Biochemical Characterization and Zinc Binding Group (Zbgs) Inhibition Studies on the Catalytic Domains of Mmp7 (Cdmmp7) and Mmp16 (Cdmmp16)

Home , ADAM28, Interstitial collagenase, Macrophage elastase, MMP1, MMP11, MMP14

MIAMI UNIVERSITY The Graduate School

Certificate for Approving the Dissertation

We hereby approve the Dissertation

Fan Meng

Candidate for the Degree

DOCTOR OF PHILOSOPHY

______

Director Dr. Michael W. Crowder ______

Dr. David L. Tierney ______

Dr. Carole Dabney-Smith ______

Dr. Christopher A. Makaroff ______

Graduate School Representative Dr. Hai-Fei Shi

ABSTRACT BIOCHEMICAL CHARACTERIZATION AND ZINC BINDING GROUP (ZBGS) INHIBITION STUDIES ON THE CATALYTIC DOMAINS OF MMP7 (CDMMP7) AND MMP16 (CDMMP16)

by Fan Meng

Matrix metalloproteinase 7 (MMP7/matrilysin-1) and membrane type matrix metalloproteinase 16 (MMP16/MT3-MMP) have been implicated in the progression of pathological events, such as cancer and inflammatory diseases; therefore, these two MMPs are considered as viable drug targets. In this work, we (a) provide a review of the role(s) of MMPs in biology and of the previous efforts to target MMPs as therapeutics (Chapter 1), (b) describe our efforts at over-expression, purification, and characterization of the catalytic domains of MMP7 (cdMMP7) and MMP16 (cdMMP16) (Chapters 2 and 3), (c) present our efforts at the preparation and initial spectroscopic characterization of Co(II)-substituted analogs of cdMMP7 and cdMMP16 (Chapters 2 and 3), (d) present inhibition data on cdMMP7 and cdMMP16 using zinc binding groups (ZBG) as potential scaffolds for future inhibitors (Chapter 3), and (e) summarize our data in the context of previous results and suggest future directions (Chapter 4). The work described in this dissertation integrates biochemical (kinetic assays, inhibition studies, limited computational methods), spectroscopic (CD, UV-Vis, 1H-NMR, fluorescence, and EXAFS), and analytical (MALDI-TOF mass spectrometry, isothermal calorimetry) methods to provide a detailed structural and mechanistic view of these MMPs. This work is part of overall effort to prepare selective and specific inhibitors against the MMPs, which are expected to be drug candidates in MMP-associated diseases.

BIOCHEMICAL CHARACTERIZATION AND ZINC BINDING GROUP (ZBGS) INHIBITION STUDIES ON THE CATALYTIC DOMAINS OF MMP7 (CDMMP7) AND MMP16 (CDMMP16)

A DISSERTATION

Presented to the Faculty of

Miami University in partial

fulfillment of the requirements

for the degree of

Doctor of Philosophy

Department of Chemistry and Biochemistry

Fan Meng

The Graduate School Miami University Oxford, Ohio

2016

Dissertation Director: Dr. Michael W. Crowder

Fan Meng

2016

Table of Contents

Chapter 1 Introduction: What are matrix metalloproteinases? 1

1.1 Classification of matrix metalloproteinases (MMPs) 2

1.2 Structural features and the catalytic mechanism of MMPs 3

1.2.1. Overall structure of MMPs. 3 1.2.2 Catalytic domains of MMPs. 4 1.2.3 Proposed reaction mechanisms of the MMPs. 6 1.3 Regulation of MMPs 7

1.3.1 Transcriptional regulation. 7 1.3.2 Cell-specific expression of MMPs. 8 1.3.3 Pre-translational regulation of the expression of MMPs 9 1.3.4 Regulation of MMP enzymatic activity 9 1.3.5 Unregulated enzymatic activities of MMPs contribute to multiple pathological processes 11 1.4 Role of MMPs in physiological and pathological conditions 12

1.4.1 Role of MMPs in ECM biology 12 1.4.2 MMPs regulate pathological and physiological events by processing signaling proteins 13 1.4.3 Protective role of MMPs in pathological processes 17 1.5 Review of MMP inhibitors 17

1.5.1 Therapeutic targeting of MMPs 17 1.5.2 Overview of ZBG-based MMPi 19 1.5.3 Challenges and new opportunities for next generation ZBG MMPi 20 1.6 Inhibition studies on MMP7 and MMP16 22

1.7 References 24

1.8 Tables and figures 46

iii

Chapter 2 Biochemical and spectroscopic characterization of the catalytic domain of MMP16 (cdMMP16) 60

2.1 Introduction 62

2.2 Materials and methods 65

2.3 Results 72

2.4 Discussion 78

2.5 References 82

2.6 Tables and figures 89

Chapter 3 Biochemical characterization and zinc binding group (ZBGs) inhibition studies on the catalytic domain of MMP7 (cdMMP7) 113

3.1 Introduction 115

3.2 Material and methods 118

3.3 Results 124

3.4 Discussion 128

3.5 Acknowledgements 134

3.6 Reference 135

3.7 Tables and figures 142

Chapter 4 Conclusions of dissertation 161

4.1 Conclusions 162

4.2 References 168

4.3 Table and figure 172

List of Tables

Table 1.1: Overview of MMP sub-classes, structural elements, and ECM substrates

Table 1.2: Overview of cell-specific expression of MMPs under normal physiological conditions 48

Table 1.3: Physiological and pathological events and identified substrates for select MMPs. 49

Table 1.4: Classes of MMP inhibitors (MMPi) 51

Table 2.1: Data fit for EXAFS traces of Zn2-cdMMP16. 89

Table 2.2: Data fit for EXAFS traces of Co2-cdMMP16. 90

Table 2.3: Steady state kinetic constants and metal content of cdMMP16 samples 91

Table 2.4: Kinetic mechanism used to fit stopped-flow fluorescence data and the Dynafit-generated microscopic rate constants. 92

Table 2.5: Secondary structural elements of Zn2-cdMMP16, Co2-cdMMP16, ZnCo-cdMMP16, and metal-free cdMMP16. 93

Table 2.6:Summary of EXAFS fits of Zn2-cdMMP16 and Co2-substituted analogs 94

Table 3.1: Steady-state kinetic constants and metal content for recombinant cdMMP7 analogs 142

Table 3.2: Secondary structural content of Zn2-cdMMP7 and ZnCo-cdMMP7 analogs 143

Table 3.3: Microscopic kinetic constants of the hydrolysis of BML-P131 by

Zn2-cdMMP7 144

Table 3.4: Microscopic kinetic constants of as-isolated Zn2-cdMMP7 and

Zn2-cdMMP16 from substrate-emission stopped-flow studies 145

Table 3.5: IC50 values of AHA, maltol, TM, and ATM as inhibitors of cdMMP16 and cdMMP7. 146

Table 3.6: ITC measurements on cdMMP7 and cdMMP16 using ATM and TM as binding groups. 147

Table 4.1: Select inhibitors of MMPs with reported IC50 values 172

List of Figures Figure 1.1: Domain structure of represented MMPs. 52

Figure 1.2: Activation of pro-MMPs via cysteine switch 53

Figure 1.3: Overlapped crystal structures of cdMMP7 (green, PDB#1MMP) and cdMMP16 (cyan, PDB#1RMB). 54

Figure 1.4: Proposed reaction mechanisms for MMPs 55

Figure 1.5: Transcriptional, translational, and physiological regulation of MMPs 56

Figure 1.6: Overview of MMPs’ physiological roles, pathological roles and protective roles 57

Figure 1.7: Review of MMPi development and current strategies of therapeutic targeting MMPs 58

Figure 1.8: Overview of experimental approach used in this dissertation. 59

Figure 2.1:Crystal structure of Zn2-cdMMP16 using the coordinates (PDB 1RM8) 95

Figure 2.2: EXAFS of Zn2-cdMMP16 96

Figure 2.3: EXAFS of Co2-cdMMP16 97

Figure 2.4: SDS-PAGE analysis of cdMMP16 purification 98

Figure 2.5: Substrates used in this study 99

Figure 2.6: Relative catalytic activity of Zn2-cdMMP16 in the presence of NaCl 100

Figure 2.7: Circular dichroism of as-isolated Zn2-cdMMP16 (solid), Reconstituted

Zn2-cdMMP16 (tight-dot), Co2-cdMMP16 (dash), ZnCo-cdMMP16 (dash dot), and metal-free cdMMP16 (dot) 101

Figure 2.8: Stopped-flow fluorescence traces of 1 M Zn2-cdMMP16 with various concentrations of DNP-Pro-Leu-Ala-Leu-Trp-Ala-Arg-OH 102

Figure 2.9: Stability of cdMMP16 samples using SDS-PAGE 103

Figure 2.10: Relative catalytic activity of metal-free cdMMP16 104

Figure 2.11: MALDI-TOF mass spectra of cdMMP16 in the presence of 2 eq of Co(II) and 5 mM Ca(II) over time 105

vii

Figure 2.12: SDS-PAGE of cdMMP16 samples using modified metal incorporation procedure. 106

Figure 2.13: Fluorescence spectra of as-isolated Zn2-cdMMP16 (solid), reconstituted

Zn2-cdMMP16 (tight dot) 107

Figure 2.14: Fourier transformed EXAFS data (solid lines) and corresponding best fits (open symbols) for Zn2-cdMMP16 and Co2-cdMMP16 108

Figure 2.15: Optical spectra of Co2-cdMMP16 (dash) and ZnCo-cdMMP16 (solid).

Both of samples were diluted in 50 mM Hepes, pH 7.0, containing 5 mM CaCl2 109

1 Figure 2.16: 200 MHz H NMR spectra of Co2-cdMMP16 (top) and ZnCo-cdMMP16 (below). Solvent exchangeable protons are marked with asterisks. 110

Figure 2.17: Fluorescence spectra of 10 M Zn2-cdMMP16 exposed in differential salt concentration 111

1 Figure 2.18: H NMR spectra of reconstituted Zn2-cdMMP16 and as-isolated

Zn2-cdMMP16. 112

Figure 3.1: Crystal structure of Zn2-cdMMP7 (PDB #1MMB) 148

Figure 3.2: Structure of substrate 149

Figure 3.3: SDS-PAGE analysis of the two cdMMP7 refolding procedures. 150

Figure 3.4: Circular dichroism spectra of Zn2-cdMMP7 (solid), ZnCo-cdMMP7

(dashed line), and inactive Zn2-cdMMP7 (dotted line) 151

Figure 3.5: Fluorescence emission spectra of ZnCo-cdMMP7 (dash) and as-isolated

Zn2-cdMMP7 (solid). 152

Figure 3.6: Fluorescence emission spectra of inactive Zn2-cdMMP7 (dash) and as-isolated, Zn2-cdMMP7 (solid). 153

Figure 3.7: Stopped-flow fluorescence traces of reactions of Zn2-cdMMP7 and with substrate BML-P131 154

Figure 3.8: Stopped-flow fluorescence traces of reactions of A Zn2-cdMMP7 and B

Zn2-cdMMP16 with FS-6 fluorescent substrate 155

viii

Figure 3.9: Uv-vis spectrum of of the ZnCo-cdMMP7 analog( right top corner) and UV-Vis difference (spectrum of ZnCo analog minus the spectrum of metal-free enzyme) spectrum 156

Figure 3.10: Structures of zinc binding group (ZBG) inhibitors used in this study 157

Figure 3.11: Inhibitory potency of AHA, maltol, TM, and ATM against cdMMP7 and cdMMP16 relative to cdMMP1. 158

Figure 3.12: ITC results of ZBG binding to cdMMP7 and cdMMP16 159

Figure 3.13: Representative binding modes from docking simulations 160

Figure 4.1: Crystal structures of (A) cdMMP1 (yellow, PDB#966C), (B) cdMMP7 (green, PDB#1MMP), (C) cdMMP9(pink, PDB#4H1Q), (D) cdMMP12 (blue,PDB#2POJ), and (E) cdMMP16 (cyan, PDB#1RM8). 175

ACKNOWLEDGEMENTS

I would like to thank my research director, Dr. Michael W. Crowder, for guiding me on each step of my research and academics at Miami University. He provided support and advice in difficult times and allowed me to progress in my research. Dr. Crowder is an excellent role model of professionalism, dedication, and diligence. Meanwhile, I’d like thank Dr. David L. Tierney and Dr. Rick Page for generously sharing their knowledge with me in the fields of biophysical and bioinorganic chemistries. Both of them have broadened my vision in research. I’d like to thank all faculty, technicians, and staff for their help and support during my time at Miami University. I would like to thank my family for their unconditional support and love. I can achieve nothing without my family’s support. Specifically, I’d like to thank my uncle, Dr. Xiao Xu, who inspired my interest in bioscience and provides me with endless support, guidance, and love while I live in the United States. Lastly, I would like to thank my girlfriend, Miss Yue Wang, for her support in this long journey. Meanwhile, I would like to thank all of my friends and colleagues at Miami University. Bonds with these people are, and always will be a valuable part in my life.

Chapter 1

Introduction: What are matrix metalloproteinases?

Fan Meng

Department of Chemistry and Biochemistry, Miami University, Oxford, Ohio 45056

*The author thanks Dr. Michael W. Crowder for his significant contribution to this chapter

Introduction

1.1 Classification of matrix metalloproteinases (MMPs)

Matrix metalloproteinases (MMPs) are Zn(II)-dependent endopeptidases, belonging to the metzincin superfamily of metalloproteinases [1]. The first MMP activity was observed in tadpole tail metamorphosis, and an enzyme exhibiting MMP activity was purified from human cells in 1968 by J. Gross and his colleagues [2, 3]. Due to its collagenase activity, the first MMP was named interstitial collagenase, which was later renamed MMP1 [3, 4]. Currently, there have been 23 MMPs identified in human [4, 5]. All of these MMPs appear to be involved with the degradation of extracellular matrix (ECM) components [4, 6] (Table 1.1). MMPs were categorized into 7 functional sub-classes, based on the preferred physiological substrate [5]. Other classification schemes have been suggested, such as a scheme that categorizes the MMPs based on domain structure [7]. Although MMPs can hydrolyze a number of common biological substrates, MMPs from each sub-class preferentially hydrolyze distinct ECM components (Table 1.1). The enzymes in the gelatinase sub-class, which includes MMP2 and MMP9, preferentially degrade insoluble elastin and gelatin [8, 9]. The enzymes in the collagenase sub-class, which includes MMP1, MMP8, MMP13, and MMP18, degrade interstitial type I, type II, and type III collagen into 1/4 and 3/4 fragments [5, 10]. The enzymes in the stromelysin sub-class, which include MMP3, MMP10, and MMP11, preferentially hydrolyze proteoglycan and fibronectin [5, 11, 12]. The enzymes in the matrilysin sub-class, which includes MMP7 and MMP26, degrade the B chain of insulin that is trapped in the ECM at two sites (Ala14-Leu, Tyr16-Leu) [13, 14]. The metalloelastase sub-class (a.k.a, macrophage elastase), which includes MMP12 and MMP19 (also known as rheumatoid arthritis synovial inflammation (RASI)-MMP), degrades soluble and insoluble elastin and components of basement members[15, 16]. The enamelysin sub-class, which includes MMP20 and MMP21, cleaves amelogenin,

ameloblastin, and enamelin, regulating dentin mineralization [5, 17]. MMP21’s role in dentinal biology is still unclear[18]. The epilysin sub-class, which contains only

MMP28, does not have an identified ECM substrate to date [19]. In addition to the 16 secreted MMPs discussed above, there are 7 membrane-bound MMPs (MT-MMPs), and these MMPs have been categorized based on differences in their membrane-associated C-termini (Table 1.1) [5, 20]. Type-I transmembrane (TM) MMPs (MMP14, MMP15, MMP16, and MMP24) contain a single transmembrane domain and a short cytoplasmic domain.

Glycosylphosphatidylinositol (GPI)-anchored MMPs (MMP17 and MMP25) attach to the membrane using GPI[21]. Transmembrane cysteine array (Cys) MMP (MMP23) does not contain the cysteine switch motif and appears to contain a type-II transmembrane segment in the N-terminal pro-domain [22]. Like secreted MMPs, the MT-MMPs can degrade components of the ECM[5, 20]: MMP14 (MT1-MMP), MMP15 (MT2-MMP), MMP16 (MT3-MMP) , and MMP25 (MT6-MMP) hydrolyze type I - III fibrillar collagens[5, 20]. MMP17 (MT4-MMP), MMP24 (MT5-MMP), and MMP23 (Cs-MMP) rapidly degrade gelatin and fibronectin [20-22]. In comparison to secreted MMPs, the MT-MMPs exhibit enhanced pericellular proteolytic activities due to their association with cellular membranes [20].

1.2 Structural features and the catalytic mechanism of MMPs

1.2.1. Overall structure of MMPs. Most MMPs contain three functional domains: an ~ 80 amino acid pro-domain, a ~ 180 – 200 amino acid catalytic domain, and a ~ 200 amino acid hemopexin-like domain (Figure 1.1) [10]. For the MT-MMPs, a membrane-anchoring domain(s) is added to the C-terminus of the hemopexin-like domain[5, 6]. The pro-domain contains a cysteine (called CS motif in Figure 1.1) in a highly-conserved PRCGxPD region (called the “bait region”) [10]. In most secreted MMPs, this cysteine coordinates the Zn(II) in the active site of the catalytic domain,

preventing substrate from coordinating the Zn(II) and resulting in a catalytically-inactive enzyme [10, 23]. Activation of MMPs from latent zymogens is commonly known as the “cysteine switch” (Figure 1.2) [23]. In the mechanism of activation, [24], the latent pro-MMPs are first transported into the extracellular space. The pro-domain of the zymogen is cleaved by endoproteinases, such as serine proteinases or other MMPs (MMP3 or MMP14) [24]. For MT-MMPs MMP-11, MMP-21, and MMP-28, a furin-recognized sequence RX[R/K]R is located in the

C-terminus of the pro-domain [5, 20]. This furin sequence can be cleaved by pro-protein convertases (PCs) in the endoplasmic reticulum (ER) to activate MT-MMPs intracellularly [20, 25]. With the exception of MMP7, MMP26, and MMP28, MMPs possess a hemopexin-like (Hpx) domain located in C-termini [5]. The Hpx domain was primarily contributes to intermolecular (protein-protein) interactions between MMPs and their target substrates or inhibitors [26]. Although the specific mechanism is unknown, the Hpx domain is essential for the unwinding of the triple helices in collagens, for the binding of MMPs to collagen, and for the hydrolysis of collagen [27]. The Hpx domain also has a role in the activation of latent MMPs [20, 26]. The hemopexin domains of pro-MMP2 and pro-MMP9 interact with the C-terminus (non-inhibitory domain) of tissue inhibitor of metalloproteinase (TIMP) to form a TIMP-proMMP complex [28]. This complex interacts with active MMP14 (MT1-MMP), which hydrolyzes the pro-domain and activates the enzyme [29, 30]. TIMP1, TIMP3, and TIMP4 can also form similar TIMP-proMMP complexes [31]; however, the biological function of these TIMP-proMMP complexes is not known [20].

1.2.2 Catalytic domains of MMPs. The catalytic domains of the MMPs share a similar spherical structure, consisting of three -helices, four to six -sheets, and random loops (Figure 1.3) [10]. Each catalytic domain binds 2 equivalents of Zn(II) and 2-3 equivalents of Ca(II) [4] . One Zn(II) is bound in a buried site by three 4

conserved His residues and one Asp[4]. Along with the Ca(II)-binding sites, this Zn(II) maintains the structure of the catalytic domain and has been called the structural Zn site[4, 32]. The other Zn(II) is coordinated by three His residues, which are part of the conserved HExxHxxGxxH motif [33]. In addition to the 3 His ligands, this Zn(II) is coordinated by 1-2 water/hydroxide ligands, resulting in a tetrahedral geometry[23]. This Zn(II) binding site is called the catalytic Zn site, and substrate binds to this site during catalysis. Adjacent to the catalytic Zn site are a number of sub-pockets (called S1-S3 and S1’–S3’), and these pockets vary in depth, charge, and hydrophobicity[34]. These pockets afford MMPs substrate/inhibitor selectivities[34, 35]. A conserved methionine is present ~ 8 amino acids from the catalytic Zn site, and this methionine is part of the “Met-turn”, which is conserved in all MMPs belonging to the metazincin family and is an essential structural feature for the MMPs [4]. The Met-turn acts as a

“plug” to stabilize the Zn(II)-binding motif for metal ion coordination [36]. The deletion of the Met-turn resulted in a protein with reduced Zn(II) binding and impaired catalytic activity [36]. All MMPs also contain a 9-12 amino acid loop (called specificity loop) in the C-terminus, which is thought to contribute to selectivity in substrate binding [4] (Figure 1.3). Although all MMPs share similar structures of their catalytic domains, different MMPs contain unique elements in these domains. For example, the gelatinases (MMP2 and MMP9) contain two or three fibronectin type II (FN-II) repeats, which are about 58 amino acids in length, within their catalytic domains [37]. The FN-II repeats, which consist of -sheet pairs, form a hydrophobic pocket that interacts with gelatin and elastin [37-39]. In the MT-MMPs, a distinct 8 amino acid sequence is inserted between -sheet II and -sheet III, which forms a loop called the

“MT-loop” [20, 40]. The MT-loop interacts with MMP14, which hydrolyzes the pro-domain from the TIMP2-MMP2 complex and activates the enzyme[40]. MT-loops are also known to localize MT-MMPs to focal adhesion sites to allow for matrix degradation and/or other pericellular activities[41]. The specific mechanism of how the MT-loop localizes MT-MMPs at focal adhesion sites is unknown[20, 41]. 5

1.2.3 Proposed reaction mechanisms of the MMPs. Compared to the enormous efforts to study the structural, biological, and inhibition properties of MMPs, very little information has been reported on the reaction mechanism of these enzymes. To date, there have been four reaction mechanisms proposed for the MMPs (Figure 1.4). Firstly, a base-catalyzed mechanism was proposed by Browner et al. after the authors solved the crystal structure of a MMP7-inhibitor complex[42]. The mechanism showed a Glu-participated formation of Zn(II)-OH, which nucleophilically attacks the substrate’s scissile bond. Secondly, a mechanism with the active site Glu serving as the nucleophile was proposed by Manzetti et al. using modeling studies on MMP3[43]. This mechanism was unique in the fact that the active site Zn(II) is not involved in the creation of the nucleophile. This mechanism also proposed that coordination number of Zn(II) switches between 4 and 5, due to the shifting of Zn(II) binding to His or Glu during the course of the reaction. Thirdly, a base-catalyzed mechanism was proposed by Bertini et al. using crystallographic “snapshots” of MMP12 during catalysis[44]. Different from the Manzetti mechanism[43], Bertini proposed that the Zn(II) coordination changes during catalysis are associated with changes in the number of bound water molecules: catalytic Zn(II) is 6-coordinate (with 3 waters bound) in the resting state, 5-coordinate in a gem-diol bound state (no water bound), 5-coordinate after peptide bond cleavage with one product bound (1 water bound), and 4-coordinate in the final product bound form (not shown in Figure 1.4). Lastly, a base-catalyzed mechanism was proposed by Pelmenschikov and Siegbahn using a two-layered ONIOM study[45]. This proposed mechanism relied heavily on previous mechanistic studies on thermolysin and carboxypeptidase [45]. This mechanism showed that Zn(II) is 4-coordinate in the resting form and 5-coordinate in all other steps. Details in each mechanism vary; however, these mechanism could be differentiated by use of rapid-freeze quench spectroscopic studies as long as suitable Co(II)-substitituted analogs can be prepared.

1.3 Regulation of MMPs

1.3.1 Transcriptional regulation. The expression of MMPs is regulated at the transcriptional and post-transcriptional levels[46]. The promoters of most MMPs often contain one TATA box (~30 bp) and two cis-regulatory elements, one activator protein 1 (AP-1) binding site (~70 bp), and an adjacent polyoma enhancer activator 3 (PEA-3) binding site [47]. AP-1/PEA-3 sites are not found in the promotor regions of MMP8, MMP11, and MMP21, which suggests a simpler transcriptional regulation pattern for these three MMPs [46]. The expression of most MMPs is induced by the binding of ECM proteins (collagen, fibronectin, or fibrinogen) with integrin transmembrane receptors [48], and this binding activates AP-1/PEA-3 expression via mitogen-activated protein kinases (MAPKs)/focal adhesion kinase (FAK) signaling pathways [48, 49]. In addition to AP-1/PEA-3, additional cis-transcription elements and signaling pathways are involved in the expression of several MMPs [46]. For MMP7, MMP12, MMP14 and MMP26, Tcf-4, a -catenin binding site, is present in the promoter region, and this expression is under control of a Wnt-signaling pathway[50]. For the expression of MMP9, the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kB) is required for transcription [51, 52], which indicates that the expression of MMP9 is under control of the inflammatory cytokine-activated pathways[53, 54] (Figure 1.5). Epigenetic regulation also appears to influence the expression of MMPs. The expression of MMP3 appears to be up-regulated in the absence of DNA methyltransferase I / 3b (Dmt-1/ Dmt 3b) in a colon cancer cell line (HCT) [55], which implies that DNA methylation of the MMP3 promoter down-regulates the expression of MMP3[55]. Interestingly, the expression of several MMPs, such as MMP1 and MMP2, was not regulated by Dmt in HCT cell lines[46, 55], possibly due to a lack of CpG islands on the promoters of these MMPs[46]. Chromatin-remodeling via histone modification is another epigenetic mechanism regulating the expression of certain MMPs[46]. Metastasis-associated protein 1 (MTA1) suppresses MMPs’

genetic expression by recruiting histone deacetylase 2/5/7 (HDAC 2/5/7) to the nucleosome remodeling deacetylase (NuRD) complex[56]. The NuRD complex deacetylates histone 3 / histone 4(H3/H4), which results in formation of condensed chromatin to halt genetic expression of MMP9 [57]. On the other hand, the expression of MMP9 and MMP1 was promoted by recruiting CREB-binding protein (CBP) histone acetyltransferase[58], ATP-dependent helicase 1 (Brg1) [59], and barrier-to-autointegration factor (BAF60e) to the Swi/SNF complex[46, 60]. The Swi/SNF complex “unpacks” the condensed chromatin by acetylating histones, unwinds the double-stranded DNA, and releases the MMP genes for expression[46, 61].

1.3.2 Cell-specific expression of MMPs. In addition to the cis-regulatory elements, expression of distinct MMPs often requires specific trans-regulator factors[46]. Most secreted MMPs are highly-expressed in stromal cells and leukocytes under normal physiological conditions, and MT-MMPs are expressed in a broader range of cell types (Table 1.2) [20, 62]. Although it is not completely clear how trans-regulatory activators control cell-specific expression, it is known that cytokines and chemokines, such as EGF, VEGF, TNF-a, and TGF-regulate the expression of certain MMPs[63]. For example, cytokines/chemokines regulate the expression of MMP1, MMP2, MMP3, MMP8, MMP9, and MMP14 in hematopoietic stem cells (HSC) and in leukocytes during the early immune response[6, 64]. The expression of MMP28 in developing germ cells is regulated by Sox-like trans-elements[65]. The expression of MMP13 in bone osteoblasts is regulated by the cis-element AP-1 and the trans- element RunT domain factor-2 (Runx2) during cartilage development[66]. In addition to regulation by cis- and trans-elements, single nucleotide polymorphisms (SNPs) in certain promoter regions appear to be another mechanism for cell-specific expression of certain MMPs[67]. The SNPs regulate the expression of MMPs by promoting or disrupting the binding of some transcription factors with the promoters. Some SNFs often up-regulate expression of MMPs in cells where MMPs 8

should not be expressed, leading to pathological processes[46, 67]. For example, the expression of MMP1 was significantly up-regulated in oral squamous cell carcinoma, due to an inserted guanosine at -1607 that forms a 2G allele and a binding site of protein c-ets-1 (EST1) [68, 69]. Insertion of an adenosine at -1171 forms a 6A binding site for zinc finger repressor 89 (ZBG 89), which reduced expression of MMP3 in a lung cancer cell line[70, 71]. Conversely, two SNPs were reported to lead to an anti-pathological mechanism by reducing the expression of MMP2 and MMP12: A nucleoside substitution at -1306 (C to T) and -1576 (G to A), along with a site for estrogen receptors  (Er), in the promoter region resulted in reduced expression of MMP2 in breast cancer[72]. An A to G substitution at the -82 position disrupts the binding of AP-1 with the promoter and reduced MMP12 expression in coronary artery disease[73].

1.3.3 Post-translational regulation of the expression of MMPs. AU-rich elements (ARE) in 3’ untranslated regions (3’-UTR) play a crucial role in post-translational regulation of MMPs [74]. Studies on transcripts of MMP9 and MMP13 showed that ARE interact with Hu protein family (HuR) proteins, which increased stability of the transcripts and directed transcripts to the ribosome and polysome [75, 76]. Interestingly, studies also showed that the ARE motif binds to the KH-ype splicing regulatory protein (KSRP/FBP2), resulting in transcript degradation via the exosome [46, 74]. These results indicate that the ARE motif is essential for maintaining homeostasis of MMP expression levels[46] (Figure 1.5).

1.3.4 Regulation of MMP enzymatic activity. Under normal physiological conditions, the enzymatic activity of MMPs are tightly-regulated via two mechanisms: (a) pro-MMP activation by proteolysis and (b) inhibition by naturally-occurring MMP inhibitors [62] (Figure 1.5). As discussed above, activation of pro-MMPs require proteolysis by serine proteinases or metazincin proteinases, including ADAM and MT-MMPs[24]. MT-MMPs can also be activated via pro-protein convertases (PCs) 9

[20, 25] (Figure 1.2, Figure 1.5). One mechanism of regulation is negative feedback of the pro-protein activation pathway. Plasminogen is the pro-protein form of plasmin, which is a serine proteinase involved in activation of multiple pro-MMPs [77]. Plasminogen can be targeted and degraded by MMP2, MMP7, MMP9, and MMP12, which prevents plasmin-mediated, pro-MMP activation [62, 78, 79]. The redox state of the cell can also affect MMP activity. Reactive oxygen species (ROS) can activate pro-MMPs by oxidizing the cysteine in the pro-domain and preventing coordination of the cysteine thiol to the active site Zn(II)[80]. Myeloperoxidase (MPO) transforms ROS to hypochloric acid (HOCl), thereby reducing the ROS-activation of MMPs in acute inflammation [81]. MMPs also regulate other MMPs by proteolysis. MT-MMPs, such as MMP14 and MMP16, are responsible for activating pro-MMP2 and pro-MMP9 in a TIMP-2-dependent or -independent manner[29, 30]. Interestingly, studies have also shown that the enzymatic activities of certain MT-MMPs can be abolished via auto-proteolysis or by proteolysis by other MT-MMPs[82]. MMP14 and MMP16 often exist as an inactive 41 kDa protein that lacks the 20 kDa catalytic domain[83]. MMP16 has been shown to hydrolyze the catalytic domain of MMP14 and partially- unfolded MMP16[82, 84]. Catalytically-active MMPs, which lack the pro-domain, can be inhibited by naturally-occurring inhibitors[62]. The two most-studied, naturally-occurring inhibitors of the MMPs are 2-macroglobulin (2-MG) and tissue inhibitor of metalloproteinases (TIMPs) (Figure 1.5), and these inhibitors have distinct modes of inhibition [10, 62]. 2-MG “traps” secreted MMPs into a macroglobulin complex, and the complex is transported into the lysosome for degradation via the low density lipoprotein receptor-related protein 1 (LRP1)-mediated endocytosis[85]. The TIMPs, on other hand, inhibit the enzymatic activities of certain MMPs by binding directly to the catalytic Zn(II) [5]. The inhibition by TIMPs is achieved by the binding of two conservative motifs (CTCV and ESVC), which are found in the N-termini of the

TIMPs, to Zn(II) [86]. In addition to 2-MG and TIMPs, proteins like -amyloid 10

precursor protein (APP) [87], and reversion-inducing-cysteine-rich protein with kazal motifs (RECK) are also potent inhibitors of MMP2, MMP9, and MMP14 [88].

