillary endothelial cells (BCE) (161, 162). Induces disassembly of focal adhesions and apoptosis of BCE cells (162). Inhibits migration of mouse melanoma cells (161).

(161, 162)

Rhodocetin

non-RGD

α2 β1

Inhibits α2β1’s interaction with collagen, and antagonises important cellular responses to type I collagen. Has diﬀerential eﬀects on MMP production, and inhibits the invasion of ﬁbrosarcoma cells into a type I collagen matrix.

(163–165)

Kistrin

RGD

αvβ, αIIbβ3 , β complexes that may contain β1 , β2 , β3 or β5

Blocks platelet activation induced by RGD dependent integrins. Blocks adhesion of smooth muscle cells and endothelial cells to matrix components.

(166–168)

α2 β1

Elicits angiogenic eﬀects. Increases human umbilical vein endothelial cell (HUVEC) proliferation as well as HUVEC migration toward immobilised aggretin. Increases VEGF production by HUVEC cells.

(169)

Aggretin

Echistatin

RGD

Activates caspase 3 in GD25 cells prior to inducing cell detachment. Also activates pp125 (FAK), prior to apoptosis/caspase 3 activation.

(170)

Jerdonin

RGD, 12 cysteine residues

Inhibits ADP- and collagen-induced human platelet aggregation. Also inhibits solid tumour growth in C57BL/6 mice.

(171)

K. Conant

Salmosin

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Integrin binding motif(s) present

Snake venom MPs interact with integrins.

May 26, 2005

Snake Venom MP

136

Table 1.

Integrin binding motif(s) present

Integrin(s) with which the MP can associate

Crovidisin

(Continued) Eﬀect(s)

Reference(s)

(172)

MLD (TMLD)

α4 β1 , α9 β1 , α4 β7

Potent inhibitor of α4β1 and α9β1, less eﬀective with α4β7.

(139)

EO5

MLD (TMLD)

α4 β1 , α9 β1 , α4 β7

Potent inhibitor of α4β1 and α9β1, less eﬀective with α4β7.

(139)

EC3

MLD (AMLD)

α4 β1 , α9 β1 , α4 β7

Potent inhibitor of α4β7, less eﬀective with α4β1 and α9β1. Inhibits neutrophil chemotaxis induced by fMet-Leu-Phe but can also act as a potent chemotactic agent. Increases actin polymerisation and induces focal adhesion kinase (FAK) and phosphoinositide 3-kinase (PI3K) activation. Inhibits Erk-2 activation and has a proapoptotic eﬀect on neutrophils.

(139, 173)

Jarastatin

RGD

αM

Inﬂuences neutrophil chemotaxis and activates FAK and PI3K. Induces Erk-2 translocation to nucleus and delays spontaneous apoptosis of neutrophils.

(173)

Jararhagin

RKKH and others

α2 β1

Mimics collagen signalling in ﬁbroblasts with eﬀects including stimulated production of MMPs.

(137)

Halydin

DECD

α2 β1

Is though to inhibit platelet aggregation by suppressing platelet adhesion to collagen via α2 β1 rather than by blocking ﬁbrinogen binding to glycoprotein (GP) IIb–IIIa on the platelet surface.

(140)

137

VLO5

Examples of Signalling by MMPs

Exhibits collagen-binding activity and matrix metalloproteinase activity. Induces detachment of osteosarcoma cells.

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Snake Venom MP

May 26, 2005

Table 1.

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

138

K. Conant

Fig. 1. MMP-1 contains an RGD sequence in its catalytic domain. Shown in 1(A) is the RGD domain in porcine MMP-1 (green stick structure indicated by arrow). The whole molecule is shown. Of the two predominant regions which can be appreciated, the bottom represents the hemopexin domain. The small area between the two regions represents the linker. 1(B) shows MMP-1 as a stick drawing with the RGD domain as the space ﬁlled portion (arrow). 1(C) shows MMP-9, which also contains an RGD domain (green stick structure noted by arrow), though it appears somewhat less exposed in this view. Images were made using the protein databank (http://www.rcsb.org/pdb/) with Protein r Chime software, Explorer 2 and Chime software, and printed with permission (MDL c MDL Information Systems, Inc.).

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

139

RGD sequence (see http://www.rcsb.org/pdb/, and search MMP-2 and -9 to reach structural and sequence information). Future studies will, however, be necessary to determine whether these potential integrin binding sequences play any role in MMP-integrin interactions. Recent work does suggest that like MMP-1, MMP-9 might associate with integrins. For example, pro-MMP-9 can interact with select leukocyte integrin I domains through its catalytic domain while the C-terminal domain of MMP-9 can interact with select activated integrin subunits (142). Of interest, the former interaction occurs in the presence of calcium and is thought to maintain pro-MMP-9 in an inactive state, while the latter appears to play a role in the activation of MMP-9. In leukocytes, elastase has also been shown to associate with an integrin, αM β2 , and to modulate cell adhesion (143). However, whether eﬀects on cell adhesion were due simply to integrin binding, as opposed to integrin receptor cleavage, is unknown. As discussed previously, MMPs can cleave select receptors including integrins. Hemopexin like domains of MMPs may also be involved in integrin binding, and in fact serum hemopexin may inhibit integrin dependent polymorphonuclear leukocyte adhesion to ﬁbrinogen (144). For example, in the study which demonstrated binding of pro-MMP-1 to keratinocytes, the use of MMP-1 chimeras further demonstrated that conservation of the linker and hemopexin regions was necessary for optimal binding to immobilised α2 (132). Moreover, a C-terminal hemopexin fragment of MMP-2 can be generated through autocatalytic activity (145). This fragment can in turn compete with MMP-2 for binding to αv β3 and may thus interfere with the cell surface activation of MMP-2. With respect to the latter, it is possible that a region which includes portions of the linker and hemopexin like domain is involved. The mammalian ADAM proteins may also interact with integrins. These proteins contain a domain homologous to the ‘disintegrins’. For example, ADAM-9 can interact with α6 β1 to stimulate ﬁbroblast Rho kinase signalling and migration in a manner that is independent of its catalytic activity (146). In terms of MMP-integrin interactions, the fact that integrins can form complexes with other cell surface receptors should be mentioned. For example, some integrins associate with tetraspanin family members, and signalling through integrin/tetraspanin complexes may be linked to the activation of Rho GTPases (147–149). Similarly, signalling through integrin/CD47 complexes may be linked to signalling through Gi (150).

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

140

K. Conant

MMPs may also associate with non-integrin cell surface proteins. For example, MMP-9 may be taken up by the low density lipoprotein receptor-related protein (LRP) (151), which has recently been implicated in tPA mediated signalling (152). In this tPA study, the authors noted that complex interactions between the signalling and scavenging actions of LRP are likely to be present. They further pointed out that the cytoplasmic portion of LRP associates with several signal transduction proteins including Disabled and Shc. Whether MMP-9 can, however, act like tPA to stimulate intracellular signalling remains to be determined. MMP like proteins might also associate with non-integrin cell surface molecules. One possibility is an interaction between the thrombospondin like domain of ADAMTS proteinases and thrombospondin receptors. Whether such interactions occur, however, is unknown. ADAMTS1 has been shown to suppress endothelial cell proliferation in vitro (153), which is intriguing in that TSP may have similar actions. Whether this eﬀect involves binding of intact proteinase, or TSP repeats which have been released through proteolysis, is not known. The region of ADAMTS-1 containing the TSP motifs and spacer has been shown to suppress mouse tumour growth (154). 4.2. Interactions between MMPs and cell surface receptors may be relevant to development and disease Abundant evidence suggests that integrins and their ligands play important roles in the development of the CNS, and emerging evidence suggest they may play a role in maintenance of the same (155). Integrins are widely expressed on CNS derived cells including neurons both before and after terminal diﬀerentiation. Moreover, integrins have been examined in studies of neuronal polarisation, neuronal migration, neuronal process outgrowth, axon guidance, neuronal survival, and synaptic stability. MMP-mediated cleavage of matrix proteins that act as integrin ligands may therefore have profound eﬀects on the CNS. MMP-integrin interactions may also have profound eﬀects. The possibility that MMP-integrin interactions might inﬂuence CNS development can be appreciated by considering other integrin binding proteins found in the brain, such as Reelin or endostatin. For example, it has been shown that through interactions with integrins, immobilised endostatin supports endothelial cell migration and survival, while soluble endostatin instead inhibits endothelial cell migration (156). Similarly,

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

141

the molecule Reelin can associate with α3 β1 and thus modulate integrin mediated cell adhesion to arrest neuronal migration and promote neuronal cortical lamination (157). Given the ability of CNS derived cells to release pro-MMP-1, and the potential for this protein to be either soluble or immobilised through association with the cell surface or ECM, MMP-1 might conceivably mimic or interfere with the eﬀects of molecules like Reelin. Select semaphorins may also interact with integrins. For example, semaphorin 7A can enhance axon outgrowth in a β1 integrin dependent manner (158). An RGD sequence within semaphorin 7A seems necessary for this eﬀect in that it was lost when the RGD region was mutated. Moreover, anti-β1 or echistatin, a viper venom RGD peptide which inhibits β1 and β3 integrins, could inhibit semaphorin 7A’s eﬀect on olfactory bulb axon growth. Whether immobilised MMPs might mimic semaphorin 7A’s eﬀects, or whether soluble MMPs might, by interacting with neuronal integrins, interfere with semaphorin’s eﬀects is, however, unknown. As to the signiﬁcance of MMP/integrin interactions in disease rather than development, one possibility is that these modulate cell migration and/or tissue remodelling in conditions are associated by increased MMP production such as inﬂammation (159). For example, if MMPs play a deadhesive role, this may complement their ability to degrade matrix in terms of facilitating cell migration. Moreover, stimulation of intracellular events linked to changes in the cytoskeleton might further facilitate this process. And in situations characterised by extremely high levels of MMP production, the apoptosis that is required for tissue remodelling may be stimulated. Insight into some of these possibilities will likely be gained from studies that are focused on MMP inhibition, and involve the use of chemical MMP inhibitors, mice that under or over express MMPs (including catalytic mutants), and small interfering RNA technology. 5. Miscellaneous As discussed elsewhere in this book, MMPs are typically secreted and their actions thus targeted extracellularly. Nonetheless, the potential for MMPs to act within cells to aﬀect signalling is not diﬃcult to imagine. As reported in a recent study, it was noted that MMP-2 could associate with condensed chromatin as determined by immunogold electron microscopy and immunostaining (160). Nuclear extracts were also found to possess both MMP-2 and gelatinolytic activity. The authors also noted that MMP-2 could be mixed with poly (ADP-ribose) polymerase (PARP) to generate

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

142

K. Conant

a cleavage product. While these results are intriguing, whether MMP-2 is active within the cell nucleus in vivo, and whether it can properly access or eﬃciently process nuclear PARP in vivo, is unknown. It is interesting, however, to consider that MMPs might in some situations be internalised and exert eﬀects from within the cell.

References 1. Chen, Z. L., Indyk, J. A., and Strickland, S. (2003) Mol Biol Cell 14, 2665–2676 2. Gary, D. S., and Mattson, M. P. (2001) J Neurochem 76, 1485–1496 3. Bozzo, C., Bellomo, G., Silengo, L., Tarone, G., and Altruda, F. (1997) Exp Cell Res 237, 326–337 4. Okamoto, M., Mori, S., and Endo, H. (1994) Brain Res 637, 57–67 5. Dedhar, S., Williams, B., and Hannigan, G. (1999) Trends Cell Biol 9, 319–323 6. Liu, S., Calderwood, D. A., and Ginsberg, M. H. (2000) J Cell Sci 113 (Pt 20), 3563–3571 7. Frisch, S. M., and Screaton, R. A. (2001) Curr Opin Cell Biol 13, 555–562 8. Brazil, D. P., and Hemmings, B. A. (2001) Trends Biochem Sci 26, 657–664 9. Datta, S. R., Brunet, A., and Greenberg, M. E. (1999) Genes Dev 13, 2905–2927 10. Pankov, R., Cukierman, E., Clark, K., Matsumoto, K., Hahn, C., Poulin, B., and Yamada, K. M. (2003) J Biol Chem 278, 18671–18681 11. Bourdoulous, S., Orend, G., MacKenna, D. A., Pasqualini, R., and Ruoslahti, E. (1998) J Cell Biol 143, 267–276 12. George, S. J., and Dwivedi, A. (2004) Trends Cardiovasc Med 14, 100–105 13. Bergin, E., Levine, J. S., Koh, J. S., and Lieberthal, W. (2000) Am J Physiol Renal Physiol 278, F758–768 14. Waterhouse, C. C., Joseph, R. R., and Stadnyk, A. W. (2001) Exp Cell Res 269, 109–116 15. Kantak, S. S., and Kramer, R. H. (1998) J Biol Chem 273, 16953–16961 16. McGill, G., Shimamura, A., Bates, R. C., Savage, R. E., and Fisher, D. E. (1997) J Cell Biol 138, 901–911 17. Grossmann, J. (2002) Apoptosis 7, 247–260 18. Cohen, P., and Frame, S. (2001) Nat Rev Mol Cell Biol 2, 769–776 19. Al-Fageeh, M., Li, Q., Dashwood, W. M., Myzak, M. C., and Dashwood, R. H. (2004) Oncogene 23, 4839–4846 20. Thomas, G. M., Frame, S., Goedert, M., Nathke, I., Polakis, P., and Cohen, P. (1999) FEBS Lett 458, 247–251 21. Werb, Z., Tremble, P. M., Behrendtsen, O., Crowley, E., and Damsky, C. H. (1989) J Cell Biol 109, 877–889 22. Forsyth, C. B., Pulai, J., and Loeser, R. F. (2002) Arthritis Rheum 46, 2368–2376

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

143

23. Huhtala, P., Humphries, M. J., McCarthy, J. B., Tremble, P. M., Werb, Z., and Damsky, C. H. (1995) J Cell Biol 129, 867–879 24. Pasco, S., Monboisse, J. C., and Kieﬀer, N. (2000) J Biol Chem 275, 32999–33007 25. Blobel, C. P. (2000) Curr Opin Cell Biol 12, 606–612 26. Bellon, G., Martiny, L., and Robinet, A. (2004) Crit Rev Oncol Hematol 49, 203–220 27. Dong, Z., Kumar, R., Yang, X., and Fidler, I. J. (1997) Cell 88, 801–810 28. Cornelius, L. A., Nehring, L. C., Harding, E., Bolanowski, M., Welgus, H. G., Kobayashi, D. K., Pierce, R. A., and Shapiro, S. D. (1998) J Immunol 161, 6845–6852 29. Lijnen, H. R., Ugwu, F., Bini, A., and Collen, D. (1998) Biochemistry 37, 4699–4702 30. Patterson, B. C., and Sang, Q. A. (1997) J Biol Chem 272, 28823–28825 31. Pozzi, A., LeVine, W. F., and Gardner, H. A. (2002) Oncogene 21, 272–281 32. Eriksson, K., Magnusson, P., Dixelius, J., Claesson-Welsh, L., and Cross, M. J. (2003) FEBS Lett 536, 19–24 33. Claesson-Welsh, L., Welsh, M., Ito, N., Anand-Apte, B., Soker, S., Zetter, B., O’Reilly, M., and Folkman, J. (1998) Proc Natl Acad Sci USA 95, 5579–5583 34. O’Reilly, M. S., Boehm, T., Shing, Y., Fukai, N., Vasios, G., Lane, W. S., Flynn, E., Birkhead, J. R., Olsen, B. R., and Folkman, J. (1997) Cell 88, 277–285 35. Pepper, M. S. (2001) Thromb Haemost 86, 346–355 36. Dickinson, S. C., Vankemmelbeke, M. N., Buttle, D. J., Rosenberg, K., Heinegard, D., and Hollander, A. P. (2003) Matrix Biol 22, 267–278 37. Iruela-Arispe, M. L., Luque, A., and Lee, N. (2004) Int J Biochem Cell Biol 36, 1070–1078 38. Adams, J. C., and Lawler, J. (2004) Int J Biochem Cell Biol 36, 961–968 39. Numakawa, T., Ishimoto, T., Suzuki, S., Numakawa, Y., Adachi, N., Matsumoto, T., Yokomaku, D., Koshimizu, H., Fujimori, K. E., Hashimoto, R., Taguchi, T., and Kunugi, H. (2004) J Biol Chem 279, 43245–43253 40. Koshimizu, H., Araki, T., Takai, S., Yokomaku, D., Ishikawa, Y., Kubota, M., Sano, S., Hatanaka, H., and Yamada, M. (2002) J Neurochem 82, 249–257 41. Hayashido, Y., Nakashima, M., Urabe, K., Yoshioka, H., Yoshioka, Y., Hamana, T., Kitano, H., Koizumi, K., and Okamoto, T. (2003) Int J Mol Med 12, 447–452 42. Imai, K., Hiramatsu, A., Fukushima, D., Pierschbacher, M. D., and Okada, Y. (1997) Biochem J 322 (Pt 3), 809–814 43. Fowlkes, J. L., Serra, D. M., Bunn, R. C., Thrailkill, K. M., Enghild, J. J., and Nagase, H. (2004) Endocrinology 145, 620–626 44. Engsig, M. T., Chen, Q. J., Vu, T. H., Pedersen, A. C., Therkidsen, B., Lund, L. R., Henriksen, K., Lenhard, T., Foged, N. T., Werb, Z., and Delaisse, J. M. (2000) J Cell Biol 151, 879–889 45. Whitelock, J. M., Murdoch, A. D., Iozzo, R. V., and Underwood, P. A. (1996) J Biol Chem 271, 10079–10086

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

144

K. Conant

46. Levi, E., Fridman, R., Miao, H. Q., Ma, Y. S., Yayon, A., and Vlodavsky, I. (1996) Proc Natl Acad Sci USA 93, 7069–7074 47. Kerkela, E., and Saarialho-Kere, U. (2003) Exp Dermatol 12, 109–125 48. Wallner, K., Li, C., Shah, P. K., Wu, K. J., Schwartz, S. M., and Shariﬁ, B. G. (2004) Arterioscler Thromb Vasc Biol 24, 1416–1421 49. VanSaun, M., and Werle, M. J. (2000) J Neurobiol 43, 140–149 50. Sole, S., Petegnief, V., Gorina, R., Chamorro, A., and Planas, A. M. (2004) J Neuropathol Exp Neurol 63, 338–349 51. Gautam, M., Noakes, P. G., Moscoso, L., Rupp, F., Scheller, R. H., Merlie, J. P., and Sanes, J. R. (1996) Cell 85, 525–535 52. Bose, C. M., Qiu, D., Bergamaschi, A., Gravante, B., Bossi, M., Villa, A., Rupp, F., and Malgaroli, A. (2000) J Neurosci 20, 9086–9095 53. Khan, A. A., Bose, C., Yam, L. S., Soloski, M. J., and Rupp, F. (2001) Science 292, 1681–1686 54. Warth, A., Kroger, S., and Wolburg, H. (2004) Acta Neuropathol (Berl) 107, 311–318 55. Yamada, H., Saito, F., Fukuta-Ohi, H., Zhong, D., Hase, A., Arai, K., Okuyama, A., Maekawa, R., Shimizu, T., and Matsumura, K. (2001) Hum Mol Genet 10, 1563–1569 56. Livesey, F. J. (1999) Cell Mol Life Sci 56, 62–68 57. Galko, M. J., and Tessier-Lavigne, M. (2000) Science 289, 1365–1367 58. Culley, F. J., Brown, A., Conroy, D. M., Sabroe, I., Pritchard, D. I., and Williams, T. J. (2000) J Immunol 165, 6447–6453 59. McQuibban, G. A., Gong, J. H., Tam, E. M., McCulloch, C. A., Clark-Lewis, I., and Overall, C. M. (2000) Science 289, 1202–1206 60. McQuibban, G. A., Butler, G. S., Gong, J. H., Bendall, L., Power, C., Clark-Lewis, I., and Overall, C. M. (2001) J Biol Chem 276, 43503–43508 61. Van den Steen, P. E., Proost, P., Wuyts, A., Van Damme, J., and Opdenakker, G. (2000) Blood 96, 2673–2681 62. Conant, K., Garzino-Demo, A., Nath, A., McArthur, J. C., Halliday, W., Power, C., Gallo, R. C., and Major, E. O. (1998) Proc Natl Acad Sci USA 95, 3117–3121 63. Rostasy, K., Egles, C., Chauhan, A., Kneissl, M., Bahrani, P., Yiannoutsos, C., Hunter, D. D., Nath, A., Hedreen, J. C., and Navia, B. A. (2003) J Neuropathol Exp Neurol 62, 617–626 64. Zou, Y. R., Kottmann, A. H., Kuroda, M., Taniuchi, I., and Littman, D. R. (1998) Nature 393, 595–599 65. Albensi, B. C., and Mattson, M. P. (2000) Synapse 35, 151–159 66. Galiano, M., Liu, Z. Q., Kalla, R., Bohatschek, M., Koppius, A., Gschwendtner, A., Xu, S., Werner, A., Kloss, C. U., Jones, L. L., Bluethmann, H., and Raivich, G. (2001) Eur J Neurosci 14, 327–341 67. Barres, B. A., Schmid, R., Sendnter, M., and Raﬀ, M. C. (1993) Development 118, 283–295 68. Lee, R., Kermani, P., Teng, K. K., and Hempstead, B. L. (2001) Science 294, 1945–1948

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

145

69. Frade, J. M., and Barde, Y. A. (1998) Bioessays 20, 137–145 70. Couture, R., Harrisson, M., Vianna, R. M., and Cloutier, F. (2001) Eur J Pharmacol 429, 161–176 71. Takano, M., Horie, M., Narahara, M., Miyake, M., and Okamoto, H. (1999) Immunopharmacology 45, 121–126 72. Yayama, K., Nagaoka, M., Takano, M., and Okamoto, H. (2000) Biochem Biophys Acta 1495, 69–77 73. Stuardo, M., Gonzalez, C. B., Nualart, F., Boric, M., Corthorn, J., Bhoola, K. D., and Figueroa, C. D. (2004) J Leukoc Biol 74. Kozik, A., Moore, R. B., Potempa, J., Imamura, T., Rapala-Kozik, M., and Travis, J. (1998) J Biol Chem 273, 33224–33229 75. Khodorova, A., Navarro, B., Jouaville, L. S., Murphy, J. E., Rice, F. L., Mazurkiewicz, J. E., Long-Woodward, D., Stoﬀel, M., Strichartz, G. R., Yukhananov, R., and Davar, G. (2003) Nat Med 9, 1055–1061 76. Fernandez-Patron, C., Zouki, C., Whittal, R., Chan, J. S., Davidge, S. T., and Filep, J. G. (2001) Faseb J 15, 2230–2240 77. Schonbeck, U., Mach, F., and Libby, P. (1998) J Immunol 161, 3340–3346 78. Griﬃn, W. S., and Mrak, R. E. (2002) J Leukoc Biol 72, 233–238 79. Grilli, M., Goﬃ, F., Memo, M., and Spano, P. (1996) J Biol Chem 271, 15002–15007 80. Viviani, B., Bartesaghi, S., Gardoni, F., Vezzani, A., Behrens, M. M., Bartfai, T., Binaglia, M., Corsini, E., Di Luca, M., Galli, C. L., and Marinovich, M. (2003) J Neurosci 23, 8692–8700 81. Yong, V. W., Krekoski, C. A., Forsyth, P. A., Bell, R., and Edwards, D. R. (1998) Trends Neurosci 21, 75–80 82. Zhang, K., M. G., Silva, C., Butler, G. S., Johnston, J. B., Holden, J., Clark-Lewis, I., Overall, C. M., and Power, C. (2003) Nature Neuroscience 6, 1064–1011 83. Gu, Z., Kaul, M., Yan, B., Kridel, S. J., Cui, J., Strongin, A., Smith, J. W., Liddington, R. C., and Lipton, S. A. (2002) Science 297, 1186–1190 84. Vos, C. M., Sjulson, L., Nath, A., McArthur, J. C., Pardo, C. A., Rothstein, J., and Conant, K. (2000) Exp Neurol 163, 324–330 85. Conant, K., McArthur, J. C., Griﬃn, D. E., Sjulson, L., Wahl, L. M., and Irani, D. N. (1999) Ann Neurol 46, 391–398 86. Lee, S. J., and Benveniste, E. N. (1999) J Neuroimmunol 98, 77–88 87. Lyons, P. D., and Benveniste, E. N. (1998) Glia 22, 103–112 88. Fiore, E., Fusco, C., Romero, P., and Stamenkovic, I. (2002) Oncogene 21, 5213–5223 89. Wang, Q., and Doerschuk, C. M. (2002) Antioxid Redox Signal 4, 39–47 90. Wang, Q., and Doerschuk, C. M. (2001) J Immunol 166, 6877–6884 91. Etienne, S., Adamson, P., Greenwood, J., Strosberg, A. D., Cazaubon, S., and Couraud, P. O. (1998) J Immunol 161, 5755–5761 92. McCabe, S. M., Riddle, L., Nakamura, G. R., Prashad, H., Mehta, A., Berman, P. W., and Jardieu, P. (1993) Cell Immunol 150, 364–375 93. Rieckmann, P., Michel, U., Albrecht, M., Bruck, W., Wockel, L., and Felgenhauer, K. (1995) J Neuroimmunol 60, 9–15

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

146

K. Conant

94. Hummel, V., Kallmann, B. A., Wagner, S., Fuller, T., Bayas, A., Tonn, J. C., Benveniste, E. N., Toyka, K. V., and Rieckmann, P. (2001) J Neuropathol Exp Neurol 60, 320–327 95. Crowe, P. D., Walter, B. N., Mohler, K. M., Otten-Evans, C., Black, R. A., and Ware, C. F. (1995) J Exp Med 181, 1205–1210 96. Garton, K. J., Gough, P. J., Blobel, C. P., Murphy, G., Greaves, D. R., Dempsey, P. J., and Raines, E. W. (2001) J Biol Chem 276, 37993–38001 97. Tsou, C. L., Haskell, C. A., and Charo, I. F. (2001) J Biol Chem 276, 44622–44626 98. Chapman, G. A., Moores, K., Harrison, D., Campbell, C. A., Stewart, B. R., and Strijbos, P. J. (2000) J Neurosci 20, RC87 99. Hundhausen, C., Misztela, D., Berkhout, T. A., Broadway, N., Saftig, P., Reiss, K., Hartmann, D., Fahrenholz, F., Postina, R., Matthews, V., Kallen, K. J., Rose-John, S., and Ludwig, A. (2003) Blood 102, 1186–1195 100. von Bredow, D. C., Nagle, R. B., Bowden, G. T., and Cress, A. E. (1997) Exp Cell Res 236, 341–345 101. Si-Tahar, M., Pidard, D., Balloy, V., Moniatte, M., Kieﬀer, N., Van Dorsselaer, A., and Chignard, M. (1997) J Biol Chem 272, 11636–11647 102. Isacke, C. M., and Yarwood, H. (2002) Int J Biochem Cell Biol 34, 718–721 103. Herrlich, P., Morrison, H., Sleeman, J., Orian-Rousseau, V., Konig, H., Weg-Remers, S., and Ponta, H. (2000) Ann NY Acad Sci 910, 106–118; discussion 118–120 104. Lee, J., Richburg, J. H., Shipp, E. B., Meistrich, M. L., and Boekelheide, K. (1999) Endocrinology 140, 852–858 105. Ju, S. T., Panka, D. J., Cui, H., Ettinger, R., el-Khatib, M., Sherr, D. H., Stanger, B. Z., and Marshak-Rothstein, A. (1995) Nature 373, 444–448 106. Newell, M. K., and Desbarats, J. (1999) Apoptosis 4, 311–315 107. Powell, W. C., Fingleton, B., Wilson, C. L., Boothby, M., and Matrisian, L. M. (1999) Curr Biol 9, 1441–1447 108. Hane, M., Lowin, B., Peitsch, M., Becker, K., and Tschopp, J. (1995) FEBS Lett 373, 265–268 109. Suzuki, I., and Fink, P. J. (1998) J Exp Med 187, 123–128 110. Choi, C., and Benveniste, E. N. (2004) Brain Res Brain Res Rev 44, 65–81 111. Manitt, C., Colicos, M. A., Thompson, K. M., Rousselle, E., Peterson, A. C., and Kennedy, T. E. (2001) J Neurosci 21, 3911–3922 112. Nishiwaki, K., Hisamoto, N., and Matsumoto, K. (2000) Science 288, 2205–2208 113. Yebra, M., Montgomery, A. M., Diaferia, G. R., Kaido, T., Silletti, S., Perez, B., Just, M. L., Hildbrand, S., Hurford, R., Florkiewicz, E., TessierLavigne, M., and Cirulli, V. (2003) Dev Cell 5, 695–707 114. Logeat, F., Bessia, C., Brou, C., LeBail, O., Jarriault, S., Seidah, N. G., and Israel, A. (1998) Proc Natl Acad Sci USA 95, 8108–8112 115. Chan, Y. M., and Jan, Y. N. (1998) Cell 94, 423–426 116. Schroeter, E. H., Kisslinger, J. A., and Kopan, R. (1998) Nature 393, 382–386

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

147

117. Kornblum, H. I., Zurcher, S. D., Werb, Z., Derynck, R., and Seroogy, K. B. (1999) Eur J Neurosci 11, 3236–3246 118. Dong, J., Opresko, L. K., Dempsey, P. J., Lauﬀenburger, D. A., Coﬀey, R. J., and Wiley, H. S. (1999) Proc Natl Acad Sci USA 96, 6235–6240 119. Prenzel, N., Zwick, E., Daub, H., Leserer, M., Abraham, R., Wallasch, C., and Ullrich, A. (1999) Nature 402, 884–888 120. Bergman, A., Laudon, H., Winblad, B., Lundkvist, J., and Naslund, J. (2004) J Biol Chem 279, 45564–45572 121. Buxbaum, J. D., Liu, K. N., Luo, Y., Slack, J. L., Stocking, K. L., Peschon, J. J., Johnson, R. S., Castner, B. J., Cerretti, D. P., and Black, R. A. (1998) J Biol Chem 273, 27765–27767 122. Lammich, S., Kojro, E., Postina, R., Gilbert, S., Pfeiﬀer, R., Jasionowski, M., Haass, C., and Fahrenholz, F. (1999) Proc Natl Acad Sci USA 96, 3922–3927 123. DiStefano, P. S., Chelsea, D. M., Schick, C. M., and McKelvy, J. F. (1993) J Neurosci 13, 2405–2414 124. Ohnishi, H., Kobayashi, H., Okazawa, H., Ohe, Y., Tomizawa, K., Sato, R., and Matozaki, T. (2004) J Biol Chem 279, 27878–27887 125. Dery, O., Corvera, C. U., Steinhoﬀ, M., and Bunnett, N. W. (1998) Am J Physiol 274, C1429–1452 126. Hollenberg, M. D. (2003) Life Sci 74, 237–246 127. Suidan, H. S., Bouvier, J., Schaerer, E., Stone, S. R., Monard, D., and Tschopp, J. (1994) Proc Natl Acad Sci USA 91, 8112–8116 128. Domotor, E., Bartha, K., Machovich, R., and Adam-Vizi, V. (2002) J Neurochem 80, 746–754 129. Ginestra, A., Monea, S., Seghezzi, G., Dolo, V., Nagase, H., Mignatti, P., and Vittorelli, M. L. (1997) J Biol Chem 272, 17216–17222 130. Andolfo, A., English, W. R., Resnati, M., Murphy, G., Blasi, F., and Sidenius, N. (2002) Thromb Haemost 88, 298–306 131. Farias-Eisner, R., Vician, L., Silver, A., Reddy, S., Rabbani, S. A., and Herschman, H. R. (2000) J Neurosci 20, 230–239 132. Stricker, T. P., Dumin, J. A., Dickeson, S. K., Chung, L., Nagase, H., Parks, W. C., and Santoro, S. A. (2001) J Biol Chem 276, 29375–29381 133. Dumin, J. A., Dickeson, S. K., Stricker, T. P., Bhattacharyya-Pakrasi, M., Roby, J. D., Santoro, S. A., and Parks, W. C. (2001) J Biol Chem 276, 29368–29374 134. Conant, K., St Hillaire, C., Nagase, H., Visse, R., Gary, D., Haughey, N., Anderson, C., Turchan, J., and Nath, A. (2004) J Biol Chem 279, 8056–8062 135. Brooks, P. C., Stromblad, S., Sanders, L. C., von Schalscha, T. L., Aimes, R. T., Stetler-Stevenson, W. G., Quigley, J. P., and Cheresh, D. A. (1996) Cell 85, 683–693 136. Yu, Q., and Stamenkovic, I. (1999) Genes Dev 13, 35–48 137. Zigrino, P., Kamiguti, A. S., Eble, J., Drescher, C., Nischt, R., Fox, J. W., and Mauch, C. (2002) J Biol Chem 277, 40528–40535 138. Ruoslahti, E. (1996) Annu Rev Cell Dev Biol 12, 697–715

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

148

K. Conant

139. Bazan-Socha, S., Kisiel, D. G., Young, B., Theakston, R. D., Calvete, J. J., Sheppard, D., and Marcinkiewicz, C. (2004) Biochemistry 43, 1639–1647 140. You, W. K., Jang, Y. J., Chung, K. H., and Kim, D. S. (2003) Biochem Biophys Res Commun 309, 637–642 141. Buhusi, M., Midkiﬀ, B. R., Gates, A. M., Richter, M., Schachner, M., and Maness, P. F. (2003) J Biol Chem 278, 25024–25031 142. Bjorklund, M., Heikkila, P., and Koivunen, E. (2004) J Biol Chem 279, 29589–29597 143. Cai, T. Q., and Wright, S. D. (1996) J Exp Med 184, 1213–1223 144. Suzuki, K., Kobayashi, N., Doi, T., Hijikata, T., Machida, I., and Namiki, H. (2003) Cell Struct Funct 28, 243–253 145. Brooks, P. C., Silletti, S., von Schalscha, T. L., Friedlander, M., and Cheresh, D. A. (1998) Cell 92, 391–400 146. Nath, D., Slocombe, P. M., Webster, A., Stephens, P. E., Docherty, A. J., and Murphy, G. (2000) J Cell Sci 113 (Pt 12), 2319–2328 147. Delaguillaumie, A., Lagaudriere-Gesbert, C., Popoﬀ, M. R., and Conjeaud, H. (2002) J Cell Sci 115, 433–443 148. Berditchevski, F. (2001) J Cell Sci 114, 4143–4151 149. Hemler, M. E. (2001) J Cell Biol 155, 1103–1107 150. Brown, E. J., and Frazier, W. A. (2001) Trends Cell Biol 11, 130–135 151. Hahn-Dantona, E., Ruiz, J. F., Bornstein, P., and Strickland, D. K. (2001) J Biol Chem 276, 15498–15503 152. Wang, X., Lee, S. R., Arai, K., Tsuji, K., Rebeck, G. W., and Lo, E. H. (2003) Nat Med 9, 1313–1317 153. Iruela-Arispe, M. L., Carpizo, D., and Luque, A. (2003) Ann NY Acad Sci 995, 183–190 154. Kuno, K., Bannai, K., Hakozaki, M., Matsushima, K., and Hirose, K. (2004) Biochem Biophys Res Commun 319, 1327–1333 155. Clegg, D. O., Wingerd, K. L., Hikita, S. T., and Tolhurst, E. C. (2003) Front Biosci 8, d723–750 156. Rehn, M., Veikkola, T., Kukk-Valdre, E., Nakamura, H., Ilmonen, M., Lombardo, C., Pihlajaniemi, T., Alitalo, K., and Vuori, K. (2001) Proc Natl Acad Sci USA 98, 1024–1029 157. Dulabon, L., Olson, E. C., Taglienti, M. G., Eisenhuth, S., McGrath, B., Walsh, C. A., Kreidberg, J. A., and Anton, E. S. (2000) Neuron 27, 33–44 158. Pasterkamp, R. J., Peschon, J. J., Spriggs, M. K., and Kolodkin, A. L. (2003) Nature 424, 398–405 159. Murphy-Ullrich, J. E. (2001) J Clin Invest 107, 785–790 160. Kwan, J. A., Schulze, C. J., Wang, W., Leon, H., Sariahmetoglu, M., Sung, M., Sawicka, J., Sims, D. E., Sawicki, G., and Schulz, R. (2004) Faseb J 18, 690–692 161. Kim, S. I., Kim, K. S., Kim, H. S., Choi, M. M., Kim, D. S., Chung, K. H., and Park, Y. S. (2004) Oncol Res 14, 227–233 162. Hong, S. Y., Lee, H., You, W. K., Chung, K. H., Kim, D. S., and Song, K. (2003) Biochem Biophys Res Commun 302, 502–508

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch05

Examples of Signalling by MMPs

149

163. Eble, J. A., Beermann, B., Hinz, H. J., and Schmidt-Hederich, A. (2001) J Biol Chem 276, 12274–12284 164. Eble, J. A., Niland, S., Dennes, A., Schmidt-Hederich, A., Bruckner, P., and Brunner, G. (2002) Matrix Biol 21, 547–558 165. Eble, J. A., and Tuckwell, D. S. (2003) Biochem J 376, 77–85 166. Juliano, D., Wang, Y., Marcinkiewicz, C., Rosenthal, L. A., Stewart, G. J., and Niewiarowski, S. (1996) Exp Cell Res 225, 132–142 167. Handagama, P., Bainton, D. F., Jacques, Y., Conn, M. T., Lazarus, R. A., and Shuman, M. A. (1993) J Clin Invest 91, 193–200 168. Yee, K. O., Ikari, Y., Bodary, S., and Schwartz, S. M. (2000) Thromb Res 97, 39–50 169. Chung, C. H., Wu, W. B., and Huang, T. F. (2004) Blood 103, 2105–2113 170. Alimenti, E., Tafuri, S., Scibelli, A., D’Angelo, D., Manna, L., Pavone, L. M., Belisario, M. A., and Staiano, N. (2004) Biochem Biophys Acta 1693, 73–80 171. Zhou, X. D., Jin, Y., Chen, R. Q., Lu, Q. M., Wu, J. B., Wang, W. Y., and Xiong, Y. L. (2004) Toxicon 43, 69–75 172. Tang, C. H., Yang, R. S., Liu, C. Z., Huang, T. F., and Fu, W. M. (2004) Toxicon 43, 11–20 173. Coelho, A. L., De Freitas, M. S., Mariano-Oliveira, A., Rapozo, D. C., Pinto, L. F., Niewiarowski, S., Zingali, R. B., Marcinkiewicz, C., and BarjaFidalgo, C. (2004) Exp Cell Res 292, 371–384

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

This page intentionally left blank

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

PART IV MMPs AND TIMPs IN DEVELOPMENT

151

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

This page intentionally left blank

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

CHAPTER 6 METALLOPROTEINASES IN DEVELOPMENT — BREAKING THINGS DOWN TO BUILD A NERVOUS SYSTEM

S. McFarlane University of Calgary, Genes and Development Research Group, HSC 2207, 3330 Hospital Dr., Calgary, AB, T2N 4N1 E-mail: [email protected]

1. Introduction During development of the nervous system, neurons are ﬁrst generated and then migrate to their ﬁnal destination. Once there, they send out a single axon, followed by multiple dendrites, to make connections with target cells. All of these events are regulated by extrinsic factors in the environment. Research over the last several years has revealed that the signalling between molecules in the environment and their receptors on the neuron can be modulated through the action of a family of zinc-dependent proteolytic enzymes, the metalloproteinases. Cleavage of either ligands or their receptors by metalloproteinases can result in activation or termination of signalling. As such, these proteases are clearly important in regulating key processes in brain development, and their involvement will be discussed in this chapter. Metalloproteinases are members of the Metzincin superfamily, and include the matrix metalloproteinase (MMP) and the A Disintegrin and Metalloproteinase (ADAM) families. The former contains over 30 members (at least 20 in mammals), and is implicated in remodelling the extracellular matrix (ECM) in diverse biological and pathological central nervous system (CNS) processes (1, 2). Most MMPs are secreted, although a subset are membrane-type MMPs (MT-MMPs). In contrast, the ADAMs, which also make up a group of at least 30 members, are mostly transmembrane Correspondence to: S. McFarlane 153

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

154

S. McFarlane

proteins. In addition to their proteolytic domain, ADAMs contain an extracellular disintegrin domain, which allows ADAMs to act as a ligand for integrin receptors, and compete for binding with ECM proteins (3). Half of the known ADAM proteins have a defective proteinase domain and are catalytically inactive (4). It has been proposed that these inactive ADAMs might act as endogenous inhibitors of catalytically active ADAMs (5). However, in C. elegans the inactive ADM-1 (UNC-71) protein does not appear to interact with the other worm ADAMs, which are active, but may act in a cell adhesion dependent fashion to aﬀect axon guidance decisions of GABAergic motoneurons (6). The inactive metalloproteinase domain may still be important as a mediator of proper protein processing or targeting (6, 7). For instance, expression of an UNC-71 with a mutated metalloproteinase domain fails to rescue the axonal defects in unc-71 mutants (6). In this chapter, I will restrict my discussion of metalloproteinases to those shown to function in neuronal development. I will outline what is known about metalloproteinases in early developmental events, such as neurogenesis and neuronal migration, and later processes such as the establishment of connections between neurons and neuronal survival. 2. Metalloproteinase Expression in the Developing Nervous System Members of both the MMP and ADAM families are expressed in the developing nervous system, where their diversity and widespread distribution suggest they may play numerous roles (2). The large number of diﬀerent family members seems daunting when considering their roles in the nervous system, but many are not expressed in neural tissues. For instance, an RT-PCR analysis indicated that while at least 16 diﬀerent ADAMs are expressed in the murine adult brain, only 10 of these are expressed at any appreciable level, and of these, 7 are predicted to be catalytically active (8). In the adult CNS, certain ADAMs are expressed by neurons, others by either astroctyes or oligodendrocytes (4, 9, 10). Unfortunately, only a few studies have reported embryonic expression (4) (Table 1). ADAM9, ADAM10, ADAM19 and one of the isoforms of ADAM23 are expressed in the nervous system of mouse embryos, primarily in neuroepithelial precursors and postmitotic neurons (4, 11). These ADAMs are diﬀerentially expressed. For instance, ADAM10 is expressed in the forebrain and sympathoadrenal and olfactory neural precursors, whereas ADAM19 is expressed in the craniofacial ganglia, dorsal root ganglia (DRG), and the ventral horns

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development Table 1. systems.

155

Expression of ADAM family members in the adult and developing nervous

ADAM

Expression pattern

Reference 8

ADAM1

Inactive

Relatively low abundance by Northern analysis

ADAM 2–5

Inactive

Very low levels in brain by RT-PCR

8

ADAM7

Inactive

Anterior pituitary in the adult

108

ADAM8

Active

In adult CNS (not found by RT-PCR analysis; 8)

109

ADAM9 (Meltrin γ)

Active

Ubiquitous in embryonic brain

110

ADAM10 (Kuzbanian)

Active

Neurons in embryonic brain

13

ADAM11

Inactive

Widespread expression in neurons of embryonic brain and PNS

111

ADAM12 (Meltrin a)

Active

Found at P10 in rat brain; not expressed in embryonic mouse

10, 112

ADAM13 (Xenopus)

Active

Embryonic neural crest cells

62

ADAM14 (C. elegans; ADM-1, UNC-71)

Inactive

D-type motoneurons

6

ADAM15 (Metargidin)

Active

Embryonic expression not described; in adult brain

113

ADAM17 (TACE)

Active

Low levels in embryonic mouse brain; in postnatal cerebellum and DRG

114

ADAM19 (Meltrin β)

Active

Embryonic cranial facial and dorsal root ganglia; ventral horns of spinal cord

12

ADAM21

Inactive

Very low levels in brain by RT-PCR

8

ADAM22

Inactive

Purkinje cells of cerebellum; neurons in neocortex

4

ADAM23

Inactive

Fetal brain Gamma isoform in embryonic brain

11

ADAM33

Active

Various brain regions

115, 116

of the spinal cord (12, 13). Of the catalytically inactive ADAMs, three are expressed at signiﬁcant levels in the adult brain, but their function is unknown (4). Only a few studies have reported the expression patterns of MMPs in the developing nervous system (Table 2). Interestingly, they show that MMPs are often expressed in migrating populations of neurons. For instance MMP-8 is expressed in migrating neural crest cells (14), and MMP-9 in radially migrating immature granule cells (15). Additionally, some MMPs are

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

156

S. McFarlane Table 2.

Expression of MMPs in the developing nervous system.

MMP

Expression pattern

Reference

MMP2

Postnatal rat cerebellar Purkinje cells; human CNS neuroepithelial stem cells

16, 19

MMP3 (stromelysin-1)

Cultured neurons of embryonic cerebral cortex; rat E15 CNS and PNS; postnatal rat cerebellar Purkinje cell dendrites

52, 68

MMP8

Migrating neural crest cells

14

MMP9

Neuroprogenitor cells; oligodendrocytes of mouse optic nerve; postnatal cerebellar granule precursors cells

15, 17, 19, 107

MMP21

Mouse embryonic neuroectoderm

117

MT-MMP5

Developing cerebellum; associated with Purkinje cell dendritic tree formation

18, 102

expressed in precursor cells. MMP-2 is expressed in human neuroepithelial CNS stem cells (16) and MMP-9 in progenitor cells of the pituitary gland and the choroid plexus (17). MMPs are expressed in structures such as the cerebellum during the postnatal period when diﬀerentiation of the various cell types occurs (15, 18, 19). What is striking, however, is how little is known about the expression of the over 20 diﬀerent MMPs in the developing nervous system. 3. Analysis of Metalloproteinase Function To assess potential roles for metalloproteinases in CNS development, investigators have taken several approaches, including the use of: (1) Genetically deﬁcient mice, (2) Speciﬁc hydroxamate inhibitors designed against the proteinase catalytic site (e.g. GM6001, IC-3, BB-94), (3) One of the four tissue inhibitors of metalloproteinases (TIMPs) that act as endogenous regulators of metalloproteinase function, and (4) Dominant negative mutant proteins against speciﬁc metalloproteinase family members, engineered by mutating and rendering inactive the proteinase domain. These studies have implicated metalloproteinases in such processes as neurogenesis, myelination, axon guidance and neural crest cell migration (2). The majority of information has been garnered using the pharmacological inhibitors, which are broad spectrum, and cannot diﬀerentiate between MMPs and ADAMs. Though eﬀective, dominant negatives are not often used, so our knowledge of the role of individual metalloproteinases has been restricted to the

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

157

analysis of transgenic mice, for the most part with null mutations of the metalloproteinase of interest. In this regard, the analysis has proven particularly disappointing for the identiﬁcation of roles for MMPs in nervous system development. Indeed, all MMP knockout mice complete embryogenesis and are born. MMP-3, MMP-7, MMP-11 and MMP-12 mutant mice have no apparent developmental defects (20–24). MMP-2 mutants display defects in lung branching (25) and MMP-14 mutants die postnatally with skeletal and connective tissue defects (26, 27). Certainly, at the gross morphological level no nervous system defects have been observed and/or reported. Given the large number of MMPs, this may either be due to compensation through up-regulation of other MMPs, or redundancy (28, 29). Interestingly, it was shown recently that ﬂies mutant for both of the known Drosophila MMP genes show no obvious nervous system or embryonic defects (30). This is despite the fact that in situ hybridisation analysis of MMP-1 and MMP-2 mRNA shows extensive expression of both genes in the embryonic nervous system. For instance, MMP-1 is expressed in repeating cells along the ventral midline of the CNS, whereas MMP-2 mRNA is found in the migrating stomatogastric nervous system, peripheral nervous system (PNS), and developing CNS. The reported analysis of the mutants, however, was restricted primarily to quantifying the percentage of embryos that survived, pupated and hatched. Thus, it is possible that roles in nervous system development may become apparent on closer examination. Indeed, the initial reports on the mouse MMP-9 knockout animals described a defect in long-bone growth (31). However, a more recent study has described subtle transient changes in the numbers, survival and migration of granule cell precursors in the postnatal cerebellum (15), consistent with the spatiotemporal expression pattern of MMP-9 in this structure (19). The role for ADAM family members in nervous system development is on much stronger ground than that of MMPs. A number of the catalytically active ADAM family members are expressed in the nervous system, though the embryonic expression has only been investigated in a few cases (reviewed in 4) (Table 2). Moreover, several of the mice homozygous null for a particular ADAM family member show defects in neural development, and will be discussed below. For both MMPs and ADAMs, it is likely that more extensive analysis of the embryonic expression patterns, and phenotypes of the mouse knockouts, will identify additional players in the near future.

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

158

S. McFarlane

4. Neurogenesis Cell–cell interactions play an important role in regulating cell fate decisions in the developing nervous system. For instance, during Drosophila neurogenesis, a single neural precursor is singled out from a group of equivalent cells through a process called lateral inhibition, in which the selected cell prevents its neighbours from taking on the same fate (reviewed in 32). The Notch gene encodes for a transmembrane receptor that along with its ligands, Serrate and Delta, is an important mediator of lateral inhibition (reviewed in 33). Loss-of-function mutations in Notch result in hyperplasia of neural cells at the expense of epidermis (reviewed in 34). Drosophila embryos that lack all maternal and zygotic product from the gene encoding the adam10 homologue, kuzbanian (kuz ), exhibit a similar phenotype, which results in embryonic lethality (35). A clonal mosaic analysis approach was taken to more speciﬁcally address roles for kuz in inhibiting neural fate during Drosophila neurogenesis. With this technique, patches of cells mutant for kuz develop in amongst wildtype cells (35). These studies showed that kuz plays a key role in regulating the Notch-mediated lateral inhibition involved in the generation of sensory bristles in the adult epidermis (Fig. 1(A)). During the development of adult sensory bristles, lateral inhibition assures that only a few cells are selected out of a group (proneural cluster) to become sensory organ precursors (SOPs). SOPs ultimately give rise to the sensory bristles. The SOP is thought to activate the Notch pathway in its surrounding neighbours and suppress them from taking on an SOP fate (32). In kuz mutant clones, supernumerary bristles were observed, similar to what is observed with the loss of the function of Notch (35). Expression of a dominant negative form of KUZ that lacked the proteinase domain, under the control of a heat shock promoter, produced the same phenotype (5). The data also showed that KUZ and Notch likely act in the same cell, as the function of both proteins is required cell autonomously for those cells that are inhibited from acquiring a neural fate (35). Presumably mutant cells in the proneural cluster are unable to receive an inhibitory signal from the emerging neural/SOP cell that normally prevents them from adopting this fate. Subsequent analysis revealed, through genetic epistasis experiments, that KUZ functions upstream of Notch, as discussed below (5). A non-cell autonomous function for kuz was also revealed by the sensory bristle mosaic analyses (35). Mutant bristle clusters were only observed

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

159

(A) kuz required cell autonomously to inhibit neural fate

_

E

Np

+

+

+

Notch

+ Wildtype

Cleaved Notch

- KUZ mutant

Fig. 1. In Drosophila, kuz acts cell autonomously and non-cell autonomously to regulate the neural fate decision. (A) In a proneural cluster, one cell, via a process of lateral inhibition, is selected out to become a neural precursor (Np). The neural precursor inhibits (arrows) its neighbours from acquiring the same fate, and they become epidermal (E) cells. This process is mediated by the transmembrane receptor Notch, and its ligand Delta. Notch needs to be proteolytically cleaved before it becomes activated; the proximal membrane ‘site 2’ cleavage of Notch requires the actions of a metalloproteinase, KUZ (5). kuz mutant cells cannot cleave Notch and thus are unable to transduce the neural inhibitory signal from the selected neural precursor and themselves become sensory bristles (35). (B) Diﬀerent fates are observed for kuz mutant cells at the edge of clones, and in their centres. Rooke and colleagues (35) postulated that normally epidermal cells produce a signal that promotes the neural precursor fate. On the edge of a clone, wildtype cells will still provide this signal to the kuz mutant cells, which respond and are pushed towards a neural fate. In contrast, in the centre of the clone, a mutant cell is surrounded by other mutant cells unable to produce the neural signal, and so will go on to acquire an epidermal fate (adapted from 35).

at the edges of kuz-/- clones, with cells in the interior of the large clones acquiring an epidermal fate (Fig. 1(B)). These data can be explained if normally the neighbours of the neural precursor produce a neural inducing signal, which is dependent on KUZ function. At the borders of clones, kuz-/- cells will receive the signal from their wildtype neighbours and go on to form sensory bristles. In contrast, in the interior of the clone there are no cells to produce the signal, and kuz-/- cells go on to acquire an epidermal fate. These data suggest that there are two signals necessary for proper sensory bristle formation. A neural inducing signal that promotes

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

160

S. McFarlane

(B) kuz required non-cell autonomously to promote neural fate

_

_

+ _

_ _

_

+

_

+ clone border

middle of clone

Fig. 1.

(Continued )

neural precursor formation, and lateral inhibition that pushes other cells in the proneural cluster towards an epidermal fate, and both signalling events require KUZ (36). kuz also inhibits the neural fate in Drosophila ommatidia (5), where Notch-mediated lateral inhibition plays a role in photoreceptor (R) development (37). KUZ function was perturbed by expressing the dominant negative KUZ under the control of the rough enhancer, which drives expression in all cells within the morphogenetic furrow, as well as transiently in the R2-R5 photoreceptors posterior to the furrow (5). The furrow is a wave of pattern formation that sweeps across the undiﬀerentiated eye disc from posterior to anterior, and is the site of organisation and diﬀerentiation of ommatidial cells (37). Knock down of KUZ function resulted in supernumerary photoreceptor cells in each ommatidium, suggesting that KUZ normally functions to limit the number of photoreceptors in the developing Drosophila eye (5). The vertebrate kuz homologue, ADAM10, may also regulate the neural fate choice. This was ﬁrst shown for primary neurogenesis in Xenopus laevis (5). In Xenopus, primary neurons are generated in three bilaterally arranged columns and can be identiﬁed by their expression of a neuralspeciﬁc β-tubulin gene (N -tubulin) (38). In this system, manipulations that disrupt Notch-mediated lateral inhibition result in an overproduction of N -tubulin positive primary neurons. This was the phenotype observed

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

161

when mRNA encoding a dominant negative version of mouse ADAM10 was injected into one side of a 2-cell stage Xenopus embryo, which indicates that ADAM10 is also required for the lateral inhibition process in Xenopus (5). ADAM10 homologues have also been cloned in human, mouse and chick (13, 39). Analysis of the chick and mouse embryos deﬁcient in ADAM10 function, along with the Xenopus data discussed above, argue for an evolutionarily conserved role for ADAM10 in Notch signalling (39, 40). Mouse mutants died at embryonic (E) day 9.5, with a foreshortened forebrain anlage and abnormal hindbrain ﬂexure. Moreover, the fusing neural tube had irregularly shaped lateral walls; a defect described previously in other mouse models with impaired Notch signalling. Similarly, electroporation of a dominant negative ADAM10 construct into the early ectoderm of stage 7–9 chick embryos led to neural speciﬁcation, characterised by a thickening of the ectoderm and expression of an early neuronal marker, N -cadherin (39). The authors of the mouse study went on to show, by using in situ hybridisation, that the mRNA expression of components of the Notch pathway was aﬀected in the ADAM10 mutants. The expression pattern of hairy enhancer of split-5 (hes-5 ), a basic helix-loop-helix transcription factor that is activated by Notch signalling, in the forebrain and brain stem was severely disorganised and reduced, and undetectable in the spinal cord. Furthermore, there was an increased expression of the Notch ligand, delta like-1 (dll-1 ), which is normally downregulated by Notch signalling via a negative-feedback mechanism. Because the adam10 -/- mice die before neurons are born, the mice cannot be used to address a role for ADAM10 in later neuronal speciﬁcation. Instead, an in vitro approach has proven informative. In the mouse PNS, sympathoadrenal precursor cells move ventrally from the neural tube and diﬀerentiate into sympathetic neurons or migrate further to the adrenal medulla, where they undergo chromaﬃn diﬀerentiation. In mouse, ADAM10 is expressed in the embryonic sympathetic ganglia and in adrenal medullary tumour cell lines, but not in the adult adrenal medulla (13). A role for ADAM10 in the decision to become a chromaﬃn cell versus a sympathetic neuron was suggested by experiments where one of the adrenal tumour cell lines, the pheochromocytoma (PC12) cell line, was transfected with a dominant negative chick adam10. The proteinase deﬁcient ADAM10 induced neurite outgrowth, which normally requires the presence of a neurotrophin such as nerve growth factor (NGF). These data are somewhat confusing in that the expression data infers that continued ADAM10

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

162

S. McFarlane

expression shifts a precursor cell to a sympathetic neuronal cell fate, whereas experimentally the opposite is observed. A number of studies have focused on how KUZ acts through the Notch pathway to regulate lateral inhibition, and, it is now clear that KUZ regulates Notch signalling via two diﬀerent mechanisms (40, 41). First, it performs the extracellular ‘site2’ cleavage of the Notch receptor close to the transmembrane domain, which is followed, upon ligand binding, by a KUZ-independent intramembranous cleavage (S3) to release the active soluble cytoplasmic domain (42). Second, KUZ acts in a cell autonomous fashion to cleave, or shed, two diﬀerent ectodomains of the Notch ligand Delta (41, 43). These soluble Delta fragments are biologically inactive both in vitro and when ectopically expressed in ﬂies, which indicates that cleavage downregulates the activity of the Notch ligand (41). A model was proposed where Delta inactivation would help to ensure the epidermal fate of cells receiving the neural inhibiting signal (41) (Fig. 2). These cells would express Notch, Delta and KUZ, and KUZ-mediated cleavage of Delta would prevent them from talking back to their neighbour that had been selected to become a neural precursor. This model relies on KUZ being diﬀerentially regulated between neighbouring cells, which has yet to be demonstrated. It should be noted that the embryonic neurogenic phenotype of a Drosophila kuz null mutant is more severe than that of a Notch null (35), suggesting that there are additional proteolytic substrates for KUZ (5). Indeed, mouse ADAM10 can cleave other substrates, such as the epidermal growth factor (EGF) receptor and the cell adhesion molecule (CAM), L1 (44, 45). In vertebrates, there has been some debate as to which ADAM is responsible for Notch cleavage. In vitro, Notch cleavage is performed by ADAM17 and not ADAM10 (46, 47). However, ADAM17-deﬁcient mice show no sign of a ‘Notch phenotype’, whereas ADAM10 null mice do (40). These data suggest that diﬀerent ADAMs may contribute to Notch ‘site2’ cleavage in a tissue speciﬁc manner. 5. Neuronal Survival During development, neuronal numbers are controlled by a balance of the generation of new neurons, and cell death via apoptosis. Only a few studies have examined changes in neuronal apoptosis after manipulation of metalloproteinase function. For instance, MMP-9 was implicated in promoting neuronal apoptosis in the developing mouse cerebellum. In this

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

(A)

(B) Wildtype

gon-1

163

(C)

mig-17

Dorsal DTC

Gonad arm

Ventral

Fig. 2. DTC migration in the C. elegans gonad is regulated by metalloproteinases. (A) DTCs migrate to form the worm gonad. First, they migrate along the ventral body muscle wall, and then make a turn and grow dorsally to reach the dorsal body wall muscle. A second turn is made and the DTCs migrate along the dorsal body wall muscle. (B) Both the DTCs and the dorsal body wall muscle secrete the metalloproteinase GON-1. It is required for cell migration, because in gon-1 mutants DTCs fail to migrate (58). This phenotype could be rescued by GON-1 expression in the dorsal muscle. (C) The metalloproteinase MIG-17 is secreted by the dorsal body wall muscle, and acts on DTCs (7). This metalloproteinase is not required for cell migration, but for directing the trajectory of DTCs after they have made the initial turn dorsally. In mig-17 mutant embryo, the DTCs turn ventrally and grow along the ventral body wall muscle (adapted from 56).

system, MMP-9 is expressed in a premigratory pool of granule cell precursors located in the inner part of the external granular layer (EGL) (15, 19). Granule cell precursors undergo apoptosis in the EGL, mainly at the beginning of the second postnatal week. In MMP-9 deﬁcient mice, fewer apoptotic TUNEL positive ﬁgures were observed in this layer at postnatal (P ) day 12 (15). Vaillant and colleagues (15) speculated that metalloproteinases might indirectly regulate apoptosis by modulating synapse formation between granule cell parallel ﬁbres and Purkinje cell dendrites over the postnatal period. Alternatively, metalloproteinases could inhibit

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

164

S. McFarlane

adhesion-dependent mechanisms, the consequence of which is known to be programmed cell death (48, 49). For instance, local disruption of the ECM by MMP-3/stromelysin-1 induced a loss of cell adhesion of mammary epithelial cells, which resulted in their elimination by selective programmed cell death (50). While MMP-9 promotes granule cell precursor apoptosis (15), an ADAM family member, ADAM8, promotes cerebellar granule cell survival (51). ADAM8 ectodomain cleavage of the CAM ‘close homologue of L1’ (CHL1) activates the molecule and promotes the survival of cultured postnatal cerebellar granule cells, a function that ADAM10 or ADAM17 were not able to fulﬁll. Thus, it is the balance of the functions of diﬀerent metalloproteinases that ultimately may decide whether a neuron survives or dies. In contrast, it is the balance between MMP-3 and the tissue inhibitor of metalloproteinase-3 (TIMP-3) that seems important to regulate neuronal sensitivity of embryonic rat cortical neurons to doxorubicin-induced apoptosis (52). Doxorubicin induces apoptosis in these neurons via the death receptor, Fas, a member of the tumour necrosis factor receptor superfamily. An interaction between Fas and the transmembrane Fas-ligand (Fas-L) at the cell surface of neighbouring cells is thought to mediate cell death in response to the drug. TIMP-3 and MMP-3 were shown to be constitutively expressed by primary cortical neurons in culture, and manipulating their function altered neuronal apoptosis induced by doxorubicin: an antibody against TIMP-3 and addition of active MMP-3, but not MMP-7, both attenuated doxorubicin-induced neuronal death. Active MMP-3 also caused diminished Fas-Fas-L interactions at the cell surface, presumably because of Fas-L cleavage by MMP-3, which provides a reasonable explanation for how MMP-3 regulates the decision between neuronal survival and death. However, while Fas is important in neuronal death in the embryonic brain (53), it is unclear how the doxorubicin experimental model of apoptosis relates to neuronal apoptosis in CNS development. Finally, ADAMs may act via the neurotrophins, which are growth and survival factors, to modulate neuronal survival (54). The low aﬃnity neurotrophin receptor, p75NTR , binds all the neurotrophins. In addition to functions in myelination and neurite outgrowth, p75NTR is important in promoting cell survival in some cells and apoptosis in others (55). A membrane proximal cleavage may be necessary for intramembrane proteolytic processing of p75NTR and thus its function. When p75NTR is transfected into murine ﬁbroblasts, its extracellular portion is shed constitutively at low levels, and this is greatly enhanced by the addition of

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

165

the phorbol ester PMA. To assess the role of candidate ADAMs in p75NTR cleavage, the release of the p75NTR ectodomain in murine ﬁbroblasts isolated from adam9 /12 /15 / , adam10 / , or adam17 / mice was assayed. Whereas PMA strongly enhanced p75NTR shedding in adam9 /1 /12 /15 / or adam10 / cells, no PMA-dependent stimulation of p75NTR shedding was observed in adam17 / cells. As yet, a role for ADAM cleavage of the p75NTR in neuronal survival has not been demonstrated. Moreover, since adam17 deﬁcient mice die perinatally, it will be necessary to generate conditional knockout animals to assess ADAM17 function in postnatal neuronal processes requiring cleavage of p75NTR .