1.3.5 Unregulated enzymatic activities of MMPs contribute to multiple pathological processes. Unregulated expression of MMPs often results in unregulated MMP activities, leading to many pathological conditions, such as metastasis and tumor-related angiogenesis in multiple cancers, auto-inflammatory diseases, or disorders in the central nervous system, lung, and liver [6, 89]. It is common to observe MAPK/FAK associating with the early development of cancer and inflammatory diseases. Increased MAPK/FAK signaling results in increased expression of MMPs via AP cis-regulation [49, 64]. For example, the expression of

MMP2 is increased in astroglioma [90]. The up-regulation of Wnt signaling pathways during cell invasion by several cancers leads to an increase in the expression of

MMP7 [91]. An up-regulated EGFR/E-cadherin signaling pathway leads to increased transcription of MMP14 in lung cancer [92]. In addition, MMPs have been implicated in the release and activation of multiple chemokines and growth factors, including

EGF, KGF, VEGF, TNF-, and TGF- [6]These chemokine and growth factors are trans-regulatory factors and are required for trans-activation of AP-1/PAE-3 that regulate the transcription of several MMPs (Figure 1.5) [93]. Elevated MMP catalytic activities facilitate the release and activation of chemokines and growth factors, which in turn increase the expression of MMPs in many pathological processes [49, 93]. Up-regulated MMP catalytic activities also affect known regulatory pathways. For example as mentioned above, plasmin hydrolyzes pro-MMPs and activates MMP catalytic activity[77]. The activation of pro-protein plasminogen is negatively-regulated by proteinase inhibitors 1-chymotrypsin(1-CT), 1-proteinase

P1(1-P1), and 2-antiplasmin (2-AP) [94]. These three inhibitors are substrates of MMPs; therefore, MMPs regulate the catalytic activity of plasmin, which regulates the activity of the MMPs [94]. In multiple pathological processes, over- expression of MMPs (including MMP 3 and MMP 9) results in the cleavage of 1-CT, 1-P1, and 11

2-AP and activation of pro-MMPs [62, 95, 96]. The potent inhibitor of MMP, 2-MG, can be proteolyzed by multiple MMPs, leading to prolonged MMPs’ pathological activity [62].

1.4 Role of MMPs in physiological and pathological conditions

1.4.1 Role of MMPs in ECM biology. MMPs have been described as the primary regulators of the extracellular matrix (ECM) [6, 97]. The ECM is the collection of extracellular molecules that are secreted by cells and that participate in cell-cell connections, cellular motility, and intercellular communication[98]. By cleaving ECM components, MMPs create “space” for cell growth, proliferation, migration, and angiogenesis, and directly contribute to biochemical processes, such as tissue growth and remolding, embryotic growth, and wound healing[97]. Moreover, the proteolysis of ECM by MMPs releases ECM-trapped bioactive molecules that trigger several signaling pathways, leading to cell motility and angiogenesis[6] (Figure 1.6). Angiogenesis is the process of forming new blood vessels from existing vessels and requires the degradation of vessel membranes and ECM to allow endothelial cells to invade surrounding tissue [99]. The vascular endothelial growth factor-VEGF ) is usually bound to heparin sulfate proteoglycan, which is part of the ECM [100]. MMP2- and MMP9-mediated proteolysis releases VEGF-to the extracellular space to initiate the signaling pathway of angiogenesis [101]. The mobilization of VEGF-by MMP9 is called the “angiogenetic switch” [102]. Transforming growth factor (TGF–) is in the latent form when bound to the latency-associated peptide (LAP) and the TGF binding protein (TGFB) in the ECM [103]. MMP2, MMP9, MMP13, and MMP-14 (MT-1 MMP) cleave LAP and TGFB, releasing TGF- from the ECM for angiogenesis [103, 104] (Figure 1.6). In addition to the growth factors released from the cleaved ECM, the fragments from the MMP-cleaved ECM (called cryptic fragments) can regulate angiogenetic processes [104, 105]. MMP2 and MMP9 cleave collagen IV, exposing a

hidden cryptic epitope site for pro-angiogenic αvβ3 integrin binding, which facilitate integrin-dependent angiogenesis [106]. The use of monoclonal antibody (HUIV26) blocks the cryptic epitope site, resulting in inhibited angiogenesis in tumor cell models[106]. Several cryptic fragments exhibit anti-angiogenetic properties [104]. For example, endostatin (cleaved from collagen XVIII by MMP7 [14]), arresten, canstain, fumstain (cleaved from collagen IV by MMP9 [107]), and endorepellin (cleaved from pelican by multiple MMPs [108]) exhibit anti-angiogenesis activities by inhibiting endothelial cell proliferation, migration, and tube formation [97, 107]. Cryptic fragments can also be involved in regulating cell motility [97]. Laminin 52 is released after cleavage by MMP2, MMP3, MMP12, MMP13, and MMP14 [109]. Laminin 52, with its epidermal-growth-factor (EGF)-like domains exposed, binds to the EGF-receptor and activates the signaling pathways to promote cell motility and other cellular biological events [110]. The cleavage of fibronectin by MMPs were assumed be associated with releasing migration- stimulating factor (MSF) [97]. MSF is thought to act as an epigenetic factor to promote stromal cell migration and possibly the development of carcinoma cells [111].

1.4.2 MMPs regulate pathological and physiological events by processing signaling proteins. In addition to the ECM components, the physiological substrates of the MMPs also include several regulatory and signaling proteins. Therefore, MMPs not only release bioactive molecules from the ECM but also directly regulate diverse biological processes by proteolytically-activating or -inactivating these bioactive molecules [62, 112] (Figure 1.6). Several non-ECM substrates and their associated physiological/pathological events of five representative MMPs (understudied in our lab) are shown in Table 1.3. In angiogenesis, MMPs can release cryptic fragments and growth factors from the ECM, as described above. MMPs can also cleave certain cytokines and growth factors and activate/inactivate them [6, 112]. For example, MMPs can enhance angiogenesis by proteolytically-processing a latent cytokine/chemokine precursor to 13

an active form: The ELR motif exists in most CXC chemokines for binding with the chemokine receptor CXCL2 to activate downstream pathways [113]. MMP1, MMP8, MMP9, MMP13, and MMP14 cleave the upstream region of the ELF motif in cytokine interleukin 8 (CXCL8/IL8) to enhance the binding potency of the chemokine to its receptor and promote pro-angiogenic activities [6, 114, 115]. MMP14 cleaves the C-terminal tail of platelet derived growth factor receptor (PGFR ), which is associated with PGFR dependent vascular smooth cells differentiation and proliferation and activates via an ERK1/2-mediated signaling cascade [116]. In addition, MMPs can promote angiogenesis by targeting and cleaving inhibitors of angiogenesis. Platelet factor-4 (PF-4) is identified aa angiogenesis inhibitors [117].

Degradation of PF-4 by MMP9 are associated with the formation of blood vessels [118]. Unregulated MMP-promoted angiogenesis has been observed in cancers and auto- inflammatory diseases [6, 99, 112]. Importantly, MMPs are also involved in anti-angiogenesis [6, 99]. MMP2 cleaves the fibroblast growth factor receptor 1 (FGR1)’s ectodomain, abrogating the FGR-1-associated signaling cascade for angiogenesis [119]. Prolactin (PRL) can be cleaved by several secreted MMPs [120]. The cleaved N-terminal fragments of PRL block the epidermal growth factor (EGF) signaling pathway, and thereby suppress endothelial cell proliferation in vascular development [62, 120]. The dual roles of MMPs in angiogenesis are important to maintain a homeostatic equilibrium in vessel remolding and development [62]. MMPs play a major role in cell motility and growth due to the ability of MMPs to proteolytically cleave the ECM and to process diverse regulatory proteins. Over-expression of MMPs in tumor cells promotes increased cell motilities, which are associated with cellular invasion, tumors growth, metastasis, and tumor-related angiogenesis [62] (Figure 1.6). MMP14 have been reported to proteolytically-process

CD44 [121]. MMP14-processed CD44 is associated with the localization of MMP2 and MMP9 on the ECM, and CD44 hyper-activates the TGF- signaling cascade, which increases tumor-related angiogeneesis and tumor cell invasion in multiple malignant cancers [122]. Transmembrane heparin sulfate proteoglycans (HSPGs) 14

stabilizes ECM’s adhesion with cell[123]. MMP9, MMP14 and MMP16 cleave

HSPGs, syndecan-1 and sydecan-4, to promote tumor cell invasion and migration [124, 125]. Moreover, MMPs promote metastasis by activating several signaling molecules: Surface-associated tissue transglutaminase (SATT) is a membrane regulatory protein that maintains cell-ECM adhesion [126]. Degradation of SATT by over-expressed MMP2, MMP14, and MMP16 has been observed in glioma and fibrosarcoma cell invasion, tumor growth, and metastasis [127, 128].

Protease-activated receptor 1 (PAR 1) correlating with cell motility and invasion [129], and MMP1-cleaved PAR 1 contributes to growth and invasion of breast cancer cells [130]. Receptor activator of nuclear factor B ligand (RANKL) triggers activation, proliferation and differentiation of osteoblasts [131]. MMP7 cleaves the transmembrane analog of RNAKL to a soluble form (sRNAKL), which promotes activation and proliferation of pre-osteoblast precursors that are associated with the development of bone tumors [132]. Lipoprotein receptor-related protein (LRP) regulates multiple cellular events(migration and proliferation) via endocytosis of various biomolecules, including proMMPs and growth factors for intracellular degradation [133]. LRP has been identified as substrate of MMP14, and MMP14 cleaves LRP and terminates LRP-mediated endocytosis [134] The loss (or large reduction) of LRP-mediated endocytosis via MMP14 has been associating with abnormal accumulations of MMP2, MMP9, and MMP13 and tissue-type plasminogen activators (PLATs) in MMPs-involved pathological conditions [135]. MMPs promote tumor cell survival and growth by providing a pathway to bypass apoptosis [6, 62](Figure 1.6). The FAS/FAS ligand (FASL) signaling pathway triggers apoptosis [136]. Transmembrane FASL is converted to a soluble analog when

MMP3 or MMP7 cleaves the membrane-anchoring domain (sFASL) [6, 137]. Interestingly, the role of soluble FASL is cell-specific[6]. In most tumor cells, sFASL inhibits apoptosis by lowering the sFAL binding potency to the FAS receptor (FASR) [138]. In endothelial cells, sFASL processed from MMP7 enhance apoptosis[139]; however, the mechanism of how MMP7 regulates apoptosis via FASL is unknown[6]. 15

Tumor necrosis factor superfamily member 21/Death receptor 6 (TNFSF 21/DR 6) is another signaling protein that triggers cell death [140, 141]. MMP14 cleaves the ectodomain of TNFSF 21, which “short circuits” the intracellular signaling pathway and abrogates this cell death process [6, 142]. Leukocyte recruitment is required in inflammation and innate immunity to combat pathogens or to remove damaged tissues. Excess leukocyte recruitment is considered the hallmark of multiple inflammatory diseases [143]. Leukocyte recruitment is triggered by chemokines, especially those in the CXC family. Recent studies demonstrate that MMPs can cleave multiple proteins that anchor chemokine to membranes, resulting in both enhanced and suppressed leukocyte recruitment[6]. Hematopoietic stem cells (HSCs) are the precursors of leukocytes[144]. HSCs must be released from bone narrow before being differentiated into leukocytes following an inflammation trigger[144,145]. Chemokine CXCL 12 and membrane-associated kit-ligand (KIT) regulate the retention of HSCs in bone marrow[146, 147]. MMP9 cleaves CXCL and KIT, which releases the HSCs from the bone marrow stromal cells [6, 147]. Moreover, MMPs can directly activate or silence chemokine signaling pathways to regulate trafficking of several different leukocytes [6]. MMP8, MMP9, MMP13, and MMP14 proteolytically-activate CXCL 8, which triggers neutrophil recruitment in damaged tissues[6, 114]. MMP1 and MMP9 proteolytically- activate CXCL5, which promotes neutrophil recruitment[114]. In contrast, macrophage-specific MMP12 inactivates CXCL1, CXCL2, CXCL3, CXCL5, CXCL6, and CXCL8 by cleaving receptor-binding motif (ELR) of these chemokines[148]. MMP12-associated proteolysis of these chemokines disrupts macrophage and polymorphonuclear leukocyte (PMN)’s influx [148]. In a related fashion, certain pro-inflammatory cytokines are also involved in MMP-regulated inflammation [6]. For example, MMP7 activates TNF, which is involved in macrophage infiltration during disc resorption[149].

1.4.3 Protective role of MMPs in pathological processes. MMPs have long been 16

considered as a drug target in the treatment of cancer and inflammatory diseases[1, 150]. However, recent studies reveal that MMPs may also have roles in anti-tumorigenesis and anti-inflammation[89, 151]. MMP8 was the first reported MMP to exhibit an anti-tumorigenetic role in pre-clinical tests [89]. A MMP8 null mouse showed higher levels of chemical-induced skin tumorigenesis and lower survival rates than wild-type mice[152]. In lung cancers, MMP7 exhibited anti-metastatic activity due to its ability to proteolytically inactivate the 64 intergrin signaling pathway, resulting in the blockage of lung tumor cell migration and adhesion to endothelial cells[153]. MMP2 cleaves the N-terminus of CX3CL1, which converts this chemokine from agonist to antagonist and lessens the pro-inflammatory response in rheumatoid arthrodesis[6, 154]. Moreover, a MMP3-deficient mouse model exhibited a higher rate of lung cancer metastasis as compared to the wild-type mouse[155]. A MMP19-deficient mouse model displayed accelerated tumor-related angiogenesis and tumor cell invasion[156]. Expression of MMP26 prolongs survival times of ductal carcinoma patients[157]. In fact, Dufour et al. published a paper describing MMPs as “targets” or “anti-targets”, depending on whether the MMPs could be successfully targeted for clinical inhibitors. Presently, six of the twenty-three MMPs have been shown to exhibit protective functions in diverse pathologies[89](Figure 1.6).

1.5 Review of MMP inhibitors

1.5.1 Therapeutic targeting of MMPs. Considerable effort has been made to develop strategies to suppress the unregulated expression and activities of MMPs in cancers and inflammatory diseases[1, 150]. These strategies involved reducing the expression of MMPs, suppressing the activation of MMPs, and inhibiting the catalytic activities of MMPs (Figure 1.7). High-level expression of MMPs is a hallmark phenotype in multiple MMP-associated pathologies[93]. Therefore, the suppression of MMP transcription

could be a strategy to reduce the catalytic activities of MMPs. In tumor cell models, chemically-engineered tetracycline was reported to inhibit MMP-related cis-transcription factors IL-8 and TNF- and suppress the expression of several MMPs[150, 158]. In an in vitro cancer model, curcumin, which is a specific inhibitor of NF-kB, lowered the expression of MMP9, leading to a reduction in tumor size [159]. In addition to the targeting of transcription factors, the transcripts of MMPs have also been targeted. Antisense oligonucleotides (ASON) designed to bind to the mRNA of a specific MMP were shown to antagonize translation of that MMP [160]. Previous studies reported that ASON for MMP7 exhibited a potent anti-metastasis effect in cancer animal models [161, 162]. As mentioned above, the hemopexin (Hpx) domain in most MMPs contributes to the ability of MMPs to interact with proteins or other macromolecules [150]. In the MT-MMPs, the Hpx domains play a role in these membrane-associated

MMPs being able to proteolytically activate MMP2 and MMP9 [20] (Figure 1.6).

Two research groups, Dufour A et al. and Remacle AG, reported small-molecule inhibitors, which selectively target the Hpx domain in MMP14, and these inhibitors were potent inhibitors of the interaction of MMP14 with pro-MMP2/9 [163]. These Hpx-targeted inhibitors reduced tumor cell volumes and invasion in a fibrotic cancer model [163, 164]. Engineered fusion proteins were used as an alternative strategy to target the HPx domain in MMPs [150]. For example, fusion protein APP-IP-TIMP2 (ten amino acid residue sequence of APP-derived MMP-2 selective inhibitory peptide (APP-IP) added to N’ terminal of TIMP2) was designed to bind to the pro- and HPx domains in MMP2[165]. This fusion protein prevented the formation of functional pro-MMP2 -TIMP2 complex, hence hinder the activation of proMMP2 via MMP14[165]. The catalytic domains, specifically the active sites, are the most common targets for MMP inhibitors[35, 150]. There are two major types of MMP inhibitors (MMPi) targeting the catalytic domain[35] (Table 1.4): (a) inhibitors that target and bind to the catalytic Zn(II) (called zinc binding groups, ZBGs) and (b) inhibitors that 18

do not target the catalytic Zn(II) but rather target the surrounding substrate/inhibitor binding pockets[35]. A.R Johnson and coworkers reported a series of MMPi, which did not target the catalytic Zn(II) and were based on pyrimidine dicarboxamide scaffolds [166]. C.K Engel’s group and Morales’s group also reported non-zinc binding inhibitors respectively, which exhibited selective inhibition on MMP13,

MMP8, and MMP12 in preclinical trials [167, 168]. The ZBG-based MMPi, on the other hand, inhibit enzymatic activity by directly binding to the catalytic Zn(II) [35]. Some of these ZBG-based MMPi will be discussed below.

1.5.2 Overview of ZBG-based MMPi. Based on chemical structure, the ZBG-based MMPi have been categorized into four major classes (Table 1.4): (a) hydroxamic acids, (b) carboxylates, (c) phosphorus-based, and (d) nitrogen-based[35]. Carboxylate-based MMPi are thought to bind to the catalytic Zn(II) via two oxygens[35]. Li and coworkers reported selective MMP12 inhibitors, and they attributed the selectivity to a dibenzofuran group that interacted strongly with the S1’ substrate binding pocket[169]. Phosphorus-based MMPi are thought to bind to the catalytic Zn(II) via two phosphonate oxygens[35]. A’Biasone and coworkers reported potent and selective inhibitors of MMP8, and some of these compounds

were used in preclinical studies for treating acute liver diseases and sclerosis [170]. Nitrogen-based MMPi have been developed by Jacobsen’s group since 2006, and

these inhibitors often have pyridine rings with substituents that can coordinate Zn(II) [171]. In fact, the mode of binding is thought to be bidentate with the pyridine nitrogen and usually a carbonyl oxygen on a substituent binding to the catalytic Zn(II) [35, 171]. An example of a nitrogen-based MMPi is pyrimidine-2,4,6 thione, and Hoffman and LaRoche showed that inhibitors based on this scaffold are potent inhibitors of MMP3 and MMP8[172, 173]. Importantly, this scaffold has been used in several FDA-approved drugs[35, 172].

Hydroxamic acids are, by far, the most commonly used scaffolds for MMPi, and several broad-spectrum inhibitors have been used in clinical trials[1, 35]. Hydroxamic acid-based MMPi can bind to the catalytic Zn(II) in a mono- or bidentate fashion[35]. In early studies, the hydroxamic acid-based inhibitors exhibited low inhibitory potencies, and it was speculated that one of the hydroxamic oxygens rotated away from Zn(II) and did not coordinate the metal ion[174, 175]. To overcome this issue, cyclic hydroxamic acid-based MMPis were developed[175]. The cyclic structure of the resulting MMPi was thought to favor bidentate binding of both hydroxamic acid oxygens with Zn(II) [175]. Squaric acid-based MMPi were the first ZBG-MMPi with a cyclic structure, and these compounds did not contain a hydroxamic acid binding group[176] Subsequently, hydroxamic-based, heterocyclic MMPi were reported [177-179], and these compounds were designed to bind in a bidentate binding mode to the catalytic Zn(II) and to increase bioavailability[35]. Cohen and coworkers developed a series of sulfur-substituted, heterocyclic MMPi, and these MMPi exhibited over a hundred-fold higher inhibitory potency on MMP3 and MMP8 than the parent compounds that lacked sulfur[180, 181]. In addition to the competitive, Zn(II)-chelating MMPi, Mobashery and coworkers developed a “mechanism-based” MMPi called SB-3CT[182] (Table 1.3). SB-3CT was shown to specifically target MMP2 and MMP9 and inhibit bone metastasis in preclinical tests on prostate cancer[182]. Unlike most MMPi, SB-3CT irreversibly inactivated MMPs by forming a covalent attachment with an active site Glu [182, 183].

1.5.3 Challenges and new opportunities for next generation ZBG MMPi. Although considerable efforts have been made developing MMPi over the last several decades, only one broad-spectrum MMP inhibitor (Periostat, CollaGenex) is FDA-approved for treating periodontal disease[1]. Previous ZBG-MMPi exhibited significant inhibitory potencies in pre-clinical studies; however, these same compounds have been

unsuccessful in clinical trials due to dose-limiting cytotoxicities and poor bioavailabilities[1]. For example, marimastat, which is a broad-spectrum MMPi, causes severe musculoskeletal pain and inflammation in patients [184]. These severe side-effects were attributed to off-target inhibition of other MMPs and MMP-like metacinzin proteinases (ADAM and ADAMT)[1, 89, 184]. Several of the best in vitro MMPi exhibit low oral bioavailabilities and poor solubilities when tested in clinical trials[1, 35, 185, 186]. In addition, several hydroxamic acid-based MMPi were reported to be hydroxylated or cleaved by exopeptidases, which lowered bioavailability of these compounds [35, 187]. Therefore, two major objectives of future inhibitor/drug discovery are to increase inhibitor selectivity and specificity and to improve bioavailability[1]. The initial idea of developing and using a broad spectrum MMPi to treat disease has proven to be problematic because MMPs have been shown to have important physiological roles as well as roles in clinical pathologies[89, 150]. As mentioned above, MMPs have been classified as drug targets and anti-drug targets[89]. To achieve the goal of developing clinical inhibitors of MMPs, it is necessary to characterize several of the MMPs and probe for structural or mechanistic differences in the enzymes, and these differences can be targeted with optimized ZBG-type MMPi. The biggest challenge in optimizing the selectivity of ZBG MMPi is the high functional and structural similarities of the catalytic domains in the MMPs. Promising targets for selective MMPi are the substrate binding pockets that surround the catalytic Zn(II) site in the MMPs[4, 35, 188]. Of these substrate binding pockets, the S1’ pocket is considered to be the prime target for optimizing MMPi because there is significant diversity (some shallow, some deep, etc.) in this pocket in the different MMPs[35, 188]. The S1’ pocket is the closest pocket to the catalytic Zn(II) site, making it the most “reachable” for substituents on small-molecule ZBG-MMPi[181, 188]. Combining X-ray crystallographic studies and computational simulations, Cohen’s group reported that significant selectivity can be achieved on MMP3 and MMP8 with small molecule ZBG MMPi without the need for bulky pseudo-peptides 21

[181]. Cohen and coworkers tested allothiomaltol (ATM)，thiomaltol (TM), and thiopyromeconic acid (TPMA) [180, 181]. The only difference between these ZBGs is the position of a methyl group on the ring structure[181]. The observed selectivity was attributed to interactions between S1’ pockets of MMPs and the methyl side-chain [180, 181]. This work was significant because MMPi selectivity could be achieved by minor chemical modifications of a MMPi scaffold[178, 180]. Cohen’s work presented a new potentially simple and economic way for achieving ZBG MMPi’ s selectivity by minor chemical modification on the MMPi’s heterocyclic scaffold[178-180]. Unfortunately, Cohen and coworkers only used two different MMPs. To test whether this improved selectivity extrapolates to MMPs from other classes, we examined the interaction of TM and ATM with matrilysin (MMP7) and Type I MT-MMP (MMP16) in Chapter 3 of this dissertation.

1.6 Characterization of and inhibition studies on mmp7 and mmp16

Most previous attempts to identify clinical inhibitors of the MMPs have involved inhibition studies on 1-3 MMPs[1, 35, 185, 186, 189]. Not surprisingly, most of the discovered inhibitors failed in clinical trials, most often due to off target inhibition of the “wrong” MMP[1, 89]. Our approach is different. We propose to characterize a representative MMP from each of the distinct MMP subclasses in an effort to identify different mechanistic and/or structural properties of the enzymes that can be targeted for inhibitor design efforts. Unfortunately, there are very few mechanistic studies on any MMP. Therefore, one major goal of our work was to conduct detailed mechanistic studies on several MMPs. In addition, we probed structure and metal binding using several biophysical techniques (Figure 1.8). The work in this dissertation describes kinetic and structural studies on two distinct MMPs: matrilysin (MMP7) and Type I MT-MMP (MMP16).

In our studies, we focused on the catalytic domains of MMP7 (cdMMP7) and MMP16 (cdMMP16). The cdMMP16 and cdMMP7 were over-expressed in E. coli and purified using optimized strategies. We characterized the resulting recombinant enzymes using steady-state kinetics, mass spectrometry, metal analyses, CD spectroscopy, and fluorescence studies. Pre-steady state kinetic studies were conducted on these two cdMMPs to probe the kinetic mechanisms of these enzymes[82, 190]. In addition to biochemical characterization, we conducted inhibition studies on cdMMP16 and cdMMP 7. Inspired by Dr. Cohen’s previous work on MMPs[178, 180, 181], we probed the interactions of AHA, maltol, TM, and ATM with cdMMP7 and cdMMP16. The inhibitory properties of these ZBGs were evaluated with IC50 measurements, and binding of these ZBGs to cdMMP7 and cdMMP16 was evaluated using ITC measurements (Figure 1.8). Docking studies of TM and ATM binding to cdMMP7 and cdMMP16 were conducted to probe how TM and ATM interact with the MMPs. We attempted to use X-ray crystallography to probe ZBG binding to cdMMP16; however, we were unable to obtain suitable crystals. We therefore prepared analogs of cdMMP7 and cdMMP16 that could be used for future spectroscopic studies to probe ZBG binding to the enzymes. MMPs bind Zn(II) in the physiologically-relevant enzyme[4]. However, very few spectroscopic techniques can be used to interrogate Zn(II) binding sites in proteins[82, 190, 191]. Therefore, we prepared Co(II)-substituted analogs of cdMMP7 and cdMMP16: (1) a CoCo-substituted analog of cdMMP16 (Co(II) in the catalytic and structural Zn(II) sites) and (2) ZnCo analogs of cdMMP16 and cdMMP7 (Co(II) in the catalytic site and Zn(II) in structural site). These Co(II)-substituted analogs were characterized using ICP-AES, UV-Vis, 1H NMR, and EXAFS spectroscopies (only cdMMP16) (Figure 1.8). These analogs are now available for future structural studies to probe MMPi binding to these MMPs. In addition as mentioned above, these analogs now allow for future rapid-freeze quench spectroscopic studies to determine the reaction mechanism of the MMPs. 23

1.7 References

1. Vandenbroucke, R. E. and Libert, C. (2014) Is there new hope for therapeutic matrix metalloproteinase inhibition? Nature Rev. Drug Dis. 13, 904-927 2. Gross, J. and Lapiere, C. M. (1962) Collagenolytic activity in amphibian tissues: a tissue culture assay. Proc. Nat. Acad. Sci. 48, 1014-1022 3. Eisen, A. Z., Jeffrey, J. J. and Gross, J. (1968) Human skin collagenase, isolation and mechanism of attack on the collagen molecule. Biochim. Biophys. Acta (BBA)-Enzymology. 151, 637-645 4. Tallant, C., Marrero, A. and Gomis-Ruth, F. X. (2010) Matrix metalloproteinases: fold and function of their catalytic domains. Biochim. Biophys. Acta. 1803, 20-28 5. Nagase, H., Visse, R. and Murphy, G. (2006) Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc. Res. 69, 562-573 6. Rodríguez, D., Morrison, C. J. and Overall, C. M. (2010) Matrix metalloproteinases: what do they not do? New substrates and biological roles identified by murine models and proteomics. Biochim. Biophys. Acta -Mol. Cell Res. 1803, 39-54 7. Adlakha, N. and Mittal, A. (2010) Multi class classification approach for classification of ADAMs, MMPs and their subclasses. Int. J. Engin. Technolo. 2, 302 8. Patterson, M. L., Atkinson, S. J., Knäuper, V. and Murphy, G. (2001) Specific collagenolysis by gelatinase A, MMP-2, is determined by the hemopexin domain and not the fibronectin-like domain. FEBS Lett. 503, 158-162 9. Toth, M., Sohail, A. and Fridman, R. (2012) Assessment of gelatinases (MMP-2 and MMP-9) by gelatin zymography. Metastasis Res. Prot., 121-135 10. Visse, R. and Nagase, H. (2003) Matrix metalloproteinases and tissue inhibitors of metalloproteinases structure, function, and biochemistry. Circul. Res. 92, 827-839 11. Sternlicht, M. D., Bissell, M. J. and Werb, Z. (2000) The matrix metalloproteinase stromelysin-1 acts as a natural mammary tumor promoter. Oncogene. 19, 1102 12. Murphy, G., Segain, J., O'Shea, M., Cockett, M., Ioannou, C., Lefebvre, O., Chambon, P. and Basset, P. (1993) The 28-kDa N-terminal domain of mouse

stromelysin-3 has the general properties of a weak metalloproteinase. J. Biol. Chem. 268, 15435-15441 13. Park, H. I., Turk, B. E., Gerkema, F. E., Cantley, L. C. and Sang, Q.-X. A. (2002) Peptide substrate specificities and protein cleavage sites of human endometase/matrilysin-2/matrix metalloproteinase-26. J. Biol. Chem. 277, 35168-35175 14. Ii, M., Yamamoto, H., Adachi, Y., Maruyama, Y. and Shinomura, Y. (2006) Role of matrix metalloproteinase-7 (matrilysin) in human cancer invasion, apoptosis, growth, and angiogenesis. Exp. Biol. Med. 231, 20-27 15. Pellicoro, A., Aucott, R. L., Ramachandran, P., Robson, A. J., Fallowfield, J. A., Snowdon, V. K., Hartland, S. N., Vernon, M., Duffield, J. S. and Benyon, R. C. (2012) Elastin accumulation is regulated at the level of degradation by macrophage metalloelastase (MMP-12) during experimental liver fibrosis. Hepatology. 55, 1965-1975 16. Stracke, J. O., Hutton, M., Stewart, M., Pendás, A. M., Smith, B., López-Otin, C., Murphy, G. and Knäuper, V. (2000) Biochemical characterization of the catalytic domain of human matrix metalloproteinase 19. Evidence for a role as a potent basement membrane degrading enzyme. J. Biol. Chem. 275, 14809-14816 17. Bartlett, J., Beniash, E., Lee, D. and Smith, C. (2004) Decreased mineral content in MMP-20 null mouse enamel is prominent during the maturation stage. J. Dent. Res. 83, 909-913 18. Zhang, S.-c. and Kern, M. (2009) The role of host-derived dentinal matrix metalloproteinases in reducing dentin bonding of resin adhesives. Int. J. Oral Sci. 1, 163 19. Saarialho-Kere, U., Kerkelä, E., Jahkola, T., Suomela, S., Keski-Oja, J. and Lohi, J. (2002) Epilysin (MMP-28) expression is associated with cell proliferation during epithelial repair. J. Invest. Dermatol. 119, 14-21 20. Itoh, Y. (2015) Membrane-type matrix metalloproteinases: Their functions and regulations. Matrix Biol. 44-46, 207-223 25

21. Sohail, A., Sun, Q., Zhao, H., Bernardo, M. M., Cho, J.-A. and Fridman, R. (2008) MT4-(MMP17) and MT6-MMP (MMP25), A unique set of membrane-anchored matrix metalloproteinases: properties and expression in cancer. Cancer Metastasis Rev. 27, 289-302 22. Galea, C. A., Nguyen, H. M., Chandy, K. G., Smith, B. J. and Norton, R. S. (2014) Domain structure and function of matrix metalloprotease 23 (MMP23): role in potassium channel trafficking. Cell. Mol. Life Sci. 71, 1191-1210 23. Van Wart, H. E. and Birkedal-Hansen, H. (1990) The cysteine switch: a principle of regulation of metalloproteinase activity with potential applicability to the entire matrix metalloproteinase gene family. Proc. Nat. Acad. Sci. 87, 5578-5582 24. Nagase, H. (1996) Activation mechanisms of matrix metalloproteinases. Biol. Chem. 378, 151-160 25. Stawowy, P., Meyborg, H., Stibenz, D., Stawowy, N. B. P., Roser, M., Thanabalasingam, U., Veinot, J. P., Chrétien, M., Seidah, N. G. and Fleck, E. (2005) Furin-like proprotein convertases are central regulators of the membrane type matrix metalloproteinase–pro-matrix metalloproteinase-2 proteolytic cascade in atherosclerosis. Circulation. 111, 2820-2827 26. Piccard, H., Van den Steen, P. E. and Opdenakker, G. (2007) Hemopexin domains as multifunctional liganding modules in matrix metalloproteinases and other proteins. J. Leukocyte Biol. 81, 870-892 27. Chung, L., Dinakarpandian, D., Yoshida, N., Lauer‐Fields, J. L., Fields, G. B., Visse, R. and Nagase, H. (2004) Collagenase unwinds triple-helical collagen prior to peptide bond hydrolysis. EMBO J. 23, 3020-3030 28. Itoh, Y. and Seiki, M. (2006) MT1-MMP: A potent modifier of pericellular microenvironment. J. Cell. Physiol. 206, 1-8 29. Itoh, Y., Takamura, A., Ito, N., Maru, Y., Sato, H., Suenaga, N., Aoki, T. and Seiki, M. (2001) Homophilic complex formation of MT1-MMP facilitates proMMP‐2 activation on the cell surface and promotes tumor cell invasion. EMBO J. 20, 4782-4793 26