6. Neuronal Migration Several observations suggest metalloproteinases may function in neuronal migration. First, in non-neuronal systems, cell migration is regulated by metalloproteinases that either act proteolytically on the substrate the cells migrate over, or, in the case of ADAMs, actually provide a substrate via the disintegrin domain (reviewed in (56, 57)). For instance, in C. elegans, a secreted member of the ADAMTS (ADAMs with thrombospondin-1 repeats) family, GON-1, functions in distal tip cell (DTC) migration (58). In the formation of the worm gonad, DTCs ﬁrst migrate along the ventral surface, and then turn dorsally and migrate over the basal lamina of the dorsal body wall muscle (Fig. 2(A)). gon-1 is expressed both by the migrating DTCs and by the dorsal muscle cells. DTCs fail to migrate in gon-1 mutants (Fig. 2(B)). This phenotype is rescued only by expression of GON-1 in the DTCs, and not the muscle cells. GON-1 may function in DTC migration by structurally remodelling the ECM, as has been reported for laminin-5 activation by MMP-2 in the case of migrating breast epithelial cells (59). In the developing vertebrate nervous system, several MMPs and ADAMs are expressed in migrating neurons. In only a handful of studies, however, has a role in neuronal migration been tested in vivo. The most concrete case for metalloproteinase participation in neuronal migration is the migration of neural crest cells (NCCs). NCCs are a population of neurons that leave the dorsal neural tube by undergoing an epithelial to mesenchymal transition (EMT), and then migrate through the ECM to reach their targets. The cells diﬀerentially express several MMP and ADAM family members prior to, during, and after the EMT. For instance, MMP-2 is expressed

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

166

S. McFarlane

in chick NCCs as they detach from the neural epithelium, but is rapidly extinguished as they disperse (60), whereas ADAM10, at least in vitro, is expressed both in early outgrowth NCCs as well as migrating melanoblasts (39). Metalloproteinases, for example MMP-8, are also expressed in NCCs in mouse embryos (14). The function of NCC-expressed metalloproteinases in the migration of NCCs has been tested for MMP-2 in chick and ADAM13 in Xenopus. Speciﬁc inhibition of MMP-2 in the dorsal neural tube with antisense morpholino oligonucleotides, prevented the EMT both in tissue culture and in vivo (60). However, NCCs that had already left the neural tube were unaﬀected. Whether MMP-2 is required in NCCs or the substrate they grow through is confusing in that an earlier study reported that MMP-2 is expressed in only a subset of chick NCCs late in their migration, but is expressed in the basement membranes deposited by mesodermal cells upon which NCCs migrate (61). In contrast to MMP-2, Xenopus ADAM13 is expressed in cranial NCCs prior to and during their migration (62). Functional analysis has indicated that ADAM13 is required in Xenopus for NCC migration (63). Cranial NCCs emerge from the neural tube and migrate in three streams towards the ventral side of the embryo. These are the mandibular, hyoid and branchial streams, with the latter subdividing further into two. An elegant transplantation approach was used to place either wildtype or proteinase defective, dominant negative ADAM13 misexpressing cranial NCCs into unmanipulated host embryos, and assay for defects in cranial NCC migration (63). No eﬀect was observed with misexpression of the wildtype ADAM13, but expression of the dominant negative construct in NCCs resulted in a retardation of the hyoid and branchial cranial NCC streams (Fig. 3). Timelapse videomicroscopy showed dominant negative expressing cranial NCC cells failing to enter the hyoid and branchial pathways. These data are consistent with a model in which ADAM13 functions to promote NCC migration by modifying a component of the ECM. ADAM13 may either help cells detach from the neuroepithelium by decreasing cell adhesion to the surrounding ECM, or may cleave ECM proteins and clear migration pathways. Alternatively, ADAM13 may function to direct cranial NCC migration by unmasking potential guidance cues along their migratory routes. The mandibular cranial NCC stream did not appear to be as aﬀected, suggesting that an alternate mechanism is able to stimulate migration of these cells in the absence of functional ADAM13.

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

167

Fig. 3. ADAM13 is required for cranial neural crest cell migration in Xenopus. mRNA encoding green ﬂuorescent protein (GFP) and either a wildtype (WT) or dominant negative (DN) form of ADAM13 were injected into a 2-cell stage Xenopus embryo (63). Later, GFP-positive NCCs were dissected from stage 17–18 embryos and transplanted into an uninjected host embryo. The trajectories of the GFP positive NCCs could then be assayed. Brightﬁeld (a–c) and epiﬂuorescent (a –c ) photomicrographs of three embryos that received transplants. GFP and GFP/WT-ADAM13 expressing NCCs migrate in four streams (mandibular, hyoid and the branchial that divides into an anterior and posterior stream). In contrast, when NCCs misexpress a proteinase deﬁcient DN-ADAM13 the cells often fail to migrate. The mandibular NCC stream is aﬀected in some individuals, and unaﬀected in others (see c ) (adapted from 63 ‘Xenopus ADAM 13 is a metalloprotease required for cranial neural crest-cell migration’, and reprinted with permission from Elsevier).

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

168

S. McFarlane

Granule cell precursors of the developing mouse cerebellum are the only other cell type for which a function of metalloproteinases in migration has been examined. In this system, granule cell precursors divide in the outer part of the external granular layer (EGL) and subsequently migrate inward through the molecular layer and past the Purkinje cells to reach the internal granular layer (IGL). The spatiotemporal pattern of MMP-9 expression is correlated with the major stage of granule cell migration (15, 19): MMP-9 is expressed in a pool of pre-migratory granule cell precursors within the EGL, and along their path in Purkinje cell bodies and dendrites, and in some cells in the IGL. When P 8 mouse cerebellar explants were cultured, axons expressing a granule cell marker, TAG-1, extended radially from the explants (15). Subsequently, TAG-1 positive cells extended along these axons. In the presence of an MMP-9 function-blocking antibody both ﬁber outgrowth and cell migration were dramatically impaired. Though interesting, it is likely that cells failed to migrate as a consequence of severely impaired neurite outgrowth. Nonetheless, delayed migration is a possible explanation for the transient increase in granule cell precursor numbers and enlargement of the EGL observed in MMP-9 deﬁcient mice. While the thicker EGL could be explained by reduced apoptosis in the mutant animals (see above), birthdating experiments provided further support for the delayed migration model. On P 10, bromodeoxyuridine (BrdU) was used to label granule cell precursors during their last round of mitosis, just prior to their migration. When the location of the BrdU-positive cells was assayed two days later, only in wildtype mice had some of the cells completed their migration to the IGL. No diﬀerence was observed in MMP-9 deﬁcient mice in the number of cells present in the IGL at P 21, indicating that though delayed, granule cells eventually reached their target.

7. Axon Extension and Guidance Found at the tips of developing axons, growth cones interpret cues in the environment in order to grow and reach the appropriate target tissue. Recent data indicate that metalloproteinases regulate either the presentation of guidance cues to the growing axon, or transduction of the cues by the receptor in the growth cone. The ﬁrst evidence that metalloproteinases might be important in regulating growth cone behaviour was the observation of metalloproteinases or proteolytic activity in the growth cones of numerous vertebrate neurons (64–67). More recent analysis has identiﬁed individual MMPs in growth

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

169

cones, including MMP-2, MMP-3, MMP-9 and MT-MMP-5 (18, 68–70). Treatment of neuronal cultures with metalloproteinase inhibitors has been used to assay for a role of metalloproteinases in neurite outgrowth. In such studies, axon elongation was impaired (67, 71–73). In addition, upon NGF treatment PC12 cell lines engineered to express antisense mRNA for stromelysin-1/MMP-3 did initiate neurites, but had diﬃculty in penetrating a basal lamina when compared to control NGF-treated cells (68). These culture data are supported by the observation that Drosophila kuz mutants exhibit CNS axon extension defects (74). KUZ is expressed in all CNS neurons during the period when they extend axons. Interestingly, the later growing axons were more aﬀected than the pioneering axons. The authors postulated that this might reﬂect the fact that axons are faced with an increasingly complex extracellular environment as development proceeds. Several recent papers have raised the possibility that metalloproteinases also regulate the guidance of axons. Two culture studies indicate that metalloproteinases cleave known axon guidance molecules and aﬀect axonal behaviour (75, 76). For example, ADAMs have been implicated in terminating the high aﬃnity interaction between the ephrins and their Eph receptors (76). This axon guidance signalling system is known to regulate the topographic mapping of retinal ganglion cell (RGC) axons in the superior colliculus in mouse, and the optic tectum in chick (77, 78). RGC axons that express Eph receptors are repulsed by ephrins expressed in the target. Hattori et al (2002) showed that ADAM10, which is widely expressed in the embryonic day 18 mouse brain, might regulate the interaction of membrane-tethered ephrins with Ephs by cleavage of the ephrin ectodomain (Fig. 4). Co-immunoprecipitation experiments suggested that ADAM10 forms a stable complex with ephrinA2. Interaction of the complex with an Eph receptor, in this case expressed by the growth cone, then results in the speciﬁc and temporally regulated cleavage of ephrinA2. The end result of the cleavage is to terminate the ligand/receptor interaction. Because the interaction that normally occurs between an ephrin and its receptor is high aﬃnity, this would allow the growth cone to withdraw from the ephrinexpressing cell it had contacted. For instance, when growth cones of Eph expressing hippocampal neurons encountered cells expressing the repulsive ephrinA2, they collapsed and withdrew (Fig. 4). In contrast, in the presence of a metalloproteinase inhibitor, GM6001, where the ephrin/Eph interaction could not be terminated, only collapse was observed. Two elegant features of this regulation were hypothesised. First, cleavage can only happen when the speciﬁc Eph receptor-ephrin ligand interaction occurs, presumably because the interaction reveals the metalloproteinase cleavage site on

Metalloproteinase Activity

Metalloproteinase Inhibition

COLLAPSE

Fig. 4. ADAM10 regulates the cleavage of ephrinA2. A growth cone expressing an Eph receptor encounters a cell in its pathway, which expresses the ligand ephrinA2 in a complex with ADAM10 (76). Interaction of the growth cone with the cell results in the formation of a tri-protein complex between ADAM10, ephrinA2 and the Eph receptor. This permits ADAM10 to cleave the ectodomain of ephrinA2, and terminate Eph-ephrin signalling. Subsequently the growth cone retracts from the cell. In the presence of metalloproteinase inhibitors the growth cone collapses upon contacting the target cell, but does not withdraw.

S. McFarlane

Eph EphrinA2 ADAM10

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

REPULSION

May 26, 2005

170

ADHESION

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

171

ephrin. Second, the released soluble ephrin might additionally inhibit further Eph signalling by binding to and preventing the receptor’s interaction with cell-anchored ephrin. The in vitro functional data that argued for a role of ADAM10 in terminating Eph-ephrin interactions used hippocampal neurons (76). However, both ephrinA2 and ADAM10 are expressed in a graded posterior-to-anterior fashion in the mouse midbrain, strongly suggesting the involvement of ADAM10 in the topographic mapping of RGC axons. The early embryonic lethality (E9.5) of ADAM10-deﬁcient mice has prevented an analysis of ADAM10 function in retinal topographic mapping, thus we will have to wait for conditional knockouts to address this issue. Determining, however, whether pharmacological inhibition of metalloproteinases has similar eﬀects on RGC growth cones as it does on hippocampal growth cones is now possible. While the Hattori study (2000) implicates ADAM10 in the control of RGC axon topographic mapping, a report in chick questions the importance of metalloproteinases in this event in vivo as metalloproteinase inhibitors had no eﬀect on the ephrin-mediated preference of RGC axons for anterior versus posterior tectal membranes (72). The interaction of another well-known pair of guidance molecules, Netrin and its receptor Deleted in Colorectal Cancer (DCC), may also be terminated by metalloproteinase activity. The metalloproteinase inhibitors IC-3 and GM6001 potentiated Netrin-1 dependent stimulation of neurite outgrowth from rat embryonic dorsal spinal cord explants (75). These inhibitors also blocked shedding of the ectodomain of DCC, as assayed biochemically and immunocytochemically. Presumably, Netrin-1 is more eﬀective at stimulating outgrowth in neurons that express more intact receptors. The importance of either the DCC or ephrin cleavage events has not been demonstrated in vivo. Further, in the case of DCC, a candidate proteinase has not been identiﬁed. Taken together these studies indicate that metalloproteinases are part of the mechanism that steers extending axons; however, until recently in vivo evidence has been lacking. The ﬁrst hint of an interaction between metalloproteinases and guidance molecules in vivo came from a study of the migrating DTCs in the C. elegans gonad (7). In this system, DTCs ﬁrst migrate ventrally, and then turn to migrate over the basal lamina of the dorsal body wall muscle (Fig. 2(A)). The dorsal turn requires a secreted member of the ADAM family, MIG-17. In the mig-17 mutant, DTCs extend normally but are misdirected shortly after they make the dorsal turn (Fig. 2(B)). This phenotype is rescued when MIG-17 is expressed either in the muscle

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

172

S. McFarlane

cells or the DTCs. The authors postulated that the muscle cells secrete MIG-17, which is then present on the surface of the gonad arms at the time DTCs make their dorsal turn. Interestingly, mig-17 interacts genetically with unc-6, the C. elegans homologue of netrin. The discovery that a metalloproteinase and a known guidance molecule act in the same genetic pathway to direct cell migration, raises the possibility that similar interactions play a role in axon guidance. Direct evidence for metalloproteinase involvement in axon guidance comes from studies of the role of kuz in the embryonic Drosophila CNS, where it is widely expressed (74, 79). In zygotic kuz mutants there was a decrease in the thickness of longitudinal CNS axon tracts, and an increase in commissural region staining due to a failure of longitudinal axon bundles to cross the midline (Fig. 5). Instead, the axons were said to stall out at the midline (74). These defects were apparently speciﬁc to the CNS, in that at least one population of motoneurons extended axons normally to innervate their muscle target. Given the importance of maternal KUZ on the neuronal cell fate decision in Drosophila, it was possible that the defects observed in the axon bundles were due to a failure of proper cell fate determination. However, this does not appear to be the case. For instance, no major changes were observed in the expression patterns of at least two CNS cell fate markers, Even-skipped and Engrailed. Moreover, kuz expression was only required in postmitotic neurons, and not neural progenitors to rescue the axon extension phenotype. KUZ function was required in the axons themselves as expression of a dominant negative KUZ under the control of a neuronal speciﬁc promoter using the GAL4-UAS system, produced the same axonal defects (5). Interestingly, targeted expression of the dominant negative KUZ to CNS midline cells by using a single-minded GAL4 driver also caused a kuz -like axonal phenotype, in a cell non-autonomous fashion (79). As a whole, the kuz mutant studies indicate that kuz has two functions during Drosophila development. First, as described earlier, maternal kuz is suﬃcient to correctly specify cell fates. Later in development, zygotic kuz functions during axon extension. The kuz mutant axonal phenotype was described as axon stalling by the authors, which led them to suggest that KUZ participates in axon extension (74). More recently, the stalling phenotype has been reinterpreted as an axon guidance defect (79). Fasciclin II-positive longitudinal axons are prevented from crossing the midline by repulsive interactions between Slit, a repellant localised to the midline, and the Slit receptor family, Roundabout

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

173

Fig. 5. KUZ is required for normal axon outgrowth in the Drosophila CNS. Axons in the embryonic Drosophila CNS can be labelled with an antibody that recognises all axons (A-B) or with anti-Fasciclin II (C-D) that labels only the longitudinal axons (arrowhead in A), which in wildtype embryos do not cross the midline (C) (74). A, C: In wildtype embryos, in addition to the longitudinal axons evident in C, commissural axons cross the midline as seen in A (small arrow). B, D: In zygotic kuz mutants, the longitudinal tracts are much thinner and the commissures are disorganised. In addition, many Fasciclin-II positive axons approach and cross the midline (D) (79). (Reproduced from 74, ‘The cell surface metalloprotease/disintegrin Kuzbanian is required for axonal extension in Drosophila’, copyright 1996, National Academy of Sciences, U.S.A., used with permission.)

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

174

S. McFarlane

(robo) (reviewed in 80) (Fig. 5(C)). Thus, in robo mutants the longitudinal axons are no longer repelled, but rather criss-cross the midline. In kuz mutants the longitudinal tracts are reduced in size and many Fasciclin II positive axons cross the CNS midline, which suggests that KUZ may interact with the Slit/Robo signalling pathway (79) (Fig. 5(D)). This idea was supported by genetic interaction experiments. Flies double mutant for both robo and kuz, have a phenotype only slightly more severe than the robo-/phenotype. Moreover, if one copy of kuz is removed in a slit mutant background a slit -/- phenotype is observed. Since both of these genes aﬀect CNS axon development, the fact that the axonal phenotype is not particularly enhanced in the double mutants argues that the two genes act in the same genetic pathway. Based on these data, KUZ function appears necessary to promote Slit-Robo signalling. It may do so by activating either Slit or one or more of the Robo receptors through proteolytic cleavage. Interestingly, KUZ also down-regulates Robo expression in commissural axons that cross the midline. When a dominant negative KUZ was targeted to midline cells by using a single-minded GAL4 driver, axons continued to express Robo (a similar phenotype was apparently observed in kuz-/ﬂies). The reason these Robo-positive commissural axons still crossed was likely a result of the requirement for KUZ in enabling repulsive Slit-Robo signalling. Not only are catalytically active metalloproteinases implicated in axon guidance, but a recent report in C. elegans suggests that a proteinaseinactive ADAM, UNC-71 (ADM-1), can also act to regulate the guidance of the axons of GABAergic D-type motoneurons (6). This group of motoneurons consists of 6 embryonically born DD neurons, and 13 VD neurons born at the end of the ﬁrst larval stage. All 19 D-type neurons have axons that follow the same basic pattern. A longitudinal process is sent out anteriorly in the ventral nerve cord. A circumferential process, or commissure, then branches oﬀ the longitudinal axon and grows dorsally. Upon reaching the dorsal nerve cord the axon bifurcates and branches extend in both anterior and posterior directions. unc-71 is a previously characterised mutant with axon defects (81–83). The unc-71 gene was cloned and found to be a member of the ADAM family of metalloproteinases. An analysis of the amino acid sequence suggests that UNC-71 (otherwise known as ADM-1) is probably catalytically inactive. Of the twenty or more known alleles of unc-71, most have mutations in either the disintegrin or cysteine rich domains, suggesting that UNC-71 may function in cell adhesion. In unc-71 mutants, some but not all of the guidance decisions of D-type neurons are aﬀected.

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

175

First, there is defasciculation of the DD and DV longitudinal axon bundles. Second, 39% of commissures fail to branch from the longitudinal axon, and in another subset a branch emerges on the wrong side of the animal. UNC-71 is not produced by D-type neurons, nor can overexpression of unc-71 in these neurons rescue the axonal defects in unc-71 mutants. Instead, rescue is achieved when unc-71 expression is driven either pan-neuronally or in epidermis. These data indicate that UNC-71 has a non-cell autonomous eﬀect on the guidance of D-type neuronal axons. How might UNC-71 be aﬀecting D-type motoneuron axons? The data argue that UNC-71 has speciﬁc actions, rather than globally aﬀecting the environment through which the axons traverse. For instance, the axon trajectories of other ventral motoneurons (e.g. DA and DB motoneurons) are hardly aﬀected in the unc-71 mutants, in contrast to what is observed for D-type motoneurons. This is despite the fact that the somata of the neurons are close, and their axons follow similar paths. Integrin- and netrindependent signalling pathways are known to participate in the fasciculation of the ventral nerve cord and circumferential navigation of motoneuron axons, respectively. Double mutants between an unc-71 null allele and null alleles of members of either the integrin (ina-1, pat-3 ) or netrin (for e.g. unc-6, unc-5 ) signalling pathways had phenotypes more severe than the single mutants. These data argue that unc-71 acts independently, but in parallel, with these pathways to aﬀect D-type motoneuron axons. Thus, as yet the molecular nature of UNC-71 function is unknown. The idea that metalloproteinases also participate in vertebrate axon guidance is supported by the expression of mouse ADAM10, the homologue of the Drosophila axon guidance molecule KUZ, in developing olfactory nerves and in the sensory trigeminal projection innervating the head (13). Moreover, in embryonic chick ADAM10 is expressed in ventral root motor ﬁbres (76). Further backing for metalloproteinase involvement in guiding vertebrate axons comes from two recent reports. First, a member of the ADAM family, ADAM23, was identiﬁed in a gene trap screen in mice for genes controlling neural connectivity (84). However, these authors only brieﬂy mentioned that ADAM23 is expressed by axons in the developing mouse CNS, and that ADAM23 homozygous mutants have tremor and are ataxic (84, 85). The second study asked whether metalloproteinases regulate the growth and guidance of RGC axons in developing visual system of Xenopus laevis. Here, a useful exposed brain preparation has allowed the role of candidate molecules in RGC axon outgrowth to be tested in vivo (86). Two

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

176

S. McFarlane

Fig. 6. Metalloproteinases are required for RGC axons in the developing visual system of Xenopus laevis to make two key guidance decisions. (A) In control, axons grow dorsally through the diencephalon (Di) to reach their midbrain (Mb) target, the optic tectum (TEC). Along the way they make two key guidance decisions. They make a turn in the mid-diencephalon (1), and they recognise the optic tectum as their target (2). (B) If metalloproteinase inhibitors are applied to the developing optic tract by using an exposed brain preparation as the axons are entering the ventral diencephalon (*), the axons fail to make the diencephalic turn, and grow dorsally towards the pineal gland (Pi) (86). (C) If metalloproteinase inhibitors are applied after RGC axons have made the turn in the diencephalon (*), the axons fail to innervate the optic tectum and instead turn and grow dorsally along its anterior border.

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

Fig. 6.

177

(Continued )

diﬀerent metalloproteinase inhibitors, GM6001 and BB-94, induced similar defects in the optic projection as RGC axons extended through the diencephalic neuroepithelium (Fig. 6). At low doses of the inhibitors, most axons extended normally. They failed, however, to make appropriate guidance decisions at two choice points. First, RGC axons grew straight in the mid-diencephalon, where they normally make a caudal turn towards their target, the optic tectum (Fig. 6(B)). Secondly, if the inhibitors were not applied until after the axons had made this turn, axons failed to recognise the optic tectum and instead turned to grow along the anterior tectal border (Fig. 6(C)). At higher concentrations of the metalloproteinase inhibitors, axon extension defects were observed, with no obvious eﬀect on the neuroepithelium. The diﬀerential sensitivity of axon guidance and extension to metalloproteinase inhibition argue that these two processes are modulated by diﬀerent metalloproteinases. Consistent with this hypothesis, diﬀerent metalloproteinases, GON-1 and MIG-17, regulate movement and guidance of DTCs (7, 58) (Fig. 2). In summary, these results clearly support a role for metalloproteinases in axon extension and guidance in vivo. It remains to be determined whether metalloproteinase function is required within the growing axons themselves or in the substrate they extend through. The literature suggests several interesting candidate targets for metalloproteinases in developing Xenopus visual system (87–91). They include

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

178

S. McFarlane

ephrins, DCC, and ﬁbroblast growth factor receptors (FGFRs) (75, 76, 92). Importantly, these molecules are implicated in Xenopus RGC axon guidance, and are cleaved by metalloproteinases in other systems. Ephrins are unlikely to be involved in that there is no data to suggest that they act in the diencephalon to direct axon trajectories. As discussed earlier, metalloproteinases cleave the netrin receptor DCC (75), a signalling pathway that is necessary for RGC axons to leave the eye and in the guidance of the optic chiasm (91, 93). DCC mRNA is expressed by neuroepithelial cells in the mid-diencephalon and at the border of the optic tectum (94), suggesting that normal metalloproteinase-dependent regulation of DCC function may be required for proper directed RGC axon growth. Finally, growth cone FGFR signalling is required for RGC axons to recognise the optic tectum as their target, and inhibiting RGC growth cone FGFRs produced targeting defects similar to those observed with inhibition of metalloproteinases (86, 90). The extracellular portion of the FGFR can be cleaved by MMP-2 (92). If metalloproteinases constitutively activated FGFR signalling as a result of cleavage, as has been demonstrated for the trkA neurotrophin receptor (95), it could explain why inhibition of metalloproteinases and FGFRs produce the same axonal phenotype. Whether the identiﬁed Xenopus MMP-2 (96) aﬀects RGC FGFR signalling has not yet been determined. The studies in this section strongly argue that ADAM family members function in axon guidance both in vertebrates and invertebrates. A direct role for MMPs in axon guidance remains to be demonstrated, though MMPs are expressed in the growth cones of a number of diﬀerent neuron types (18, 68–70). Moreover, inhibition of MMP-3/stromelysin-1 in PC12 cells by an antisense approach, inhibited neurite outgrowth (68). Finally, it is possible that MMPs are required for Xenopus RGC axon guidance, as the inhibitors used in the study would have blocked both MMPs and ADAMs (86). 8. Mechanisms of Metalloproteinase Function How do metalloproteinases regulate the developmental events in which they are involved? As discussed earlier, metalloproteinases regulate the neural fate decision via cleavage of the Notch receptor and its ligand Delta. Whereas, the only study that addressed this issue for neuronal survival, implicated the Fas-ligand for the Fas death receptor as the cleavage target (52). We know little about how metalloproteinases regulate neuronal migration and axon outgrowth, but given the similarity of the molecular mechanisms involved in the two processes, it is likely that metalloproteinases

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

179

function in a like manner for both. The classical view for how metalloproteinases regulate migration of neuronal somata and axons is that they chew up the ECM and clear a passage (67). Further insight into how metalloproteinases may function comes from in vitro studies that examined the actions of metalloproteinases on speciﬁc axon guidance signalling pathways, and on non-neuronal cell migration (57, 75, 76). These studies now suggest that metalloproteinases may speciﬁcally interact with growth and guidance signalling pathways — via either the ligands and/or their receptors — with either pathway activation or repression as the outcome. Many transmembrane proteins are proteolytically cleaved to reach their biologically active conﬁguration. The ligand targets for metalloproteinases can be secreted or bound to cell membranes, and cleavage reveals or terminates the protein’s activity. Our knowledge is currently restricted to ﬁbronectin and ephrins as metalloproteinase substrates implicated in neuronal migration and axon guidance, respectively (63, 76). In vitro, Xenopus ADAM13 can cleave ﬁbronectin, and mouse ADAM10 targets ephrinA2 (Fig. 4). In either case, however, it has not been demonstrated that the protein is targeted by metalloproteinases in vivo. The ephrin cleavage is an example where metalloproteinases terminate signalling. Ligand activation is also possible, though this has yet to be demonstrated for molecules that participate in either neuronal migration or axon outgrowth. Nonetheless, molecules that you would expect to function in axon outgrowth and neuronal migration, such as growth factors and CAMs, are activated by metalloproteinases in other systems. For instance, metalloproteinases release heparin binding-EGF (HB-EGF), transforming growth factor α, L1 and several neurotrophins in activated forms (1, 45). The activated ligands then have actions either in the soluble form, or bound to the ECM substrate. In the case of L1, cleavage produces a soluble form that can bind either integrin receptors or laminin (45). Interestingly, soluble L1 is produced in the developing brain, though its function is not understood. Finally, depending on the substrate it cleaves, the same MMP can activate one pathway and inhibit another, sometimes with the same ultimate consequence. For example, MMP-2 facilitates cell motility by revealing an encrypted form of laminin-5 in the ECM (59), and inactivating a neurite inhibiting chondrotin sulfate proteoglycan present in sections of adult sciatic nerve (70). Two alternate, though less well-studied, mechanisms by which metalloproteinases may regulate cell behavior include cleavage of receptors and cell adhesion. A number of cell surface receptors undergo cleavage near the

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

180

S. McFarlane

transmembrane domain, a process called ectodomain shedding. Receptor cleavage would either terminate or activate receptor signalling. Termination appears to be the case with the Netrin receptor, DCC (75), while cleavage of the ectodomain of the trkA neurotrophin receptor results in activation by generating cell-bound receptor fragments whose intracellular domain is constitutively tyrosine phosphorylated (95). Moreover, the soluble released ectodomains of the receptors themselves may have additional functions, such as inhibitors that bind to and prevent ligands from interacting with cell surface receptors. The second mechanism, cell adhesion, would be speciﬁc to ADAM family members as they contain a disintegrin domain (3). For instance, while ADAM-9 causes the shedding of HB-EGF, it also binds to α6β1 integrin on cultured ﬁbroblast cell lines, enhancing their motility (97). The metalloproteinase and disintegrin functions are likely independent of each other as the binding of an ADAM-9/Fc protein to integrin, and enhanced cell motility, were unaﬀected by an inhibitor of proteolytic activity, BB-94. It is likely the adhesion function of UNC-17 (adm-1 ), a proteolytically inactive ADAM in C. elegans that is important for the guidance of a subset of ventral motoneurons (6). Indeed, the mutations in the various unc-17 alleles are mostly found either in the disintegrin or cysteine-rich domain. Metalloproteinases modulate growth and guidance signalling, but are themselves regulated by the ECM. This has been demonstrated for the migration of human endothelial cells (98). These cells locomote on gelatin, but not on β1-integrin dependent substrates such as collagen and ﬁbronectin. Endothelial cells that are quiescent express inactive MT1-MMP that is associated with β1-integrin at the site of cell-to-cell contacts. In motile cells, however, an active form of MT1-MMP is localised by clustered αvβ3 integrin at structures involved in motility, such as ﬁlopodia. These data indicate that the ECM substrate regulates both the location and the activity of the metalloproteinases. A model was proposed whereby MMPs would be activated at the onset of cell migration, where they would have two roles. First they would cleave adhesion receptors and allow cells to separate from one another. Second, the MMPs would regulate cell motility by ECM degradation. This study starts to address the issue of what turns on the metalloproteinases in the ﬁrst place. In other cases, growth factors and signalling through G-protein coupled receptors can activate metalloproteinases (67, 99).

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

181

9. Dendrite Development If metalloproteinases are important for soma and axonal migrations, it is likely that they are also important in dendrite formation during development. While this has not yet been investigated, a recent study has implicated MMP-9 in dendrite remodelling in the adult rat hippocampus (100). Systemic administration of the glutamate receptor agonist, kainate, causes a generalised increase in synaptic activity and seizures followed by tissue remodelling in the dentate gyrus (101). Szklarczyk and colleagues (100) found that kainate administration up-regulated MMP-9 protein and mRNA expression in the dendritic layer of the dentate gyrus. These data argue for a role of MMP-9 in activity-dependent remodelling and/or formation of dendritic structures in the adult CNS. As yet, these data are correlative, and do not test a role for MMP-9 in adult dendritogenesis. Moreover, whether MMP-9 functions similarly in development, where activity modulates the formation of both axonal and dendritic arbors, is speculative. Nonetheless, it would not be surprising if in the future metalloproteinases were also found to be important in regulating dendritic outgrowth in the embryonic and postnatal brain. Certainly, at least one membrane-type MMP, MT5MMP, is expressed in the dendrites of postnatal rat Purkinje cells during a period of extensive dendritic arborisation (102). Along these lines, a recent paper has indicated that the secreted growth factor neuregulin-2 (NRG-2) is targeted to the proximal primary dendrites of hippocampal and Purkinje cerebellar neurons in vivo (103). NRGs (also known as acetylcholine receptor inducing activity, glial growth factor, heregulin or neu diﬀerentiation factor) are a family of growth factors that arise from alternative splicing of a single gene. They mediate a variety of biological functions, which include stimulation of glial Schwann cell growth and synthesis of acetylcholine receptors in skeletal muscle (104). NRGs act through the ErbB family of tryosine receptor kinases. Most soluble NRGs are derived from membrane-anchored precursor proteins via proteolytic cleavage of the extracellular region. This cleavage may act as a critical regulatory mechanism of NRG function and recent data suggest that ADAM19 processes NRGs (105). During mouse development, ADAM19 is co-expressed along with NRGs in the craniofacial and dorsal root ganglia, and in the ventral horns of the spinal cord (105). For instance, the majority of neuroﬁlament 160 positive DRG neurons express both ADAM19 and NRGs simultaneously. The

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

182

S. McFarlane

mouse L929 ﬁbroblast cell line expresses low levels of ADAM19, mainly within the cell. When these cells were transfected with a construct encoding NRG-B1, a soluble active form of this growth factor was released into the media. Transfection of the cells with wildtype ADAM19 stimulated proteolytic processing, whereas protease-deﬁcient mutants of ADAM19 inhibited basal release of the soluble form of NRG-B1. It remains to be seen if NRG processing by ADAM19, or another metalloproteinase, is important in the regulation of dendrite outgrowth. ADAM19 deﬁcient mice have been generated and die perinatally of cardiac defects (106). However, no mention was made of nervous system defects. 10. Overall Summary Though somewhat limited, the data thus far argue convincingly that metalloproteinases play essential roles in the developing nervous system. Moreover, metalloproteinases may have late functions in events such as myelination (107). To determine which metalloproteinases are important, and what subset of molecules they regulate, future studies will need to focus on individual metalloproteinases. The variety of metalloproteinases expressed in developing brain, and the growing list of targets, illustrates the enormity of this task (1, 4). Fortunately, neural speciﬁcation, neuronal migration, neuronal survival, and axon guidance defects are associated with deﬁciencies in individual metalloproteinases, which means high speciﬁcity and limited redundancy are likely (6, 79, 84). Indeed, UNC-71 diﬀerentially regulates the guidance of diﬀerent subsets of ventral motoneurons in the worm, despite the fact that their cell bodies are close and their axon trajectories similar (6). How widespread will be the involvement of metalloproteinases in the processes of neural development remains to be determined. Acknowledgments The author would like to thank J. Hocking, R. Parker, K. AtkinsonLeadbeater and C. Webber for their helpful comments on this manuscript. A Canada Research Chair in Developmental Neurobiology, and the Alberta Heritage Foundation for Medical Research provided salary support for S. McFarlane. References 1. Kaczmarek, L., Lapinska-Dzwonek, J., and Szymczak, S. (2002) Embo J 21, 6643–6648

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

183

2. Yong, V. W., Power, C., Forsyth, P., and Edwards, D. R. (2001) Nat Rev Neurosci 2, 502–511 3. Schlondorﬀ, J., and Blobel, C. P. (1999) J Cell Sci 112 (Pt 21), 3603–3617 4. Novak, U. (2004) J Clin Neurosci 11, 227–235 5. Pan, D., and Rubin, G. M. (1997) Cell 90, 271–280 6. Huang, X., Huang, P., Robinson, M. K., Stern, M. J., and Jin, Y. (2003) Development 130, 3147–3161 7. Nishiwaki, K., Hisamoto, N., and Matsumoto, K. (2000) Science 288, 2205–2208 8. Karkkainen, I., Rybnikova, E., Pelto-Huikko, M., and Huovila, A. P. (2000) Mol Cell Neurosci 15, 547–560 9. Goddard, D. R., Bunning, R. A., and Woodroofe, M. N. (2001) Glia 34, 267–271 10. Bernstein, H. G., Keilhoﬀ, G., Bukowska, A., Ziegeler, A., Funke, S., Dobrowolny, H., Kanakis, D., Bogerts, B., and Lendeckel, U. (2004) J Neurosci Res 75, 353–360 11. Sun, Y. P., Deng, K. J., Wang, F., Zhang, J., Huang, X., Qiao, S., and Zhao, S. (2004) Gene 325, 171–178 12. Kurisaki, T., Masuda, A., Osumi, N., Nabeshima, Y., and Fujisawa-Sehara, A. (1998) Mech Dev 73, 211–215 13. Yavari, R., Adida, C., Bray-Ward, P., Brines, M., and Xu, T. (1998) Hum Mol Genet 7, 1161–1167 14. Giambernardi, T. A., Sakaguchi, A. Y., Gluhak, J., Pavlin, D., Troyer, D. A., Das, G., Rodeck, U., and Klebe, R. J. (2001) Matrix Biol 20, 577–587 15. Vaillant, C., Meissirel, C., Mutin, M., Belin, M. F., Lund, L. R., and Thomasset, N. (2003) Mol Cell Neurosci 24, 395–408 16. Frolichsthal-Schoeller, P., Vescovi, A. L., Krekoski, C. A., Murphy, G., Edwards, D. R., and Forsyth, P. (1999) Neuroreport 10, 345–351 17. Canete-Soler, R., Gui, Y. H., Linask, K. K., and Muschel, R. J. (1995) Dev Dyn 204, 30–40 18. Hayashita-Kinoh, H., Kinoh, H., Okada, A., Komori, K., Itoh, Y., Chiba, T., Kajita, M., Yana, I., and Seiki, M. (2001) Cell Growth Diﬀer 12, 573–580 19. Vaillant, C., Didier-Bazes, M., Hutter, A., Belin, M. F., and Thomasset, N. (1999) J Neurosci 19, 4994–5004 20. Itoh, T., Ikeda, T., Gomi, H., Nakao, S., Suzuki, T., and Itohara, S. (1997) J Biol Chem 272, 22389–22392 21. Masson, R., Lefebvre, O., Noel, A., Fahime, M. E., Chenard, M. P., Wendling, C., Kebers, F., LeMeur, M., Dierich, A., Foidart, J. M., Basset, P., and Rio, M. C. (1998) J Cell Biol 140, 1535–1541 22. Mudgett, J. S., Hutchinson, N. I., Chartrain, N. A., Forsyth, A. J., McDonnell, J., Singer, II, Bayne, E. K., Flanagan, J., Kawka, D., Shen, C. F., Stevens, K., Chen, H., Trumbauer, M., and Visco, D. M. (1998) Arthritis Rheum 41, 110–121 23. Shipley, J. M., Wesselschmidt, R. L., Kobayashi, D. K., Ley, T. J., and Shapiro, S. D. (1996) Proc Natl Acad Sci USA 93, 3942–3946 24. Wilson, C. L., Heppner, K. J., Labosky, P. A., Hogan, B. L., and Matrisian, L. M. (1997) Proc Natl Acad Sci USA 94, 1402–1407

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

184

S. McFarlane

25. Kheradmand, F., Rishi, K., and Werb, Z. (2002) J Cell Sci 115, 839–848 26. Holmbeck, K., Bianco, P., Caterina, J., Yamada, S., Kromer, M., Kuznetsov, S. A., Mankani, M., Robey, P. G., Poole, A. R., Pidoux, I., Ward, J. M., and Birkedal-Hansen, H. (1999) Cell 99, 81–92 27. Zhou, Z., Apte, S. S., Soininen, R., Cao, R., Baaklini, G. Y., Rauser, R. W., Wang, J., Cao, Y., and Tryggvason, K. (2000) Proc Natl Acad Sci USA 97, 4052–4057 28. Ducharme, A., Frantz, S., Aikawa, M., Rabkin, E., Lindsey, M., Rohde, L. E., Schoen, F. J., Kelly, R. A., Werb, Z., Libby, P., and Lee, R. T. (2000) J Clin Invest 106, 55–62 29. Rudolph-Owen, L. A., Hulboy, D. L., Wilson, C. L., Mudgett, J., and Matrisian, L. M. (1997) Endocrinology 138, 4902–4911 30. Page-McCaw, A., Serano, J., Sante, J. M., and Rubin, G. M. (2003) Dev Cell 4, 95–106 31. Vu, T. H., Shipley, J. M., Bergers, G., Berger, J. E., Helms, J. A., Hanahan, D., Shapiro, S. D., Senior, R. M., and Werb, Z. (1998) Cell 93, 411–422 32. Simpson, P. (1990) Trends Genet 6, 343–345 33. Artavanis-Tsakonas, S., Matsuno, K., and Fortini, M. E. (1995) Science 268, 225–232 34. Campos-Ortega, J. A. (1993) J Neurobiol 24, 1305–1327 35. Rooke, J., Pan, D., Xu, T., and Rubin, G. M. (1996) Science 273, 1227–1231 36. Rooke, J. E., and Xu, T. (1998) Bioessays 20, 209–214 37. Sawamoto, K., and Okano, H. (1996) Neurosci Res 26, 205–214 38. Chitnis, A., and Kintner, C. (1995) Perspect Dev Neurobiol 3, 3–15 39. Hall, R. J., and Erickson, C. A. (2003) Dev Biol 256, 146–159 40. Hartmann, D., de Strooper, B., Serneels, L., Craessaerts, K., Herreman, A., Annaert, W., Umans, L., Lubke, T., Lena Illert, A., von Figura, K., and Saftig, P. (2002) Hum Mol Genet 11, 2615–2624 41. Mishra-Gorur, K., Rand, M. D., Perez-Villamil, B., and Artavanis-Tsakonas, S. (2002) J Cell Biol 159, 313–324 42. Lieber, T., Kidd, S., and Young, M. W. (2002) Genes Dev 16, 209–221 43. Qi, H., Rand, M. D., Wu, X., Sestan, N., Wang, W., Rakic, P., Xu, T., and Artavanis-Tsakonas, S. (1999) Science 283, 91–94 44. Dong, J., Opresko, L. K., Dempsey, P. J., Lauﬀenburger, D. A., Coﬀey, R. J., and Wiley, H. S. (1999) Proc Natl Acad Sci USA 96, 6235–6240 45. Mechtersheimer, S., Gutwein, P., Agmon-Levin, N., Stoeck, A., Oleszewski, M., Riedle, S., Postina, R., Fahrenholz, F., Fogel, M., Lemmon, V., and Altevogt, P. (2001) J Cell Biol 155, 661–673 46. Brou, C., Logeat, F., Gupta, N., Bessia, C., LeBail, O., Doedens, J. R., Cumano, A., Roux, P., Black, R. A., and Israel, A. (2000) Mol Cell 5, 207–216 47. Mumm, J. S., Schroeter, E. H., Saxena, M. T., Griesemer, A., Tian, X., Pan, D. J., Ray, W. J., and Kopan, R. (2000) Mol Cell 5, 197–206 48. Lukashev, M., and Werb, Z. (1998) Trends Cell Biol 8, 437–441 49. Chen, R. H., and Chang, T. Y. (1997) Cell Growth Diﬀer 8, 821–827

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

185

50. Boudreau, N., Sympson, C. J., Werb, Z., and Bissell, M. J. (1995) Science 267, 891–893 51. Naus, S., Richter, M., Wildeboer, D., Moss, M., Schachner, M., and Bartsch, J. W. (2004) J Biol Chem 279, 16083–16090 52. Wetzel, M., Rosenberg, G. A., and Cunningham, L. A. (2003) Eur J Neurosci 18, 1050–1060 53. Raoul, C., Henderson, C. E., and Pettmann, B. (1999) J Cell Biol 147, 1049–1062 54. Weskamp, G., Schlondorﬀ, J., Lum, L., Becherer, J. D., Kim, T. W., Saftig, P., Hartmann, D., Murphy, G., and Blobel, C. P. (2004) J Biol Chem 279, 4241–4249 55. Gentry, J. J., Barker, P. A., and Carter, B. D. (2004) Prog Brain Res 146, 25–39 56. Lehmann, R. (2001) Curr Opin Genet Dev 11, 457–463 57. Murphy, G., and Gavrilovic, J. (1999) Curr Opin Cell Biol 11, 614–621 58. Blelloch, R., and Kimble, J. (1999) Nature 399, 586–590 59. Giannelli, G., Falk-Marzillier, J., Schiraldi, O., Stetler-Stevenson, W. G., and Quaranta, V. (1997) Science 277, 225–228 60. Duong, T. D., and Erickson, C. A. (2004) Dev Dyn 229, 42–53 61. Cai, D. H., Vollberg, T. M., Sr., Hahn-Dantona, E., Quigley, J. P., and Brauer, P. R. (2000) Anat Rec 259, 168–179 62. Alfandari, D., Wolfsberg, T. G., White, J. M., and DeSimone, D. W. (1997) Dev Biol 182, 314–330 63. Alfandari, D., Cousin, H., Gaultier, A., Smith, K., White, J. M., Darribere, T., and DeSimone, D. W. (2001) Curr Biol 11, 918–930 64. Krystosek, A., and Seeds, N. W. (1981) Science 213, 1532–1534 65. Pittman, R. N. (1985) Dev Biol 110, 91–101 66. McGuire, P. G., and Seeds, N. W. (1990) Neuron 4, 633–642 67. Muir, D. (1994) Exp Cell Res 210, 243–252 68. Nordstrom, L. A., Lochner, J., Yeung, W., and Ciment, G. (1995) Mol Cell Neurosci 6, 56–68 69. Chambaut-Guerin, A. M., Herigault, S., Rouet-Benzineb, P., Rouher, C., and Lafuma, C. (2000) J Neurochem 74, 508–517 70. Zuo, J., Ferguson, T. A., Hernandez, Y. J., Stetler-Stevenson, W. G., and Muir, D. (1998) J Neurosci 18, 5203–5211 71. Hawkins, R. L., and Seeds, N. W. (1989) Brain Res Dev Brain Res 45, 203–209 72. Schlosshauer, B., Walter, J., and Bonhoeﬀer, F. (1990) Neurosci Lett 113, 333–338 73. Sheﬃeld, J. B., Krasnopolsky, V., and Dehlinger, E. (1994) Dev Dyn 200, 79–88 74. Fambrough, D., Pan, D., Rubin, G. M., and Goodman, C. S. (1996) Proc Natl Acad Sci USA 93, 13233–13238 75. Galko, M. J., and Tessier-Lavigne, M. (2000) Science 289, 1365–1367 76. Hattori, M., Osterﬁeld, M., and Flanagan, J. G. (2000) Science 289, 1360–1365

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

186

S. McFarlane

77. Roskies, A., Friedman, G. C., and O’Leary, D. D. (1995) Perspect Dev Neurobiol 3, 63–75 78. Drescher, U., Bonhoeﬀer, F., and Muller, B. K. (1997) Curr Opin Neurobiol 7, 75–80 79. Schimmelpfeng, K., Gogel, S., and Klambt, C. (2001) Mech Dev 106, 25–36 80. Guthrie, S. (2001) Curr Biol 11, R300–303 81. Brenner, S. (1974) Genetics 77, 71–94 82. McIntire, S. L., Garriga, G., White, J., Jacobson, D., and Horvitz, H. R. (1992) Neuron 8, 307–322 83. Siddiqui, S. S., and Culotti, J. G. (1991) J Neurogenet 7, 193–211 84. Leighton, P. A., Mitchell, K. J., Goodrich, L. V., Lu, X., Pinson, K., Scherz, P., Skarnes, W. C., and Tessier-Lavigne, M. (2001) Nature 410, 174–179 85. Sagane, K., Yamazaki, K., Mizui, Y., and Tanaka, I. (1999) Gene 236, 79–86 86. Webber, C. A., Hocking, J. C., Yong, V. W., Stange, C. L., and McFarlane, S. (2002) J Neurosci 22, 8091–8100 87. O’Leary, D. D., and Wilkinson, D. G. (1999) Curr Opin Neurobiol 9, 65–73 88. Nakagawa, S., Brennan, C., Johnson, K. G., Shewan, D., Harris, W. A., and Holt, C. E. (2000) Neuron 25, 599–610 89. McFarlane, S., McNeill, L., and Holt, C. E. (1995) Neuron 15, 1017–1028 90. McFarlane, S., Cornel, E., Amaya, E., and Holt, C. E. (1996) Neuron 17, 245–254 91. Deiner, M. S., Kennedy, T. E., Fazeli, A., Seraﬁni, T., Tessier-Lavigne, M., and Sretavan, D. W. (1997) Neuron 19, 575–589 92. Levi, E., Fridman, R., Miao, H. Q., Ma, Y. S., Yayon, A., and Vlodavsky, I. (1996) Proc Natl Acad Sci USA 93, 7069–7074 93. Deiner, M. S., and Sretavan, D. W. (1999) J Neurosci 19, 9900–9912 94. de la Torre, J. R., Hopker, V. H., Ming, G. L., Poo, M. M., Tessier-Lavigne, M., Hemmati-Brivanlou, A., and Holt, C. E. (1997) Neuron 19, 1211–1224 95. Diaz-Rodriguez, E., Cabrera, N., Esparis-Ogando, A., Montero, J. C., and Pandiella, A. (1999) Eur J Neurosci 11, 1421–1430 96. Jung, J. C., Leco, K. J., Edwards, D. R., and Fini, M. E. (2002) Dev Dyn 223, 402–413 97. Nath, D., Slocombe, P. M., Webster, A., Stephens, P. E., Docherty, A. J., and Murphy, G. (2000) J Cell Sci 113 (Pt 12), 2319–2328 98. Galvez, B. G., Matias-Roman, S., Yanez-Mo, M., Sanchez-Madrid, F., and Arroyo, A. G. (2002) J Cell Biol 159, 509–521 99. Yan, Y., Shirakabe, K., and Werb, Z. (2002) J Cell Biol 158, 221–226 100. Szklarczyk, A., Lapinska, J., Rylski, M., McKay, R. D., and Kaczmarek, L. (2002) J Neurosci 22, 920–930 101. Cronin, J., and Dudek, F. E. (1988) Brain Res 474, 181–184 102. Sekine-Aizawa, Y., Hama, E., Watanabe, K., Tsubuki, S., Kanai-Azuma, M., Kanai, Y., Arai, H., Aizawa, H., Iwata, N., and Saido, T. C. (2001) Eur J Neurosci 13, 935–948 103. Longart, M., Liu, Y., Karavanova, I., and Buonanno, A. (2004) J Comp Neurol 472, 156–172

May 26, 2005

16:3 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch06

Metalloproteinases in Development

187

104. Gassmann, M., and Lemke, G. (1997) Curr Opin Neurobiol 7, 87–92 105. Shirakabe, K., Wakatsuki, S., Kurisaki, T., and Fujisawa-Sehara, A. (2001) J Biol Chem 276, 9352–9358 106. Zhou, H. M., Weskamp, G., Chesneau, V., Sahin, U., Vortkamp, A., Horiuchi, K., Chiusaroli, R., Hahn, R., Wilkes, D., Fisher, P., Baron, R., Manova, K., Basson, C. T., Hempstead, B., and Blobel, C. P. (2004) Mol Cell Biol 24, 96–104 107. Oh, L. Y., Larsen, P. H., Krekoski, C. A., Edwards, D. R., Donovan, F., Werb, Z., and Yong, V. W. (1999) J Neurosci 19, 8464–8475 108. Cornwall, G. A., and Hsia, N. (1997) Endocrinology 138, 4262–4272 109. Schlomann, U., Rathke-Hartlieb, S., Yamamoto, S., Jockusch, H., and Bartsch, J. W. (2000) J Neurosci 20, 7964–7971 110. Weskamp, G., Cai, H., Brodie, T. A., Higashyama, S., Manova, K., Ludwig, T., and Blobel, C. P. (2002) Mol Cell Biol 22, 1537–1544 111. Rybnikova, E., Karkkainen, I., Pelto-Huikko, M., and Huovila, A. P. (2002) Neuroscience 112, 921–934 112. Kurisaki, T., Masuda, A., Sudo, K., Sakagami, J., Higashiyama, S., Matsuda, Y., Nagabukuro, A., Tsuji, A., Nabeshima, Y., Asano, M., Iwakura, Y., and Sehara-Fujisawa, A. (2003) Mol Cell Biol 23, 55–61 113. Horiuchi, K., Weskamp, G., Lum, L., Hammes, H. P., Cai, H., Brodie, T. A., Ludwig, T., Chiusaroli, R., Baron, R., Preissner, K. T., Manova, K., and Blobel, C. P. (2003) Mol Cell Biol 23, 5614–5624 114. Marcinkiewicz, M., and Seidah, N. G. (2000) J Neurochem 75, 2133–2143 115. Yoshinaka, T., Nishii, K., Yamada, K., Sawada, H., Nishiwaki, E., Smith, K., Yoshino, K., Ishiguro, H., and Higashiyama, S. (2002) Gene 282, 227–236 116. Gunn, T. M., Azarani, A., Kim, P. H., Hyman, R. W., Davis, R. W., and Barsh, G. S. (2002) BMC Genet 3, 2 117. Ahokas, K., Lohi, J., Lohi, H., Elomaa, O., Karjalainen-Lindsberg, M. L., Kere, J., and Saarialho-Kere, U. (2002) Gene 301, 31–41

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

This page intentionally left blank

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

CHAPTER 7 MATRIX METALLOPROTEINASES IN MYELIN FORMATION

P.H. Larsen & V.W. Yong∗ Hotchkiss Brain Institute, 3330 Hospital Drive, University of Calgary, Calgary, Alberta, Canada T2N 4N1 E-mail: ∗[email protected]

Emerging evidence demonstrate that matrix metalloproteinase (MMP) activity has many beneﬁcial properties during development and also after injuries. Examples include the observations that angiogenesis (1–3) and the migration of neuronal and oligodendrocyte precursor cells (4, 5) require metalloproteinase activity. A growing list has ascribed novel functions of metalloproteinases in normal development, such as in cytokine activation/inactivation and controlling apoptotic pathways (reviewed in (6)). These ﬁndings have left us with a more complex picture of the functional roles metalloproteinases have in diﬀerent cell types during developmental processes. Early during postnatal development of the central nervous system axons become myelinated. Investigations of this event have implicated a role for proteolytic activity in oligodendrocyte (OL) process extension and, furthermore, MMP mutant animals demonstrate a failure in the maturation of OLs and myelination. Other than developmental processes, MMP activity has also been implicated in remyelination after an injury in adulthood. Altogether, these studies suggest that MMPs are important in various aspects of developmental myelination and remyelination. These data and the beneﬁcial versus detrimental properties of MMPs during injury will be discussed in this chapter.

1. Regulation of Myelin Formation Myelination is a developmentally regulated event that begins postnatally in the rat and mouse CNS. At birth, oligodendrocyte precursor cells (OPCs) have already migrated to the vicinity of the axons that are to be myelinated, and around postnatal day 5, the last round of OPC proliferation is seen and initiation of the myelination program commences. This includes Correspondence to: V.W. Yong 189

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

190

P.H. Larsen & V.W. Yong

diﬀerentiation of OPCs into myelin producing cells, process outgrowth, axonal contact and spiralling of the OL processes around the axon (7, 8). A whole battery of molecules is involved in the diﬀerentiation and survival of OLs (reviewed in (9)). Some of these are growth factors {e.g. platelet-derived growth factor (PDGF), ﬁbroblast growth factor (FGF) and insulin-like growth factor-1 (IGF-1)} but signalling pathways such as Notch/Jagged are also important for the accurate development of the OL lineage (10–12). The roles of PDGF and FGF-2 have been extensively studied with respect to proliferation, diﬀerentiation and survival of OLs. In OPCs, PDGF acts as a mitogen, and it also enhances their diﬀerentiation and survival (9, 13). As a mitogen, PDGF is optimal in combination with other factors, such as FGF-2 (14). FGF-2 is a potent mitogen for the more mature pro-OLs and, besides being a mitogen, it prevents their diﬀerentiation into mature OLs (14–16). Insulin and IGFs are also potent regulators of OL development. IGF-1 increases the proliferation of OPCs and the number of mature OLs (17). Furthermore, overexpression of IGF-1 inhibits natural apoptosis in OPCs and gives rise to an overproduction of OLs (18). Conversely, IGF-1 null mice have less myelin and a proportional reduction in brain weight and number of axons (19, 20), indicating a role for IGF-1 in myelination and OL survival in vivo. In addition, IGF signalling has been shown to be required for the survival, proliferation, and diﬀerentiation of OPCs during remyelination (21). It is important to note that the bioavailability and actions of IGF-1 are tightly regulated by six high-aﬃnity IGF binding proteins (IFGBPs), which can either inhibit or promote IGF-1 actions (22–26). The presence of IGFBP-1 and -2 is known to reduce survival and diﬀerentiation of OPCs both in the absence and presence of IGF-1 (27). Likewise, IGFBP-6 is developmentally regulated in OL cultures where it decreases during early diﬀerentiation but increases by about 250% at the end of diﬀerentiation on day 7 (25). This coincides with the increased need for IGF-1 to stimulate survival during the ﬁrst 5 days of OL cell maturation. In addition, IGFBP-6 added exogenously to cultures leads to a reduction in OPC survival and expression of myelin proteins (25). MMPs are capable of degrading or processing a wide selection of growth factors (28). These properties allow MMPs to have important functions in many developmental processes including the formation of the CNS; for this reason, it is interesting that MMPs have been shown to degrade IGFBPs and thereby release bioactive IGF-1 and -2 (29, 30).

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

Matrix Metalloproteinases in Myelin Formation

191

Altogether, it seems important to address whether MMPs could play a role in myelination by regulating growth factors such as those described above, and through other mechanisms. Furthermore, when an OL extends its processes early in the myelinogenic program, extensive arborisation occurs and we have postulated that proteolytic activity is required to remodel the surrounding extracellular matrix (ECM) to allow the OL processes to come into contact with axons that are to be myelinated. 2. Requirement of MMPs in Myelin Formation 2.1. MMP-9 in OL process extension It is known that neurons are dependent on metalloprotease activity for neurite elongation through ECM proteins (31–33). For that reason, it was postulated that OLs may utilise an analogous proteolytic mechanism for process extension (34). Indeed, data from our laboratory indicated that MMP-9 was elevated during the early phases of myelination in the corpus callosum and the optic nerve (34, 35). Furthermore, the correlation between increased process extension and MMP-9 protein levels was associated with protein kinase C activation (34, 35). Using in situ zymography, proteolytic activity was detected on OLs where it was localised in the cell soma and on the tip of growing processes (35). Furthermore, these studies showed that applying MMP-9 function blocking antibodies to OL cultures reduced their ability to extend processes. Similarly, data from MMP-9 null mice demonstrated a 50% reduction in OL process extension in cells cultured from null mice compared to wildtype animals (35). These data clearly suggest a role for MMP-9 in OL process outgrowth. 2.2. MMP-12 and its role in OL biology MMP-12, which is also known as macrophage metalloelastase, has most often been studied in vascular diseases and lung inﬂammation, where an increase in MMP-12 has been associated with disease progression (36, 37). MMP-12 has also been shown to play a role in macrophage inﬁltration after emphysema (38) and leukocyte and eosinophil recruitment to the lung (39). With respect to the CNS, the literature on MMP-12 is sparse. A study by Vos and colleagues reported that MMP-12 is present in active MS lesions where its expression was localised to phagocytic macrophages (40). Recently, a favourable role for MMP-12 has been described in experimental autoimmune encephalomyelitis (EAE), an animal model of MS. In that

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

192

P.H. Larsen & V.W. Yong

study, MMP-12 null mice exhibited more severe disease scores than wildtype controls indicating that MMP-12 plays a beneﬁcial role in EAE (Weaver et al, submitted). While most studies indicate that cells of monocytoid lineage (i.e. macrophages and microglia) are responsible for MMP-12 expression, we have demonstrated that OLs express this protein in large amounts (41). Furthermore, we showed that MMP-12 has functional roles for both the maturation and morphological diﬀerentiation (process extension) of cells of the OL lineage (41). Our data has addressed the functional importance of MMP-12 activity for OL development and shows that MMP-12 has a similar function on OL process extension as observed for MMP-9. Finally, 3 antibodies targeted against diﬀerent domains of MMP-12 demonstrated function blocking eﬀects and process extension by OLs was signiﬁcantly reduced (41). Investigations of the number of mature OLs present after 48 hours in maturation experiments showed a signiﬁcant reduction in the number of mature cells isolated from MMP-12 null mice versus wildtype mice. Further investigations of the role of MMP-12 on OL maturation were conducted by examining morphological features. Using this approach, we found signiﬁcantly fewer cells with highly branched processes in the MMP-12 null cultures compared to wildtype cells (Fig. 1), and this could be rescued by the exogenous administration of MMP-12 protein to the null cultures (41). Overall, as with MMP-9, MMP-12 is needed for process extension and furthermore plays a role in the maturation of OLs.

Fig. 1. MMP-12 regulates the morphological diﬀerentiation of OLs. As cells of the OL lineage mature, their degree of process arborisation becomes increasingly complex. Panel A shows that the processes of OLs from a wildtype culture are highly branched while cells from MMP-12 null cultures have less extensive arborisation (B) (Larsen and Yong, 2004).

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

Matrix Metalloproteinases in Myelin Formation

193

2.3. MMP-9 and -12 in developmental myelination The functional signiﬁcance of MMP-9 and -12 in developmental myelination has been addressed in vivo. Our studies show that MMP-9 and MMP-12 transcripts are up-regulated in the optic nerve at periods correlating with myelination. The elevation of MMP-9 transcripts is transient while that for MMP-12 is more prolonged. Further evidence that this up-regulation of MMPs is important comes from immunohistochemistry and western blot analysis of myelin basic protein (MBP), which indicates that there is a delay in the formation of myelin in animals deﬁcient in MMP-9 and -12. This delay was already observed at postnatal day 7 (P 7) and signiﬁcant in all mutant genotypes at P 10; however, the impairment was transient as all genotypes exhibited similar MBP immunostaining at P 14 (Larsen and Yong, submitted). In order to demonstrate that these observations were not a result of delay in overall development, axonal proﬁles were compared. There were no diﬀerences among genotypes in neuroﬁlament immunoreactivity, indicating that the observed delay in myelination was not due to reduced axonal populations. The mechanisms for the delay in myelination in MMP-9 and -12 null mice were examined. Investigations of OL cell numbers in MMP-9 and -12 deﬁcient mice revealed a 50% reduction in the number of mature OLs compared to wildtype controls. As there was no diﬀerence in PDGFαR positive precursor cell counts, this suggests that OL precursor cells were not diﬀerentiating into mature myelinating OLs in appropriate numbers for developmental myelination to proceed at a normal rate. Thus, MMP-9 and -12 aﬀects a mechanism that results in OL maturation. Several molecules are involved in the maturation of OLs and some of these can act as substrates for MMP activity. As mentioned above, IGF-1 is a potent diﬀerentiation factor for oligodendrocyte precursor cells and is critical for myelination (18–20, 42, 43). Since the bioavailability of IGF-1 is regulated by IGFBPs, and as MMPs have been reported to cleave IGFBPs without degrading bound IGF-1 (29, 30), this could be a common strategy in which MMPs control the bioavailability of IGF-1. Our data reveal that MMP-9 and -12 deﬁcient animals have a signiﬁcant higher level of IGFBP-6 at P 10 compared to wildtype mice, correspondent with the period of a delay in myelination (Larsen and Yong, submitted). This suggests that MMP-9 and -12 are important in regulating IGFBP-6. The lack of MMP-9 or -12 could result in excessive amounts of IGFBP-6, which would then limit the access of IGF-1 to drive OL diﬀerentiation and myelination.

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

194

P.H. Larsen & V.W. Yong

Redundant mechanisms may account for the transient phenotypes observed during myelination. It is possible that other MMP members or members of other proteinase families can substitute for MMP-9 and MMP-12 in these mutant animals. MMP null mutations have generally produced subtle phenotypes most of which are transient. For example, MMP-9 null mice show a transient delay in skeletal growth plate and ossiﬁcation due to a delayed release of an angiogenic activator, vascular endothelial cell growth factor (2). Similarly, MMP-9 null mice exhibit a transient delay in osteoclast recruitment (44). With respect to neovascularisation, MMP-9 null mice showed reduced levels of choroidal neovascularisation (45); however, this was more severely aﬀected by the simultaneous deletion of MMP-9 and MMP-2 (46), suggesting that MMPs can have redundant functions to some degree. In addition, a novel report demonstrates that MMP-2 and -14 exerts redundant mechanisms as double mutant mice cannot survive early postnatal life in contrast to what is demonstrated for either of the single mutant mice (47). These observations indicate that careful studies have to be performed in order to uncover novel functions of MMPs. In conclusion, MMP-9 and -12 have been investigated for their involvement during myelination. An up-regulation of both these MMPs is detected during myelin formation. Both MMPs show eﬀects on OL process extension and maturation in vitro. Examination of the corpus callosum showed a transient delay in the developmental myelination in mice deﬁcient for MMP-9 and -12 due to a reduced number of mature OLs. Furthermore, an increase in IGFBP-6 protein was found in MMP-9 and -12 deﬁcient animals at a time when myelination was delayed suggesting important roles for MMP-9 and -12 in regulating the concentration of IGFBP-6 during initial developmental myelination. Altogether, MMP activity seems critical for the timing of OL maturation and myelin formation (Fig. 2).

3. MMPs in Remyelination 3.1. MMP-9 has a role in the remyelination process Various types of CNS insults lead to demyelination and/or axon loss, resulting in disorders such as multiple sclerosis (MS). The CNS does attempt recovery from such insults, and remyelination occurs in MS lesions, albeit in limited extent (48, 49). Thus, if the intrinsic CNS repair process could be enhanced, remyelination might be facilitated. While several factors contribute to remyelination, of paramount importance are the recruitment and

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

Development

Matrix Metalloproteinases in Myelin Formation

MMP-9 MMP-12 Maturation Process extension

MMP-9 MMP-12

195

IGFBP-6 IGF

Myelin formation

Fig. 2. Summary of the role of MMP-9 and -12 in regulating myelin formation in development. Both MMP-9 and MMP-12 play important roles for the maturation and process extension of OLs and subsequent myelination is partly controlled by MMP activity.

Fig. 3. Remyelination after a lysolecithin injury is deﬁcient in MMP-9 null mice. Panel (A) shows a lysolecithin induced injury and the production of extensive demyelination in the dorsal column of the spinal cord at 1 week post-injury. By 2 weeks of the lysolecithin insult, remyelination has ensued in wildtype mice (B) but not in MMP-9 null mice (C). Myelin was revealed using luxol fast blue, and the slides were additionally subjected to hematoxylin and eosin stain.

maturation of OL precursors and the ensheathment of axons by their processes (50). In this regard, proteases that are capable of remodelling the CNS matrix to allow for precursor cell migration and for the elongation of OL processes are likely to be critical. Since MMP-9 was shown to be important for developmental myelin formation, we asked whether MMP-9 would play a role in remyelination following an insult to adult animals. Indeed, in chemically-induced demyelination insults caused by lysolecithin, our data show that MMP-9 null mice have an impaired ability to remyelinate compared to that seen in wildtype mice (Fig. 3) (51).

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

196

P.H. Larsen & V.W. Yong

3.2. MMP-9 regulates remyelination in part by processing the inhibitory NG2 proteoglycan The ability to remyelinate after a demyelinating injury depends on a variety of factors, including the inﬂux of mononuclear phagocytes to clear myelin debris, the responsiveness of axons, a favourable extracellular environment including the ECM, and the availability of progenitor cells that diﬀerentiate into mature OLs to initiate the remyelinating program. Added to this list is the presence of proteolytic activity, since the absence of MMP-9 impairs the remyelinating program after lysolecithin-induced demyelination. MMP-9 does not appear to be required for the initial entry of phagocytes into the CNS, since equivalent numbers of mononuclear phagocytes are present in the lesions of both wildtype and MMP-9 null mice, and because the extent of demyelination is comparable between both groups at 1 week after injury (51). However, a decline in the number of mature OLs in lesion areas was observed in the MMP-9 null mice compared to wildtype controls. This did not appear to be due to an intrinsic deﬁciency of OLs in MMP-9 null mice, as tissue away from the lesion in MMP-9 deﬁcient animals exhibited OL densities similar to those observed in wildtype counterparts (51). The MMPs should be considered as candidate molecules for enabling remyelination, as they can degrade all protein components of the ECM (28). Although MMPs are known to have detrimental roles after injury (52), their consistent up-regulation in the damaged CNS invites the hypothesis that these proteases have subtle but important functions in the repair process, particularly in remyelination. Upon injury to the CNS, various ECM components such as proteoglycans are up-regulated at the lesion sites. For example, the NG2 chondroitin sulfate proteoglycan accumulates as a result of both axonal injury (53, 54) and demyelinating insults, including in MS (55, 56). This proteoglycan has also been found to have inhibitory eﬀects on axon elongation (57, 58). Examination of NG2 expression upon the demyelinating injury showed an intense ECM-associated NG2 immunoreactivity in the injury site. The dense NG2 accumulation seen in the dorsal column of both wildtype and MMP-9 null mice at 1 week after injury was resolved in wildtype mice by 2 weeks postinjury, leaving NG2 associated principally with progenitor cells. However, dense NG2 deposits remained in the lesions of MMP-9 null mice (51). The persistence of NG2 immunoreactivity in MMP-9 deﬁcient mice prompted the investigation of NG2 as a substrate for MMP-9 proteolysis. Previously, MMP-9 was found capable of binding to the core protein of CSPGs (59). We extended this observation by demonstrating that the

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

Matrix Metalloproteinases in Myelin Formation

197

incubation of NG2 with MMP-9 resulted in new fragments whose formation could be blocked by a metalloproteinase inhibitor. Furthermore, the subtle processing of NG2 by MMP-9 is suﬃcient to alter its inhibitory eﬀect on OL maturation in vitro. Another potential explanation for the lack of remyelination could be that NG2 is inhibiting OL process extension since NG2 is known to inhibit neuronal process extension. We did investigate this potential mechanism and found that NG2 did not aﬀect OL process extension in vitro (51). Altogether, these results demonstrate that NG2 accumulates following injury; that MMP-9 activity is required to clear the NG2 deposition, and that MMP-9 processing overcomes the negative impact of NG2 on OL maturation and remyelination. The current result that NG2 is a substrate for MMP-9 is of relevance not only to remyelination, but also to the recovery from various types of CNS insults in which proteoglycans are deposited in lesions. This work suggests that when MMP-9 is up-regulated upon CNS injury, some of its activity may be aimed at degrading inhibitory ECM molecules to enable axonal regeneration and remyelination. The OL maturation is inﬂuenced primarily by the NG2 core protein, since NG2 with or without the chondroitin sulfate side chain inhibited OL maturation in vitro. In conclusion, we have deﬁned for the ﬁrst time a useful property of an MMP member following CNS insult. These results show that MMP-9 plays a signiﬁcant role in clearing the NG2 chondroitin sulfate proteoglycan, and this then allows OPCs to mature and diﬀerentiate into myelin-forming OLs at the site of injury.

4. MMPs in Demyelinating Diseases MMPs have been suggested to play an important role in diﬀerent CNS diseases such as Alzheimer’s disease, stroke, MS and gliomas (52). With respect to demyelinating disorders such as MS and its experimental model, EAE, evidence is emerging that MMPs may be involved in the disruption of the blood brain barrier. One of the hallmarks of both MS and EAE is the presence of activated T -cells that can synthesise MMPs and, via secreted cytokines, induce their production by other inﬂammatory and CNS resident cells. Also, the breakdown of the blood brain barrier in MS, which results in leakage of plasma proteins into the CNS parenchyma, may inﬂuence MMP activities through integrin-mediated mechanisms. Furthermore, proteases including MMPs have long been known to participate in CNS demyelination

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

198

P.H. Larsen & V.W. Yong

(60–64). It has been demonstrated that proteases capable of catalysing the degradation of myelin are present in the cerebrospinal ﬂuid (CSF) of MS patients and in EAE (65); some of these have been identiﬁed as MMPs. Both MMP-1, -2, -3, -7, -9 and -12 all have been shown to degrade MBP and some can produce encephalitogenic products (66–68). Several MMPs such as MMP-2, -3, -7 and -9 are expressed in and around MS plaques (65, 69, 70). In acute MS lesions MMP-2, -7 and -9 have been found to be up-regulated in microglia/macrophages (71). In actively demyelinating MS lesions, strong activity for MMP-7 has been detected (70). Studies on CSF of MS patients have revealed a selective elevation of MMP-9 whereas MMP-2, -3 and -7 were not elevated or not detectable (72). Likewise, elevated levels of MMP-9 (73) and MMP-7 (74, 75) have been detected in the CSF during EAE. In another demyelinating disease, experimental autoimmune neuritis (EAN), an animal model of Guillain-Barr´e syndrome, MMP-9 and MMP-7 have been found to be up-regulated during worsening of the clinical course of the disease, whereas the mRNA expression level of other MMPs (MMP-2, -3, -11 and -13) was not altered (76). Another study has found MMP-7, -9 and -12 mRNA levels to be up-regulated during EAN (77). The inhibition of activated MMPs using broad-spectrum inhibitors has been shown to reverse ongoing EAE in rats (78, 79) and produce beneﬁcial eﬀects in chronic relapsing MS. The mechanism of action of MMP inhibitors in EAE and MS may be due to the prevention of inﬂammatory cells from crossing the basement membrane or ECM barrier that surrounds cerebral endothelium. Indeed, it has been shown that the inhibition of MMP-9 activity of T -cells is a major mechanism of action of interferon-β, a drug used clinically in MS (80, 81). Thus, there is considerable evidence suggesting involvement of the MMPs in the pathogenesis of both the inﬂammatory and demyelinating components of MS lesions. The elevation of MMPs alone, however, does not necessary reﬂect that they are harmful. MMPs may also be involved in a series of beneﬁcial eﬀects after an insult. As already described, a null mutation in the MMP-12 gene renders mice to fare worse after induction of EAE, suggesting that MMP-12 is playing a beneﬁcial role (Weaver et al, submitted). In other aspects of CNS diseases MMPs have been suggested to have beneﬁcial roles. For example, the formation of new blood vessels is required to facilitate recovery from various CNS insults and MMPs may play a role in angiogenesis (2). Furthermore, it is possible that several neural progenitor cell types require MMPs to migrate into lesion sites to replenish lost cells and to attempt repair. In

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

Matrix Metalloproteinases in Myelin Formation

199

addition, MMPs could act as guidance proteins while they are very tightly regulated and while they are present on the tip of OLs when they extend their processes (see above, (35)). As mentioned earlier, MMP-9 and -12 have important roles in developmental myelin formation and MMP-9 (MMP-12 has not been tested) facilitates remyelination. 5. MMPs: Detrimental and Beneﬁcial After Injury Evaluating the beneﬁcial versus detrimental properties of MMPs after a CNS insult is no easy task, complicated by the fact that not only do several diﬀerent cell types express MMPs but the timing of their expression after an initial insult is also variable. Historically, the over-expression of MMPs in the injured adult CNS has largely been considered to be harmful to the tissue (52). However, although MMPs clearly have detrimental roles, several lines of investigation now suggest that MMPs have beneﬁcial functions when expressed in the CNS. For example, metalloproteinase activity has been proposed to regulate axon elongation of cultured neurons (82, 83) as well as play a role in modulating axon guidance cues (84, 85) in vitro. Furthermore, Webber and colleagues have shown that the application of MMP inhibitors during development results in a decrease in axon elongation and misguidance of the retinotectal tract in Xenopus (86). While the above cited literature promotes the beneﬁcial aspects of MMPs this must be balanced with reported detrimental eﬀects. In this regard, MMPs have been shown to be directly neurotoxic (87), and they produce axonal injury (88) as well as demyelination (71). In addition, in studies involving MMP-9 null mice, both brain (89) and spinal cord (90) injuries have resulted in fewer deﬁcits in functional outcome compared to wildtype animals. Similarly, MMP-12 null mice have shown better recovery rates and less secondary damage after spinal cord injury, indicating deleterious eﬀects of MMP-12 in response to injury of the spinal cord (91). In the same way, dysfunctions of MMPs are thought to contribute to the pathology of various diseases that include MS, stroke and gliomas; in these conditions, MMPs are thought to impair blood-brain barrier dysfunction, promote inﬂammation, and produce neurotoxicity (52, 92–95). Clearly, the complex roles of MMPs in the developing or injured CNS require further attention. The spatial and temporal expression of speciﬁc MMP family members by identiﬁed cell types and the interaction of MMPs with other molecules present at that precise location and time are critical factors that could determine the beneﬁcial or detrimental functions of MMPs. The use of non-speciﬁc

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

200

P.H. Larsen & V.W. Yong

inhibitors of MMPs to treat diseases, while important in alleviating the harmful properties of MMPs, may also impair the beneﬁcial functions of MMPs in the long term. The future of MMP inhibition in therapeutics may be one in which very speciﬁc inhibitors to particular MMP members are used at particular times, or it may be an approach to identify and modulate particular substrates that detrimental MMPs are acting upon.

6. Conclusion Overall, MMP-9 and -12 seem to have overlapping functions in various aspects of OL biology. There must be diﬀerential roles, too, since MMP-12 did not substitute for MMP-9 in MMP-9 null after lysolecithin injury or during development, and vice versa. We have not studied all members of the MMP family and it is very likely that other members not yet examined have functions in OL biology and myelin formation. Similarly, it is possible that other proteinase families have roles in myelination. For example, it has not been addressed whether ADAMs (A disintegrin and metalloproteinase family) or ADAM with thrombospondin modules (ADAM-TS) are involved in the process of myelination. A large number of these proteinases, which have similarities to MMPs in their domain structure, are expressed in the CNS during development (96–99). They are known to be involved in ectodomain shedding and they degrade many of the same substrates as MMPs, including proteoglycans (100, 101). Furthermore, important developmental processes are attributed to ADAMs. For example, ADAM-10 and -17 are known to process the Notch receptor (102–104), which have critical importance for fetal development. Interestingly, a recent study shows that ADAM-28 is activated by MMP-7 to degrade IGFBP-3 (105). This creates the idea that ADAMs similarly to MMPs could play a role in controlling the bioavailability of IGF-1 and -2; thus, investigations into their expression in OLs may be of interest. Furthermore, ADAM-12 has recently been detected in the rat and human brain where it is conﬁned exclusively to OLs (106). Overall, emerging evidence suggests that some ADAMs and ADAM-TSs could have roles for OL biology and CNS biology in general. In conclusion, we have in this chapter discussed the importance for MMP-9 and -12 in developmental myelination and furthermore we have shown how MMP-9 is critical during remyelination. The mechanisms contribute, at least in part, to both the developmental delay and the recovery after a demyelinating insult.

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

Matrix Metalloproteinases in Myelin Formation

201

This work has raised the awareness that MMPs can have beneﬁcial properties not only during developmental myelination but also for remyelination after a demyelinating insult. Can we use this knowledge to promote remyelination? As more research is performed, we can start to elucidate more completely the mechanisms involved and use this to promote remyelination. It is important to note that excess MMP activity can result in tissue destruction and therefore MMPs themselves cannot be injected when trying to promote remyelination. However, by exploiting the mechanisms of how MMP activity beneﬁts tissue remodelling after wound healing, it is possible that these may be used to promote remyelination in the CNS after injury. Altogether, these studies open new doors for further investigation into aspects of MMP involvement during myelin formation in the CNS. References 1. Canete Soler, R., Gui, Y. H., Linask, K. K., and Muschel, R. J. (1995) Brain Res Dev Brain Res 88, 37–52 2. Vu, T. H., Shipley, J. M., Bergers, G., Berger, J. E., Helms, J. A., Hanahan, D., Shapiro, S. D., Senior, R. M., and Werb, Z. (1998) Cell 93, 411–422 3. Werb, Z., Vu, T. H., Rinkenberger, J. L., and Coussens, L. M. (1999) Apmis 107, 11–18 4. Vaillant, C., Meissirel, C., Mutin, M., Belin, M. F., Lund, L. R., and Thomasset, N. (2003) Mol Cell Neurosci 24, 395–408 5. Amberger, V. R., Avellana-Adalid, V., Hensel, T., Baron-van Evercooren, A., and Schwab, M. E. (1997) Eur J Neurosci 9, 151–162 6. Sternlicht, M. D., and Werb, Z. (2001) Annu Rev Cell Dev Biol 17, 463–516 7. Skoﬀ, R. P., Price, D. L., and Stocks, A. (1976) J Comp Neurol 169, 291–312 8. Dangata, Y. Y., and Kaufman, M. H. (1997) Eur J Morphol 35, 3–17 9. Grinspan, J. (2002) J Neuropathol Exp Neurol 61, 297–306 10. Givogri, M. I., Costa, R. M., Schonmann, V., Silva, A. J., Campagnoni, A. T., and Bongarzone, E. R. (2002) J Neurosci Res 67, 309–320 11. John, G. R., Shankar, S. L., Shaﬁt-Zagardo, B., Massimi, A., Lee, S. C., Raine, C. S., and Brosnan, C. F. (2002) Nat Med 8, 1115–1121 12. Wang, S., Sdrulla, A. D., diSibio, G., Bush, G., Nofziger, D., Hicks, C., Weinmaster, G., and Barres, B. A. (1998) Neuron 21, 63–75 13. Calver, A. R., Hall, A. C., Yu, W. P., Walsh, F. S., Heath, J. K., Betsholtz, C., and Richardson, W. D. (1998) Neuron 20, 869–882 14. McKinnon, R. D., Matsui, T., Dubois-Dalcq, M., and Aaronson, S. A. (1990) Neuron 5, 603–614 15. Gard, A. L., and Pfeiﬀer, S. E. (1993) Dev Biol 159, 618–630 16. Baron, W., Metz, B., Bansal, R., Hoekstra, D., and de Vries, H. (2000) Mol Cell Neurosci 15, 314–329

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

202

P.H. Larsen & V.W. Yong

17. McMorris, F. A., and Dubois-Dalcq, M. (1988) J Neurosci Res 21, 199–209 18. Barres, B. A., Schmid, R., Sendnter, M., and Raﬀ, M. C. (1993) Development 118, 283–295 19. Cheng, C. M., Joncas, G., Reinhardt, R. R., Farrer, R., Quarles, R., Janssen, J., McDonald, M. P., Crawley, J. N., Powell-Braxton, L., and Bondy, C. A. (1998) J Neurosci 18, 5673–5681 20. Ye, P., Li, L., Richards, R. G., DiAugustine, R. P., and D’Ercole, A. J. (2002) J Neurosci 22, 6041–6051 21. Mason, J. L., Xuan, S., Dragatsis, I., Efstratiadis, A., and Goldman, J. E. (2003) J Neurosci 23, 7710–7718 22. Spagnoli, A., and Rosenfeld, R. G. (1996) Endocrinol Metab Clin North Am 25, 615–631 23. Kelley, K. M., Oh, Y., Gargosky, S. E., Gucev, Z., Matsumoto, T., Hwa, V., Ng, L., Simpson, D. M., and Rosenfeld, R. G. (1996) Int J Biochem Cell Biol 28, 619–637 24. Jones, J. I., and Clemmons, D. R. (1995) Endocr Rev 16, 3–34 25. Kuhl, N. M., Hoekstra, D., De Vries, H., and De Keyser, J. (2003) Glia 44, 91–101 26. Baxter, R. C. (2000) Am J Physiol Endocrinol Metab 278, E967–976 27. Kuhl, N. M., De Keyser, J., De Vries, H., and Hoekstra, D. (2002) J Neurosci Res 69, 207–216 28. McCawley, L. J., and Matrisian, L. M. (2001) Curr Opin Cell Biol 13, 534–540 29. Fowlkes, J. L., and Winkler, M. K. (2002) Cytokine Growth Factor Rev 13, 277–287 30. Fowlkes, J. L., Serra, D. M., Bunn, R. C., Thrailkill, K. M., Enghild, J. J., and Nagase, H. (2004) Endocrinology 145, 620–626 31. Machida, C. M., Rodland, K. D., Matrisian, L., Magun, B. E., and Ciment, G. (1989) Neuron 2, 1587–1596 32. Muir, D. (1994) Exp Cell Res 210, 243–252 33. Nordstrom, L. A., Lochner, J., Yeung, W., and Ciment, G. (1995) Mol Cell Neurosci 6, 56–68 34. Uhm, J. H., Dooley, N. P., Oh, L. Y., and Yong, V. W. (1998) Glia 22, 53–63 35. Oh, L. Y., Larsen, P. H., Krekoski, C. A., Edwards, D. R., Donovan, F., Werb, Z., and Yong, V. W. (1999) J Neurosci 19, 8464–8475 36. Matsumoto, S., Kobayashi, T., Katoh, M., Saito, S., Ikeda, Y., Kobori, M., Masuho, Y., and Watanabe, T. (1998) Am J Pathol 153, 109–119 37. Annabi, B., Shedid, D., Ghosn, P., Kenigsberg, R. L., Desrosiers, R. R., Bojanowski, M. W., Beaulieu, E., Nassif, E., Moumdjian, R., and Beliveau, R. (2002) J Vasc Surg 35, 539–546 38. Shipley, J. M., Wesselschmidt, R. L., Kobayashi, D. K., Ley, T. J., and Shapiro, S. D. (1996) Proc Natl Acad Sci USA 93, 3942–3946 39. Lanone, S., Zheng, T., Zhu, Z., Liu, W., Lee, C. G., Ma, B., Chen, Q., Homer, R. J., Wang, J., Rabach, L. A., Rabach, M. E., Shipley, J. M., Shapiro, S. D., Senior, R. M., and Elias, J. A. (2002) J Clin Invest 110, 463–474

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

Matrix Metalloproteinases in Myelin Formation

203

40. Vos, C. M., van Haastert, E. S., de Groot, C. J., van der Valk, P., and de Vries, H. E. (2003) J Neuroimmunol 138, 106–114 41. Larsen, P. H., and Yong, V. W. (2004) J Neurosci 24, 7597–7603 42. Goddard, D. R., Berry, M., and Butt, A. M. (1999) J Neurosci Res 57, 74–85 43. Ye, P., Carson, J., and D’Ercole, A. J. (1995) J Neurosci 15, 7344–7356 44. Engsig, M. T., Chen, Q. J., Vu, T. H., Pedersen, A. C., Therkidsen, B., Lund, L. R., Henriksen, K., Lenhard, T., Foged, N. T., Werb, Z., and Delaisse, J. M. (2000) J Cell Biol 151, 879–889 45. Lambert, V., Munaut, C., Jost, M., Noel, A., Werb, Z., Foidart, J. M., and Rakic, J. M. (2002) Am J Pathol 161, 1247–1253 46. Lambert, V., Wielockx, B., Munaut, C., Galopin, C., Jost, M., Itoh, T., Werb, Z., Baker, A., Libert, C., Krell, H. W., Foidart, J. M., Noel, A., and Rakic, J. M. (2003) Faseb J 17, 2290–2292 47. Oh, J., Takahashi, R., Adachi, E., Kondo, S., Kuratomi, S., Noma, A., Alexander, D. B., Motoda, H., Okada, A., Seiki, M., Itoh, T., Itohara, S., Takahashi, C., and Noda, M. (2004) Oncogene 23, 5041–5048 48. Prineas, J. W., Barnard, R. O., Kwon, E. E., Sharer, L. R., and Cho, E. S. (1993) Ann Neurol 33, 137–151 49. Chang, A., Tourtellotte, W. W., Rudick, R., and Trapp, B. D. (2002) N Engl J Med 346, 165–173 50. Franklin, R. J. (2002) Nat Rev Neurosci 3, 705–714 51. Larsen, P. H., Wells, J. E., Stallcup, W. B., Opdenakker, G., and Yong, V. W. (2003) J Neurosci 23, 11127–11135 52. Yong, V. W., Power, C., Forsyth, P., and Edwards, D. R. (2001) Nat Rev Neurosci 2, 502–511 53. Zhang, Y., Tohyama, K., Winterbottom, J. K., Haque, N. S., Schachner, M., Lieberman, A. R., and Anderson, P. N. (2001) Mol Cell Neurosci 17, 444–459 54. Jones, L. L., Yamaguchi, Y., Stallcup, W. B., and Tuszynski, M. H. (2002) J Neurosci 22, 2792–2803 55. Chang, A., Nishiyama, A., Peterson, J., Prineas, J., and Trapp, B. D. (2000) J Neurosci 20, 6404–6412 56. Watanabe, M., Toyama, Y., and Nishiyama, A. (2002) J Neurosci Res 69, 826–836 57. Fidler, P. S., Schuette, K., Asher, R. A., Dobbertin, A., Thornton, S. R., Calle-Patino, Y., Muir, E., Levine, J. M., Geller, H. M., Rogers, J. H., Faissner, A., and Fawcett, J. W. (1999) J Neurosci 19, 8778–8788 58. Dou, C. L., and Levine, J. M. (1994) J Neurosci 14, 7616–7628 59. Winberg, J. O., Kolset, S. O., Berg, E., and Uhlin-Hansen, L. (2000) J Mol Biol 304, 669–680 60. Einstein, E. R., Csejtey, J., Dalal, K. B., Adams, C. W., Bayliss, O. B., and Hallpike, J. F. (1972) J Neurochem 19, 653–662 61. Hirsch, H. E. (1981) J Histochem Cytochem 29, 425–430 62. Allen, I. V. (1981) Neuropathol Appl Neurobiol 7, 169–182 63. Cammer, W., Bloom, B. R., Norton, W. T., and Gordon, S. (1978) Proc Natl Acad Sci USA 75, 1554–1558

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

204

P.H. Larsen & V.W. Yong

64. Banik, N. L. (1992) Crit Rev Neurobiol 6, 257–271 65. Cuzner, M. L., Gveric, D., Strand, C., Loughlin, A. J., Paemen, L., Opdenakker, G., and Newcombe, J. (1996) J Neuropathol Exp Neurol 55, 1194–1204 66. Proost, P., Van Damme, J., and Opdenakker, G. (1993) Biochem Biophys Res Commun 192, 1175–1181 67. Chandler, S., Coates, R., Gearing, A., Lury, J., Wells, G., and Bone, E. (1995) Neurosci Lett 201, 223–226 68. Chandler, S., Cossins, J., Lury, J., and Wells, G. (1996) Biochem Biophys Res Commun 228, 421–429 69. Maeda, A., and Sobel, R. A. (1996) J Neuropathol Exp Neurol 55, 300–309 70. Cossins, J. A., Clements, J. M., Ford, J., Miller, K. M., Pigott, R., Vos, W., Van der Valk, P., and De Groot, C. J. (1997) Acta Neuropathol (Berl) 94, 590–598 71. Anthony, D. C., Miller, K. M., Fearn, S., Townsend, M. J., Opdenakker, G., Wells, G. M., Clements, J. M., Chandler, S., Gearing, A. J., and Perry, V. H. (1998) J Neuroimmunol 87, 62–72 72. Leppert, D., Ford, J., Stabler, G., Grygar, C., Lienert, C., Huber, S., Miller, K. M., Hauser, S. L., and Kappos, L. (1998) Brain 121 (Pt 12), 2327–2334 73. Gijbels, K., Proost, P., Masure, S., Carton, H., Billiau, A., and Opdenakker, G. (1993) J Neurosci Res 36, 432–440 74. Clements, J. M., Cossins, J. A., Wells, G. M., Corkill, D. J., Helfrich, K., Wood, L. M., Pigott, R., Stabler, G., Ward, G. A., Gearing, A. J., and Miller, K. M. (1997) J Neuroimmunol 74, 85–94 75. Kieseier, B. C., Kiefer, R., Clements, J. M., Miller, K., Wells, G. M., Schweitzer, T., Gearing, A. J., and Hartung, H. P. (1998) Brain 121 (Pt 1), 159–166 76. Kieseier, B. C., Clements, J. M., Pischel, H. B., Wells, G. M., Miller, K., Gearing, A. J., and Hartung, H. P. (1998) Ann Neurol 43, 427–434 77. Hughes, P. M., Wells, G. M., Clements, J. M., Gearing, A. J., Redford, E. J., Davies, M., Smith, K. J., Hughes, R. A., Brown, M. C., and Miller, K. M. (1998) Brain 121 (Pt 3), 481–494 78. Gijbels, K., Galardy, R. E., and Steinman, L. (1994) J Clin Invest 94, 2177–2182 79. Hewson, A. K., Smith, T., Leonard, J. P., and Cuzner, M. L. (1995) Inﬂamm Res 44, 345–349 80. Leppert, D., Waubant, E., Burk, M. R., Oksenberg, J. R., and Hauser, S. L. (1996) Ann Neurol 40, 846–852 81. Stuve, O., Dooley, N. P., Uhm, J. H., Antel, J. P., Francis, G. S., Williams, G., and Yong, V. W. (1996) Ann Neurol 40, 853–863 82. Zuo, J., Ferguson, T. A., Hernandez, Y. J., Stetler-Stevenson, W. G., and Muir, D. (1998) J Neurosci 18, 5203–5211 83. Hayashita-Kinoh, H., Kinoh, H., Okada, A., Komori, K., Itoh, Y., Chiba, T., Kajita, M., Yana, I., and Seiki, M. (2001) Cell Growth Diﬀer 12, 573–580 84. Galko, M. J., and Tessier-Lavigne, M. (2000) Science 289, 1365–1367

May 26, 2005

16:4 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch07

Matrix Metalloproteinases in Myelin Formation

205

85. Hattori, M., Osterﬁeld, M., and Flanagan, J. G. (2000) Science 289, 1360–1365 86. Webber, C. A., Hocking, J. C., Yong, V. W., Stange, C. L., and McFarlane, S. (2002) J Neurosci 22, 8091–8100 87. Gu, Z., Kaul, M., Yan, B., Kridel, S. J., Cui, J., Strongin, A., Smith, J. W., Liddington, R. C., and Lipton, S. A. (2002) Science 297, 1186–1190 88. Newman, T. A., Woolley, S. T., Hughes, P. M., Sibson, N. R., Anthony, D. C., and Perry, V. H. (2001) Brain 124, 2203–2214 89. Wang, X., Jung, J., Asahi, M., Chwang, W., Russo, L., Moskowitz, M. A., Dixon, C. E., Fini, M. E., and Lo, E. H. (2000) J Neurosci 20, 7037–7042 90. Noble, L. J., Donovan, F., Igarashi, T., Goussev, S., and Werb, Z. (2002) J Neurosci 22, 7526–7535 91. Wells, J. E., Rice, T. K., Nuttall, R. K., Edwards, D. R., Zekki, H., Rivest, S., and Yong, V. W. (2003) J Neurosci 23, 10107–10115 92. Yong, V. W., Krekoski, C. A., Forsyth, P. A., Bell, R., and Edwards, D. R. (1998) Trends Neurosci 21, 75–80 93. Lo, E. H., Wang, X., and Cuzner, M. L. (2002) J Neurosci Res 69, 1–9 94. Rosenberg, G. A. (2002) Glia 39, 279–291 95. Rao, J. S. (2003) Nat Rev Cancer 3, 489–501 96. Cal, S., Freije, J. M., Lopez, J. M., Takada, Y., and Lopez-Otin, C. (2000) Mol Biol Cell 11, 1457–1469 97. Cal, S., Obaya, A. J., Llamazares, M., Garabaya, C., Quesada, V., and Lopez-Otin, C. (2002) Gene 283, 49–62 98. Rybnikova, E., Karkkainen, I., Pelto-Huikko, M., and Huovila, A. P. (2002) Neuroscience 112, 921–934 99. Sun, Y. P., Deng, K. J., Wang, F., Zhang, J., Huang, X., Qiao, S., and Zhao, S. (2004) Gene 325, 171–178 100. Blobel, C. P. (2002) Inﬂamm Res 51, 83–84 101. Apte, S. S. (2004) Int J Biochem Cell Biol 36, 981–985 102. Sotillos, S., Roch, F., and Campuzano, S. (1997) Development 124, 4769–4779 103. Brou, C., Logeat, F., Gupta, N., Bessia, C., LeBail, O., Doedens, J. R., Cumano, A., Roux, P., Black, R. A., and Israel, A. (2000) Mol Cell 5, 207–216 104. Hartmann, D., de Strooper, B., Serneels, L., Craessaerts, K., Herreman, A., Annaert, W., Umans, L., Lubke, T., Lena Illert, A., von Figura, K., and Saftig, P. (2002) Hum Mol Genet 11, 2615–2624 105. Mochizuki, S., Shimoda, M., Shiomi, T., Fujii, Y., and Okada, Y. (2004) Biochem Biophys Res Commun 315, 79–84 106. Bernstein, H. G., Keilhoﬀ, G., Bukowska, A., Ziegeler, A., Funke, S., Dobrowolny, H., Kanakis, D., Bogerts, B., and Lendeckel, U. (2004) J Neurosci Res 75, 353–360

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

This page intentionally left blank

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

CHAPTER 8 TIMPs IN CNS DEVELOPMENT

D. Jaworski Department of Anatomy & Neurobiology, University of Vermont College of Medicine, 149 Beaumont Avenue, HSRF 418, Burlington, VT 05450 E-mail: [email protected]

Our understanding of the role of MMPs and TIMPs (Tissue Inhibitor of Metalloproteinases) in mature nervous system function is constantly evolving as reﬂected by recent reviews (1–4). However, the study of TIMPs during nervous system development is at its infancy. A paucity of information regarding the substrates to which TIMPs bind, and their mechanism of action within the nervous system exists. This chapter will highlight what is known about TIMPs during nervous system development as well as their role in the mature nervous system. Finally, the participation of TIMPs in neuropathological disorders will be considered.

1. Developmental Regulation of TIMPs Much of what is known about the role of TIMPs in the developing nervous system is based on supposition according to temporal and spatial expression patterns. However, there are few comprehensive studies examining the temporal regulation of both TIMP mRNA and protein within the nervous system throughout the timecourse of development. Moreover, as TIMP-1, -2, and -4 are secreted molecules, it is important to examine the spatial distribution of the protein relative to that of the mRNA. Given that differences in TIMP expression patterns between mice and rats exist, both species will be highlighted here. Very low basal levels of TIMP-1 mRNA are present in developing (5) and mature (6) murine brain. In a number of systems, TIMP-1 expression

Correspondence to: D. Jaworski 207

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

208

D. Jaworski

is induced by c-Fos (AP-1) (7–9). The lack of basal levels of TIMP-1 may be due to the fact that c-Fos is not expressed in the unstimulated adult nervous system. In contrast, TIMP-1 is the most abundantly expressed TIMP in the embryonic rat brain (10). During embryogenesis, TIMP-1 expression is enriched in the neuroepithelium, the site of neuro- and gliogenesis (10). After birth, TIMP-1 mRNA is dramatically down-regulated (10) and expressed primarily in the hippocampal formation and cerebellum (11). Therefore, it is possible that TIMP-1 regulates cell growth in the nervous system as has been observed for other systems (12, 13). TIMP-1 binds to the cell surface of MCF-7 breast carcinoma cell and is translocated to the nucleus (14). Nuclear accumulation of TIMP-1 in ﬁbroblasts occurs in a cell cycle-dependent manner (15). Whether TIMP-1 exerts similar growth eﬀects directly within the nucleus of neurons or glia is yet to be determined. Owing to the very low levels of TIMP-1 mRNA expression, the detection of TIMP-1 protein in developing rat cerebellum required immunocytochemical enhancement with tyramide signal ampliﬁcation (16). TIMP-1 labelled presumptive granule cell precursors, Purkinje cells, some granule cells, and Bergmann glia. The expression of TIMP-1 coincides with the formation of climbing ﬁbre aﬀerent projections, suggesting that it may play a role in the formation of olivocerebellar connections, particularly neuronal migration and synaptogenesis (16). TIMP-2 mRNA expression is detected as early as embryonic day 11.5 (E11.5) in the murine brain and signiﬁcantly increases from E13.5 to E15.5 (5). Expression then gradually increases to peak at postnatal day 3 (P 3) and declines slightly to P 28, the last timepoint examined. Nonetheless, TIMP-2 mRNA is expressed at high levels in the adult murine brain (6, 17). The developmental regulation of TIMP-2 mRNA in the rat brain diﬀers in that expression continues to increase throughout development and is the most abundantly expressed TIMP in the adult rat brain (10). Embryonically, TIMP-2 mRNA is expressed in the choroid plexus and ependymal cells lining the lateral ventricles, but is not expressed in the adjacent neuroepithelium (17). Thus, in contrast to TIMP-1, TIMP-2 expression is enriched in post-mitotic neurons (10). The expression of TIMP-2 in post-mitotic neurons in divergent regions of the nervous system (e.g. cerebral cortical and hippocampal pyramidal cells, cerebellar Purkinje cells, spinal motor neurons) suggests it likely regulates the acquisition of the general features of a neuronal phenotype (10, 18). The best characterised role for TIMP-2 in other systems is growth regulation. TIMP-2 exerts both growth promoting (19, 20) and growth inhibiting (21–23) activities depending on the cellular

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

TIMPs in CNS Development

209

context. We recently reported that TIMP-2 inhibits the proliferation of rat PC12 cells which diﬀerentiate into a neuronal phenotype in the presence of nerve growth factor (NGF) (24). In addition, the neocortical neuroepithelium of Timp-2 -/- mice contains more nestin-positive progenitors (24). Together, these data indicate that TIMP-2 serves as an anti-mitogenic signal during nervous system development. This is further substantiated by the enrichment of TIMP-2 protein in presumptive granule cell precursors during the peak of granule cell proliferation (16). TIMP-3 mRNA is expressed at low levels in developing and adult murine (5, 6) and rat (10) nervous system. Embryonically, TIMP-3 mRNA is expressed in the choroid plexus and olfactory epithelium (25, 26). TIMP-3 mRNA is expressed throughout the neuraxis of the rat central nervous system (CNS) and, in sensory ganglion, including the trigeminal and dorsal root ganglion, in the peripheral nervous system (PNS) (26). Neonatally, TIMP-3 mRNA is expressed in the cerebellum, several brainstem nuclei, the cerebral cortex and thalamus, and the olfactory bulb (26). Like TIMP-1, TIMP-3 mRNA is expressed in the neuroepithelium within the ventricular zone. TIMP-3 mRNA is also expressed in the subventricular zone (SVZ) which generates glia and olfactory bulb interneurons. Although TIMP-3 is expressed by astrocytes (18, 26–28), its temporal regulation in the SVZ does not correlate with gliogenesis in that TIMP-3 expression precedes glial proliferation and diﬀerentiation. In addition, TIMP-3 expression declines dramatically during the ﬁrst two postnatal weeks, a time when the majority of olfactory bulb interneurons are generated. Thus, TIMP-3 likely does not play a role in neurogenesis. TIMP-3 is known to induce apoptosis in a number of systems (29–32), including the nervous system (33, 34). Given that ﬁfty percent of progenitors undergo apoptosis in the SVZ (35) during the ﬁrst two postnatal weeks (36), it is possible that TIMP-3 plays a role in cell death and not cell birth in the ventricular and subventricular zones. In addition to its expression in the SVZ, TIMP-3 mRNA is expressed in the rostral migratory stream (RMS) (26). In contrast to radial glial guided migration in the cerebral cortex (37), neurons migrate from the SVZ to the olfactory bulb along the RMS without the aid of glial guides (38, 39). TIMP-3 decreases migration (30, 40) by reducing attachment to the extracellular matrix (ECM) (30, 41). By preventing inappropriate attachment to the ECM of the RMS, TIMP-3 could facilitate neuronal migration towards the olfactory bulb. The expression of TIMP-3 protein by cerebellar granule cells coincide with migration (16) further substantiates the proposed role for TIMP-3 in migration. Interestingly, the N -terminal domain of TIMPs

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

210

D. Jaworski

share structural homology with proteins involved in guidance (42), suggesting that all TIMPs may play a role in neuronal migration and axon guidance. In addition to expression on Purkinje cell soma, TIMP-3 protein, unlike the other TIMPs, is detected on Purkinje cell dendritic processes (16). Thus, in addition to migration, TIMP-3 may play a role in terminal neuronal diﬀerentiation, including arborisation and synaptogenesis. TIMP-4 is the most abundantly expressed TIMP in developing (5) and mature (6) murine brain and the second most abundant TIMP in the rat brain (10). However, nothing is known or supposed about its function in the nervous system. In the mouse, TIMP-4 has been detected at E15.5 by RT-PCR (5), but its expression is not detectable during embryogenesis by in situ hybridisation or northern blot analysis (6). Expression increases to peak at P 14, then declines (5). TIMP-4 mRNA expression in the rat is detected in the neuroepithelium and dorsal root ganglion at E14 and expression increases throughout embryonic and postnatal development (10). In the postnatal rat brain, TIMP-4 mRNA expression is largely restricted to cerebellar Purkinje cells. The elucidation of TIMP-4’s function in developing the nervous system requires additional immunohistochemical data and/or analysis of a TIMP-4 knockout mouse. The expression of TIMP-1, -3, and -4 in the neuroepithelium (10) suggests that TIMPs are involved in very early stages of neurogenesis. This observation is corroborated by the detection of TIMP-1, -2, and -3 in E10.5 rat neural stem cells (43). In contrast to MMP-9, whose expression decreases during diﬀerentiation, the expression of MMP-2 as well as the three TIMPs is not altered upon diﬀerentiation to progenitors. Similarly, TIMP-1, -2, and -3 expression is unaltered during diﬀerentiation of human embryonic CNS stem cells (44). In contrast, TIMP-4 mRNA expression (6.4-fold) and protein levels in conditioned media (23-fold) decreased upon diﬀerentiation and was associated with a similar decrease in MMP-2. Given the lack of regulation for most TIMPs upon stem cell diﬀerentiation, the functional signiﬁcance of TIMP expression in stem cells and their progenitor progeny remains to be elucidated. 2. TIMPs in the Mature Nervous System Because TIMPs are traditionally recognised for their MMP-inhibitory activity, their role in the maintenance of matrix integrity during morphogenesis is well established. As discussed above, proposed functions of TIMPs during nervous system development include the regulation of proliferation,

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

TIMPs in CNS Development

211

apoptosis, neuronal migration, axon guidance, neuritic arborisation and synaptogenesis. Given the limited matrix remodelling in the mature nervous system, the potential functions of TIMPs are less obvious and therefore have largely gone uncharacterised. One possibility is that TIMPs play a role in synaptic plasticity. The ability to learn and form memories depends on speciﬁc patterns of synaptic activity that produces rapid and long lasting modiﬁcations of synaptic structure. ECM molecules are thought to participate in activitydependent plasticity by providing a new microenvironment subsequent to synaptic transmission (reviewed in (45, 46)). In particular, proteases may play a role in the structural rearrangements in the synapse/spine complex associated with neural plasticity (47, 48). Kainate (KA) treatment provides a robust model of synaptic plasticity in that it is associated with structural reorganisation of existing synapses and the induction of new synapse formation (49, 50). In its role as a glutamate receptor agonist KA is both pro-convulsive and neurotoxic. TIMP-1 expression is up-regulated in the hippocampal formation following KA-induced seizures (9, 11, 51). The upregulation of TIMP-1 in response to KA is not solely due to neurotoxicity. Treatment with pentylenetetrazole (PTZ), which induces hippocampal neuronal excitation without adverse eﬀects on neuronal survival, similarly induces TIMP-1 expression (9). Furthermore, TIMP-1 is increased in response to KCl depolarisation (52). The up-regulation of MMPs in response to KA treatment (53–55) suggests that TIMP-1 may function in a traditional MMP-dependent manner during dendritic remodelling. Alternatively, TIMP-1 could exert MMP-independent neuroprotective eﬀects by inhibiting glutamate-induced calcium entry (56). TIMP-1 is also upregulated in Pavlovian fear conditioning (51), a model of synaptic plasticity (57, 58) which is non-excitotoxic and thus more physiologic. After a few pairings of a neutral stimulus (e.g. tone) with an aversive stimulus (e.g. foot shock), the tone comes to elicit a variety of responses, including an increased startle reﬂex, that are indicative of conditioned fear. The synaptic plasticity associated with conditioned fear is mediated by the basolateral nuclear complex of the amygdala (59). The up-regulation of TIMP-1 in fear conditioning is unexpected in that TIMP-1 mRNA is not expressed in the basolateral nucleus (60). It has also been proposed that TIMP-3 plays a role in cerebellar synaptic plasticity (16) due to its expression on Purkinje cell dendrites; however, this has not been veriﬁed physiologically. In addition to structural alterations, synaptic plasticity is associated with an increase in synaptic transmission eﬃcacy. Interestingly, TIMP genes

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

212

D. Jaworski

are nested within the genes for synapsins and this organisation is evolutionarily conserved (61, 62). The single fruit-ﬂy TIMP is nested within the intron of the synapsin gene in reverse orientation. A similar association of TIMP-1 within synapsin 1, TIMP-3 within synapsin 3 and TIMP-4 within synapsin 2 is observed in humans (62) and mice (63). While TIMP-2 is not located within a synapsin gene, it is located near synapsin 4 (61). Synapsins are neuron-speciﬁc phosphoproteins associated with the membranes of synaptic vesicles and regulate neurotransmitter release. In addition, synapsins play a broad role during neuronal development (64). While this ﬁnding is intriguing, its signiﬁcance is uncertain. The gene structure could be happenstance, reﬂecting evolutionary multiplication of genome sections, or it may represent a speciﬁc organisational or regulatory relationship between TIMPs and synapsins. The elucidation of deﬁnitive functions for TIMPs in developing and mature nervous system requires phenotypic analysis of knockout mice. TIMP-1 deletion is associated with impaired reproductive physiology (65, 66), cardiovascular remodelling after myocardial infarction (67) and nutritionally induced obesity (68). In addition, TIMP-1 deﬁcient mice are hyper-resistant to bacterial infection (69). Thus far, the only phenotype reported for TIMP-2 mutant mice is the impairment of proMMP-2 activation (70, 71). This reﬂects the unique ability of TIMP-2 to regulate both the activation of latent proMMPs to the proteolytically active form and the inhibition of MMP activity. TIMP-3 knockout mice have altered lung development (72) and progressive decline in lung function (73) and are moribund at 1 year of age. In addition, TIMP-3 null mice display accelerated apoptosis in the mammary gland (74). TIMP-4 knockout mice have not yet been reported, but should be forthcoming soon given the recent characterisation of the murine TIMP-4 promoter (5, 63). Thus far, no overt phenotype has been observed in the nervous system of TIMP knockout mice. One possibility is that physiological phenotypes are present. Hypothalamic expression of TIMP-3 increases in response to food deprivation (75). TIMP-2 expression is greater in mouse lines selected for heat loss (i.e. animals have greater food intake and locomotor activity with reduced body fat) (76). Therefore, TIMP knockout mice may have metabolic alterations which require physiological analysis to be detected. Given the signiﬁcant expression of TIMP-2 in neurons of the developing and mature nervous system, we recently conducted an in-depth analysis of the nervous system of TIMP-2 knockout mice. Histologically, the neocortical neuroepithelium of neonatal Timp-2 -/- mice contains more nestin-positive

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

TIMPs in CNS Development

213

progenitors than wild-type littermates. In addition, neuritic arborisation was signiﬁcantly reduced in vivo and in vitro. As these diﬀerences are no longer apparent by P 14, these data indicate a delay in initial and terminal neuronal diﬀerentiation in the absence of TIMP-2 (24). TIMP-2 is abundantly expressed in spinal cord motor neurons throughout development and expression is maintained into adulthood (18). In contrast, TIMP-2 expression in muscle is down-regulated after P 21 (5). Thus, we examined motor function in Timp-2 -/- mice using a motorised rotating rod (RotaRod). Timp-2 -/- mice fell oﬀ the RotaRod much faster than wild-type or heterozygous littermates (77). Furthermore, the distribution of acetylcholine receptors on knockout muscle was altered. TIMP-2 mRNA and protein are also expressed in the basolateral nucleus of the amygdala (60). Inasmuch no other TIMP is expressed in the amygdala, TIMP-2 deletion would not likely be compensated by other TIMPs. Thus, we examined the acquisition of conditioned fear in Timp-2 -/- mice. Timp-2 -/- mice did not acquire conditioned fear (60). This observation substantiates the need to re-examine the other TIMP knockout mice for developmental histological abnormalities and/or subtle phenotypes which would only be revealed with behavioural methodologies. 3. TIMPs in Neuropathology The only known neuropathological state associated with a TIMP mutation is Sorby’s fundus dystrophy (SFD), which is associated with mutations in TIMP-3. SFD is a dominantly inherited degenerative disease of the retina producing loss of vision in middle age. Although SFD is a relatively rare disease, it is of signiﬁcant interest due to its resemblance to the wet form of age-related macular degeneration, the leading cause of permanent visual impairment among the elderly in developed countries. Histologically, the disease is characterised by subretinal neovascularisation and an accumulation of TIMP-3 within and thickening of Bruch’s membrane, the basement membrane separating the retinal pigment epithelium from the choriocapillaris, its blood supply. Several mutations in the TIMP-3 gene have been identiﬁed in individuals aﬀected by SFD. Most commonly, missense point mutations in the C-terminal region of exon 5 replace conserved amino acids with cysteines (78–82). In addition, a point mutation that results in deletion of most of the C-terminal domain of TIMP-3 (83), a splice site mutation (84), and a mutation resulting in a premature termination codon (85) have been identiﬁed. However, a pedigree with SFD but no causative mutation

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

214

D. Jaworski

in the TIMP-3 gene has also been reported (86). The mechanism by which TIMP-3 mutation leads to neovascularisation and Bruch’s membrane thickening is not well understood. Interestingly, S156C TIMP-3 induced angiogenesis in a chorioallantoic membrane assay, suggesting it may play a role in choroidal neovascularisation (87). The mutant protein had reduced MMP inhibitory activity in retinal pigment epithelial cells. This is in contrast to other studies which demonstrated that multiple forms of mutant TIMP-3 proteins were localised to the extracellular matrix and displayed appropriate MMP inhibitory activity (88, 89). It has been proposed that the accumulation of TIMP-3 in Bruch’s membrane is not due to increased TIMP-3 mRNA synthesis (90). Rather, aberrant protein-protein interactions and altered cell adhesion (89), including the formation of TIMP-3 dimers (88) may alter TIMP-3 metabolism. Dysregulation of MMP/TIMP balance is present in a variety of neuropathological disorders. The roles of MMPs in these disorders is discussed elsewhere in this text. Here, the roles of TIMPs in tumourigenesis, neuronal injury, and neurodegenerative diseases will be presented. Tumourigenesis involves aberrant cell growth, angiogenesis and invasion into the surrounding parenchyma. While these processes may occur due to an imbalance in MMPs and TIMPs (91), they may also be mediated by TIMPs in an MMP-independent manner. Paradoxically, TIMP-1 is associated with malignancy promotion rather than suppression. TIMP-1 levels are signiﬁcantly higher in glioblastoma than anaplastic astrocytomas (92–94). In contrast, TIMP-2 (95) and TIMP-4 (93) negatively correlate with glioma malignancy and are expressed at lower levels in glioma than normal brain. TIMPs are known to exert both growth promoting and inhibiting activities depending on the cellular context. In a number of systems, the growth regulatory functions of TIMPs are independent of MMP inhibition (20, 22, 96). This may explain the less resounding success of the MMP inhibitor clinical trials (97). Angiogenesis is another key feature of tumorigenesis. TIMP-1 is localised to the walls of neovessels (98) and promotes vascular endothelial growth factor (VEGF)-mediated vascularisation in the retina (99). TIMP-2 inhibits ﬁbroblast growth factor or VEGF-induced angiogenesis in an MMP-independent manner by blocking the interaction of HSP60 and SHP-1 to α3 β1 integrin (23). TIMP-3 also inhibits angiogenesis (100) by serving as antagonist to the VEGF-2 receptor (101). Thus, anti-angiogenic neoplastic therapies could be developed to either TIMP-2 or TIMP-3 (102, 103) since they regulate angiogenesis by an MMP-independent mechanism. Finally, local invasion is a characteristic

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

TIMPs in CNS Development

215

feature of glioblastoma which prevents complete surgical resection and accounts, in large part, for its poor prognosis. While collagenolysis may promote invasion in non-neural systems, little collagen exists in the mature nervous system. Therefore, other molecules are likely to play a role. Integrins have been shown to serve as a docking system for the pro-MMP-2, TIMP-2, MT1-MMP complex (104–106) and promote the activation of pro-MMP-2 and pro-MT1-MMP (107–109). Inasmuch as integrins regulate cell cycle progression and migration, it is possible that MMPs and TIMPs exert their tumorigenic actions via integrins. Another candidate molecule regulating tumour invasion is brevican. Brevican expression is up-regulated in glioma (110), but requires proteolytic cleavage to induce invasion (111). Cleavage is mediated by ADAM-TS4 (112), which is inhibited by TIMP-3 (113). Hence, TIMP-3 could be used therapeutically to reduce invasion as well as angiogenesis. Interestingly, the TIMP-3 promoter is hypermethylated in a number of cancers (114), including astrocytomas (115). Hypermethylation is known to inactivate tumour suppressor genes; thus, it is possible that TIMP-3 functions as a tumour suppressor. Successful regeneration, consequent to neuronal injury, is dependent upon maintaining an intricate balance between proteolysis and its inhibition. TIMP expression is altered in a number of injury models. TIMP-1 expression (116) and activity (117) is increased following sciatic nerve crush and spinal root avulsion (118). Expression is detected in inﬁltrating macrophages and Schwann cells (116). Increased TIMP-1 expression is not restricted to PNS injuries. TIMP-1 mRNA is also increased in response to spinal cord (119) and cerebral (52) contusion, and a penetrating stab wound through the cortex, hippocampus and thalamus (120). In the cerebral contusion model TIMP-1 expression is restricted to neurons, while in the stab wound model expression is primarily detected in reactive astrocytes. TIMP-1 expression is also detected in reactive astrocytes and microglia in wobbler mutant mice (121). TIMP-1 can be induced by pro-inﬂammatory cytokines, including IL-1β and TNF-α (27). Disruption of the ventricular system and the inﬂuence of cytokines in the CSF may be responsible for the presence of TIMP-1 in reactive astrocytes in the stab model, but not in the contusion model. The induction of IL-1β and TNF-α in the spinal cord and brainstem of wobbler mice (122) may explain TIMP-1 expression in this model. Given the protective and homeostatic functions of TIMP-1 (3), it may limit inﬂammation (52) and/or protect the basement membrane from uncontrolled degradation following injury (116). TIMP-2 expression is up-regulated following spinal root avulsion (118) and bulbectomy (removal

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

216

D. Jaworski

of the olfactory bulb) (123). In contrast, TIMP-2 is not up-regulated in the cerebral contusion injury model (52) or wobbler mice (121). Two different cerebral cortical injury models have demonstrated increased TIMP-2 expression, but the cell types in which TIMP-2 is expressed diﬀers. In one model, a penetrating lesion, described above, is made with a small gauge needle. In this case, TIMP-2 expression is detected, but not increased, in neurons, and is up-regulated in microglia and inﬁltrated macrophages (120). This is in contrast to the incision lesion (3–4 mm deep, 5–6 mm long) model in which injury is restricted to the cerebral cortex. In this case, TIMP-2 expression is up-regulated in neurons and reactive astrocytes (28). Although TIMP-2 is generally expressed in a constitutive manner, taken together these data suggest that it can indeed be induced in response to injury. While a clear consensus for TIMP-1 and a general consensus for TIMP-2 in response to injury exists, the data for TIMP-3 is less clear. TIMP-3 is expressed in reactive astrocytes in the incision injury (28) and in wobbler mice (121). In contrast, no change in TIMP-3 is detected in either penetrating stab (120) or cerebral contusion (52) injuries. Moreover, TIMP-3 activity is actually down-regulated in sciatic nerve crush (117). Given that TIMP-3 has been suggested to play a role in neuronal migration and process outgrowth (26), it is somewhat surprising that its expression decreases during the period of greatest axonal growth following injury. The decreased expression of TIMP-3 may be due to the fact that IL-1β and TNF-α, which are induced during injury, down-regulate TIMP-3 (27). In all the injury models examined thus far, no change in TIMP-4 expression has been observed (28, 120, 121). While the increased expression of TIMPs following injury may serve a protective role to curb excessive MMP-mediated matrix degradation, it is also possible that TIMPs may make the glial scar and injured white matter more inhibitory than adjacent grey matter (28). Examination of injury responses in TIMP knockout mice may resolve some of the discrepancies observed in various injury models and better deﬁne the function and mechanism of action for TIMPs in response to injury. Parkinson’s disease (PD), Huntington’s disease (HD), Alzheimer’s disease (AD) and amyotrophic lateral sclerosis (ALS) are progressive movement disorders with dementia, except in the case of ALS. Histologically, these disorders are characterised by neuronal degeneration within the substantia nigra, striatum, nucleus basalis of Meynert, and upper/lower motor neurons, respectively. The mechanism mediating this selective neuronal death is unknown. Recently, it was reported that TIMP-1 is signiﬁcantly