30. Strongin, A. Y., Collier, I., Bannikov, G., Marmer, B. L., Grant, G. A. and Goldberg, G. I. (1995) Mechanism of cell surface activation of 72-kDa type IV collagenase isolation of the activated form of the membrane metalloprotease. J. Biol. Chem. 270, 5331-5338 31. Brew, K. and Nagase, H. (2010) The tissue inhibitors of metalloproteinases (TIMPs): an ancient family with structural and functional diversity. Biochim. Biophys. Acta -Mol. Cell Res. 1803, 55-71 32. Massova, I., Kotra, L. P. and Mobashery, S. (1998) Structural insight into the binding motifs for the calcium ion and the non-catalytic zinc in matrix metalloproteases. Bioorg. Med. Chem. Lett. 8, 853-858 33. Bode, W., Gomis-Rüth, F.-X. and Stöckler, W. (1993) Astacins, serralysins, snake venom and matrix metalloproteinases exhibit identical zinc‐binding environments (HEXXHXXGXXH and Met‐turn) and topologies and should be grouped into a common family, the ‘metzincins’. FEBS Lett. 331, 134-140 34. Aureli, L., Gioia, M., Cerbara, I., Monaco, S., Fasciglione, G. F., Marini, S., Ascenzi, P., Topai, A. and Coletta, M. (2008) Structural bases for substrate and inhibitor recognition by matrix metalloproteinases. Curr. Med. Chem. 15, 2192-2222 35. Jacobsen, J. A., Jourden, J. L. M., Miller, M. T. and Cohen, S. M. (2010) To bind zinc or not to bind zinc: an examination of innovative approaches to improved metalloproteinase inhibition. Biochim. Biophys. Acta -Mol. Cell Res. 1803, 72-94 36. Tallant, C., García-Castellanos, R., Baumann, U. and Gomis-Rüth, F. X. (2010) On the relevance of the Met-turn methionine in metzincins. J. Biol. Chem. 285, 13951-13957 37. Berton, A., Rigot, V., Huet, E., Decarme, M., Eeckhout, Y., Patthy, L., Godeau, G., Hornebeck, W., Bellon, G. and Emonard, H. (2001) Involvement of fibronectin type II repeats in the efficient inhibition of gelatinases A and B by long-chain unsaturated fatty acids. J. Biol. Chem. 276, 20458-20465

38. Steffensen, B., Xu, X., Martin, P. A. and Zardeneta, G. (2002) Human fibronectin and MMP-2 collagen binding domains compete for collagen binding sites and modify cellular activation of MMP-2. Matrix Biol. 21, 399-414 39. Ozhogina, O. A., Trexler, M., Bányai, L., Llinás, M. and Patthy, L. (2001) Origin of fibronectin type II (FN2) modules: Structural analyses of distantly‐related members of the kringle family idey the kringle domain of neurotrypsin as a potential link between FN2 domains and kringles. Protein Sci. 10, 2114-2122 40. English, W. R., Holtz, B., Vogt, G., Knäuper, V. and Murphy, G. (2001) Characterization of the Role of the “MT-loop”. An eight-amino acid insertion specific to progelatinase A (MMP2) activating membrane-type matrix metalloproteinases. J. Biol. Chem. 276, 42018-42026 41. Woskowicz, A. M., Weaver, S. A., Shitomi, Y., Ito, N. and Itoh, Y. (2013) MT-LOOP-dependent localization of membrane type I matrix metalloproteinase (MT1-MMP) to the cell adhesion complexes promotes cancer cell invasion. J. Biol. Chem. 288, 35126-35137 42. Browner, M. F., Smith, W. W. and Castelhano, A. L. (1995) Matrilysin-inhibitor complexes: common themes among metalloproteases. Biochemistry. 34, 6602-6610 43. Manzetti, S., McCulloch, D. R., Herington, A. C. and van der Spoel, D. (2003) Modeling of enzyme–substrate complexes for the metalloproteases MMP-3, ADAM-9, and ADAM-10. J. Comput.-Aided Mol. Des. 17, 551-565 44. Bertini, I., Calderone, V., Fragai, M., Luchinat, C., Maletta, M. and Yeo, K. J. (2006) Snapshots of the reaction mechanism of matrix metalloproteinases. Angew. Chem. Int. Ed. 45, 7952-7955 45. Pelmenschikov, V. and Siegbahn, P. E. (2002) Catalytic mechanism of matrix metalloproteinases: two-layered ONIOM study. Inorg. Chem. 41, 5659-5666 46. Yan, C. and Boyd, D. D. (2007) Regulation of matrix metalloproteinase gene expression. J. Cell. Physiol. 211, 19-26 47. Benbow, U. and Brinckerhoff, C. E. (1997) The AP-1 site and MMP gene regulation: What is all the fuss about? Matrix Biol. 15, 519-526 28

48. DeMali, K. A., Wennerberg, K. and Burridge, K. (2003) Integrin signaling to the actin cytoskeleton. Curr. Opin. Cell Biol. 15, 572-582 49. Overall, C. M. and López-Otín, C. (2002) Strategies for MMP inhibition in cancer: innovations for the post-trial era. Nature Rev. Cancer. 2, 657-672 50. Ilyas, M. (2005) Wnt signalling and the mechanistic basis of tumour development. J. Pathol. 205, 130-144 51. Yan, C., Wang, H. and Boyd, D. D. (2001) KiSS-1 represses 92-kDa type IV collagenase expression by down-regulating NF-κB binding to the promoter as a consequence of IκBα-induced block of p65/p50 nuclear translocation. J. Biol. Chem. 276, 1164-1172 52. Huang, S., Pettaway, C. A., Uehara, H., Bucana, C. D. and Fidler, I. J. (2001) Blockade of NF-κB activity in human prostate cancer cells is associated with suppression of angiogenesis, invasion, and metastasis. Oncogene. 20 53. Yokoo, T. and Kitamura, M. (1996) Dual regulation of IL-1 beta-mediated matrix metalloproteinase-9 expression in mesangial cells by NF-kappa B and AP-1. American J. Physiol.-Renal Physiol. 270, F123-F130 54. Collins, T., Read, M., Neish, A., Whitley, M., Thanos, D. and Maniatis, T. (1995) Transcriptional regulation of endothelial cell adhesion molecules: NF-kappa B and cytokine-inducible enhancers. FASEB J. 9, 899-909 55. Couillard, J., Demers, M., Lavoie, G. and St-Pierre, Y. (2006) The role of DNA hypomethylation in the control of stromelysin gene expression. Biochem. Biophys. Res. Commun. 342, 1233-1239 56. Knoepfler, P. S. and Eisenman, R. N. (1999) Sin meets NuRD and other tails of repression. Cell. 99, 447-450 57. Yan, C., Wang, H., Toh, Y. and Boyd, D. D. (2003) Repression of 92-kDa type IV collagenase expression by MTA1 is mediated through direct interactions with the promoter via a mechanism, which is both dependent on and independent of histone deacetylation. J. Biol. Chem. 278, 2309-2316

58. Ma, Z., Chang, M. J., Shah, R. C. and Benveniste, E. N. (2005) Interferon-γ-activated STAT-1α suppresses MMP-9 gene transcription by sequestration of the coactivators CBP/p300. J. Leukocyte Biol. 78, 515-523 59. Ma, Z., Shah, R. C., Chang, M. J. and Benveniste, E. N. (2004) Coordination of cell signaling, chromatin remodeling, histone modifications, and regulator recruitment in human matrix metalloproteinase 9 gene transcription. Mol. Cell. Biol. 24, 5496-5509 60. Ito, T., Yamauchi, M., Nishina, M., Yamamichi, N., Mizutani, T., Ui, M., Murakami, M. and Iba, H. (2001) Identification of SWI· SNF complex subunit BAF60a as a determinant of the transactivation potential of Fos/Jun dimers. J. Biol. Chem. 276, 2852-2857 61. Sudarsanam, P. and Winston, F. (2000) The Swi/Snf family: nucleosome-remodeling complexes and transcriptional control. Trends Genet. 16, 345-351 62. Kessenbrock, K., Plaks, V. and Werb, Z. (2010) Matrix metalloproteinases: regulators of the tumor microenvironment. Cell. 141, 52-67 63. Fini, M. E., Cook, J. R., Mohan, R. and Brinckerhoff, C. (1998) Regulation of matrix metalloproteinase gene expression. Matrix metalloproteinases, 299-356 64. Burno,C. (2008) Extracellular remodeling and matrix metlloproteinase in the liver, in Matrix metalloproteinases in tissue remodelling and inflammation (Lagente, V., ed.). pp.154-156. Springer 65. Illman, S. A., Keski-Oja, J. and Lohi, J. (2001) Promoter characterization of the human and mouse epilysin (MMP-28) genes. Gene. 275, 185-194 66. Hess, J., Porte, D., Munz, C. and Angel, P. (2001) AP-1 and Cbfa/runt physically interact and regulate parathyroid hormone-dependent MMP13 expression in osteoblasts through a new osteoblast-specific element 2/AP-1 composite element. J. Biol. Chem. 276, 20029-20038

67. Ye, S. (2000) Polymorphism in matrix metalloproteinase gene promoters: implication in regulation of gene expression and susceptibility of various diseases. Matrix Biol. 19, 623-629 68. Rutter, J. L., Mitchell, T. I., Butticè, G., Meyers, J., Gusella, J. F., Ozelius, L. J. and Brinckerhoff, C. E. (1998) A single nucleotide polymorphism in the matrix metalloproteinase-1 promoter creates an Ets binding site and augments transcription. Cancer Res. 58, 5321-5325 69. Wyatt, C. A., Coon, C. I., Gibson, J. J. and Brinckerhoff, C. E. (2002) Potential for the 2G single nucleotide polymorphism in the promoter of matrix metalloproteinase to enhance gene expression in normal stromal cells. Cancer Res. 62, 7200-7202 70. Ye, S., Whatling, C., Watkins, H. and Henney, A. (1999) Human stromelysin gene promoter activity is modulated by transcription factor ZBP‐89. FEBS Lett. 450, 268-272 71. Fang, S., Jin, X., Wang, R., Li, Y., Guo, W., Wang, N., Wang, Y., Wen, D., Wei, L. and Zhang, J. (2005) Polymorphisms in the MMP1 and MMP3 promoter and non-small cell lung carcinoma in North China. Carcinogenesis. 26, 481-486 72. Harendza, S., Lovett, D. H., Panzer, U., Lukacs, Z., Kühnl, P. and Stahl, R. A. (2003) Linked common polymorphisms in the gelatinase a promoter are associated with diminished transcriptional response to estrogen and genetic fitness. J. Biol. Chem. 278, 20490-20499 73. Jormsjö, S., Ye, S., Moritz, J., Walter, D. H., Dimmeler, S., Zeiher, A. M., Henney, A., Hamsten, A. and Eriksson, P. (2000) Allele-specific regulation of matrix metalloproteinase-12 gene activity is associated with coronary artery luminal dimensions in diabetic patients with manifest coronary artery disease. Circul. Res. 86, 998-1003 74. Knapinska, A. M., Irizarry-Barreto, P., Adusumalli, S., Androulakis, I. and Brewer, G. (2005) Molecular mechanisms regulating mRNA stability: physiological and pathological significance. Curr. Genomics. 6, 471-486 31

75. Huwiler, A., Akool, E.-S., Aschrafi, A., Hamada, F. M., Pfeilschifter, J. and Eberhardt, W. (2003) ATP potentiates interleukin-1β-induced MMP-9 expression in mesangial cells via recruitment of the ELAV protein HuR. J. Biol. Chem. 278, 51758-51769 76. Rydziel, S., Delany, A. M. and Canalis, E. (2004) AU-rich elements in the collagenase 3 mRNA mediate stabilization of the transcript by cortisol in osteoblasts. J. Biol. Chem. 279, 5397-5404 77. Collen, D. (1999) The plasminogen (fibrinolytic) system. Thromb. Haemost. 82, 259-270 78. Dong, Z., Kumar, R., Yang, X. and Fidler, I. J. (1997) Macrophage-derived metalloelastase is responsible for the generation of angiostatin in Lewis lung carcinoma. Cell. 88, 801-810 79. O’Reilly, M. S., Wiederschain, D., Stetler-Stevenson, W. G., Folkman, J. and Moses, M. A. (1999) Regulation of angiostatin production by matrix metalloproteinase-2 in a model of concomitant resistance. J. Biol. Chem. 274, 29568-29571 80. Weiss, S., Peppin, G., Ortiz, X., Ragsdale, C. and Test, S. (1985) Oxidative autoactivation of latent collagenase by human neutrophils. Science. 227, 747-749 81. Fu, X., Kassim, S. Y., Parks, W. C. and Heinecke, J. W. (2003) Hypochlorous Acid Generated by Myeloperoxidase Modifies Adjacent Tryptophan and Glycine Residues in the Catalytic Domain of Matrix Metalloproteinase-7 (Matrilysin). An oxidative mechanism for restraining proteolytic activity during inflammation. J. Biol. Chem. 278, 28403-28409 82. Meng, F., Yang, H., Aitha, M., George, S., Tierney, D. L. and Crowder, M. W. (2016) Biochemical and spectroscopic characterization of the catalytic domain of MMP16 (cdMMP16). J. Biol. Inorg. Chem., 1-13 83. Lehti, K., Jouko, L., Valtanen, H. and Keski-Oja, J. (1998) Proteolytic processing of membrane-type-1 matrix metalloproteinase is associated with gelatinase A activation at the cell surface. Biochem. J. 334, 345-353 32

84. Tatti, O., Arjama, M., Ranki, A., Weiss, S. J., Keski-Oja, J. and Lehti, K. (2011) Membrane-type-3 matrix metalloproteinase (MT3-MMP) functions as a matrix composition-dependent effector of melanoma cell invasion. PLoS One. 6, e28325 85. Yen, F. T., Mann, C. J., Guermani, L. M., Hannouche, N. F., Hubert, N., Hornick, C. A., Bordeau, V. N., Agnani, G. and Bihain, B. E. (1994) Identification of a lipolysis-stimulated receptor that is distinct from the LDL receptor and the LDL receptor-related protein. Biochemistry. 33, 1172-1180 86. Gomis-R, F.-X., Maskos, K., Betz, M., Bergner, A., Huber, R., Suzuki, K., Yoshida, N., Nagase, H., Brew, K. and Bourenkov, G. P. (1997) Mechanism of inhibition of the human matrix metalloproteinase stromelysin-1 by TIMP-1. Nature. 389, 77-81 87. Higashi, S. and Miyazaki, K. (2003) Identification of a region of β-amyloid precursor protein essential for its gelatinase A inhibitory activity. J. Biol. Chem. 278, 14020-14028 88. Oh, J., Takahashi, R., Kondo, S., Mizoguchi, A., Adachi, E., Sasahara, R. M., Nishimura, S., Imamura, Y., Kitayama, H. and Alexander, D. B. (2001) The membrane-anchored MMP inhibitor RECK is a key regulator of extracellular matrix integrity and angiogenesis. Cell. 107, 789-800 89. Dufour, A. and Overall, C. M. (2013) Missing the target: matrix metalloproteinase antitargets in inflammation and cancer. Trends Pharmacol. Sci. 34, 233-242 90. Qin, H., Sun, Y. and Benveniste, E. N. (1999) The transcription factors Sp1, Sp3, and AP-2 are required for constitutive matrix metalloproteinase-2 gene expression in astroglioma cells. J. Biol. Chem. 274, 29130-29137 91. Le Floch, N., Rivat, C., De Wever, O., Bruyneel, E., Mareel, M., Dale, T. and Gespach, C. (2005) The proinvasive activity of Wnt-2 is mediated through a noncanonical Wnt pathway coupled to GSK-3β and c-Jun/AP-1 signaling. FASEB J. 19, 144-146

92. Nawrocki‐Raby, B., Gilles, C., Polette, M., Bruyneel, E., Laronze, J. Y., Bonnet, N., Foidart, J. M., Mareel, M. and Birembaut, P. (2003) Upregulation of MMPs by soluble E‐cadherin in human lung tumor cells. Int. J. Cancer. 105, 790-795 93. Yan, C. and Boyd, D. D. (2007) Regulation of matrix metalloproteinase gene expression. J. Cell. Physiol. 211, 19-26 94. Woessner, J. F. and Nagase, H. (2000) Matrix metalloproteinases and TIMPs. Oxford University Press 95. Parks, W. C., Wilson, C. L. and López-Boado, Y. S. (2004) Matrix metalloproteinases as modulators of inflammation and innate immunity. Nature Rev. Immunol. 4, 617-629 96. Lijnen, H. R., Van Hoef, B. and Collen, D. (2001) Inactivation of the serpin α 2-antiplasmin by stromelysin-1. Biochim. Biophys. Acta-Prot. Struct. Mol. Enzymol.. 1547, 206-213 97. Mott, J. D. and Werb, Z. (2004) Regulation of matrix biology by matrix metalloproteinases. Curr. Opin. Cell Biol. 16, 558-564 98. Frantz, C., Stewart, K. M. and Weaver, V. M. (2010) The extracellular matrix at a glance. J. Cell Sci. 123, 4195-4200 99. Rundhaug, J. E. (2005) Matrix metalloproteinases and angiogenesis. J. Cell. Mol. Med. 9, 267-285 100. Park, J. E., Keller, G.-A. and Ferrara, N. (1993) The vascular endothelial growth factor (VEGF) isoforms: differential deposition into the subepithelial extracellular matrix and bioactivity of extracellular matrix-bound VEGF. Mol. Biol. Cell. 4, 1317-1326 101. Belotti, D., Paganoni, P., Manenti, L., Garofalo, A., Marchini, S., Taraboletti, G. and Giavazzi, R. (2003) Matrix Metalloproteinases (MMP9 and MMP2) induce the release of Vascular Endothelial Growth Factor (VEGF) by ovarian carcinoma cells implications for ascites formation. Cancer Res. 63, 5224-5229

102. Bergers, G., Brekken, R., McMahon, G., Vu, T. H., Itoh, T., Tamaki, K., Tanzawa, K., Thorpe, P., Itohara, S. and Werb, Z. (2000) Matrix metalloproteinase-9 triggers the angiogenic switch during carcinogenesis. Nat. Cell Biol. 2, 737-744 103. Annes, J. P., Munger, J. S. and Rifkin, D. B. (2003) Making sense of latent TGFβ activation. J. Cell Sci. 116, 217-224 104. Mott, J. D. and Werb, Z. (2004) Regulation of matrix biology by matrix metalloproteinases. Curr. Opin. Cell Biol. 16, 558-564 105. Werb, Z. (1997) ECM and cell surface proteolysis: regulating cellular ecology. Cell. 91, 439-442 106. Xu, J., Rodriguez, D., Petitclerc, E., Kim, J. J., Hangai, M., Yuen, S. M., Davis, G. E. and Brooks, P. C. (2001) Proteolytic exposure of a cryptic site within collagen type IV is required for angiogenesis and tumor growth in vivo. J. Cell Biol. 154, 1069-1080 107. Hamano, Y. and Kalluri, R. (2005) Tumstatin, the NC1 domain of α3 chain of type IV collagen, is an endogenous inhibitor of pathological angiogenesis and suppresses tumor growth. Biochem. Biophys. Res. Commun. 333, 292-298 108. Gialeli, C., Theocharis, A. D. and Karamanos, N. K. (2011) Roles of matrix metalloproteinases in cancer progression and their pharmacological targeting. FEBS J. 278, 16-27 109. Pirilä, E., Sharabi, A., Salo, T., Quaranta, V., Tu, H., Heljasvaara, R., Koshikawa, N., Sorsa, T. and Maisi, P. (2003) Matrix metalloproteinases process the laminin-5 γ2-chain and regulate epithelial cell migration. Biochem. Biophys. Res. Commun. 303, 1012-1017 110. Schenk, S., Hintermann, E., Bilban, M., Koshikawa, N., Hojilla, C., Khokha, R. and Quaranta, V. (2003) Binding to EGF receptor of a laminin-5 EGF-like fragment liberated during MMP-dependent mammary gland involution. J. Cell Biol. 161, 197-209 111. Schor, S. L., Ellis, I. R., Jones, S. J., Baillie, R., Seneviratne, K., Clausen, J., Motegi, K., Vojtesek, B., Kankova, K. and Furrie, E. (2003) Migration-stimulating 35

factor a genetically truncated onco-fetal fibronectin isoform expressed by carcinoma and tumor-associated stromal cells. Cancer Res. 63, 8827-8836 112. Butler, G. S. and Overall, C. M. (2013) Matrix metalloproteinase processing of signaling molecules to regulate inflammation. Periodontol. 2000. 63, 123-148 113. Strieter, R. M., Polverini, P. J., Kunkel, S. L., Arenberg, D. A., Burdick, M. D., Kasper, J., Dzuiba, J., Van Damme, J., Walz, A. and Marriott, D. (1995) The functional role of the ELR motif in CXC chemokine-mediated angiogenesis. J. Biol. Chem. 270, 27348-27357 114. Van den Steen, P. E., Wuyts, A., Husson, S. J., Proost, P., Van Damme, J. and Opdenakker, G. (2003) Gelatinase B/MMP-9 and neutrophil collagenase/MMP-8 process the chemokines human GCP-2/CXCL6, ENA-78/CXCL5 and mouse GCP-2/LIX and modulate their physiological activities. Eur. J. Biochem. 270, 3739-3749 115. Tester, A. M., Cox, J. H., Connor, A. R., Starr, A. E., Dean, R. A., Puente, X. S., López-Otín, C. and Overall, C. M. (2007) LPS responsiveness and neutrophil chemotaxis in vivo require PMN MMP-8 activity. PLoS One. 2, e312 116. Lehti, K., Allen, E., Birkedal-Hansen, H., Holmbeck, K., Miyake, Y., Chun, T.-H. and Weiss, S. J. (2005) An MT1-MMP-PDGF receptor-β axis regulates mural cell investment of the microvasculature. Genes Dev. 19, 979-991 117. Bikfalvi, A. (2004) Platelet factor 4: an inhibitor of angiogenesis. In Semin. Thromb. Hemost. 30. 379-385 118. Opdenakker, G., Van den Steen, P. E. and Van Damme, J. (2001) Gelatinase B: a tuner and amplifier of immune functions. Trends Immunol. 22, 571-579 119. Levi, E., Fridman, R., Miao, H.-Q., Ma, Y.-S., Yayon, A. and Vlodavsky, I. (1996) Matrix metalloproteinase 2 releases active soluble ectodomain of fibroblast growth factor receptor 1. Proc. Nat. Acad. Sci. 93, 7069-7074 120. Macotela, Y., Aguilar, M. B., Guzmán-Morales, J., Rivera, J. C., Zermeño, C., López-Barrera, F., Nava, G., Lavalle, C., de la Escalera, G. M. and Clapp, C. (2006)

Matrix metalloproteases from chondrocytes generate an antiangiogenic 16 kDa prolactin. J. Cell Sci. 119, 1790-1800 121. Suenaga, N., Mori, H., Itoh, Y. and Seiki, M. (2005) CD44 binding through the hemopexin-like domain is critical for its shedding by membrane-type 1 matrix metalloproteinase. Oncogene. 24, 859-868 122. Yu, Q. and Stamenkovic, I. (1999) Localization of matrix metalloproteinase 9 to the cell surface provides a mechanism for CD44-mediated tumor invasion. Genes Dev. 13, 35-48 123. Sarrazin, S., Lamanna, W. C. and Esko, J. D. (2011) Heparan sulfate proteoglycans. Cold Spring Harb. Perspect. Biol. 3, a004952 124. Endo, K., Takino, T., Miyamori, H., Kinsen, H., Yoshizaki, T., Furukawa, M. and Sato, H. (2003) Cleavage of syndecan-1 by membrane type matrix metalloproteinase-1 stimulates cell migration. J. Biol. Chem. 278, 40764-40770 125. Brule, S., Charnaux, N., Sutton, A., Ledoux, D., Chaigneau, T., Saffar, L. and Gattegno, L. (2006) The shedding of syndecan-4 and syndecan-1 from HeLa cells and human primary macrophages is accelerated by SDF-1/CXCL12 and mediated by the matrix metalloproteinase-9. Glycobiology. 16, 488-501 126. Zemskov, E. A., Janiak, A., Hang, J., Waghray, A. and Belkin, A. M. (2006) The role of tissue transglutaminase in cell-matrix interactions. Front. Biosci. 11, 1057-1076 127. Belkin, A. M., Zemskov, E. A., Hang, J., Akimov, S. S., Sikora, S. and Strongin, A. Y. (2004) Cell-surface-associated tissue transglutaminase is a target of MMP-2 proteolysis. Biochemistry. 43, 11760-11769 128. Belkin, A. M., Akimov, S. S., Zaritskaya, L. S., Ratnikov, B. I., Deryugina, E. I. and Strongin, A. Y. (2001) Matrix-dependent proteolysis of surface transglutaminase by membrane-type metalloproteinase regulates cancer cell adhesion and locomotion. J. Biol. Chem. 276, 18415-18422

129. Shi, X., Gangadharan, B., Brass, L. F., Ruf, W. and Mueller, B. M. (2004) Protease-activated receptors (PAR1 and PAR2) contribute to tumor cell motility and metastasis. Mol. Cancer Res. 2, 395-402 130. Boire, A., Covic, L., Agarwal, A., Jacques, S., Sherifi, S. and Kuliopulos, A. (2005) PAR1 is a matrix metalloprotease-1 receptor that promotes invasion and tumorigenesis of breast cancer cells. Cell. 120, 303-313 131. Udagawa, N., Takahashi, N., Jimi, E., Matsuzaki, K., Tsurukai, T., Itoh, K., Nakagawa, N., Yasuda, H., Goto, M. and Tsuda, E. (1999) Osteoblasts/stromal cells stimulate osteoclast activation through expression of osteoclast differentiation factor/RANKL but not macrophage colony-stimulating factor. Bone. 25, 517-523 132. Lynch, C. C., Hikosaka, A., Acuff, H. B., Martin, M. D., Kawai, N., Singh, R. K., Vargo-Gogola, T. C., Begtrup, J. L., Peterson, T. E. and Fingleton, B. (2005) MMP-7 promotes prostate cancer-induced osteolysis via the solubilization of RANKL. Cancer Cell. 7, 485-496 133. Lillis, A., Mikhailenko, I. and Strickland, D. (2005) Beyond endocytosis: LRP function in cell migration, proliferation and vascular permeability. J. Thromb. Haemost. 3, 1884-1893 134. Rozanov, D. V., Hahn-Dantona, E., Strickland, D. K. and Strongin, A. Y. (2004) The low density lipoprotein receptor-related protein LRP is regulated by membrane type-1 matrix metalloproteinase (MT1-MMP) proteolysis in malignant cells. J. Biol. Chem. 279, 4260-4268 135. Emonard, H., Bellon, G., de Diesbach, P., Mettlen, M., Hornebeck, W. and Courtoy, P. J. (2005) Regulation of matrix metalloproteinase (MMP) activity by the low-density lipoprotein receptor-related protein (LRP). A new function for an “old friend”. Biochimie. 87, 369-376 136. Waring, P. and Müllbacher, A. (1999) Cell death induced by the Fas/Fas ligand pathway and its role in pathology. Immunol. Cell Biol. 77, 312-317 137. Tanaka, M., Itai, T., Adachi, M. and Nagata, S. (1998) Downregulation of Fas ligand by shedding. Nat. Med. 4, 31-36 38

138. Strand, S., Vollmer, P., van den Abeelen, L., Gottfried, D., Alla, V., Heid, H., Kuball, J., Theobald, M., Galle, P. R. and Strand, D. (2004) Cleavage of CD95 by matrix metalloproteinase-7 induces apoptosis resistance in tumour cells. Oncogene. 23, 3732-3736 139. Powell, W. C., Fingleton, B., Wilson, C. L., Boothby, M. and Matrisian, L. M. (1999) The metalloproteinase matrilysin proteolytically generates active soluble Fas ligand and potentiates epithelial cell apoptosis. Curr. Biol. 9, 1441-1447 140. Pan, G., Bauer, J. H., Haridas, V., Wang, S., Liu, D., Yu, G., Vincenz, C., Aggarwal, B. B., Ni, J. and Dixit, V. M. (1998) Identification and functional characterization of DR6, a novel death domain‐containing TNF receptor. FEBS Lett. 431, 351-356 141. Wajant, H., Pfizenmaier, K. and Scheurich, P. (2003) Tumor necrosis factor signaling. Cell Death Differ. 10, 45-65 142. Tam, E. M., Morrison, C. J., Wu, Y. I., Stack, M. S. and Overall, C. M. (2004) Membrane protease proteomics: Isotope-coded affinity tag MS identification of undescribed MT1–matrix metalloproteinase substrates. Proc. Natl. Acad. Sci. U. S. A. 101, 6917-6922 143. Ulbrich, H., Eriksson, E. E. and Lindbom, L. (2003) Leukocyte and endothelial cell adhesion molecules as targets for therapeutic interventions in inflammatory disease. Trends Pharmacol. Sci. 24, 640-647

144. Kamel-Reid, S. and Dick, J. (1988) Engraftment of immune-deficient mice with human hematopoietic stem cells. Science. 242, 1706-1709 145. Birbrair, A. and Frenette, P. S. (2016) Niche heterogeneity in the bone marrow. Ann. N. Y. Acad. Sci. 1370, 82-96 146. Nagasawa, T. (2007) The chemokine CXCL12 and regulation of HSC and B lymphocyte development in the bone marrow niche. In Osteoimmunology. pp. 69-75, Springer 147. Heissig, B., Hattori, K., Dias, S., Friedrich, M., Ferris, B., Hackett, N. R., Crystal, R. G., Besmer, P., Lyden, D. and Moore, M. A. (2002) Recruitment of stem and 39

progenitor cells from the bone marrow niche requires MMP-9 mediated release of kit-ligand. Cell. 109, 625-637 148. Dean, R. A., Cox, J. H., Bellac, C. L., Doucet, A., Starr, A. E. and Overall, C. M. (2008) Macrophage-specific metalloelastase (MMP-12) truncates and inactivates ELR+ CXC chemokines and generates CCL2,-7,-8, and-13 antagonists: potential role of the macrophage in terminating polymorphonuclear leukocyte influx. Blood. 112, 3455-3464 149. Haro, H., Crawford, H. C., Fingleton, B., Shinomiya, K., Spengler, D. M. and Matrisian, L. M. (2000) Matrix metalloproteinase-7–dependent release of tumor necrosis factor-α in a model of herniated disc resorption. J. Clin. Invest. 105, 143-150 150. Cathcart, J., Pulkoski-Gross, A. and Cao, J. (2015) Targeting Matrix Metalloproteinases in Cancer: Bringing New Life to Old Ideas. Genes & Diseases. 2, 26-34 151. Murphy, G. and Nagase, H. (2008) Reappraising metalloproteinases in rheumatoid arthritis and osteoarthritis: destruction or repair? Nature Clin. Prac. Rheumatol.. 4, 128-135 152. Balbín, M., Fueyo, A., Tester, A. M., Pendás, A. M., Pitiot, A. S., Astudillo, A., Overall, C. M., Shapiro, S. D. and López-Otín, C. (2003) Loss of collagenase-2 confers increased skin tumor susceptibility to male mice. Nat. Genet. 35, 252-257 153. Abdel-Ghany, M., Cheng, H.-C., Elble, R. C. and Pauli, B. U. (2001) The breast cancer β4 integrin and endothelial human CLCA2 mediate lung metastasis. J. Biol. Chem. 276, 25438-25446 154. Dean, R. A. and Overall, C. M. (2007) Proteomic discovery of metalloproteinase substrates in the cellular context by iTRAQ0 labeling reveals a diverse MMP-2 substrate degradome. Mol. Cell. Proteomics. 6, 611-623 155. McCawley, L. J., Crawford, H. C., King, L. E., Mudgett, J. and Matrisian, L. M. (2004) A protective role for matrix metalloproteinase-3 in squamous cell carcinoma. Cancer Res. 64, 6965-6972

156. Jost, M., Folgueras, A. R., Frérart, F., Pendas, A. M., Blacher, S., Houard, X., Berndt, S., Munaut, C., Cataldo, D. and Alvarez, J. (2006) Earlier onset of tumoral angiogenesis in matrix metalloproteinase-19–deficient mice. Cancer Res. 66, 5234-5241 157. Savinov, A. Y., Remacle, A. G., Golubkov, V. S., Krajewska, M., Kennedy, S., Duffy, M. J., Rozanov, D. V., Krajewski, S. and Strongin, A. Y. (2006) Matrix metalloproteinase 26 proteolysis of the NH2-terminal domain of the estrogen receptor β correlates with the survival of breast cancer patients. Cancer Res. 66, 2716-2724 158. Perdigão, J., Reis, A. and Loguercio, A. D. (2013) Dentin adhesion and MMPs: a comprehensive review. J. Esthet. Restor. Dent. 25, 219-241 159. Yan, C., Wang, H., Aggarwal, B. and Boyd, D. D. (2004) A novel homologous recombination system to study 92 kDa type IV collagenase transcription demonstrates that the NF-κB motif drives the transition from a repressed to an activated state of gene expression. FASEB J. 18, 540-541 160. Dias, N. and Stein, C. (2002) Antisense oligonucleotides: basic concepts and mechanisms. Mol. Cancer Ther. 1, 347-355 161. Hasegawa, S., Koshikawa, N., Momiyama, N., Moriyama, K., Ichikawa, Y., Ishikawa, T., Mitsuhashi, M., Shimada, H. and Miyazaki, K. (1998) Matrilysin‐ specific antisense oligonucleotide inhibits liver metastasis of human colon cancer cells in a nude mouse model. Int. J. Cancer. 76, 812-816 162. Yonemura, Y., Endo, Y., Fujita, H., Kimura, K., Sugiyama, K., Momiyama, N., Shimada, H. and Sasaki, T. (2001) Inhibition of peritoneal dissemination in human gastric cancer by MMP-7-specific antisense oligonucleotide. J.Exp. Clin. Cancer Res. 20, 205-212 163. Remacle, A. G., Golubkov, V. S., Shiryaev, S. A., Dahl, R., Stebbins, J. L., Chernov, A. V., Cheltsov, A. V., Pellecchia, M. and Strongin, A. Y. (2012) Novel MT1-MMP small-molecule inhibitors based on insights into hemopexin domain function in tumor growth. Cancer Res. 72, 2339-2349

164. Dufour, A., Sampson, N. S., Li, J., Kuscu, C., Rizzo, R. C., DeLeon, J. L., Zhi, J., Jaber, N., Liu, E. and Zucker, S. (2011) Small-molecule anticancer compounds selectively target the hemopexin domain of matrix metalloproteinase-9. Cancer Res. 71, 4977-4988 165. Higashi, S., Hirose, T., Takeuchi, T. and Miyazaki, K. (2013) Molecular design of a highly selective and strong protein inhibitor against matrix metalloproteinase-2 (MMP-2). J. Biol. Chem. 288, 9066-9076 166. Johnson, A. R., Pavlovsky, A. G., Ortwine, D. F., Prior, F., Man, C.-F., Bornemeier, D. A., Banotai, C. A., Mueller, W. T., McConnell, P. and Yan, C. (2007) Discovery and characterization of a novel inhibitor of matrix metalloprotease-13 that reduces cartilage damage in vivo without joint fibroplasia side effects. J. Biol. Chem. 282, 27781-27791 167. Morales, R., Perrier, S., Florent, J.-M., Beltra, J., Dufour, S., De Mendez, I., Manceau, P., Tertre, A., Moreau, F. and Compere, D. (2004) Crystal structures of novel non-peptidic, non-zinc chelating inhibitors bound to MMP-12. J. Mol. Biol. 341, 1063-1076 168. Engel, C. K., Pirard, B., Schimanski, S., Kirsch, R., Habermann, J., Klingler, O., Schlotte, V., Weithmann, K. U. and Wendt, K. U. (2005) Structural basis for the highly selective inhibition of MMP-13. Chem. Biol. 12, 181-189 169. Li, W., Li, J., Wu, Y., Wu, J., Hotchandani, R., Cunningham, K., McFadyen, I., Bard, J., Morgan, P. and Schlerman, F. (2009) A selective matrix metalloprotease 12 inhibitor for potential treatment of chronic obstructive pulmonary disease (COPD): discovery of (S)-2-(8-(methoxycarbonylamino) dibenzo [b, d] furan-3-sulfonamido)-3-methylbutanoic acid (MMP408). J. Med. Chem. 52, 1799-1802 170. Biasone, A., Tortorella, P., Campestre, C., Agamennone, M., Preziuso, S., Chiappini, M., Nuti, E., Carelli, P., Rossello, A. and Mazza, F. (2007) α-Biphenylsulfonylamino 2-methylpropyl phosphonates: Enantioselective synthesis and selective inhibition of MMPs. Biorg. Med. Chem. 15, 791-799 42

171. Jacobsen, F. E., Lewis, J. A. and Cohen, S. M. (2006) A new role for old ligands: discerning chelators for zinc metalloproteinases. J. Am. Chem. Soc. 128, 3156-3157 172. Grams, F., Brandstetter, H., DAlò, S., Geppert, D., Krell, H.-W., Leinert, H., Livi, V., Menta, E., Oliva, A. and Zimmermann, G. (2001) Pyrimidine-2, 4, 6-triones: a new effective and selective class of matrix metalloproteinase inhibitors. Biol. Chem. 382, 1277-1285 173. Brandstetter, H., Grams, F., Glitz, D., Lang, A., Huber, R., Bode, W., Krell, H.-W. and Engh, R. A. (2001) The 1.8-Å crystal structure of a matrix metalloproteinase 8-barbiturate inhibitor complex reveals a previously unobserved mechanism for collagenase substrate recognition. J. Biol. Chem. 276, 17405-17412 174. Nielsen, B. B., Frydenvang, K. and Larsen, I. K. (1993) Structures of two ribonucleotide reductase inhibitors: 1-hydroxy-1-methylurea and 1-hydroxy-3-methylurea. Acta Crystallogr. Sect. C: Cryst. Struct. Commun. 49, 1018-1022 175. Campestre, C., Agamennone, M., Tortorella, P., Preziuso, S., Biasone, A., Gavuzzo, E., Pochetti, G., Mazza, F., Hiller, O. and Tschesche, H. (2006) N-Hydroxyurea as zinc binding group in matrix metalloproteinase inhibition: mode of binding in a complex with MMP-8. Bioorg. Med. Chem. Lett. 16, 20-24 176. Onaran, M. B., Comeau, A. B. and Seto, C. T. (2005) Squaric acid-based peptidic inhibitors of matrix metalloprotease-1. J. Org. Chem. 70, 10792-10802 177. Jacobsen, F. E., Lewis, J. A., Heroux, K. J. and Cohen, S. M. (2007) Characterization and evaluation of pyrone and tropolone chelators for use in metalloprotein inhibitors. Inorg. Chim. Acta. 360, 264-272 178. Agrawal, A., Johnson, S. L., Jacobsen, J. A., Miller, M. T., Chen, L. H., Pellecchia, M. and Cohen, S. M. (2010) Chelator fragment libraries for targeting metalloproteinases. ChemMedChem. 5, 195-199 179. Puerta, D. T., Griffin, M. O., Lewis, J. A., Romero-Perez, D., Garcia, R., Villarreal, F. J. and Cohen, S. M. (2006) Heterocyclic zinc-binding groups for use in

next-generation matrix metalloproteinase inhibitors: potency, toxicity, and reactivity. J. Biol. Inorg. Chem. 11, 131-138 180. Puerta, D. T., Lewis, J. A. and Cohen, S. M. (2004) New beginnings for matrix metalloproteinase inhibitors: identification of high-affinity zinc-binding groups. J. Am. Chem. Soc. 126, 8388-8389 181. Martin, D. P., Blachly, P. G., Marts, A. R., Woodruff, T. M., de Oliveira, C. s. A., McCammon, J. A., Tierney, D. L. and Cohen, S. M. (2014) ‘Unconventional’coordination chemistry by metal chelating fragments in a metalloprotein active site. J. Am. Chem. Soc. 136, 5400-5406 182. Brown, S., Bernardo, M. M., Li, Z.-H., Kotra, L. P., Tanaka, Y., Fridman, R. and Mobashery, S. (2000) Potent and selective mechanism-based inhibition of gelatinases. J. Am. Chem. Soc. 122, 6799-6800 183. Kleifeld, O., Kotra, L. P., Gervasi, D. C., Brown, S., Bernardo, M. M., Fridman, R., Mobashery, S. and Sagi, I. (2001) X-ray absorption studies of human matrix metalloproteinase-2 (MMP-2) bound to a highly selective mechanism-based inhibitor comparison with the latent and active forms of the enzyme. J. Biol. Chem. 276, 17125-17131 184. Renkiewicz, R., Qiu, L., Lesch, C., Sun, X., Devalaraja, R., Cody, T., Kaldjian, E., Welgus, H. and Baragi, V. (2003) Broad‐spectrum matrix metalloproteinase inhibitor marimastat–induced musculoskeletal side effects in rats. Arthritis Rheum. 48, 1742-1749 185. Whittaker, M., Floyd, C. D., Brown, P. and Gearing, A. J. (1999) Design and therapeutic application of matrix metalloproteinase inhibitors. Chem. Rev. 99, 2735-2776 186. Rao, B. G. (2005) Recent developments in the design of specific matrix metalloproteinase inhibitors aided by structural and computational studies. Curr. Pharm. Des. 11, 295-322 187. Stewart, A. O., Bhatia, P. A., Martin, J. G., Summers, J. B., Rodriques, K. E., Martin, M. B., Holms, J. H., Moore, J. L., Craig, R. A. and Kolasa, T. (1997) 44

Structure-activity relationships of N-hydroxyurea 5-lipoxygenase inhibitors. J. Med. Chem. 40, 1955-1968 188. Fabre, B., Ramos, A. and de Pascual-Teresa, B. (2014) Targeting matrix metalloproteinases: Exploring the dynamics of the s1′ pocket in the design of selective, small molecule inhibitors: Miniperspective. J. Med. Chem. 57, 10205-10219 189. Prouty, K. V. (1992) MMPI patterns in codependency: Before and after treatment. 190. Yang, H., Makaroff, K., Paz, N., Aitha, M., Crowder, M. W. and Tierney, D. L. (2015) Metal ion dependence of the matrix metalloproteinase-1 mechanism. Biochemistry. 54, 3631-3639 191. Bertini, I., Fragai, M., Lee, Y. M., Luchinat, C. and Terni, B. (2004) Paramagnetic metal ions in ligand screening: the Co(II) matrix metalloproteinase 12. Angew. Chem. Int. Ed. 43, 2254-2256 192. Bookbinder, L., Frost, G. I., Keller, G., Frey, G. J., Chang, H. W., & Short, J. M.

(2010) Temperature sensitive mutants of matrix metalloproteases and uses thereof. U.S. Patent Application No. 12/660,894. 193. Lee, M., Villegas‐Estrada, A., Celenza, G., Boggess, B., Toth, M., Kreitinger, G., Forbes, C., Fridman, R., Mobashery, S. and Chang, M. (2007) Metabolism of a highly selective gelatinase inhibitor generates active metabolite. Chem. Biol. Drug Des. 70, 371-382 194. Terp, G. E., Cruciani, G., Christensen, I. T. and Jørgensen, F. S. (2002) Structural differences of matrix metalloproteinases with potential implications for inhibitor selectivity examined by the GRID/CPCA approach. J. Med. Chem. 45, 2675-2684 195. Lang, R., Braun, M., Sounni, N. E., Noël, A., Frankenne, F., Foidart, J.-M., Bode, W. and Maskos, K. (2004) Crystal structure of the catalytic domain of MMP-16/MT3-MMP: characterization of MT-MMP specific features. J. Mol. Biol. 336, 213-225 45

1.8 Tables and figures

Table 1.1: Overview of MMP sub-classes, structural elements, and ECM substrates

Structural element c,d

a,b MMP sub-class Pro cd Hpx TMD ECM substrate

CS Furin FN-II Type-I GpI Cys- array

Gelatinases

MMP2 + - + + - - - *gelatin, *elastin, collagen(I-XI),

MMP9 + - + + - - - fibronectin, laminin, proteoglycan

Collagenases

MMP1 + - - + - - - *collagen (I-XIV), gelatin,

MMP8 + - - + - - - elastin, fibronectin,

MMP13 + - - + - - - glycoprotein, proteoglycan

MMP18 + - - + - - -

Stromelysins

MMP3 + - - + - - - *proteoglycan, *fibronectin,

MMP10 + - - + - - - gelatin, elastin,

MMP11 - + - + - - - collagen (III- XI),laminin,

Matrilysins

MMP7 + ------*Insulin, Proteoglycan,

MMP26 + ------fibronectin, gelatin, elastin,

collagen (I, IV)

Metalloelastases

MMP12 + - - + - - - *elastin, Proteoglycan, fibronectin

MMP19 + - - + - - - gelatin, collagen( I, IV,V),laminin

Enamelysins

MMP20 + + - + - - - *amelogenin, *ameloblastin,

MMP21 + - - + - - - *enamelin, gelatin

Epilysin

MMP28 + + - + - - - Unidentified

Type I MT-MMPs

MMP14 + + - + + - - *collagen ( I-III),

MMP15 + + - + + - - proteoglycan,

MMP16 + + - + + - - fibronectin, gelatin,

MMp24 + + - + + - - glycoprotein

GPI MT-MMP

MMP17 + + - + - + - *gelatin, *fibronectin,

MMP25 + + - + - + - collagen IV

CS MT-MMP

MMP23 - + - + - - + *gelatin, fibronectin

a Preferred physiological substrate indicated by * b. MMPs’ ECM substrates were referred to [5, 192] c “ +” indicates the element’s appearance in structure , “ –“ indicates the element’s absent in structure

d. Structural elements abbreviation : Pro = Pro-domain; CS = cysteine – switch moiety

(bait region: PRCGxPD) in pro-domain; Furin = furin-recognized sequence

(RX[R/K]R) in pro-domain; cd = catalytic domain, FN-II = FN-II repeat moiety in catalytic domain; Hpx = hemopexin-like domain; TMD = transmembrane-anchoring

domain; Type-I= type-I transmembrane domain; GPI=glycosylphosphatidylinositol

membrane-anchoring domain; Cys array = cysteine array transmembrane domain.

Table 1.2: Overview of cell-specific expression of MMPs under normal physiological conditions

Host Cell localized MMPs

Neutrophils MMP8, MMP9 Macrophages MMP1, MMP2, MMP7, MMP9, MMP12, MMP14 Lymphocytes progenitor MMP3, MMP9, MMP25 Mast Cells MMP2, MMP9 Hematopoietic stem cells MMP2, MMP9, MMP14 Endothelial Cells MMP2, MMP3, MMP7, MMP14, MMP15, MMP16, MMP19 Dendritic cells MMP1, MMP2, MMP3, MMP9, MMP19 Fibroblasts MMP1, MMP2, MMP3, MMP9, MMP11, MMP13, MMP14, MMP15, MMP19 Osteoblasts MMP9, MMP13, MMP14 Chondrocytes MMP14 Epithelial cells MMP14 Adipocytes MMP14 Myeloid progenitor MMP14 Glial cells MMP7, MMP14,MMP16, MMP24 T-cells MMP3, MMP14, B-cells MMP14 Melanocytes MMP16 Germ cells MMP17, MMP28

Table 1.3: Physiological and pathological events and identified substrates for select MMPs.

MMP Physiological events Pathological events Identified substratea,b

MMP1 embryonic Cancer CCL3/8/7/2/16/15/23/4,HNFa,

development SAAC, FN1, HPLN1

FIBA/FIBG/FIBB,A2AP

IGFBP3, GSN, SAA2, PRL

SPARC, IGHG1, SNCA, TAC1

BGN, CIQ, DCN

MMP7 embryotic Cancer CXCL9/11, IGFBP1, IGFBP2

development tissue damage IGFBP3, TNF-, ADAM28

immune response ANXA2, FGA, TFP1, FASLG

neurotransmission PLAU, SNAP2, FAS, SPRAC

SPP1, TNC, CDH1, AIAT, CTG,

DCN

MMP9 Immune response Cancer CXCL9/10/11/12, VPAR, VWF, A2M

Inflammation abdominal sepsis, TFP1, FN1, IFNB, IGFBP1, IL8,I NS

autoimmunity KISS1, LGALS3, MBP, PRL

brain edema, SERPINE2, SNCA, TAC1

hemorrhage AIAT, APP, BGN

MMP12 host defense Cancer CXCL9/10/11, VPAR, VWF, FGA

Inflammation Autoimmunity FN1, FMOD, TFP1, LPA, PLG,

Immune response PRELP, MBP, SPRAC, ELN, FA12,

AIAT, BGN, DCN

MMP16 Unidentified Cancer KISS1, APP, MMP14, proMMP2/9

aMMPs substrates from Topfinder 3.1 bAbbreviations: CCL = chemokine (C-C motif) ligand, CXCL8/9/10/11/12= chemokine (C-X-C motif) ligand, BGN = biglycan, VPARP = Vault poly(ADP-ribose) polymerase, ADAM = A disintegrin and metalloproteinase, HNFa = hepatocyte nuclear factors a, VWF = Von Willebrand factor, ANXA1 = Annexin A1, A2M = alpha-2-macroglobulin, FGA = fibrinogen alpha chain, TFP1 = tissue factor pathway inhibitor, FN1 = fibronectin 1, HPLN1 = hyaluronan and proteoglycan link protein 1, IFNB1 = Interferon, Beta 1, fibroblast, FMOD = fibromodulin, FASLG = FAS-ligand, FIBA = fibril family structure subunit, IGFBP 1/2/3 = Insulin-like growth factor-binding protein 1/2/3, A2AP = alpha 2-antiplasmin, LPA = lipoprotein(a) , INS = insulin, PLG = plasminogen, GSN = gelsolin , KISS1 = kisspeptin, PRELP = , proline/arginine-Rich End Leucine-Rich Repeat Protein, PLAU = plasminogen Activator, SNAP2 = soluble NSF attachment protein, SAA2 = serum amyloid A2, MBP = myelin basic protein, FASR = apoptosis antigen 1 receptor, PRL = prolactin, SRC = proto-oncogene tyrosine-protein kinase, IGHG1 = immunoglobulin heavy constant gamma 1, SERPINE2 = serpin peptidase inhibitor, SPP1 = secreted phosphoprotein 1, SNCA = synuclein alpha, TAC1 = tachykinin precursor 1, CDH1 = cadherin-1, AIAT = alpha-1-antitrypsin, APP = amyloid beta A4 protein precursor, C1Q = complement component 1, q subcomponent, DCN = decorin

Table 1.4: Classes of MMP inhibitors (MMPi)

MMPi MMPi Structure a References

HA ZBGsb [180]

* * Cyclic HA-ZBG [180]

* Carboxylate ZBG * [35]

* Phosphorus-based ZBG [35]

* * Nitrogen-based ZBG [35]

* * Mechanism-based MMPi [193]

Structural-Based MMPi [166]

aAtoms thought to bind catalytic Zn(II) are labeled with “*” bAbbreviations: MMPi = MMP inhibitor(s), ZBG = zinc binding group, HA = hydroxamic acids

Figure 1.1: Domain structure of represented MMPs. Abbreviations: Structural elements abbreviation : Pro = Pro-domain; CS = cysteine – switch moiety (bait region: PRCGxPD) on pro-domain; Furin = furin-recognize sequence (RX[R/K]R) on pro-domain; CD = catalytic domain, FN-II = FN-II repeat moiety om catlytic domain; Hpx = hemopexin-like domain; TMD = membrane-anchoring domain; Type-I= type-I transmembrane domain GPI=glycosylphosphatidylinositol membrane-anchoring domain; Cys array = cysteine array transmembrane domain,

Figure 1.2: Activation of pro-MMPs via cysteine switch. A. Extracellular activation of secreted MMPs: Metalloproteinase and serine protease cleave CS-motif and hinge region to release catalytic domain from pro-domain’s auto-inhibition. B. Intracellular activation of MT-MMPs. PC can cleave the furin-recognize sequence to activate MMPs intracellularly. Abbreviations: Pro = Pro-domain; CS = cysteine – switch moiety (bait region: PRCGxPD) on pro-domain; Furin = furin-recognize sequence (RX[R/K]R) on pro-domain; CD = catalytic domain, Hpx = hemopexin-like domain;

TMD = membrane-anchoring domain; PC = pro-protein convertases, SS= signaling sequence

Figure 1.3: Overlapped crystal structures of cdMMP7 (green, PDB#1MMP) and cdMMP16 (cyan, PDB#1RMB). Zn(II) ions are colored gray, and Ca(II) ions are colored blue. Right: Zn(II) binding sites are shown with metal binding amino acids labeled and S1’ substrate binding pockets shown[194, 195]. The Met-turn is colored orange, the specificity loop is colored magenta, and the MT-loop is colored purple.

Figure 1.4: Proposed reaction mechanisms for MMPs: Mechanism 1: Browner[42], mechanism 2: Manzetti[43], mechanism 3: Bertini[44], and mechanism 4: Siegbahn[45].

Figure 1.5: Transcriptional, translational, and physiological regulation of MMPs. [6, 62, 93] Pathways that up-regulate the expression or activities of MMPs are labeled with a black arrow and a symbol “+”. Pathways that down-regulate the expression or activities of MMPs are labeled with a red arrow and a symbol of “-“. Abbreviations: ECM = extracellular matrix, MT-MMPs = membrane-bound MMPs. AP1 = activator protein 1, NF-kB = nuclear factor kappa-light-chain-enhancer of activated B cells, ARE = AU-rich elements, Tcf-4 = transcription factor-4, HuR = Hu protein family, PCs = proprotein convertases, KSRP/FBP2 = KH-ype splicing regulatory protein, TIMP = tissue inhibitor of metalloproteinases, APP = -amyloid precursor protein. RECK = reversion-inducing-cysteine-rich protein with kazal motifs, ADAMT = disintegrin and metalloproteinase with thrombospondin motif, 2-MG = 2-macroglobulin, Dmt = DNA methyltransferase.

Figure 1.6: Overview of MMPs’ physiological roles, pathological roles and protective roles[6, 89, 104]. MMPs regulate multiple bio-functional events (purple rectangles) via three regulatory mechanisms: A: directly remodeling ECM components to create space for cell motility and angiogenesis (middle arrow), B: releasing bioactive cryptic fragment (in cyan rectangle) from cleaved ECM components to regulate cell motility and angiogenesis (left arrow), and C: releasing and processing multiple bioactive molecules (in light-blue rectangles) to regulate cellular activities (right arrows). MMPs releases / processes growth factors cytokine to regulate cell proliferation, growth, survival and angiogenesis (purple rectangles). MMPs release / process chemokines to regulate angiogenesis and leukocyte recruitment. MMPs process certain regulatory proteins (such as KISS1 or FASL) to regulate cell survival / growth and migration (purple rectangles). Dysregulated MMPs activities are shown with red arrows and result in multiple tumorigenesis events (in pink rectangles) and inflammatory diseases (in dark-red rectangle). Six MMPs have been identified as anti-drug targets (in green rectangle), showing protective roles in certain pathological processes (green arrows): MMP3/7 against metastasis, MMP8/19/26 potently inhibit tumor cell growth and invasion, and MMP2 reduces inflammation in arthritis.

Figure 1.7: Review of MMPi development and current strategies of therapeutic targeting MMPs: 1st generation MMPi (cyan rectangle) and second generation MMPi (blue rectangle) failed in Phase II/III clinical test majorly due to poor bioavailability, low clinical efficacy and severe side-effects (grey rectangles). For current developing strategies of therapeutic inhibiting MMPs (in red arrows), it involved disrupting MMPs transcription/ translation by TFi and ASONs (green rectangles) respectively, inhibiting proMMP’s activation by Hpx inhibitor (orange rectangle), and selective inhibiting MMPs catalytic domain( pink rectangle) by ZBG MMPi, mechanism-based MMPi, or structural-based MMPi (purple rectangle). Abbreviations: MMPi = MMP inhibitors, ZBG = zinc-binding group, cdMMP = MMPs’ catalytic domain, Hpx = hemopexin-like domain, Pro= pro-domain, TFi = transcriptional factor inhibitor, ASONs= antisense oligonucleotides

Figure 1.8: Overview of experimental approach used in this dissertation. *EXAFS was used to characterize the Co(II) and Zn(II)-containing analogs of cdMMP16s. Abbreviations: cdMMP = catalytic domain of MMP. Co(II)-subs cdMMP = Co(II) substituted cdMMP, ITC = isothermal titration calorimetry, CD = circular dichroism spectroscopy, ICP-AES = inductively coupled plasma atomic emission spectroscopy, MALDI-TOF = matrix-assisted laser desorption/ionization mass spectroscopy time of flight, EXAFS = extended X-ray absorption fine structure, NMR = nuclear magnetic resonance spectroscopy

Chapter 2

Biochemical and spectroscopic characterization of the catalytic domain of MMP16 (cdMMP16)

Fan Meng, Hao Yang, Mahesh Aitha, Sam George, David L. Tierney, and Michael W. Crowder. J Biol Inorg Chem. 2016 Jul 21(4):523-35. Department of Chemistry and Biochemistry, Miami University, Oxford, Ohio 45056

†This work was supported by the National Institutes of Health (P30-EB-009998 to the Center for Synchrotron Biosciences from the NIBIB, which supports beamline X3B at the NSLS) and the National Science Foundation (CHE-1151658 to MWC and DLT). The authors thank Sameer Al-Abdul-Wahid and Theresa Ramelot for assistance in obtaining 1H NMR spectra on the enzymes.

AUTHOR CONTRIBUTIONS FM, DLT, MWC conceived and coordinated the study and wrote the paper. FM performed work of preparing and characterizing as-isolated cdMMP7 and ZnCo-cdMMP7 with assistance from SG. FM and HY performed the stopped flow intrinsic fluorescence emission experiments, CD, UV-vis, and NMR studies on cdMMP16. DLT and MA performed the EXAFS studies and fitted EXAFS data

ABSTRACT

Membrane-bound matrix metalloproteinase 16 (MMP16/MT3-MMP) is considered a drug target due to its role(s) in disease processes such as cancer and inflammation. Biochemical characterization of MMP16 is critical for developing new generations of MMP inhibitors (MMPi), which exhibit high efficacies and selectivities. Herein, a modified over-expression and purification protocol was used to prepare the catalytic domain of MMP16 (cdMMP16). The resulting recombinant enzyme exhibited steady state kinetic constants of Km = 10.6 + 0.7 M and kcat = 1.14 + 0.02 s-1, when using FS-6 as substrate, and the enzyme bound 1.8 + 0.1 eq of Zn(II). The enzymatic activity of cdMMP16 is salt concentration-dependent, and cdMMP16 exhibits autoproteolytic activity under certain conditions, which may be related to an in vivo regulatory mechanism of MMP16 and of other membrane-type MMPs

(MT-MMPs). Co(II)-substituted analogs (Co2- and ZnCo) of cdMMP16 were prepared and characterized using several spectroscopic techniques, such as UV-Vis, 1H NMR, and EXAFS spectroscopies. A well-characterized cdMMP16 is now available for future inhibitor screening efforts.

KEYWORDS: membrane-bound matrix metalloproteinases (MT-MMPs), autoproteolysis, cobalt-substitution, zinc

2.1 Introduction

Matrix metalloproteinases (MMPs) are members of the zinc-dependent proteinase family, and the MMP family consists of 23 distinct enzymes, all of which are found in humans[1]. Among the 23 MMPs, sixteen of the enzymes are secreted and have been sub-divided into seven classes based on substrate specificities: gelatinases (MMP2 and MMP9), collagenases (MMP1, MMP8, MMP13, and MMP18), stromelysins (MMP3 and MMP10), metalloelastases (MMP12 and MMP19), enamelysins (MMP20 and MMP21), epilysin (MMP28), and matrilysins (MMP7, MMP11, and MMP26)[1]. Seven membrane-bound MMPs (also called MT (membrane type)-MMPs) have been sub-divided into three classes based on how the enzymes associate with the membrane: type-I transmembrane (TM) MMPs (MMP14, MMP15, MMP16, and MMP24), type-II glycosylphosphatidylinositol (GPI)-attached MMPs (MMP17 and MMP25), and type-III transmembrane cysteine array (Cs) MMP

(MMP23) [1, 2]. Initially, the main role of MMPs was thought to be hydrolysis of the components of the extracellular matrix (ECM), and MMPs were shown to be involved with the regulation of various physiological activities, such as tissue remodeling, new blood vessel formation, and tissue repair[3]. Several MMPs were shown to be expressed at high levels during tumorigenesis, notably during metastasis, malignancy, tumor cell invasion, and angiogenesis [4, 5]. Therefore, considerable efforts have been made to discover inhibitors of MMPs as anticancer agents. However, almost all MMP inhibitors have failed in clinical trials due to lack of specificity (multiple MMPs inhibited by a single inhibitor)[4, 6-8]. In addition, recent studies have shown that MMPs are involved in a number of important physiological processes, including the immune response, inflammation, and apoptosis[9-11]. In fact, Gutiérrez-Fernández et al. have classified MMPs as targets and anti-targets, signifying that MMPs have both beneficial and negative roles intracellularly[12]. If MMP inhibitors are going to be used as anti-cancer or anti-inflammation drugs, inhibitors that specifically target one

MMP are needed. In order to accomplish this task, detailed information about an MMP from each of the major subclasses is needed. Most MMPs share a common structure, consisting of three domains: N-terminal propeptide domain (including a signaling peptide and one zymogenic propeptide), catalytic domain, and C-terminal domain that often contains hemopexin sub-domains, and in the case of MT-MMPs, membrane-anchoring sub-domains[13]. The overall structure of the catalytic domains of MMPs is similar sharing a common spherical structure made up by two or three-helices, five -sheets, and variable random loops. The catalytic domains most often bind 2 equivalents of Zn(II) and 2-3 equivalents of Ca(II). One Zn(II) binding site is called the structural site, and Zn(II) is coordinated by 3 conserved His residues and one monodentate-bound Asp residue. The second Zn(II) binding site is called the catalytic site, and Zn(II) is coordinated by 3 conserved His residues[14, 15]. Depending on the enzyme studied, the fourth ligand to the catalytic Zn(II) is either a water/hydroxide in the catalytically-active form of the enzyme or a cysteine thiol from a Cys residue in the propeptide domain in the catalytically-inactive pro-MMP zymogen[16]. The 2-3 Ca(II) ions have a structural role and are coordinated by Asp and Glu residues[17]. The substrate specificities of the MMPs arise from substrate binding pockets (called (S1, S2, S3, S1’, S2’, and S3’ sites) adjacent to the catalytic Zn(II) site[8, 18, 19]. In most MMPs, the C-terminal domain consists of at least one hemopexin-like sub-domain, which contributes to protein/substrate interactions and protein/protein dimerization[20, 21]. In the case of MT-MMPs, the C-terminal domain contains TM, CS, or GPI sub-domains that anchor the MT-MMPs to the membranes[2]. MMP16, which belongs to the type-I MT MMP family, mainly localizes on melanocytes, microglial cells, and endothelial cells[2, 22-25]. High-level expression of MMP16 has been observed in melanomas, renal carcinomas, breast carcinomas, glioma, astrocytic cancers, and cardiomyopathy. These pathological processes were attributed to unregulated MMP16 proteolytic activity, possibly by hydrolyzing and activating MMP2 or MMP9 [22, 23, 26-29]. MMP16 has also been implicated in the 63

hydrolysis of the metastasis suppressor kisspeptin (KISS1)[30]. Reduced levels of KISS1 have been observed in non-small cell lung cancers and in melanomas and mediate several gonadotropin-involved pathways[31, 32]. MMP16 may also be involved in the hydrolysis of amyloid precursor protein, implying that MMP16 may have a role in development of Alzheimer's disease [33]. The crystal structure of the catalytic domain of MMP16 (cdMMP16) has been reported, and the structure showed the common Zn(II) and Ca(II) sites (Figure 2.1). The substrate binding pockets (S1, S2, S2’, and S3) in MMP16 are similar to those of other MT-MMPs[34]. The S1’ site, which contains Met271, Ile263, and Gln269, is more narrow and shallow than the S1’ site in most other MMPs[34]. The crystal structure also showed the characteristic MT loop (Tyr171 to Lys178), which is found in all MT-MMPs but not in any of the secreted MMPs. Although there have been many structural, physiological, and inhibition studies published on MMPs [7, 8, 10] , there are very few mechanistic studies on any MMP, particularly on the MT-MMPs. Herein, we describe detailed studies on cdMMP16 by multiple kinetic and spectroscopic methods. First, we demonstrate an auto-proteolytic activity of cdMMP16 when metal ions are added to a metal-free (apo) analog of the enzyme. Two different cobalt-substituted analogs of cdMMP16 were prepared and characterized: (1) a diCo (Co2-cdMMP16) analog with both Zn(II) ions replaced by Co(II) and (2) a ZnCo (ZnCo-cdMMP16) analog with Co(II) in the catalytic site and Zn(II) in the structural site. These Co(II)-substituted analogs of cdMMP16 were shown to be structurally and catalytically similar to the naturally-occurring, dinuclear Zn(II)-containing enzyme. The Co(II)-substituted analogs allow for future rapid freeze quench spectroscopic studies to characterize the enzyme during catalysis and for the characterization of MMP16/inhibitor complexes.