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

TIMPs in CNS Development

217

elevated in CSF samples from these four disorders (124). This substantiates previous work that detected increased TIMP-1 in the substantia nigra, but not cortex or hippocampus of PD patients (125) and TIMP labelling in plaques and neuroﬁbrillary tangles in AD (126). However, it is in conﬂict with other reports that found no alteration in TIMP-1 in AD (127) or ALS (128). The basis for this discrepancy is unclear. Unlike TIMP-1, TIMP-2 was signiﬁcantly increased in the CSF only of AD and HD patients (124). One of the hallmarks of AD is the presence of senile plaques, extracellular aggregates consisting of the amyloid β (Aβ) peptide. Aβ is derived from the amyloid precursor protein (APP) by a series of cleavages involving α-secretase (TACE [ADAM-17] or ADAM-10), β-secretase (BACE) and γ-secretase (presenillin complex) proteases (129). Interestingly, the APP gene contains a TIMP-like MMP inhibitor domain (130). However, the signiﬁcance of it or the increased presence of TIMPs in AD is currently unknown. In contrast to the extracellular protein aggregates present in AD, intracellular proteinaceous inclusions are present in PD and HD. Presenillin is known to be proteolytically active in the endoplasmic reticulum and trans Golgi. Whether TIMPs play a role in proteolytic inhibition intracellularly or at the cell surface, or subserve some MMP-independent function in PD and HD is yet to be determined. Altered levels of MMPs and TIMPs have been implicated in several pathological states of the nervous system. Although largely recognised for their MMP-inhibitory activities, it is clear that TIMPs are multifaceted molecules that exert functions independently of MMP inhibition. Future studies need to be focused on the identiﬁcation of the MMP-independent ligands to which TIMPs bind in the hopes of developing target therapeutics that do not negatively inﬂuence matrix integrity. References 1. Yong, V. W., Power, C., Forsyth, P., and Edwards, D. R. (2001) Nat Rev Neurosci 2, 502–511 2. Baker, A. H., Edwards, D. R., and Murphy, G. (2002) J Cell Sci 115, 3719–3727 3. Gardner, J., and Ghorpade, A. (2003) J Neurosci Res 74, 801–806 4. Crocker, S. J., Pagenstecher, A., and Campbell, I. L. (2004) J Neurosci Res 75, 1–11 5. Young, D. A., Phillips, B. W., Lundy, C., Nuttall, R. K., Hogan, A., Schultz, G. A., Leco, K. J., Clark, I. M., and Edwards, D. R. (2002) Biochem J 364, 89–99

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

218

D. Jaworski

6. Leco, K. J., Apte, S. S., Taniguchi, G. T., Hawkes, S. P., Khokha, R., Schultz, G. A., and Edwards, D. R. (1997) FEBS Lett 401, 213–217 7. Edwards, D. R., Rocheleau, H., Sharma, R. R., Wills, A. J., Cowie, A., Hassell, J. A., and Heath, J. K. (1992) Biochem Biophys Acta 1171, 41–55 8. Botelho, F. M., Edwards, D. R., and Richards, C. D. (1998) J Biol Chem 273, 5211–5218 9. Jaworski, J., Biedermann, I. W., Lapinska, J., Szklarczyk, A., Figiel, I., Konopka, D., Nowicka, D., Filipkowski, R. K., Hetman, M., Kowalczyk, A., and Kaczmarek, L. (1999) J Biol Chem 274, 28106–28112 10. Fager, N., and Jaworski, D. M. (2000) Mech Dev 98, 105–109 11. Rivera, S., Tremblay, E., Timsit, S., Canals, O., Ben-Ari, Y., and Khrestchatisky, M. (1997) J Neurosci 17, 4223–4235 12. Docherty, A. J., Lyons, A., Smith, B. J., Wright, E. M., Stephens, P. E., Harris, T. J. R., Murphy, G., and Reynolds, J. J. (1985) Nature 318, 66–69 13. Hayakawa, T., Yamashita, K., Tanzawa, K., Uchijima, E., and Iwata, K. (1992) FEBS Lett 298, 29–32 14. Ritter, L. M., Garﬁeld, S. H., and Thorgeirsson, U. P. (1999) Biochem Biophys Res Commun 257, 494–499 15. Zhao, W. Q., Li, H., Yamashita, K., Guo, X. K., Hoshino, T., Yoshida, S., Shinya, T., and Hayakawa, T. (1998) J Cell Sci 111, 1147–1153 16. Vaillant, C., Didier-Baz`es, M., Hutter, A., Belin, M.-F., and Thomasset, N. (1999) J Neurosci 19, 4994–5004 17. Blavier, L., and DeClerck, Y. A. (1997) Mol Biol Cell 8, 1513–1527 18. Pagenstecher, A., Stalder, A. K., Kincaid, C. L., Shapiro, S. D., and Campbell, I. L. (1998) Am J Pathol 152, 729–741 19. Stetler-Stevenson, W. G., Bersch, N., and Golde, D. W. (1992) FEBS Lett 296, 231–234 20. Hayakawa, T., Yamashita, K., Ohuchi, E., and Shinagawa, A. (1994) J Cell Sci 107, 2373–2379 21. Murphy, A. N., Unsworth, E. J., and Stetler-Stevenson, W. G. (1993) J Cell Physiol 157, 351–358 22. Hoegy, S. E., Oh, H. R., Corcoran, M. L., and Stetler-Stevenson, W. G. (2001) J Biol Chem 276, 3203–3214 23. Seo, D. W., Li, H., Guedez, L., Wingﬁeld, P. T., Diaz, T., Salloum, R., Wei, B. Y., and Stetler-Stevenson, W. G. (2003) Cell 114, 171–80 24. P´erez-Mart´ınez, L., and Jaworski, D. M. (2004) submitted 25. Apte, S. S., Hayashi, K., Seldin, M. F., Mattei, M. G., Hayashi, M., and Olsen, B. R. (1994) Dev Dynamics 200, 177–197 26. Jaworski, D. M., and Fager, N. (2000) J Neurosci Res 61, 396–408 27. Bugno, M., Witek, B., Bereta, J., Bereta, M., Edwards, D. R., and Kordula, T. (1999) FEBS Lett 448, 9–14 28. Muir, E. M., Adcock, K. H., Morgenstern, D. A., Clayton, R., von Stillfried, N., Rhodes, K., Ellis, C., Fawcett, J. W., and Rogers, J. H. (2002) Mol Brain Res 100, 103–117

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

TIMPs in CNS Development

219

29. Bian, J., Wang, Y., Smith, M. R., Kim, H., Jacobs, C., Jackman, J., Kung, H. F., Colburn, N. H., and Sun, Y. (1996) Carcinogenesis 17, 1805–1811 30. Ahonen, M., Baker, A. H., and K¨ ah¨ ari, V.-M. (1998) Cancer Res 58, 2310–2315 31. Baker, A. H., George, S. J., Zaltsman, A. B., Murphy, G., and Newby, A. C. (1999) British Journal of Cancer 79, 1347–1355 32. Ahonen, M., Poukkula, M., Baker, A. H., Kashiwagi, M., Nagase, H., Eriksson, J. E., and Kahari, V. M. (2003) Oncogene 22, 2121–2134 33. Wallace, J. A., Alexander, S., Estrada, E. Y., Hines, C., Cunningham, L. A., and Rosenberg, G. A. (2002) J Cereb Blood Flow Metab 22, 1303–1310 34. Wetzel, M., Rosenberg, G. A., and Cunningham, L. A. (2003) Eur J Neurosci 18, 1050–1060 35. Morshead, C. M., and van der Kooy, D. (1992) J Neurosci 12, 249–256 36. Thomaidou, D., Mione, M. C., Cavanagh, J. F. R., and Parnavelas, J. G. (1997) J Neurosci 17, 1075–1085 37. Rakic, P. (1971) Brain Res 33, 471–476 38. Kishi, K., Peng, J. Y., Kakuta, S., Murakami, K., Kuroda, M., Yokota, S., Hayakawa, S., Kuge, T., and Asayama, T. (1990) Arch Histol Cytol 53, 219–226 39. Wichterle, H., Garcia-Verdugo, J. M., and Alvarez-Buylla, A. (1997) Neuron 18, 779–791 40. Anand-Apte, B., Bao, L., Smith, R., Iwata, K., Olsen, B. R., Zetter, B., and Apte, S. S. (1996) Biochem Cell Biol 74, 853–862 41. Yang, T.-T., and Hawkes, S. P. (1992) Proc Natl Acad Sci 89, 10676–10680 42. B´ anyai, L., and Patthy, L. (1999) Protein Sci 8, 1636–1642 43. Luo, Y., Cai, J., Liu, Y., Xue, H., Chrest, F. J., Wersto, R. P., and Rao, M. (2002) J Neurochem 83, 1481–1497 44. Fr¨ olichsthal-Schoeller, P., Vescovi, A. L., Krekoski, C. A., Murphy, G., Edwards, D. R., and Forsyth, P. (1999) Neuroreport 10, 345–351 45. Kaczmarek, L., Lapinska-Dzwonek, J., and Szymczak, S. (2002) EMBO J 21, 6643–6648 46. Dityatev, A., and Schachner, M. (2003) Nat Rev Neurosci 4, 456–468 47. Shiosaka, S., and Yoshida, S. (2000) Neurosci Res 37, 85–89 48. Tomimatsu, Y., Idemoto, S., Moriguchi, S., Watanabe, S., and Nakanishi, H. (2002) Life Sci 72, 355–361 49. Ben-Ari, Y., and Represa, A. (1990) Trends Neurosci 13, 312–318 50. Nedivi, E., Hevroni, D., Naot, D., Israeli, D., and Citri, Y. (1993) Nature 363, 718–722 51. Ressler, K. J., Paschall, G., Zhou, X. L., and Davis, M. (2002) J Neurosci 22, 7892–7902 52. von Gertten, C., Holmin, S., Mathiesen, T., and Nordqvist, A. C. (2003) J Neurosci Res 73, 803–810 53. Zhang, J. W., Deb, S., and Gottschall, P. E. (1998) Eur J Neurosci 10, 3358–3368

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

220

D. Jaworski

54. Zhang, J. W., Deb, S., and Gottschall, P. E. (2000) Neurosci Lett 295, 9–12 55. Szklarczyk, A., Lapinska, J., Rylski, M., McKay, R. D. G., and Kaczmarek, L. (2002) J Neurosci 22, 920–930 56. Tan, H. K., Heywood, D., Ralph, G. S., Bienemann, A., Baker, A. H., and Uney, J. B. (2003) Mol Cell Neurosci 22, 98–106 57. Davis, M., Falls, W. A., Campeau, S., and Kim, M. (1993) Behav Brain Res 58, 175–198 58. Davis, M. (1989) Behav Neurosci 103, 495–503 59. Davis, M. (1997) J Neuropsychiatry Clin Neurosci 9, 382–402 60. Jaworski, D. M., Boone, J., Caterina, J., Soloway, P., and Falls, W. A. (2004) submitted 61. Pohar, N., Godenschwege, T. A., and Buchner, E. (1999) Genomics 57, 293–296 62. Yu, W. P., Brenner, S., and Venkatesh, B. (2003) Trends Genet 19, 180–183 63. Rahkonen, O. P., Koskivirta, I. M., Oksjoki, S. M., Jokinen, E., and Vuorio, E. I. (2002) Biochem Biophys Acta 1577, 45–52 64. Ferreira, A., and Rapoport, M. (2002) Cell Mol Life Sci 59, 589–595 65. Nothnick, W. B., Soloway, P., and Curry, J., T.E. (1997) Biol Reprod 56, 1181–1188 66. Nothnick, W. B. (2000) Biol Reprod 63, 905–912 67. Creemers, E. E., Davis, J. N., Parkhurst, A. M., Leenders, P., Dowdy, K. B., Hapke, E., Hauet, A. M., Escobar, P. G., Cleutjens, J. P., Smits, J. F., Daemen, M. J., Zile, M. R., and Spinale, F. G. (2003) Am J Physiol Heart Circ Physiol 284, H364–371 68. Lijnen, H. R., Demeulemeester, D., Van Hoef, B., Collen, D., and Maquoi, E. (2003) Thromb Haemost 89, 249–255 69. Osiewicz, K., McGarry, M., and Soloway, P. D. (1999) Ann NY Acad Sci 878, 494–496 70. Wang, Z., Juttermann, R., and Soloway, P. D. (2000) J Biol Chem 275, 26411–26415 71. Caterina, J. J., Yamada, S., Caterina, N. C. M., Longenecker, G., Holmb¨ack, K., Shi, J., Yermovsky, A. E., Engler, J. A., and Birkedal-Hansen, H. (2000) J Biol Chem 275, 26416–26422 72. Gill, S. E., Pape, M. C., Khokha, R., Watson, A. J., and Leco, K. J. (2003) Dev Biol 261, 313–23 73. Leco, K. J., Waterhouse, P., Sanchez, O. H., Gowing, K. L. M., Poole, A. R., Wakeman, A., and Khokha, R. (2001) J Clin Invest 108, 817–829 74. Fata, J. E., Leco, K. J., Voura, E. B., Yu, H. Y. E., Waterhouse, P., Murphy, G., Moorehead, R. A., and Khokha, R. (2001) J Clin Invest 108, 831–841 75. Reizes, O., Benoit, S. C., Strader, A. D., Clegg, D. J., Akunuru, S., and Seeley, R. J. (2003) Ann NY Acad Sci 994, 66–73 76. Wesolowski, S. R., Allan, M. F., Nielsen, M. K., and Pomp, D. (2003) Physiol Genomics 13, 129–137

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

TIMPs in CNS Development

221

77. Jaworski, D. M., Soloway, P., Caterina, J., and Falls, W. A. (2004) submitted 78. Weber, B. H., Vogt, G., Pruett, R., Stohr, H., and Felbor, U. (1994) Nat Genet 8, 352–356 79. Felbor, U., Stohr, H., Amann, T., Schonherr, U., and Weber, B. H. (1995) Hum Mol Genet 4, 2415–2416 80. Felbor, U., Stohr, H., Amann, T., Schonherr, U., Apfelstedt-Sylla, E., and Weber, B. H. (1996) J Med Genet 33, 233–236 81. Felbor, U., Suvanto, E. A., Forsius, H. R., Eriksson, A. W., and Weber, B. H. (1997) Am J Hum Genet 60, 57–62 82. Jacobson, S. G., Cideciyan, A. V., Bennett, J., Kingsley, R. M., Sheﬃeld, V. C., and Stone, E. M. (2002) Arch Ophthalmol 120, 376–379 83. Langton, K. P., McKie, N., Curtis, A., Goodship, J. A., Bond, P. M., Barker, M. D., and Clarke, M. (2000) J Biol Chem 275, 27027–27031 84. Tabata, Y., Isashiki, Y., Kamimura, K., Nakao, K., and Ohba, N. (1998) Hum Genet 103, 179–182 85. Clarke, M., Mitchell, K. W., Goodship, J., McDonnell, S., Barker, M. D., Griﬃths, I. D., and McKie, N. (2001) Br J Ophthalmol 85, 1429–1431 86. Assink, J. J., de Backer, E., ten Brink, J. B., Kohno, T., de Jong, P. T., Bergen, A. A., and Meire, F. (2000) Br J Ophthalmol 84, 682–686 87. Qi, J. H., Ebrahem, Q., Yeow, K., Edwards, D. R., Fox, P. L., and Anand-Apte, B. (2002) J Biol Chem 277, 13394–13400 88. Langton, K. P., Barker, M. D., and McKie, N. (1998) J Biol Chem 273, 16778–16781 89. Yeow, K. M., Kishnani, N. S., Hutton, M., Hawkes, S. P., Murphy, G., and Edwards, D. R. (2002) Matrix Biol 21, 75–88 90. Chong, N. H., Kvanta, A., Seregard, S., Bird, A. C., Luthert, P. J., and Steen, B. (2003) Am J Ophthalmol 136, 954–955 91. Chang, C., and Werb, Z. (2001) Trends Cell Biol 11, S37–43 92. Nakagawa, T., Kubota, T., Kabuto, M., Sato, K., Arai, Y., and Kodera, T. (1995) Neurol Med Chir (Tokyo) 35, 728–731 93. Groft, L. L., Muzik, H., Rewcastle, N. B., Johnston, R. N., Knauper, V., Laﬂeur, M. A., Forsyth, P. A., and Edwards, D. R. (2001) Br J Cancer 85, 55–63 94. Lampert, K., Machein, U., Machein, M. R., Conca, W., Peter, H. H., and Volk, B. (1998) Am J Pathol 153, 429–437 95. Mohanam, S., Wang, S. W., Rayford, A., Yamamoto, M., Sawaya, R., Nakajima, M., Liotta, L. A., Nicolson, G. L., Stetler-Stevenson, W. G., and Rao, J. S. (1995) Clin Exp Metastasis 13, 57–62 96. Chesler, L., Golde, D. W., Bersch, N., and Johnson, M. D. (1995) Blood 86, 4506–4515 97. Pavlaki, M., and Zucker, S. (2003) Cancer Metastasis Rev 22, 177–203 98. Pagenstecher, A., Wussler, E. M., Opdenakker, G., Volk, B., and Campbell, I. L. (2001) J Neuropath Exp Neurol 60, 598–612 99. Yamada, E., Tobe, T., Yamada, H., Okamoto, N., Zack, D., Werb, Z., Soloway, P. D., and Campochiaro, P. A. (2001) Histol Histopathol 16, 87–97

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

222

D. Jaworski

100. Anand-Apte, B., Pepper, M. S., Voest, E., Montesano, R., Olsen, B., Murphy, G., Apte, S. S., and Zetter, B. (1997) Invest Ophthalmol Vis Sci 38, 817–823 101. Qi, J. H., Ebrahem, Q., Moore, N., Murphy, G., Claesson-Welsh, L., Bond, M., Baker, A., and Anand-Apte, B. (2003) Nat Med 9, 407–415 102. Ahonen, M., Ala-Aho, R., Baker, A. H., George, S. J., Grenman, R., Saarialho-Kere, U., and Kahari, V. M. (2002) Mol Ther 5, 705–715 103. Spurbeck, W. W., Ng, C., Vanin, E. F., and Davidoﬀ, A. M. (2003) Cancer Gene Ther 10, 161–167 104. Brooks, P. C., Stromblad, S., Sanders, L. C., von Schalscha, T. L., Aimes, R. T., Stetler-Stevenson, W. G., Quigley, J. P., and Cheresh, D. A. (1996) Cell 85, 683–693 105. VanMeter, T. E., Rooprai, H. K., Kibble, M. M., Fillmore, H. L., Broaddus, W. C., and Pilkington, G. J. (2001) J Neurooncol 53, 213–235 106. Hornebeck, W., Emonard, H., Monboisse, J. C., and Bellon, G. (2002) Semin Cancer Biol 12, 231–241 107. Stanton, H., Gavrilovic, J., Atkinson, S., d’Ortho, M. P., Yamada, K. M., Zardi, L., and Murphy, G. (1998) J Cell Sci 111, 2789–2798 108. Hofmann, U. B., Westphal, J. R., Van Kraats, A. A., Rutter, D. J., and Van Muijen, G. N. P. (2000) Int J Cancer 87, 12–19 109. Deryugina, E. I., Ratnikov, B., Monosov, E., Postnova, T. I., DiScipio, R., Smith, J. W., and Strongin, A. Y. (2001) Exp Cell Res 263, 209–223 110. Jaworski, D. M., Kelly, G. M., Piepmeier, J. M., and Hockﬁeld, S. (1996) Cancer Res 56, 2293–2298 111. Zhang, H., Kelly, G., Zerillo, C., Jaworski, D. M., and Hockﬁeld, S. (1998) J Neurosci 18, 2370–2376 112. Matthews, R. T., Gary, S. C., Zerillo, C., Pratta, M., Solomon, K., Arner, E. C., and Hockﬁeld, S. (2000) J Biol Chem 275, 22695–22703 113. Kashiwagi, M., Tortorella, M., Nagase, H., and Brew, K. (2001) J Biol Chem 276, 12501–12504. 114. Esteller, M., Corn, P. G., Baylin, S. B., and Herman, J. G. (2001) Cancer Res 61, 3225–3229 115. Gonzalez-Gomez, P., Bello, M. J., Arjona, D., Lomas, J., Alonso, M., De Campos, J. M., Vaquero, J., Isla, A., Gutierrez, M., and Rey, J. A. (2003) Int J Oncol 22, 601–608 116. La Fleur, M., Underwood, J. L., Rappolee, D. A., and Werb, Z. (1996) J Exp Med 184, 2311–2326 117. Platt, C. I., Krekoski, C. A., Ward, R. V., Edwards, D. R., and Gavrilovic, J. (2003) J Neurosci Res 74, 417–429 118. Hu, J., Fink, D., and Mata, M. (2002) Eur J Neurosci 16, 1409–1416 119. Tachibana, T., Noguchi, K., and Ruda, M. A. (2002) Neurosci Lett 327, 133–137 120. Jaworski, D. M. (2000) Glia 30, 199–208 121. Rathke-Hartlieb, S., Budde, P., Ewert, S., Schlomann, U., Staege, M. S., KJockusch, H., Bartsch, J. W., and Frey, J. (2000) FEBS Lett 481, 227–234

May 26, 2005

16:5 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch08

TIMPs in CNS Development

223

122. Ait-Ikhlef, A., Hantaz-Ambroise, D., Jacque, C., Belkadi, L., and Rieger, F. (1999) Cell Mol Biol 45, 393–400 123. Tsukatani, T., Fillmore, H. L., Hamilton, H. R., Holbrook, E. H., and Costanzo, R. M. (2003) Neuroreport 14, 1135–1140 124. Lorenzl, S., Albers, D. S., LeWitt, P. A., Chirichigno, J. W., Hilgenberg, S. L., Cudkowicz, M. E., and Beal, M. F. (2003) J Neurol Sci 207, 71–76 125. Lorenzl, S., Albers, D. S., Narr, S., Chirichigno, J., and Beal, M. F. (2002) Exp Neurol 178, 13–20 126. Peress, N., Perillo, E., and Zucker, S. (1995) J Neuropathol Exp Neurol 54, 16–22 127. Wollmer, M. A., Papassotiropoulos, A., Streﬀer, J. R., Grimaldi, L. M., Kapaki, E., Salani, G., Paraskevas, G. P., Maddalena, A., de Quervain, D., Bieber, C., Umbricht, D., Lemke, U., Bosshardt, S., Degonda, N., Henke, K., Hegi, T., Jung, H. H., Pasch, T., Hock, C., and Nitsch, R. M. (2002) Psychiatr Genet 12, 155–60 128. Beuche, W., Yushchenko, M., Mader, M., Maliszewska, M., Felgenhauer, K., and Weber, F. (2000) Neuroreport 11, 3419–3422 129. Ling, Y., Morgan, K., and Kalsheker, N. (2003) Int J Biochem Cell Biol 35, 1505–1535 130. Miyazaki, K., Hasegawa, M., Funahashi, K., and Umeda, M. (1993) Nature 362, 839–841

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

This page intentionally left blank

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

PART V MMPs AND TIMPs IN DISEASE

225

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

This page intentionally left blank

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

CHAPTER 9 MATRIX METALLOPROTEINASES IN CEREBRAL ISCHEMIA

M. Wetzel, L.A. Cunningham and G.A. Rosenberg∗ Departments of Neurology, Neurosciences, Cell Biology and Physiology University of New Mexico Health Sciences Center, Albuquerque, New Mexico 87131, USA E-mail: ∗[email protected]

1. Introduction Cerebral ischemia initiates a cascade of molecular events that leads to central nervous system (CNS) injury. Neurons are most vulnerable to hypoxia/ischemia and die by necrosis and apoptosis (1–3). Astrocytes and blood vessels are more resistant to cerebral ischemia, but eventually succumb to the loss of oxygen. Loss of energy stores results in cellular swelling. Cytotoxic edema begins very early, and damages the vasculature, which results in vasogenic edema with hemorrhage (4). Temporary cerebral artery occlusion with restoration of blood ﬂow, or reperfusion produces more profound damage on cerebral blood vessels than permanent occlusion, resulting in compromise of the blood-brain barrier (BBB) (5–9). As the molecular cascade of injury unfolds, proteases and free radicals are produced, acting as a ﬁnal common pathway for cerebral damage. Several neutral protease gene families play a critical role in this process, including serine proteases (plasmin), cysteine proteases (caspases) and zinc-dependent proteases (matrix metalloproteinases, MMPs). Plasminogen/plasmin are secreted serine proteases important in intravascular and extravascular coagulation (10). Caspases are cysteine proteases that act intracellularly and their activation is a hallmark of apoptotic cell death (3). MMPs are a gene family of over 20 zinc-dependent extracellular endopeptidases widely recognised for their potent ability to degrade all structural components Correspondence to: G.A. Rosenberg 227

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

228

M. Wetzel et al

of the extracellular matrix (ECM) (11). Studies by others and us have demonstrated that the MMPs play several important roles in cerebral ischemia (12). We have shown that a major pathological eﬀect of MMP activity is the disruption of cerebral blood vessels, increasing their permeability and subsequently compromising the BBB (13). Emerging evidence reveals another important role of MMPs in the CNS, as MMP activity inﬂuences neuronal survival and death decisions (14, 15). MMPs aﬀect neuronal death by either the direct action on perineuronal ECM (16) or by aﬀecting proteolytic shedding of extracellular death receptors and death ligands on the cell surface (15). The ﬁrst MMPs described in brain injury were gelatinases A and B (72and 92-kDa type IV collagenases or MMP-2 and MMP-9, respectively) (17). A number of laboratories have demonstrated the importance of MMPs in neuroinﬂammation (18, 19). The action of MMPs is balanced by tissue inhibitors to metalloproteinases (TIMPs) (20). In addition to their role in controlling excessive MMP proteolysis, TIMPs regulate cell survival and apoptosis (21). TIMP-3 is expressed in ischemic brain (14) where it may facilitate neuronal apoptosis (15). This review will focus on the role of MMPs in proteolytic disruption of the BBB and of TIMP-3 in neuronal apoptosis. Several reviews are available on MMPs in general and on the role of MMPs in the central nervous system (12, 18, 19, 22).

2. Matrix Metalloproteinase Biology Structurally, MMPs are organised into distinct domains. MMPs contain an N -terminal short signal sequence that directs synthesis to the endoplasmic reticulum. A propeptide domain that maintains the enzyme in an inactive state follows the signal sequence. Next is the catalytic domain that is linked by a hinge region to the C-terminal domain that is similar to hemopexin domain and is involved in ECM substrate binding (23). MMPs are classiﬁed into four major groups based on protein structure (11). Matrilysin (MMP-7) is the smallest of the MMPs, containing only the propeptide region and catalytic site. Stromelysins (MMP-3 and -10) add a hemopexin domain. Gelatinases (MMP-2 and -9) have ﬁbronectin binding sites that directs them to the basal lamina around the blood vessels, which contains ﬁbronectin and laminin. While stromelysins and gelatinases are secreted extracellularly, membrane-type MMPs (MT-MMPs) are not secreted, but anchored to the membrane by glycosylphosphatidylinositol (GPI) anchor

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

Matrix Metalloproteinases in Cerebral Ischemia

229

or exist as transmembrane proteins. MT-MMPs are important in the activation of constitutively produced MMP-2 (24). Rampant MMP activity leads to widespread tissue destruction. Hence, MMPs are regulated at multiple levels — transcriptional, posttranscriptional, post-translational, and endogenous inhibitors have been described. Transcription of most MMPs is activated or inhibited by cytokines, growth factors, chemokines, phorbol esters, oncogenes, cell-cell or cell-ECM interactions, cell stress, and changes in cell shape (18, 25). In the central nervous system (CNS), MMP-2 is constitutively expressed, and is normally found in resting adult brain tissue and cerebrospinal ﬂuid (26). MMP-2 promoter region structure suggests it is a housekeeping gene, involved in the slow turnover of the extracellular matrix. On the contrary, MMP-9 is normally present at low levels and becomes up-regulated in response to inﬂammatory stimuli. MMP-3 is also present normally at low concentrations, but is induced by cerebral injury. MMP-3 and -9 contain AP-1 and NF-κB sites in the promoter regions which respond to inﬂammatory mediators, such as, tumour necrosis factor-α (TNF-α) and interleukin-1β (Il-1β). Post-transcriptionally, mRNA encoding MMP-3 is stabilised by phorbol esters and epidermal growth factor (EGF) (27). Once translated, most MMPs are constitutively secreted as inactive proenzymes (zymogens), and maintained in a latent state by the interaction of the active site zinc and the cysteine near the C-terminal end of the pro-peptide (28). MMP activation requires that the cysteine to zinc bond be disrupted by proteolytic removal of the pro-peptide domain (Fig. 1). Pro-MMP-2 is activated by membrane bound MT-MMP (24). Plasminogen/plasmin or the proconvertase, furin, initiates the process by the activation of MT-MMP (29). Plasmin activates pro-MMP-3 (30, 31). Pro-MMP-9 is activated by MMP-3 (32). Recent evidence implicates the free radicals, nitric oxide (NO) and reactive oxygen species (ROS), in the activation of the MMPs. NO activates MMP-9 by a nitrosylation process that appears to involve the propeptide region (16). 3. MMPs and Neuroinﬂammation Numerous studies have demonstrated the importance of MMPs in neuroinﬂammation secondary to ischemia, trauma, infection, and immunological reactions (18, 19). A major pathological eﬀect of MMP activity is the disruption of cerebral blood vessel integrity, increasing their permeability, and resulting in BBB compromise. One consequence of disrupting

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

230

M. Wetzel et al

Fig. 1. Mechanism of activation of MMP-2, -3 and -9. In the gene promoter region in the nucleus genes for MMP-2 have AP-2 and SP-1 sites, and genes for MMP-3 and -9 have AP-1 and NF-κB sites. After the mRNA is made, it is transcribed into latent protein. Pro-MMP-2 is activated by MT-MMP, which is activated by plasmin and furin. ProMMP-3 is activated by plasmin, and pro-MMP-9 by MMP-3 or nitric oxide (NO). They act on the basal lamina around the cerebal vessels in the extracellular matrix (ECM) to disrupt the blood-brain barrier (BBB).

the cerebral vascular basal lamina is the breakdown of the blood-brain barrier (BBB), which separates the CNS from the periphery. In support of this, intracerebral injection of MMP-2 results in the opening of the BBB with subsequent hemorrhage around the blood vessels (13). Tissue inhibitor of metalloproteinases-2 (TIMP-2) blocks this opening of the BBB, demonstrating that MMPs are involved in this process. MMPs attack macromolecules of the basal lamina surrounding blood vessels, including type IV collagen, laminin, ﬁbronectin, and heparan sulfate. For example, laminin is degraded in the infarct region in a nonhuman primate model of stroke (33). The loss of laminin, was correlated with the extent of hemorrhagic transformation; MMP-9 levels were associated with hemorrhagic transformation, while MMP-2 levels correlated with neuronal injury (34). In addition to the cerebral vascular basal lamina, the integrity of the BBB is maintained by tight junctions between endothelial cells. These cerebral endothelial tight junctions are composed of structural proteins, including

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

Matrix Metalloproteinases in Cerebral Ischemia

231

occludins and claudin (35, 36). Recently, MMP-9 was shown to degrade proteins that form the tight junctions between endothelial cells, such as zona occludens-1 (37). Biochemical studies of MMPs in permanent and temporary ischemia have shown that the gelatinases contribute to the disruption of the BBB that leads to vasogenic cerebral edema and hemorrhage (12). Permanent middle cerebral artery occlusion (pMCAO) in spontaneously hypertensive rats (SHR) resulted in the production of MMP-9 by 24 hrs, and a marked increase in MMP-2 by 5 days (38). The increase in MMP-9 correlates with the time of maximal damage to the BBB, while the later increase in MMP-2 is related to the increase in reactive astrocytes around the cyst. Transient middle cerebral artery occlusion (tMCAO) for 90 min with reperfusion in SHR caused a biphasic opening of the BBB with a transient opening at 3 hrs and a second more severe injury at 48 hrs; the early opening at 3 hrs correlated with an increase in MMP-2, while a later opening at the 48 hrs correlated with increased expression of MMP-9 (Fig. 2) (39). A hydroxymate MMP inhibitor, BB-1101 (British Biotechnology, Oxon, UK) blocked the edema at 24 hrs and the early opening of the BBB at 3 hrs, but failed to alter the secondary opening of the BBB at 48 hrs. Pharmacological and gene deletion studies provide further support of elevated MMP activity contributing to the pathophysiology of cerebral ischemia. MMP-9 knock out mice are resistant to transient focal ischemia and exhibit reduced edema (40). Alternatively, MMP-2 knock out and wild type mice are equivalently vulnerable to transient focal ischemia (41), suggesting that MMP-mediated proteolytic damage following stroke may not simply be a consequence of general MMP activity, but may depend on speciﬁc MMPs. Thus, there is considerable evidence to support a role for the MMPs in pathological changes associated with cerebral ischemia, particularly to the changes that occur in the cerebral vasculature. Human studies have shown the presence of MMPs in the tissues of stroke patients (42, 43). MMP-2 and MMP-9 levels were measurable by ELISA in the serum of stroke patients with cardioembolic stroke. Patients with severe strokes had signiﬁcantly higher levels of MMP-9 at baseline, and the elevated levels persisted for 48 hrs (44). The highest levels of MMP-9 were found in patients with delayed hemorrhagic transformation (45). Currently, there is one approved treatment for stroke patients, recombinant tissue plasminogen activator (rtPA). While rtPA improves outcome in patients with cerebral ischemia by proteolytically digesting clots, it increases the risk of intracerebral hemorrhage (46). Several recent studies have revealed that

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

232

M. Wetzel et al

Fig. 2. (A) Astrocyte immunostained with MMP-2 and counterstained with diaminobenzadine (DAB). The foot process is seen surrounding a small blood vessel, while the cell body is not stained. (B) The time course of the ischemic cascade in reperfusion injury. Time is shown on the vertical axis. Ischemia starts at time zero with reperfusion in several hours. At the start of ischemia there is activation of pro-MT-MMP and MMP-2, which causes a reversible opening of the blood-brain barrier (BBB). If the hypoxia/ischemia is severe, amplification of the reaction occurs with the formation cytokines, the deposition of fibrin, and the extravasation of white blood cells. Induction of MMP-3 and -9 causes a more severe opening of the BBB with brain edema and hemorrhage. Adapted from (54).

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

Matrix Metalloproteinases in Cerebral Ischemia

233

MMP inhibitors reduce injury to the BBB and hemorrhage secondary to rtPA. Hemorrhagic infarctions secondary to multiple emboli infused into the carotid artery of rabbits was reduced with a MMP inhibitor, BB-94 (47). Prolonged ischemia prior to reperfusion increased the risk of hemorrhages with a high incidence of mortality in the animals with hemorrhages. In a recent study in rodents, BB-94 dramatically reduced mortality after reperfusion (48). Treatment of rats with rtPA in a rat model of stroke with a 6-hr middle cerebral artery occlusion and 18 hrs of reperfusion resulted in a high mortality rate, which could be dramatically reduced by closing the BBB with the inhibitor to MMPs, BB-94 (49). Thus, this demonstrates the importance of MMPs in cerebral ischemia and its treatment. 4. MMPs and the Neurovascular Unit The role of the ECM in maintaining the integrity of the cerebral vessels and the BBB emphasises the involvement of MMPs in modifying these ECM properties. Several components are thought to make up the BBB. These include endothelial cells with tight junctions, basal lamina around the epithelial-like brain blood vessels, pericytes within the basal lamina, and astrocytes (50). Astrocytes are critical for the development of tight-junctions cerebral blood vessels (51). This has been referred to as the gliovascular or neurovascular unit. While the function of each of the components remains uncertain, it appears that the damage to any of the components impacts the whole. Cell culture studies have aided in the delineation of the role of MMPs in these individual cell types (52). Zymography has shown that astrocytes are a rich source of MMP-2 and when stimulated with noxious substrates, they can form MMP-9. Microglia secrete MMP-9 into the culture medium when stimulated by lipopolysaccharide (LPS) (53). Immunohistochemical studies show that MMP-2 is present in the astrocytic foot processes that form the glial sheath of capillaries (54). The pericyte next to the capillary forms MMP-3 and MMP-9, and has a macrophage function. Neurons have MMP-2 and MMP-3 and can form a small amount of MMP-9 (Liu et al, unpublished data). In ischemia endogenous MMP-9 is produced by CNS resident cells, such as astrocytes and microglial, while neutrophils can extravasate into the CNS, release proMMP-9 and activate it with free radicals released during the respiratory burst (55). The cellular localisation of MMPs suggests that normally MMP-2 and MT-MMP in the basal lamina region acts in

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

234

M. Wetzel et al

a spatially constrained manner to modulate the permeability of the BBB. Following an ischemic insult, there is release of MMP-9 and MMP-3 which amplify the damage to the BBB directly after undergoing activation or MMP-3 does this indirectly through the activation of MMP-9. The role of the neuron in this process remains to be determined. During the ischemic phase, MMPs are no longer spatially constrained to a discreet region of the membrane, and the resulting extracellular matrix disruption is more extensive. Free radicals are more likely to play a signiﬁcant role at this stage. Hence, MMPs play a critical role altering cerebral vascular permeability and BBB integrity.

5. Cell Death Following Cerebral Ischemia Away from the neurovascular interface at the neuronal surface, extracellular proteolysis inﬂuences neuronal outcome following cerebral ischemia. Ischemia initiates immediate and irreversible necrotic cell death at the ischemic core. Organelle and cell swelling, ruptured cell membranes, and release of intracellular contents into the surrounding area characterise this rapid necrotic cell death. The spilling of cellular contents ampliﬁes the initial damage through the initiation of deleterious secondary cascades, including the inﬂammatory response (2). The penumbral zone surrounding the initial infarct is at risk of progressing towards neuronal death and enhancing the ultimate infarct size. In the penumbral region, apoptotic cell death is delayed and characterised by internucleosomal cleavage of DNA, nuclear condensation and cell shrinkage, and the organelles and plasma membranes remain intact. The cytoplasmic contents and fragmented DNA are partitioned into membrane bound vesicles that are rapidly phagocytised. Thus, by avoiding cell rupture and promptly removing self-contained cellular debris, apoptotic cell death minimises damage to adjacent tissue (56). There are similarities shared by the two cell death processes which may lead to a blurring of the distinction between necrosis and apoptosis (57). Brain tissue of the ischemic core undergoing necrotic cell death is lethally damaged. However, the penumbral tissue may be salvaged, as apoptosis is an active process that may be modulated and provides an opportunity for therapeutic intervention. In the case of cerebral ischemia, the apoptotic cascade is an attractive target for pharmacological intervention. The ultrastructural changes displayed during apoptosis are a consequence of intricate biochemical processes performed by a series of caspases, which are cysteine proteases with an absolute speciﬁcity to cleave

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

Matrix Metalloproteinases in Cerebral Ischemia

235

after aspartate residues; they mediate apoptosis by inactivating proteins that protect cells, directly disassembling cellular structures, and deregulating protein activity (58). Caspase activation may be induced by an array of signals and executed by two principal intracellular signal transduction cascades. One route originates from mitochondria. Once the apoptotic signal is registered by the mitochondria, the pro-apoptotic Bcl-2 family member, Bax releases cytochrome c from the mitochondria into the cytosol. When cytochrome c is present in the cytosol, it forms a complex with apoptotic protease activating factor-1 (APAF-1), deoxyadenosine trisphosphate, and procaspase-9 to form the apoptosome (59). The assembly of the apoptosome results in caspase-9 activation, subsequently leading to caspase-3 activation and cellular demise. The mitochondrial pathway is inhibited by anti-apoptotic Bcl-2 family members, which interfere with the release of cytochrome c from the mitochondria or with the formation of the apoptosome. The second pathway is induced by the activation of cell surface transmembrane death receptors (Fig. 3). Death receptors are members of the

FasL Fas

FADD Procaspase-8

MITOCHONDRIA Cytochrome c

Bid

Caspase-8

Procaspase-3

Cytochrome c APAF-1 Procaspase-9 Caspase-9

Caspase-3 APOPTOSIS Fig. 3. Two apoptotic pathways lead to caspase activation and cellular demise. Apoptosis may be activated at the level of cell surface death receptors or the mitochondria. Both apoptotic pathways lead to activation of distinct initiator caspases that ultimately lead to capase-3 activation and cellular disassembly. The Bcl-2 family member, Bid, links the two pathways and intensiﬁes the apoptotic signal.

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

236

M. Wetzel et al

tumour necrosis factor receptor (TNFR) superfamily that transduce apoptosis when bound by their respective ligands via intracellular signal transduction cascades. Members of the TNFR superfamily of death receptors include, TNFR1, Fas (APO-1/CD95), TNF related apoptosis inducing ligand receptor1 (TRAILR1/DR4), TRAILR2 (DR5) (60). The Fas ligandreceptor system is one of the most well characterised death receptor pathways. Fas activation is a consequence of homotrimeric Fas ligand (FasL) binding which results in receptor trimerisation and clustering of intracellular death domain (DD) motifs. DDs enable death receptors to transmit the extracellular apoptotic signal to the cell’s intracellular apoptotic machinery. Upon clustering of the DD, the adaptor molecule, Fas associated death domain (FADD) is recruited to the intracellular portion of Fas. FADD is a cofactor for procaspase-8 and recruits this initiator caspase to Fas. The association of Fas, FADD, and procaspase-8 results in the formation of the intracellular death-inducing signalling complex (DISC), leading to caspase-8 autocatalysis and activation, which results in caspase-3 activation and cellular disassembly (60, 61). This step is independent of mitochondrial pathway activation and cannot be inhibited by Bcl-2 (62). Signals originating from Fas may be linked to the mitochondria by Bid, an apoptogenic Bcl-2 family member, which upon cleavage by caspase-8 releases cyctochrome c from the mitochondria. Thus, Bid links the two apoptotic pathways and ampliﬁes the apoptotic signal (63, 64). 6. MMP Modulation of Neuronal Death A multiplicity of control mechanisms regulates cellular vulnerability to death receptor-mediated apoptosis. Death receptor activation can operate in cis or in trans resulting in cell suicide when both ligand and receptor are expressed in the same cell or cell fratricide when these components are expressed in neighbouring cells (65). While ligand-receptor interaction is an obligatory step for cell death induced by many types of stressful stimuli, co-expression of receptor and ligand is not a sine qua non for death receptor-mediated apoptosis. One level of death receptor regulation is the metalloproteinase-mediated proteolytic shedding of the receptor and ligand extracellular domain (ectodomain) from the cell surface (66). Metalloproteinases that shed protein ectodomains from the cell surface are termed sheddases, and are members of the MMP and the a disintegrin and metalloproteinase (ADAM) families. Cell surface FasL is cleaved by MMP-3 and MMP-7 (67–70). In a human chondrocyte cell line, both MMP-3 and

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

Matrix Metalloproteinases in Cerebral Ischemia

237

MMP-7 process transmembrane FasL to its soluble counterpart. However, MMP-3 cleaves FasL more eﬃciently than MMP-7 (67). MMP-7 mediated cell surface release of transmembrane FasL from prostate epithelial cells results in enhanced FasL bioactivity and initiation of Fas-mediated apoptosis (69). However, the proteolytic cleavage of FasL from Ewing’s sarcoma cells and cultured cerebellar granular cells leads to cellular desensitisation to the apoptotic signal (68, 70). Conversion of membrane bound FasL to the soluble form may diminish or enhance its death-promoting activity, depending on the target cell type (67, 69–71). Similarly, both TNFα and its death receptor, TNFR1, are released from the cell surface by the sheddase TNFα converting enzyme (TACE/ADAM-17) (72). Fas and FasL are expressed in the resting adult nervous system of rat, mouse, and human, and are quickly up-regulated following injury (73–77). Fas pathway activation is implicated in neuronal death as a consequence of CNS ischemia (75, 77, 78). FasL and Fas expression are upregulated in the penumbral zone following transient focal cerebral ischemia. Caspase-8 is activated in response to CNS ischemia further implicating the Fas pathway in ischemic neuronal demise (76). Mice with a non-functional Fas pathway due to spontaneous mutations in the gene encoding FasL or Fas display reduced ischemic infarct size concomitant with increased functional outcome (78). Moreover, neutralising antibodies against FasL injected into wild-type mice achieve dramatic neuroprotection and enhance functional recovery following transient cerebral ischemia. MMP-3, which is not detected in normal adult rat cerebral cortex, is dramatically up-regulated in the adult rat brain following 90 min focal transient cerebral ischemia (Fig. 4) (54). The up-regulation of MMP-3 following cerebral ischemia is associated with delayed apoptotic death of cortical pyramidal neurons, and spatio-temporally parallels the induction of FasL, suggesting a role for MMP-3 in death receptor-mediated neuronal apoptosis in vivo. MMP-3 is constitutively expressed in cortical neurons in culture and modulates neuronal sensitivity to Fas-mediated apoptosis induced by the chemotherapeutic drug, doxorubicin (15). Addition of exogenous active MMP-3 markedly attenuated Fas-mediated apoptosis and dampened Fas-FasL interactions at the neuronal surface (15). In cultured embryonic neurons, the transmembrane form of Fas-L is more eﬃcient at inducing apoptosis compared to its soluble counterpart (15, 76), thus ectodomain shedding of Fas-L would be expected to impart neuroprotection. Furthermore, a recent report demonstrated that the addition of active MMP-7 to cultured cerebellar granular neurons protected them from β-amyloid neurotoxicity

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

238

M. Wetzel et al

Fig. 4. MMP-3 is up-regulated by post-ischemic adult pyramidal cortical neurons, and is neuroprotective against Fas-mediated neuronal death in culture. (A–D) MMP-3 immunoreactivity is up-regulated in post-ischemic brain. (A–B) non-ischemic hemisphere (C–D) ischemic hemisphere; neuronal marker, NeuN in red, MMP-3 green is up in ischemic neurons, (E) MMP-3 immunoreactivity is expressed by cultured cortical neurons. MMP-3 in red, neuronal marker MAP-2 green. (F) MMP-3 activity dampens doxorubicin-induced Fas-mediated neuronal apoptosis. From (14, 15).

by proteolytic shedding of FasL (68). These two pieces of evidence suggest that particular MMP-mediated extracellular proteolysis may promote cell survival decisions and implicates a physiological role for MMP-3 modulating death receptor pathways at the neuronal surface following cerebral ischemia. In addition to modulating death receptor-mediated apoptosis, metalloproteinase activity can directly inﬂuence neuronal survival following cerebral ischemia potentially by regulating neuron-extracellular matrix (ECM) interactions. MMP-9 expression is up-regulated in hippocampal CA

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

Matrix Metalloproteinases in Cerebral Ischemia

239

pyramidal and dentate granule neurons, in addition to astrocytes following both global cerebral ischemia and kainate-induced excitotoxicity (79). Proteolytic activity is enhanced in the hippocampus in these injury models, and is primarily attributable to MMP-9 (80). MMP-9 appears to play a detrimental function during global ischemia and excitotoxicity, as both pharmacological inhibition and gene deletion of MMP-9 are neuroprotective in vivo (81). In further support of this, the addition of exogenous active MMP-9 to hippocampal slices and primary cortical neurons in culture results in neuronal death (16). In addition to neuronal death, the exposure of cultured cortical neurons to active MMP-9 also results in a subsequent cellular detachment from the culture dish. While the mechanisms by which MMP-9 kills neurons is yet to be resolved, it has been suggested to involve disruption of neuron-ECM interaction via anoikis or the induction of apoptosis by the detachment of cells from the ECM. Importantly, MMP-9 neurotoxicity appears to be neuron type-speciﬁc, as dentate granule cells, which are resistant to neuronal death in these models, are also resistant to MMP-9-induced death despite high expression levels following ischemic or excitotoxic insult. Hippocampal astrocytes also up-regulate expression of MMP-9 following global ischemia, but in a delayed fashion, suggesting a more prominent role in remodelling events during later phases of the injury response. In contrast to the global ischemia model, MMP-9 expression following focal cerebral ischemia is primarily vascular, suggesting that resistance of MMP-9 knock-out mice in this model may be indirect, mediated by stabilising the blood-brain barrier (82). 7. TIMP Regulation of Neuronal Death Metalloproteinase activity at the cell surface inﬂuences neuronal sensitivity to extrinsic death signals through proteolytic shedding of neuronal surface signalling molecules. MMP activity is, in turn, tightly regulated by a family of small homologous two domain proteins designated TIMPs. TIMPs are non-covalent MMP inhibitors governing the pericellular environment of tissue (20). All four TIMPs are able to inhibit all known MMP family members. However, individual TIMPs diﬀer in their eﬃcacy for various MMPs. TIMP-3 possesses exclusive properties that set it apart from other family members. While all TIMPs are secreted into the extracellular space, TIMP-3 binds tightly to the extracellular matrix, suggesting that its actions are localised to the cell surface (83). TIMP-3 potently inhibits numerous shedding events, including IL-6R, L-selectin, syndecans-1 and -4, c-met,

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

240

M. Wetzel et al

CD30, and TNFR1 (84–90). TIMP-3 overexpression is apoptogenic in nonneuronal cell types and cancer cell lines. While the mechanism by which TIMP-3 promotes cell death has not been fully elucidated, a link between metalloproteinase inhibition by TIMP-3 and death receptor-mediated apoptosis has emerged. Over-expression of an altered TIMP-3 protein containing a mutation in the amino-terminal results in blunting of metalloproteinase inhibition and attenuation of apoptosis. This report indicates that metalloproteinase inhibition is required for TIMP-3 to induce apoptosis in vascular smooth muscle cells and some cancer cell lines (91). TIMP-3 induced apoptosis is closely associated by activating the cell surface death receptor pathways. TIMP-3 overexpression sensitises cells to apoptosis induced by toxic stimuli via stabilising TNFR1, and is implicated in the stabilisation of Fas and TRAILR1 (84, 92). Further linking TIMP-3 induced apoptosis to the Fas pathway, a dominant negative form of FADD that inhibits Fasmediated apoptosis, also inhibits TIMP-3 promoted apoptosis, suggesting TIMP-3 works in concert with a death receptor pathway to promote cell death (93). TIMP-3 expression in the resting adult rat brain is relatively low. However, following 90 min transient focal cerebral ischemia, TIMP-3 mRNA and protein are dramatically up-regulated, and associated with ischemic cortical pyramidal neurons undergoing apoptosis (Fig. 5) (14). The TIMP-3 upregulation pattern following cerebral ischemia spatio-temporally parallels that of the death ligands FasL and TRAIL (77), suggesting a physiological role for TIMP-3 in neuronal apoptosis in vivo and an association with death receptor pathways following cerebral ischemia. During CNS development, TIMP-3 expression is transient and spatio-temporally parallels the pattern of developmentally regulated neuronal death and Fas expression in the post-natal cerebral cortex, potentially linking TIMP-3 to naturally occurring death receptor-mediated neuronal apoptosis (94, 95). (Wetzel et al unpublished observations) TIMP-3 is constitutively, and ubiquitously expressed by primary embryonic cortical neurons in culture (15). While physiological expression of TIMP-3 alone is not suﬃcient to induce neuronal apoptosis, it is required to sensitise cortical neurons in culture to death receptor-mediated apoptotic signals via MMP inhibition. Hence, metalloproteinase inhibition by TIMP-3 is necessary for Fas-mediated neuronal apoptosis induced by the chemotherapeutic drug, doxorubicin (15). These studies implicate a physiologic role for TIMP-3 in the regulation of receptormediated cell death in the nervous system. Further support for this is the observation that strain-dependent diﬀerences in doxorubicin sensitivity

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

Matrix Metalloproteinases in Cerebral Ischemia

241

Fig. 5. TIMP-3 is up-regulation following ischemia and is required for Fas-mediated neuronal apoptosis in culture. (A–B) TIMP-3 is up-regulated in post-ischemic adult pyramidal cortical neurons (A) Non-ischemic hemisphere (B) Ischemic hemisphere neuronal marker, NeuN in green and TIMP-3 in red. (C) TIMP-3 is constitutively expressed by cortical neurons in culture. NeuN in red and TIMP-3 in green. (D) Metalloproteinase inhibition restores neuronal sensitivity to doxorubicin-induced Fas-mediated apoptosis. From (14, 15).

correlate with diﬀerences in the pattern of expression of TIMP-3, Fas, FasL, and MMP-3 mRNA levels (Wetzel et al, in press). The convergence of MMP/TIMP balance and death receptor signalling at the neuronal surface may inﬂuence neuronal death versus survival decisions following cerebral ischemia (Fig. 6). Components of the death receptor pathways (ligand and receptor) and modulators of the death receptor pathways (TIMP-3 and MMP-3) are constitutively expressed at low levels in the resting adult brain, except the Fas-FasL interaction is below the threshold required

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

242

M. Wetzel et al

Fig. 6. Paradigm of metalloproteinase regulation of death-receptor mediated neuronal apoptosis. Panel (A), represents neurons in the resting adult brain. The molecules required to induce (Fas/FasL) and regulate (TIMP-3/MMP-3) neuronal apoptosis are expressed at low levels by cortical neurons; hence, FasL is not ligating Fas and the microenvironment favours neuronal survival. Panel (B), metalloproteinase inhibition may potentiate death receptor-mediated neuronal apoptosis following cerebral ischemia. Cerebral ischemia triggers ligation of Fas by FasL and the activation of the Fas pathway leading to neuronal apoptosis. The up-regulation of TIMP-3 may provide metalloproteinase inhibition that dominates the cell surface TIMP-3/MMP-3 balance, hence net MMP inhibition may prevail and facilitate neuronal apoptosis. Panel (C), illustrates net MMP proteolysis that may attenuate death receptor-mediated neuronal apoptosis. If MMP inhibition by TIMP-3 is dampened by either down-regulation of TIMP-3 or up-regulation of MMP-3 activity, then net proteolysis would facilitate neuronal survival. See Ref. (15).

to trigger apoptosis and the microenvironment favours neuronal survival. Consequently, cerebral ischemia would up-regulate death receptor-ligand interaction, possibly by increasing ligand and receptor expression levels, resulting in receptor trimerisation and triggering of neuronal apoptosis. Since MMP inhibition by TIMP-3 dominates the cell surface TIMP-3/MMP-3 balance, net MMP-3 inhibition prevails and apoptosis proceeds. Removal of TIMP-3 or addition of exogenous MMP-3 therefore would lead to local proteolytic shedding of FasL, reduced receptor trimerisation and neuroprotection. This model is supported by evidence that sustained receptor trimerisation is required for eﬀective activation of the Fas pathway, which is dependent on local ligand and receptor concentrations (96). These studies identify novel

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

Matrix Metalloproteinases in Cerebral Ischemia

243

factors, the TIMP-3/MMP-3 balance, regulating death receptor-mediated neuronal demise. The cell surface localisation of the MMP-3/TIMP-3 system provides pharmacologically accessible therapeutic targets for the treatment of neurological disorders in which death-receptor mediated neuronal apoptosis is implicated, such as cerebral ischemia.

8. Conclusion Emerging from the mass of accumulated data is a picture of both detrimental and beneﬁcial aspects of the MMPs and TIMPs in stroke. The constitutively expressed MMP-2 is important in the early disruption of the BBB after the onset of ischemia and reperfusion. This appears to be reversible, and may be important as an alerting mechanism to prepare the brain to reduce subsequent injury. Expression of these MMPs is driven by AP-1 and NF-B sites in the gene promoter regions, and the damage done is more widespread rather than restricted to small membrane domains as in the case of the MMP-2, which requires membrane-bound MMPs for its activation. The more extensive damage is reversible, but in a much longer timeframe, and there may be long-term consequences, such as extensive extracellular matrix remodelling. Tissue inhibitors to metalloproteinases limit proteolytic damage to the extracellular matrix. MMP-9 has been implicated in neuronal death, possibly by breaking down laminin and other extracellular macromolecules around neurons. On the other hand, MMPs protect from apoptosis; both MMP-3 and MMP-7 act as sheddases that release death receptors from the cell surface. TIMP-3 prevents the action of the sheddase, facilitating cell death. Many aspects of the complex action of the MMPs and their inhibitors are still to be determined before it is possible to formulate a comprehensive scheme. Further studies will be needed with genetically manipulated animals and with agents that speciﬁcally inhibit MMPs to complete the picture and allow for a rational approach to inhibition of unwanted actions of the MMPs while preserving those needed for recovery.