2.2 Materials and methods

Materials

The plasmid containing the full-length Homo sapiens MMP-16 was purchased from OriGene (Rockville, MD). Taq PCR reaction kit, NdeI, HindIII, and T4 ligation kit were purchased from New England Biolabs (Ipswich, MA). Pet26b and E. coli BL21(DE3) cells were purchased from Novagen (Madison, WI). Luria-Bertani (LB) medium was purchased from Invitrogen (Carlsbad, CA). Isopropyl-β-D-thiogalactoside (IPTG) was purchased from Anatrace (Maumee, OH). Tris(hydroxymethyl)aminomethane (Tris),

4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (Hepes), urea, NaCl, CaCl2,

ZnCl2, CoCl2, ethylenediaminetetraacetic acid (EDTA), phenanthroline (OP), D2O, arginine, and glutamate were purchased from Fisher Scientific (Hampton, New Hampshire). All buffer solutions and growth media were made with Barnstead NANOpure water. Chelex 100 resin was purchased from Biorad laboratories (Hercules, CA). Protein samples were concentrated by using Centricon ultrafiltration units purchased from EMD Millipore (Billerica, MA) or Amicon centrifuge units (YM-10 membranes) from Sigma Aldrich. Fluorogenic peptide FS-6

(MCA-Lys-Pro-Leu-Gly-Leu-DNP-Dpa-Ala-Arg-NH2) was purchased from Sigma Aldrich (St. Louis, MO). The MMP substrate α BML128 (DNP-Pro-Leu-Ala-Leu-Trp-Ala-Arg-OH) and reference peptide for FS-6 (Mca-Pro-Leu-OH) were purchased from Enzo Life Science (Farmingdale, NY).

Methods

Over-expression, refolding, and purification of cdMMP16 a cDNA, which encodes for the catalytic subunit of MMP16 (amino acid residues 124 to 292)[34] was PCR-amplified from a commercially-available plasmid containing the full-length, human MMP16 gene with Forward primer

(5’-AAAAACATATGGGACAGAAATGGCAGCACA-3’) and reverse primer (5’-AAAAAGCTTCTATGGACCATATATCTTCTGGATGC-3’), which introduced NdeI and HindIII restriction sites at the 5’ and 3’ ends, respectively. The amplified PCR products were digested and ligated into the pET26b vector to construct the expression vector pET26b/cdMMP16. The DNA sequence for cdMMP16 was confirmed by DNA sequencing. BL21(DE3) E. coli cells were transformed with pET26b/cdMMP16, and the resulting cells were plated on a Luria-Bertani (LB) agar plates containing 25 g/mL of kanamycin. Single colonies were used to inoculate a starter culture of 50 mL LB-Kan, and the starter culture was shaken overnight at 37 oC. Ten milliliters of the starter culture was transferred into 1 L of LB-Kan, and the culture was allowed to

o shake at 37 C for 3 hours. When the culture reached an OD600nm > 0.7, one milliliter of 1 M isopropyl--D-galactopyranoside (IPTG) was added to the culture, and protein production was induced at 37 oC for 3-4 hours. The resulting E. coli cells were collected by centrifugation at 7,000 rpm for 10 minutes at 4 oC. The cell pellet was resuspended in 50 mM Tris, pH 7.5, containing 0.1% Brij-35 and 50 mM KCl, and the suspension was passed through a French press four times at a pressure of 1,500 psi. The ruptured cell mixture was centrifuged at 12,000 rpm for 20 minutes at 4 oC. With this over-expression system, cdMMP16 was processed into insoluble inclusion bodies. The inclusion bodies were washed by adding 35 mL of 50 mM Tris, pH 7.5, containing 2 M NaCl and 0.1% Brij-35, followed by centrifugation at 12,000 rpm for 20 minutes at 4 oC. The washed inclusion bodies were re-solubilized in 17.5 mL of 50 mM Tris, pH 8.5, containing 8 M urea per liter of growth culture. The inclusion body mixture was centrifuged at 12,000 rpm for 20 min in 4 oC to remove any insoluble components, and the cleared mixture was diluted with 5.5 mL of 50 mM Tris, pH 8.5, per liter of growth culture to reduce the concentration of urea to 6 M. The diluted inclusion body mixture was then dialyzed against 3X 500 mL (per liter of growth culture) of 50 mM

o Tris, pH 8.5, containing 100 mM NaCl, 10 mM CaCl2, 200 μM ZnCl2 at 4 C. The 66

dialysis buffer was changed every 3-4 hours. A final dialysis step was conducted overnight using 50 mM Tris, pH 7.5, containing 100 mM NaCl, 5 mM CaCl2, and 4

μM ZnCl2. The refolded protein solution was centrifuged at 12000 rpm for 20 min at 4 oC to remove any insoluble species, and the solution was concentrated to < 40 mL using an Amicon fitted with a YM10 membrane. The concentrated solution was applied to 50 mL Q-Sepharose column, which was pre-equilibrated with Buffer A (50 mM Tris, pH 7.5, containing 5 mM CaCl2)). Bound protein was eluted from the column using a linear gradient (100 mL) of Buffer B (50 mM Tris, pH 7.5, containing 500 mM NaCl). The column fractions were analyzed with UV-Vis spectrophotometry

(A280nm), and SDS-PAGE was used to evaluate the purity of the fractions. Fractions containing cdMMP16 were pooled and concentrated using an Amicon equipped with a YM-10 membrane. The concentration of the purified cdMMP16 was determined by

-1 -1 using Beer’s law and an extinction coefficient (A280nm) of 25,440 M cm . Each enzyme preparation was evaluated for metal content and catalytic activity. Only enzyme stocks that exhibited sufficient metal content (+ ~ 0.1 eqv cobalt) and catalytic activity (+ 10%) were used for subsequent studies.

Measurement of metal content in cdMMP16 samples. The amount of zinc and cobalt in the cdMMP16 samples was determined by using Inductively-Coupled Plasma with Atomic Emission Spectroscopy (ICP-AES, Perkin Elmer optima 7300DV). The cdMMP16 samples were diluted to 2 5 M with Chelex-treated 20 mM Hepes, pH

7.0, containing 5 mM CaCl2. A standard curve was established by using serially-diluted metal standards buffer (Zn, Co, Cu, and Fe) with concentrations from 1 to 10 M. The correlation coefficient of the calibration curse was approximately 0.999. The emission wavelengths were set to 213.856 nm for measuring zinc and 238.892 nm for measuring cobalt.

Steady state kinetic studies on cdMMP16. Steady state kinetic assays on cdMMP16 samples were conducted on a Synergy HT plate reader (BioTEK). Eexcitation and

Eemission were set to 330 nm and 420 nm, respectively. The total reaction volume was 67

100 M for each assay, and each reaction contained 10 nM of cdMMP16 and various concentrations (final concentrations of 10 – 60 M) of the FS-6 (Sigma Aldrich) substrate. The buffer was 50 mM Hepes, pH 7.0, containing 100 mM NaCl and 5 mM

o CaCl2, and the reactions were thermostated to 25 C. Each reaction was run for 10 minutes, and emission readings were taken every 30 seconds. Steady-state kinetic constants (kcat, Km) were obtained by fitting the data using GraphPad Prism 5.

Effect of salt on the catalytic activity of cdMMP16. Kinetic studies on cdMMP16 were conducted as described above, except the concentrations of cdMMP16 and FS-6 were held constant at 10 nM and 10 M, respectively. The reaction buffer was 20 mM

HEPES, pH 7.0, containing 5 mM CaCl2. The concentration of NaCl was varied from 0 to 800 mM in the reaction wells. The catalytic activity (defined as relative fluorescence units (RFU) per min) of cdMMP16 in buffer containing no added NaCl was set to 100% activity. The ratio of observed activity in buffers containing added NaCl and observed activity in buffer with no added NaCl was defined as relative activity.

Circular dichroism (CD) and fluorescence spectroscopic studies of cdMMP16. CD spectra of cdMMP16s were obtained on a JASCO J-810 CD spectropolarimeter using 1 cm cylindrical sample holders. The protein samples (final concentration approximately 10 M) were diluted with 20 mM sodium phosphate, pH 7.0, containing 100 mM NaCl [35]. Spectra (10 scans) were collected from 190 to 260 nm at 25 oC. CD spectra were fitted with DiChroWeb [36] to obtain estimations for secondary structure elements. Fluorescence emission spectra on cdMMP16 samples were obtained on a Perkin Elmer Luminescence spectrometer (Model LS-55). Protein concentrations were 10 M, and the buffer was 20 mM sodium phosphate, pH 7.0, containing 100 mM NaCl. The excitation wavelength was 280 nm, and emission spectra were obtained from 290 nm to 500 nm at 25 oC. Five scans were averaged to obtain the final spectra.

Stopped-flow kinetic studies on cdMMP16. Stopped-flow fluorescence studies on cdMMP16 were conducted at 25 oC on an Applied Photophysics spectrophotometer, equipped with a fluorescence detector. The excitation wavelength (Eexcitation) was set to 280 nm, and a 320 nm cut-off filter was used to exclude all emission wavelengths under 320 nm. After mixing, the final concentration of cdMMP16 was 1 M, and final concentrations of substrate BML-128 were varied between 3.2 and 103 M. The buffer was 20 mM Hepes, pH 7.0, containing 100 mM NaCl and 5 mM CaCl2. The reaction was monitored over 6 seconds, using a 125 s time interval. The resulting fluorescence emission spectra were fitted using a single exponential with slope (y = e(-kx) + bx + c) equation in the Applied Photophysics software package to obtain rate constants (kobs) for fluorescence decay. The determined kobs values were plotted against substrate concentration in KaleidaGraph to obtain values for kon, koff. and Ks. Dynafit 3.0 was used to simulate progress curves to obtain microscopic rate constants

(k1, k-1, and k2 ).

Preparation of metal-free and diCo(II)-substituted cdMMP16. Two methods were used to prepare metal-free cdMMP16: (1) a published method, which involved the use of EDTA as a metal chelator, to produce metal-free MMP1 and 3[37, 38] and (2) a revised method that involved the use of phenanthroline as the metal chelator and that maintained the presence of Ca(II) in the protein sample. In the revised method, purified cdMMP16 (40 M enzyme in 5 mL total volume), was quickly mixed with 5 mL of 4 mM 1,10-phenanthroline (OP) in 20 mM Hepes, pH 7.0, containing 5 mM

CaCl2. This mixture was incubated for 30 mins on ice and then dialyzed against 3X1L (3 hours each dialysis step) 20 mM Hepes, pH 7.0, containing 2 mM OP and 5 mM

CaCl2. Then, the mixture was dialyzed 2X1L (4 hours each dialysis step) of

Chelex-treated 20 mM Hepes, pH 7.0, containing 5 mM CaCl2. To the 10 mL

Zn(II)-free cdMMP16 sample, 100 L of 100 mM CoCl2 was added, and the sample was incubated for 1 hour at 4 oC. The sample was then centrifuged for 20 minutes at 12,000 rpm to remove any insoluble components and dialyzed 2X1L (8 hours each

dialysis step) against 20 mM Hepes, pH 6.8, containing 5 mM CaCl2 and 1 mM CoCl2. The sample was then dialyzed 3X1L (3-4 hours per dialysis step) against 20 mM

Hepes, pH 6.8, containing 5 mM CaCl2 and 1 M CoCl2. Metal content was determined by using ICP-AES, as described above. A reconstituted Zn2-cdMMP16 sample was made using this revised method except that CoCl2 was substituted by

ZnCl2.

Preparation of ZnCo-cdMMP16. The ZnCo heterodimetallic analog of cdMMP16 was prepared by using an exhaustive dialysis method. Purified cdMMP16 (150 M in 10 ml total volume) was dialyzed against 2X1L (12 hours each dialysis step)

Chelex-treated 20 mM Hepes, pH 7.0, containing 5 mM CaCl2 and 1 mM CoCl2. The sample was then dialyzed against 2X1L (3-4 hours for each step) 20 mM Hepes, pH

6.8, containing 5 mM CaCl2 and 1 M CoCl2. For samples requiring high concentrations, ZnCo-cdMMP16 was dialyzed against 1X1L 20 mM Hepes, pH 7.0, containing 100 mM NaCl, 5 mM CaCl2, 50 mM arginine, 50 mM glutamate, and 1 M

CoCl2.

Stability of cdMMP16 samples using SDS-PAGE and MALDI-TOF MS. Metal-free cdMMP16 was prepared using EDTA/phenanthroline as described above and previously[37, 39]. ZnCl2 (total concentration was 4 equivalents) or CoCl2 (total concentration of 4 equivalents) were added to 100 M metal-free cdMMP16 analog in 20 mM Hepes, pH 7.0, containing 5 mM CaCl and 100 mM NaCl, and the mixtures were allowed to incubate at 4 oC for 1, 2, 3, 4, 5, and 20 hours. The samples were evaluated using a 15% SDS-PAGE gel.

CoCl2 (final concentration of 2 equivalents) was added to 100 M metal-free cdMMP16 in 20 mM Hepes, pH 7.0, containing 5 mM CaCl and 100 mM NaCl, and the mixtures were allowed to incubate at 4 oC for 1, 3, and 20 hours. The resulting mixtures were diluted 1:10 with salt-free 50% acetonitrile (ACN), and the diluted

sample was mixed with 5 mg/mL 2-cyano-4-hydroxycinnamic acid (HCCA) in volume ratio of 1:8. The sample was analyzed using an AutoFlex MALDI-TOF mass spectrometer.

EXAFS spectroscopy for cdMMP16 samples. Zn2-cdMMP16 and Co2-cdMMP16 were concentrated to approximately 1 mM in 50 mM Hepes, pH 7.5, containing 100 mM NaCl, 5 mM CaCl2, and 10% glycerol. The concentrated samples were loaded into Lucite cuvettes with 6 m polypropylene windows and rapidly frozen. X-ray absorption data were collected with a Si(111) double-crystal monochromator at the National Synchrotron Light Source (NSLS) beamline X3B, and protein samples were kept at approximately 15 K in a Displex cryostat. Fluorescence excitation spectra were obtained with a 31-element solid-state Ge detector array, and Harmonic rejection was accomplished with a Ni focusing mirror. Data processing (collection and reduction) was accomplished using previously published protocols[40]. Each dataset consists of six (Zn K-edge) or eight (Co K-edge) scans, with duplicate data sets obtained on independently prepared samples. Averaged EXAFS data were converted from energy to k-space (E0 = 9680 eV for Zn; E0 = 7730 eV for Co) using EXAFSPAK[41] (EXAFSPAK is available free of charge from http://ssrl.slac.stanford.edu/exafspak.html) and fitted using Sixpack[42] (Sixpack is available free of charge from http://www.sams-xrays.com/#!sixpack/rovht). Detailed fitting results are summarized in Figures 2.2-2.3 and Table 2.1-2.2.

Optical spectroscopy of Co(II)-substituted cdMMP16 samples. Optical spectroscopic studies on Co(II)-substituted cdMMP16 samples were conducted on a Lambda 850

UV-vis spectrophotometer. For the Co2-cdMMP16 samples, metal-free cdMMP16, prepared using the phenanthroline/Ca(II) method, were diluted to approximately 200

M in 50 mM Hepes, pH 7.0, containing 5 mM CaCl2. CoCl2 was added directly to this sample, and the sample was allowed to sit on ice for one hour and was centrifuged at 12,000 rpm for 10 minutes before collecting the optical spectra. For ZnCo-cdMMP16, the samples were diluted to approximately 200 M in 50 mM 71

Hepes, pH 7.0, containing 5 mM CaCl2. Difference spectra were obtained by subtracting the spectrum of metal-free cdMMP16 from the spectra of the Co(II)-substituted cdMMP16.

NMR spectroscopy of Co(II)-substituted cdMMP16 samples. 1H NMR spectra of

ZnCo-cdMMP16 and Co2-cdMMP16 were conducted on on a Bruker ASX200 spectrometer (H = 200.13 MHz).The Co-substitution samples were concentrated to 1 mM in 50 mM Hepes, pH 7.0, containing 100 mM NaCl, 5 mM CaCl2 and 10%

D2O , and conducting to NMR studies. Spectra of samples collected in this manner were the average of approximately 300,000 transients consisting of 8k points over a 75 kHz spectral window. The presaturation pulse was typically 100 – 150 ms (approximately 1 W), centered at the water frequency, while the acquisition pulse was 3 s at full power, typically centered between +50 and +200 ppm. Prior to Fourier transformation, all FIDs were apodized using an exponential decay that introduced an additional line width of 30 Hz (0.15 ppm).

2.3 Results

Over-expression, refolding, purification, and characterization of the catalytic domain of MMP16 (cdMMP16). The gene for the cdMMP16 (amino acid Gly124 to Pro292) was ligated into pET26b, resulting in a plasmid called pET26b/cdMMP16, which was transformed into BL21(DE3) E. coli cells. The cells were cultured at 37 oC, and protein production was induced by addition of IPTG. As previously reported [34], cdMMP16 was produced in inclusion bodies, which were solubilized with 8M urea. Soluble cdMMP16 was obtained by multiple dialysis steps, as described in Materials and Methods, and the enzyme was purified with single step Q-Sepharose column chromatography (Figure 2.4). The procedure described above yielded 4-5 mg of soluble, purified cdMMP16 per liter of growth medium. The MALDI-TOF mass spectrum of the purified protein showed a dominant peak at 18,820 m/z. ICP-AES revealed that cdMMP16 contains 1.8 + 0.1 equivalents of Zn(II); the amount of Ca(II)

was not determined using ICP-AES due to the high background amounts of Ca(II) in our buffer. Steady-state kinetic studies using FS-6 (Figure 2.5A)[43] as substrate

-1 revealed that cdMMP16 exhibits a Km of 10.6 + 0.7 M and a kcat of 1.14 + 0.02 s with this substrate (Table 2.3). Unlike previous studies on cdMMP7 [39], cdMMP16 loses significant catalytic activity in the presence of NaCl at concentrations greater than 100 M (Figure 2.6). The analysis of CD spectra of cdMMP16 reveal that the purified, recombinant enzyme has 26% -helix, 16% -sheet, and 58% random coil, which are similar values to other reported MMPs [37, 38] (Figure 2.7 and Table 2.5).

Stopped-flow fluorescence kinetic studies. In a previous study on cdMMP1[37], the fluorescence emission of Trp159 changed during the hydrolysis reaction. Trp159 is highly-conserved in all MMPs, including MMP16, and it is approximately 10.8 Å away from the catalytic zinc site in the enzyme [34, 37]. Therefore, we used stopped-flow fluorescence to monitor the reaction of Zn2-cdMMP16 with six different concentrations (3.2, 6.4, 12.9, 56.5, and 103.5 M) (Figure 2.8) of the fluorescent substrate DNP-Pro-Leu-Ala-Leu-Trp-Ala-Arg-OH (Figure 2.5B). During the first 500 sec of the reaction, there was a rapid decrease in intrinsic tryptophan fluorescence, and a much slower increase over the next 1 second. We attributed the rapid emission decrease to quenching of the conserved Trp159, upon substrate binding. The subsequent increase in fluorescence is due to fluorescence of the product and to the conserved tryptophan, as the substrate/product is no longer bound to the enzyme. Simultaneous fitting of the concentration-dependent stopped-flow fluorescence data

(Figure 2.8A) to the kinetic mechanism in Table 2.4, yielded values for k1, k-1, and k2

(Table 2.4 lists k-1, k2 and KS). The rate of intrinsic tryptophan fluorescence decay

(kobs) was plotted against the substrate concentrations to yield a straight line,

-1 -1 -1 indicating values of kon = 0.38 M s , koff = 1.42 s , and a KS = 3.71 M [44, 45] (Figure 2.8B).

Generation of metal-substituted analogs of cdMMP16. Previously, metal-substituted analogs of MMPs 1, 3, 7 and others have been prepared by using metal chelators to 73

make apo (metal-free) analogs, and Co(II) was added to the apoenzyme to prepare the Co(II)-substituted enzymes[37, 39, 46, 47]. In all cases, except MMP12 and our recent study on MMP1, the resulting metal-substituted enzymes were only characterized by steady-state kinetics, without spectroscopic examination [37, 47]. There are no such studies on any membrane-bound MMP; therefore, we wished to prepare Co(II)-substituted cdMMP16 for future kinetic/spectroscopic studies. Our initial attempt to prepare a metal-free analog of cdMMP16 involved mixing purified cdMMP16 with an equal volume of solution containing 2 mM phenanthroline and 3 mM EDTA, followed by dialysis. Metal analyses demonstrated that the resulting enzyme contains < 0.1 equivalents of Zn(II), and SDS-PAGE showed that the apoenzyme is stable at 4 oC for more than 20 hours (Figure 2.9A). The addition of 4 equivalents of Zn(II) or Co(II) (and 5 mM Ca(II)) resulted in significant hydrolysis of cdMMP-16 over time (Figure 2.9B), with cdMMP16 being cleaved into at least 3 smaller pieces. The catalytic activity of the cleaved cdMMP-16 could not be rescued by addition of Zn(II) or Co(II) to the metal-free analog (Figure 2.10). We also monitored the hydrolysis time course using MALDI-TOF mass spectrometry (Figure 2.11). The MALDI-TOF mass spectrum of cdMMP16, which was incubated for 5 minutes with 2 eq of Co(II) and 5 mM Ca(II), shows a dominant peak at 18,816 m/z and a smaller peak at 19,023 m/z. The latter peak is most likely due to cdMMP16 bound to two equivalents of Co(II) and 2 equivalents of Ca(II). MALDI-TOF mass spectra of the sample over time shows a decrease in the 18,816 and 19,023 m/z peaks and a corresponding increase in a signal at 15,922 m/z. Interestingly, cdMMP16 is also hydrolyzed when only Ca(II) is added to the metal-free enzyme (data not shown). The addition of acetohydroxamic acid (AHA), which is a known inhibitor of other MMPs, did not prevent the observed auto-proteolysis. This result demonstrates that metal-substituted analogs of cdMMP16 cannot be generated using the previously reported method [37, 39, 46, 47].

Modified procedure to prepare Co2-cdMMP16. To prepare Co2-cdMMP16, we made three significant changes to the published procedure: (1) we excluded EDTA from the preparation of metal-free cdMMP16, (2) we maintained CaCl2 at a concentration of 5 mM in all steps, and (3) we excluded additional NaCl in any of the buffers. Metal-free cdMMP16, using the modified procedure, was prepared as described in Materials and Methods, and metal analyses showed that the enzyme binds < 0.1 eq of Zn(II). The most active Co(II)-substituted cdMMP16 was recovered when 1 mM CoCl2 was added to the metal-free enzyme (40 M), and the mixture was allowed to incubate for

1 hour. Any incubations with less than 1 mM CoCl2 resulted in lower catalytic activities. SDS-PAGE gels showed, however, that there was hydrolysis of the enzyme during the incubation period (Figure 2.12). Interestingly if the sample containing full-length and cleaved cdMMP16 was centrifuged (for example, sample in lane 7 of Figure 2.12), we could obtain a sample that was predominantly full-length cdMMP16 (lanes 8 and 12, Figure 2.12). We believe that the hydrolyzed enzymes require the presence of NaCl, which we excluded in all steps with this procedure, for solubility. The resulting full-length Co(II)-substituted cdMMP16 binds 1.8 eq of Co(II) and < 0.1 eq of Zn(II), suggesting that Co(II) is bound in the structural and catalytic sites.

The Co2-cdMMP16 analog exhibits a similar Km value but a lower kcat as compared to the Zn(II)-containing analog (Table 2.3). The Zn(II)-containing analog of cdMMP16

(called reconstituted Zn2-cdMMP16 in Table 2.3), which retains 95% of the catalytic activity of “as-isolated” cdMMP16, can also be recovered using the same method. Along with activity assays, metal analyses, and fluorescence and CD spectra, 1H

NMR spectra demonstrate that the as-isolated Zn2-cdMMP16 and the reconstituted

Zn2-cdMMP16 samples share similar structures (Tables 2.3 and 2.5 and Figures 2.13, 2.7, and 2.18). While this new procedure only allows for a 7-10% yield of reconstituted Zn2-cdMMP16 and a 30-35% yield of Co2-cdMMP16, it does yield a full-length analog for spectroscopic studies. We believe that the inclusion of Ca(II) and exclusion of EDTA maintains the proper structure of cdMMP16 during the dialysis steps. A fluorescence emission 75

spectrum of cdMMP16 directly after purification showed an intense peak at 350 nm (Figure 2.13), while the emission spectrum of the metal-free enzyme, which was prepared with EDTA, yielded no significant corresponding peak (not shown). With our new method to prepare metal-substituted cdMMP16, the fluorescence emission spectra of the metal-free and Co(II)-substituted enzymes are very similar (Figure 2.13). Similarly, the CD spectra of cdMMP16 directly after purification, metal-free cdMMP16 prepared with EDTA, metal-free cdMMP16 prepared with no EDTA, and Co(II)-substituted cdMMP16 show all CD spectra are similar in shape and in composition (Figure 2.7, Table 2.5), albeit with different relative values of the ellipticity. These results are not surprising given that Ca(II) has a structural role in the MMPs [17].

EXAFS of Zn2- and Co2-cdMMP16. EXAFS spectroscopy was used to characterize

Zn2- and Co2-cdMMP16, in order to demonstrate that the Co(II) in Co2-cdMMP16 is bound similarly to Zn(II) in the Zn2-cdMMP16 enzyme. Fourier-transformed k3-weighted EXAFS of these two analogs, along with their best fits, are presented in Figure 2.14 and summarized in Table 2.6; detailed fitting results can be found in

Tables 2.1-2.2 and Figures 2.2-2.3. The Zn2-cdMMP16 EXAFS data were best modeled with 4 N/O ligands at an average of 2.00 Å, a distance that suggests both Zn(II) ions are 4-coordinate. Multiple scattering analyses indicate the presence of three histidine ligands per metal ion, and inclusion of 0.5 carbon scatterers at 2.53 Å improved the fit residual by 15%. These data are consistent with previous crystallographic results on MMP16 [34].

For Co2-cdMMP16, the EXAFS data were best modeled using 5 N/O ligands at an average distance of 2.12 Å, suggesting that both of the Co(II) sites are 5-coordinate. In Co(II)-substituted proteins, it is common that solvent-exposed Co(II) ions bind an additional solvent molecule[37, 48, 49]. However, the EXAFS cannot distinguish an average metal site made up of two five-coordinate metal ions or one four- and one six-coordinate metal. Multiple scattering analysis is consistent with

three histidine ligands per metal ion. Inclusion of 0.5 carbon scatterers, as above for

Zn2-cdMMP16, refined to an unreasonably short distance (2.31 Å) and only improved the fit residual by 5%. In sum, the EXAFS data show that the Zn2- and Co2-cdMMP16 analogs bind metal in similar sites, and that the Co2-cdMMP16 is an appropriate surrogate for the Zn(II)-analog, with the caveat that the Co(II) ion binds one or more extra water molecules.

Modified procedure to prepare ZnCo-cdMMP16. A ZnCo-cdMMP16 heterodimetallic analog with Zn(II) in the structural site and Co(II) in the catalytic site is preferred to probe for substrate and inhibitor binding to the catalytic site when using spectroscopic methods. Previously, ZnCo-analogs of MMP-1 and MMP-7 were prepared by using a phenanthroline-based procedure [37, 39]. However, the same procedure applied to cdMMP16 resulted in complete self-hydrolysis during the dialysis steps. We also attempted to prepare ZnCo-cdMMP16 by first preparing metal-free cdMMP16, using the method described above, and adding 500 M Zn(II) and 500 M Co(II) to the metal-free enzyme. Only ZnZn-cdMMP16 (8 % overall yield) was recovered using this method. Therefore, we resorted to a dialysis-based method in which Zn2-cdMMP16 was exhaustively dialyzed against 1 mM CoCl2. The rationale was that Co(II) would substitute for the more loosely-bound Zn(II), which we hypothesized to be at the catalytic zinc site. After dialysis to remove any unbound metal ions, the resulting enzyme was shown to be full-length and to contain 0.7 equivalents of Co(II) and 0.8 equivalents of Zn(II). The overall yield from this procedure was 70-75%. The ZnCo-analog obtained exhibited steady-state kinetic constants (Table 2.3), a fluorescence emission spectrum (Figure 2.13), and a CD spectrum similar to those of Zn2-cdMMP16 (Figure 2.7; Table 2.5).

Optical and NMR spectroscopy of Co(II)-substituted cdMMP-16 analogs. The optical difference spectra (Co(II)-substituted enzyme – metal-free enzyme) for

Co2-cdMMP16 and ZnCo-cdMMP16 are shown in Figure 2.15. The spectrum of

Co2-cdMMP16 showed two signals: (1) a small peak at 475 nm, which we attribute to 77

Co(III) [48, 50, 51], and (2) a broad peak at 530 nm with shoulders at 550 and 585 nm, which we attribute to ligand field bands of high-spin Co(II). The molar absorptivity of the peak at 530 nm was 267 M-1cm-1, suggesting that this sample contains a mixture of 4/5 coordinate high-spin Co(II) [37, 49, 52, 53]. The optical spectrum of ZnCo-cdMMP16 shows the peak at 475 nm and a broad, unresolved envelope of

-1 -1 ligand field bands (530 approximately 50 M cm ). The peak at 530 nm suggests 6-coordinate Co(II).

The metal ligand environments in Co2-cdMMP16 and ZnCo-cdMMP16 were also examined using 1H NMR spectroscopy (Figure 2.16). Due to protein stability issues, we were unable to obtain NMR spectra of cdMMP16 samples in 90% D2O.

However, the NMR spectrum of Co2-cdMMP16 is very similar to that of

Co2-cdMMP1[37], providing the best evidence that Co(II) occupies both sites in this analog. Following the previous assignments for Co2-cdMMP1, we tentatively assign the peaks at 63 and 48 ppm to NH protons on histidines bound to Co(II) in the structural site and peaks at 86 and 78 ppm to NH protons on histidines bound to Co(II) in the catalytic site. Like ZnCo-cdMMP1[37], the 1H NMR spectrum of ZnCo-cdMMP16 was of poorer quality, although there is evidence for the two His resonances assigned to the catalytic site above ( > 80 ppm), suggesting the Co(II) in this analog is indeed bound at the catalytic site.