References 1. Ginsberg, M. D. and Pulsinelli, W. A. (1994) Annals of Neurology 36, 553–554 2. Dirnagl, U., Iadecola, C., and Moskowitz, M. A. (1999) Trends Neurosci 22, 391–397 3. Graham, S. H. and Chen, J. (2001) J Cereb Blood Flow Metab 21, 99–109

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

244

M. Wetzel et al

4. Klatzo, I. (1967) J Neuropathol Exp Neurol 26, 1–14 5. Fisher, M. and Adams, R. D. (1951) J Exper Neurol Neuropath 10, 92–96 6. Fujimoto, T., Walker, J. T., Spatz, M., and Klatzo, I. (1976) Pathophysiologic aspects of ischemic edema. In Pappius, H. M. and Feindel, W., eds. Dynamics of Brain Edema, Springer-Verlag, Berlin 7. Hallenbeck, J. M. and Dutka, A. J. (1990) Archives of Neurology 47, 1245–1254 8. Yang, G. Y. and Betz, A. L. (1994) Stroke 25, 1658–64; discussion 1664–1665 9. Belayev, L., Busto, R., Zhao, W., and Ginsberg, M. D. (1996) Brain Res 739, 88–96 10. Cuzner, M. L. and Opdenakker, G. (1999) J Neuroimmunol 94, 1–14 11. Nagase, H. and Woessner, J. F., Jr. (1999) Journal of Biological Chemistry 274, 21491–21494 12. Mun-Bryce, S. and Rosenberg, G. A. (1998) Journal of Cerebral Blood Flow & Metabolism 18, 1163–1172 13. Rosenberg, G. A., Kornfeld, M., Estrada, E., Kelley, R. O., Liotta, L. A., and Stetler-Stevenson, W. G. (1992) Brain Res 576, 203–207 14. Wallace, J. A., Alexander, S., Estrada, E. Y., Hines, C., Cunningham, L. A., and Rosenberg, G. A. (2002) J Cereb Blood Flow Metab 22, 1303–1310 15. Wetzel, M., Rosenberg, G. A., and Cunningham, L. A. (2003) Eur J Neurosci 18, 1050–1060 16. Gu, Z., Kaul, M., Yan, B., Kridel, S. J., Cui, J., Strongin, A., Smith, J. W., Liddington, R. C., and Lipton, S. A. (2002) Science 297, 1186–1190 17. Rosenberg, G. A., Dencoﬀ, J. E., McGuire, P. G., Liotta, L. A., and StetlerStevenson, W. G. (1994) Lab Invest 71, 417–422 18. Yong, V. W., Power, C., Forsyth, P., and Edwards, D. R. (2001) Nat Rev Neurosci 2, 502–511 19. Rosenberg, G. A. (2002) Glia 39, 279–291 20. Brew, K., Dinakarpandian, D., and Nagase, H. (2000) Biochim Biophys Acta 1477, 267–283 21. Baker, A. H., George, S. J., Zaltsman, A. B., Murphy, G., and Newby, A. C. (1999) British Journal of Cancer 79, 1347–1355 22. Yong, V. W., Kerkoski, C. A., Forsyth, P. A., Bell, R., and Edwards, D. R. (1998) TINS 21, 75–80 23. Bode, W., Fernandez-Catalan, C., Grams, F., Gomis-Ruth, F. X., Nagase, H., Tschesche, H., and Maskos, K. (1999) Ann NY Acad Sci 878, 73–91 24. Sato, H., Takino, T., Okada, Y., Cao, J., Shinagawa, A., Yamamoto, E., and Seiki, M. (1994) Nature 370, 61–65 25. Sternlicht, M. D. and Werb, Z. (2001) Annual Review of Cell & Developmental Biology 17, 463–516 26. Gijbels, K., Masure, S., Carton, H., and Opdenakker, G. (1992) J Neuroimmunol 41, 29–34 27. Delany, A. M. and Brinckerhoﬀ, C. E. (1992) Journal of Cellular Biochemistry 50, 400–410 28. Van Wart, H. E. and Birkedal Hansen, H. (1990) Proc Natl Acad Sci USA 87, 5578–5582

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

Matrix Metalloproteinases in Cerebral Ischemia

245

29. Sato, H., Kinoshita, T., Takino, T., Nakayama, K., and Seiki, M. (1996) FEBS Lett 393, 101–104 30. Baricos, W. H., Cortez, S. L., el-Dahr, S. S., and Schnaper, H. W. (1995) Kidney International 47, 1039–1047 31. Nagase, H. (1997) Biological Chemistry 378, 151–160 32. Ramos-DeSimone, N., Hahn-Dantona, E., Sipley, J., Nagase, H., French, D. L., and Quigley, J. P. (1999) Journal of Biological Chemistry 274, 13066–13076 33. Hamann, G. F., Okada, Y., Fitridge, R., and del Zoppo, G. J. (1995) Stroke 26, 2120–2126 34. Heo, J. H., Lucero, J., Abumiya, T., Koziol, J. A., Copeland, B. R., and del Zoppo, G. J. (1999) Journal of Cerebral Blood Flow & Metabolism 19, 624–633 35. Watson, P. M., Anderson, J. M., Vanltallie, C. M., and Doctrow, S. R. (1991) Neurosci Lett 129, 6–10 36. Furuse, M., Fujita, K., Hiiragi, T., Fujimoto, K., and Tsukita, S. (1998) J Cell Biol 141, 1539–1550 37. Harkness, K. A., Adamson, P., Sussman, J. D., Davies-Jones, G. A., Greenwood, J., and Woodroofe, M. N. (2000) Brain 123 (Pt 4), 698–709 38. Rosenberg, G. A., Navratil, M., Barone, F., and Feuerstein, G. (1996) Journal of Cerebral Blood Flow & Metabolism 16, 360–366 39. Rosenberg, G. A., Estrada, E. Y., and Dencoﬀ, J. E. (1998) Stroke 29, 2189–2195 40. Asahi, M., Wang, X., Mori, T., Sumii, T., Jung, J. C., Moskowitz, M. A., Fini, M. E., and Lo, E. H. (2001) J Neurosci 21, 7724–7732 41. Asahi, M., Sumii, T., Fini, M. E., Itohara, S., and Lo, E. H. (2001) Neuroreport 12, 3003–3007 42. Clark, A. W., Krekoski, C. A., Bou, S. S., Chapman, K. R., and Edwards, D. R. (1997) Neurosci Lett 238, 53–56 43. Anthony, D. C., Ferguson, B., Matyzak, M. K., Miller, K. M., Esiri, M. M., and Perry, V. H. (1997) Neuropathology & Applied Neurobiology 23, 406–415 44. Montaner, J., Alvarez-Sabin, J., Molina, C., Angles, A., Abilleira, S., Arenillas, J., Gonzalez, M. A., and Monasterio, J. (2001) Stroke 32, 1759–1766 45. Montaner, J., Alvarez-Sabin, J., Molina, C. A., Angles, A., Abilleira, S., Arenillas, J., and Monasterio, J. (2001) Stroke 32, 2762–2767 46. Lo, E. H., Broderick, J. P., and Moskowitz, M. A. (2004) Stroke 35, 354–356 47. Lapchak, P. A., Chapman, D. F., and Zivin, J. A. (2000) Stroke 31, 3034–3040 48. Sumii, T. and Lo, E. H. (2002) Stroke 33, 831–836 49. Pfeﬀerkorn, T. and Rosenberg, G. A. (2003) Stroke 34, 2025–2030 50. Simard, M., Arcuino, G., Takano, T., Liu, Q. S., and Nedergaard, M. (2003) J Neurosci 23, 9254–9262 51. Stewart, P. A. and Wiley, M. J. (1981) Dev Biol 84, 183–192 52. Gottschall, P. E. and Deb, S. (1996) Neuroimmunomodulation 3, 69–75 53. Colton, C. A., Keri, J. E., Chen, W. T., and Monsky, W. L. (1993) J Neurosci Res 35, 297–304

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

246

M. Wetzel et al

54. Rosenberg, G. A., Cunningham, L. A., Wallace, J., Alexander, S., Estrada, E. Y., Grossetete, M., Razhagi, A., Miller, K., and Gearing, A. (2001) Brain Res 893, 104–112 55. Peppin, G. J. and Weiss, S. J. (1986) Proc Natl Acad Sci USA 83, 4322–4326 56. Kerr, J. F., Wyllie, A. H., and Currie, A. R. (1972) Br J Cancer 26, 239–257 57. Colbourne, F., Sutherland, G. R., and Auer, R. N. (1999) J Neurosci 19, 4200–4210 58. Thornberry, N. A. and Lazebnik, Y. (1998) Science 281, 1312–1316 59. Kroemer, G. and Reed, J. C. (2000) Nat Med 6, 513–519 60. Wallach, D., Kovalenko, A. V., Varfolomeev, E. E., and Boldin, M. P. (1998) Curr Opin Immunol 10, 279–288 61. Algeciras-Schimnich, A., Barnhart, B. C., and Peter, M. E. (2002) Curr Opin Cell Biol 14, 721–726 62. Scaﬃdi, C., Fulda, S., Srinivasan, A., Friesen, C., Li, F., Tomaselli, K. J., Debatin, K. M., Krammer, P. H., and Peter, M. E. (1998) EMBO J 17, 1675–1687 63. Bossy-Wetzel, E. and Green, D. R. (1999) J Biol Chem 274, 17484–17490 64. Yin, X. M., Wang, K., Gross, A., Zhao, Y., Zinkel, S., Klocke, B., Roth, K. A., and Korsmeyer, S. J. (1999) Nature 400, 886–891 65. Dhein, J., Walczak, H., Baumler, C., Debatin, K. M., and Krammer, P. H. (1995) Nature 373, 438–441 66. Kayagaki, N., Kawasaki, A., Ebata, T., Ohmoto, H., Ikeda, S., Inoue, S., Yoshino, K., Okumura, K., and Yagita, H. (1995) J Exp Med 182, 1777–1783 67. Matsuno, H., Yudoh, K., Watanabe, Y., Nakazawa, F., Aono, H., and Kimura, T. (2001) J Rheumatol 28, 22–28 68. Ethell, D. W., Kinloch, R., and Green, D. R. (2002) Curr Biol 12, 1595–1600 69. Powell, W. C., Fingleton, B., Wilson, C. L., Boothby, M., and Matrisian, L. M. (1999) Curr Biol 9, 1441–1447 70. Mitsiades, N., Yu, W. H., Poulaki, V., Tsokos, M., and Stamenkovic, I. (2001) Cancer Res 61, 577–581 71. Tanaka, M., Itai, T., Adachi, M., and Nagata, S. (1998) Nat Med 4, 31–36 72. Reddy, P., Slack, J. L., Davis, R., Cerretti, D. P., Kozlosky, C. J., Blanton, R. A., Shows, D., Peschon, J. J., and Black, R. A. (2000) J Biol Chem 275, 14608–14614 73. Park, C., Sakamaki, K., Tachibana, O., Yamashima, T., Yamashita, J., and Yonehara, S. (1998) Biochem Biophys Res Commun 252, 623–628 74. Bechmann, I., Mor, G., Nilsen, J., Eliza, M., Nitsch, R., and Naftolin, F. (1999) Glia 27, 62–74 75. Felderhoﬀ-Mueser, U., Taylor, D. L., Greenwood, K., Kozma, M., Stibenz, D., Joashi, U. C., Edwards, A. D., and Mehmet, H. (2000) Brain Pathol 10, 17–29 76. Qiu, J., Whalen, M. J., Lowenstein, P., Fiskum, G., Fahy, B., Darwish, R., Aarabi, B., Yuan, J., and Moskowitz, M. A. (2002) J Neurosci 22, 3504–3511 77. Martin-Villalba, A., Herr, I., Jeremias, I., Hahne, M., Brandt, R., Vogel, J., Schenkel, J., Herdegen, T., and Debatin, K. M. (1999) J Neurosci 19, 3809–3817

May 26, 2005

16:6 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch09

Matrix Metalloproteinases in Cerebral Ischemia

247

78. Martin-Villalba, A., Hahne, M., Kleber, S., Vogel, J., Falk, W., Schenkel, J., and Krammer, P. H. (2001) Cell Death Diﬀer 8, 679–686 79. Rivera, S., Ogier, C., Jourquin, J., Timsit, S., Szklarczyk, A. W., Miller, K., Gearing, A. J., Kaczmarek, L., and Khrestchatisky, M. (2002) Eur J Neurosci 15, 19–32 80. Lee, S. R., Tsuji, K., Lee, S. R., and Lo, E. H. (2004) J Neurosci 24, 671–678 81. Jourquin, J., Tremblay, E., Decanis, N., Charton, G., Hanessian, S., Chollet, A. M., Le Diguardher, T., Khrestchatisky, M., and Rivera, S. (2003) Eur J Neurosci 18, 1507–1517 82. Asahi, M., Asahi, K., Jung, J. C., del Zoppo, G. J., Fini, M. E., and Lo, E. H. (2000) J Cereb Blood Flow Metab 20, 1681–1689 83. Pavloﬀ, N., Staskus, P. W., Kishnani, N. S., and Hawkes, S. P. (1992) J Biol Chem 267, 17321–17326 84. Smith, M. R., Kung, H., Durum, S. K., Colburn, N. H., and Sun, Y. (1997) Cytokine 9, 770–780 85. Hargreaves, P. G., Wang, F., Antcliﬀ, J., Murphy, G., Lawry, J., Russell, R. G., and Croucher, P. I. (1998) Br J Haematol 101, 694–702 86. Borland, G., Murphy, G., and Ager, A. (1999) J Biol Chem 274, 2810–2815 87. Fitzgerald, M. L., Wang, Z., Park, P. W., Murphy, G., and Bernﬁeld, M. (2000) J Cell Biol 148, 811–824 88. Hansen, H. P., Dietrich, S., Kisseleva, T., Mokros, T., Mentlein, R., Lange, H. H., Murphy, G., and Lemke, H. (2000) J Immunol 165, 6703–6709 89. Nath, D., Williamson, N. J., Jarvis, R., and Murphy, G. (2001) J Cell Sci 114, 1213–1220 90. Matsushita, N., Kashiwagi, M., Wait, R., Nagayoshi, R., Nakamura, M., Matsuda, T., Hogger, P., Guyre, P. M., Nagase, H., and Matsuyama, T. (2002) Clin Exp Immunol 130, 156–161 91. Bond, M., Murphy, G., Bennett, M. R., Amour, A., Knauper, V., Newby, A. C., and Baker, A. H. (2000) J Biol Chem 275, 41358–41363 92. Ahonen, M., Poukkula, M., Baker, A. H., Kashiwagi, M., Nagase, H., Eriksson, J. E., and Kahari, V. M. (2003) Oncogene 22, 2121–2134 93. Bond, M., Murphy, G., Bennett, M. R., Newby, A. C., and Baker, A. H. (2002) J Biol Chem 277, 13787–13795 94. Jaworski, D. M. and Fager, N. (2000) J Neurosci Res 61, 396–408 95. Cheema, Z. F., Wade, S. B., Sata, M., Walsh, K., Sohrabji, F., and Miranda, R. C. (1999) J Neurosci 19, 1754–1770 96. Ashkenazi, A. and Dixit, V. M. (1998) Science 281, 1305–1308

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

This page intentionally left blank

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

CHAPTER 10 ARE MATRIX METALLOPROTEINASES DETRIMENTAL OR BENEFICIAL TO ATHEROSCLEROSIS?

V. Lemaˆıtre and J. D’Armiento∗ Columbia University College of Physicians & Surgeons, Department of Medicine, Division of Molecular Medicine, 630 West 168th Street, P&S 8-401, New York, NY10032 E-mail: ∗[email protected]

1. The Atherosclerotic Plaque: Characteristics and Evolution Atherosclerosis, a complex inﬂammatory process that aﬀects the vessel wall of large and medium-sized arteries, is the primary cause of heart disease and stroke in the Western world (for reviews, see 1–3). Genetic and environmental factors have been associated with the promotion of atherosclerosis, including elevated LDL and VLDL, obesity, diabetes, high blood pressure, a high-fat diet, and smoking (1–3). The development of atherosclerosis has been studied in animal models being fed a high fat and cholesterol diet. In these experimental models, the initiation of atherosclerosis occurs when lipoprotein particles accumulate in the vessel wall. The particles trigger the abnormal attachment of circulating leukocytes, including monocytes and lymphocytes, on specialised receptors of the endothelium (2). After they migrate through the subendothelium, monocytes diﬀerentiate into macrophages and ingest the trapped lipoproteins. The early atherosclerotic lesion is characterised by the presence of inﬂammatory cells, mainly lipid-laden macrophages or foam cells, forming fatty streaks in the neointima of the vessel. Factors secreted by the cells of the neointima are believed to stimulate the migration and replication of medial smooth muscle cells into the lesion, where they produce an extracellular matrix of collagen, proteoglycans, and elastin (1–3). Evolution of the Correspondence to: J. D’Armiento 249

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

250

V. Lemaˆıtre & J. D’Armiento

disease results in more complex plaques containing foam cells, lymphocytes, smooth muscle cells, necrotic cores, and a prominent extracellular matrix. The mature atherosclerotic plaque is covered by a ﬁbrous cap formed primarily of smooth muscle cells between collagen and elastin ﬁbres. The acute coronary syndrome, including myocardial infarction and stroke, is a severe complication of atherosclerosis. This syndrome results from erosion or rupture of the lesion, leading to the liberation of thrombogenic molecules on the surface of the lesion and into the circulating blood, with subsequent thrombus formation (1). Studies have determined that plaques susceptible to rupture are characterised by the presence of macrophages and inﬂammatory cells, a high amount of lipids, and a thin ﬁbrous cap. In contrast, the stable plaque has a thick ﬁbrous cap, fewer inﬂammatory cells, and a dense extracellular matrix (3). These characteristics have led to the hypothesis that activated macrophages, present in large numbers in unstable lesions, release proteases that degrade the connective tissue and destabilise the ﬁbrous cap (1–3). For the past decade, a substantial number of studies have been devoted to better understand the roles of proteases, in particular matrix metalloproteinases (MMPs), in atherosclerosis and its complications. 2. MMPs in Atherosclerosis In 1991, Henney and collaborators (4) reported the expression of stromelysin-1 (MMP-3), a protease with a wide range of substrates, in the human atherosclerotic plaque. Stromelysin was absent in the normal vessel, but both its mRNA and protein were detected in the atheroma, co-localising with lipid-ﬁlled macrophages and inﬁltrating smooth muscle cells. This study suggested for the ﬁrst time that elevated MMP activity in the diseased artery might contribute to the destabilisation of the extracellular matrix and eventual plaque rupture. Further support for the involvement of MMPs in atherosclerosis was generated when investigators (5) showed immunoreactive interstitial collagenase (MMP-1) and gelatinase B (MMP-9) within the plaque, together with gelatinolytic and caseinolytic activities, demonstrating that activated MMPs were present in the atherosclerotic lesion (5). Importantly, MMP activity was localised within the ﬁbrous cap, shoulders of the lesion and lipid core, all areas prone to rupture (5). Expression of MMP-1 and MMP-9 in macrophages and smooth muscle cells of human atherosclerosis was conﬁrmed in subsequent studies (6–8). In particular, increased MMP-9 expression was shown in coronary atherectomy

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

Are Matrix Metalloproteinases Detrimental or Beneﬁcial to Atherosclerosis?

251

specimens from patients with unstable angina (8), suggesting that gelatinase B contributes to the degradation of the ﬁbrous cap. The possible contribution of MMPs in rupture of atherosclerotic lesions was further supported when Halpert and collaborators (9) described prominent expression of matrilysin (MMP-7) and metalloelastase (MMP-12) in macrophages and in cells conﬁned at the border between acellular lipid cores and overlying ﬁbrous areas. Acting together in the lesion, the over-expressed MMPs could potentially digest the various components of the extracellular matrix of the ﬁbrous cap, contributing to rupture. Type I and III ﬁbrillar collagens are major components of the extracellular matrix of the atheroma and its ﬁbrous cap. At physiological pH, the triple-helical domain of the ﬁbrillar collagen molecule is resistant to most proteases, with the exception of interstitial collagenases. Shah and collaborators (10) demonstrated that culture of human macrophages in the presence of ﬁbrous cap extracts increased the expression of MMP-1 with concommitant breakdown of collagen in the cap, suggesting that macrophages contribute to ﬁbrous cap destabilisation in vivo. More recently, collagenase-3 (MMP-13) was found to co-localise with MMP-1 in macrophages of the human lesion (11). MMP-13 has a broader range of substrate speciﬁcity than MMP-1, including elastin and type IV collagen (12). The expression of MMP-1 and MMP-13, together with increased collagenolysis, is signiﬁcantly higher in the atheromatous plaques that are more susceptible to rupture (11). A third collagenase, the neutrophil collagenase (MMP-8), has been detected in macrophages of the shoulder of the lesion, in smooth muscle cells of the ﬁbrous cap, and in the overlying endothelium (13). Other MMPs that are potentially critical in the remodelling of the plaque extracellular matrix include membrane-type-1 and -3 matrix metalloproteinases (MMP-14 and MMP-16) (14, 15) and stromelysin-3 (MMP-11) (16), all expressed in the smooth muscle cells and macrophages of the human atheroma. MMP-14 may play a role in the localised activation of MMP-13 and gelatinase A (MMP-2), which is also overexpressed in atherosclerosis (17). 3. Regulation of MMPs in Atherosclerosis MMP expression and activity are regulated at several levels — gene transcription and protein secretion, activation of the latent zymogen, and regulation of enzyme activity by endogenous inhibitors (12). Numerous

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

252

V. Lemaˆıtre & J. D’Armiento

biological mediators present in the atherosclerotic lesion, such as lipids, modiﬁed LDL, cytokines and reactive oxygen species, have been shown to regulate the expression of MMPs in vitro. In a rabbit model of atherosclerosis, lipid-laden macrophages from aortic lesions, but not macrophages from the lungs, express MMP-1 and MMP-3 (18). By lowering the lipid content of the diet, MMP-1 expression in macrophages was reduced, and collagen deposition in the plaques was increased (19). Accumulation of oxidised LDL in the vessel wall is believed to play a critical role in various cellular processes involved in atherogenesis. In vitro, oxidised LDL, but not native LDL, activates transcription of MMP-1 in endothelial cells (20), of MMP-14 in smooth muscle cells (14), and of MMP-9 and MMP-16 in macrophages (15, 21). Several cytokines and growth factors present in the atheroma have also been shown to regulate MMP expression in cultured macrophages, smooth muscle cells, and endothelial cells. The pro-inﬂammatory cytokines IL-1β and TNF-α induce expression of MMP-13 (11), MMP-8 (13), MMP-14 and -16 (14, 15). IL-18 up-regulates MMP-1, -9 and -13 (22), while MCSF induced macrophage expression of MMP-16 (15). An interaction between the inﬂammatory factor CD40L and its receptor CD40, which is expressed by most cells in the lesion, has been shown to up-regulate expression of MMP-1 (23), MMP-2 and -9 (24), MMP-11 (16), and MMP-8 (13). Downregulation of MMP expression by several cytokines of the atheroma has also been reported. For instance, IL-10 inhibits the transcription of MMP-9 in macrophages (25). Other potential inducers of MMP expression in the plaque include increased circumferential stress (26), bacterial infection (27), contact of macrophages with type I collagen matrix (28) or with endothelial cells (29, 30), and the extracellular matrix metalloproteinase inducer EMMPRIN (31). Most MMPs are secreted as latent zymogens, requiring the removal of the N -terminal pro-domain to be active (12). Several proteases, which are able to cleave the pro-domain in vitro, are believed to contribute to in vivo activation of MMPs. These proteases include plasmin, various activated MMPs, and membrane-type MMPs (12). MMP activation can also be achieved independently from proteolytic cleavage, through disruption of the interaction between the pro-domain and the catalytic site of the MMP — reactive oxygen species, released by lipid-laden macrophages isolated from atherosclerotic tissues, activate MMP-2 and -9 in vitro (32). The treatment of foam cells with N -acetyl-cysteine, a scavenger of reactive oxygen species, decreases both the expression and activity of MMP-9 (33).

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

Are Matrix Metalloproteinases Detrimental or Beneﬁcial to Atherosclerosis?

253

Once activated, MMPs have a high aﬃnity for their endogenous inhibitors, the tissue inhibitors of metalloproteinases (TIMPs), which are overexpressed in atherosclerosis (34). Four human TIMPs have been cloned, and most cells of the human atheroma express TIMP-1, -2, and -3 (5, 35). Studies have shown that secretion of TIMPs is regulated by factors found in the atheroma: TIMP-1 production is down-regulated by oxidised LDL and IL-8 (36), but up-regulated by IL-10 (25) in cholesterol-loaded human macrophages. The ﬁbrogenic mediators PDGF and TGF-β increased the expression of TIMP-1 and -3 in vascular smooth muscle cells in vitro (35). Another MMP inhibitor, the tissue factor pathway inhibitor-2, has been detected in the atheroma, where it could contribute to the regulation of collagenases and gelatinases (37). 4. Functions of MMPs in Atherosclerosis Because MMPs are expressed in areas where the lesion is prone to rupture, it has been assumed that they contribute to the expansion of the atheroma and its subsequent breakdown. However, in vivo studies using genetically modiﬁed mice indicate that the role of MMPs in the biology of atherosclerosis is far more complex. Apolipoprotein-E (apoE) and LDL-receptor deﬁcient mouse models have been widely used in atherosclerosis research (38, 39). In the absence of apoE, lipoprotein remnants are not carried to the liver, where they are normally metabolised, and the mice become hypercholesterolemic and develop lesions of atherosclerosis (40, 41). Atherogenesis in apoE knockout mice is similar to that observed in humans, from the early fatty streaks to the mature atheroma (42, 43). However, murine lesions do not spontaneously rupture or erode, preventing luminal thrombus formation seen in the human disease (38). Intraplaque hemorrhages in the brachiocephalic arteries of one-year-old apoE knockout mice have been reported (44, 45), but their association with the actual destabilisation and segmentation of the ﬁbrous cap is undetermined (46). To study the in vivo function of MMPs in atherosclerosis, several MMP knockout and transgenic models have been crossed into the apoE-null background. In 2001, our laboratory showed that apoE knockout mice overexpressing human interstitial collagenase (MMP-1) in their macrophages had less mature lesions of atherosclerosis (47). Mice do not naturally express MMP-1, so atherosclerosis in this transgenic model resembled the human disease, in which lesion macrophages secrete MMP-1. Since we hypothesised that an increase in MMP-1 would destabilise the plaque, it was surprising

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

254

V. Lemaˆıtre & J. D’Armiento

to ﬁnd that the atheromatous lesions in the transgenic mice were less extensive and less mature compared to their littermate controls, with no evidence of plaque rupture. This study suggested that other MMPs or other classes of proteases, such as cathepsins, might be involved in the digestion of the cap, and that MMP-1 was protective in lesion development (47). Although the exact mechanism by which MMP-1 delays atherogenesis is unknown, several hypotheses may account for this observation. The interaction of peripheral blood monocytes with the vascular matrix is important for their subsequent diﬀerentiation into resident lipid-laden macrophages. In particular, interaction with type I collagen enhances the in vitro diﬀerentiation of monocytes into macrophages and increases their intracellular lipid accumulation (28). Digestion of type I collagen by MMP-1 could therefore aﬀect the diﬀerentiation process and delay atherogenesis. The unexpected observation that plaques from MMP-1 transgenic mice are smaller, have less collagen, and do not rupture, suggested for the ﬁrst time that an MMP was playing a protective role in atherosclerosis, probably through tissue repair and remodelling. The hypothesis that several MMPs might be athero-protective was further supported by studies on apoE knockout mice deﬁcient in TIMP-1 and MMP-3 — the TIMP-1 knockout model had smaller lesions (48), indicating that increased MMP activity delays atherosclerotic progression, while the MMP-3 knockout had bigger plaques (49), suggesting that stromelysin-1 activity, like MMP-1, was athero-protective. The protective role of MMP-3 in lesion formation is also supported by studies analysing the genetic variants of its promoter — patients who are homozygous for the weaker promoter of MMP-3 had more rapid progression of coronary atherosclerosis (50). Contrary to MMP-1 and -3, evidence suggests that MMP-9 expression in atherosclerosis correlates with the expansion and severity of the plaque. A sequence variant of the human MMP-9 promoter has been associated with increased expression of the enzyme, and presence of one or two copies of this genetic variant correlated with a higher severity of atherosclerosis (51). The loss of MMP-9 in apoE knockout mice reduced atherosclerosis, impaired macrophage inﬁltration and decreased collagen deposition, while MMP-12 deﬁciency did not aﬀect lesion development (52). Moreover, smooth muscle cells from MMP-9-deﬁcient mice had impaired migration in vitro (53, 54), indicating that MMP-9 activity is necessary for smooth muscle cells inﬁltration, a critical step in lesion maturation. Impaired migration of smooth muscle cells may account for the decreased deposition of collagen observed in lesions of the MMP-9 knockout mice. Transgenic studies in our laboratory

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

Are Matrix Metalloproteinases Detrimental or Beneﬁcial to Atherosclerosis?

255

demonstrated that mice with increased MMP-9 expression in their lesion have a higher deposition of ﬁbrillar collagen in the plaque (Lemaˆıtre et al, unpublished data). Together, these studies indicate that gelatinase B is critical in the progression of atherosclerosis, through macrophage and smooth muscle cells migration and regulation of collagen deposition. Increased collagen deposition in the MMP-9 transgenic mice could be attributed to the ability of MMPs to cleave, in addition to the various components of the extracellular matrix, cell-surface receptors and cytokines, thus potentially modulating the inﬂammatory process. For example, MMP-1, -3, and -9 can generate active IL-1β from its precursor (55), and recent studies have shown that MMP-9 can activate latent TGF-β, providing a novel mechanism for TGF-β activation (56). TGF-β signalling is believed to be critical in atherosclerosis, where it mediates decreased inﬂammation and increased ﬁbrosis (57, 58). Matrix metalloproteinases may play an important role during early atherogenesis, when cholesterol accumulates in the cells of the intima. A major component for cholesterol eﬄux from foam cells is the high density lipoprotein 3 (HDL3). In vitro, MMP-3, -12, and -7 each cleave the carboxyterminus of apoA-I, a component of HDL3, resulting in decreased cholesterol eﬄux from macrophage foam cells, suggesting that MMPs are directly involved in the accumulation of cholesterol in atherosclerotic lesions (59). After rupture or erosion of the plaque, MMPs could be involved in thrombus formation, through their ability to cleave plasma proteins involved in the control of coagulation. MMP-12, -13, and -14, all expressed in the plaque, degrade ﬁbrinogen in vitro and inactivate the coagulation factor XII (60), suggesting that in vivo these MMPs may impair thrombosis. The role of MMPs in neovascularisation is critical — they are up-regulated during vessel formation and MMP inhibition stops angiogenesis in vitro (for a review, see (61)). It has been shown recently that several MMPs can cleave connective tissue growth factor, which binds and inactivates vascular endothelial growth factor (VEGF) (62). It is also known that cleavage of ﬁbrillar collagen at collagenase-speciﬁc sites is a rate-limiting event in growth factor-stimulated angiogenesis in vivo (63). Therefore, the formation of small vessels inside the atheroma could be modulated by MMPs. 5. MMPs and Aneurysm Degeneration Aneurysms are generally associated with atherosclerosis and are characterised by the degradation of the media and subsequent dilation of the vessel

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

256

V. Lemaˆıtre & J. D’Armiento

wall. Correlative studies have demonstrated expression of MMPs in human abdominal aortic aneurysms — MMP-9 (64, 65, 66), MMP-2 (65, 66), MMP-3 (67), and MMP-12 (68). Several animal models of aortic aneurysm have been developed (for a review, see (69)). In apoE-null mice, atherosclerotic lesions are sometimes associated with pseudo micro-aneurysms, characterised by the complete disruption of the underlying media and macrophage inﬁltration from the lesion into the adventitia (70). The extravasation of blood is prevented by a thickened adventitia, rich in extracellular matrix, which forms a cap around the destroyed media. The frequency of these micro-aneurysms is low and depends on both the age and the genetic background of the animals (70, 71). Both the destruction of the media and the macrophage inﬁltration into the adventitia are thought to involve MMPs. In the mouse, MMP-3, -9, -12, and -13 have been detected in inﬁltrating macrophages, and in the inﬂamed adventitia surrounding sites of medial destruction (70). ApoE-null mice deﬁcient in urokinase-type activator of plasminogen (uPA) are resistant to micro-aneurysm formation (70). Since plasmin cannot degrade ﬁbrillar collagens or elastin, but can activate MMPs in vitro, the absence of medial disruption in uPA-deﬁcient mice is believed to be due to a lack of plasmin-activated MMPs in the plaque, preventing the migration of lesion macrophages into the adventitia (70). Other models have provided evidence for the role of MMPs in medial degeneration. Mice deﬁcient in MMP-9 are resistant to elastase-induced aortic aneurysm formation (72). In rats, decreased MMP activity associated with the retroviral expression of TIMP-1 in smooth muscle cells prevents degradation of the media (73). The importance of MMPs in medial degradation during atherosclerosis was further demonstrated in apoE knockout mice deﬁcient in TIMP-1 (48, 74). In these studies, the lesions of TIMP-1 knockout mice develop significantly more micro-aneurysms, with in situ zymographies showing increased gelatinolytic activity at the site of medial degeneration (74). These studies demonstrate that during atherosclerosis, TIMP-1 plays a key role in preventing micro-aneurysm formation, decreasing the activity of MMPs involved in the degradation of the media underlying the atheroma. Compound apoE knockout mice deﬁcient in either MMP-3 (49), MMP-9, or MMP-12 (52) were protected from medial degradation and ectasia, suggesting that these MMPs work in concert during aneurysm formation. Resistance to medial degradation, and formation of bigger plaques in the MMP-3 knockout model indicate that stromelysin-1 plays a dual role in atherosclerosis, by reducing the progression of the lesion, but degrading the underlying media.

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

Are Matrix Metalloproteinases Detrimental or Beneﬁcial to Atherosclerosis?

257

6. Conclusion A strong body of evidence indicates that matrix metalloproteinases are involved in the biology of the atherosclerotic plaque. MMPs are crucial to lesion growth, smooth muscle cell migration, matrix deposition, and remodelling of the lesion. Studies of genetically modiﬁed mice have revealed that, surprisingly, several MMPs are athero-protective, reducing the severity of the lesion possibly through repair and remodelling of the diseased tissue. Animal studies have also conﬁrmed that MMPs are necessary in the degradation of the media underlying the atherosclerotic plaque as the aneurysm forms. Although strong evidence supports the involvement of MMPs in aneurysm degeneration, there is so far no direct evidence for the role of MMPs in plaque destabilisation and rupture. Other families of proteases that are present in the plaque could be responsible for these events. Cathepsin S and K are both expressed in vulnerable regions of the human plaque and are proteolytic candidates in lesion destabilisation (75). The absence of an experimental mouse model presenting spontaneous plaque rupture constitutes a setback in the ability to determine which proteases or mechanisms are involved in this severe and often lethal complication of atherosclerosis. Diﬀerences in enzyme repertoire, in coagulation regulation and in hemodynamic forces present in the mice might explain in part the absence of plaque rupture in rodents (39). Recently, Welch and collaborators (76) discovered that lesions of apoEnull mice deﬁcient in the Niemann-Pick C gene (Npc1) frequently develop large intraluminal thrombi. This new model of lesion erosion may provide important insight into the role of proteases in plaque destabilisation and thrombosis. In the future, the use of metalloproteinases inhibitors may constitute an attractive pharmacotherapy aimed at stabilising atherosclerotic plaques or preventing their development. However, such therapy would necessitate inhibitors that speciﬁcally target the pro-atherogenic MMPs, while permitting the activity of athero-protective ones. The need for speciﬁcity has already been demonstrated in vivo. Use of a broad-spectrum MMP inhibitor in LDL-receptor deﬁcient mice signiﬁcantly reduced medial degradation, but had no eﬀect on the extent of aortic atherosclerosis (77), probably due to indiscriminate inhibition of both pro- and anti-atherogenic MMPs. Therefore, the development of MMP inhibitors will require further research into the precise role of each individual MMP in atherosclerosis.

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

258

V. Lemaˆıtre & J. D’Armiento

References 1. 2. 3. 4.

5. 6. 7. 8. 9.

10.

11. 12. 13.

14.

15.

16.

17.

18. 19. 20.

Ross, R. (1999) N Engl J Med 340, 115–126 Lusis, A. J. (2000) Nature 407, 233–241 Libby, P. (2002) Nature 420, 868–874 Henney, A. M., Wakeley, P. R., Davies, M. J., Foster, K., Hembry, R., Murphy, G., and Humphries, S. (1991) Proc Natl Acad Sci USA 88, 8154–8158 Galis, Z. S., Sukhova, G. K., Lark, M. W., and Libby, P. (1994) J Clin Invest 94, 2493–2503 McMillan, W. D., Patterson, B. K., Keen, R. R., Shively, V. P., Cipollone, M., and Pearce, W. H. (1995) Arterioscler Thromb Vasc Biol 15, 1139–1144 Nikkari, S. T., O’Brien, K. D., Ferguson, M., Hatsukami, T., Welgus, H. G., Alpers, C. E., and Clowes, A. W. (1995) Circulation 92, 1393–1398 Brown, D. L., Hibbs, M. S., Kearney, M., Loushin, C., and Isner, J. M. (1995) Circulation 91, 2125–2131 Halpert, I., Sires, U. I., Roby, J. D., Potter-Perigo, S., Wight, T. N., Shapiro, S. D., Welgus, H. G., Wickline, S. A., and Parks, W. C. (1996) Proc Natl Acad Sci USA 93, 9748–9753 Shah, P. K., Falk, E., Badimon, J. J., Fernandez-Ortiz, A., Mailhac, A., Villareal-Levy, G., Fallon, J. T., Regnstrom, J., and Fuster, V. (1995) Circulation 92, 1565–1569 Sukhova, G. K., Schonbeck, U., Rabkin, E., Schoen, F. J., Poole, A. R., Billinghurst, R. C., and Libby, P. (1999) Circulation 99, 2503–2509 Visse, R. and Nagase, H. (2003) Circ Res 92, 827–839 Herman, M. P., Sukhova, G. K., Libby, P., Gerdes, N., Tang, N., Horton, D. B., Kilbride, M., Breitbart, R. E., Chun, M., and Schonbeck, U. (2001) Circulation 104, 1899–1904 Rajavashisth, T. B., Xu, X. P., Jovinge, S., Meisel, S., Xu, X. O., Chai, N. N., Fishbein, M. C., Kaul, S., Cercek, B., Shariﬁ, B., and Shah, P. K. (1999) Circulation 99, 3103–3109 Uzui, H., Harpf, A., Liu, M., Doherty, T. M., Shukla, A., Chai, N. N., Tripathi, P. V., Jovinge, S., Wilkin, D. J., Asotra, K., Shah, P. K., and Rajavashisth, T. B. (2002) Circulation 106, 3024–3030 Schonbeck, U., Mach, F., Sukhova, G. K., Atkinson, E., Levesque, E., Herman, M., Graber, P., Basset, P., and Libby, P. (1999) Exp Med 189, 843–853 Li, Z., Li, L., Zielke, H. R., Cheng, L., Xiao, R., Crow, M. T., StetlerStevenson, W. G., Froehlich, J., and Lakatta, E. G. (1996) Am J Pathol 148, 121–128 Galis, Z. S., Sukhova, G. K., Kranzhofer, R., Clark, S., and Libby, P. (1995) Proc Natl Acad Sci USA 92, 402–406 Aikawa, M., Rabkin, E., Voglic, S. J., Shing, H., Nagai, R., Schoen, F. J., and Libby, P. (1998) Circ Res 83, 1015–1026 Huang, Y., Mironova, M., and Lopes-Virella, M. F. (1999) Arterioscler Thromb Vasc Biol 19, 2640–2647

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

Are Matrix Metalloproteinases Detrimental or Beneﬁcial to Atherosclerosis?

259

21. Xu, X. P., Meisel, S. R., Ong, J. M., Kaul, S., Cercek, B., Rajavashisth, T. B., Shariﬁ, B., and Shah, P. K. (1999) Circulation 99, 993–998 22. Gerdes, N., Sukhova, G. K., Libby, P., Reynolds, R. S., Young, J. L., and Schonbeck, U. (2002) J Exp Med 195, 245–257 23. Schonbeck, U., Mach, F., Bonnefoy, J. Y., Loppnow, H., Flad, H. D., and Libby, P. (1997) J Biol Chem 272, 19569–74 24. Mach, F., Schonbeck, U., Bonnefoy, J. Y., Pober, J. S., and Libby, P. (1997) Circulation 96, 396–399 25. Lacraz, S., Nicod, L. P., Chicheportiche, R., Welgus, H. G., and Dayer, J. M. (1995) J Clin Invest 96, 2304–2310 26. Lee, R. T., Schoen, F. J., Loree, H. M., Lark, M. W., and Libby, P. (1996) Arterioscler Thromb Vasc Biol 16, 1070–1073 27. Kol, A., Bourcier, T., Lichtman, A. H., and Libby, P. (1999) J Clin Invest 103, 571–577 28. Wesley, R. B., Meng, X., Godin, D., and Galis, Z. S. (1998) Arterioscler Thromb Vasc Biol 18, 432–440 29. Amorino, G. P. and Hoover, R. L. (1998) Am J Pathol 152, 199–207 30. Hojo, Y., Ikeda, U., Takahashi, M., Sakata, Y., Takizawa, T., Okada, K., Saito, T., and Shimada, K. (2000) J Mol Cell Cardiol 32, 1459–68 31. Major, T. C., Liang, L., Lu, X., Rosebury, W., and Bocan, T. M. (2002) Arterioscler Thromb Vasc Biol 22, 1200–1207 32. Rajagopalan, S., Meng, X. P., Ramasamy, S., Harrison, D. G., and Galis, Z.S. (1996) J Clin Invest 98, 2572–2579 33. Galis, Z. S., Asanuma, K., Godin, D., and Meng, X. (1998) Circulation 97, 2445–2453 34. Zaltsman, A. B., George, S. J., and Newby, A. C. (1999) Arterioscler Thromb Vasc Biol 19, 1700–1707 35. Fabunmi, R. P., Sukhova, G. K., Sugiyama, S., and Libby, P. (1998) Circ Res 83, 270–278 36. Moreau, M., Brocheriou, I., Petit, L., Ninio, E., Chapman, M. J., and Rouis, M. (1999) Circulation 99, 420–426 37. Herman, M. P., Sukhova, G. K., Kisiel, W., Foster, D., Kehry, M. R., Libby, P., and Schonbeck, U. (2001) J Clin Invest 107, 1117–1126 38. Breslow, J. L. (1996) Science 272, 685–688 39. Carmeliet, P., Moons, L., and Collen, D. (1998) Cardiovasc Res 39, 8–33 40. Zhang, S. H., Reddick, R. L., Piedrahita, J. A., and Maeda, N. (1992) Science 258, 468–471 41. Plump, A. S., Smith, J. D., Hayek, T., Aalto-Setala, K., Walsh, A., Verstuyft, J. G., Rubin, E. M., and Breslow, J. L. (1992) Cell 71, 343–353 42. Zhang, S. H., Reddick, R. L., Burkey, B., and Maeda, N. (1994) J Clin Invest 94, 937–945 43. Nakashima, Y., Plump, A. S., Raines, E. W., Breslow, J. L., and Ross, R. (1994) Arterioscler Thromb Vasc Biol 14, 133–140 44. Williams, H., Johnson, J. L., Carson, K. G., and Jackson, C. L. (2002) Arterioscler Thromb Vasc Biol 22, 788–792

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

260

V. Lemaˆıtre & J. D’Armiento

45. Rosenfeld, M. E., Polinsky, P., Virmani, R., Kauser, K., Rubanyi, G., and Schwartz, S. M. (2000) Arterioscler Thromb Vasc Biol 20, 2587–2592 46. Lutgens, E., van Suylen, R. J., Faber, B. C., Gijbels, M. J., Eurlings, P. M., Bijnens, A. P., Cleutjens, K. B., Heeneman, S., Daemen, M. J. (2003) Arterioscler Thromb Vasc Biol 23, 2123–2130 47. Lemaˆıtre, V., O’Byrne, T. K., Borczuk, A. C., Okada, Y., Tall, A. R., and D’Armiento, J. (2001) ApoE knockout mice expressing human matrix metalloproteinase-1 in their macrophages have less advanced atherosclerosis. J Clin Invest 107, 1227–1234 48. Silence, J., Collen, D., and Lijnen, H. R. (2002) Circ Res 90, 897–903 49. Silence, J., Lupu, F., Collen, D., and Lijnen, H. R. (2001) Arterioscler Thromb Vasc Biol 21, 1440–1445 50. Ye, S., Eriksson, P., Hamsten, A., Kurkinen, M., Humphries, S. E., and Henney, A. M. (1996) J Biol Chem 271, 13055–13060 51. Zhang, B., Ye, S., Herrmann, S. M., Eriksson, P., de Maat, M., Evans, A., Arveiler, D., Luc, G., Cambien, F., Hamsten, A., Watkins, H., and Henney, A. M. (1999) Circulation 99, 1788–1794 52. Luttun, A., Lutgens, E., Manderveld, A., Maris, K., Collen, D., Carmeliet, P., and Moons, L. (2004) Circulation 109, 1408–1414 53. Galis, Z. S., Johnson, C., Godin, D., Magid, R., Shipley, J. M., Senior, R. M., and Ivan, E. (2002) Circ Res 91, 852–859 54. Cho, A., and Reidy, M. A. (2002) Circ Res 91, 845–851 55. Schonbeck, U., Mach, F., and Libby, P. (1998) J Immunol 161, 3340–3346 56. Yu, Q. and Stamenkovic, I. (2000) Genes Dev 14, 163–176 57. Lutgens, E., Gijbels, M., Smook, M., Heeringa, P., Gotwals, P., Koteliansky, V. E., and Daemen, M. J. (2002) Arterioscler Thromb Vasc Biol 22, 975–982 58. Mallat, Z., Gojova, A., Marchiol-Fournigault, C., Esposito, B., Kamate, C., Merval, R., Fradelizi, D., and Tedgui, A. (2001) Circ Res 89, 930–934 59. Lindstedt, L., Saarinen, J., Kalkkinen, N., Welgus, H., and Kovanen, P. T. (1999) J Biol Chem 274, 22627–22634 60. Hiller, O., Lichte, A., Oberpichler, A., Kocourek, A., and Tschesche, H. (2000) J Biol Chem 275, 33008–33013 61. Pepper, M. S. (2001) Arterioscler Thromb Vasc Biol 21, 1104–1117 62. Hashimoto, G., Inoki, I., Fujii, Y., Aoki, T., Ikeda, E., and Okada, Y. (2002) J Biol Chem 277, 36288–36295 63. Seandel, M., Noack-Kunnmann, K., Zhu, D., Aimes, R. T., and Quigley, J. P. (2001) Blood 97, 2323–2332 64. Thompson, R. W., Holmes, D. R., Mertens, R. A., Liao, S., Botney, M. D., Mecham, R. P., Welgus, H. G., and Parks, W. C. (1995) J Clin Invest 96, 318–326 65. Freestone, T., Turner, R. J., Coady, A., Higman, D. J., Greenhalgh, R. M., and Powell, J. T. (1995) Arterioscler Thromb Vasc Biol 15, 1145–1151 66. Goodall, S., Crowther, M., Hemingway, D. M., Bell, P. R., and Thompson, M. M. (2001) Circulation 104, 304–309

May 26, 2005

16:7 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch10

Are Matrix Metalloproteinases Detrimental or Beneﬁcial to Atherosclerosis?

261

67. Carrell, T. W., Burnand, K. G., Wells, G. M., Clements, J. M., and Smith, A. (2002) Circulation 105, 477–482 68. Curci, J. A., Liao, S., Huﬀman, M. D., Shapiro, S. D., and Thompson, R. W. (1998) J Clin Invest 102, 1900–1910 69. Daugherty, A. and Cassis, L. A. (2004) Arterioscler Thromb Vasc Biol 24, 429–434 70. Carmeliet, P., Moons, L., Lijnen, R., Baes, M., Lemaˆıtre, V., Tipping, P., Drew, A., Eeckhout, Y., Shapiro, S., Lupu, F., and Collen, D. (1997) Nat Genet 17, 439–444 71. Shi, W., Brown, M. D., Wang, X., Wong, J., Kallmes, D. F., Matsumoto, A. H., Helm, G. A., Drake, T. A., and Lusis, A. J. (2003) Arterioscler Thromb Vasc Biol 23, 1901–1906 72. Pyo, R., Lee, J. K., Shipley, J. M., Curci, J. A., Mao, D., Ziporin, S. J., Ennis, T. L., Shapiro, S. D., Senior, R. M., and Thompson, R. W. (2000) J Clin Invest 105, 1641–1649 73. Allaire, E., Forough, R., Clowes, M., Starcher, B., and Clowes, A. W. (1998) J Clin Invest 102, 1413–1420 74. Lemaˆıtre, V., Soloway, P. D., and D’Armiento, J. (2003) Circulation 107, 333–338 75. Sukhova, G. K., Shi, G. P., Simon, D. I., Chapman, H. A., and Libby, P. (1998) J Clin Invest 102, 576–583 76. Welch, C. L., Lemaˆıtre, V., Sharma, N., Gorelik, A., Kuriakose, G., Tabas, I., D’Armiento, J., Dekelbaum, R., and Tall, A. R. (2003) Circulation 108, IV–224 77. Prescott, M. F., Sawyer, W. K., Von Linden-Reed, J., Jeune, M., Chou, M., Caplan, S. L., and Jeng, A. Y. (1999) Ann NY Acad Sci 878, 179–190

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

This page intentionally left blank

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

CHAPTER 11 ROLE AND REGULATION OF MATRIX METALLOPROTEASES IN BRAIN TUMOURS

S.S. Lakka and J.S. Rao∗ University of Illinois College of Medicine, Departments of Biomedical and Therapeutic Sciences and Neurosurgery, One Illini Drive, Box 1649, Peoria, IL 61656 E-mail: ∗[email protected]

Gliomas, a type of devastating primary brain tumour, are distinct from other solid, non-neural primary neoplasms in that they display extensive inﬁltrative invasive behaviour but seldom metastasise to distant organs. This invasiveness into the surrounding normal brain tissue makes gliomas a major challenge for clinical intervention. Total resection of gliomas is impossible, and recurrence of tumour growth is a common phenomenon. Patients have a mean survival time of 8–12 months. Although in recent years substantial progress has been made toward understanding the invasive behaviour of gliomas in in vitro and in vivo models. The factors responsible for the extensive inﬁltration are still poorly documented. Factors in the invasion process have been ascertained through clariﬁcation of the key roles played by the extra-cellular matrix (ECM) composition, cell adhesive molecules and the degrading enzymes. Primarily, three classes of proteases play a signiﬁcant role during the progression and invasive behaviour of human gliomas. This book chapter focuses mainly on the recent developments concerning the role and regulation of MMPs in the invasiveness of human gliomas.

1. Introduction The invasive and destructive characteristics of malignant neoplasms in the central nervous system (CNS) are of great clinical importance. The prognosis for patients with higher-grade tumours, such as glioblastomas, is poor with a mean survival of only 8–12 months following chemotherapy or radiotherapy (1). This poor prognosis is due in part to the resistance of tumour

Correspondence to: J.S. Rao 263

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

264

S.S. Lakka & J.S. Rao

cells to radiation (2), cytostatics (3) and the diﬃculty in achieving complete tumour resection (4). The inﬁltrative growth pattern of gliomas contrasts with that of brain metastases, which most often grow with self-deﬁned borders to the adjacent neural tissue. Whereas the unregulated proliferation of primary brain tumour cells within the skull is a major factor in the prognosis of aﬄicted patients, the ability of neoplastic cells to migrate and invade surrounding normal brain tissue presents a major hurdle to successful therapeutic intervention. Although astrocytic tumours very rarely metastasise to distal structures, the tumour cells often display the ability to migrate within the brain, which eventually leads to local or distal recurrence. The ability of tumour cells to invade normal tissue has been the focus of numerous studies of many diﬀerent human cancers. Invasion by primary brain tumour cells seems to have some distinct properties. A clear illustration of this phenomenon is the diﬀerence between cells from systemic lesions (e.g. melanoma, breast, colon and lung carcinoma), which readily metastasise to the brain but are rarely invasive there, and primary brain tumours, which are invasive and can migrate to the opposite, unaﬀected hemisphere of the brain. Diﬀusive individual cell inﬁltration of gliomas involves changes in adhesion, cell motility and proteolytic activity. The underlying mechanism of invasion is a complex process in which the tumour cells at the invasive front detach from the growing primary tumour mass and reattach to extracellular matrix (ECM) components or surrounding tissue elements. This process involves migration, which is increased cell motility along the surface or through the three-dimensional structure, and invasion which involves potential to hydrolyse proteins, proteoglycans and carbohydrate resulting in damage to the tissue or structures reﬂecting the ability of tumour cells to translocate through ECM barriers (5). Brain tumour cells utilise proteinases capable of degrading proteoglycans and collagens for tumour progression, such as matrix metalloproteases (MMPs) and membrane-type metalloproteases (MT-MMPs). Matrix metalloproteases are neutral proteases that attack all the components of the ECM. 2. The ECM of the Brain The invasive nature of malignant brain tumours leads to the local destruction and dissolution of the adjacent normal brain tissue. Invasion of cells depends mainly on the composition and degradation of the ECM, which in the CNS is restricted to the glial limitans externa and the walls of large blood vessels (6). The ECM of the brain parenchyma is ﬁlled with an

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

Role and Regulation of Matrix Metalloproteases in Brain Tumours

265

amorphous matrix containing a few ﬁbrous proteins, hyaluronic acid and other glycosaminoglycans. The ECM components that have been identiﬁed in normal brain and brain tumours include collagen types I, III, IV, V, VI, VII and VIII, laminin, and ﬁbronectin (6–12). Malignant gliomas are also known to contain increased levels of hyaluronic acid (13). The ECM plays an important role in tumour cell behaviour in vitro. Although coating tissue culture ﬂasks with ﬁbronectin and laminin had no eﬀect on the growth of U343MG-A glioma cells, cells grown in ﬂasks coated with types I and IV collagen showed decreased proliferation, formation of stellate cells and increased production of glial ﬁbrillary acidic protein compared with glioma cells growing on plastic (6). Immunohistochemical localisation of ECM proteins during brain tumour invasion in BD1X rats showed an overproduction of collagen types I & IV and ﬁbronectin, and in particular, enrichment of type I collagen and ﬁbronectin in the zone of invasion (10). Several recent studies, including our own (14–16), have shown that various ECM protein components play signiﬁcant roles in the migration and invasion of glioma cells (12, 17–19). We have also demonstrated that certain ECM components are present in greater amounts in glioblastomas as compared with low-grade glioma or normal brain tissue (15). 3. Matrix Metalloproteinases Matrix metalloproteinases (MMPs) constitute a family of zinc-containing enzymes with more than 20 members identiﬁed to date. They are divided into interstitial collagenases, stromelysins, gelatinases and membrane type metalloproteinases (MT-MMPs) based on protein structure (20). Unlike MMPs, MT-MMPs are membrane-bound and studies have shown that they are activated either intracellularly or on the cell surface (21). Another group of proteolytic metalloproteases is the ADAMs family, which consists of at least 33 diﬀerent enzymes (22). More recently, a new group of metalloproteases called ADAMs has been identiﬁed. Like the ADAMs, these enzymes contain a disintegrin and metalloprotease domain. However, instead of a transmembrane domain, the ADAMs express one or several thrombospondin 1-like repeats (23). Although their physiological functions are not yet completely understood, nineteen diﬀerent members have been identiﬁed (24). MMPs are secreted as pro-MMPs and then activated following the sequential cleavage steps (25, 26). Removal of the signal peptide and pro-peptide domains or a change in conﬁguration activates the enzymes.

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

266

S.S. Lakka & J.S. Rao

MMP expression and proteolytic activity are tightly regulated at three stages — gene transcription, proenzyme activation and activity of natural inhibitors (tissue inhibitors of metalloproteinase known as TIMPs). The balance between production, activation and inhibition prevents excessive proteolysis or inhibition. Several factors like cytokines, growth factors, phorbol esters; cell-cell and cell-matrix interactions are thought to control MMP expression (27). Most MMPs are secreted as inactive zymogens, which may be proteolytically activated by diﬀerent proteinases such as other MMPs, plasmin, trypsin, chymotrypsin and cathepsins. Several cell types produce MMPs including monocytes, macrophages, neutrophils (28, 29), T -lymphocytes (30), endothelial cells (31), ﬁbroblasts (32) and in the CNS by microglia, astrocytes, oligodendrocytes and neurons (33–35). MMP-2 and MMP-9 are secreted by microglia and astrocytes as active forms (36). MMPs have been shown to regulate tumour cell invasion through their interactions with extracellular matrix components including cell matrix embedded growth factors and cell adhesion molecules (37, 38). 4. MMP Expression in Gliomas Local invasive growth is one of the key features of primary malignant brain tumours and is accompanied by remodelling of the vasculature and destruction of normal tissue (39). The expression of several forms of metalloprotease inhibitors by fetal astrocytes and glioma cell lines has been reported (40, 41). Furthermore, a metalloprotease secreted by the rat glioma cell line BT5C in serum-free medium was observed to be capable of degrading fetal rat brain aggregates (11, 39). MMP-2 (gelatinase A) and MMP-9 (gelatinase B) transcript and activity levels were found to be higher in glioma cell lines as well as surgical specimens (40, 42–52). MMP-9 expression has been strongly correlated with tumour grade. MMP-2 has been observed to be localised in malignant glioma cells and blood vessels along with MT1-MMP, whereas MMP-9 was strongly expressed in blood vessels at proliferating margins (43) as well as in the tumour cells (49). Furthermore, glioblastoma cells injected intracranially into nude mice resulted in signiﬁcantly increased levels of MMP-2 and MMP-9 (53, 54). In addition, astroglioma cells constitutively express high levels of MMP-2 mRNA, protein and bioactivity as assessed by ribonuclease protection assay, immunoblotting and zymography assay respectively (55). From the subfamily of collagenases, MMP-1 has been found in gliomas (43) and MMP-13 in childhood astrocytomas (56) and medulloblastomas (57).

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

Role and Regulation of Matrix Metalloproteases in Brain Tumours

267

MMP-3, a stromelysin, has been shown to be expressed in glioma progression in vitro (58). MMP-3 is expressed during the insidious invasiveness of astrocytoma either directly through degradation of speciﬁc matrix macromolecules, or indirectly via activation of other MMPs that might elicit an enhanced scope of ECM digestion (59). MMP-7, 10 and 11 were strongly expressed by glial tumour cells, showing a signiﬁcantly stronger immunoreaction in neoplastic astrocytes as compared to oligodendroglial tumour cells (60). Membrane-type MMPs (MT1-MMP, MT2-MMP, MT3MMP and MT4-MMP) have shown proteolytic activity towards several ECM molecules including ﬁbronectin, vitronectin, laminin, tenascin, proteoglycans and collagen (61, 62). Several studies reported the evidence of MT-MMP expression in glioma cells lines. MT1-MMP was overexpressed in D54, LG11, T98 and U87MG cells as compared to A173, Hs683, U251MG and U373 (63). High levels of MT1-MMP and MT3-MMP mRNA were detected in U87MG whereas MT2-MMP was expressed in low levels (64). MT1-MMP mRNA was absent or barely detectable in normal brain and was overexpressed in high-grade gliomas. Further, its expression was shown to correlate with the expression and activation of MMP-2 (46). MT2-MMP was shown to increase accordingly with the histological grade of malignancy, whereas MT3-MMP was variably expressed in low levels in brain tumour samples (65, 66). Although MT5-MMP transcripts were not detected in normal brain, studies have demonstrated that it is overexpressed in a number of glioblastomas, astrocytomas and anaplastic astrocytomas (67). ADAM8 (CD156) is present in neurons and oligodendrocytes and has been implicated in cell adhesion in neurodegenerative disorders (68). 5. Regulation of MMPs MMPs are very elaborately controlled, particularly by enzyme activation to produce a functional form, their endogenous inhibitors (TIMPs) and at the level of gene expression (69). Other underlying mechanisms aﬀect mRNA stability, protein secretion, and speciﬁc degradation and clearance (69). Growth factors, such as endothelial growth factor (EGF), basic ﬁbroblast growth factor (b-FGF), transforming growth factor (TGF-β1 and β2) and vascular endothelial growth factor (VEGF), have been shown to upregulate MMP-2 and MMP-9 in gliomas. This eﬀect was most prominent with TGF-β1 and β-2 (70). MMP-2 is constitutively present in the brain at low basal levels. Basal gene transcription is mediated via normal cytokine signalling events, which cause basal levels of transcriptional activation at

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

268

S.S. Lakka & J.S. Rao

SP1, ASP3 and AP-2 binding sites (71). The use of a 5 deletional MMP-2 reporter construction in conjunction with site-directed mutagenesis of the human MMP-2 promoter demonstrated that SP1 and AP-2 elements are critical for constitutive expression of MMP-2 gene in astroglioma cells (71). Angiopoietin-2 (Ang-2), a naturally occurring antagonist of Ang-1, is capable of causing MMP-2 activation and thereby inducing glioma cell invasion (72). MMP-9 and stromelysin (MMP-3) have been shown to be chieﬂy transcribed under the inﬂuence of various transcription factors commonly found to be involved with cellular stress responses and tissue morphogenesis, including NF-κβ, ETS family members and AP-1 (73). Epidermal growth factor variant subtype III promoted activation of MMP-9 possibly through the activation of MAPK/ERK in GBM (74). We have previously shown that MMP-9 production is induced by cytoskeletal changes involving protein kinase C activation mediated by NF-κβ (75). Another study demonstrated that vanadate and phenyl arsenic oxide inhibited migration and invasion of glioma cells by their eﬀects on the cytoskeleton and inhibition of MMP-9 expression (76). Factors that have been reported to induce MT1-MMP expression in glioma cell lines include EGFR activation (73), EMM-PRIN (CD147) (77), HGF/SF stimulation (78) PMA (79), and possibly the β-amyloid peptide (80). Matrilysin (MMP-7) was also reported to be overexpressed in gliomas. However, this protease is generally overexpressed in comparison to other MMPs in the brain (81). High basal levels of MMP-7 could increase the expression of other MMPs involved in tumour progression in gliomas since MMP-7 was shown to activate MMP-9 by a proteolytic interaction which dissociates the inhibitory pro-MMP-9/TIMP complex (82). TGF-β1 stimulated the expression of matrylisin in two human glioma cell lines and activated MMP-9 in C6 rat glioma cells (42). Recombinant TGF-β2 additions to U87MG and Ln-229 glioma cell lines resulted in a marked increase in MMP-2 expression and suppression of TIMP-2 protein. Further, MT1-MMP-mediated activation of pro-MMP-2, may be an important factor during glioma invasion (46, 83). Protein kinase C (PKC) also plays an important role in the regulation of glioma MT1-MMP mRNA expression, MMP-2 activation and invasion (84). The mitogen-activated kinase/extracellular signal-regulated kinase (MEK/ERK) signalling pathway is essential for MMP-9 up-regulation in astrocytes after PMA [PKC induction and TNF-α (cytokine) stimulation]. It has been reported that SNB19 cells transfected with dominant negative JNK, MEKK and ERK1 expression vectors decreased MMP-9 expression as well as promoter activity (85). The mt-ERK stable SNB19 cells showed

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

Role and Regulation of Matrix Metalloproteases in Brain Tumours

269

decreased levels of MMP-9 and less invasiveness as compared to parental and vector-transfected stable clones (86). A second level of regulation of MMPs is accomplished after secretion as a pro-peptide. The pro-peptide mechanism was thought to provide an intrinsic regulatory mechanism by inhibiting rapid activation or by activating co-factors. Pro-peptide cleavage of gelatinase A to its active form appears to be a critical step in invasion and metastasis in gliomas with the potential of initiating further proteolytic cascades via cleavage of the pro-forms of collagenase 3 and MMP-9 (87). However, there are no reports so far on the intermolecular cleavage among the secreted MMPs in gliomas. A third level of MMP regulation is accomplished by the expression of tissue inhibitors of metalloproteinases (TIMPs). The balance, or more precisely the lack thereof, between MMP and TIMP activity may contribute to malignant behaviour of many cancers (88). The TIMP family consist of 4 gene products (TIMP 1–4) that inhibit speciﬁc MMPs (e.g. TIMP-2) or a broad spectrum of MMPs (89–91). Several studies indicate that the pattern of TIMP expression and localisation is very diﬀerent in gliomas and, consequently, could inﬂuence tumour progression by diﬀerent mechanisms. Studies utilising TIMPs have demonstrated that down-regulation of both TIMP-1 and TIMP-2 contributes signiﬁcantly to the invasive potential of gliomas (92). Other studies demonstrate that TIMP-1 expression increases dramatically with glioma progression (93, 94). The actual mechanism of TIMP-1 is not clear in brain tumour progression, although the increased expression of TIMP-1 could be due to the elevated expression of growth factors and cytokines in the same manner that these factors inﬂuence the induction of MMP-9 or activate MMP-2. Since TIMP-1 binds to MMP-9, TIMP-1 overexpression could be due to the regulatory mechanism by the host to neutralise MMP-9 overexpression in glioblastomas. Alternately, TIMP-1 may stimulate glioma tumour progression via proliferation of erythroid precursors (95). TIMP-1 and TIMP-2 positively inﬂuence the proliferation of numerous cell types (96). However, one group has reported that TIMP-3 expression did not correlate with glioma grade (97) and another study demonstrated reduced expression of TIMP-3 by gene silencing via methylation (98). In situ hybridisation studies have demonstrated that TIMP-4 expression was negatively correlated with glioma malignancy (97). Recombinant or overexpressed TIMP-4 had no eﬀect on cellular viability and proliferation but decreased the invasive capacity of glioma cells to invade through matrigel. The increase in TIMP-4 was also shown to inhibit MMP-2 activation by its ability to inhibit activation of other MMPs (97).

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

270

S.S. Lakka & J.S. Rao

The decreased expression of TIMP-4 in glioblastoma samples as compared with increased expression of other MMPs such as MMP-9 probably favours increased cellular invasion in brain tumours. MMPs function via intracellular signalling. This group of enzymes has been shown to interact with a broad range of non-matrix proteins including growth factors and their receptors, mediators of apoptosis, and cell adhesion molecules. Since cells have receptors for structural ECM components (for example, integrins), cleavage of ECM proteins by MMPs also aﬀects cellular signalling and functions (99). For example, MMPs mediate cellsurface-receptor cleavage and release, cytokine and chemokine activation and inactivation, and the release of apoptotic ligands (100). These cellular processes are all involved in promoting the aspects of tumour growth, such as cell proliferation, adhesion and dispersion, migration, diﬀerentiation, angiogenesis, apoptosis and host defence evasion. β1 integrin is highly expressed in invasive gliomas, and in vitro invasion of β1-integrin expressing glioma cells can be blocked with the exogenous application of speciﬁc antisera (70). We have previously observed that stimulation of intracellular β1 integrin can increase the activation of MMP-2 and the invasiveness of glioma cells in vitro (101). In addition to cleaving structural ECM components, MMPs and the related proteinases, such as the ADAMs (a disintegrin and metalloproteinase), participate in the release of cell-membrane-bound precursor forms of many growth factors, including transforming growth factor-α (TGF-α) (102). 6. Therapy Inhibition of MMP activity should be a rational approach to glioma therapy. Thus far, a number of strategies have been utilised to modulate MMP expression/activity and assess subsequent changes in invasive potential. U87MG cells engineered to overexpress TIMP-2 demonstrated increased MMP-2 activation, indicating that an increase in physiological levels of TIMP-2 can promote MMP-2 activation and invasion in glioblastoma cells. However, exogenous administration or recombinant overexpression of higher amounts of TIMP-2 in U87MG cells resulted in the inhibition of MMP-2 activation. These results demonstrate that the complex balance between TIMP-2 and MMP-2 is a critical determinant of glioblastoma invasion as well as indicating that increasing TIMP-2 in glioblastoma patients may potentially cause adverse eﬀects, particularly in tumours containing high levels of MT1-MMP and MMP-2 (103). Studies have demonstrated

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

Role and Regulation of Matrix Metalloproteases in Brain Tumours

271

that regulation of pericellular proteolysis by MT-MMPs could provide an additional target for the inhibition of tumour cell invasion by inhibiting the activators of MMP. Glioma cells transfected with MT1-MMP cDNA constructs displayed increased cell surface expression of MT1-MMP and TIMP-2 and resulted in increased migration in spheroid outgrowth assays (104) as well as remodeling of extracellular matrix in vitro (105). Studies have demonstrated that cytokines, tumour necrosis factor-α and interferon-γ inhibit MMP-2 expression in glioma cells, resulting in decreased invasion (106). We have demonstrated that glioblastoma cells expressing antisense MMP-9 exhibited decreased migration and invasion in vitro and did not form tumours when injected intracranially in nude mice (107). Intracranial injections of glioblastoma cells (SNB19) infected with an adenovirus expressing antisense MMP-9 did not produce tumours in nude mice (86). A bicistronic Ad-construct with antisense uPAR and MMP-9 had more eﬀect in regard to the inhibition of invasion, angiogenesis and tumour growth in vivo (108). Our recent studies demonstrated that siRNA bicistronic construct for cathepsin B and MMP-9 completely repressed pre-established intracranial tumours (109). Stable transfection of PTEN (Phosphatase and tensin homologue) reduced MMP-9 secretion caused by hyaluronic acid-induced phosphorylation of focal adhesion kinase and ERK1/ERK2 signalling (110). Studies utilising native MMP inhibitors, such as the TIMPs, have demonstrated that down-regulation of both TIMP-1 and TIMP-2 contributes signiﬁcantly to the invasive potential of gliomas (92–94). Introduction of TIMP-1 cDNA into an invasive astrocytoma cell line reduced its invasive potential (111). Also, synthetic MMP inhibitors Batimastat and Marimastat eﬀectively reduce glioma invasion in vitro (112). Several reports demonstrated that glioma invasion could be suppressed by synthetic MMP inhibitors in vitro and in vivo. The peptidomimetic MMP inhibitors mimic the cleavage sites of MMP substrates and include Batimastat (BB-94) and Marimastat (BB-2516). The addition of these synthetic inhibitors eﬀectively reduced the invasion of glioma cells administered into matrigel and spheroid co-culture assays (112). However, Batimastat cannot be given orally and is no longer tested for the treatment of human cancer. The non-peptidomimetic MMP inhibitors are synthesised on the basis of the conformation of the MMP active site. Prinomastat/AG3340 signiﬁcantly inhibited glioma growth in a mouse model, but an in vitro study showed no signiﬁcant eﬀect on cell proliferation (113). In another model, AG3340 (100 mg/kg, once daily, i.p.) markedly inhibited U87 glioma

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

272

S.S. Lakka & J.S. Rao

growth and increased animal survival (114). BE16627B (BE), an inhibiter of zinc metalloproteinase, decreased MMP-2 and MMP-9 expression, and thus invasion, in a dose-dependent manner in the following malignant glioma cell lines: U87MG, U251MG and U373MG. Notably, the concentrations of BE required for this eﬀect were not cytotoxic (115). SI-27, a low molecular derivative of BE16627B, was found to be ten-fold greater in inhibiting MMP activity and four-fold greater in inhibiting tumour growth (116). The main target of these anti-MMP agents is the process of activation from pro-MMP to active MMP (117). It has been reported that post-surgical administration of PEX or PF-4/CTF signiﬁcantly reduced the incidence human malignant glioma recurrences for a signiﬁcant period of time (118). Recent studies demonstrated that testican 2 may contribute to glioma invasion by inactivating other testican family members and MT-MMPs (119). Chlorotoxin inhibits glioma cell invasion by inhibiting MMP-2 enzymatic activity of MMP-2 and causes a reduction in the surface expression of MMP-2 (120). The main objective of glioma therapy is to arrest tumour invasion and convert it to a controlled, localised disease. This type of control would contribute to enhanced surgical removal and more exact targeting for radiation and local delivery of therapeutic agents to the tumour margin. Accumulated lines of evidence indicate that MMPs play an essential role in brain tumour invasion. Therapeutic strategies that can inhibit a broad spectrum of MMPs may be beneﬁcial for retarding or preventing tumour growth and angiogenesis in brain tumour patients. To achieve this, further investigation and understanding of proteases at the molecular level should have an important eﬀect on the future development of new, target-selective treatments.

References 1. Russell, D. S. and Rubenstein, L. J. (1989) Pathology of the Tumours of the Nervous System, Edward Arnold, London 2. Sheline, G. E. (1986) Normal tissue tolerance and radiation therapy of gliomas of the adult brain (Blehen, N. M., ed.). Tumours of the Brain, Springer-Verlag, Berlin 3. Shapiro, J. R. (1986) Semin Oncol 13, 4–15 4. Gleave, J. R. (1986) Surgery of primary brain tumours (Blehen, N. M., ed.). Tumours of the Brain, Springer-Verlag, Berlin, Germany 5. De Clerck, Y. A., Shimada, H., Gonzalez-Gomez, I., and Raﬀel, C. (1994) J Neurooncol 18, 111–121 6. Rutka, J. T., Myatt, C. A., Giblin, J. R., Davis, R. L., and Rosenblum, M. L. (1987) Can J Neurol Sci 14, 25–30

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

Role and Regulation of Matrix Metalloproteases in Brain Tumours

273

7. Rutka, J. T., Giblin, J. R., Hoifodt, H. K., Dougherty, D. V., Bell, C. W., McCulloch, J. R., Davis, R. L., Wilson, C. B., and Rosenblum, M. L. (1986) Cancer Res 46, 5893–5902 8. McComb, R. D., Moul, J. M., and Bigner, D. D. (1987) J Neuropathol Exp Neurol 46, 623–633 9. Bjerkvig, R., Laerum, O. D., and Rucklidge, G. J. (1989) Cancer Res 49, 5424–5428 10. Rucklidge, G. J., Dean, V., Robins, S. P., Mella, O., and Bjerkvig, R. (1989) Cancer Res 49, 5419–5423 11. Lund-Johansen, M., Rucklidge, G. J., Milne, G., and Bjerkvig, R. (1991) Anticancer Res 11, 1001–1006 12. Paulus, W., Baur, I., Liszka, U., Drlicek, M., Leigh, I., and BrucknerTuderman, L. (1995) Virchows Arch 426, 199–202 13. Delpech, B., Maingonnat, C., Girard, N., Chauzy, C., Maunoury, R., Olivier, A., Tayot, J., and Creissard, P. (1993) Eur J Cancer 29A, 1012– 1017 14. Chintala, S. K., Gokaslan, Z. L., Go, Y., Sawaya, R., Nicolson, G. L., and Rao, J. S. (1996) Clin Exp Metastasis 14, 358–366 15. Chintala, S. K., Sawaya, R., Gokaslan, Z. L., Fuller, G., and Rao, J. S. (1996) Cancer Lett 101, 107–114 16. Chintala, S. K., Sawaya, R., Gokaslan, Z. L., and Rao, J. S. (1996) Cancer Lett 102, 57–63 17. Reith, A., Bjerkvig, R., and Rucklidge, G. J. (1994) Anticancer Res 14, 1071–1076 18. Giese, A., Loo, M. A., Rief, M. D., Tran, N., and Berens, M. E. (1995) Neurosurgery 37, 294–301 19. Koochekpour, S., Merzak, A., and Pilkington, G. J. (1995) Neurosci Lett 186, 53–56 20. Romanic, A. M. and Madri, J. A. (1994) Brain Pathol 4, 145–156 21. Yana, I. and Weiss, S. J. (2000) Mol Biol Cell 11, 2387–2401 22. Primakoﬀ, P. and Myles, D. G. (2000) Trends Genet 16, 83–87 23. Tang, B. L. (2001) Int J Biochem Cell Biol 33, 33–44 24. Cal, S., Obaya, A. J., Llamazares, M., Garabaya, C., Quesada, V., and Lopez-Otin, C. (2002) Gene 283, 49–62 25. Murphy, G., Knauper, V., Cowell, S., Hembry, R., Stanton, H., Butler, G., Freije, J., Pendas, A. M., and Lopez-Otin, C. (1999) Ann NY Acad Sci 878, 25–39 26. Shapiro, S. D., Fliszar, C. J., Broekelmann, T. J., Mecham, R. P., Senior, R. M., and Welgus, H. G. (1995) J Biol Chem 270, 6351–6356 27. Westermarck, J. and Kahari, V. M. (1999) FASEB J 13, 781–792 28. Nielsen, B. S., Timshel, S., Kjeldsen, L., Sehested, M., Pyke, C., Borregaard, N., and Dano, K. (1996) Int J Cancer 65, 57–62 29. Welgus, H. G., Campbell, E. J., Cury, J. D., Eisen, A. Z., Senior, R. M., Wilhelm, S. M., and Goldberg, G. I. (1990) J Clin Invest 86, 1496–1502 30. Leppert, D., Waubant, E., Galardy, R., Bunnett, N. W., and Hauser, S. L. (1995) J Immunol 154, 4379–4389

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

274

S.S. Lakka & J.S. Rao

31. Haas, T. L. and Madri, J. A. (1999) Trends Cardiovasc Med 9, 70–77 32. Unemori, E. N., Bair, M. J., Bauer, E. A., and Amento, E. P. (1991) J Biol Chem 266, 23477–23482 33. Gottschall, P. E. and Deb, S. (1996) Neuroimmunomodulation 3, 69–75 34. Maeda, A. and Sobel, R. A. (1996) J Neuropathol Exp Neurol 55, 300–309 35. Uhm, J. H., Dooley, N. P., Oh, L. Y., and Yong, V. W. (1998) Glia 22, 53–63 36. Muir, E. M., Adcock, K. H., Morgenstern, D. A., Clayton, R., von Stillfried, N., Rhodes, K., Ellis, C., Fawcett, J. W., and Rogers, J. H. (2002) Brain Res Mol Brain Res 100, 103–117 37. Kruger, A., Soeltl, R., Lutz, V., Wilhelm, O. G., Magdolen, V., Rojo, E. E., Hantzopoulos, P. A., Graeﬀ, H., Gansbacher, B., and Schmitt, M. (2000) Cancer Gene Ther 7, 292–299 38. Itoh, T., Tanioka, M., Yoshida, H., Yoshioka, T., Nishimoto, H., and Itohara, S. (1998) Cancer Res 58, 1048–1051 39. Bjerkvig, R., Laerum, O. D., and Mella, O. (1986) Cancer Res 46, 4071–4079 40. Apodaca, G., Rutka, J. T., Bouhana, K., Berens, M. E., Giblin, J. R., Rosenblum, M. L., McKerrow, J. H., and Banda, M. J. (1990) Cancer Res 50, 2322–2329 41. Rucklidge, G. J., Lund-Johansen, M., Milne, G., and Bjerkvig, R. (1990) Biochem Biophys Res Commun 172, 544–550 42. Nakano, A., Tani, E., Miyazaki, K., Furuyama, J., and Matsumoto, T. (1993) Biochem Biophys Res Commun 192, 999–1003 43. Nakagawa, T., Kubota, T., Kabuto, M., Sato, K., Kawano, H., Hayakawa, T., and Okada, Y. (1994) J Neurosurg 81, 69–77 44. Nakagawa, T., Kubota, T., Kabuto, M., Fujimoto, N., and Okada, Y. (1996) J Neurooncol 28, 13–24 45. Uhm, J. H., Dooley, N. P., Villemure, J. G., and Yong, V. W. (1996) Clin Exp Metastasis 14, 421–433 46. Yamamoto, M., Mohanam, S., Sawaya, R., Fuller, G. N., Seiki, M., Sato, H., Gokaslan, Z. L., Liotta, L. A., Nicolson, G. L., and Rao, J. S. (1996) Cancer Res 56, 384–392 47. Forsyth, P. A., Laing, T. D., Gibson, A. W., Rewcastle, N. B., Brasher, P., Sutherland, G., Johnston, R. N., and Edwards, D. R. (1998) J Neurooncol 36, 21–29 48. Forsyth, P. A., Wong, H., Laing, T. D., Rewcastle, N. B., Morris, D. G., Muzik, H., Leco, K. J., Johnston, R. N., Brasher, P. M., Sutherland, G., and Edwards, D. R. (1999) Br J Cancer 79, 1828–1835 49. Raithatha, S. A., Muzik, H., Muzik, H., Rewcastle, N. B., Johnston, R. N., Edwards, D. R., and Forsyth, P. A. (2000) Neuro-oncol 2, 145–150 50. Sawaya, R. E., Yamamoto, M., Gokaslan, Z. L., Wang, S. W., Mohanam, S., Fuller, G. N., McCutcheon, I. E., Stetler-Stevenson, W. G., Nicolson, G. L., and Rao, J. S. (1996) Clin and Exp Metastasis 14, 35–42 51. Rao, J. S., Steck, P. A., Mohanam, S., Stetler-Stevenson, W. G., Liotta, L. A., and Sawaya, R. (1993) Cancer Res 53, 2208–2211 52. Rao, J. S., Yamamoto, M., Mohaman, S., Gokaslan, Z. L., Fuller, G. N., Stetler-Stevenson, W. G., Rao, V. H., Liotta, L. A., Nicolson, G. L., and Sawaya, R. E. (1996) Clin Exp Metastasis 14, 12–18

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

Role and Regulation of Matrix Metalloproteases in Brain Tumours

275

53. Sawaya, R. E., Yamamoto, M., Gokaslan, Z. L., Wang, S. W., Mohanam, S., Fuller, G. N., McCutcheon, I. E., Stetler-Stevenson, W. G., Nicolson, G. L., and Rao, J. S. (1996) Clin Exp Metastasis 14, 35–42 54. Sawaya, R., Go, Y., Kyritisis, A. P., Uhm, J., Venkaiah, B., Mohanam, S., Gokaslan, Z. L., and Rao, J. S. (1998) Biochem Biophys Res Commun 251, 632–636 55. Qin, H., Moellinger, J. D., Wells, A., Windsor, L. J., Sun, Y., and Benveniste, E. N. (1998) J Immunol 161, 6664–6673 56. Bodey, B., Bodey, B., Jr., Siegel, S. E., and Kaiser, H. E. (2000) Anticancer Res 20, 3287–3292 57. Bodey, B., Bodey, B., Jr., Siegel, S. E., and Kaiser, H. E. (2000) In Vivo 14, 667–673 58. Vince, G. H., Wagner, S., Pietsch, T., Klein, R., Goldbrunner, R. H., Roosen, K., and Tonn, J. C. (1999) Int J Dev Neurosci 17, 437–445 59. Mercapide, J., Lopez, D. C., Castresana, J. S., and Klein-Szanto, A. J. (2003) Int J Cancer 106, 676–682 60. Thorns, V., Walter, G. F., and Thorns, C. (2003) Anticancer Res 23, 3937–3944 61. d’Ortho, M. P., Will, H., Atkinson, S., Butler, G., Messent, A., Gavrilovic, J., Smith, B., Timpl, R., Zardi, L., and Murphy, G. (1997) Eur J Biochem 250, 751–757 62. Ohuchi, E., Imai, K., Fujii, Y., Sato, H., Seiki, M., and Okada, Y. (1997) J Biol Chem 272, 2446–2451 63. Hur, J. H., Park, M. J., Park, I. C., Yi, D. H., Rhee, C. H., Hong, S. I., and Lee, S. H. (2000) J Korean Med Sci 15, 309–314 64. Shofuda, K., Moriyama, K., Nishihashi, A., Higashi, S., Mizushima, H., Yasumitsu, H., Miki, K., Sato, H., Seiki, M., and Miyazaki, K. (1998) J Biochem (Tokyo) 124, 462–470 65. Kachra, Z., Beaulieu, E., Delbecchi, L., Mousseau, N., Berthelet, F., Moumdjian, R., Del Maestro, R., and Beliveau, R. (1999) Clin Exp Metastasis 17, 555–566 66. Nakada, M., Nakamura, H., Ikeda, E., Fujimoto, N., Yamashita, J., Sato, H., Seiki, M., and Okada, Y. (1999) Am J Pathol 154, 417–428 67. Llano, E., Pendas, A. M., Freije, J. P., Nakano, A., Knauper, V., Murphy, G., and Lopez-Otin, C. (1999) Cancer Res 59, 2570–2576 68. Satoh, K., Suzuki, N., and Yokota, H. (2000) Neurosci Lett 289, 177–180 69. Sternlicht, M. D. and Werb, Z. (2001) Annu Rev Cell Dev Biol 17, 463–516 70. Rooprai, H. K., Rucklidge, G. J., Panou, C., and Pilkington, G. J. (2000) Br J Cancer 82, 52–55 71. Qin, H., Sun, Y., and Benveniste, E. N. (1999) J Biol Chem 274, 29130–29137 72. Hu, B., Guo, P., Fang, Q., Tao, H. Q., Wang, D., Nagane, M., Huang, H. J., Gunji, Y., Nishikawa, R., Alitalo, K., Cavenee, W. K., and Cheng, S. Y. (2003) Proc Natl Acad Sci USA 100, 8904–8909 73. VanMeter, T. E., Rooprai, H. K., Kibble, M. M., Fillmore, H. L., Broaddus, W. C., and Pilkington, G. J. (2001) J Neurooncol 53, 213–235

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

276

S.S. Lakka & J.S. Rao

74. Choe, G., Park, J. K., Jouben-Steele, L., Kremen, T. J., Liau, L. M., Vinters, H. V., Cloughesy, T. F., and Mischel, P. S. (2002) Clin Cancer Res 8, 2894–2901 75. Chintala, S. K., Sawaya, R., Aggarwal, B. B., Majumder, S., Giri, D. K., Kyritsis, A. P., Gokaslan, Z. L., and Rao, J. S. (1998) J Biological Chemistry 273, 13545–13551 76. Chintala, S. K., Kyritsis, A. P., Mohan, P. M., Mohanam, S., Sawaya, R., Gokslan, Z., Yung, W. K., Steck, P., Uhm, J. H., Aggarwal, B. B., and Rao, J. S. (1999) Molecular Carcinogenesis 26, 274–285 77. Sameshima, T., Nabeshima, K., Toole, B. P., Yokogami, K., Okada, Y., Goya, T., Koono, M., and Wakisaka, S. (2000) Cancer Lett 157, 177–184 78. Hamasuna, R., Kataoka, H., Moriyama, T., Itoh, H., Seiki, M., and Koono, M. (1999) Int J Cancer 82, 274–281 79. Park, M. J., Park, I. C., Hur, J. H., Rhee, C. H., Choe, T. B., Yi, D. H., Hong, S. I., and Lee, S. H. (2000) Neurosci Lett 290, 201–204 80. Deb, S., Zhang, J. W., and Gottschall, P. E. (1999) J Neurosci Res 55, 44–53 81. Pagenstecher, A., Stalder, A. K., and Campbell, I. L. (1997) J Immunol Methods 206, 1–9 82. von Bredow, D. C., Cress, A. E., Howard, E. W., Bowden, G. T., and Nagle, R. B. (1998) Biochem J 331 (Pt 3), 965–972 83. Uhm, J. H., Dooley, N. P., Villemure, J. G., and Yong, V. W. (1996) Clin Exp Metastasis 14, 421–433 84. da Rocha, A. B., Mans, D. R., Lenz, G., Fernandes, A. K., de Lima, C., Monteiro, V. F., Goncalves, D., Moreira, J. C., Brunetto, A. L., Rodnight, R., and Schwartsmann, G. (2000) Pathobiology 68, 113–123 85. Lakka, S. S., Jasti, S. L., Kyritsis, A. P., Yung, W. K., Ali-Osman, F., Nicolson, G. L., and Rao, J. S. (2000) Clin Exp Metastasis 18, 245–252 86. Lakka, S. S., Rajan, M., Gondi, C., Yanamandra, N., Chandrasekar, N., Jasti, S. L., Adachi, Y., Siddique, K., Gujrati, M., Olivero, W., Dinh, D. H., Kouraklis, G., Kyritsis, A. P., and Rao, J. S. (2002) Oncogene 21, 8011–8019 87. Fridman, R., Toth, M., Pena, D., and Mobashery, S. (1995) Cancer Res 55, 2548–2555 88. Chambers, A. F. and Matrisian, L. M. (1997) J Natl Cancer Inst 89, 1260–1270 89. Apte, S. S., Olsen, B. R., and Murphy, G. (1995) J Biol Chem 270, 14313–14318 90. Greene, J., Wang, M., Liu, Y. E., Raymond, L. A., Rosen, C., and Shi, Y. E. (1996) J Biol Chem 271, 30375–30380 91. Blavier, L., Henriet, P., Imren, S., and DeClerck, Y. A. (1999) Ann NY Acad Sci 878, 108–119 92. Mohanam, S., Wang, S. W., Rayford, A., Yamamoto, M., Sawaya, R., Nakajima, M., Liotta, L. A., Nicolson, G. L., Stetler-Stevenson, W. G., and Rao, J. S. (1995) Clin Exp Metastasis 13, 57–62 93. Lampert, K., Machein, U., Machein, M. R., Conca, W., Peter, H. H., and Volk, B. (1998) Am J Pathol 153, 429–437

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

Role and Regulation of Matrix Metalloproteases in Brain Tumours

277

94. Nakano, A., Tani, E., Miyazaki, K., Yamamoto, Y., and Furuyama, J. (1995) J Neurosurg 83, 298–307 95. Gasson, J. C., Golde, D. W., Kaufman, S. E., Westbrook, C. A., Hewick, R. M., Kaufman, R. J., Wong, G. G., Temple, P. A., Leary, A. C., and Brown, E. L. (1985) Nature 315, 768–771 96. Wingﬁeld, P. T., Sax, J. K., Stahl, S. J., Kaufman, J., Palmer, I., Chung, V., Corcoran, M. L., Kleiner, D. E., and Stetler-Stevenson, W. G. (1999) J Biol Chem 274, 21362–21368 97. Groft, L. L., Muzik, H., Rewcastle, N. B., Johnston, R. N., Knauper, V., Laﬂeur, M. A., Forsyth, P. A., and Edwards, D. R. (2001) Br J Cancer 85, 55–63 98. Bachman, K. E., Herman, J. G., Corn, P. G., Merlo, A., Costello, J. F., Cavenee, W. K., Baylin, S. B., and Graﬀ, J. R. (1999) Cancer Res 59, 798–802 99. Streuli, C. (1999) Curr Opin Cell Biol 11, 634–640 100. Egeblad, M. and Werb, Z. (2002) Nat Rev Cancer 2, 161–174 101. Chintala, S. K., Sawaya, R., Gokaslan, Z. L., and Rao, J. S. (1996) Cancer Lett 103, 201–208 102. Peschon, J. J., Slack, J. L., Reddy, P., Stocking, K. L., Sunnarborg, S. W., Lee, D. C., Russell, W. E., Castner, B. J., Johnson, R. S., Fitzner, J. N., Boyce, R. W., Nelson, N., Kozlosky, C. J., Wolfson, M. F., Rauch, C. T., Cerretti, D. P., Paxton, R. J., March, C. J., and Black, R. A. (1998) Science 282, 1281–1284 103. Lu, K. V., Jong, K. A., Rajasekaran, A. K., Cloughesy, T. F., and Mischel, P. S. (2004) Lab Invest 84, 8–20 104. Hotary, K., Allen, E., Punturieri, A., Yana, I., and Weiss, S. J. (2000) J Cell Biol 149, 1309–1323 105. Deryugina, E. I., Bourdon, M. A., Luo, G. X., Reisfeld, R. A., and Strongin, A. (1997) J Cell Sci 110, 2473–2482 106. Qin, H., Moellinger, J. D., Wells, A., Windsor, L. J., Sun, Y., and Benveniste, E. N. (1998) J Immunol. 161, 6664–6673 107. Kondraganti, S., Mohanam, S., Chintala, S. K., Kin, Y., Jasti, S. L., Nirmala, C., Lakka, S. S., Adachi, Y., Kyritsis, A. P., Ali-Osman, F., Sawaya, R., Fuller, G. N., and Rao, J. S. (2000) Cancer Res 60, 6851–6855 108. Lakka, S. S., Gondi, C. S., Yanamandra, N., Dinh, D. H., Olivero, W. C., Gujrati, M., and Rao, J. S. (2003) Cancer Res 63, 2454–2461 109. Lakka, S. S., Gondi, C. S., Yanamandra, N., Olivero, W. C., Dinh, D. H., Gujrati, M., and Rao, J. S. (2004) Oncogene In press 110. Park, M. J., Kim, M. S., Park, I. C., Kang, H. S., Yoo, H., Park, S. H., Rhee, C. H., Hong, S. I., and Lee, S. H. (2002) Cancer Res 62, 6318–6322 111. Matsuzawa, K., Fukuyama, K., Hubbard, S. L., Dirks, P. B., and Rutka, J. T. (1996) J Neuropathol Exp Neurol 55, 88–96 112. Tonn, J. C., Kerkau, S., Hanke, A., Bouterfa, H., Mueller, J. G., Wagner, S., Vince, G. H., and Roosen, K. (1999) Int J Cancer 80, 764–772 113. Price, A., Shi, Q., Morris, D., Wilcox, M. E., Brasher, P. M., Rewcastle, N. B., Shalinsky, D., Zou, H., Appelt, K., Johnston, R. N., Yong, V. W., Edwards, D., and Forsyth, P. (1999) Clin Cancer Res 5, 845–854

May 26, 2005

16:8 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch11

278

S.S. Lakka & J.S. Rao

114. Shalinsky, D. R., Brekken, J., Zou, H., McDermott, C. D., Forsyth, P., Edwards, D., Margosiak, S., Bender, S., Truitt, G., Wood, A., Varki, N. M., and Appelt, K. (1999) Ann NY Acad Sci 878, 236–270 115. Watanabe, M., Higashi, T., Hashimoto, M., Tomoda, I., Tominaga, S., Hashimoto, N., Morimoto, S., Yamauchi, Y., Nakatsukasa, H., and Kobayashi, M. (1987) Hepatogastroenterology 34, 120–122 116. Naito, K., Kanbayashi, N., Nakajima, S., Murai, T., Arakawa, K., Nishimura, S., and Okuyama, A. (1994) Int J Cancer 58, 730–735 117. Yoshida, D., Watanabe, K., Noha, M., Takahashi, H., and Teramoto, A. (2002) J Neurooncol 58, 1–11 118. Bello, L., Giussani, C., Carrabba, G., Pluderi, M., Lucini, V., Pannacci, M., Caronzolo, D., Tomei, G., Villani, R., Scaglione, F., Carroll, R. S., and Bikfalvi, A. (2002) Clin Cancer Res 8, 3539–3548 119. Nakada, M., Miyamori, H., Yamashita, J., and Sato, H. (2003) Cancer Res 63, 3364–3369 120. Deshane, J., Garner, C. C., and Sontheimer, H. (2003) J Biol Chem 278, 4135–4144

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

CHAPTER 12 MATRIX METALLOPROTEINASES AND RELATED PROTEINS IN ALZHEIMER’S DISEASE, PARKINSON’S DISEASE AND OTHER NEURODEGENERATIVE DISORDERS

C.H. Parsons1 , I. Koolwijk2 , T. Roy1 , C. Roy1 , D. Johnstone1 , C.M.P. Vos3 , K. Conant4 and E.A. Milward1,∗ 1

School of Biomedical Sciences, University of Newcastle, Callaghan, NSW 2308, Australia 2

Maatzorg de Werven, Parent and Child Care, pobox 603, 2600 AP, Delft, The Netherlands

3

Department of Pathology, Vrije Universiteit Medical Centre, De Boelelaan 1117, 1007 MB, Amsterdam, The Netherlands 4

Johns Hopkins University, Department of Neurology, Meyer 6-109, 600 North Wolfe Street, Baltimore, MD 21287 E-mail: ∗[email protected]

1. Introduction The main neurodegenerative diseases can be loosely grouped into two broad categories — memory disorders or ‘dementias’ such as Alzheimer’s disease (AD) and movement disorders such as Parkinson’s disease (PD). As reviewed elsewhere (1–6), in general, these diseases involve progressive neuronal dysfunction over periods of up to a decade or more, usually with eventual loss of neurons or synapses aﬀecting certain speciﬁc, vulnerable brain regions earlier and more severely than other regions. There can be substantial overlap between these disease groups. For example, some AD patients develop movement deﬁcits and some patients presenting with PD develop memory impairment and sometimes AD (7–9). To some extent this reﬂects overlap in the regions aﬀected by these disorders but, in part, it is also because, despite diﬀerent etiologies, neurodegenerative Correspondence to: E.A. Milward 279

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

280

C.H. Parsons et al

diseases often involve common neurochemical and morphological perturbances, for example changes in cytoskeletal ﬁlamentous proteins such as tau or disrupted cell-cell communications (e.g. dendrite pruning, synapse loss) (1, 2, 9). The same broad molecular networks are often involved in the course of neurodegeneration, including mediators of inﬂammatory or oxidative damage (10–14), caspases involved in cell death (3, 15–18) and, as reviewed here, molecules involved in the re-organisation of the extracellular matrix after cell damage or death. The physical alterations in brain structures that occur when synapses and cells are destroyed in any neurodegenerative disease necessarily entail alterations in the extracellular matrix, so it is not surprising that MMPs alter in several neurodegenerative diseases, including both AD and PD. 2. Alzheimer’s Disease and Other Dementias The dementias, including AD, can be deﬁned as syndromes involving sustained decline of memory and at least one other cognitive domain, that signiﬁcantly impede social or occupational functioning (19, 20). Any of multiple cognitive or behavioural domains can be impaired, including judgment and abstract thinking, time and space orientation, changes in personality and other neuropsychiatric symptoms (21–23). An integrative analysis of 47 surveys across 17 countries estimates rates of under 1% for dementia from any cause in persons aged 60 to 69 years, rising to approximately 40% in persons 90 to 95 years old (24). The leading cause of dementia in most Western nations is AD. Deﬁnitive diagnosis of AD requires post-mortem or (rarely) biopsy veriﬁcation of abnormally high densities of two characteristic lesions — amyloid plaques and tau neuroﬁbrillary tangles. Alzheimer’s disease is a chronic, progressive and ultimately fatal disorder. Non-Alzheimer’s dementias such as vascular dementia (25–28), dementia with Lewy bodies (DLB) (7, 29–31) and other dementias such as frontotemporal dementia or alcohol-related dementia are responsible for 25% to 50% of all cases of dementia and again are usually progressive and ultimately fatal (25, 32). Dementia potentially attributable to cerebrovasculopathy is a contentious area and since other chapters discuss changes in MMPs accompanying stroke and other cerebrovasculopathy, we will not address this speciﬁcally here. However, it is noted that vascular pathologies frequently co-exist with AD or PD and that, while the ramiﬁcations are still poorly understood, there is the potential for confounding eﬀects when considering phenomena relating to MMPs.

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

281

3. Parkinson’s Disease and Dementia with Lewy Bodies Parkinson’s disease is also a chronic, progressive and ultimately fatal disorder. It is caused by the degeneration of dopamine-producing neurons particularly in brain areas allied with motor function (5, 6). There is gradual onset of tremors, muscle rigidity and bradykinesia (slowing of movements). As many as 40% of people with PD develop dementia (5). In PD, spherical, intracytoplasmic, eosinophilic neuronal inclusion bodies called Lewy bodies are present in subcortical nuclei, as reviewed elsewhere (33). Abnormalities in various proteins contribute to Parkinson’s disease (34), including β-synuclein, which can be present in lesions in both AD and PD (33, 35–38). A related disorder, dementia with Lewy bodies (DLB), is distinguished by cortical Lewy bodies and has features of both AD and PD (e.g. Aβ lesions, Parkinsonian symptoms), which can provide valuable comparative insights (7, 29–31, 39, 40). There are few if any studies of MMPs changes in DLB but this is likely to be a fruitful area for future investigations. The MMPs have been studied in a few rarer neurodegenerative disorders but so far most research has been conducted on AD and we will review this ﬁrst.

4. Expression of Matrix Metalloproteinases in Alzheimer’s Disease and Related Disorders In diseases such as AD and PD, which are more frequent in older people, it is important to distinguish speciﬁc disease-related eﬀects from nonspeciﬁc, aging-related eﬀects. The MMPs alter with aging in diverse tissues, including skin, bones, teeth, eyes, kidneys, lungs and pancreas. The structural remodelling of the cardiovascular system that occurs with aging or disease is also associated with alterations in MMPs. This remodelling involves phenomena also believed to be important in the aging brain such as age-linked glycation and stable intermolecular cross-linking of MMP substrates (41–43). However, while aging-related alterations in MMP systems could contribute indirectly to dementia (44), this chapter will focus primarily on speciﬁc disease-related eﬀects. The inﬂammatory changes in AD and related neurodegenerative disorders (45) are in general relatively limited compared to diseases such as multiple sclerosis or HIV-dementia. Consistent with this, MMP changes in AD and related disorders are also often not as great and can be obscured by variability in small or inadequately characterised patient groups. Nonetheless, several MMPs have been reported to alter in AD (Table 1).

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

282

C.H. Parsons et al

Table 1. Matrix metalloproteinases and related enzymes which have been proposed to be associated with Alzheimer’s disease. The table summarises the main MMPs, ADAMs and TIMPs reported to be associated with Alzheimer’s disease, with select substrates of potential relevance, from Woessner and Nagase (118) or as cited, and proposed associations. The list should not be considered exhaustive. Abbreviations: α1 -ACT alpha1 -antichymotrypsin; AD Alzheimer’s disease; ADAM a disintegrin and metalloproteinase; AGG aggrecan; α2 M alpha2 -macroglobulin, C1q complement component 1q; CSF cerebrospinal ﬂuid; DEC decorin; ENT entactin; FGF-R1 ﬁbroblast growth factor receptor 1, IL interleukin; LAM laminin; link protein LP; KiSS-1 metastasis suppressor gene product (229); MBP myelin basic protein; MT-MMP membrane-type metalloproteinase; TACE TNFα convertase enzyme; TEN tenascin; TIMP tissue inhibitor of metalloproteinase; TNFα tumour necrosis factor alpha; SPARC secreted protein acidic and rich in cysteine, also known as osteonectin. Enzyme

Selected Substrates

Disease Associations

MMP-1 (interstitial collagenase)

Collagen types I–III+, LAM, TEN, ENT, AGG, C1q, link protein, α1 -ACT, MBP, proTNFα, IL-1β, α2 M

Increased in AD brain (46), neurotoxic in vitro (47)

MMP-2 (gelatinase A)

Collagen IV, LAM 5, FGF-R1, Aβ, DEC, proTNFα, TEN, ENT, IL-1β, AGG, LP, MBP, α1 -ACT, KiSS1

Induced by Aβ in vitro (103), possibly increased (208) in AD CSF (see text)

MMP-3 (transin, stromelysin)

LAM, pro-MMPs, proTNFα, AGG, IL-1β, α1 -ACT, ENT, TEN, link protein, MBP, α2 M

Increased in AD brain (57), induced by Aβ in vitro (103, 104)

MMP-9 (gelatinase B)

Collagen IV/V/others, vitronectin, gelatin-1, elastin, plasminogen, link protein, AGG, α2 M, SPARC, LAM, substance P, MBP, KiSS-1 (229)

Increased in AD brain (50, 58, 59) & plasma (51), in canine Aβ plaques (130), induced by Aβ in vitro (103)

MMP-14 (MT1-MMP)

MMPs e.g. pro-MMP-2, MMP-13 α2 M, proTNFα, collagen I–III, LAM, ENT, AGG, KiSS-1 (229)

Increased in AD brain (53)

MMP-24 (MT5-MMP)

Pro-MMP-2 (230, 231), KiSS-1 (229)

Brain speciﬁc and co-localised with Aβ in AD brain (56)

ADAM-1 & 2 ADAM9, 10, 15, 17 (TACE) & others

Bind integrin alpha(9)beta(1) (232) APP; TNFα; Notch-related; others

Possibly increased in AD hippocampal neurons (233) α-Secretion of APP, reviewed (84), Notch/presenilin roles (66, 234, 235); ADAM-15 reduced in AD brain (96)

TIMPsreviewed (217)

TIMP-1: All bar MT-MMPs, ADAM-10 TIMP-2: Inhibits all MMPs tested

Increased in AD brain (98), induced by Aβ in vitro (106), possibly increased (61) in AD CSF (see text)

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

283

As determined by enzyme linked immunosorbent assay (ELISA), MMP-1 has been found to increase by about 50% in AD cortex compared to controls (46). Since MMP-1 can be neurotoxic in vitro (47), it will be important to conﬁrm this and assess whether MMP-1 changes in speciﬁc regions correlate with synaptic loss, other cell dysfunctions or death in neurodegenerative diseases. Changes associated with MMP-2 appear more modest. While increases in MMP-2 have been reported after stroke (48), no changes were observed in white matter of patients with vascular dementia (49) or in AD hippocampi (50) or plasma (51). However, Yamada and colleagues (52) reported microglial MMP-2 is more intense in parietal white matter in AD compared to normal or stroke brains although no diﬀerences in MMP-2 transcripts were observed by reverse transcriptase PCR, suggesting a posttranscriptional eﬀect (53). Yoshiyama and colleagues (53) found MMP-14 (MT1-MMP), which participates in MMP-2 activation, is increased in AD brains so active MMP-2 might increase in some regions. While no changes were observed in levels of MMP-16 (MT3-MMP) (53), there is some evidence for involvement of another membrane-associated MMP, MMP-24 (MT5-MMP), which is expressed in subsets of hippocampal, cortical and cerebellar neurons (54–56) and may preferentially co-localise with Aβ deposits in AD (56). While little information is available for MMP-3, it is present in white matter in normal and AD brains, and MMP-3 staining has been observed in some senile plaques, more frequently and intensely in the parietal cortex than in the hippocampus (57). The signiﬁcance of this is unclear. In contrast, various groups report abnormalities in MMP-9 expression (51, 58–60). The hippocampus, a region particularly vulnerable in AD, shows elevated levels of pro-MMP-9 in AD patients compared to healthy, age-matched subjects (58). Abnormal MMP-9 immunoreactivity has been reported to be associated with neuroﬁbrillary tangles and senile plaques, with neuronal cytoplasm and in the vascular walls in AD brains (59). To what extent this MMP-9 is active is unknown. Elevated MMP-9 levels have also been observed in the plasma of AD patients (51). Whether this is of CNS or peripheral origin or both is unclear. While one study failed to detect MMP-9 in the CSF of AD patients (61), a later study (60) did detect low levels of MMP-9 and MMP-2 in AD CSF but these were comparable to control values. In contrast patients with clinically diagnosed multi-infarct or small vessel vascular dementia were found to have signiﬁcantly elevated CSF MMP-9 levels in this study.

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

284

C.H. Parsons et al

In summary, AD appears to show a proﬁle of MMP-9 elevation in the CNS accompanied by changes in one or more other MMPs, probably including MMP-1, but with relative little change in MMP-2. 5. Production and Clearance of Amyloid in Alzheimer’s Disease While it appears likely that one or more MMPs are speciﬁcally elevated in AD, the functional and pathogenic consequences of abnormal MMP expression in AD are unknown. One possibility that has sometimes been put forward is that MMPs may contribute to either the production or the clearance of amyloid. The 4 kDa Aβ peptide which accumulates in amyloid plaques in AD is released from amyloid precursor protein (APP) by proteolytic enzymes dubbed secretases (62–68). Cutting APP within the Aβ domain, at the ‘α-secretase’ site (Fig. 1), prevents release of full-length intact Aβ. If proteolysis occurs instead at the ‘β-secretase’ site at the N -terminus of the Aβ domain, soluble, potentially amyloidogenic Aβ peptide is generated. The main β-site APP cleavage enzyme, BACE, is an endosomal transmembrane aspartyl protease and atypical pepsin family member (69–75). The C-terminus of Aβ is primarily produced by ‘γ-secretase’ cleavage involving an aspartyl protease complex containing presenilin (PS1 or PS2), nicastrin (Nct), anterior pharynx-defective-1 (APH-1) and presenilin enhancer-2 (PEN-2) (76, 77). Various MMPs can digest APP or related peptides and could alter Aβ production in some scenarios (58, 78–80) but a range of studies in mouse deletion mutants (81) and other systems argue against prominent MMP involvement in γ- or β-cleavage (62–75). The MMPs also do not appear to have prominent primary roles in clearing Aβ in comparison to Aβ-degrading metalloproteinases such as neprilysin and insulin-degrading enzyme (IDE), which inhibit plaque formation in APP transgenic mice (82). However, MMPs could potentially contribute to human Aβ clearance if other enzymes fail — MMP-2 hydrolyses human Aβ1–40 and Aβ1–42 (83) and MMP-9 can also degrade Aβ (58). 6. The ADAMs Family in Alzheimer’s Disease While the MMPs are not prominent in γ- or β-cleavage, related metalloproteases of the ADAMs family (a disintegrin and metalloproteinase) appear likely to be involved in APP α-secretion (84), and possibly also in clearing Aβ. These enzymes might have protective or defensive roles

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

(A)

285

Membrane

α-secretases

APP N

C

A Domain Aβ γ -secretases

Extracellular

Intracellular

CT83

sAPP sAPPα N

C

p3

sAPP sAPPα

CT57-59

N

C Nucleus Extracellular Environment

β-secretases

(B)

Membrane

APP N

C

A Domain Aβ γ -secretases

Extracellular N

A Domain Aβ

sAPP sAPPβ N

Intracellular

CT99

sAPP sAPPβ

A Aβ40−442 A Aβ Domain

C

CT57-59 C Nucleus

Extracellular Environment

Fig. 1. Overview of APP processing pathways. (A) Cleavage of APP by α-secretases within the Aβ domain, precluding formation of intact Aβ and releasing a truncated, secreted APP species (sAPPα) and other products (p3 and C-terminal CT fragments) following subsequent γ-secretase cleavage. (B) Alternatively APP can be cleaved ﬁrst by β-secretases then by γ-secretases to release the Aβ peptide (usually containing 40– 42 amino acid residues), a shorter truncated, secreted APP species (sAPPβ) and a CT fragment. Both (A) and (B) illustrate cleavage at the cell membrane, releasing soluble APP species or Aβ into the extracellular environment and Aβ can also be produced by APP cleavage within intracellular compartments. Modiﬁed from (228).

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

286

C.H. Parsons et al

in AD since stimulating APP α-secretion reduces Aβ formation in many systems (85–90). One ADAM able to catalyse both constitutive and pharmacologically stimulated endogenous α-secretase cleavage of APP is ADAM-10 (91–95). While ADAM-10 is unlikely to be the only ADAM involved in AD, even moderate neuronal over-expression of ADAM-10 in a mouse model expressing human APP carrying the AD-associated V717I mutation leads to increased APP α cleavage with near complete abolition of plaque deposition (95). Conversely, expressing a catalytically inactive dominant ADAM-10 mutant in these mice increased amyloid plaque load up to three-fold (95). Expression of ADAM-10 in human brain is reported to increase during aging and to be associated with amyloid deposits in AD and Down syndrome brains (96), possibly indicating an inadequate compensatory response. 7. The TIMPs in Alzheimer’s Disease The systems regulating the various secretases are unknown. However, factors controlling interactions between ADAMs and their natural inhibitors, the TIMPs, could potentially become an important research focus for treatment strategies that seek to redirect APP processing down nonamyloidogenic pathways. Both TIMP-1 and TIMP-3 can inhibit the α-secretase activity of ADAM-10 (97) and, as described elsewhere in this book, the TIMPs are also natural inhibitors of MMPs. Various studies have investigated the presence of TIMPs in AD. Immunoreactivity for TIMPs is observed in dendrites and cell bodies of cerebellar Purkinje cells in normal and AD brains and in plaques and tau neuroﬁbrillary tangles in the hippocampus and cerebral cortex of AD brains (98). This could increase build-up of amyloid or ECM proteins (see below). It is unclear whether TIMPs contribute to neuropathology in AD or instead curb excessive and potentially damaging activity of MMPs. It would be informative, although diﬃcult, to conduct comparative studies to determine how TIMP, MMP and ADAM activities change relative to each other in diﬀerent regions at diﬀerent disease stages. Plasma and CSF levels of TIMPs are probably not a useful guide to brain changes. While Lorenzl and colleagues using ELISA found TIMP-1 and TIMP-2 are increased in AD CSF but not plasma (61), neither Wollmer and colleagues, who also measured TIMP protein levels (99) nor Adair and colleagues, who examined TIMP activity by reverse zymography (60), found evidence for changes in TIMP-1 and TIMP-2 levels or activities in AD CSF.

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

287

8. Regulation of the MMPs in Alzheimer’s Disease The MMPs can be up-regulated in response to inﬂammatory cytokines, cell damage and other factors associated with tissue degeneration. Glial pro-inﬂammatory cytokines (e.g. TNFα, IL-1β) can be induced by Aβ and increase in AD (100–102). Both MMPs and TIMPs can be induced by these cytokines as well as by some Aβ peptide forms (103–109) but caution is required as MMP regulation can diﬀer substantially between cell systems and with aging. For example, MMP induction by IL-1β is reduced in ﬁbroblasts from older donors (110–112). There may be other regulatory mechanisms speciﬁc to AD. Kunitz serine protease inhibitor (KPI/‘serpin’) domains in some APP isoforms may inhibit plasmin and suppress activation of MMPs by plasmin cascades (113). In addition, APP can inhibit MMP-2 and possibly other MMPs directly (78, 114). In HT1080 ﬁbrosarcoma cells, MMP-14 (MT1-MMP) generates an unusual, truncated form of soluble APP lacking the domain that inhibits MMP-2 (115). As MMP-2 in turn can be activated by MMP-14 (115) and, as above, can degrade Aβ in vitro (83), regulation by MMP-14 could increase Aβ degradation. Another potential regulatory mechanism involves C-Jun, which is induced by Aβ and in turn induces MMP transcription through AP-1 regulatory sites (103). Presenilin 1 can suppress C-Jun (116) and could potentially down-regulate MMPs. Many MMPs strongly bind α2 -macroglobulin (α2 M), which is a 150-fold better MMP-1 substrate than type I collagen and out-competes equimolar quantities of TIMP-1 for MMP binding (117, 118). The α2 M receptor is the low density lipoprotein receptor-related protein (LRP), also known as CD91 (119), a major neuronal receptor for apolipoprotein E (APOE), Aβ and tissue plasminogen activator (tPA) (120, 121). Many cytokines and growth factors (e.g. TNFα, IL-1β, TGF-β) are bound by α2 M activated by plasmin or other proteases (122). While the size of α2 M (725 kDa) usually conﬁnes it to the ﬂuid phase, it is a critical MMP regulator in blood and could be relevant to cerebral amyloid angiopathy or other vascular abnormalities in AD, as well as entering tissues after injury (117, 118). The MMPs that bind α2 M and could indirectly aﬀect handling of Aβ, APOE or tPA by LRP include MMP-1, 2, 3, 7, 8, 9, 11, 12 and 14 (117, 118). Cellular uptake of MMPs by LRP independent of α2 M may also regulate extracellular levels of several MMPs as LRP can mediate cellular internalisation of (i) MMP-13 as a complex with a speciﬁc 170 kDa receptor (123), (ii) a thrombospondin 2-MMP-2 complex (124) and (iii) a complex of

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

288

C.H. Parsons et al

TIMP-1 and MMP-9 (125). The converse may also be true — MMPs may act as LRP ‘convertases’ or ‘sheddases’, regulating the amount of LRP available to internalise extracellular ligands. Quinn and colleagues identiﬁed a soluble form of LRP (sLRP) in human plasma produced by cleavage of the extracellular domain of cell surface LRP by one or more unknown proteases (126, 127). This soluble form retains LRP ligand-binding capabilities and could potentially act as a competitive antagonist against membranebound LRP, reducing eﬀectiveness of membrane-bound LRP as a scavenger receptor. Shedding of sLRP is inhibited by a hydroxamic acid compound that inhibits MMPs (127) and other broad-spectrum MMP inhibitors (DJ unpublished data). In addition, the catalytic domain of MMP-14 (MT1MMP) can cleave LRP, generating a soluble product (128). However, this appears to diﬀer from the cleavage reported by Quinn and colleagues, suggesting more than one MMP may be capable of catalysing LRP ectodomain shedding. The reciprocal regulatory interactions between MMPs and LRP could provide a dynamic system contributing to modulation of matrix modelling in diﬀerent conditions. As discussed further below, excessive MMP activity can contribute to events usually considered harmful, such as BBB breakdown or neuronal death (47). Yet such events might be advantageous in some circumstances. For example, in CNS infection or other advanced disease, facilitating entry of peripheral immune cells into the brain by perturbing the blood-brain barrier might constitute a ‘last-resort’ defence mechanism. It is also feasible that MMP-mediated neurotoxicity might allow targeting and destruction of unhealthy neurons interacting inappropriately with their environment. It will be important, although diﬃcult, to determine if any MMP up-regulation that does occur in AD or other neurodegenerative diseases reﬂects aberrant induction or is instead part of an appropriate defence response that may have beneﬁcial components. If the latter, the possibility must then be considered that MMP up-regulation in AD is suboptimal. For example, it has been proposed that in AD a local lack of urokinase plasminogen activator (uPA) may reduce MMP-9 activation by the plasminogen-activating pathway (59). As MMP-9 can degrade Aβ (58), this could reduce Aβ turnover and accelerate amyloid accumulation. Active MMPs can be rapidly degraded and hard to assess in human studies (129) but canine studies suggest inadequate MMP-9 activation might reduce Aβ clearance (130). Inappropriate changes in TIMPs or other naturally occurring MMP inhibitors in AD might also limit MMP activities.

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

289

9. The Extracellular Matrix and MMPs in Alzheimer’s Disease Accumulation of MMP substrates in Alzheimer’s disease. The MMPs can be viewed as molecular machetes used by cells to control the jungle of molecules comprising the ECM. Reactive glia in or around brain lesions in AD may secrete ECM proteoglycans in response to Aβ exposure (131). Components of the ECM accumulate in or near plaques, neuroﬁbrillary tangles and cerebrovascular amyloid (Table 2) and may block access of enzymes or cells such as microglia to Aβ in lesion cores. These components include laminin, collagen type IV and the sulfated proteoglycans (PGs) such as heparan (HSPG), chondroitin (CSPG), dermatan and keratan sulfate PGs (132–142). This is also consistent with the possibility that there is insuﬃcient net activity of some MMPs in AD since most of these ECM molecules are normally degraded by MMPs. Just one example of an MMP substrate which accumulates in AD, agrin, one of the main brain basement membrane HSPGs, is prominent in AD plaques, tangles and cerebrovascular lesions (142–147). Agrin occurs in brain vessels and synapses, where it is thought to participate in axon pathway formation and synaptogenesis (148–150). Long-term potentiation (LTP), which contributes to memory formation, can be blocked by disrupting HSPG such as agrin that may transmit extracellular signals to intracellular kinase pathways (151, 152). Agrin can be cleared from neuromuscular synapses by MMP-3 activated in response to synaptic activity (153), suggesting active roles for MMPs in normal memory functions as well as in disease. Other MMP substrates in the ECM which may have roles in AD are described in Table 2 below.

10. Relationships with Aβ Species The ECM proteoglycans interact in a variety of ways with Aβ peptides or amyloid deposits, in manners that diﬀer depending on the aggregation state of Aβ (e.g. (154) and Table 2 below). Besides promoting initial Aβ aggregate nucleation (‘seeding’) and ﬁbril deposition (Table 2), PGs may subsequently shield Aβ deposits from direct degradation by matrix proteases and inhibit microglial phagocytosis of Aβ. High levels of Aβ can interfere with PG degradation by intracellular heparanases so binding by Aβ may in turn block PG turnover in a vicious cycle (155). This may also occur for vascular amyloid. In transgenic mice overproducing

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

290

C.H. Parsons et al

Table 2. ECM components proposed to interact with Aβ or tau species. The table shows the main ECM species proposed to interact with pathogenic Aβ or tau species in Alzheimer’s disease or congophilic amyloidogenic angiopathy of the Dutch type (HCHWA-D), together with some of the reported associations, relevant reviews and other references. Again the list is not exhaustive. The ECM species listed include chondroitin (CSPG), dermatan (DSPG), heparan (HSPG) and keratan (KSPG) sulfate proteoglycans (PG) and glycosaminoglycans (GAG). Form of Association

ECM Components

Eﬀects involving tau proteins or NFT

GAGs (236), HSPG (237) incl. agrin (145), DSPG (238)

Increase α-synuclein aggregation in vitro

Speciﬁc GAGs e.g. heparan but not keratan sulfate PGs (165)

Co-localised with neuropil amyloid in AD

DSPG (238), HSPGs with agrin typically greater than glypican and syndecans e.g. (144, 145), laminin (146), glial hyaluronic acid protein complexes related to aggrecan (142, 239). All major HSPG bar perlecan have been reported in diﬀuse cerebellar plaques (147)

Localised with amyloid in Hereditary Congophilic Angiopathy, Dutch Type or sporadic amyloid angiopathy in AD

Collagen types III and IV, laminin and ﬁbronection (139), HSPG (139, 141, 146, 240), including agrin (146) and perlecan (141, 142)

Binding of Aβ peptide and/or amyloid

GAGs and HSPG (154, 240–243) including agrin (244), vascular proteoglycans (240) with vascular perlecan (138, 245, 246) usually greater than biglycan, decorin (138)

Promotion of initial Aβ nucleation and/or amyloid ﬁbre deposition in vitro or in vivo

KSPG (247), CSPG (247), HSPG (137, 155, 247, 248), notably agrin (244), perlecan or related (137, 246) and other sulfated ECM components (249)

Docking of Aβ to CNS neural cells (proposed to be necessary for toxicity)

Glypican (250)

Enhanced CNS cell uptake of Aβ

HSPG (168, 251)

Enhancement of Aβ toxicity

Perlecan (252), glypican-1 (243)

Decreased microglial phagocytosis of Aβ

PG and GAG (253, 254)

Inhibition of Aβ degradation

HSPG & CSPG (255)

Reduction of Aβ toxicity of fragments

HSPG and GAGs (256, 257), laminins (258)

Aβ induction of ECM

CSPG (259)

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

291

TGF-β1 in activated perivascular astrocytes, cerebrovascular amyloidosis and microvascular degeneration are associated with buildup of MMP substrates (e.g. laminin, perlecan), possibly due to MMP suppression by TGF-β1 (156). Neuropathological lesions in patients have the potential to build up PG coatings over far longer times than in mouse models, making amyloid harder to dissociate. Note that conversely, a glycoprotein coating could serve to reduce the toxicity of Aβ species. Integrins are transmembrane glycoproteins involved in cell-ECM binding (157–160). Speciﬁc integrins (e.g. integrin α5 β1 ) bind to nonﬁbrillar Aβ and this can lead to increased Aβ uptake and reduced apoptotic toxicity of Aβ (161). Proteoglycan shielding may also occur with α-synuclein, which aggregates abnormally in Lewy bodies diseases such as PD and DLB and can also occur in association with Aβ deposition in AD (35, 36, 162) and transgenic animal models of AD (163, 164). α-Synuclein binds certain glycosaminoglycans (165) and may build up PG coatings when present in extracellular lesions (Table 2). Speciﬁc glycosaminoglycans (e.g. heparin and heparan sulfate PGs but not keratan sulfate PGs) alter the kinetics of α-synuclein ﬁbril formation in vitro, with heparin increasing the rate of aggregation and the aggregate mass (165). Using MMPs to clear excess PGs in conjunction with other amyloid removal strategies might ultimately have therapeutic beneﬁts although ﬁnesse may be needed, as some PGs and integrins can reduce Aβ toxicity (161, 166) and can also increase cellular uptake of APOE (167), perhaps blocking initial docking of Aβ or APOE on the cell surface. 11. Eﬀects of MMPs on Neuronal Biology There are direct interactions between ECM molecules and lipid-handling molecules, notably including APOE, a major genetic susceptibility factor for AD (discussed further below). The HSPGs also enhance lipoprotein remnant binding to APOE, appear vital for APOE and Aβ uptake by LRP and can also act alone as an APOE receptor (120, 168). As the enzymes primarily responsible for HSPG degradation, MMPs are therefore likely to profoundly inﬂuence all these processes. While few, if any, CNS studies have examined this directly — MMPs participate in lipoprotein-handling systems outside the CNS (169–171). Several MMPs increase in conjunction with both atherosclerosis and aneurysm development in either LDL receptor-deﬁcient or ApoE-deﬁcient mice on high fat diets (170).