2.4 Discussion

Previously, Lang et al. described a procedure to over-express and purify cdMMP16 [34]. This method involved over-expression of the enzyme in BL21(DE3) E. coli cells and a pCTBR2 over-expression plasmid. Using this method, cdMMP16 was processed into inclusion bodies, the enzyme was solubilized using 8 M urea and folded by slow removal of the urea in several dialysis steps. We used this published method early on in our studies; however, our percent recovery of active cdMMP16 was low. Therefore, we optimized the procedure by (1) solubilizing the inclusion

bodies in 8 M urea, diluting the urea to 6 M (to produce an unfolded enzyme concentration of 1 mg mL-1), and centrifuging the resulting sample before the first dialysis step[54] and (2) increasing the Zn(II) concentration to 200 M in the refolding dialysis buffers. The method described in Materials and Methods yielded 4-5 mg of cdMMP16 per liter of growth medium. cdMMP16 was shown to bind 1.8 equivalents of Zn(II) and exhibit a molecular mass of 18.8 kDa, steady state kinetic

-1 constants of Km = 10.6 + 0.7 M and kcat = 1.14 + 0.02 s , and a secondary structure of 26% -helix, 16% -sheet, and 58% random coil. While the previously-reported study[34] on cdMMP16 did not provide biochemical analysis of the recombinant enzyme, our data show that cdMMP16 is similar to cdMMP1[37]. Nonetheless, our analyses revealed two significant differences with our purified, recombinant cdMMP16 as compared to other characterized MMPs. First, cdMMP16 exhibited maximal catalytic activity in low salt containing buffers (Figure 2.6). In contrast, cdMMP-7 exhibited maximal activities in buffers containing > 1 M NaCl[55]. Fluorescence emission spectra of cdMMP16 in the presence of different concentrations of salt do not show a shift in the fluorescence emission spectrum, indicating unfolding of the enzyme[56]. However, there is a reduction in the intensities of the fluorescence emission spectra in the presence of salt (Figures 2.17). It is not clear why high salt conditions affect the microenvironment around Trp159, but it is clear that the environment around Trp159 (Figure 2.8) affects catalysis. Second, cdMMP16 appears to exhibit significant autoproteolytic activity under certain conditions (Figures 2.9 and 2.11), and the hydrolyzed enzyme is catalytically-inactive

(Figure 2.10). Purified Zn2-cdMMP16 is stable over long periods of time. However when Zn(II) or Co(II) is added to metal-free cdMMP16, which was prepared by using previously-published procedures, the enzyme is cleaved into smaller pieces. Similar autoproteolytic activities have not been reported for other cdMMP’s. Efforts to determine the cleavage sites using mass spectrometry have been unsuccessful so far. Previous studies have suggested that MT-MMPs are responsible for hydrolyzing and activating other MMPs, especially MMP2 and MMP9 [2]. MMP16 has been reported 79

to inactivate MMP14 and MMP16 in vivo, presumbably by hydrolyzing the catalytic domain [57, 58]. The autoproteolysis that we observed with cdMMP16 in vitro is consistent with this observation; however, we only observed this proteolysis with metal-free cdMMP16, which was prepared using previously-published procedures, and not with fully-load Zn2-cdMMP16. This result suggests that there might be a population of metal-free MMP16 in vivo and that the autoproteolysis of metal-free MMP16 is an important way to regulate the activity of MMP16 (and possibly MMP14). Previously, efforts to prepare Co(II)-substituted analogs of the MMPs have been reported for cdMMP1, 3, 7, and 12 [37, 39, 46, 47]. These Co(II)-substituted analogs were prepared by generating metal-free analogs using EDTA and phenanthroline, and adding Co(II) to the metal-free enzymes. The formation of Co(II)-substituted MMP3 was confirmed only by activity assays, while Co(II)-substituted MMP1 and MMP12 were characterized using metal analyses and spectroscopic studies [37, 47]. Co(II)-substituted MMP7 was confirmed by metal analyses and activity assays [39]. Our studies show that Co(II)-substituted MMP16 cannot be generated using this approach, due to autoproteolysis. ZnCo-analogs of cdMMP1 and cdMMP7 have been prepared by using phenanthroline as the sole chelator, and results with cdMMP1 showed that only the catalytic Zn(II) was removed from the enzyme and that that the structural Zn(II) and Ca(II) ions remained in the enzyme [37]. For cdMMP16, we optimized this protocol to successfully prepare the

Co2-cdMMP16 analog with low yields. We believe that the presence of Ca(II), which is not chelated by phenanthroline, is required to prepare the Co2-MMP16 analog. Unlike with cdMMP1 and cdMMP7, our modified method using phenanthroline cannot be used to prepare a ZnCo-analog of cdMMP16 because phenanthroline removed both Zn(II) ions from cdMMP16. Therefore, we developed a dialysis-based method to prepare the ZnCo-cdMMP16 analog. This method yielded an analog with roughly equal equivalents of Co(II) and Zn(II), although the total metal equivalents were less than 2. 80

This analog exhibited a Km value similar to those of Zn2- and Co2-cdMMP16, and a kcat value that is roughly half of that of the Zn2-cdMMP16 analog (Table 2.3). UV-Vis of the ZnCo-cdMMP16 showed broad ligand field bands and weak 1H NMR, similar ZnCo-cdMMP1, with their appearance suggesting that Co(II) is bound to the catalytic site in this analog [37]. Efforts to more fully characterize the ZnCo analog of cdMMP16 are underway. While it is clear that the MMPs are important potential drug targets, most of the MMPs are challenging enzymes with which to work because the enzymes are often over-expressed in insoluble forms and the stability of the enzymes varies. Indeed, MMP16 (and possibly other MT-MMPs) is particularly challenging because of autoproteolysis. This work describes efforts to overcome these challenges and describe our efforts to prepare a stable, well-characterized analog of cdMMP16, which is now available for future inhibition studies. In addition, Co(II)-substituted analogs (Co2- and ZnCo-) are now available for characterization of MMPi-MMP16 inhibitor complexes.

2.5 References

1. Nagase, H., Visse, R. and Murphy, G. (2006) Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc Res. 69, 562-573 2. Itoh, Y. (2015) Membrane-type matrix metalloproteinases: Their functions and regulations. Matrix Biol. 44–46, 207-223 3. Rodriguez, D., Morrison, C. J. and Overall, C. M. (2010) Matrix metalloproteinases: what do they not do? New substrates and biological roles identified by murine models and proteomics. Biochim. Biophys. Acta. 1803, 39-54 4. Gialeli, C., Theocharis, A. D. and Karamanos, N. K. (2011) Roles of matrix metalloproteinases in cancer progression and their pharmacological targeting. FEBS J. 278, 16-27 5. Kessenbrock, K., Plaks, V. and Werb, Z. (2010) Matrix metalloproteinases: Regulators of the tumor microenvironment. Cell. 141, 52-67 6. Fingleton, B. (2008) MMPs as therapeutic targets—Still a viable option? Semin. Cell Dev. Biol. 19, 61-68 7. Elisa, N., Tiziano, T. and Armando, R. (2007) Matrix Metalloproteinase Inhibitors: New challenges in the era of post broad-spectrum inhibitors. Curr. Pharm. Des. 13, 2087-2100 8. Jacobsen, J. A., Major Jourden, J. L., Miller, M. T. and Cohen, S. M. (2010) To bind zinc or not to bind zinc: An examination of innovative approaches to improved metalloproteinase inhibition. Biochim. Biophys. Acta. 1803, 72-94 9. Van Lint, P. and Libert, C. (2007) Chemokine and cytokine processing by matrix metalloproteinases and its effect on leukocyte migration and inflammation. J. Leukoc. Biol. 82, 1375-1381 10. Dufour, A. and Overall, C. M. (2013) Missing the target: matrix metalloproteinase antitargets in inflammation and cancer. Trends Pharmacol. Sci. 34, 233-242

11. Butler, G. S. and Overall, C. M. (2013) Matrix metalloproteinase processing of signaling molecules to regulate inflammation. Periodontology 2000. 63, 123-148 12. Gutierrez-Fernandez, A., Fueyo, A., Folgueras, A. R., Garabaya, C., Pennington, C. J., Pilgrim, S., Edwards, D. R., Holliday, D. L., Jones, J. L., Span, P. N., Sweep, F. C., Puente, X. S. and Lopez-Otin, C. (2008) Matrix metalloproteinase-8 functions as a metastasis suppressor through modulation of tumor cell adhesion and invasion. Cancer Res. 68, 2755-2763 13. Visse, R. and Nagase, H. (2003) Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. Circ. Res. 92, 827-839 14. Bode, W., Fernandez-Catalan, C., Tschesche, H., Grams, F., Nagase, H. and Maskos, K. Structural properties of matrix metalloproteinases. Cell Mol Life Sci CMLS. 55, 639-652 15. Tallant, C., Marrero, A. and Gomis-Rüth, F. X. (2010) Matrix metalloproteinases: Fold and function of their catalytic domains. Biochim. Biophys. Acta. 1803, 20-28 16. Van Wart, H. E. and Birkedal-Hansen, H. (1990) The cysteine switch: a principle of regulation of metalloproteinase activity with potential applicability to the entire matrix metalloproteinase gene family. Proc. Natl. Acad. Sci. 87, 5578-5582 17. Massova, I., Kotra, L. P. and Mobashery, S. (1998) Structural insight into the binding motifs for the calcium ion and the non-catalytic zinc in matrix metalloproteases. Bioorg. Med. Chem. Lett. 8, 853-858 18. Rao, B. G. (2005) Recent developments in the design of specific matrix metalloproteinase inhibitors aided by structural and computational studies. Curr. Pharm. Des. 11, 295-322 19. Loretta, A., Magda, G., Ilaria, C., Susanna, M., Giovanni Francesco, F., Stefano, M., Paolo, A., Alessandra, T. and Massimo, C. (2008) Structural bases for substrate and inhibitor recognition by matrix metalloproteinases. Curr. Med. Chem. 15, 2192-2222 20. Tochowicz, A., Goettig, P., Evans, R., Visse, R., Shitomi, Y., Palmisano, R., Ito, N., Richter, K., Maskos, K. and Franke, D. (2011) The dimer interface of the 83

membrane type 1 matrix metalloproteinase hemopexin domain. Crystal structure and biological functions. J. Biol. Chem. 286, 7587-7600 21. Verma, R. P. and Hansch, C. (2007) Matrix metalloproteinases (MMPs): chemical–biological functions and (Q) SARs. Bioorg. Med. Chem. 15, 2223-2268 22. Ueno, H., Nakamura, H., Inoue, M., Imai, K., Noguchi, M., Sato, H., Seiki, M. and Okada, Y. (1997) Expression and tissue localization of membrane-types 1, 2, and 3 matrix metalloproteinases in human invasive breast carcinomas. Cancer Res. 57, 2055-2060 23. Nakada, M., Nakamura, H., Ikeda, E., Fujimoto, N., Yamashita, J., Sato, H., Seiki, M. and Okada, Y. (1999) Expression and tissue localization of membrane-type 1, 2, and 3 matrix metalloproteinases in human astrocytic tumors. Am. J. Pathol. 154, 417-428 24. Plaisier, M., Kapiteijn, K., Koolwijk, P., Fijten, C., Hanemaaijer, R., Grimbergen, J. M., Mulder-Stapel, A., Quax, P. H., Helmerhorst, F. M. and van Hinsbergh, V. W. (2004) Involvement of membrane-type matrix metalloproteinases (MT-MMPs) in capillary tube formation by human endometrial microvascular endothelial cells: role of MT3-MMP. Clin. Endocrinol. Metab. 89, 5828-5836 25. Yoshiyama, Y., Sato, H., Seiki, M., Shinagawa, A., Takahashi, M. and Yamada, T. (1998) Expression of the membrane-type 3 matrix metalloproteinase (MT3-MMP) in human brain tissues. Acta Neuropathol. 96, 347-350 26. Tatti, O., Gucciardo, E., Pekkonen, P., Holopainen, T., Louhimo, R., Repo, P., Maliniemi, P., Lohi, J., Rantanen, V., Hautaniemi, S., Alitalo, K., Ranki, A., Ojala, P. M., Keski-Oja, J. and Lehti, K. (2015) MMP16 mediates a proteolytic switch to promote cell-cell adhesion, collagen alignment, and lymphatic invasion in melanoma. Cancer Res. 75, 2083-2094 27. Cho, N. H., Shim, H. S., Rha, S. Y., Kang, S. H., Hong, S. H., Choi, Y. D., Hong, S. J. and Cho, S. H. (2003) Increased expression of matrix metalloproteinase 9 correlates with poor prognostic variables in renal cell carcinoma. Eur. Urol. 44, 560-566 84

28. Li, Y., Wang, Y., Yu, L., Sun, C., Cheng, D., Yu, S., Wang, Q., Yan, Y., Kang, C., Jin, S., An, T., Shi, C., Xu, J., Wei, C., Liu, J., Sun, J., Wen, Y., Zhao, S. and Kong, Y. (2013) miR-146b-5p inhibits glioma migration and invasion by targeting MMP16. Cancer Lett. 339, 260-269 29. Liu, J., van Mil, A., Aguor, E. N., Siddiqi, S., Vrijsen, K., Jaksani, S., Metz, C., Zhao, J., Strijkers, G. J., Doevendans, P. A. and Sluijter, J. P. (2012) MiR-155 inhibits cell migration of human cardiomyocyte progenitor cells (hCMPCs) via targeting of MMP-16. J. Cell. Mol. Med. 16, 2379-2386 30. Takino, T., Koshikawa, N., Miyamori, H., Tanaka, M., Sasaki, T., Okada, Y., Seiki, M. and Sato, H. (2003) Cleavage of metastasis suppressor gene product KiSS-1 protein/metastin by matrix metalloproteinases. Oncogene. 22, 4617-4626 31. Lee, J.-H., Miele, M. E., Hicks, D. J., Phillips, K. K., Trent, J. M., Weissman, B. E. and Welch, D. R. (1996) KiSS-1, a novel human malignant melanoma metastasis-suppressor gene. J. Natl. Cancer Inst. 88, 1731-1737 32. Skorupskaite, K., George, J. T. and Anderson, R. A. (2014) The kisspeptin-GnRH pathway in human reproductive health and disease. Hum. Reprod. Update. 20, 485-500 33. Ahmad, M., Takino, T., Miyamori, H., Yoshizaki, T., Furukawa, M. and Sato, H. (2006) Cleavage of amyloid-beta precursor protein (APP) by membrane-type matrix metalloproteinases. J. Mol. Biol. 139, 517-526 34. Lang, R., Braun, M., Sounni, N. E., Noel, A., Frankenne, F., Foidart, J. M., Bode, W. and Maskos, K. (2004) Crystal structure of the catalytic domain of MMP-16/MT3-MMP: characterization of MT-MMP specific features. J. Mol. Biol. 336, 213-225 35. Greenfield, N. J. (2006) Using circular dichroism spectra to estimate protein secondary structure. Nature Protoc. 1, 2876-2890 36. Whitmore, L. and Wallace, B. A. (2004) DICHROWEB, an online server for protein secondary structure analyses from circular dichroism spectroscopic data. Nucleic Acids Res. 32, W668-673 85

37. Yang, H., Makaroff, K., Paz, N., Aitha, M., Crowder, M. W. and Tierney, D. L. (2015) Metal ion dependence of the matrix metalloproteinase-1 mechanism. Biochemistry. 54, 3631-3639 38. Wetmore, D. R. and Hardman, K. D. (1996) Roles of the propeptide and metal ions in the folding and stability of the catalytic domain of stromelysin (matrix metalloproteinase 3). Biochemistry. 35, 6549-6558 39. Cha, J., Pedersen, M. V. and Auld, D. S. (1996) Metal and pH dependence of heptapeptide catalysis by human matrilysin. Biochemistry. 35, 15831-15838 40. Tierney, D. L. and Schenk, G. (2014) X-ray absorption spectroscopy of dinuclear metallohydrolases. Biophys. J. 107, 1263-1272 41. George, G. and Pickering, I. (1995) EXAFSPAK: A suite of computer programs for analysis of X-ray absorption spectra. SSRL, Stanford 42. Webb, S. (2005) SIXpack: a graphical user interface for XAS analysis using IFEFFIT. Physica Scripta. 2005, 1011 43. Neumann, U., Kubota, H., Frei, K., Ganu, V. and Leppert, D. (2004) Characterization of Mca-Lys-Pro-Leu-Gly-Leu-Dpa-Ala-Arg-NH2, a fluorogenic substrate with increased specificity constants for collagenases and tumor necrosis factor converting enzyme. Anal. Biochem. 328, 166-173 44. de Mol, N. J. and Fischer, M. J. (2008) Kinetic and thermodynamic analysis of ligand-receptor interactions: SPR applications in drug development. Handbook of Surface Plasmon Resonance, 123-172 45. Kuzmic, P. (2009) DynaFit--a software package for enzymology. Methods Enzymol. 467, 247-280 46. Cha, J., Sørensen, V. M., Ye, Q.-Z. and Auld, S. D. Selective replacement of the catalytic zinc of the human stromelysin-1 catalytic domain. J. Biol, Inorg, Chem, 3, 353-359 47. Bertini, I., Fragai, M., Lee, Y. M., Luchinat, C. and Terni, B. (2004) Paramagnetic metal ions in ligand screening: the Co(II) matrix metalloproteinase 12. Angew Chem (Int. Ed.). 43, 2254-2256 86

48. Garmer, D. R. and Krauss, M. (1993) Ab initio quantum chemical study of the cobalt d-d spectroscopy of several substituted zinc enzymes. J. Am. Chem. Soc. 115, 10247-10257 49. Breece, R. M., Costello, A., Bennett, B., Sigdel, T. K., Matthews, M. L., Tierney, D. L. and Crowder, M. W. (2005) A five-coordinate metal center in Co(II)-substituted VanX. J. Biol. Chem. 280, 11074-11081 50. Sabel, C. E., Carbone, R., Dabous, J. R., Lo, S. Y. and Siemann, S. (2011) Preparation and characterization of cobalt-substituted anthrax lethal factor. Biochem. Biophys. Res. Commun. 416, 106-110 51. Lichty, J., Allen, S. M., Grillo, A. I., Archibald, S. J. and Hubin, T. J. (2004) Synthesis and characterization of the cobalt(III) complexes of two pendant-arm cross-bridged cyclams. Inorg. Chim. Acta. 357, 615-618 52. Bertini, I. and Luchinat, C. (1983) Cobalt (II) as a probe of the structure and function of carbonic anhydrase. Acc. Chem. Res. 16, 272-279 53. Banci, L., Bencini, A., Benelli, C., Gatteschi, D. and Zanchini, C. (1982) Spectral-structural correlations in high-spin cobalt (II) complexes. In Structures versus Special Properties. pp. 37-86, Springer 54. Vallejo, L. F. and Rinas, U. (2004) Strategies for the recovery of active proteins through refolding of bacterial inclusion body proteins. Microb. Cell. Fact. 3, 11 55. Inouye, K., Tanaka, H. and Oneda, H. (2000) States of tryptophyl residues and stability of recombinant human matrix metalloproteinase 7 (matrilysin) as examined by fluorescence. J Biochem. 128, 363-369 56. Royer, C. A. (2006) Probing protein folding and conformational transitions with fluorescence. Chem Rev. 106, 1769-1784 57. Lehti, K., Lohi, J., Valtanen, H. and Keski-Oja, J. (1998) Proteolytic processing of membrane-type-1 matrix metalloproteinase is associated with gelatinase A activation at the cell surface. Biochem. J. 334 ( Pt 2), 345-353

58. Tatti, O., Arjama, M., Ranki, A., Weiss, S. J., Keski-Oja, J. and Lehti, K. (2011) Membrane-type-3 matrix metalloproteinase (MT3-MMP) functions as a matrix composition-dependent effector of melanoma cell invasion. PloS One. 6, e28325 59. DeLano, W. L. (2002) Pymol: An open-source molecular graphics tool. CCP4 Newsletter On Protein Crystallography. 40, 82-92

2.6 Tables and figures

a Table 2.1: Data fit for EXAFS traces of Zn2-cdMMP16.

Fit Model Zn-N/O Zn-His Zn-C Rf Ru

S1-1 4 N/O 2.00 (3.7) 18 163

S1-2 4 N/O (3His) 2.00 (3.5) 2.89(7.1) 3.12(3.0) 40 40

4.15(16) 4.40(16)

S1-3 4 N/O (3His) 2.00 (3.6) 2.89(7.5) 3.12(3.3) 2.53 (2.9) 34 35

+ 0.5 C 4.15(16) 4.40(16)

a Distances (Å) and disorder parameters (in parentheses, 2 (10-3 Å2)) shown derive from integer or half-integer coordination number fits to filtered EXAFS data [k = 1.5-11 Å-1; R = 0.5 – 2.0 Å (fit 1), 0.3-4.0 Å (fits 2-3)]. Imidazole multiple scattering paths represent combined paths, as described previously (see

Materials and Methods).

a Table 2.2: Data fit for EXAFS traces of Co2-cdMMP16.

Fit Model Co-N/O Co-His Co-C Rf Ru

S2-1 4.5 N/O 2.12 (5.1) 26 125

S2-2 4.5 N/O (3His) 2.12 (5.1) 2.99(14) 3.28(7.9) 60 55

4.13(31) 4.38(4.4)

S2-3 4.5 N/O (3His) 2.12 (4.9) 2.99(14) 3.29(8.3) 2.31 (8.6) 57 52

+ 0.5 C 4.13(30) 4.39(4.4)

Table 2.3: Steady state kinetic constants and metal content of cdMMP16 samples

K (μM) -1 -1 -1 Protein/Metal (eq) m kcat(s ) kcat/KM(s μM ) Zn Co

Zn2-cdMMP16 10.6 + 0.7 1.14 + 0.02 0.11 1.8 <0.1 (as-isolated)

Zn2-cdMMP16 12.0 + 1.0 1.11 + 0.07 0.093 1.8 <0.1 (reconstituted)

Co2-cdMMP16 10.0 + 0.4 0.36 + 0.07 0.032 <0.1 1.8

ZnCo-cdMMP16 11.2 + 0.9 0.72 + 0.07 0.070 0.8 0.7

Steady-state kinetic experiments were conducted in 50 mM Hepes, pH 7.0, containing

o 100 mM NaCl and 5 mM CaCl2 at 25 C. The substrate was FS-6.

Table 2.4: Kinetic mechanism used to fit stopped-flow fluorescence data and the Dynafit-generated microscopic rate constants.

-1 -1 -1 -1 -1 k1 (μM s ) k-1 (s ) k2 (s ) KS (μM )

0.30 1.0 X 10-6 2.7 3.7

Table 2.5: Secondary structural elements of Zn2-cdMMP16, Co2-cdMMP16, ZnCo-cdMMP16, and metal-free cdMMP16.

The concentration of protein samples were diluted to 10 mM in 20 mM phosphate, pH

Alpha-helix Beta sheet Random loop

As-isolated 26% 16% 58%

Zn2-cdMMP16

Reconstituted 29% 16% 55%

Zn2-cdMMP16

Co2-cdMMP16 29% 15% 56%

ZnCo-cdMMP16 31% 12% 57%

cdMMP16 33% 15% 53%

(metal free)

7.5, containing 100 mM NaCl. The CD spectra were analysis by DichroWeb32 for estimating the secondary structural elements

a Table 2.6: Summary of EXAFS fits of Zn2-cdMMP16 and Co2-substituted analogs

Model Metal-N/O Metal-His Metal-C Rf Ru

4.5 N/O (3His) 2.12 (4.9) 2.99(14) 2.31 (8.6) 57 52 Co2-cdMMP16 + 0.5 C 3.29(8.3) 4.13(30) 4.39(4.4)

Zn2-cdMMP16 4 N/O (3His) 2.00 (3.6) 2.89(7.5) 2.53 (2.9) 34 35 + 0.5 C 3.12(3.3) 4.15(16) 4.40(16) aSee Supporting Information, Figures S1-S2 and Tables S1-S2 for detailed fitting results.

Figure 2.1: Crystal structure of Zn2-cdMMP16 using the coordinates (PDB 1RM8) [34]. -Helices are labeled in red, -sheets are labeled in yellow, and random loops are labeled in green. The structural site (top zoomed in figure) and catalytic site (bottom zoomed in figure) are shown to the right. This figure was generated using Pymol Version 0.99[59].

Figure 2.2: EXAFS of Zn2-cdMMP16: a Fourier transforms and, b k-space data. The experimental data are represented by solid lines, fits are represented by open symbols.

Figure 2.3: EXAFS of Co2-cdMMP16: a Fourier transforms and b k-space data. The experimental data are represented by solid lines, and fits are represented by open symbols.

Figure 2.4: SDS-PAGE analysis of cdMMP16 purification. Lane 1: Perfect protein molecular weight markers, lane 2: boiled cell fraction of BL21(DE3) E. coli cells containing the pET26b/cdMMP16 plasmid before induction, lane 3: boiled cell fraction of BL21(DE3) E. coli cells containing the pET26b/cdMMP16 plasmid after 4 hour induction with IPTG, lane 4: resuspended cell pellet of BL21(DE3) E. coli cells containing the pET26b/cdMMP16 plasmid after 4 hour induction with IPTG, lane 5: resuspended inclusion bodies after E. coli cell lysis, washing, and centrifugation, lane 6: inclusion bodies after suspension in 8 M urea, Lane 7: inclusion bodies after centrifugation and diluted with 50 mM Tris, pH 8.5 (final concentration of urea is 6 M), lane 8: inclusion bodies after dialysis (final concentration of urea is 0 M), lane 9, purified cdMMP16 after Q-Sepharose column chromatography.

Figure 2.5: Substrates used in this study: FS-6 (a) and BML128 (b). The scissile bonds are labeled with red-dashed boxes.

Figure 2.6: Relative catalytic activity of Zn2-cdMMP16 in the presence of NaCl. The substrate used in these assays was FS-6, and the buffer was 20 mM Hepes, pH 7.0, containing 5 mM CaCl2.

100

Figure 2.7: Circular dichroism of as-isolated Zn2-cdMMP16 (solid), Reconstituted

Zn2-cdMMP16 (tight-dot), Co2-cdMMP16 (dash), ZnCo-cdMMP16 (dash dot), and metal-free cdMMP16 (dot)

101

8 50 A B

7 3.2 M 40

6.4 M 6 30 12.9 M

5 20 56.5 M

4 10 103.5 M

3 0 0 0.5 1 1.5 0 20 40 60 80 100 120

Time (s) Substrate (mM)

Figure 2.8: A: Stopped-flow fluorescence traces of 1 M Zn2-cdMMP16 with various concentrations of DNP-Pro-Leu-Ala-Leu-Trp-Ala-Arg-OH. Experimental data are represented with open symbols with the corresponding substrate concentration, and Dynafit-generated simulated lines are shown as overlays of the data points. B: Plot of decay of intrinsic Trp fluorescence (kobs) versus the substrate concentration. The data were fitted to the equation kobs = koff + kon[S], Ks = koff/kon.

[44] Reactions is conducted with buffer in 50 mM Hepes, pH 7.0, containing 100 mM

NaCl, 5 mM CaCl2

102

Figure 2.9: Stability of cdMMP16 samples using SDS-PAGE. a Lane 1, Perfect protein molecular weight markers, lanes 2-8: cdMMP16 (containing 1.8 equivalents of Zn(II) and 5 mM Ca(II)) stored at 4 oC for 0, 1, 2, 3, 4, 5, and 20 hours, and lanes 9-15: Metal-free cdMMP16 (containing <0.1 equivalents of Zn(II) and no added Ca(II)) stored at 4 oC for 0, 1, 2, 3, 4, 5, and 20 hours. b Lane 1, Perfect protein molecular weight markers, lanes 2-8: metal-free cdMMP16 with 4 eq of added Co(II) stored at 4 oC for 0, 1, 2, 3, 4, 5, and 20 hours, and lanes 9-15: metal-free cdMMP16 with 4 eq of added Zn(II) stored at 4 oC for 0, 1, 2, 3, 4, 5, and 20 hours.

103

Figure 2.10: Relative catalytic activity of metal-free cdMMP16, prepared with previously published procedures, upon addition of 4 equivalents of Zn(II) or Co(II) over time. Metal-free cdMMP16 was incubated with 4 equivalents of Zn(II) () or

Co(II) () and assayed in 20 mM Hepes, pH 7.0, containing 5 mM CaCl2 and 100 mM NaCl. The substrate was FS-6. For comparison, the activity of cdMMP16 (), after purification, is shown.

104

Figure 2.11: MALDI-TOF mass spectra of cdMMP16 in the presence of 2 eq of Co(II) and 5 mM Ca(II) over time. The concentration of enzyme was 20 M.

105

Figure 2.12: SDS-PAGE of cdMMP16 samples using modified metal incorporation procedure. Lane 1: Perfect Protein markers, Lanes 2-3: Metal-free cdMMP16 after 1 hr, 3 hrs, and 20 hrs incubation at 4 oC. Lanes 5-7: Metal-free cdMMP16 after incubating with 1 mM Zn(II) for 1 hr, 3 hrs, and 20 hrs at 4 oC. Lane 8: Reconstituted Zn(II)-containing cdMMP16 after centrifugation. Lanes 9-11: Metal-free cdMMP16 after incubating with 1 mM Co(II) for 1 hr, 3 hrs, and 20 hrs at 4 oC. Lane 12: Co(II)-substituted cdMMP16 after centrifugation.

106

Figure 2.13: Fluorescence spectra of as-isolated Zn2-cdMMP16 (solid), reconstituted

Zn2-cdMMP16 (tight dot), ZnCo-cdMMP16 (dash) Co2-cdMMP16 (dash-dot), and metal-free cdMMP16 (dot). The spectra of Co-substituted analogs were amplified (X

1.4 for ZnCo- and X 2.5 for Co2- and metal-free enzymes) and normalized.

107

Figure 2.14: Fourier transformed EXAFS data (solid lines) and corresponding best fits (open symbols) for Zn2-cdMMP16 and Co2-cdMMP16.

108

Figure 2.15: Optical spectra of Co2-cdMMP16 (dash) and ZnCo-cdMMP16 (solid).

Both of samples were diluted in 50 mM Hepes, pH 7.0, containing 5 mM CaCl2

109

1 Figure 2.16: 200 MHz H NMR spectra of Co2-cdMMP16 (top) and ZnCo-cdMMP16

(below). Solvent exchangeable protons are marked with asterisks.

110

Figure 2.17: Fluorescence spectra of 10 M Zn2-cdMMP16 exposed in differential salt concentration of 50 mM Hepes, pH 7.5 containing 5 mM CaCl2. The salt-concertation of each spectra trace was indicated on right-up corner of the figure.

111

1 Figure 2.18: Top: H NMR spectra of reconstituted Zn2-cdMMP16 and as-isolated

1 o Zn2-cdMMP16. One-dimensional H NMR spectra were recorded for 10 min at 25 C on 500 L samples of ~0.1 mM MMP16 on a Bruker 500 MHz Avance NMR system with a 5 mm BBO probe using WATERGATE for water suppression. The reconstituted Zn2-cdMMP16 sample was in 50 mM Hepes, pH 7.0, containing 5 mM

CaCl2. The as-isolated Zn2-cdMMP16 sample was in 50 mM Tris, pH 7.5, containing

100 mM NaCl and 5 mM CaCl2. Bottom: SDS-PAGE gel of Zn2-cdMMP16 samples used for NMR studies: Lane 1: Molecular weight markers; Lanes 2-5: FPLC fractions of as-isolated Zn2-cdMMP16; Lane 6: concentrated, as-isolated Zn2-cdMMP16 sample used for NMR spectrum; Lane 7: concentrated, reconstituted Zn2-cdMMP16 sample used for NMR spectrum.

112

Chapter 3

Biochemical characterization and zinc binding group (ZBGs) inhibition studies on the catalytic domain of MMP7 (cdMMP7)

Fan Meng, Hao Yang, Colin Jack, Hua-Qun Zhang, Abraham Moller, Devin Spivey, Richard C. Page, David L. Tierney and Michael W. Crowder

Department of Chemistry and Biochemistry, Miami University, Oxford, Ohio 45056

†This work was supported by the National Science Foundation (CHE-1509285 to MWC and DLT, and MCB-1552113 to RCP).

AUTHOR CONTRIBUTIONS

FM, DLT, RP, and MWC wrote the paper. CJ performed DNA cloning work on cdMMP7. FM prepared and characterized as-isolated cdMMP7 and ZnCo-cdMMP7 with assistance from DS. FM and HY performed the stopped flow fluorescence emission experiments, and ZBG-IC50 assays on cdMMP7/16. FM performed the stopped flow substrate-hydrolysis experiments. AM calculated theoretical equations of kcat and Km by using the King-Altman method. FM and HQZ performed the ITC experiments. RP performed computational docking studies.