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

292

C.H. Parsons et al

In the CNS, analogous confederacies of lipid-handling and ECMremodelling molecules will necessarily be central to neurogenesis and neuritogenesis, axon guidance, synaptic plasticity, LTP, memory and injury repair — any process of reorganisation of the lipid-rich neuronal surface and the matrix structures embedding it. While direct studies are lacking, collateral evidence ties neuronal plasticity and related processes to multiplex interactions of MMPs, ECM components and AD-related species such as APP which promotes neurite outgrowth (172–174) and is up-regulated strongly and consistently in neuronal injury (175, 176). Various MMP substrates in the ECM can promote or inhibit neurite outgrowth and in some systems, neurite penetration into the matrix depends on MMP induction by growth factors (177, 178). As we described in a previous chapter, speciﬁc MMPs are also upregulated in response to nerve injury (in a context-dependent manner). Disrupting the ECM can also aﬀect neurons by disturbing functions such as growth factor sequestration or synaptic support, in addition to direct eﬀects on attachment (179). Both APP and APP-like proteins (e.g. APLP-2) can carry CSPG side-chains that could potentially be cleaved by MMPs and may alter ECM adhesion (180–182). However, these forms are rare (183). There can be substantial cross-play between MMP substrates. For example, neurite promotion by laminin is inhibited by CSPGs elevated in endoneurial basal laminae after peripheral nerve injury (184). Laminin and other ECM proteins can aﬀect neuronal viability through survival signals mediated by integrins and integrin signal transduction may be modulated by both MMPs and APP, which binds laminin strongly and co-localises with integrins (179, 185–187). 12. The Blood-Brain Barrier and MMPs in Alzheimer’s Disease Few studies have looked at how MMPs aﬀect the BBB in AD. This is surprising, as in conditions such as MS, meningitis, HIV dementia and reperfusion injury after strokes, it is well-recognised that the capillary basal lamina of the BBB can be degraded by MMPs, facilitating peripheral immune cell entry into the neuropil (188). While the BBB usually remains relatively intact in AD, this depends partly on the patient’s age and whether the barrier features in question involve molecular or cellular passage. With increasing age, BBB dysfunction becomes more common in AD and vascular dementia, as indicated by endothelial cell organelle

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

293

alterations, collagen accumulation in vascular basement membranes (189) and extravasation of serum protein markers (190, 191), with increases in the CSF/serum albumin ratio (192, 193). The disruption increases in parallel with AD severity (193) and cerebrovasculopathy. Approximately a third or more of AD patients have co-existing vasculopathy and associated BBB disruption (194, 195). Inﬁltration of peripheral immune cells can occur in AD (191), although typically to a lesser degree than in the diseases above. While the evidence is primarily indirect, increases in MMPs triggered by Aβ probably contribute to BBB disruption. Interaction of Aβ1–40 with the basolateral surface of brain endothelial cells is proposed to initiate signalling leading to traﬃcking of peripheral blood monocytes from the apical to the basolateral side of the BBB (196). In murine cerebral endothelial cells, Aβ1–40 induces increased expression, release and activation of MMP-9 (197). Abnormal fragmentation of agrin and laminin in the AD cerebrovascular basement membrane, suggestive of aberrant MMP activity, is associated with increased BBB leakage, as gauged by prothrombin immunoreactivity (143, 146). Furthermore, aged APPsw transgenic mice expressing human APP with an amyloidproducing mutation display MMP-9 immunoreactivity in association with microhemorrhage in vessels with cerebral amyloid angiopathy (197). In sum, MMPs are likely to inﬂuence BBB integrity in AD and this in turn could inﬂuence Aβ passage between the neuropil and the periphery. 13. Genetic Associations between MMPs and Related Proteins and Alzheimer’s Disease Approximately 10% of AD patients have relatively early disease onset (< 65 years of age) and often show strong familial inheritance patterns. At least half of these have aggressive, highly penetrant mutations in the presenilin 1 gene (198). In contrast, late onset patients (≥ 65 years) often do not have clear-cut inheritance patterns, instead appearing to occur ‘sporadically’. Nonetheless, late onset, ‘sporadic’ AD is also inﬂuenced by genetic factors, although with lower penetrance. Notably the apolipoprotein E gene (APOE) ε4 allele is a strong and well-established risk factor accounting for as much as 50% or more of the total genetic risk of sporadic AD (199–201). Genetic polymorphisms inﬂuencing MMP expression have been tested for eﬀects on sporadic AD. Because the strong APOE eﬀect can override weaker associations it is important to take it into account such studies. The MMP-9 C-1562T promoter polymorphism aﬀects MMP-9 gene

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

294

C.H. Parsons et al

transcription and high MMP-9 expressors (TT or CT) lacking an APOE ε4 allele showed a reduced risk of dementia (odds ratio 0.5, 95% conﬁdence interval 0.3–0.9) in a French Caucasian population (202). This is again consistent with the possibility that there could be too little MMP-9 activity in AD. There does not appear to be any relationship between plasma MMP-9 levels and APOE genotype (51). It will be important to extend these data into other population samples. A full genome scan for late onset AD reported linkage disequilibrium between AD and a region around the TIMP-1 gene (203). Wollmer and colleagues (99) studied two single nucleotide polymorphisms (SNPs) in this region. One was the TIMP-1 C124T exonic SNP. The second was ∼22 kb upstream of the TIMP-1 gene initiation site, within intron 1 of the synapsin SYN1 gene. The TIMP-1 gene is contained within intron 6 of SYN1 (204) and is probably transcribed in the opposite direction (99). Neither SNP altered the risk of AD in two independent populations of diﬀerent ethnicity or inﬂuenced TIMP-1 levels in AD CSF compared to CSF in other neurological disorders or healthy controls (99). In addition, the intron 6 BamH1 polymorphism in the HSPG2 (perlecan) gene was reported to increase the risk of AD and increase tau paired helical ﬁlament formation in Finnish APOE ε4 carriers (205). However, the same polymorphism failed to show association with AD irrespective of APOE genotype in an Israeli Jewish population sample (206). Neither study examined the polymorphism in conjunction with any MMP SNPs. In general, in view of the large potential for interactive or compensatory eﬀects and redundancies among diﬀerent ECM proteoglycans, MMPs and related molecules, multifactorial analyses which also take into account interactive eﬀects are probably more likely to provide insights than single SNP studies. 14. Other Neurodegenerative Diseases Alterations in MMPs and TIMPs have been observed in other neurodegenerative diseases, including Parkinson’s disease (PD), progressive supranuclear palsy (PSP) and amyelotrophic lateral sclerosis (ALS). While studies to date have been observational rather than mechanistic in nature, the proﬁles of MMP alterations in particular regions diﬀer between diseases, suggesting MMP alterations do not simply reﬂect generalised, non-speciﬁc responses to neurodegeneration. The gelatinases MMP-2 and MMP-9 have been studied in various neurodegenerative diseases. As described above, the pattern seen in AD tends to be one of relatively strong increases in MMP-9 in regions most aﬀected by the disease with relatively modest and restricted

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

295

changes in MMP-2. This bears resemblances to patterns seen in PSP and ALS but appears to diﬀer from PD and the spinocerebellar ataxia SCA3, as described below. 15. Progressive Supranuclear Palsy In PSP, there is neurodegeneration and gliosis in the basal ganglia, forebrain and brainstem. Pathology is characterised by tau NFT resembling those in AD as well as straight neuroﬁbrillary ﬁlaments. The characteristic clinical feature is supranuclear gaze palsy, together with dementia, parkinsonism and other movement, postural and neurological abnormalities. In PSP patients, MMP-1 has been reported to be elevated in the substantia nigra, with increased MMP-9 in the substantia nigra and frontal cortex (207). There was no evidence for altered MMP-2 levels in any of the brain regions studied but signiﬁcant increases in both TIMP-1 and TIMP-2 in the substantia nigra (207). 16. Parkinson’s Disease In contrast, in PD brains, the same group reported decreased MMP-2 levels in the substantia nigra, which is vulnerable to cell damage in PD, but no change in the cortex or hippocampus (208). Levels of MMP-1 and MMP-9 were not signiﬁcantly diﬀerent to those in age-matched control subjects. As MMP-9 appeared localised to neurons, it was suggested that failure to observe up-regulation of MMP-9 in PD might be due to neuronal loss, since samples obtained post-mortem typically represent end-stage disease. (However MMP-9 is up-regulated in post-mortem AD samples). It was further proposed that MMP-9 levels might be elevated at earlier stages of PD. However, the same group’s later studies of CSF obtained antemortem from PD patients failed to detect appreciable quantities of MMP-9 monomer, although a 130 kDa band with gelatinase activity was observed by zymography and TIMP-1 levels were elevated (61). The 130 kDa band may be an MMP-9/TIMP-1 heterodimer (209) or a covalent heterodimer of MMP-9 with a lipocalin (210). Lipocalins are small, extracellular proteins with roles that include protein transport (211) and guidance of pioneer neurons during neurogenesis (212). 17. Amyotrophic Lateral Sclerosis In ALS, upper and lower motor neurons are aﬀected, with degeneration of cortical Betz cells, corticospinal tracts and anterior horn cells in the

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

296

C.H. Parsons et al

spinal cord and weakness in limb movements, muscle ﬁbre denervation and atrophy. Changes in MMPs relative to controls have been observed in post-mortem tissue from ‘sporadic’ ALS patients (213) and in serum but not CSF from ALS patients (214). Elevated MMP-9 and the 130 kDa gelatinase complex above have been detected by immunohistochemistry and zymography in CNS areas most aﬀected in ALS, with MMP-9 present in the cytoplasm, dendrites and axons of upper and lower motor neurons and increased MMP-9 activity by zymography in frontal and occipital cortex and throughout the spinal cord (213). Serum MMP-9 in ALS is comparable to levels in viral meningoencephalitis or bacterial meningitis. However, this MMP-9 is thought to originate systemically, from degenerating muscle or peripheral nerves in ALS (214). The situation for MMP-2 in ALS is less clear, with immunoreactivity largely restricted to astrocytes and average activity reduced in ALS motor cortex but not other regions compared to controls (213). A few specimens showed high levels of MMP-2 in thoracic and lumbar cord. The authors did not identify reasons for the variability, which did not correlate with gliosis. However, the patient group (n = 9) ranged in age from 47 to 94 years (213). While cytokines may cause short-term up-regulation of glial MMP-2, in chronic disease with longstanding inﬂammation there may be eventual reductions in astrocyte MMP-2 (215). This could explain the reduced MMP-2 in the substantia nigra and motor cortex in endstage PD and in the motor cortex in endstage ALS. The consequences of these reductions in MMP-2 are unknown. 18. Trinucleotide Repeat Disorders In contrast, up-regulation of MMP-2 mRNA and increased MMP-2 immunoreactivity in pontine neurons containing nuclear inclusions has been observed in the neurodegenerative polyglutamine repeat disorder, spinocerebellar ataxia type 3 (SCA3; Machado-Joseph disease), in conjunction with Aβ immunoreactivity in the pons and general increases in APP and pro- and anti-inﬂammatory cytokines (216). There was no evidence for changes in CSF MMP-2 or -9 in Huntington’s disease (HD), the archetypal polyglutamine repeat disorder (61). However, as noted above, CSF and brain changes do not always correlate well. Relative to control levels, increases in TIMP-1 have been reported in CSF from PD, ALS and HD patients whereas increases in TIMP-2 were found in HD and AD CSF but not in CSF from PD or ALS patients (61).

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

297

More studies are again required to determine whether or not similar alterations are seen in the neuropil in these diseases and whether there are diﬀerences in activities of enzymes such as ADAM-10 and the membrane type MMPs (MT-MMPs), that are diﬀerently regulated by TIMP-1 and TIMP-2, as reviewed elsewhere (217). Additional evidence for involvement of the MMPs and TIMPs in neurodegenerative disease comes from wobbler (WR) mice. These mice undergo progressive neurodegeneration of large motoneurons and other neurons, causing a muscular atrophy resembling ALS. There is reactive astrocytosis and microglial activation, increases in pro-inﬂammatory cytokines (e.g. interleukin-1β, TNF-α) and increased expression of MT1-MMP and TIMP-1 and -3 transcripts but no changes in TIMP-2 or -4 transcripts (218). Together, all these studies suggest that changes in MMPs and probably also TIMPs in neurodegenerative conditions are not purely part of a generalised, inﬂammatory response but could have characteristic, disease-speciﬁc ramiﬁcations. 19. Therapeutic Implications More research is required to determine whether there may be potential beneﬁts to be gained from inhibiting MMPs in AD and related disorders or whether this might be deleterious in some circumstances. If activity of key metalloproteinases is insuﬃcient in some neurodegenerative disorders, controlled up-regulation of MMPs or ADAMs may instead be called for, as previously suggested for ADAM-10 (95). Numerous epidemiological studies show nonsteroidal anti-inﬂammatory drugs (NSAIDs) can protect against AD. However, protection may only occur if used some time (e.g. 2 years or more) before symptom onset so it has been proposed that the mechanisms involved may be distinct from the prototypal anti-inﬂammatory actions of these drugs (219–222). Antiinﬂammatory drugs are known to be able to down-regulate MMPs (223). It remains to be seen whether regulation of MMP expression contributes to the protective eﬀects of NSAIDs and whether this alters at diﬀerent disease stages. It could be interesting to investigate MMP expression in the CNS of brains from patients with histories of using these drugs. Another AD therapy currently under investigation is metal chelation, in particular zinc chelation (224, 225). Chelation has strong potential to perturb MMPs. While chelators do not usually strip ions from proteins except at high concentrations, they bind ions within metalloproteins to

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

298

C.H. Parsons et al

form inactive complexes, and sop up or redistribute free ions, aﬀecting expression of metal-containing proteins (226, 227). To what extent the zinc-dependent activities of the many MMPs, ADAMs and other metalloenzymes are impaired by zinc re-distributing chelators is unknown but as these enzymes have so many ﬁnely modulated functions in the CNS, indiscriminate interference with their activity is potentially extremely dangerous. Agents perturbing biodistribution of one metal ion species often also aﬀect others since ion transport and cellular uptake systems are frequently not monospeciﬁc.

20. Conclusions Synaptic and neuronal loss in neurodegenerative diseases is likely to be accompanied by ongoing restructuring of the neuropil, including clearance of damaged tissue, destruction of dysfunctional neurons or opening of the blood-brain barrier, allowing entry of peripheral immune cells. The MMPs can do great damage if dysregulated but also have the potential to contribute to neuroprotection or regeneration. There is little available information on the complex MMP regulatory mechanisms operating in the nervous system or on what mechanisms dominate under particular circumstances. The orchestrated proﬁles of MMP and TIMP activities and the nature and consequences of their actions are likely to vary in diﬀerent disease stages and between diﬀerent brain regions. Treatment strategies based on na¨ıve or overly simplistic reasoning are likely to be at best ineﬀective and a waste of time, money and resources for patients, clinicians and researchers or, at worst, dangerous. Much more research is needed in this area to enable eﬀective and appropriately targeted treatments to be developed.

References 1. 2. 3. 4. 5. 6. 7. 8. 9.

Coleman, P. D. and Yao, P. J. (2003) Neurobiol Aging 24(8), 1023–7 Mattson, M. P. (2004) Nature 430(7000), 631–9 Ross, C. A. and Poirier, M. A. (2004) Nat Med 10(Suppl), S10–7 Bossy-Wetzel, E., Schwarzenbacher, R., and Lipton, S. A. (2004) Nat Med 10(Suppl), S2–9 Samii, A., Nutt, J. G., and Ransom, B. R. (2004) The Lancet 363, 1783–93 Dauer, M. and Prezedborski, S. (2003) Neuron 39, 889–909 Cummings, J. L. (1995) Brain & Cognition 28(3), 266–80 Emre, M. (2003) Lancet Neurol 2(4), 229–37 Emre, M. (2003) Mov Disord 18(Suppl 6), S63–71

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

22. 23. 24. 25.

26. 27. 28. 29. 30.

31.

299

Markesbery, W. R. and Carney, J. M. (1999) Brain Pathol 9(1), 133–46 Gonzalez-Scarano, F. and Baltuch, G. (1999) Annu Rev Neurosci 22, 219–40 Allan, S. M. (2000) Ann NY Acad Sci 917, 84–93 Halliday, G., Robinson, S. R., Shepherd, C., and Kril, J. (2000) Clin Exp Pharmacol Physiol 27(1–2), 1–8 Andersen, J. K. (2004) Nat Med 10(Suppl), S18–25 Leavitt, B. R., Wellington, C. L., and Hayden, M. R. (1999) Semin Neurol 19(4), 385–95 Tatton, W. G., Chalmers-Redman, R., Brown, D., and Tatton, N. (2003) Ann Neurol 53(Suppl 3), S61–70; discussion S70–2 Przedborski, S. (2004) Neurologist 10(1), 1–7 Cribbs, D. H., Poon, W. W., Rissman, R. A., and Blurton-Jones, M. (2004) Am J Pathol 165(2), 353–5 McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., and Stadlan, E. M. (1984) Neurology 34(7), 939–44 Association, A. P. (1994) Diagnostic and Statistical Manual of Mental Disorders, 4th Ed., Washington, D.C. Roman, G. C., Tatemichi, T. K., Erkinjuntti, T., Cummings, J. L., Masdeu, J. C., Garcia, J. H., Amaducci, L., Orgogozo, J. M., Brun, A., Hofman, A., Moody, D. M., O’Brien, M. D., Yamaguchi, T., Grafman, J., Drayer, B. P., Bennett, D. A., Fisher, M., Ogata, J., Kokmen, E., Bermejo, F., Wolf, P. A., Gorelick, P. N., Bick, K. L., Pajeau, A. K., Bell, M. A., DeCarli, C., Culebras, A., Korczyn, A. D., Bogousslavsky, J., Hartmann, A., and Scheinberg, P. (1993) Neurology 43(2), 250–60 Birkenhager, W. H., Forette, F., Seux, M. L., Wang, J. G., and Staessen, J. A. (2001) Arch Intern Med 161(2), 152–6 Boustani, M., Peterson, B., Hanson, L., Harris, R., and Lohr, K. N. (2003) Annals of Internal Medicine 138(11), 927–37 Henderson, S. (1998) Ann Med Interne (Paris) 149(4), 181–6 Ott, A., Breteler, M. M., van Harskamp, F., Claus, J. J., van der Cammen, T. J., Grobbee, D. E., and Hofman, A. (1995) Bmj 310(6985), 970–3 McPherson, S. C., JL. (1996) Brain and Cognition 31, 269–82 Chui, H. (2001) Clin Cornerstone 3, 40–51 Kalaria, R. (2002) J Neurol Sci 203–204, 29–34 F¨ orstl, H. (1999) European Archives of Psychiatry & Clinical Neuroscience 249(9), pS64, 1p McKeith, I., Galasko, D., Kosaka, K., Perry, E., Dickson, D., Hansen, L., Salmon, D., Lowe, J., Mirra, S., Byrne, E., Lennox, G., Quinn, N., Edwardson, J., Ince, P., Bergeron, C., Burns, A., Miller, B., Lovestone, S., Jansen, E., Ballard, C., Vos, R. D., Wilcock, G., Jellinger, K., and Perry, R. (1996) Neurology 47, 1113–1124 McKeith, I., Mintzer, J., Aarsland, D., Burn, D., Chiu, H., Cohen-Mansﬁeld, J., Dickson, D., Dubois, B., Duda, J. E., Feldman, H., Gauthier, S., Halliday, G., Lawlor, B., Lippa, C., Lopez, O. L., Carlos Machado, J., O’Brien, J., Playfer, J., and Reid, W. (2004) Lancet Neurol 3(1), 19–28

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

300

C.H. Parsons et al

32. Verghese, J., Lipton, R. B., Hall, C. B., Kuslansky, G., Katz, M. J., and Buschke, H. (2002) New England Journal of Medicine 347(22), 1761–1768 33. Michotte, A. (2003) Acta Neurol Belg 103(3), 155–8 34. Huang, Y., Cheung, L., Rowe, D., and Halliday, G. (2004) Brain Res Brain Res Rev 46(1), 44–70 35. Mukaetova-Ladinska, E. B., Hurt, J., Jakes, R., Xuereb, J., Honer, W. G., and Wischik, C. M. (2000) J Neuropathol Exp Neurol 59(5), 408–17 36. Takeda, A., Hashimoto, M., Mallory, M., Sundsumo, M., Hansen, L., and Masliah, E. (2000) Acta Neuropathol (Berl) 99(3), 296–304 37. Wirths, O. and Bayer, T. A. (2003) Prog Neuropsychopharmacol Biol Psychiatry 27(1), 103–8 38. Fu, C., Chute, D. J., Farag, E. S., Garakian, J., Cummings, J. L., and Vinters, H. V. (2004) Archives of Pathology and Laboratory Medicine 128(1), 32–8 39. Graham, C., Ballard, C., and Saad, K. (1997) Int J Geriatr Psychiatry 12, 314–8 40. Luis, C. A., Barker, W. W., Gajaraj, K., Harwood, D., Petersen, R., Kashuba, A., Waters, C., Jimison, P., Pearl, G., Petito, C., Dickson, D., and Duara, R. (1999) Int J Geriatr Psychiatry 14(7), 526–33 41. Bailey, A. J., Sims, T. J., Avery, N. C., and Halligan, E. P. (1995) Biochem J 305 (Pt 2), 385–90 42. Bruno, G., Todor, R., Lewis, I., and Chyatte, D. (1998) J Neurosurg 89(3), 431–40 43. Robert, V., Besse, S., Sabri, A., Silvestre, J. S., Assayag, P., Nguyen, V. T., Swynghedauw, B., and Delcayre, C. (1997) Lab Invest 76(5), 729–38 44. Zhang, J. W., Deb, S., and Gottschall, P. E. (2000) Neurosci Lett 295(1–2), 9–12 45. Ringheim, G. E. and Conant, K. (2004) J Neuroimmunol 147(1–2), 43–9 46. Leake, A., Morris, C. M., and Whateley, J. (2000) Neurosci Lett 291(3), 201–3 47. Vos, C. M., Sjulson, L., Nath, A., McArthur, J. C., Pardo, C. A., Rothstein, J., and Conant, K. (2000) Exp Neurol 163(2), 324–30 48. Clark, A. W., Krekoski, C. A., Bou, S. S., Chapman, K. R., and Edwards, D. R. (1997) Neurosci Lett 238, 53–6 49. Rosenberg, G. A., Sullivan, N., and Esiri, M. M. (2001) Stroke 32(5), 1162–8 50. Backstrom, J. R., Miller, C. A., and Tokes, Z. A. (1992) J Neurochem 58(3), 983–92 51. Lorenzl, S., Albers, D. S., Relkin, N., Ngyuen, T., Hilgenberg, S. L., Chirichigno, J., Cudkowicz, M. E., and Beal, M. F. (2003) Neurochemistry International 43(3), 191–196 52. Yamada, T., Miyazaki, K., Koshikawa, N., Takahashi, M., Akatsu, H., and Yamamoto, T. (1995) Acta Neuropathol (Berl) 89(3), 199–203 53. Yoshiyama, Y., Sato, H., Seiki, M., Shinagawa, A., Takahashi, M., and Yamada, T. (1998) Acta Neuropathol (Berl) 96(4), 347–50 54. Jaworski, D. M. (2000) Brain Research 860(1–2), 174–7

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

301

55. Hayashita-Kinoh, H., Kinoh, H., Okada, A., Komori, K., Itoh, Y., Chiba, T., Kajita, M., Yana, I., and Seiki, M. (2001) Cell Growth & Diﬀerentiation 12(11), 573–80 56. Sekine-Aizawa, Y., Hama, E., Watanabe, K., Tsubuki, S., Kanai-Azuma, M., Kanai, Y., Arai, H., Aizawa, H., Iwata, N., and Saido, T. C. (2001) Eur J Neurosci 13(5), 935–48 57. Yoshiyama, Y., Asahina, M., and Hattori, T. (2000) Acta Neuropathol (Berl) 99(2), 91–5 58. Backstrom, J. R., Lim, G. P., Cullen, M. J., and Tokes, Z. A. (1996) J Neurosci 16(24), 7910–9 59. Asahina, M., Yoshiyama, Y., and Hattori, T. (2001) Clin Neuropathol 20(2), 60–3 60. Adair, J. C., Charlie, J., Dencoﬀ, J. E., Kaye, J. A., Quinn, J. F., Camicioli, R. M., Stetler-Stevenson, W. G., and Rosenberg, G. A. (2004) Stroke 61. Lorenzl, S., Albers, D. S., LeWitt, P. A., Chirichigno, J. W., Hilgenberg, S. L., Cudkowicz, M. E., and Beal, M. F. (2003) J Neurol Sci 207(1–2), 71–6 62. Howlett, D. R., Simmons, D. L., Dingwall, C., and Christie, G. (2000) Trends Neurosci 23(11), 565–70 63. Nunan, J. and Small, D. H. (2000) FEBS Lett 483(1), 6–10 64. Vassar, R. and Citron, M. (2000) Neuron 27(3), 419–22 65. Walter, J., Kaether, C., Steiner, H., and Haass, C. (2001) Curr Opin Neurobiol 11(5), 585–90 66. Wolfe, M. S. and Haass, C. (2001) J Biol Chem 276(8), 5413–6 67. Sisodia, S. S. and St George-Hyslop, P. H. (2002) Nat Rev Neurosci 3(4), 281–90 68. Wilson, C. A., Doms, R. W., and Lee, V. M. (2003) J Neurosci Res 74(3), 361–9 69. Sinha, S., Anderson, J. P., Barbour, R., Basi, G. S., Caccavello, R., Davis, D., Doan, M., Dovey, H. F., Frigon, N., Hong, J., Jacobson-Croak, K., Jewett, N., Keim, P., Knops, J., Lieberburg, I., Power, M., Tan, H., Tatsuno, G., Tung, J., Schenk, D., Seubert, P., Suomensaari, S. M., Wang, S., Walker, D., John, V., and et al (1999) Nature 402(6761), 537–40 70. Vassar, R., Bennett, B. D., Babu-Khan, S., Kahn, S., Mendiaz, E. A., Denis, P., Teplow, D. B., Ross, S., Amarante, P., Loeloﬀ, R., Luo, Y., Fisher, S., Fuller, J., Edenson, S., Lile, J., Jarosinski, M. A., Biere, A. L., Curran, E., Burgess, T., Louis, J. C., Collins, F., Treanor, J., Rogers, G., and Citron, M. (1999) Science 286(5440), 735–41 71. Yan, R., Bienkowski, M. J., Shuck, M. E., Miao, H., Tory, M. C., Pauley, A. M., Brashier, J. R., Stratman, N. C., Mathews, W. R., Buhl, A. E., Carter, D. B., Tomasselli, A. G., Parodi, L. A., Heinrikson, R. L., and Gurney, M. E. (1999) Nature 402(6761), 533–7 72. Haniu, M., Denis, P., Young, Y., Mendiaz, E. A., Fuller, J., Hui, J. O., Bennett, B. D., Kahn, S., Ross, S., Burgess, T., Katta, V., Rogers, G., Vassar, R., and Citron, M. (2000) J Biol Chem 275(28), 21099–106

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

302

C.H. Parsons et al

73. Huse, J. T., Pijak, D. S., Leslie, G. J., Lee, V. M., and Doms, R. W. (2000) J Biol Chem 275(43), 33729–37 74. Cai, H., Wang, Y., McCarthy, D., Wen, H., Borchelt, D. R., Price, D. L., and Wong, P. C. (2001) Nat Neurosci 4(3), 233–4 75. Luo, Y., Bolon, B., Kahn, S., Bennett, B. D., Babu-Khan, S., Denis, P., Fan, W., Kha, H., Zhang, J., Gong, Y., Martin, L., Louis, J. C., Yan, Q., Richards, W. G., Citron, M., and Vassar, R. (2001) Nat Neurosci 4(3), 231–2 76. Haass, C. and Steiner, H. (2002) Trends Cell Biol 12(12), 556–62 77. De Strooper, B. (2003) Neuron 38(1), 9–12 78. Miyazaki, K., Hasegawa, M., Funahashi, K., and Umeda, M. (1993) Nature 362(6423), 839–41 79. Evin, G., Cappai, R., Li, Q. X., Culvenor, J. G., Small, D. H., Beyreuther, K., and Masters, C. L. (1995) Biochemistry 34(43), 14185–92 80. LePage, R. N., Fosang, A. J., Fuller, S. J., Murphy, G., Evin, G., Beyreuther, K., Masters, C. L., and Small, D. H. (1995) FEBS Lett 377(2), 267–70 81. Itoh, T., Ikeda, T., Gomi, H., Nakao, S., Suzuki, T., and Itohara, S. (1997) J Biol Chem 272(36), 22389–92 82. Leissring, M. A., Farris, W., Chang, A. Y., Walsh, D. M., Wu, X., Sun, X., Frosch, M. P., and Selkoe, D. J. (2003) Neuron 40(6), 1087–93 83. Roher, A. E., Kasunic, T. C., Woods, A. S., Cotter, R. J., Ball, M. J., and Fridman, R. (1994) Biochem Biophys Res Commun 205(3), 1755–61 84. Allinson, T. M., Parkin, E. T., Turner, A. J., and Hooper, N. M. (2003) J Neurosci Res 74(3), 342–52 85. Caporaso, G. L., Gandy, S. E., Buxbaum, J. D., Ramabhadran, T. V., and Greengard, P. (1992) Proc Natl Acad Sci USA 89(7), 3055–9 86. Buxbaum, J. D., Koo, E. H., and Greengard, P. (1993) Proc Natl Acad Sci USA 90, 9195–9198 87. Gandy, S. E., Caporaso, G. L., Buxbaum, J. D., de Cruz Silva, O., Iverfeldt, K., Nordstedt, C., Suzuki, T., Czernik, A. J., Nairn, A. C., and Greengard, P. (1993) Ann NY Acad Sci 695, 117–21 88. Gabuzda, D., Busciglio, J., and Yankner, B. (1993) J Neurochem 61, 2326–2329 89. Hung, A. Y., Haass, C., Nitsch, R. M., Qiu, W. Q., Citron, M., Wurtman, R. J., Growdon, J. H., and Selkoe, D. J. (1993) J Biol Chem 268, 22959–22962 90. Jacobsen, J. S., Spruyt, M. A., Brown, A. M., Sahasrabudhe, S. R., Blume, A. J., Vitek, M. P., Muenkel, H. A., and Sonnenberg-Reines, J. (1994) J Biol Chem 269, 8376–8382 91. Lammich, S., Kojro, E., Postina, R., Gilbert, S., Pfeiﬀer, R., Jasionowski, M., Haass, C., and Fahrenholz, F. (1999) Proc Natl Acad Sci USA 96(7), 3922–7 92. Skovronsky, D. M., Moore, D. B., Milla, M. E., Doms, R. W., and Lee, V. M. (2000) J Biol Chem 275(4), 2568–75 93. Kojro, E., Gimpl, G., Lammich, S., Marz, W., and Fahrenholz, F. (2001) Proc Natl Acad Sci USA 98(10), 5815–20

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

303

94. Asai, M., Hattori, C., Szabo, B., Sasagawa, N., Maruyama, K., Tanuma, S., and Ishiura, S. (2003) Biochem Biophys Res Commun 301, 231–235 95. Postina, R., Schroeder, A., Dewachter, I., Bohl, J., Schmitt, U., Kojro, E., Prinzen, C., Endres, K., Hiemke, C., Blessing, M., Flamez, P., Dequenne, A., Godaux, E., van Leuven, F., and Fahrenholz, F. (2004) J Clin Invest 113, 1456–1464 96. Bernstein, H. G., Bukowska, A., Krell, D., Bogerts, B., Ansorge, S., and Lendeckel, U. (2003) J Neurocytol 32(2), 153–60 97. Amour, A., Knight, C. G., Webster, A., Slocombe, P. M., Stephens, P. E., Knauper, V., Docherty, A. J., and Murphy, G. (2000) FEBS Lett 473(3), 275–9 98. Peress, N., Perillo, E., and Zucker, S. (1995) J Neuropathol Exp Neurol 54(1), 16–22 99. Wollmer, M. A., Papassotiropoulos, A., Streﬀer, J. R., Grimaldi, L. M. E., Kapaki, E., Salani, G., Paraskevas, G. P., Maddelena, A., de Quervain, D., Bieber, C., Umbricht, D., Lemke, U., Bosshardt, S., Degonda, N., Henke, K., Hegi, T., Jung, H. H., Pasch, T., Hock, C., and Nitsch, R. M. (2002) Psychiatric Genetics 12(3), 155–160 100. Dickson, D. W., Lee, S. C., Mattiace, L. A., Yen, S. H., and Brosnan, C. (1993) Glia 7(1), 75–83 101. Johnstone, M., Gearing, A. J., and Miller, K. M. (1999) J Neuroimmunol 93(1–2), 182–93 102. Meda, L., Baron, P., Prat, E., Scarpini, E., Scarlato, G., Cassatella, M. A., and Rossi, F. (1999) J Neuroimmunol 93(1–2), 45–52 103. Deb, S. and Gottschall, P. E. (1996) J Neurochem 66(4), 1641–7 104. Estus, S., Tucker, H. M., van Rooyen, C., Wright, S., Brigham, E. F., Wogulis, M., and Rydel, R. E. (1997) J Neurosci 17(20), 7736–45 105. Tucker, H. M., Kihiko, M., Caldwell, J. N., Wright, S., Kawarabayashi, T., Price, D., Walker, D., Scheﬀ, S., McGillis, J. P., Rydel, R. E., and Estus, S. (2000) J Neurosci 20(11), 3937–46 106. Deb, S., Zhang, J. W., and Gottschall, P. E. (1999) J Neurosci Res 55(1), 44–53 107. Chong, Y. H., Sung, J. H., Shin, S. A., Chung, J. H., and Suh, Y. H. (2001) J Biol Chem 276(26), 23511–7 108. Deb, S., Wenjun Zhang, J., and Gottschall, P. E. (2003) Brain Res 970(1–2), 205–13 109. Jung, S. S., Zhang, W. B., and Van Nostrand, W. E. (2003) J Neurochem 85(5), 1208–15 110. West, M. D., Pereira-Smith, O. M., and Smith, J. R. (1989) Exp Cell Res 184(1), 138–47 111. Zeng, G. and Millis, A. J. (1996) Exp Cell Res 222(1), 150–6 112. Jarisch, A., Krieg, T., and Hunzelmann, N. (1996) Acta Derm Venereol 76(4), 287–90 113. Rao, C. N., Mohanam, S., Puppala, A., and Rao, J. S. (1999) Biochem Biophys Res Commun 255(1), 94–8 114. Higashi, S. and Miyazaki, K. (2003) J Biol Chem 278(16), 14020–8

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

304

C.H. Parsons et al

115. Higashi, S. and Miyazaki, K. (2003) Biochemistry 42(21), 6514–26 116. Imafuku, I., Masaki, T., Waragai, M., Takeuchi, S., Kawabata, M., Hirai, S., Ohno, S., Nee, L. E., Lippa, C. F., Kanazawa, I., Imagawa, M., and Okazawa, H. (1999) J Cell Biol 147(1), 121–34 117. Vincenti, M. P., White, L. A., Schroen, D. J., Benbow, U., and Brinckerhoﬀ, C. E. (1996) Crit Rev Eukaryot Gene Expr 6(4), 391–411 118. Woessner, J. F. and Nagase, H. (2000) Matrix Metalloproteinases and TIMPs, Oxford University Press, New York 119. Herz, J., Hamann, U., Rogne, S., Myklebost, O., Gausepohl, H., and Stanley, K. K. (1988) EMBO J 7(13), 4119–27 120. Mahley, R. W. and Ji, Z. S. (1999) J Lipid Res 40(1), 1–16 121. Hyman, B. T., Strickland, D., and Rebeck, G. W. (2000) Arch Neurol 57(5), 646–50 122. LaMarre, J., Wollenberg, G. K., Gonias, S. L., and Hayes, M. A. (1991) Lab Invest 65(1), 3–14 123. Barmina, O. Y., Walling, H. W., Fiacco, G. J., Freije, J. M., Lopez-Otin, C., Jeﬀrey, J. J., and Partridge, N. C. (1999) J Biol Chem 274(42), 30087–93 124. Yang, Z., Strickland, D. K., and Bornstein, P. (2001) J Biol Chem 276(11), 8403–8. Epub 2000 Dec 11 125. Hahn-Dantona, E., Ruiz, J. F., Bornstein, P., and Strickland, D. K. (2001) J Biol Chem 276(18), 15498–503. Epub 2001 Feb 2 126. Quinn, K. A., Grimsley, P. G., Dai, Y. P., Tapner, M., Chesterman, C. N., and Owensby, D. A. (1997) J Biol Chem 272(38), 23946–51 127. Quinn, K. A., Pye, V. J., Dai, Y. P., Chesterman, C. N., and Owensby, D. A. (1999) Exp Cell Res 251(2), 433–41 128. Rozanov, D. V., Hahn-Dantona, E., Strickland, D. K., and Strongin, A. Y. (2004) J Biol Chem 279(6), 4260–8. Epub 2003 Nov 25 129. Yu, Q. and Stamenkovic, I. (1999) Genes Dev 13(1), 35–48 130. Lim, G. P., Russell, M. J., Cullen, M. J., and Tokes, Z. A. (1997) J Neurochem 68(4), 1606–11 131. Canning, D. R., Hoke, A., Malemud, C. J., and Silver, J. (1996) Int J Dev Neurosci 14(3), 153–75 132. Snow, A. D., Mar, H., Nochlin, D., Sekiguchi, R. T., Kimata, K., Koike, Y., and Wight, T. N. (1990) Am J Pathol 137(5), 1253–70 133. Perlmutter, L. S., Barron, E., Saperia, D., and Chui, H. C. (1991) J Neurosci Res 30(4), 673–81 134. Su, J. H., Cummings, B. J., and Cotman, C. W. (1992) Neuroscience 51(4), 801–13 135. DeWitt, D. A., Silver, J., Canning, D. R., and Perry, G. (1993) Exp Neurol 121(2), 149–52 136. Eikelenboom, P., Zhan, S. S., Kamphorst, W., van der Valk, P., and Rozemuller, J. M. (1994) Virchows Arch 424(4), 421–7 137. Snow, A. D., Sekiguchi, R. T., Nochlin, D., Kalaria, R. N., and Kimata, K. (1994) Am J Pathol 144(2), 337–47 138. Snow, A. D., Kinsella, M. G., Parks, E., Sekiguchi, R. T., Miller, J. D., Kimata, K., and Wight, T. N. (1995) Arch Biochem Biophys 320(1), 84–95

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

305

139. van Duinen, S. G., Maat-Schieman, M. L., Bruijn, J. A., Haan, J., and Roos, R. A. (1995) Lab Invest 73(2), 183–9 140. Small, D. H., Williamson, T., Reed, G., Clarris, H., Beyreuther, K., Masters, C. L., and Nurcombe, V. (1996) Ann NY Acad Sci 777, 316–21 141. Verbeek, M. M., Otte-Holler, I., Veerhuis, R., Ruiter, D. J., and De Waal, R. M. (1998) Acta Neuropathol (Berl) 96(6), 628–36 142. van Horssen, J., Otte-Holler, I., David, G., Maat-Schieman, M. L. C., van den Heuvel, L. P. W. J., Wesseling, P., de Waal, R. M. W., and Verbeek, M. M. (2001) Acta Neuropathol 102, 604–14 143. Barber, A. J. and Lieth, E. (1997) Dev Dyn 208(1), 62–74 144. Verbeek, M. M., Otte-Holler, I., van den Born, J., van den Heuvel, L. P., David, G., Wesseling, P., and de Waal, R. M. (1999) Am J Pathol 155(6), 2115–25 145. Donahue, J. E., Berzin, T. M., Raﬁi, M. S., Glass, D. J., Yancopoulos, G. D., Fallon, J. R., and Stopa, E. G. (1999) Proc Natl Acad Sci USA 96(11), 6468–72 146. Berzin, T. M., Zipser, B. D., Raﬁi, M. S., Kuo-Leblanc, V., Yancopoulos, G. D., Glass, D. J., Fallon, J. R., and Stopa, E. G. (2000) Neurobiol Aging 21(2), 349–55 147. van Horssen, J., Kleinnijenhuis, J., Maass, C. N., Rensink, A. A., OtteHoller, I., David, G., van den Heuvel, L. P., Wesseling, P., de Waal, R. M., and Verbeek, M. M. (2002) Neurobiol Aging 23(4), 537–45 148. So, K. L., Pun, S., Wan, D. C., and Tsim, K. W. (1996) FEBS Lett 379(1), 63–8 149. Cohen, N. A., Kaufmann, W. E., Worley, P. F., and Rupp, F. (1997) Neuroscience 76(2), 581–96 150. Li, Z., Hilgenberg, L. G., O’Dowd, D. K., and Smith, M. A. (1999) J Neurobiol 39(4), 547–57 151. Lauri, S. E., Kaukinen, S., Kinnunen, T., Ylinen, A., Imai, S., Kaila, K., Taira, T., and Rauvala, H. (1999) J Neurosci 19(4), 1226–35 152. Wright, J. W., Kramar, E. A., Meighan, S. E., and Harding, J. W. (2002) Peptides 23(1), 221–246 153. VanSaun, M. and Werle, M. J. (2000) J Neurobiol 44(3), 369 154. Watson, D. J., Lander, A. D., and Selkoe, D. J. (1997) J Biol Chem 272(50), 31617–24 155. Bame, K. J., Danda, J., Hassall, A., and Tumova, S. (1997) J Biol Chem 272(27), 17005–11 156. Wyss-Coray, T., Lin, C., Sanan, D. A., Mucke, L., and Masliah, E. (2000) Am J Pathol 156(1), 139–50 157. Aharoni, D., Meiri, I., Atzmon, R., Vlodavsky, I., and Amsterdam, A. (1997) Curr Biol 7(1), 43–51 158. Bozzo, C., Bellomo, G., Silengo, L., Tarone, G., and Altruda, F. (1997) Exp Cell Res 237(2), 326–37 159. Mooney, A., Jackson, K., Bacon, R., Streuli, C., Edwards, G., Bassuk, J., and Savill, J. (1999) Am J Pathol 155(2), 599–606 160. Hynes, R. O. (1992) Cell 69(1), 11–25

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

306

C.H. Parsons et al

161. Matter, M. L., Zhang, Z., Nordstedt, C., and Ruoslahti, E. (1998) J Cell Biol 141(4), 1019–30 162. Ueda, K., Fukushima, H., Masliah, E., Xia, Y., Iwai, A., Yoshimoto, M., Otero, D. A., Kondo, J., Ihara, Y., and Saitoh, T. (1993) Proc Natl Acad Sci USA 90(23), 11282–6 163. Masliah, E., Rockenstein, E., Veinbergs, I., Sagara, Y., Mallory, M., Hashimoto, M., and Mucke, L. (2001) Proc Natl Acad Sci USA 98(21), 12245–50. Epub 2001 Sep 25 164. Terai, K., Iwai, A., Kawabata, S., Tasaki, Y., Watanabe, T., Miyata, K., and Yamaguchi, T. (2001) Neuroscience 104(2), 299–310 165. Cohlberg, J. A., Li, J., Uversky, V. N., and Fink, A. L. (2002) Biochemistry 41(5), 1502–11 166. Woods, A. G., Cribbs, D. H., Whittemore, E. R., and Cotman, C. W. (1995) Brain Res 697(1–2), 53–62 167. Hoy, A., Leininger-Muller, B., Jolivalt, C., and Siest, G. (2000) Neurosci Lett 285(3), 173–6 168. Winkler, K., Scharnagl, H., Tisljar, U., Hoschutzky, H., Friedrich, I., Hoﬀmann, M. M., Huttinger, M., Wieland, H., and Marz, W. (1999) J Lipid Res 40(3), 447–55 169. Lindstedt, L., Saarinen, J., Kalkkinen, N., Welgus, H., and Kovanen, P. T. (1999) J Biol Chem 274(32), 22627–34 170. Jeng, A. Y., Chou, M., Sawyer, W. K., Caplan, S. L., Von Linden-Reed, J., Jeune, M., and Prescott, M. F. (1999) Ann NY Acad Sci 878, 555–8 171. Moreau, M., Brocheriou, I., Petit, L., Ninio, E., Chapman, M. J., and Rouis, M. (1999) Circulation 99(3), 420–6 172. Milward, E. A., Papadopoulos, R., Fuller, S. J., Moir, R. D., Small, D., Beyreuther, K., and Masters, C. L. (1992) Neuron 9, 129–137 173. Mattson, M. P. (1997) Physiol Rev 77(4), 1081–132 174. Dodart, J. C., Mathis, C., and Ungerer, A. (2000) Rev Neurosci 11(2–3), 75–93 175. Graham, D. I., Gentleman, S. M., Nicoll, J. A., Royston, M. C., McKenzie, J. E., Roberts, G. W., and Griﬃn, W. S. (1996) Acta Neurochir Suppl (Wien) 66, 96–102 176. Smith, D. H., Chen, X. H., Iwata, A., and Graham, D. I. (2003) J Neurosurg 98(5), 1072–7 177. Muir, D. (1994) Exp Cell Res 210(2), 243–52 178. Nordstrom, L. A., Lochner, J., Yeung, W., and Ciment, G. (1995) Mol Cell Neurosci 6(1), 56–68 179. Narindrasorasak, S., Lowery, D. E., Altman, R. A., GonzalezDeWhitt, P. A., Greenberg, B. D., and Kisilevsky, R. (1992) Lab Invest 67(5), 643–52 180. Pangalos, M. N., Shioi, J., and Robakis, N. K. (1995) J Neurochem 65(2), 762–9 181. Shioi, J., Pangalos, M. N., Ripellino, J. A., Vassilacopoulou, D., Mytilineou, C., Margolis, R. U., and Robakis, N. K. (1995) J Biol Chem 270(20), 11839–44

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

307

182. Li, X. F., Thinakaran, G., Sisodia, S. S., and Yu, F. S. (1999) J Biol Chem 274(38), 27249–56 183. Oohira, A., Kushima, Y., Matsui, F., and Watanabe, E. (1995) Neurosci Lett 189(1), 25–8 184. Zuo, J., Ferguson, T. A., Hernandez, Y. J., Stetler-Stevenson, W. G., and Muir, D. (1998) J Neurosci 18(14), 5203–11 185. Kibbey, M. C., Jucker, M., Weeks, B. S., Neve, R. L., Van Nostrand, W. E., and Kleinman, H. K. (1993) Proc Natl Acad Sci USA 90(21), 10150–3 186. Jucker, M., Walker, L. C., Kibbey, M. C., Kleinman, H. K., and Ingram, D. K. (1993) Neuroscience 56(4), 1009–22 187. Yamazaki, T., Koo, E. H., and Selkoe, D. J. (1997) J Neurosci 17(3), 1004–10 188. Huber, J. D., Egleton, R. D., and Davis, T. P. (2001) Trends Neurosci 24(12), 719–25 189. Claudio, L. (1996) Acta Neuropathol (Berl) 91(1), 6–14 190. Akiguchi, I., Tomimoto, H., Suenaga, T., Wakita, H., and Budka, H. (1998) Acta Neuropathol (Berl) 95(1), 78–84 191. Fiala, M., Liu, Q. N., Sayre, J., Pop, V., Brahmandam, V., Graves, M. C., and Vinters, H. V. (2002) Eur J Clin Invest 32(5), 360–71 192. Skoog, I., Wallin, A., Fredman, P., Hesse, C., Aevarsson, O., Karlsson, I., Gottfries, C. G., and Blennow, K. (1998) Neurology 50(4), 966–71 193. Wada, H. (1998) Intern Med 37(6), 509–13 194. Kalaria, R. N. (1999) Ann NY Acad Sci 893, 113–25 195. Wallin, A., Sjogren, M., Edman, A., Blennow, K., and Regland, B. (2000) Eur Neurol 44(4), 229–35 196. Giri, R., Selvaraj, S., Miller, C. A., Hofman, F., Yan, D. S., Stern, D., Zlokovic, B. V., and Kalra, V. K. (2002) Am J Physiol Cell Physiol 283, C895–904 197. Lee, J.-M., Yin, K.-J., Hsin, I., Chen, S., Fryer, J. D., Holtzman, D. M., Hsu, C. Y., and Xu, J. (2003) Ann Neurol 54(3), 379–82 198. Sherrington, R., Rogaev, E. I., Liang, Y., Rogaeva, E. A., Levesque, G., Ikeda, M., Chi, H., Lin, C., Li, G., Holman, K., and et al (1995) Nature 375(6534), 754–60 199. Strittmatter, W. J., Saunders, A. M., Schmechel, D., Pericak-Vance, M., Enghild, J., Salvesen, G. S., and Roses, A. D. (1993) Proc Natl Acad Sci USA 90(5), 1977–81 200. Corder, E. H., Saunders, A. M., Strittmatter, W. J., Schmechel, D. E., Gaskell, P. C., Small, G. W., Roses, A. D., Haines, J. L., and Pericak-Vance, M. A. (1993) Science 261(5123), 921–3 201. Saunders, A. M., Trowers, M. K., Shimkets, R. A., Blakemore, S., Crowther, D. J., Mansﬁeld, T. A., Wallace, D. M., Strittmatter, W. J., and Roses, A. D. (2000) Biochim Biophys Acta 1502(1), 85–94 202. Helbecque, N., Hermant, X., Cottel, D., and Amouyel, P. (2003) Neurosci Lett 350, 181–3 203. Kehoe, P., Wavrant-De Vrieze, F., Crook, R., Wu, W. S., Holmans, P., Fenton, I., Spurlock, G., Norton, N., Williams, H., Williams, N., Lovestone, S.,

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

308

204. 205. 206.

207. 208. 209. 210. 211. 212. 213. 214. 215.

216.

217. 218. 219. 220. 221. 222.

223. 224. 225.

C.H. Parsons et al

Perez-Tur, J., Hutton, M., Chartier-Harlin, M. C., Shears, S., Roehl, K., Booth, J., Van Voorst, W., Ramic, D., Williams, J., Goate, A., Hardy, J., and Owen, M. J. (1999) Hum Mol Genet 8(2), 237–45 Sudhof, T. C. (1990) J Biol Chem 265(14), 7849–52 Iivonen, S., Helisalmi, S., Mannermaa, A., Alafuzoﬀ, I., Lehtovirta, M., Soininen, H., and Hiltunen, M. (2003) Neurosci Lett 352(2), 146–50 Rosenmann, H., Meiner, Z., Kahana, E., Aladjem, Z., Friedman, G., Ben-Yehuda, A., Grenader, T., Wertman, E., and Abramsky, O. (2004) Am J Med Genet Part B (Neuropsychiatric Genet) 128B, 123–5 Lorenzl, S., Albers, D. S., Chirichigno, J. W., Augood, S. J., and Beal, M. F. (2004) J Neurol Sci 218(1–2), 39–45 Lorenzl, S., Albers, D. S., Narr, S., Chirichigno, J., and Beal, M. F. (2002) Exp Neurol 178(1), 13–20 Moll, U. M., Youngleib, G. L., Rosinski, K. B., and Quigley, J. P. (1990) Cancer Res 50, 6162–70 Goldberg, G. I., Strongin, A., Collier, I. E., Genrich, L. T., and Marmer, B. L. (1992) J Biol Chem 267(7), 4583–91 Kjeldsen, L., Johnsen, A. H., Sengelov, H., and Borregaard, N. (1993) J Biol Chem 268, 10425–10432 Sanchez, D., Ganfornina, M. D., and Bastiani, M. J. (2000) Biochim Biophys Acta 1482(1–2), 102–9 Lim, G. P., Backstrom, J. R., Cullen, M. J., Miller, C. A., Atkinson, R. D., and Tokes, Z. A. (1996) J Neurochem 67(1), 251–9 Beuche, W., Yushchenko, M., Mader, M., Maliszewska, M., Felgenhauer, K., and Weber, F. (2000) Neuroreport 11(16), 3419–22 Suryadevara, R., Holter, S., Borgmann, K., Persidsky, R., Labenz-Zink, C., Persidsky, Y., Gendelman, H. E., Wu, L., and Ghorpade, A. (2003) Glia 44(1), 47–56 Evert, B. O., Vogt, I. R., Kindermann, C., Ozimek, L., de Vos, R. A., Brunt, E. R., Schmitt, I., Klockgether, T., and Wullner, U. (2001) J Neurosci 21(15), 5389–96 Crocker, S. J., Pagenstecher, A., and Campbell, I. L. (2004) J Neurosci Res 75(1), 1–11 Rathke-Hartlieb, S., Budde, P., Ewert, S., Schlomann, U., Staege, M. S., Jockusch, H., Bartsch, J. W., and Frey, J. (2000) FEBS Lett 481(3), 227–34 Stewart, W. F., Kawas, C., Corrada, M., and Metter, E. J. (1997) Neurology 48(3), 626–32 in t’ Veld, B. A., Launer, L. J., Hoes, A. W., and et al (1998) Neurobiol Aging 19, 607–11 Breitner, J. (2001) N Engl J Med 345, 1567–8 in t’ Veld, B. A., Ruitenberg, A., Hofman, A., Launer, L. J., van Duijn, C. M., Stijnen, T., Breteler, M. M., and Stricker, B. H. (2001) N Engl J Med 345(21), 1515–21 Pan, M. R., and Hung, W. C. (2002) J Biol Chem 277(36), 32775–80 Richardson, D. R. (2004) Ann NY Acad Sci 1012, 326–41 Doraiswamy, P. M., and Finefrock, A. E. (2004) Lancet Neurol 3(7), 431–4

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

Matrix Metalloproteinases and Related Proteins in Alzheimer’s Disease

226. 227. 228. 229. 230. 231. 232.

233.

234. 235.

236.

237.

238. 239. 240. 241. 242. 243. 244. 245.

246.

247.

309

Barrett, A. J. (1994) Methods Enzymol 244, 1–15 Auld, D. S. (1995) Methods Enzymol 248, 228–42 Suh, Y. H. and Checler, F. (2002) Pharmacol Rev 54(3), 469–525 Takino, T., Koshikawa, N., Miyamori, H., Tanaka, M., Sasaki, T., Okada, Y., Seiki, M., and Sato, H. (2003) Oncogene 22(30), 4617–26 Llano, E., Pendas, A. M., Freije, J. P., Nakano, A., Knauper, V., Murphy, G., and Lopez-Otin, C. (1999) Cancer Res 59(11), 2570–6 Pei, D. (1999) J Biol Chem 274(13), 8925–32 Eto, K., Huet, C., Tarui, T., Kupriyanov, S., Liu, H. Z., PuzonMcLaughlin, W., Zhang, X. P., Sheppard, D., Engvall, E., and Takada, Y. (2002) J Biol Chem 277(20), 17804–10. Epub 2002 Mar 6 Gerst, J. L., Raina, A. K., Pirim, I., McShea, A., Harris, P. L., Siedlak, S. L., Takeda, A., Petersen, R. B., and Smith, M. A. (2000) J Neurosci Res 59(5), 680–4 Qi, H., Rand, M. D., Wu, X., Sestan, N., Wang, W., Rakic, P., Xu, T., and Artavanis-Tsakonas, S. (1999) Science 283(5398), 91–4 Brou, C., Logeat, F., Gupta, N., Bessia, C., LeBail, O., Doedens, J. R., Cumano, A., Roux, P., Black, R. A., and Israel, A. (2000) Mol Cell 5(2), 207–16 Dudas, B., Cornelli, U., Lee, J. M., Hejna, M. J., Walzer, M., Lorens, S. A., Mervis, R. F., Fareed, J., and Hanin, I. (2002) Neurobiol Aging 23(1), 97–104 Perry, G., Siedlak, S. L., Richey, P., Kawai, M., Cras, P., Kalaria, R. N., Galloway, P. G., Scardina, J. M., Cordell, B., Greenberg, B. D., and et al (1991) J Neurosci 11(11), 3679–83 Snow, A. D., Mar, H., Nochlin, D., Kresse, H., and Wight, T. N. (1992) J Histochem Cytochem 40(1), 105–13 Bignami, A., LeBlanc, A., and Perides, G. (1994) Acta Neuropathol (Berl) 87(3), 308–12 Snow, A. D., Mar, H., Nochlin, D., Kimata, K., Kato, M., Suzuki, S., Hassell, J., and Wight, T. N. (1988) Am J Pathol 133(3), 456–63 Leveugle, B., Ding, W., Buee, L., and Fillit, H. M. (1995) J Neuroimmunol 60(1–2), 151–60 Gupta-Bansal, R. and Brunden, K. R. (1998) J Neurochem 70(1), 292–8 Watanabe, N., Araki, W., Chui, D. H., Makifuchi, T., Ihara, Y., and Tabira, T. (2004) FASEB J 18(9), 1013–5 Cotman, S. L., Halfter, W., and Cole, G. J. (2000) Mol Cell Neurosci 15(2), 183–98 Castillo, G. M., Ngo, C., Cummings, J., Wight, T. N., Snow, A. D., Mar, H., Nochlin, D., Kimata, K., Kato, M., Suzuki, S., and Hassell, J. (1997) J Neurochem 69(6), 2452–65 Castillo, G. M., Lukito, W., Wight, T. N., Snow, A. D., Ngo, C., Cummings, J., Mar, H., Nochlin, D., Kimata, K., Kato, M., Suzuki, S., and Hassell, J. (1999) J Neurochem 72(4), 1681–7 McLaurin, J., Franklin, T., Zhang, X., Deng, J., and Fraser, P. E. (1999) Eur J Biochem 266(3), 1101–10

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch12

310

C.H. Parsons et al

248. Miller, J. D., Cummings, J., Maresh, G. A., Walker, D. G., Castillo, G. M., Ngo, C., Kimata, K., Kinsella, M. G., Wight, T. N., and Snow, A. D. (1997) Glia 21(2), 228–43 249. McLaurin, J., Franklin, T., Kuhns, W. J., and Fraser, P. E. (1999) Amyloid 6(4), 233–43 250. Schulz, J. G., Megow, D., Reszka, R., Villringer, A., Einhaupl, K. M., and Dirnagl, U. (1998) Eur J Neurosci 10(6), 2085–93 251. Scharnagl, H., Tisljar, U., Winkler, K., Huttinger, M., Nauck, M. A., Gross, W., Wieland, H., Ohm, T. G., and Marz, W. (1999) Lab Invest 79(10), 1271–86 252. Holcomb, L. A., Gordon, M. N., Benkovic, S. A., and Morgan, D. G. (2000) Mech Aging Dev 112(2), 135–52 253. Shaﬀer, L. M., Dority, M. D., Gupta-Bansal, R., Frederickson, R. C., Younkin, S. G., and Brunden, K. R. (1995) Neurobiol Aging 16(5), 737–45 254. Kopec, K. K. and Carroll, R. T. (1998) J Neurochem 71(5), 2123–31 255. Gupta-Bansal, R., Frederickson, R. C., and Brunden, K. R. (1995) J Biol Chem 270(31), 18666–71 256. Woods, A. G., Cribbs, D. H., Whittemore, E. R., and Cotman, C. W. (1995) Brain Res 697(1–2), 53–62 257. Walzer, M., Lorens, S., Hejna, M., Fareed, J., Hanin, I., Cornelli, U., and Lee, J. M. (2002) Eur J Pharmacol 445(3), 211–20 258. Morgan, C., Bugueno, M. P., Garrido, J., and Inestrosa, N. C. (2002) Peptides 23(7), 1229–40 259. Hoke, A., Canning, D. R., Malemud, C. J., and Silver, J. (1994) Exp Neurol 130(1), 56–66

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

CHAPTER 13 TISSUE INHIBITORS OF MATRIX METALLOPROTEINASES, INFLAMMATION, AND THE CENTRAL NERVOUS SYSTEM

S.J. Crocker1,∗ and I.L. Campbell1,2,† 1

Department of Neuropharmacology, The Scripps Research Institute, La Jolla, California, USA 2 School of Molecular and Microbial Biosciences, The University of Sydney, Sydney, NSW, Australia E-mails: ∗[email protected] † [email protected]

1. Introduction Many facets of the innate and acquired immune system are regulated by the proteolytic activities of matrix metalloproteinases (MMPs). Regulation of MMPs is mediated, in part, through the expression and actions of the tissue inhibitors of matrix metalloproteinases (TIMPs). Discord in the potent proteolytic balance between MMPs and TIMPs, often represented by an increased MMP:TIMP ratio, has led to current hypotheses that increased MMP-related activity may be involved in immunopathologies and diseases of the central nervous system (CNS). Although TIMPs are recognised as the principal inhibitors of the MMPs, increasing evidence also indicates that TIMP family proteins have inherent immunomodulatory activities distinct from their interactions with or inhibition of the MMPs, such as co-activators of some MMPs, growth factors, diﬀerentiation, proliferation and inducers of intracellular signalling pathways. These less recognised functions of TIMPs have stimulated renewed interest in this family of proteins as potential modulators and mediators of immune responses (1, 2). In this chapter, we provide an overview of our current understanding of the TIMPs and their roles in immune system responses, with speciﬁc Correspondence to: S.J. Crocker and I.L. Campbell 311

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

312

S.J. Crocker & I.L. Campbell

attention to: (1) expression of TIMPs in the nervous system and immune cell types, (2) changes in TIMP gene expression in infection and inﬂammatory diseases of the CNS, and (3) the potential roles for TIMP proteins in immune-related CNS pathophysiology. 2. Properties of the TIMPs The inherent physiological signiﬁcance of TIMPs is underscored by the presence of TIMP and TIMP-like proteins in organisms ranging from invertebrates to mammals. The mammalian TIMP gene family presently consists of 4 members, named timp-1 through -4 (Table 1). TIMP-1 was ﬁrst identiﬁed as a serum protein that exhibited trophic activity for erythroid cells, though its role as a metalloproteinase inhibiting protein has eclipsed its initial function in notoriety. It is therefore postulated that additional functions of eukaryotic TIMP proteins evolved from ancestral homolog(s) coincident with the development of complex organ systems (2, 3). The genomic organisation of all mammalian timp genes is conserved with 5 coding exons within the loci of each timp gene that are either nested within the introns of, or located in close proximity to the synapsin gene family (4, 5). It is not presently known whether the nesting of TIMP within synapsin genes represents a physiological interdependence in either the regulation of these genes or their gene products, although it has been proposed that this gene linkage represents an evolutionary specialisation of individual protein functions (subfunctionalisation) through duplication events (5). All mammalian TIMP proteins are composed of approximately 180 amino acids that comprise two structural domains — an amino-terminal containing the MMP inhibitory domain, and a carboxy-terminal domain that mediates important protein-protein interactions such as the hemopexin domains of pro-MMPs (6, 7). Mature TIMP proteins are held together by six disulﬁde bonds (3 in each) that provide a wedge-shaped conformation resembling the Fab of immunoglobulins. For an in-depth survey of structure-function relationship of TIMPs, see (2). Most TIMPs inhibit all known MMPs through non-covalent associations in a 1:1 stoichiometry (8). The physiological functions of the four known TIMP proteins are recognised to be largely redundant-the broadly-targeted and direct inhibition of MMPs and related proteases (Table 1). However, it is important to note that accumulating evidence also points to TIMP proteins as multifarious, as not all activities of TIMP proteins are accountable to inhibition of MMPs. For instance, TIMP-2 is implicated as an important co-activator

May 26, 2005

TIMP-1 (EPA)3

TIMP-4

1

Protein size Extracellular (amino acids) localisation

X11p11.23-11.4 Human

169 aa

X 6.2 cM

Mouse

205 aa

Human

220 aa

Mouse

201

22q12.1-q13.2

Human

2115

10 47.0 cM

Mouse

211

3p25

Human

224

6 46.0 cM

Mouse

224

TIMP-2 17q25 (CSC-21K) 11 72.0 cM TIMP-3 (MIG-5)

Species

Linked synapsin1

Proteases inhibited

Receptor

Additional known functions Trophic factor Nuclear translocation Anti-apoptotic

Secreted2

I (intron V)

All MMPs tested not MT1-MMP not MMP-14 ADAM-10

None identiﬁed

Secreted4

IV

Inhibits all MMPs tested

pro-MMP-2 Trophic factor α3β1 integrin Activation MMP-2, MT1-MMP Angiogenesis inhibitor

Membrane

Secreted

III (intron V) MMP-1, -2, -3, -7, -9 KDR MMP-14 ADAM-10, -12S, -17 ADAM-TS4, -TS5

VEGF-R2 antagonist Stabilisation of death Domain receptors Tumour suppressor

II (intron V)

None currently known

MMP-1, -2, -3, -7, -9 None identiﬁed

The speciﬁc intron of each synapsin gene in which TIMPs-1, -3 and -4 are nested are indicated in brackets. TIMP-2 is not nested within a synapsin, but is located close to synapsin IV on chromosome 17. 2 Reported to have receptor and may have intranuclear functions following stimulation and/or during cell cycle progression. 3 TIMP-1 was initially identiﬁed as a serum factor with erythroid potentiating activity (EPA) and was so named until it became classiﬁed as a metalloproteinase inhibitor protein in 1985. Additional alternate names for other TIMP family proteins are given in brackets. 4 TIMP-2 is considered membrane associated when bound to pro-MMP-2. 5 Reported variation in size of TIMP-3 related to post-translational modiﬁcations including glycosylations (ref 22).

Tissue Inhibitors of MMPs, Inﬂammation, and the CNS

Gene locus

Properties of the human and mouse ortholog TIMP protein family.

313

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

Table 1.

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

314

S.J. Crocker & I.L. Campbell

of pro-MMP-2 (9) and a potent anti-angiogenic factor (10), while TIMP-1 was ﬁrst identiﬁed as a growth factor for erythroid cells (11). Similarly, TIMPs-1, -2 and -4 are secreted proteins but TIMP-3, while exuded from the cell, remains tethered to the ECM through sulfated glycosaminoglycans (12). TIMP-3 also is a more potent inhibitor of membrane-associated proteases, including the membrane type matrixins (MT-MMPs) (13) and members of the metalloproteinase disintegrin (ADAM) and aggrecanase families (14, 15). Thus, our views on the functions of TIMPs is broadening as novel functions are being elucidated for this protein family. 3. Anatomical Distribution of the TIMPs The developmental expression of TIMP proteins will be examined in detail by our colleagues in the chapter following and thus we will not cover this topic here. However, we will explore what is known of the patterns of expression of timp genes in the adult nervous and immune systems as these descriptions are particularly germane to our later discussions on models of disease and neuroinﬂammation. Analysis of timp gene expression by semi-quantitative reverse transciptase-PCR (qRT-PCR) has shown that each TIMP mRNA is diﬀerentially expressed in relative abundance in the major organ systems of adult mice (16). Examination of speciﬁc TIMP mRNA expression patterns in the adult murine brain reveal that timps-2, -3 and -4 genes are abundantly expressed (17, 18) while TIMP-1 RNA is typically non-detectable to very low by either RT-PCR (16) or RNase protection assay (RPA) (18). Similarly, while expression of TIMPs-2 and -3 mRNA are found in heart, lung and kidney of adult mice, the TIMPs-1 and -4 are not readily detected (19). These generalised patterns of relative expression for each timp gene appear to be valid for most species examined to date, though some diﬀerences have been observed. For example, Northern blot analysis of human tissues by Greene et al (1996) would suggest that expression of the timp-4 gene is limited to only the cardiac tissues of adults (20). However, RPA analyses of mouse tissues performed by Rahkonen et al (2002) demonstrate a slightly wider distribution of timp-4 gene expression that includes the heart as well as the CNS and ovaries (21). While the extent and nature of any underlying interspecies diﬀerences in timp gene expression is not presently clear, further characterisation of any potential disparities between species will be necessary for extrapolating experimental ﬁndings in mice to human disease conditions.

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

Tissue Inhibitors of MMPs, Inﬂammation, and the CNS

315

Within the adult CNS, the expression of each timp gene also exhibits individual variation. In general, the timp-2 gene is widely and constitutively expressed by both neurons and astrocytes. More speciﬁcally, TIMP-2 protein and mRNA is expressed within the deep layers of the cerebral cortex, olfactory cortex, gigantocellular neurons of the brain stem, spinal motoneurons of the anterior horn and granular neurons of the cerebellum (18, 22, 23). TIMP-3 mRNA is expressed by astrocytes as well as neurons and is predominant in the thalamus, subventricular zone (SVZ) and olfactory bulb, choroid plexus, and neurons within the hilar region of dentate gyrus of the hippocampus (18, 24). The CNS distribution and cellular phenotype(s) expressing TIMP-4 RNA or protein within the developing or adult CNS have not been extensively studied, although the level of TIMP-4 expression is known to be down-regulated in brain arteriovenous malformations (25). The persistent expression of TIMPs-2 and -3 within sites enriched with populations of neuronal progenitors in the adult CNS (e.g. the rostral migratory tract) has suggested a role for these proteins in neurogenesis and/or cellular migration (23, 24). The coincident expression of MMPs in the same regions of cellular proliferation in the adult CNS where TIMPs-2 and -3 are expressed suggests a functional balance in dynamic proteolysis in the mature brain where regulation of cellular adhesion, mobilisation of growth factors and/or neurite outgrowth participate in the limited regenerative capacity of the CNS (26). Although knockout mice for the timps-1, -2, and -3 genes have been developed, there have not yet been any reports describing phenotypes associated with either the development or responsiveness of the CNS to injury or infection. However, phenotypes have been noted for other organ systems in timp knockout mice. The limited basal expression of timp-1, aside from the female reproductive organs has left the timp-1 knockout mouse without an overt CNS phenotype. However, these animals have been reported to exhibit disrupted female reproductive cyclicity associated with reduced uterus size and low progesterone production during estrus and increased branching of lumen precocious folds (27). TIMP-1 knockout mice have also been reported to be resistant to bacterial infection (discussed below). In contrast, mice lacking TIMP-3 develop a progressive impairment in lung function resulting in premature lethality after one year of age (19). Although timp-2 deﬁcient mice display a reduction in MMP-2 activity, these mice are described as otherwise normal (9, 28). This latter knockout also reﬂects the paradoxical role for this inhibitor protein in the activation of an enzyme it also inhibits (9, 29).

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

316

S.J. Crocker & I.L. Campbell

Of the four known TIMP proteins, it is widely acknowledged that TIMP-1 exhibits the lowest (very little to not detectable) expression in the adult nervous system. It has also been observed that TIMP-1 is readily and profoundly induced by astrocytes and/or neurons within the adult CNS under a variety of conditions where the brain is stressed and/or injured. It is of particular relevance to add that while the CNS rapidly induces TIMP-1 protein, expression of the remaining TIMP family proteins remains unchanged or is concurrently decreased by the stressor or injury. This observation is central to our current understanding that TIMP-1 protein functions as a primary protective response of the CNS to injury. Accordingly, TIMP-1 likely performs a role as guardian of the CNS against inﬁltrating cells following permeation of the blood-brain barrier (BBB) during injury and/or inﬂammation. Regulation of timp-1 gene expression is also considered to be sensitive to ‘danger signals’ associated with inﬂammation, including pro-inﬂammatory cytokines (as discussed below). However, it is also important in the context of neuroimmune system interactions and disease to discuss timp gene expression in the immune system since TIMP proteins have also been reported to be involved in the physiological regulation of immune responses. Moreover, since immune system plays a primary role in defence from pathogen infection it is also integral in the development of CNS viral infections and autoimmune disorders aﬀecting the CNS (30–32). With respect to organs of the mature immune system, expression of each timp gene is also varied and distinct. For instance, expression of timps-2 and -3 are most prominent in the thymus, while expression of timp-4 is less than either timps-2 or -3 but of comparatively greater relative abundance than timp-1. In the secondary lymphoid organs, expression of timps in spleen have revealed timps-2 and -3 to be most abundant, with modest timp-1 expression and timp-4 exhibiting the lowest levels of relative expression (33). These organs of the immune system provide sites for immune cell production, maturation and/or antigen-presentation which play a major role in the development of autoimmune disorders aﬀecting the CNS. Although our understanding of the precise function(s) of TIMP proteins in the physiology of these organs and the immune system is incomplete at this time, studies described below provide compelling evidence to suggest direct participation of TIMPs in the modulation of immune system responses. Complementing this notion is the identiﬁcation that cells of the immune system also exhibit phenotype-speciﬁc patterns of timp expression. Cellular immunity is achieved through the complex and coordinated interplay between the diﬀerentiation and maturation of many distinct cell types including, but not limited to, lymphocytes, macrophages, neutrophils

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

Tissue Inhibitors of MMPs, Inﬂammation, and the CNS

317

and granulocytes. In a recent survey of timp gene expression in immune cell types, Bar-Or et al (2003) determined using qPCR that the ordinal expression levels of all TIMPs in human peripheral blood mononuclear cells (PBMCs) is: TIMP-1, TIMP-2, TIMP-3 with minimal TIMP-4 detected in these cells. When examined with respect to speciﬁc subtypes of PBMCs, including B-cells, monocytes and T -cells, the ordinal pattern of TIMP expression is analogous to overall PBMCs (34), although the expression of TIMP-1 in T -cells is approximately 5 fold higher than in B-cells (35). TIMP-1 mRNA was found to be enriched in monocytes by several fold versus any other timp gene (e.g. TIMP-1 expression is 16 fold higher than TIMP-2). TIMP-4 was not expressed by either B-cells or T -cells, and only expressed at very low levels in monocytes (34). Since these ﬁndings clearly demonstrate that expression each timp gene is variably expressed in cell types and tissues, understanding the key factors that govern this diversity and regulate TIMP expression patterns may serve to elucidate their roles in neuroimmune interactions, homeostasis and disease. Accumulating evidence indicates that expression of timp genes is modulated by infection and inﬂammation and thus mediated by the selective stimulatory or repressive transcriptional actions of chemokines and cytokines. For instance, in cells of the immune system, such as cultured human granulocytes, constitutive expression and spontaneous release of TIMP-1 protein can be enhanced with exposure to phorbol esters (PMA) or fMLP (36). In contrast, stimulation of murine B-cells with IgGs, monocytes with IFN-γ or T -cells with LPS resulted in the consistent down-regulation of TIMPs-1, -2 and -4. While expression of TIMP-3 was also reduced in B-cells following IgG stimulation, expression of TIMP-3 was modestly increased by stimulation in monocytes and T -cells. This ﬁnding is consistent with previous studies on the regulation of timp genes where the comparatively high levels of expression of TIMPs-1 and -2 are signiﬁcantly down-regulated by treatment with tumour necrosis factor-α (TNFα), whilst concurrent expression of TIMP-3 is profoundly up-regulated (37). This inverse regulation of TIMP-3 with other TIMPs is also observed in cultured brain microvascular endothelial cells or astrocytes following treatment with combinations of IL-1β and TNFα or IFNγ (22, 38). From these studies, it is also intuitive that systemic pro-inﬂammatory cytokines could coincidently modulate timp gene expression in cells of both the immune and nervous systems (38–40). It is important, therefore to reiterate that while TIMP-1 expression in cells of the immune system is frequently robust and can be negatively regulated, in the adult CNS, there is little constitutive expression of TIMP-1 although TIMPs-2, -3 and -4 are

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

318

S.J. Crocker & I.L. Campbell

expressed widely. Hence, in terms of neuroinﬂammation, the blood-brain barrier may provide an eﬀective shield to basal, regulatory blood-borne factors that support constitutive expression of timp-1 in the immune system compared with the CNS. As an inherent response to permeation of the blood-brain barrier during disease or infection, induction of TIMP-1 in astrocytes as a protective response may be acutely governed by serum (40) and cytokines such as IL-1β, TNFα or IFN-γ (38, 40). Hence, ‘danger’ factors, such as cytokines induced by infection or injury likely play dual roles as important physiological regulators of timp expression in cells of both the nervous and immune systems. As cytokines are also key determinants of immune cell activation and/or maturation during immune responses, it is relevant to note that granulocytes spontaneously release TIMP-1 and their activation enhances its release (36). Given the important role MMPs play in the activation of cytokines and that MMPs are regulated by these same molecules, it could be postulated that the MMPs are integral to immune cell maturation and/or activation. In this scenario, increased expression of TIMPs could modulate the activation status of T -cells, by limiting or inhibiting the production of Th1 cytokines by MMPs. However, the paucity of data at the present time regarding the complex relationships between timp gene expression and TIMP protein functions associated with immune physiology limits our interpretations on the involvement of TIMPs in immune responses. In future, establishing functional roles for the observed changes in expression of timps during acute pathogen infections and/or involved in T -cell memory, for instance, may extend our understanding of the eﬀects of infections on the TIMP/MMP axis, immune system function and even their contribution to related CNS pathologies. The diﬀerential regulation and expression of TIMPs in the cells of the nervous and immune systems likely correlate with their involvement and the speciﬁc cell types associated with inﬂammatory injury to the CNS. In the following sections, we will explore several neuroinﬂammatory conditions and the role(s) of TIMP proteins in these situations.

4. TIMPs in Immune, Inﬂammatory and Infectious Disorders of the CNS 4.1. Demyelinating disorders Accumulating evidence supports a role for MMPs in the processes mediating demyelination [for review see (Cuzner and Opdenakker, 1999)].

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

Tissue Inhibitors of MMPs, Inflammation, and the CNS

319

Dysregulation of the timp gene expression and modiﬁcation of the proteolytic balance between TIMPs and MMP gene expression by circulating lymphocytes have been proposed as an event leading to demyelination (Fig. 1). In experimental autoimmune encephalomyelitis (EAE), an animal model of demyelination, signiﬁcant changes in MMP and TIMP-1 expression coincide with the clinical phase of disability associated with demyelination in mice (18, 41). Moreover, the expression of TIMP-1 by resident astrocytes surrounding inﬂammatory lesions in the spinal cord suggests that the brain combats the propagation of inﬂammation-generated injury by the selective induction of TIMP-1 (but not TIMPs-2 or -3) in response to signals brought about by the migration of activated lymphocyte across the BBB (18). Recent reports have also identiﬁed that TIMP expression may not be limited to resident cells within the CNS, but as

Fig. 1. Dynamic regulation of matrix metalloproteinases (MMP) and the MMP inhibitor (TIMP-1) in CNS inflammation. MMPs are produced predominantly by infiltrating leukocytes and facilitate extravasation and migration of these cells across the blood-brain barrier (BBB) and into the brain parenchyma leading to tissue destruction and inflammation. Cytokines and possibly other factors associated with the inflammatory lesion and processes associated with BBB permeability (e.g. serum) induce TIMP-1 in surrounding astrocytes. TIMP-1 acts to inhibit MMP activity and is proposed to promote containment and resolution of the inflammatory lesion as well as contribute to tissue repair.

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

320

S.J. Crocker & I.L. Campbell

discussed above, monocytes in peripheral blood may also represent a principal source of TIMPs (and MMPs) (34, 42). It is important to note, however, that analysis of TIMP expression in the CNS of EAE mice by in situ hybridisation indicated that astrocytes and not cells inﬁltrating into the CNS, were the primary source of timp gene expression in this disease model (18). Importantly, the involvement of the TIMP/MMP proteolytic axis is not limited to immune-mediated demyelination since genetic models of spontaneous demyelination also report perturbation in the expression of MMPs and TIMPs. Development of spontaneous demyelination in the DM20 transgenic mouse is accompanied by a signiﬁcant and sustained increase of TIMP-1 protein expression (but not the other three TIMPs) (43). Moreover, crossing the DM20 mouse with a TIMP-1 transgenic mouse resulted in modest diminution of the clinical eﬀects of DM20 overexpression. Indeed, these ﬁndings support the notion that inhibiting MMPs is therapeutically beneﬁcial in preventing demyelination as ﬁrst supported by the observation that synthetic MMP inhibitors ameliorate EAE in mice (44, 45). Moreover, these results also support the idea that localised expression of TIMP-1 (by astrocytes in the CNS, for instance) function in models of white matter injury to contain the pools of activated leukocytes in the brain parenchyma and thereby limit the expansion of MMP-mediated inﬂammatory lesions and may also act to mitigate tissue destruction (18). If the ﬁndings regarding the role of TIMPs (and MMPs) in demyelinating disease are valid, one would predict analogous changes to be observed in equivalent human disease conditions, such as multiple sclerosis (MS). And, the evolving story of TIMPs and MMPs derived from in vivo animal studies reﬂects the changes observed clinically in human cases of multiple sclerosis (MS). In the CSF as well as brain tissue of patients suﬀering from MS increased MMP activities are reported (44, 46, 47), though in contrast with ﬁndings from the animal models, expression of the four timp genes may not be signiﬁcantly altered in tissues from MS patients (48). Clearly, more studies are needed to address the status of the TIMPs in the MS aﬄicted CNS. However, measurements of TIMPs-1 and -2 (and MMP-9) levels in serum of MS patients in other studies also suggest decreases in TIMP-1 expression may predict new inﬂammatory lesions (49, 50). Therapeutically, the observation that treatment of secondary progressive MS patients with interferon-β coincidently modulates the MMP/TIMP ratio by enhancing expression of TIMP-1 lends further support to the views that increased MMP activities are deleterious in this autoimmune disease (50) and TIMP-1 has an important protective role via inhibition of MMPs.

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

Tissue Inhibitors of MMPs, Inﬂammation, and the CNS

321

4.2. Cerebral ischemia Inﬂammation is an almost ubiquitous feature accompanying CNS injury. Activation of resident microglia and/or the transmigration of inﬂammatory cells into the brain parenchyma are commonplace in conditions resulting in neurodegeneration. One situation that evokes a breakdown in the BBB and subsequent leukocytic invasion into regions of the CNS is the prolonged or permanent interruption of cerebral blood ﬂow, called cerebral ischemia. Ischemic brain injury most often results when there is insuﬃcient cranial blood pressure that evokes the loss of neurons in brain regions particularly sensitive to hypoxia (global cerebral ischemia), or when an interruption in blood ﬂow to a speciﬁc area (e.g. blood clot) triggers a focused cerebral ischemic event resulting in loss of brain tissue emanating from the site of the vascular blockage. In experimental models of stroke, ischemia induces the expression of MMPs, and inhibition of MMP-9 has been shown to result in a reduction in ischemic infarct volume (51, 52). The increased expression of MMPs by leukocytes following cerebral ischemia conforms with the theme that activation of MMPs without physiologically antagonistic changes in the expression of TIMPs results in a shift in the net proteolytic balance. Enhanced MMP activities thus precipitate an increase in BBB permeability and facilitate the migration of inﬂammatory cells into the CNS (51). The results described for MMP-9 involve a severe focal model of cerebral ischemia that lacks induction of timp genes (52). Where the focal model of cerebral ischemia produces a large core of necrotic tissue that evolves quickly during the ischemic event, a more gradual model of cerebral ischemia is transient global ischemia where the majority of neuronal loss is delayed and involves apoptosis rather than necrosis. In global ischemia, TIMP-1 expression is initially expressed by neurons in the dentate gyrus of the hippocampus, a region recognised for its resilience to this form of ischemic injury (51). The early induction of TIMP1 mRNA (by 4 hrs following the ischemic event) is then followed by a second, delayed phase of TIMP-1 expression within the CA1 hippocampus by 24 hrs post-infarction (51). Interestingly, the CA1 region of the hippocampus is particularly vulnerable to hypoxic injury and can exhibit signiﬁcant neuronal death following brief periods of ischemia (53). Rivera et al (2002) also reported that induction of TIMP-1 spreads over time to encompass the entire hippocampus. In this context, the later phase of TIMP-1 expression in the CA1 may represent an endogenous protective response of the hippocampus to the ischemic injury, perhaps initiated by inﬂammation or signals emanating from dying hippocampal neurons.

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

322

S.J. Crocker & I.L. Campbell

Although it is tempting to speculate on the potential beneﬁcial eﬀects of increased TIMP expression following cerebral ischemia, the precise function(s) of TIMP proteins in neurodegenerative conditions are poorly understood. Previous work in primary hippocampal neuron cultures has demonstrated that adenovirus-mediated expression of TIMP-1 can attenuate glutamate-induced excitotoxicity by modulating receptormediated calcium inﬂux — an eﬀect not reproduced by synthetic MMP inhibitors (54). Taken together, the role of increased TIMP-1 expression observed in vivo by Rivera et al (2002) may represent a dual role of TIMP1 as both a regulator of MMP activity (parrying inﬂammatory cells) and/or as a neuroprotective or neurotrophic molecule (preventing delayed neuronal death). The temporal pattern of hippocampal TIMP-1 expression following ischemia is also coincident with reported bi-phasic increases in BBB permeability, indicating that TIMP-1 induction following ischemia may also represent a secondary response, or marker, of serum leakage and immune cell inﬁltration into the CNS. As we had mentioned above, the regulation of TIMP-3 is often inversely related to TIMP-1. Moreover, the biological activities of TIMP-3 also seem to counter the functions of TIMP-1. For instance, expression of TIMP-3 can sensitise cortical neuron cultures to Fas-mediated injury through the stabilisation of cell death domain receptors (55, 56). This eﬀect is unrelated to the function of TIMP-3 as an inhibit of MMPs (55). It is of particular interest on this point that TIMP-3 is also induced in the rodent hippocampus in response to cerebral hypoxia within areas that are most vulnerable to ischemic injury, and more importantly, expression of TIMP-3 co-localises with markers of neuronal death (56). Therefore, identifying the factors involved in the regulation of each TIMP protein following cerebral ischemia and determining their roles in ischemic injury or recovery will be pertinent to our understanding of the physiological roles of TIMP proteins in hypoxic injury and other CNS pathologies.

4.3. Infectious disorders The brain has evolved many strategies to minimise the intensity of an immunological storm resulting from an infection. This is to avoid catastrophic consequences of injury to and loss of irreplaceable neurons. Nevertheless, mounting evidence suggests that in certain circumstances the nervous system itself may promote immune cell recruitment across the BBB

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

Tissue Inhibitors of MMPs, Inﬂammation, and the CNS

323

following injury (57) or even as a part of routine immunosurveillance of the normal CNS (58). On these points, increasing evidence suggests that TIMPs, and TIMP-1 in particular, actively participate in the regulation of immune cell traﬃcking and BBB permeability (35, 59). An accumulating number of studies have reported increased incidence of antibodies to pathogens in individuals with neurodegenerative illnesses, including Parkinson’s disease and Multiple Sclerosis. These correlative ﬁndings have led to the notion that some pathogens may either increase the susceptibility of an individual (or population) to a particular disorder, or that certain pathogenic infections may precipitate neurological and/or neurodegenerative illnesses in humans (60, 61). Hence, should TIMP proteins, and TIMP-1 in particular, function to modulate viral or bacterial pathogenesis or provide a ﬁrst line defensive reaction of the CNS to these agents, understanding the physiological roles of TIMP protein may not only lead to an understanding of how the brain responds to infections but also may provide a novel link between infections and neurological disorders. In the following sections, we discuss what is known of the role of TIMP proteins in the immune response to viral and bacterial pathogens.

5. TIMPs and Viral Infections The production of MMPs by activated lymphocytes is thought to be a primary trigger of immunopathology associated with viral infections (62). The selective temporal and spatial induction of TIMP expression following viral infections (Table 2) has therefore led to the proposal that the physiological function of TIMP proteins in inﬂammatory diseases are a cytoprotective response by the infected tissues to cell-mediated tissue injury (see Fig. 1). Perhaps one of the most studied viruses in human history, the Human Immunodeﬁciency Virus (HIV)-1 has well characterised eﬀects on expression of timp genes and in particular, TIMP-1 protein. In the human, HIV infection can lead to encephalitis (HIVE). Analysis of cerebrospinal ﬂuid or brain tissues from HIVE infection led to an initial ﬁnding that HIV infection is associated with increased levels of TIMP-1 (63). However, analysis of human brain samples from HIVE and controls suggested that TIMP-1 expression was actually signiﬁcantly reduced in HIV infection (64). Infection of cultured human astrocytes was also reported to acutely up-regulate TIMP-1 mRNA expression (64). From the results of these studies on HIV infection and HIVE, we can begin to devise a model describing

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

324

S.J. Crocker & I.L. Campbell Table 2.

Pathogens reported to signiﬁcantly up-regulate expression of TIMPs.

Pathogen

Variant or

Viruses

strain

Tissue

sindbis virus

dsTE12H

Brainc

sindbis virus

dsTE12Q

Brainc

simian immunodeﬁciency virus human immunodeﬁciency virus vaccinia

mac182

Brainb

HIV-1

Braina

MVA

DCsd

canine distemper virus

Onderstepoort

Brainc

hepatitis virus

DM/JHMV

Brainc

hepatitis virus

HVA, B or C

Livera

n/ae

Newcastle disease virus

Reference

Johnston et al (2001) (ref 82) Johnston et al (2001) (ref 82) Roberts et al (2003) (ref 72) Suryadevara et al (2003) (ref 64) Osman et al (2002) (ref 37) Khuth et al (2001) (ref 75) Zhou et al (2002) (ref 83) Flisiak et al (2002) (ref 84) Koulentaki et al (2002) (ref 85) Leroy et al (2004) (ref 86) Gewert et al (1987) (ref 87)

Bacteria M. tuberculosis Staphylococcus aureus Salmonella typhi Escherichia coli Enterohemoragic E. coli Pseduomonas aeruginosa

Erdman ISP794 Quailes sd-4 0157:H7

macrophagesd macrophagesd macrophagesd macrophagesd macrophagesd corneac

Nau et al (2002) (ref 77) Nau et al (2002) (ref 77) Nau et al (2002) (ref 77) Nau et al (2002) (ref 77) Nau et al (2002) (ref 77) Kernacki et al (2004) (ref 88)

(a) samples from human subjects; (b) sample from non-human primate; (c) sample from rodent; (d) results derived from in vitro human cells or cell line experiment; (e) results from rodent-derived cell line.

our understanding of the role of TIMP-1 protein function and immunity to viral infection(s) in the CNS. Infection of monocytes is thought to be a primary site of HIV infection, where HIV virions can be transported about the body into a variety of tissues, including CNS (65). Infection of monocytes acutely triggers expression of TIMP-1 (66). Administration of TIMP-1 protein to HIV-infected monocytes in culture prevents cell migration across endothelial cells pointing to a prominent role for TIMP-1 expression as

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

Tissue Inhibitors of MMPs, Inﬂammation, and the CNS

325

limiting inﬁltration of cells into the CNS (66, 67). Further, Alter et al (2003) recently reported that in an in vitro cell migration assay application of recombinant TIMP-1 protein also prevented B-cell migration across cultured human brain-derived endothelial cells in response to TNFα and/or IFNγ. Since macrophage and leukocyte transmigration are relevant components in the development of CNS pathology (68), the actions of TIMP-1 in this context are of particular relevance to HIV and HIVE. Interestingly, and in contrast to ﬁndings following acute post-infection time intervals, chronic infection with HIV has been reported to down-regulate TIMP-1 protein production in many cell types including macrophages and vascular endothelial cells (69). This observation is consistent with the increased motility of monocytes by HIV-tat induced chemotaxis and the ability of recombinant TIMP-1 protein or TIMP-1 with TIMP-2 protein to significantly retard this invasiveness (70). Thus, from the proposed model, a decrease in TIMP-1 expression over time may potentially contribute to an increased propensity of viral infection within the brain parenchyma (71). Indeed, a reduction in TIMP-1 protein levels are observed in both the brain tissues and CSF of chronically-infected HIV infected patients in comparison with seronegative controls (64). It is important to note that these changes in TIMP-1 expression following extended exposure to HIV are also observed in cultured glial cells, where HIV infection initially mimics an enhancement of timp-1 production that is followed by reduced expression (64). Together, these ﬁndings implicate a potential key role for TIMP-1 protein function(s) in entry and spread of viral infections of the CNS, and HIV in particular. Notably, these changes in human HIV are also modelled in non-human primates as in a simian immunodeﬁciency virus (SIV) model in macaque monkeys, gene array analysis indicated that TIMP-1 mRNA was one of the most abundantly induced by acute infections (72). Increased expression of MMP-9 and TIMP-1 are also detected in the CSF of bacterial meningitis and viral meningoencephalitis (73, 74) (Table 1). Interestingly, in the CSF of patients suﬀering from viral meningitis the expression level of TIMP-1 was 43-fold higher as compared to MMP-9 suggesting that all MMP-9 molecules are bound to TIMP-1 (74). In a murine model of canine distemper virus (CDV) infection, site speciﬁc induction of pro-inﬂammatory cytokines such as IFN-γ, TNF-α and IL-6 in diﬀerent CNS regions was found to correlate with increased mt1-Mmp and timp-1 and timp-3 gene expression (75), indicating that a number of Mmp as well as timp genes are coincidently regulated by lesion-associated factors such as the cytokines. However, changes in Mmp and timp gene expression

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

326

S.J. Crocker & I.L. Campbell

are not always in concert. For instance, in systemic lupus erythematosus (SLE) induced peripheral neuropathy, Mawrin et al (2003) have recently reported mononuclear inﬁltrates immunoreactive for several MMPs in 11 of 12 patients while there were only ﬁve cases immunoreactive for TIMP-1. Conversely, administration of the bacterial endotoxin, lipopolysaccharide (LPS) induces a condition of endotoxemia wherein low doses of LPS lead to induction of timp-1 gene expression in the brain while the expression of MMP-genes remain unaltered (76). 6. TIMPs and Bacterial Infections In comparison with the diversity of studies examining changes in TIMPs (and MMPs) in viral infections, little is known of the fate of TIMPs in response to the plethora of potential bacterial pathogens. In a recent study by Nau et al (2002), changes in the gene expression proﬁles of macrophages were examined in response to bacterial infections using DNA microarray technology (77). This study revealed that TIMP-1 is robustly induced by both Gram-positive and Gram-negative bacteria (Table 2), and thus may participate in the innate immune response to infection by these agents. Administration of the Gram-negative bacterial endotoxin lipopolysaccharide (LPS) can also modulate expression of MMPs and TIMPs. LPS evokes production and release of pro-inﬂammatory cytokines mediated by its heteromeric receptor complex (containing CD14 and TLR4) which is expressed on a variety of cell types including cells of the immune and nervous systems (78, 79). When administered to mice, LPS can induce endotoxemia and is often used as a model of sepsis. Previous work has demonstrated that when administered to mice LPS induces a signiﬁcant up-regulation of mRNA for TIMPs-1 and -2 in kidney and spleen, with TIMP-1 mRNA also induced in liver and brain (76). While high doses of LPS were required to induce MMP mRNA expression in the brain, even low doses of LPS were capable of triggering expression of TIMP-1 mRNA in brain suggesting that this gene in particular is a sensitive marker of changes in cytokines within the neurovasculature (76). Interestingly, analysis of timp expression in LPS-resistant mouse strain (C3H/HeJ) also exhibited an up-regulation of TIMP-1 in peripheral organs in response to LPS administration. Aside from the obvious contribution these changes in gene expression (Mmps and timps) likely play in LPS-induced immunopathology, these ﬁndings also point to LPS-induced pro-inﬂammatory cytokines as potent inducers of timp-1 gene expression (76), particularly IL-1 and TNFα. Consistent

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

Tissue Inhibitors of MMPs, Inﬂammation, and the CNS

327

with the observation of LPS-induces timp gene expression in vivo, work by Pagenstecher et al (1998) has also described selective increases in TIMP-1 and -3 in the brains of transgenic mice with astrocyte (GFAP)-targeted production of TNFα (GFAP-TNF) or IL-3 (GFAP-IL3) (76). While the reports above indicate signiﬁcant changes in timp gene expression and hence likely to reﬂect a shift in the TIMP/MMP axis, what is the functional evidence that changes in TIMP-1 have a physiological eﬀect in response to bacterial infection? A study by Soloway and colleagues has indicated that mice deﬁcient of TIMP-1 have increased resistance to pseudomonas infection (80). This bacteria resistant phenotype of the timp-1 knockout mice has been suggested to be associated with a change in neutrophil functioning (80), although it has been reported elsewhere that neutrophils do not produce TIMP-1 protein (81). Taken together, these ﬁndings indicate that this phenotype is not associated with the loss of timp-1 expression in the neutrophils but is clearly a consequence of loss of TIMP-1 function in an associated cell type that might inﬂuence the neutrophil. Similarly, as each of the above mentioned transgenic mouse lines (GFAP-TNF and GFAP-IL3) develop signiﬁcant clinical and neuropathological features that culminate in premature lethality, it would be revealing to determine a functional role for TIMP proteins in these models by exploring the consequence(s) of blocking their expression in vivo. 7. Conclusion From the above discussion, it is clear that our understanding of the physiological role(s) of TIMP proteins as both endogenous metalloproteinase inhibitors and proteins with MMP-independent activities is continuing to evolve. Consequently, the function(s) these proteins play in the immune system response to infection or inﬂammatory responses involving the CNS is also emerging. Nevertheless, we know that TIMP proteins are widely and diﬀerentially expressed in the nervous and immune systems. Further, the observation that increased expression of timp genes, and timp-1 in particular, as a common (and almost ubiquitous) response of organs, including the CNS, to inﬂammatory processes elicited by viruses, bacteria, ischemia or autoimmunity underlies the emerging importance of these proteins and the inherent value in elucidating their physiological role(s) in these diseases. Given that the expression of individual timp genes appears to be diﬀerentially regulated in the immune and nervous system in response to injury and infection, this is likely to reﬂect unique properties and functions

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

328

S.J. Crocker & I.L. Campbell

speciﬁc to each TIMP protein. Thus, our understanding of the implications of modulating the TIMP:MMP proteolytic balance and even the multipotency of TIMP family proteins is presently limited. In the context of human diseases and infections eﬀecting the CNS, it is consequently unclear what precise roles TIMP proteins may have on pathology and/or survival in the variety of possible infections and diseases that aﬀect neuro-immune system. Future directions in research on TIMP proteins should address the outstanding question of the functional importance of the TIMP proteins in inﬂammation and diseases of the central nervous system.

Acknowledgments The authors’ work described in this article was supported by NIH grants MH62231, NS36979, MH62261 and NS36979 to I.L.C. S.J.C is recipient of an Advanced Postdoctoral Fellowship from the National Multiple Sclerosis Society (USA). This is manuscript No. 16643-NP from The Scripps Research Institute.

References 1. Mannello, F. and Gazzanelli, G. (2001) Apoptosis 6, 479–482 2. Brew, K., Dinakarpandian, D., and Nagase, H. (2000) Biochim Biophys Acta 1477, 267–283 3. Blelloch, R. and Kimble, J. (1999) Nature 399, 586–590 4. Pohar, N., Godenschwege, T. A., and Buchner, E. (1999) Genomics 57, 293–296 5. Yu, W. P., Brenner, S., and Venkatesh, B. (2003) Trends Genet 19, 180–183 6. Kai, H. S., Butler, G. S., Morrison, C. J., King, A. E., Pelman, G. R., and Overall, C. M. (2002) J Biol Chem 277, 48696–48707 7. Strongin, A. Y., Collier, I., Bannikov, G., Marmer, B. L., Grant, G. A., and Goldberg, G. I. (1995) J Biol Chem 270, 5331–5338 8. Gomis-Ruth, F. X., Maskos, K., Betz, M., Bergner, A., Huber, R., Suzuki, K., Yoshida, N., Nagase, H., Brew, K., Bourenkov, G. P., Bartunik, H., and Bode, W. (1997) Nature 389, 77–81 9. Wang, Z., Juttermann, R., and Soloway, P. D. (2000) J Biol Chem 275, 26411–26415 10. Seo, D. W., Li, H., Guedez, L., Wingﬁeld, P. T., Diaz, T., Salloum, R., Wei, B. Y., and Stetler-Stevenson, W. G. (2003) Cell 114, 171–180 11. Gasson, J. C., Golde, D. W., Kaufman, S. E., Westbrook, C. A., Hewick, R. M., Kaufman, R. J., Wong, G. G., Temple, P. A., Leary, A. C., Brown, E. L., and et al (1985) Nature 315, 768–771

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

Tissue Inhibitors of MMPs, Inﬂammation, and the CNS

329

12. Yu, W. H., Yu, S., Meng, Q., Brew, K., and Woessner, J. F., Jr. (2000) J Biol Chem 275, 31226–31232 13. Butler, G. S., Will, H., Atkinson, S. J., and Murphy, G. (1997) Eur J Biochem 244, 653–657 14. Amour, A., Slocombe, P. M., Webster, A., Butler, M., Knight, C. G., Smith, B. J., Stephens, P. E., Shelley, C., Hutton, M., Knauper, V., Docherty, A. J., and Murphy, G. (1998) FEBS Lett 435, 39–44 15. Kashiwagi, M., Tortorella, M., Nagase, H., and Brew, K. (2001) J Biol Chem 276, 12501–12504 16. Young, D. A., Phillips, B. W., Lundy, C., Nuttall, R. K., Hogan, A., Schultz, G. A., Leco, K. J., Clark, I. M., and Edwards, D. R. (2002) Biochem J 364, 89–99 17. Leco, K. J., Apte, S. S., Taniguchi, G. T., Hawkes, S. P., Khokha, R., Schultz, G. A., and Edwards, D. R. (1997) FEBS Lett 401, 213–217 18. Pagenstecher, A., Stalder, A. K., Kincaid, C. L., Shapiro, S. D., and Campbell, I. L. (1998) Am J Pathol 152, 729–741 19. Leco, K. J., Waterhouse, P., Sanchez, O. H., Gowing, K. L., Poole, A. R., Wakeham, A., Mak, T. W., and Khokha, R. (2001) J Clin Invest 108, 817–829 20. Greene, J., Wang, M., Liu, Y. E., Raymond, L. A., Rosen, C., and Shi, Y. E. (1996) J Biol Chem 271, 30375–30380 21. Rahkonen, O. P., Koskivirta, I. M., Oksjoki, S. M., Jokinen, E., and Vuorio, E. I. (2002) Biochim Biophys Acta 1577, 45–52 22. Muir, E. M., Adcock, K. H., Morgenstern, D. A., Clayton, R., von Stillfried, N., Rhodes, K., Ellis, C., Fawcett, J. W., and Rogers, J. H. (2002) Brain Res Mol Brain Res 100, 103–117 23. Vaillant, C., Didier-Bazes, M., Hutter, A., Belin, M. F., and Thomasset, N. (1999) J Neurosci 19, 4994–5004 24. Jaworski, D. M. and Fager, N. (2000) J Neurosci Res 61, 396–408 25. Hashimoto, T., Wen, G., Lawton, M. T., Boudreau, N. J., Bollen, A. W., Yang, G. Y., Barbaro, N. M., Higashida, R. T., Dowd, C. F., Halbach, V. V., and Young, W. L. (2003) Stroke 34, 925–931 26. Chang, C. and Werb, Z. (2001) Trends Cell Biol 11, S37–43 27. Nothnick, W. B. (2000) Biol Reprod 63, 905–912 28. Caterina, J., Caterina, N., Yamada, S., Holmback, K., Longenecker, G., Shi, J., Netzel-Arnett, S., Engler, J., Yermovski, A., Windsor, J., and Birkedal-Hansen, H. (1999) Ann NY Acad Sci 878, 528–530 29. Emmert-Buck, M. R., Emonard, H. P., Corcoran, M. L., Krutzsch, H. C., Foidart, J. M., and Stetler-Stevenson, W. G. (1995) FEBS Lett 364, 28–32 30. Nguyen, M. D., Julien, J. P., and Rivest, S. (2002) Nat Rev Neurosci 3, 216–227 31. Cross, A. H., Trotter, J. L., and Lyons, J. (2001) J Neuroimmunol 112, 1–14 32. McGavern, D. B., Homann, D., and Oldstone, M. B. (2002) J Infect Dis 186 Suppl 2, S145–151 33. Nuttall, R. K., Sampieri, C. L., Pennington, C. J., Gill, S. E., Schultz, G. A., and Edwards, D. R. (2004) FEBS Lett 563, 129–134

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

330

S.J. Crocker & I.L. Campbell

34. Bar-Or, A., Nuttall, R. K., Duddy, M., Alter, A., Kim, H. J., Ifergan, I., Pennington, C. J., Bourgoin, P., Edwards, D. R., and Yong, V. W. (2003) Brain 126, 2738–2749 35. Alter, A., Duddy, M., Hebert, S., Biernacki, K., Prat, A., Antel, J. P., Yong, V. W., Nuttall, R. K., Pennington, C. J., Edwards, D. R., and Bar-Or, A. (2003) J Immunol 170, 4497–4505 36. Cataldo, D., Munaut, C., Noel, A., Frankenne, F., Bartsch, P., Foidart, J. M., and Louis, R. (2001) Allergy 56, 145–151 37. Osman, M., Tortorella, M., Londei, M., and Quaratino, S. (2002) Immunology 105, 73–82 38. Bugno, M., Witek, B., Bereta, J., Bereta, M., Edwards, D. R., and Kordula, T. (1999) FEBS Lett 448, 9–14 39. Galboiz, Y., Shapiro, S., Lahat, N., and Miller, A. (2002) J Neuroimmunol 131, 191–200 40. Crocker, S. J., Milner, R., Pham-Mitchell, N., and Campbell, I. L. (2003) at the Society for Neuroscience Meeting, New Orleans, LA. 41. Teesalu, T., Hinkkanen, A. E., and Vaheri, A. (2001) Am J Pathol 159, 2227–2237 42. Kouwenhoven, M., Ozenci, V., Tjernlund, A., Pashenkov, M., Homman, M., Press, R., and Link, H. (2002) J Neuroimmunol 126, 161–171 43. D’Souza, C. A., Mak, B., and Moscarello, M. A. (2002) J Biol Chem 277, 13589–13596 44. Gijbels, K., Galardy, R. E., and Steinman, L. (1994) J Clin Invest 94, 2177–2182 45. Hewson, A. K., Smith, T., Leonard, J. P., and Cuzner, M. L. (1995) Inﬂamm Res 44, 345–349 46. Cuzner, M. L., Gveric, D., Strand, C., Loughlin, A. J., Paemen, L., Opdenakker, G., and Newcombe, J. (1996) J Neuropathol Exp Neurol 55, 1194–1204 47. Leppert, D., Ford, J., Stabler, G., Grygar, C., Lienert, C., Huber, S., Miller, K. M., Hauser, S. L., and Kappos, L. (1998) Brain 121 (Pt 12), 2327–2334 48. Lindberg, R. L., De Groot, C. J., Montagne, L., Freitag, P., van der Valk, P., Kappos, L., and Leppert, D. (2001) Brain 124, 1743–1753 49. Lee, M. A., Palace, J., Stabler, G., Ford, J., Gearing, A., and Miller, K. (1999) Brain 122, 191–197 50. Waubant, E., Goodkin, D., Bostrum, A., Bacchetti, P., Hietpas, J., Lindberg, R., and Leppert, D. (2003) Neurology 60, 52–57 51. Rivera, S., Ogier, C., Jourquin, J., Timsit, S., Szklarczyk, A. W., Miller, K., Gearing, A. J., Kaczmarek, L., and Khrestchatisky, M. (2002) Eur J Neurosci 15, 19–32 52. Romanic, A. M., White, R. F., Arleth, A. J., Ohlstein, E. H., and Barone, F. C. (1998) Stroke 29, 1020–1030 53. Pulsinelli, W. A., Brierley, J. B., and Plum, F. (1982) Ann Neurol 11, 491–498

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

Tissue Inhibitors of MMPs, Inﬂammation, and the CNS

331

54. Tan, H. K., Heywood, D., Ralph, G. S., Bienemann, A., Baker, A. H., and Uney, J. B. (2003) Mol Cell Neurosci 22, 98–106 55. Ahonen, M., Poukkula, M., Baker, A. H., Kashiwagi, M., Nagase, H., Eriksson, J. E., and Kahari, V. M. (2003) Oncogene 22, 2121–2134 56. Wallace, J. A., Alexander, S., Estrada, E. Y., Hines, C., Cunningham, L. A., and Rosenberg, G. A. (2002) J Cereb Blood Flow Metab 22, 1303–1310 57. Brack, A. and Stein, C. (2003) Exp Neurol 184, 44–49 58. Hickey, W. F. (2001) Glia 36, 118–124 59. Uhm, J. H., Dooley, N. P., Stuve, O., Francis, G. S., Duquette, P., Antel, J. P., and Yong, V. W. (1999) Ann Neurol 46, 319–324 60. Ringheim, G. E. and Conant, K. (2004) J Neuroimmunol 147, 43–49 61. Langford, D. and Masliah, E. (2003) Exp Neurol 184, 553–555 62. Giraudon, P., Szymocha, R., Buart, S., Bernard, A., Cartier, L., Belin, M. F., and Akaoka, H. (2000) J Immunol 164, 2718–2727 63. Mastroianni, C. M., Liuzzi, G. M., D’Ettorre, G., Lichtner, M., Forcina, G., Di Campli, N. F., Riccio, P., and Vullo, V. (2002) HIV Clin Trials 3, 310–315 64. Suryadevara, R., Holter, S., Borgmann, K., Persidsky, R., Labenz-Zink, C., Persidsky, Y., Gendelman, H. E., Wu, L., and Ghorpade, A. (2003) Glia 44, 47–56 65. Wiley, C. A., Schrier, R. D., Nelson, J. A., Lampert, P. W., and Oldstone, M. B. (1986) Proc Natl Acad Sci USA 83, 7089–7093 66. Williams, K., Alvarez, X., and Lackner, A. A. (2001) Glia 36, 156–164 67. Dhawan, S., Weeks, B. S., Soderland, C., Schnaper, H. W., Toro, L. A., Asthana, S. P., Hewlett, I. K., Stetler-Stevenson, W. G., Yamada, S. S., Yamada, K. M., and et al (1995) J Immunol 154, 422–432 68. Eugenin, E. A. and Berman, J. W. (2003) Methods 29, 351–361 69. Argyris, E. G., Acheampong, E., Nunnari, G., Mukhtar, M., Williams, K. J., and Pomerantz, R. J. (2003) J Virol 77, 12140–12151 70. Lafrenie, R. M., Wahl, L. M., Epstein, J. S., Hewlett, I. K., Yamada, K. M., and Dhawan, S. (1996) J Immunol 157, 974–977 71. Liu, Y., Tang, X. P., McArthur, J. C., Scott, J., and Gartner, S. (2000) J Neurovirol 6 Suppl 1, S70–81 72. Roberts, E. S., Zandonatti, M. A., Watry, D. D., Madden, L. J., Henriksen, S. J., Taﬀe, M. A., and Fox, H. S. (2003) Am J Pathol 162, 2041–2057 73. Beuche, W., Yushchenko, M., Mader, M., Maliszewska, M., Felgenhauer, K., and Weber, F. (2000) Neuroreport 11, 3419–3422 74. Kolb, S. A., Lahrtz, F., Paul, R., Leppert, D., Nadal, D., Pﬁster, H. W., and Fontana, A. (1998) J Neuroimmunol 84, 143–150 75. Khuth, S. T., Akaoka, H., Pagenstecher, A., Verlaeten, O., Belin, M. F., Giraudon, P., and Bernard, A. (2001) J Virol 75, 8268–8282 76. Pagenstecher, A., Stalder, A. K., Kincaid, C. L., Volk, B., and Campbell, I. L. (2000) Am J Pathol 157, 197–210 77. Nau, G. J., Richmond, J. F., Schlesinger, A., Jennings, E. G., Lander, E. S., and Young, R. A. (2002) Proc Natl Acad Sci USA 99, 1503–1508

May 26, 2005

16:9 WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System ch13

332

S.J. Crocker & I.L. Campbell

78. Lacroix, S., Feinstein, D., and Rivest, S. (1998) Brain Pathol 8, 625–640 79. Triantaﬁlou, M. and Triantaﬁlou, K. (2002) Trends Immunol 23, 301–304 80. Osiewicz, K., McGarry, M., and Soloway, P. D. (1999) Ann NY Acad Sci 878, 494–496 81. Starckx, S., Van den Steen, P. E., Wuyts, A., Van Damme, J., and Opdenakker, G. (2002) Leuk Lymphoma 43, 233–241 82. Johnston, C., Jiang, W., Chu, T., and Levine, B. (2001) J Virol 75, 10431–10445 83. Zhou, J., Stohlman, S. A., Atkinson, R., Hinton, D. R., and Marten, N. W. (2002) J Virol 76, 7374–7384 84. Flisiak, R., Maxwell, P., Prokopowicz, D., Timms, P. M., and Panasiuk, A. (2002) Hepatogastroenterol 49, 1369–1372 85. Koulentaki, M., Valatas, V., Xidakis, K., Kouroumalis, A., Petinaki, E., Castanas, E., and Kourroumalis, E. (2002) J. Viral Hepat 9, 189–193 86. Leroy, V., Monier, F., Bottari, S., Trocme, C., Sturm, N., Hillert, M. N., Morel, F., and Zarski, J. P. (2004) Am J Gastroenterol 99, 271–279 87. Gewert, D. R., Coulombe, B., Castelino, M., Skup, D., and Williams, B. R. (1987) EMBO J 6, 651–657 88. Kernacki, K. A., Chunta, J. L., Barrett, R. P., and Hazlett, L. D. (2004) Exp Eye Res 78, 1155–1162

May 26, 2005

16:28WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System index

INDEX cAMP response element binding protein (CREB), 28 caspase 9, 26, 227, 234, 235 Cathepsin S and K, 257 CCAAT, 32, 33, 37 CD44, 95, 108 Cdc42, 121 cell adhesion, 154, 164, 166, 174, 179, 180 cell adhesion molecule (CAM), L1, 162 cholesterol diet, 249 chromatin, 25, 33 CNTF, 42, 47 collagen type IV, 90–92 collagen XVIII, 92 conditional knockout, 165, 171 crovidisin, 137 cryptic site, 88, 89 CXCR4, 76, 78 cyclooxygenase, 20 cysteine proteases, 227, 234 cysteine thiol group, 67 cytochrome c, 235

ACK, 121 adamalyins (ADAMs), 10, 87 ADAMTS, 10, 13, 87 age-related macular degeneration, 213 aggrecan, 94–101, 103 aggretin, 136 agrin, 92, 119, 124, 125 alcohol-related dementia, 280 Alzheimer’s disease (AD), 216 β-amyloid, 92, 102, 108, 110 amyotrophic lateral sclerosis (ALS), 216 androgens, 21 anoikis, 80, 81, 120–122, 134 AP-1, 25, 27, 30–34, 36–38, 45, 48, 50, 53 AP-2, 30, 31, 33, 34, 37, 47 apolipoprotein-E, 253 apoptosis, 67, 69, 79–81, 108 apoptosome, 235 apoptotic protease activating factor-1, 235 arachidonic acid, 20 ARK5, 26 astrocytomas, 266, 267 axon, 153, 168, 169, 171–179, 182 axon guidance, 154, 156, 168, 169, 172, 174, 175, 177–179, 182 axonal guidance, 97, 104

D-type motoneurons, 155, 174, 175 decysin ADAMDEC 1, 35 deleted in colorectal cancer (DCC), 171, 125, 132 delta, 158, 159, 162, 178 demyelinating injury, 196 dendrites, 153, 156, 163, 168, 181, 182 DGEA, 135 disintegrin domain, 154, 165, 180 distal tip cell (DTC), 132, 165 dorsal root ganglia (DRG), 154, 155, 181 Drosophila, 157–160, 162, 169, 172, 173, 175

Bad, 26 basal lamina, 88–90, 92, 106, 107 Bax, 235 blood brain barrier, 12 brevican, 10, 94–102, 215 bromodeoxyuridine (BrdU), 168 cadherins, 120–122 cAMP, 20, 23, 27, 28, 32, 33, 39, 47 333

May 26, 2005

16:28WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System index

334

α-dystroglycan, 107 β-dystroglycan, 107 dystroglycan, 106, 107 eNOS, 26 echistatin, 136, 141 EGF, 20, 23, 36, 39, 48 endostatin, 123, 140 endothelin-1, 128 endothelin-converting enzyme, 128 engrailed, 172 entactin, 90, 92, 93 Eph receptor, 169, 170 ephrins, 169, 178, 179 ErbB, 181 estrogen, 21, 43, 52 even-skipped, 172 exosites, 110 experimental autoimmune encephalomyelitis (EAE), 191 experimental autoimmune neuritis (EAN), 198 Fas, 122, 131, 132 Fas-ligand, 164, 178 FGF, 45, 92 ﬁbroblast growth factor receptors (FGFRs), 178 ﬁbronectins, 90, 92, 96, 104–106 focal adhesion kinase (FAK), 120, 137 forkhead transcription factors, 26 furin, 101 furin-like protease, 101 G-protein coupled receptors, 180 gelatinases, 6 glial hyaluronic acid binding protein (GHAP), 99 glial scar, 10, 11 gliomas, 263–271 gliosis, 97 glucocorticoids, 21, 23, 24, 30, 36, 42 glycogen synthase kinase (GSK)-3β, 26 glycosylphosphatidylinositol (GPI) anchor, 228

Index

GON-1, 163, 165, 177 granule cells, 155, 157, 163, 164, 168 growth cones, 168–171, 178 Guillain-Barr´e syndrome, 198 hairy enhancer of split-5 (hes-5), 161 halydin, 137 HB-EGF, 133 hemopexin-like domain, 4 heparan sulfate proteoglycans, 90–92, 109 heparin binding-EGF (HB-EGF), 179 heregulin, 181 hexabrachion, 104 high density lipoprotein 3 (HDL3), 255 high fat, 249 hinge region, 4, 5 HIV-associated dementia, 67, 75, 76, 78, 81 HIV-dementia, 281 human T cell lymphotropic virus type 1 (HTLV-I), 21 human immunodeﬁciency virus HIV-1 glycoprotein 120, 21 Huntington’s disease (HD), 216 hyalectans, 94–99, 101–105 hydroxamate inhibitors, 156 IFGBPs, 190, 193, 194 IFN, 23, 25, 28, 42, 43, 47, 52 IGF-1, 27 αv β3 integrin, 48 integrin receptors, 154, 179 integrin-linked kinase (ILK), 120 integrins, 120, 121, 130, 133–136, 139–141 interleukin 1α, 20 interleukin 1β, 20 ischemia/reperfusion, 67, 69, 74, 78 jararhagin, 137 jarastatin, 137 jerdonin, 136 JNK, 24, 30, 36, 47

May 26, 2005

16:28WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System index

Index

Kainate (KA), 211 kallikreins, 127 kininogens, 127 kinins, 127 kistrin, 136 kuzbanian (kuz), 158 L-selectin, 239 laminin, 90–92, 107, 179 laminin-1, 91 laminin-5, 91 LDL, 249, 252, 253 LDL receptor, 92 LDL-receptor deﬁcient mouse, 253, 257 Lewy bodies, 280, 281, 291 link proteins, 95, 98 lipid-laden macrophages, 249, 252, 254 lipopolysaccharide, 21, 34, 42, 43 lipoxygenase, 20 long-term potentiation, 105 α2 -macroglobulin, 9, 100 macrophages, 191, 192, 198, 249–254, 256 MALDI-TOF analysis, 71, 73, 75 malignant brain tumours, 264, 266 MAPK, 22, 24, 26–28, 30, 36, 45, 47 MCPs, 12 medulloblastomas, 266 metastasise, 263, 264 microglia, 192, 198 MIG-17, 163, 171, 172, 177 MMP-1, 5, 24, 135, 198, 250–254 MMP-2, 91, 92, 101, 106, 109, 110 MMP-3, 91, 92, 106, 108 MMP-7, 91, 92, 106, 108, 109 MMP-8, 5, 6 MMP-9, 91, 108–110 MMP-10, 5, 6 MMP-11, 91 MMP-12, 91, 92 MMP-13, 91, 92 MMP-14, 91, 92 MMP-15, 91, 92

335

MMP-16, 91 MMP-24 (MT5-MMP), 40 MMP-25, 91 MS, 191, 194, 196–199 MT-MMP, 229, 230, 233 MT-MMPs, 5, 6, 8, 91 myelin basic protein (MBP), 80 myelination, 97, 104, 105, 156, 164, 182 N -acetyl-cysteine, 252 N -terminal catalytic domain, 4 N-terminal G1-globular domain, 99 neoepitope antisera, 103 neprilysin, 110 nerve growth factor response element (NGFRE), 31, 32 netrin, 9, 171, 172, 175, 178, 180 neu diﬀerentiation factor, 181 neural cell adhesion molecule (N-CAM), 92, 105 neural crest cells, 155, 156, 165, 167 neuregulin-2 (NRG-2), 181 neurite outgrowth, 93, 101, 105 neurocan, 10, 94, 96–99, 101, 105 neuroﬁbrillary tangles, 92 neurogenesis, 104, 105 neuronal migration, 104 neurotransmitters, 21 NF-Y , 32, 38 NG2, 94, 109 NG2 chondroitin sulfate proteoglycan, 196, 197 NGF, 20, 45 nidogen, 11, 90, 92 Nitric oxide (NO), 67, 69 Notch, 158, 159, 161, 162, 178 Notch/Jagged, 190 NRGs, 181 NSAID, 23 occludin, 12 oligodendrocyte, 189 oligodendrocyte precursor cells (OPCs), 189

May 26, 2005

16:28WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System index

336

Index

oncogenes, 21 optic nerve, 191, 193

RGD sites, 11 rhodocetin, 136

p53, 23, 31, 33, 34, 47, 48 p75 receptor, 108 Parkinson’s disease (PD), 216 PDGF, 92 PDGF-BB, 11 PEA-3/ETS, 30, 32, 37, 38, 45 peptide mass ﬁngerprinting, 71, 72 perineuronal (pericellular) nets, 88, 93, 94, 97, 98, 104, 105 perlecan, 90–92, 107 phorbol esters, 21, 24, 30 phosphacan, 94 PI3K/Akt, 26, 27 plasmin, 227, 229, 230 plasminogen, 227, 229, 231 plasminogen activators, 102, 103 plasticity, 89, 98, 101, 102, 106, 110 polydendrocyte, 109 polymorphism, 25, 51, 52 PPAR gamma, 20 pro-peptide domain, 4, 5 procaspase-9, 235 progesterone, 21 prostaglandins, 20 protein kinase C, 20, 36, 47 proteinase activated receptors (PARs), 134 proteoglycans, 10 Purkinje cell, 155, 156, 163, 168, 181

S-nitrosylated proteins, 69 salmosin, 136 Schwann cells, 41, 44, 49 SDF-1, 76–78 α secretase, 108 β-secretase, 108 γ secretase, 108 secreted member of the ADAMTS (ADAMs with thrombospondin-1 repeats) family, GON-1, 165 semaphorins 7A, 141 serotonin, 21, 22 shedding of syndecan-1, 109 SHPS-1, 133 signal peptide, 4, 5 Slit receptor family, Roundabout (robo), 174 Slit/Robo signalling, 174 Smad, 26, 27, 38 snake venom metalloproteinases, 135 Sorby’s fundus dystrophy (SFD), 213 SPARC (secreted protein acidic and rich in cysteine), 90 ST3β, 34 STATs, 28 stroke, 67, 73, 75, 78–81 substance P, 110 subventricular zone (SVZ), 209 synantocyte, 109 synapsin gene, 212 synaptic plasticity, 93, 101, 105 syndecan-2, 109 syndecan-4, 109

reactive nitrogen species, 20 reactive oxygen species, 20, 48 redox state, 69 reelin, 140, 141 retinal ganglion cell (RGC) axons, 169 retinoic acid receptors, 34 retinoids, 20, 23, 24, 27, 30, 36 reversion-inducing cysteine-rich protein with kazal motifs (RECK), 9

T -lymphocytes, 36 tenascin-C, 96, 104–106 tenascin-R, 96, 98, 104, 105 tenascin-W, 104 tenascin-X, 104 tetracycline, 21 TGF-α, 10 the low aﬃnity neurotrophin receptor, p75NTR , 164

May 26, 2005

16:28WSPC/Book-SPI-B282 (9in x 6in) Matrix Metalloproteinases in the Central Nervous System index

Index

337

thrombospondin motifs, 88 thrombospondin type I repeats, 100 thrombospondins, 9, 123, 140 thyroid hormone, 21 TIMPs, 3, 4, 8, 9, 36–39, 207–217, 239–243, 253, 311–328 TNF-α, 10 triple helical molecules, 90 trkA neurotrophin receptor, 127, 178, 180 β-tubulin gene (N -tubulin), 160

vascular dementia, 280, 283, 292 VEGF, 11 versican, 10, 94–101 vonWillebrand factor, 100

UNC-71, 154, 155, 174, 175, 182 uPA, 134 uPAR, 134

ZO-1, 80 zonae occludens-1 (ZO-1), 80 zymogen, 101

wobbler mutant mice, 215 Xenopus, 155, 160, 161, 166, 167, 175–179 YB-1, 33, 34, 48