113

ABSTRACT

Matrix metalloproteinase 7 (MMP7 / matrilysin-1) has been implicated in many pathological conditions, such as in cancer and inflammatory diseases; therefore, MMP7 has been considered as drug target. Success in developing a clinical inhibitor, which exhibits suitable specificity and selectivity, will likely require structural and/or kinetic evaluation of enzyme/inhibitor interactions. To enable these future studies we herein describe the over-expression, purification, and characterization of the catalytic domain of MMP7 (cdMMP7). cdMMP7 was over-expressed in an E. coli over-expression system, and the resulting enzyme was processed into inclusion bodies, which were subsequently solubilized, enabling the enzyme to be re-folded into a catalytically-active form. cdMMP7 was shown to bind 1.8 eq of Zn(II), exhibit

-1 steady-state kinetic constants of 0.4 s for kcat and 23 M for Km, and yield CD and fluorescence spectra that are consistent with a properly-folded enzyme. Pre-steady state kinetic studies yielded kinetic mechanisms of cdMMP7, and these mechanisms are similar to those of other MMPs. Inhibition studies on cdMMP7 with four zinc binding group (ZBG) inhibitors showed that maltol, thiomaltol, and allothiomaltol are better inhibitors with lower IC50 values and lower Kd values against cdMMP7 and cdMMP16 than the commonly-used ZBG acetohydroxamic acid. Docking studies suggest that improved inhibitory character may be due to interactions with the S1' substrate binding pocket. Finally, a ZnCo-heterobimetallic analog of cdMMP7 with Co(II) bound in the catalytic site was prepared and characterized. This study describes a well-characterized analog of MMP7 that is available for future inhibitor redesign efforts.

KEYWORDS Matrix Metalloproteinases (MMPs), Pre-steady-state Kinetic Study, Cobalt(II)-substitution, Zinc-binding-group MMP Inhibitors( ZBG-MMPis)

114

3.1 Introduction

Matrix metalloproteinases (MMPs) are Zn(II)-dependent metalloproteases, and there are 23 distinct MMPs in humans[1]. The MMPs can be broadly categorized into two classes: secreted MMPs and membrane-bound MMPs (MT-MMPs). Matrix metalloproteinase 7 (MMP7) is a member of the matrilysin subclass of secreted MMPs[1, 2]. Full-length MMP7 consists of two conserved functional domains: a N-terminal pro-domain and a C-terminal catalytic domain [3, 4]. The function of the pro-domain is to maintain MMP7 in a latent form (proMMP7), and this latency is achieved by coordination of the active site Zn(II) with a cysteine residue, which is located in the highly-conserved PRCGPD “bait region” motif in the pro-domain[2, 5]. When proMMP7 is secreted into the extracellular space, the enzyme is cleaved by one of several serine or metzincin proteases, leading to the release of the pro-domain and activation of MMP7. The cleavage of the pro-domain from the catalytic domain results in the dissociation of the cysteine-catalytic Zn(II) bond and activation of the MMP; this mechanism is known as the “cysteine switch” and is a common regulatory mechanism in secreted MMPs [5-7]. The catalytic domain has a ball-like structure consisting of three -helices, five-sheets, and multiple loops [8](Figure 3.1). The catalytic domain contains 2 Zn(II) binding sites: a structural site with Zn(II) coordinated by 3 conserved histidines and an aspartic acid and a catalytic site with Zn(II) coordinated by 3 conserved histidines found in a conserved HExxHxxGxxH motif, and 1-2 solvent molecules (Figure 3.1)[8]. Adjacent to the active site are a set of substrate binding pockets, termed S1-S3 and S1’-S3’. The substrate binding pockets of MMPs vary in shape, charge, and hydrophobicity and are considered key factors in selectivity and specificity for substrates and inhibitors[8-10]. A conserved methionine, Met237, is present 8 amino acids from the catalytic Zn site as part of “Met-turn”, which is conserved in all MMPs belonging to the metzincin family and is essential for stabilizing the Zn(II)-binding motif for metal ion coordination [8, 11]. MMP7 also contains a ten amino acid loop termed the

115

specificity loop (Tyr241- Leu251) in the C-terminus, which is thought to contribute to selectivity in substrate binding [12] (Figure 3.1). The physiological substrates for MMP7 are components of the extracellular matrix (ECM), such as proteoglycan, fibronectin, casein, and elastin[13, 14]. Studies have shown that MMP7 affects cell-ECM adhesion and cell-cell interactions, which allows for matrix remodeling, cellular motility, cellular growth, cellular proliferation, and angiogenesis[13, 15]. In addition to its role in regulation of ECM biology, MMP7 has been implicated in mediating multiple signaling pathways by processing key molecules. For example, MMP7 appears to hydrolyze pro-apoptotic ligands (FASL, FAS), chemokines (CXCL-9, CXCl-11), growth factors (IGFBP -1,-2,-3), and metalloproteinases (MMP2, MMP9, ADAM28)[13, 16-19]. With this wide range of substrates, MMP7 has been implicated in the regulation of multiple biological processes, such as early immune response and inflammation[13, 14, 20]. Moreover, MMP7 plays a role in neurotransmitter exocytosis by degrading the SNARE protein (SNAP 25)[21]. No other MMP appears to be involved in neurotransmitter processing[18]. Up-regulation of MMP7 directly contributes to metastasis, abnormal angiogenesis, immune surveillance evasion, and cancer related inflammation in tumorigenesis of breast, colon, and pancreatic cancers [13, 22-25]. Expression of MMP7 has also been reported in diseases like arthritis and chronic ulcers[26-28]. Through these multitude of targets and role in myriad biological processes and pathways MMP7 is considered to be a potential therapeutic target[29]. Previous attempts to use broad-spectrum MMP inhibitors (MMPis) in clinical trials have failed, due primarily to reported severe side effects like musculoskeletal syndrome (MSS). In studies with inhibitors targeting MMP7, “off-target” inhibition of other MMPs (and perhaps other proteases) was used to explain the failure of the inhibitors clinically[10, 30]. The side effects of “off-target” inhibition included increased infection risk and delay in wound healing[31]. Therefore, improved selectivity and specificity are required if MMPis are to be used as effective and safe therapeutics[30, 32].

116

Due to the highly similar catalytic domains of MMPs (cdMMPs), particularly at the active sites, the development of a high specificity MMPi has been challenging[9]. One common approach to improve MMPi selectivity and efficacy is to optimize the interactions of the MMPi with adjacent substrate binding pockets, most notably the S1’ pocket[10, 33]. The architecture of S1’ pockets vary significantly across MMPs; therefore, the S1’ pocket is considered to be the most important site influencing substrate and inhibitor selectivity of the MMPs[34]. A previous comparison of the S1’ pockets demonstrated that the depths of the pockets can be described as shallow, mid-size, and deep: the S1’ pockets in MMP2, MMP8, and MMP9 are mid-size, the S1’ pockets in MMP3, MMP11, MMP13, MMP14, and MMP16 are deep with a wide entrance to the pockets[10, 35, 36], and the S1’ pocket in MMP7 is shallow with the opening surrounded by three aromatic amino acid resides (Tyr214, His218, and Tyr240 ). The side-chains of these amino acids extend into the pocket[8]. Cohen et al. reported structure-activity relationship (SAR) studies with hydroxamate-containing zinc binding groups (ZBGs) containing substituents targeting the S1’ pocket. These (SAR) studies showed that inhibitor selectivity can be achieved by targeting the S1’ pocket in MMP3[16]. By analogy, we hypothesized that selective MMP7 inhibitors can be achieved by targeting the S1’ pocket in MMP7. To conduct these studies, a well-characterized analog of MMP7 is needed. Herein, we report the over-expression, purification, and characterization of the catalytic domain of MMP7 (cdMMP7). We also report the preparation and characterization of Co(II)-substituted analogs of cdMMP7 for future spectroscopic interrogation of MMP7-inhibitor complexes.

117

3.2 Material and methods

Materials

Taq PCR reaction kit, restriction enzymes (NdeI, HindIII), and T4 ligation kit were purchased from New England Biolabs (Ipswich, MA). Luria-Bertani (LB) medium was obtained from Invitrogen (Carlsbad, CA). Isopropyl-β-D-thiogalactoside (IPTG) was procured from Anatrace (Maumee, OH). Tris(hydroxymethyl)aminomethane (Tris),

4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (Hepes), urea, NaCl, CaCl2,

ZnCl2, CoCl2, ethylenediaminetetraacetic acid (EDTA), Brij 35, Triton X100, sucrose, phenanthroline (OP), D2O, tris(2-carboxyethyl)phosphine (TCEP), arginine, and glutamate were purchased from Fisher Scientific (Hampton, New Hampshire). Barnstead NANOPure water was used to make all buffers. Chelex100 resin was procured from Biorad (Hercules, CA). Centricon ultrafiltration units were purchased from EMD Millipore (Billerica, MA), and Amicon centrifuge units (YM-10 membranes) were obtained from Sigma Aldrich (St. Louis, MO). Fluorogenic peptide

FS-6 (MCA-Lys-Pro-Leu-Gly-Leu-DNP-Dpa-Ala-Arg-NH2 Figure 2A) was purchased from Sigma Aldrich (St. Louis, MO) for pre-steady state kinetic, steady state kinetic, and inhibition studies. Fluorogenic peptide BML-P131

(Dnp-Pro-Leu-Gly-Leu-Trp-Ala-Arg-NH2, Figure 2B) and reference peptide for FS-6 (Mca-Pro-Leu-OH) were purchased from Enzo Life Science (Farmingdale, NY). Acetohydroxamic acid (AHA) and maltol were procured from Sigma Aldrich. Thiomaltol (TM) and allothiomaltol (ATM) were kindly provided by Dr. Seth Cohen (University of California at San Diego, CA)

Cloning, over-expression, refolding, and purification of cdMMP7. A pGEM7-based plasmid, containing the cDNA for full-length MMP7, was generously provided by Dr. Carole L. Wilson (University of Washington). The gene for the catalytic domain of

118

MMP7 (cdMMP7, corresponding to amino acids 100 to 271) was subcloned by PCR using the primers, 5'-aaa aaa cat atg tac tca cta ttt cca-3' and 5'-aaa aaa aag ctt cta ttt ctt tct tga-3'. These primers introduced NdeI and HindIII restriction sites at the 5’ and 3’ ends of the resulting gene, respectively. The PCR product was then digested and ligated into pET26b to yield pET26b/cdMMP7. DNA sequencing was used to confirm the cdMMP7 gene, and the over-expression plasmid was transformed into E. coli BL21(DE3) cells. A single colony was used to prepare a 50 mL overnight culture in Luria-Bertani (LB) medium containing 25 g/mL of kanamycin (kan). The overnight culture was used to inoculate 4X1 L of LB-kan (1:100 dilution), and the cultures were shaken at 37 oC with a 180 rpm shaking rate until the cultures reached an optical density at 600 nm of 0.7. Protein production was induced by making adding isopropyl-β-D-thiogalactoside (IPTG) to 1 mM final concentration followed by shaking for 4 hours at 37 oC. Cells were collected by centrifugation at 14,000 rpm for 20 minutes at 4 oC and resuspended in lysis buffer (50 mM Tris, pH 7.5, containing 200 mM NaCl, 10% sucrose, and 1 mM EDTA) containing 0.5 mM lysozyme. Resuspended cells were ruptured by passing the suspension through a French press with a pressure of 1,500 psi three to four times, and the lysed mixture was centrifuged at 15,000 rpm for 25 minutes at 4 oC. The resulting pellet, which contained cell debris and inclusion bodies, was washed three times with 30 mL of lysis buffer containing 0.1% Triton X100. Inclusion bodies were dissolved in 35 mL of 50 mM Tris, pH 7.5, containing 8 M urea at room temperature for at least one hour with gentle stirring. Solubilized inclusion bodies were centrifuged at 4 oC for 10 minutes at 14,000 rpm and then diluted to a final concentration of 0.3 mg/mL with 50 mM Tris, pH 7.5, containing 6 M urea. The concentration of dissolved inclusion bodies was determined by using a Pierce protein assay[37]. The diluted inclusion bodies (approximately 100 ml) were dialyzed twice against 2L 50 mM Tris, pH 7.5, containing 5 mM CaCl2, 100

o M ZnCl2, and 0.1% Brij 35. Each dialysis step lasted 3-4 hours at 4 C, and a final dialysis step was allowed to proceed for approximately 12 hours, using 50 mM Tris, pH 7.5, containing 20 M ZnCl2, 5 mM CaCl2, and 0.1% Brij 35. If refolded protein 119

was prepared for isothermal titration calorimetry, inductively-couple plasma analysis or Co(II)-substitution an additional dialysis step was required against 50 mM Tris, pH

7.5, containing 5 mM CaCl2, 1 M ZnCl2, and 0.1% Brij 35 for 4 hours,. The purity of refolded protein was determined by SDS-PAGE and MALDI-TOF mass spectrometry. The concentration of purified cdMMP7 was determined by measuring the absorbance at 280 nm and using a calculated extinction coefficient of 30,000 M-1cm-1 [38]. Each enzyme preparation was evaluated for metal content and catalytic activity. Only enzyme stocks that exhibited sufficient metal content (+ ~ 0.1 eqv cobalt) and catalytic activity (+ 10%) were used for subsequent studies.

Metal analyses of cdMMP7 analogs. The metal content of cdMMP7 samples was determined using a Perkin Elmer Optima 7300DV Inductively-Coupled Plasma with Atomic Emission Spectrometer. The emission wavelengths were set to 213.8 nm and 238.9 nm for measuring zinc and cobalt, respectively, as previously described[39, 40].

Steady state kinetic studies on cdMMP7. Steady-state kinetic studies were conducted on cdMMP7 using a Synergy HT plate reader. The fluorescent substrate

FS6 (Sigma Aldrich) was used in the assay, and excitation and emission wavelenghts were set to 330 nm and 420 nm, respectively. The concentration of substrate was varied between 10 and 60 M, and the concentration of Zn2-cdMMP7 (analog of cdMMP7 with Zn(II) in the catalytic and structural sites) was fixed to 10 nM (or to 20 nM with ZnCo-cdMMP7). The final volume for each reaction was 100L, and each reaction was allowed to proceed for 5 minutes. GraphPad Prism 5 was used to fit the resulting kinetic data to the Michaelis equation to obtain Km and kcat[41, 42].

Stopped-flow kinetic studies on cdMMP7 and cdMMP16. An Applied Photophysics spectrophotometer was used to conduct stopped-flow fluorescence studies. The excitation wavelength was set to 280 nm. Two different fluorescence cut-off filters were used depending on whether intrinsic tryptophan fluorescence or substrate fluorescence was monitored over time. In experiments monitoring intrinsic tryptophan

120

fluorescence, a 320 nm cut-off filter was used to exclude all emission wavelengths under 320 nm. In these experiments at 25 oC, the concentration of cdMMP7 was fixed at 10 M, and BML-P131 was used as the substrate (concentrations varied between 12 and 100 M). The reaction buffer was 50 mM Hepes, pH 7.0, containing 100 mM

NaCl and 5 mM CaCl2. The reaction was monitored over 20 seconds using a 125 s time interval. Dynafit 3.0 was used to simulate progress curves to obtain microscopic rate constants (k1, k-1, and k2 ). In reactions in which the fluorescence of substrate was monitored, a 400 nm cutoff filter was used to filter out emission wavelengths below 400 mm, including fluorescence emission of protein tryptophans. The concentrations of cdMMP7 and cdMMP16 samples were 0.5 M, and the buffer was 50 mM Hepes, pH 7.5, containing 100 mM NaCl and 5 mM CaCl2. The final concentrations of substrate FS-6 were varied between 70 M and 4.9 M for studies with cdMMP7 and 32 M to 1.98 M for studies with cdMMP16. After mixing, emission intensities were obtained every 125 s for 1 second. The resulting progress curves were simulated to a standard uni-bi hydrolytic mechanism[43] by using Dynafit 3.0 to obtain microscopic rate constants[44]. The King-Altman method was used to generate theoretical expressions for kcat and Km[45], and the theoretically-calculated steady-state kinetic constants were compared to those kinetic constant determined experimentally. The theoretical expressions were kcat =[k2(k3 + k-2)] / (k3 + k-2 + k2) and Km = (k2k3 + k-2k-1 + k3k-1) / (k3k1 + k1k-2 + k1k2)]. Preparation of the ZnCo-cdMMP7 analog. An analog of cdMMP7 that contains ~1 equivalent of Zn(II) was prepared using a previously-published procedure[46]. Briefly, purified cdMMP7 was diluted to ~100 M in a total volume of ~5 mL. The sample was dialyzed against three changes of1L of 20 mM Hepes, pH 7.0, containing 2 mM

o 1,10-phenanthroline (OP) and 5 mM CaCl2 at 4 C. The dialysis buffer was changed every 3 hours. The sample was then dialyzed against 2X1L of Chelex100-treated, 20 mM Hepes, pH 7.0, containing 5 mM CaCl2 for (3 hours for each step). The resulting sample was made 1 mM in CoCl2, allowed to incubate for at least 40 minutes on ice, 121

and dialyzed for 10 hours versus 1L of 20 mM Hepes, pH 7.0, containing 1 mM

CoCl2 and 5 mM CaCl2 and twice against 1L of Chelex100-treated, 20 mM Hepes, pH

7.0, containing 5 mM CaCl2 (3 hours each step). Finally, the sample was dialyzed versus 1L of 50 mM Hepes, pH 7.5, containing 100 mM NaCl, 5 mM CaCl2, and 0.4 mM TCEP for 4 hours before the sample was concentrated using a Centricon unit to a final concentration of approximately 100 M.

Circular dichroism and fluorescence emission spectroscopic studies of cdMMP7. cdMMP7 was diluted to 10 M with 20 mM sodium phosphate, pH 7.0, containing 100 mM NaCl. CD spectra were obtained on a JASCO J-810 CD spectropolarimeter using a 1 cm cylindrical cell. The spectra were recorded in the range of 190 – 260 nm, and 10 scans were signal-averaged at 25 oC. DiChroWeb was used to estimate the secondary structure composition[47]. Fluorescence emission spectra were collected on a Perkin Elmer Luminescence spectrometer (Model LS-55), using an excitation wavelength of 280 nm. Emission spectra from 290 to 500 nm were signal-averaged (5 scans) at room temperature.

Optical spectroscopy of ZnCo-cdMMP7. UV-Vis spectra on cdMMP7 samples were collected on a Lambda 850 UV-vis spectrophotometer from 200 – 800 nm, operating at approximately 25 oC. ZnCo-cdMMP7 samples were concentrated to 100 M using a Centricon, and the buffer was 50 mM Hepes, pH 7.0, containing 100 mM NaCl and

5 mM CaCl2. Difference spectra were obtained by subtracting the spectrum of metal-free cdMMP7 from those of the ZnCo-cdMMP7 analogs.

Determination of IC50 values for ZBGs against cdMMP7 and cdMMP16. Four zinc binding group (ZBG) inhibitors[48] were used in inhibitor studies with cdMMP16 and cdMMP7: acetohydroxamic acid (AHA), maltol, thiomaltol (TM), and allothiomaltol

(ATM). Steady-state kinetic assays for determining IC50 were conducted on a Synergy HT plate reader (BioTEK). The concentration of the cdMMPs was 10 nM, and the

122

concentration of substrate FS-6 was 10 M. The reaction volume was 100 L, and the buffer was 20 mM Hepes, pH 7.0, containing 100 mM NaCl and 5 mM CaCl2. The cdMMP samples were incubated with various concentrations of the ZBGs at room temperature for no more than 30 minutes, and each reaction was monitored for 10 minutes. The concentrations of AHA, maltol, TM, and ATM were varied between 0 -

1.2 M, 0 - 620 mM, 0 – 12.3 mM, and 0 - 12.3 mM, respectively. The values for IC50 were calculated by using GraphPad Prism 5. Isothermal titration calorimetry (ITC) on cdMMP 16 and cdMMP7. The binding of ZBGs to the cdMMPs was monitored using a NanoITC (TA Instruments) isothermal titration calorimeter. Samples of cdMMP16 and cdMMP7 were diluted to concentrations of 50 M using 50 mM Hepes, pH 7.5, containing 100 mM NaCl and

5 mM CaCl2 for cdMMP16 and 50 mM Tris, pH 7.5, containing 5 mM ZnCl2, and 0.1% (w/v) Brij35 for cdMMP7. Inhibitors were diluted with the same buffers used to dilute the enzyme to minimize any heat of dilution effects[49]. Concentrations of ATM and TM stocks were 500 M and 350 M, respectively. The enzymes were titrated with 1.9 L ligand per injection for a total of 24 injections. The NanoITC program (TA Instruments) was used to fit the experimental data to yield values for enthalpy (△H), entropy (△S), dissociation constant (Kd), and binding stoichiometry (n). The free Gibbs energy (△G) for the binding of each ZBG was calculated using △G = △H – T△S.

ZBG docking studies. Models of cdMMP7 and cdMMP16 were prepared from PDB depositions 1mmp and 1rm8 [8, 50], respectively, using the “prepack” protocol within the Rosetta modeling software[51]. Rosetta ligand parameter files were prepared from the final geometries and charges for conformers of thiomaltol or allothiomaltol generated using OMEGA[52]. Ligands placed in a starting pose near the catalytic zinc site were paired with prepacked cdMMP7 and cdMMP16 coordinates for RosettaLigand docking calculations[53, 54]. Docking runs allowed for backbone and side-chain flexibility within cdMMP7 or cdMMP16 with extra Chi1 and

123

aromatic-Chi2 rotamers and permitted ligand translations away from the starting pose of up to ± 5 Å along the x, y and z axes. An ensemble of 10,000 decoys was generated for each docking calculation, and the top 50 were clustered based upon overall Rosetta energy score (total_score). A representative model for MMP/ZBG pair was selected from the low energy ensemble.

3.3 Results

Over-expression, refolding, and purification of cdMMP7. In order to obtain large quantities of catalytically-active, recombinant cdMMP7, we used a pET26b-based vector containing the gene for cdMMP7, and we over-expressed the enzyme in BL21(DE3) E. coli cells, in a method similar to that of Cha et al[46]. Recombinant cdMMP7 formed inclusion bodies after over-expression in E. coli. Previously, Cha et al. solubilized the inclusion bodies in 6 M Gdn-HCl and removed the Gdn-HCl using a single dialysis step with buffer containing 0.05% Brij35, 10 mM CaCl2, and 20 M

ZnCl2 [46]. The resulting protein was purified using a hemopexin-affinity column, and fractions containing cdMMP7 were identified by activity assays[46]. Our initial efforts at purifying cdMMP7 used a modified Cha method[46], in which we replaced Gdn-HCl with urea. In our hands, we obtained low levels of cdMMP7 when using urea, and SDS-PAGE showed that the enzyme was hydrolyzed (Figure 3.3A). Therefore, we solubilized the inclusion bodies with 8 M urea, and the enzyme was folded by slow removal of urea using dialysis. Three significant changes in the dialysis steps were used: (1) the concentration of Brij35 was increased from 0.05% (w/v) to 0.10% (w/v)[55] to increase the solubility of refolding intermediates during dialysis, (2) the single dialysis step used in the Cha method[46] was replaced by three shorter steps (each dialysis step was 3 to 4 hours besides the final step), (3) the concentration of ZnCl2 was 100 M in the first two dialysis steps, 20 M in the third step, and the resulting cdMMP7 was shown to be intact and pure by SDS-PAGE (Figure 3.3B). The yield of cdMMP7 was ~ 2.4 mg per liter of LB growth medium.

124

ICP-AES showed that the recombinant enzyme binds 1.8 + 0.1 equivalents of Zn(II)

(Table 3.1). Steady-state kinetic studies using FS-6 as the substrate yielded a kcat of

-1 0.40 + 0.02 s and a Km of 23 + 2 M (Table 3.1). CD spectra of the recombinant enzyme (denoted as Zn2-cdMMP7) showed 27% -helix and 33% -sheet (Figure 3.4 and Table 3.2). The fluorescence spectrum of the refolded cdMMP7 showed a single emission peak centered at ~340 nm (Figure 3.5) The initial concentration (< 0.3 mg/mL) of inclusion bodies was critical in obtaining properly-folded, catalytically-active cdMMP7. If high concentrations (>0.5 mg/ml) of solubilized cdMMP7 inclusion bodies were used, the resulting enzyme

(called inactive Zn2-cdMMP7) exhibited a slightly shifted CD spectrum (Figure 3.4), lower metal to protein ratios (~1.5 Zn(II) equivalents), a significantly-altered fluorescence spectrum that had two emission peaks (Figure 3.6), and no catalytic activity.

Stopped-flow fluorescence studies. Previously, we reported that the fluorescence emission properties of presumably highly-conserved Trp154 in cdMMP1 and cdMMP16 changed during the hydrolysis of substrate BML-P131 (Figure 3.2B). Trp154 is approximately 10 Å away from the catalytic Zn(II) in MMP1, MMP7, and MMP16[39, 40]. Therefore, we conducted stopped-flow fluorescence studies with cdMMP7 using BML-P131 as substrate (Figure 3.7). BML-P131 is hydrolyzed slowly by cdMMP7, so we followed the reaction for 20 seconds. During the first 5 seconds of the reaction, there was a decrease in fluorescence emission, followed by an increase in fluorescence over the subsequent 15 seconds. The progress curves were fitted to a simple kinetic mechanism[44] (Table 3.3), and we were able to obtain microscopic rate constants (k1, k-1, and k2) for the hydrolysis of BML-P131by cdMMP7. These microscopic rate constants are much lower than those for MMP1 and MMP16[39, 40], demonstrating that BML-P131 is a poor substrate for cdMMP7 (Table 3.3). The rate of fluorescence decay (first phase of progress curve in Figure 3.7) was plotted against

125

the substrate concentration as previously reported[39], and the Ks for BML-P131 for cdMMP7 is 30.5 M (Table 3.3). In an effort to further probe the reaction mechanism of the MMPs, we conducted stopped-flow fluorescence studies on cdMMP7 (Figure 3.8A) and cdMMP16 (Figure 3.8B) with fluorescent substrate FS-6. The progress curves for both cdMMPs were similar, exhibiting a burst phase during the initial 0.2 sec of the reaction for cdMMP7 and during the initial 0.1 sec of the reaction for cdMMP16 (Figure 3.8). The burst, which had an amplitude that was substrate-concentration dependent, was followed by a relatively slower phase. The stopped-flow progress curves were fitted using Dynafit to a uni-bi hydrolytic mechanism[43] (see lines in Figure 3.8). The three-step kinetic mechanism proposed for cdMMP7 and cdMMP16 is shown in Table 3.4. The best fit to the data for both MMPs assumed rate-limiting chemistry steps (k2) and slow reverse chemistry steps (k-2) (Table 3.4). To test the validity of our proposed kinetic mechanism, we used the King-Altman method to calculate the theoretical expressions for kcat and Km. The theoretical kcat values for

-1 -1 cdMMP7 and cdMMP16 were 0.58 s and 1.6 s , and the theoretical Km values were 27 μM and 13 μM, respectively. These theoretical steady-state kinetic constants are

-1 similar to those determined experimentally: cdMMP7 has a kcat of 0.40 s and a Km of

-1 23 M (Table 3.1) and cdMMP16 has a kcat of 1.1 s and a Km of 10.6 M[39]. Preparation and characterization of heterobimetallic CoZn-cdMMP7. Using a previously-published procedure[46], we prepared an analog of cdMMP7, which contains ~ 1 equivalent of Zn(II). The addition of Co(II) to this sample resulted in a CoZn-cdMMP7 analog that binds 0.7 eq of Zn(II) and 1.2 eq of Co and exhibits a

-1 kcat of 0.17 + 0.04 s and a Km of 16 + 3 M when using FS-6 as substrate (Table 3.1). In our hands, the yield of CoZn-cdMMP7 was about 25%. The CD spectrum of the heterobimetallic analog showed a similar secondary structural composition as the ZnZn analog (Figure 3.4 and Table 3.2), and the fluorescence spectrum of

CoZn-cdMMP7 was similar to the as-isolated Zn2-cdMMP7 analog (Figure 3.5) suggesting similar tertiary structure for both analogs. A UV-vis difference spectrum 126

of the CoZn-cdMMP7 analog showed a small, broad peak at 475 nm, which accounts for ~10% of the cobalt in the sample, which we attribute to Co(III), and a broad peak at 530 nm, which has an extinction coefficient of ~ 50 M-1cm-1 (Figure 3.9)[39, 40, 56, 57]. We assign the latter peak to ligand field transitions of high-spin Co(II), and the intensity of this peak suggests 6-coordinate Co(II), suggesting that the Co(II) binds in the catalytic zinc site. A similar 6-coordinate Co(II) was previously reported for Co(II) bound in the catalytic metal binding site of cdMMP16[39].

Inhibition studies on cdMMP7 and cdMMP16 with zinc binding group (ZBGs) inhibitors. Four ZBGs, acetohydroxamic acid (AHA), maltol, thiomaltol (TM), and allothiomaltol (ATM) (Figure 3.10), were evaluated as inhibitors of cdMMP7 and cdMMP16. AHA, which is the most common ZBG used in MMPis[10, 30, 32], is a very weak inhibitor of cdMMP7 and cdMMP16, exhibiting IC50 values around 50 mM (Table 3.5). Maltol was a much better inhibitor, exhibiting IC50 values near 10 mM. The replacement of the keto oxygen in maltol with sulfur in thiomaltol resulted in a significant reduction in IC50 values, with cdMMP7 exhibiting an IC50 of 134 M and cdMMP16 exhibiting an IC50 of 30 M (Table 3.5). Even though TM is a relatively small inhibitor in size, there is >4-fold difference in the IC50 values for cdMMP7 and cdMMP16 (Table 3.5). Allothiomaltol, which is different from thiomaltol in the fact that the methyl group is trans to the hydroxyl group rather than cis, is a slightly better inhibitor of cdMMP7 and worse inhibitor of cdMMP16. Maltol, thiomaltol, and allothiomaltol are better inhibitors of cdMMP7 and cdMMP16 than the commonly used acetohydroxamic acid. Comparison of our IC50 values with those of cdMMP1 (Y. Hao, unpublished) demonstrates selectivity among the four tested ZBGs (Figure 3.11). ITC studies on cdMMP7 and cdMMP16 with thiomaltol and allothiomaltol. To further probe binding of the thiomaltol and allothiomaltol to cdMMP7 and cdMMP16, ITC studies were conducted (Figure 3.12; Table 3.6). TM binds to cdMMP7 and

127

cdMMP16 with Kd values of near 3 µM while affinities for ATM against both cdMMP7 and cdMMP16 are 8.5 and 5.8 µM, respectively. Although the affinities are within the same order of magnitude, thermodynamic parameters determined by ITC highlight a potential difference in binding. The enthalpic contributions to binding for TM against cdMMP7 and cdMMP16 are similar to those for ATM against cdMMP7

-1 and cdMMP16. The range of Kd values and negative H values near 24 kJ mol suggest that although binding of ATM to cdMMP16 is weaker, and binding to cdMMP7 is weaker still, there are fundamental similarities in binding modes for TM/cdMMP7, TM/cdMMP16, ATM/cdMMP7, and ATM/cdMMP16. However, while the TM/cdMMP7, TM/cdMMP16 and ATM/cdMMP16 interactions feature positive S values around 25 J mol-1K-1, the S value for ATM against cdMMP7 is smaller and suggests a difference in binding mode compared to TM/cdMMP7, TM/cdMMP16, and ATM/cdMMP16.

3.4 Discussion

MMP7, which is the smallest full-length MMP in terms of size[2, 3], is involved in a number of physiological and pathological processes, due to its involvement in ECM processing and regulation of chemokines and growth factors in various signaling cascades[13, 18]. Since MMP7 lacks a hinge region and a hemopexin-like domain in its C-terminus, the catalytic domain (cdMMP7) is the physiologically-relevant analog of the enzyme, once the pro-domain is removed[5].

Therefore, biochemical studies on Zn2- and metal-substituted analogs of cdMMP7 could shed light on MMP7’s role in vivo and provide information needed to design and prepare MMPis in the future. The NMR solution structures of cdMMP7, MMP7 containing the propeptide, and MMP7/inhibitor complexes have been reported, and all enzymes were prepared using mammalian over-expression cell lines[8]. Two E. coli based over-expression systems were subsequently reported: one system used pET22b[58] and yielded the

128

propeptide, and the other system used pET11a and produced the catalytic domain (cdMMP7)[46]. Both systems resulted in MMP7 being processed into inclusion bodies and required optimized refolding procedures. Compared to cdMMP1[40] and cdMMP16[39], we discovered that the refolding intermediates of cdMMP7 were relatively unstable during dialysis, as the majority of the protein precipitated when using our previous methods. Therefore, we used the method published by Cha to refold cdMMP7, and this method included Brij35 during dialysis steps[46]. Although we successfully obtained active cdMMP7 using this method, the yield of cdMMP7 was low, and cdMMP7 was hydrolyzed (Figure 3.3). To address these issues, we modified the Cha method[46] by increasing the Brij35 concentration from 0.05% to 1% and by increasing the number of dialysis steps from 1 step to 3 steps with higher concentrations of Zn(II) (from 20 to 100 M). Moreover, we found that the starting concentration of inclusion bodies is critical, with any initial concentrations greater than 0.3 mg/mL resulted in cdMMP7 that exhibited fluorescence spectra that suggested a mixture of folded and unfolded protein and no catalytic activity (Figure 3.6). We speculate that the higher concentrations of inclusion bodies may have resulted in aggregated forms of the enzyme. The kinetic mechanism of cdMMP7 was probed using stopped-flow fluorescence studies. In the first experiment, we monitored intrinsic tryptophan fluorescence emission during hydrolysis of substrate BML-P131. Like earlier studies with cdMMP1 and cdMMP16[39], the fluorescence emission decreased during the initial part of the reaction, and there was an increase in fluorescence emission during the subsequent time period. In the early phase of the reaction, we attribute the fluorescence change solely to Trp154, which is conserved in most MMPs such as MMP1, MMP7, and MMP16 and is roughly 10 Å from the catalytic Zn(II) binding site[39, 40]. The slower regain in fluorescence is due to product (formation) and to a regain in Trp fluorescence as substrate/product is no longer bound to the enzyme. Fitting of the progress curves to a simple binding kinetic mechanism resulted in microscopic rate constants for cdMMP7 that are much lower than those for MMP1 129

and MMP16. Steady-state kinetic studies of cdMMP7 with substrate BML-P131 revealed that BML-P131 is a poor substrate for MMP7, most likely due, in part, to the shallow S1’ pocket found in MMP7 but not in MMP1 and MMP16[50, 59]. This hypothesis is supported by the fact that the Ks for BML-P131 binding to cdMMP7 is 10 times larger than that for cdMMP16 (Table 3.3)[39]. In the second experiment, we monitored the increase in fluorescence emission from product as the substrate FS-6 was hydrolyzed by cdMMP7 and cdMMP16 (Figure 3.8). The resulting progress curves were biphasic, with a fast initial “burst” phase followed by a slower phase. The amplitude of the “burst” was substrate concentration-dependent, and the data could be fitted to a uni-bi (one substrate, two product) hydrolytic mechanism. The “burst” can be explained by a relatively fast binding of substrate to the enzymes that must result in the rapid increase in fluorescence, presumably due to the loss of quenching between the donor and acceptor in substrate. The much slower hydrolysis of the amide bond in substrate results in the second, slower phase seen in the progress curves in Figure 3.8. While the kinetic mechanism in Table 3.4 is the simplest mechanism that models the progress curves, it is likely that there is an ordered released of products, with the product that interacts with the S1’ pocket releasing slower. In fact, a kinetic mechanism with ordered product release does in fact simulate the progress curves well, and we believe that the product MCA-L-P-L, which fluoresces, interacts with the S1’ pocket and releases after the other hydrolysis product, which does not fluoresce. Co(II)-substituted analogs of MMP1, MMP3, MMP12, and MMP16 have been reported[39, 40, 60, 61]. To prepare heterobimetallic analogs (ZnCo) of MMP, two methods have been used. For MMP1 and MMP16, we dialyzed the as-isolated enzymes against phenanthroline, which chelated Zn(II) out of the catalytic site but not out of the structural site, and Co(II) was added to the resulting enzymes[39, 40, 60]. Spectroscopic studies were used to confirm that Co(II) was bound in the catalytic site in cdMMP1 and cdMMP16[39, 40]. Cha et al. used this approach to prepare 130

Co(II)-substituted MMP3[60]. Bertini and coworkers prepared a ZnCo analog of a mutated cdMMP12 by dialyzing the as-isolated enzyme versus millimolar concentrations of Co(II), and NMR spectra, albeit of the diamagnetic region, were reported[61]. In this work, the ZnCo analog of cdMMP7 was prepared by using the former method, and CD and fluorescence spectra demonstrate that this analog retains the secondary and tertiary structure of the as-isolate ZnZn analog, respectively. UV-Vis spectra revealed broad, weak ligand field transitions, indicating that Co(II) is 6-coordinate and is bound in the catalytic site. Co(II) bound in the structural site would have been expected to yield much more intense ligand field transitions[39, 40]. The ZnCo analog is now available to conduct spectroscopic studies on enzyme-inhibitor complexes. MMPs have been targets for inhibitors for decades[31, 35]. A wide variety of inhibitors have been reported[10, 32], and one common strategy to offer inhibitors is to use a zinc binding group (ZBG) to target the catalytic Zn(II). Substituents have been added to the ZBGs to target substrate binding pockets near the catalytic Zn(II) in an effort to afford inhibitor selectivity and specificity[30, 33]. The most common ZBG used is AHA, and most MMPs have been tested against AHA-based inhibitors[30]. However, most of the MMPs reported to date do not exhibit suitable selectivities or specificities to be clinically-useful. In addition, our studies herein and the studies of others[62, 63] demonstrate that AHA is a very weak binding inhibitor of the MMPs, warranting efforts to test other ZBGs. Indeed, our data presented herein demonstrate that maltol, thiomaltol, and allothiomaltol are all better inhibitors of cdMMP7, cdMMP16, and cdMMP1 (Table 3.5 and Figure 3.11). ITC measurements demonstrate that thiomaltol and allothiomaltol bind to cdMMP7 and cdMMP16 with low micromolar Kd values (Table 3.6). Despite similar structures, there does appear to be, albeit modest, selectivity of maltol, thiomaltol, and allothiomaltol binding to cdMMP7 and cdMMP16 (Figure 3.11), suggesting that these compounds could serve as better ZBGs, when compared to AHA, for future MMP inhibitors. The trends observed in the ITC data suggest a commonality between TM binding to cdMMP7 131

and cdMMP16. Docking studies using RosettaDock found that TM binding to both cdMMP7 and cdMMP16 feature thione sulfur binding to the catalytic zinc, with possible bidentate binding to zinc via the TM hydroxyl oxygen and burying of the TM methyl group into the S1’ pocket (Figure 3.13). In contrast, the ITC data suggests a weaker, yet similar affinity for ATM against cdMMPP16 and again weaker affinity of ATM against cdMMP7. The docking studies suggest that decreased affinity of ATM against cdMMP16 may be due to monodentate zinc binding through the ATM thione sulfur paired with burying of the ATM methyl group into the S1’ pocket of cdMMP16 (Figure 3.13B). In contrast, the docking studies indicate that the further decrease in affinity observed by ITC for ATM against cdMMP7 may be due to a binding orientation that precludes burying of the ATM methyl group into the cdMMP7 S1’ pocket, despite a possible bidentate zinc binding mode via the ATM thione sulfur and hydroxyl oxygen. Differences in binding mode for ATM/cdMMP7, in comparison to TM/cdMMP7, TM/cdMMP16, and ATM/cdMMP16 is consistent with a different entropic contribution to binding for ATM/cdMMP7 despite similarities in enthalpic contributions across the four ZBGs. In combination, the ITC data and docking studies indicate that inhibitor binding to MMPs is dictated both by inhibitor structure and the local protein structure surrounding the S1’ pocket. Future structural, biophysical, and spectroscopic studies of ZBG/cdMMP7 and ZBG/cdMMP16 complexes will be an important step to capitalizing upon these different binding modes to provide a potential avenue for introducing additional selectivity and specificity against different MMPs. With the preparation of a heterobimetallic analog of cdMMP7, we can now probe inhibitor binding using future spectroscopic studies, rather than relying completely on crystal structures of enzyme-inhibitor complexes. We hypothesize that the successful preparation of suitably specific and selective MMPis will require iterative design of inhibitors using tandem kinetic and structural studies on one or more MMPs from each of the distinct MMP subclasses. The generation of a stable heterobimetallic ZnCo analog of cdMMP7 will yield less complicated spectroscopic 132

results and remove the need to subtract spectroscopic signals associated with the structural metal binding site. With the preparation and characterization of a stable, well-characterized analog of MMP7 (and our previous work on MMP1[40], MMP16[39], and MMP3 (unpublished)), we need only a MMP from the two remaining classes of MMPs (gelatinases and elastases) to possess a MMP from each of representative MMP classes. Efforts to prepare and characterize cdMMP9 and cdMMP12 are underway.

133

3.5 Acknowledgements

We thank Dr. Carole L. Wilson (University of Washington) for providing pGEM7-based plasmid containing full-length MMP7 sequence. We thank Dr. Seth M. Cohen (University of California San Diego) for providing ZBGs. We thank Dr. Neil D. Danielson (Miami University Ohio) for assistance with fluorescence spectroscopic experiments. Experiments of ITC, ICP-AES and CD were supported by facilities of Department of Chemistry and Biochemistry in Miami University Ohio

134

3.6 Reference

1. Tallant, C., Marrero, A. and Gomis-Ruth, F. X. (2010) Matrix metalloproteinases: fold and function of their catalytic domains. Biochim. Biophys. Acta. 1803, 20-28 2. Nagase, H., Visse, R. and Murphy, G. (2006) Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc. Res. 69, 562-573 3. Rodgers, W. H., Matrisian, L. M., Navre, M., Giudice, L. C. and Gorstein, F. (1993) Expression and localization of matrilysin, a matrix metalloproteinase, in human endometrium during the reproductive cycle. Am. J. Obstet. Gynecol. 168, 253-260 4. Dunsmore, S. E., Saarialho-Kere, U. K., Roby, J. D., Wilson, C. L., Matrisian, L. M., Welgus, H. G. and Parks, W. C. (1998) Matrilysin expression and function in airway epithelium. J. Clin. Invest. 102, 1321 5. Crabbe, T., Willenbrock, F., Eaton, D., Hynds, P., Carne, A. F., Murphy, G. and Docherty, A. J. (1992) Biochemical characterization of matrilysin. Activation conforms to the stepwise mechanisms proposed for other matrix metalloproteinases. Biochemistry. 31, 8500-8507 6. Nagase, H., Enghild, J. J., Suzuki, K. and Salvesen, G. (1990) Stepwise activation mechanisms of the precursor of matrix metalloproteinase 3 (stromelysin) by proteinases and (4-aminophenyl) mercuric acetate. Biochemistry. 29, 5783-5789 7. Park, A., Matrisian, L., Kells, A., Pearson, R., Yuan, Z. and Navre, M. (1991) Mutational analysis of the transin (rat stromelysin) autoinhibitor region demonstrates a role for residues surrounding the" cysteine switch". J. Biol. Chem. 266, 1584-1590 8. Browner, M. F., Smith, W. W. and Castelhano, A. L. (1995) Matrilysin-inhibitor complexes: common themes among metalloproteases. Biochemistry. 34, 6602-6610 9. Terp, G. E., Cruciani, G., Christensen, I. T. and Jørgensen, F. S. (2002) Structural differences of matrix metalloproteinases with potential implications for inhibitor selectivity examined by the GRID/CPCA approach. J. Med. Chem. 45, 2675-2684

135

10. Jacobsen, J. A., Jourden, J. L. M., Miller, M. T. and Cohen, S. M. (2010) To bind zinc or not to bind zinc: an examination of innovative approaches to improved metalloproteinase inhibition. Biochim. Biophys. Acta-Mol. Cell Res. 1803, 72-94 11. Tallant, C., García-Castellanos, R., Baumann, U. and Gomis-Rüth, F. X. (2010) On the relevance of the Met-turn methionine in metzincins. J. Biol. Chem. 285, 13951-13957 12. Engel, C. K., Pirard, B., Schimanski, S., Kirsch, R., Habermann, J., Klingler, O., Schlotte, V., Weithmann, K. U. and Wendt, K. U. (2005) Structural basis for the highly selective inhibition of MMP-13. Chem. Biol. 12, 181-189 13. Ii, M., Yamamoto, H., Adachi, Y., Maruyama, Y. and Shinomura, Y. (2006) Role of matrix metalloproteinase-7 (matrilysin) in human cancer invasion, apoptosis, growth, and angiogenesis. Exp. Biol. Med. 231, 20-27 14. Wilson, C. L. and Matrisian, L. M. (1996) Matrilysin: an epithelial matrix metalloproteinase with potentially novel functions. Int. J. Biochem. Cell Biol. 28, 123-136 15. Mott, J. D. and Werb, Z. (2004) Regulation of matrix biology by matrix metalloproteinases. Curr. Opin. Cell Biol. 16, 558-564 16. Agrawal, A., Romero-Perez, D., Jacobsen, J. A., Villarreal, F. J. and Cohen, S. M. (2008) Zinc ‐ Binding Groups Modulate Selective Inhibition of MMPs. ChemMedChem. 3, 812-820 17. Butler, G. S. and Overall, C. M. (2013) Matrix metalloproteinase processing of signaling molecules to regulate inflammation. Periodontol. 2000. 63, 123-148 18. Dufour, A. and Overall, C. M. (2013) Missing the target: matrix metalloproteinase antitargets in inflammation and cancer. Trends Pharmacol. Sci. 34, 233-242 19. Fortelny, N., Yang, S., Pavlidis, P., Lange, P. F. and Overall, C. M. (2015) Proteome TopFIND 3.0 with TopFINDer and PathFINDer: database and analysis tools for the association of protein termini to pre-and post-translational events. Nucleic Acids Res. 43, D290-D297

136

20. Yokoyama, Y., Grünebach, F., Schmidt, S. M., Heine, A., Häntschel, M., Stevanovic, S., Rammensee, H.-G. and Brossart, P. (2008) Matrilysin (MMP-7) is a novel broadly expressed tumor antigen recognized by antigen-specific T cells. Clin. Cancer. Res. 14, 5503-5511 21. Szklarczyk, A., Oyler, G., McKay, R., Gerfen, C. and Conant, K. (2007) Cleavage of neuronal synaptosomal-associated protein of 25 kDa by exogenous matrix metalloproteinase‐7. J. Neurochem. 102, 1256-1263 22. Prior, S. H., Fulcher, Y. G., Koppisetti, R. K., Jurkevich, A. and Van Doren, S. R. (2015) Charge-triggered membrane insertion of matrix metalloproteinase-7, supporter of innate immunity and tumors. Structure. 23, 2099-2110 23. Chaturvedi, S. and Hass, R. (2011) Extracellular signals in young and aging breast epithelial cells and possible connections to age-associated breast cancer development. Mech. Ageing Dev. 132, 213-219 24. Wilson, C. L., Heppner, K. J., Labosky, P. A., Hogan, B. L. and Matrisian, L. M. (1997) Intestinal tumorigenesis is suppressed in mice lacking the metalloproteinase matrilysin. Proc. Nat. Acad. Sci. 94, 1402-1407 25. Fukuda, A., Wang, S. C., Morris, J. P., Folias, A. E., Liou, A., Kim, G. E., Akira, S., Boucher, K. M., Firpo, M. A. and Mulvihill, S. J. (2011) Stat3 and MMP7 contribute to pancreatic ductal adenocarcinoma initiation and progression. Cancer Cell. 19, 441-455 26. Konttinen, Y. T., Ainola, M., Valleala, H., Ma, J., Ida, H., Mandelin, J., Kinne, R. W., Santavirta, S., Sorsa, T. and López-Otín, C. (1999) Analysis of 16 different matrix metalloproteinases (MMP-1 to MMP-20) in the synovial membrane: different profiles in trauma and rheumatoid arthritis. Ann. Rheum. Dis. 58, 691-697 27. Saarialho-Kere, U. (1998) Patterns of matrix metalloproteinase and TIMP expression in chronic ulcers. Arch. Dermatol. Res. 290, S47-S54 28. Saarialho-Kere, U. K., Vaalamo, M., Puolakkainen, P., Airola, K., Parks, W. C. and Karjalainen-Lindsberg, M.-L. (1996) Enhanced expression of matrilysin, collagenase, and stromelysin-1 in gastrointestinal ulcers. Am.J. Ppathol. 148, 519 137

29. Wielockx, B., Libert, C. and Wilson, C. (2004) Matrilysin (matrix metalloproteinase-7): a new promising drug target in cancer and inflammation? Cytokine Growth Factor Rev. 15, 111-115 30. Vandenbroucke, R. E. and Libert, C. (2014) Is there new hope for therapeutic matrix metalloproteinase inhibition? Nature Rev. Drug Dis. 13, 904-927 31. Fingleton, B. (2008) MMPs as therapeutic targets—still a viable option? In Semin. Cell Dev. Biol. 19. 61-68 32. Cathcart, J., Pulkoski-Gross, A. and Cao, J. (2015) Targeting Matrix Metalloproteinases in Cancer: Bringing New Life to Old Ideas. Genes Dis. 2, 26-34 33. Fabre, B., Ramos, A. and de Pascual-Teresa, B. (2014) Targeting Matrix Metalloproteinases: Exploring the Dynamics of the S1′ Pocket in the Design of Selective, Small Molecule Inhibitors: Miniperspective. J. Med. Chem. 57, 10205-10219 34. Aureli, L., Gioia, M., Cerbara, I., Monaco, S., Fasciglione, G. F., Marini, S., Ascenzi, P., Topai, A. and Coletta, M. (2008) Structural bases for substrate and inhibitor recognition by matrix metalloproteinases. Curr. Med. Chem. 15, 2192-2222 35. Skiles, J. W., Gonnella, N. C. and Jeng, A. Y. (2004) The design, structure, and clinical update of small molecular weight matrix metalloproteinase inhibitors. Curr. Med. Chem. 11, 2911-2977 36. Rowsell, S., Hawtin, P., Minshull, C. A., Jepson, H., Brockbank, S. M., Barratt, D. G., Slater, A. M., McPheat, W. L., Waterson, D. and Henney, A. M. (2002) Crystal structure of human MMP9 in complex with a reverse hydroxamate inhibitor. J. Mol. Biol. 319, 173-181 37. Antharavally, B. S., Mallia, K. A., Rangaraj, P., Haney, P. and Bell, P. A. (2009) Quantitation of proteins using a dye–metal-based colorimetric protein assay. Anal. Biochem. 385, 342-345 38. Gasteiger, E., Hoogland, C., Gattiker, A., Duvaud, S. e., Wilkins, M. R., Appel, R. D. and Bairoch, A. (2005) Protein identification and analysis tools on the ExPASy server. Springer 138

39. Meng, F., Yang, H., Aitha, M., George, S., Tierney, D. L. and Crowder, M. W. (2016) Biochemical and spectroscopic characterization of the catalytic domain of MMP16 (cdMMP16). J. Biol. Inorg. Chem., 1-13 40. Yang, H., Makaroff, K., Paz, N., Aitha, M., Crowder, M. W. and Tierney, D. L. (2015) Metal Ion Dependence of the Matrix Metalloproteinase-1 Mechanism. Biochemistry. 54, 3631-3639 41. Michaelis, L. and Menten, M. L. (1913) Die kinetik der invertinwirkung. Biochem. z. 49, 352 42. Motulsky, H. (2007) Prism 5 statistics guide, 2007. GraphPad Software, 1-26 43. Galvez, J., Varon, R., Canovas, F. G. and Carmona, F. G. (1982) IV. Transient phase of the Uni-Bi mechanisms. J. Theor. Biol. 94, 413-420 44. Kuzmič, P. (2009) DynaFit—a software package for enzymology. Methods Enzymol. 467, 247-280 45. King, E. L. and Altman, C. (1956) A schematic method of deriving the rate laws for enzyme-catalyzed reactions. J. Phys. Chem. 60, 1375-1378 46. Cha, J., Pedersen, M. V. and Auld, D. S. (1996) Metal and pH dependence of heptapeptide catalysis by human matrilysin. Biochemistry. 35, 15831-15838 47. Whitmore, L. and Wallace, B. (2004) DICHROWEB, an online server for protein secondary structure analyses from circular dichroism spectroscopic data. Nucleic Acids Res. 32, W668-W673 48. Martin, D. P., Blachly, P. G., Marts, A. R., Woodruff, T. M., de Oliveira, C. s. A., McCammon, J. A., Tierney, D. L. and Cohen, S. M. (2014) ‘Unconventional’Coordination Chemistry by Metal Chelating Fragments in a Metalloprotein Active Site. J. Am. Chem. Soc. 136, 5400-5406 49. Freyer, M. W. and Lewis, E. A. (2008) Isothermal titration calorimetry: experimental design, data analysis, and probing macromolecule/ligand binding and kinetic interactions. Methods Cell Biol. 84, 79-113 50. Lang, R., Braun, M., Sounni, N. E., Noël, A., Frankenne, F., Foidart, J.-M., Bode, W. and Maskos, K. (2004) Crystal structure of the catalytic domain of 139

MMP-16/MT3-MMP: characterization of MT-MMP specific features. J. Mol. Biol. 336, 213-225 51. Chaudhury, S., Berrondo, M., Weitzner, B. D., Muthu, P., Bergman, H. and Gray, J. J. (2011) Benchmarking and analysis of protein docking performance in Rosetta v3. 2. PLoS One. 6, e22477 52. Hawkins, P. C., Skillman, A. G., Warren, G. L., Ellingson, B. A. and Stahl, M. T. (2010) Conformer generation with OMEGA: algorithm and validation using high quality structures from the Protein Databank and Cambridge Structural Database. J. Chem. Inf. Model. 50, 572-584 53. Davis, I. W. and Baker, D. (2009) RosettaLigand docking with full ligand and receptor flexibility. J. Mol. Biol. 385, 381-392 54. Meiler, J. and Baker, D. (2006) ROSETTALIGAND: Protein–small molecule docking with full side-chain flexibility. Proteins: Struct., Funct., Bioinf. 65, 538-548 55. Oneda, H. and Inouye, K. (1999) Refolding and recovery of recombinant human matrix metalloproteinase 7 (matrilysin) from inclusion bodies expressed by Escherichia coli. J. Biochem. 126, 905-911 56. Breece, R. M., Costello, A., Bennett, B., Sigdel, T. K., Matthews, M. L., Tierney, D. L. and Crowder, M. W. (2005) A five-coordinate metal center in Co (II)-substituted VanX. J. Biol. Chem. 280, 11074-11081 57. Bertini, I. and Luchinat, C. (1983) Cobalt (II) as a probe of the structure and function of carbonic anhydrase. Acc. Chem. Res. 16, 272-279 58. Muta, Y., Yasui, N., Matsumiya, Y., Kubo, M. and Inouye, K. (2010) Expression in Escherichia coli, refolding, and purification of the recombinant mature form of human matrix metalloproteinase 7 (MMP-7). Biosci., Biotechnol., Biochem. 74, 2515-2517 59. Lovejoy, B., Welch, A. R., Carr, S., Luong, C., Broka, C., Hendricks, R. T., Campbell, J. A., Walker, K. A., Martin, R. and Van Wart, H. (1999) Crystal structures

140

of MMP-1 and-13 reveal the structural basis for selectivity of collagenase inhibitors. Nat. Struct. Mol. Biol. 6, 217-221 60. Cha, J., Sørensen, M. V., Ye, Q.-Z. and Auld, D. S. (1998) Selective replacement of the catalytic zinc of the human stromelysin-1 catalytic domain. J. Biol. Inorg. Chem. 3, 353-359 61. Bertini, I., Fragai, M., Lee, Y. M., Luchinat, C. and Terni, B. (2004) Paramagnetic metal ions in ligand screening: the CoII matrix metalloproteinase 12. Angew. Chem. Int. Ed. 43, 2254-2256 62. Puerta, D. T., Lewis, J. A. and Cohen, S. M. (2004) New beginnings for matrix metalloproteinase inhibitors: identification of high-affinity zinc-binding groups. J. Am. Chem. Soc. 126, 8388-8389 63. Puerta, D. T., Griffin, M. O., Lewis, J. A., Romero-Perez, D., Garcia, R., Villarreal, F. J. and Cohen, S. M. (2006) Heterocyclic zinc-binding groups for use in next-generation matrix metalloproteinase inhibitors: potency, toxicity, and reactivity. J. Biol. Inorg. Chem. 11, 131-138

141

3.7 Tables and figures

Table 3.1: Steady-state kinetic constants and metal content for recombinant cdMMP7 analogs

-1 -1 -1 kcat (s ) Km (μM) kcat/KM (s μM ) Metal/protein ratio

Zn Co

Zn2-cdMMP7 0.40 + 0.02 23 + 2 0.017 1.8 + 0.1 <0.1 (as-isolated)

ZnCo-cdMMP7 0.17 + 0.04 16 + 3 0.011 0.7 1.2

Steady-state kinetic reactions were conducted in 50 mM Hepes, pH 7.5, containing

o 100 mM NaCl and 5 mM CaCl2 at 25 C. The substrate was FS-6 (Sigma Aldrich

MCA-fluorescence peptide), and the concentration of Zn2-cdMMP7 was 10 M, and the concentration of ZnCo-cdMMP7 was 20 M

142

Table 3.2: Secondary structural content of Zn2-cdMMP7 and ZnCo-cdMMP7 analogs

-helix -sheet Random loop

Zn2-cdMMP7 27% 33% 40%

Inactive Zn2-cdMMP7 27% 33% 40%

ZnCo-cdMMP7 28% 32% 40%

Enzyme concentration was 10 mM, and the buffer was 50 mM phosphate, pH 7.5, containing 100 mM NaCl. The secondary structure composition was estimated by using diChroWeb[47].

143

Table 3.3: Microscopic kinetic constants of the hydrolysis of BML-P131 by

Zn2-cdMMP7

-1 -1 -1 -1 k1 (μM s ) k-1 (s ) k2 (s ) Ks (M) cdMMP7 0.0079 0.83 0.001 30.5 cdMMP1 0.35 10.7 2.0 NR cdMMP16 0.30 1X10-6 2.7 3.71

Stopped-flow fluorescence reactions were conducted in 50 mM Hepes, pH 7.5,

o containing 100 mM NaCl and 5 mM CaCl2 at 25 C. Progress curves for data in Figure 3.7 were generated by using Dynafit 3.4[44]. Data on cdMMP1 and cdMMP16 were previously published[39, 40]. NR – not reported.

144

Table 3.4: Microscopic kinetic constants of as-isolated Zn2-cdMMP7 and

Zn2-cdMMP16 from substrate-emission stopped-flow studies

cdMMP7 cdMMP16

-1 -1 k1 (μM s ) 374 524

-1 k-1 (s ) 10,000 6,835

-1 k2 (s ) 0.6 1.6 k-2 (s-1) 0.0008 0.01

-1 k3 (s ) 23.3 21

-1 -1 k-3 (μM s ) 0.01 0.001

Reactions were conducted in 50 mM Hepes, pH 7.5, containing 100 mM NaCl and 5

o mM CaCl2 at 25 C. Progress curves for data in Figure 3.8 were produced by using

Dynafit 3.4 [44].

145

Table 3.5: IC50 values of AHA, maltol, TM, and ATM as inhibitors of cdMMP16 and cdMMP7.

cdMMP7 cdMMP16

IC50 (M) Potency vs IC50 (M) Potency vs AHA AHA

AHA 55,800 + 1,730 ------41,000 + 1,000 -----

Maltol 7,308 + 890 7.6 10,300 + 1,900 4

TM 134 + 13 416 30 + 5 1370

ATM 103 + 9 544 43 + 3 953

Inhibition studies were conducted with 10 M enzyme and 10 M MCA substrate at

o 25 C in 50 mM Hepes, pH 7.5, containing 100 mM NaCl and 5 mM CaCl2.

146

Table 3.6: ITC measurements on cdMMP7 and cdMMP16 using ATM and TM as binding groups.

MMPs ZBGs Kd △H △S △G n

M) (kJ/mol) (J/molK) (kJ/mol)

TM 3.0 -22 33 -32 1.0

ATM 5.8 -24 21 -30 0.9

Zn2-MMP7 TM 3.2 -23 25 -30 1.1

ATM 8.5 -26 11 -29 1.0

The buffer used in these experiments was 50 M using 50 mM Hepes, pH 7.5, containing 100 mM NaCl and 5 mM CaCl2 for cdMMP16 and 50 mM Tris, pH 7.5, containing 5 mM ZnCl2, and 0.1% (w/v) Brij35 for cdMMP7. The enzyme concentration for MMP16 and MMP7 was 50 M.

147

Figure 3.1: Crystal structure of Zn2-cdMMP7 (PDB #1MMB) [8]. Zn(II) and Ca are colored gray and blue, respectively. Right: Zn(II) binding sites are shown with metal binding amino acids labeled and S1’ substrate binding pockets shown in magenta color [12]. The Met-turn is shown in orange[8]. \

148

Figure 3.2: Structure of substrate A. FS-6 substrate and B. BML-P131 substrate. The scissile bonds are labeled with the red-dashed box.

149

Figure 3.3: SDS-PAGE analysis of the two cdMMP7 refolding procedures. A: cdMMP7 prepare by using the Cha method[46]: Lane 1: Perfect protein molecular weight markers, Lane 2: boiled cell fraction of E. coli cells after 4 hours of induction with 1 mM IPTG, Lane 3: supernatant of culture after lysis by French press and centrifugation, Lane 4: resuspended inclusion bodies after French press and washing, Lane 5: inclusion bodies after solubilization in 8 M urea, Lane 6: solubilized inclusion bodies after centrifugation, Lane 7: inclusion bodies after one-step dialysis (final concentration of urea is 0 M), Lane 8: refolding product after centrifugation, and Lane 9: cdMMP7 after hemoglobin column. B: cdMMP7 prepared by optimized method: Lane 1: Perfect protein molecular weight markers, Lane 2: boiled cell fractionref of E. coli cells cells after 4 hours of induction with 1 mM IPTG, Lane 3: re- solubilized inclusion bodies in 8 M urea after centrifugation, Lanes 4-5: refolded cdMMP7 after optimized dialysis method, Lanes 6-7: refolded cdMMP7 after centrifugation, and Lanes 8-9: concentrated, refolded cdMMP7 after concentration using ultrafiltration unit.

150

Figure 3.4: Circular dichroism spectra of Zn2-cdMMP7 (solid), ZnCo-cdMMP7

(dashed line), and inactive Zn2-cdMMP7 (dotted line)

151

Figure 3.5: Fluorescence emission spectra of ZnCo-cdMMP7 (dash) and as-isolated

Zn2-cdMMP7 (solid).

152

Figure 3.6: Fluorescence emission spectra of inactive Zn2-cdMMP7 (dash) and as-isolated, Zn2-cdMMP7 (solid).

153

Figure 3.7: Stopped-flow fluorescence traces of reactions of Zn2-cdMMP7 and with substrate BML-P131. The buffer was 50 mM Hepes, pH 7.5, containing 100 mM

o NaCl and 5 mM CaCl2, and the reactions were conducted at 25 C. The simulated progress curves, which were generated by using Dynafit 3.4 and the kinetic mechanism in Table 2, are represented by the solid lines, and the stopped-flowed fluorescence data are represented by the symbols. The final concentration of cdMMP7 in these reactions was 10 M.

154

Figure 3.8: Stopped-flow fluorescence traces of reactions of A Zn2-cdMMP7 and B

Zn2-cdMMP16 with FS-6 fluorescent substrate. The buffer was 50 mM Hepes, pH 7.5,

o containing 100 mM NaCl and 5 mM CaCl2, and the reactions were conducted at 25 C. The simulated progress curves are represented by the solid lines, and the stopped-flow fluorescence data are represented by the symbols. The final concentrations of the cdMMPs were 0.5 M.

155

Figure 3.9: Uv-vis spectrum of of the ZnCo-cdMMP7 analog( right top corner) and UV-Vis difference (spectrum of ZnCo analog minus the spectrum of metal-free enzyme) spectrum in 50 mM Hepes, pH 7.5, containing 100 mM NaCl and 5 mM

CaCl2.

156

Figure 3.10: Structures of zinc binding group (ZBG) inhibitors used in this study (from left to right): acetohydroxamic acid (AHA), maltol, thiomaltol (TM), and allothiomaltol (ATM).

157

Figure 3.11: Inhibitory potency of AHA, maltol, TM, and ATM against cdMMP7 and cdMMP16 relative to cdMMP1 (H. Yang, unpublished data).

158

Figure 3.12: ITC results of ZBG binding to cdMMP7 and cdMMP16. Titration of

TM (A) and ATM (B) into 50 M Zn2-cdMMP16 in 50 mM Hepes, pH 7.5,

o containing 100 mM NaCl and 5 mM CaCl2 at 25 C. Titrations of TM (C) / ATM

(D) into 50M Zn2-cdMMP7 in 50 mM Tris, pH 7.5, containing 5 mM CaCl2 and 0.1% Brij 35 at 25 oC.

159

Figure 3.13: Representative binding modes from docking simulations. RosettaDock simulations of cdMMP7 (A) with TM (cyan) and ATM (blue). RosettaDock simulations of cdMMP16 (B) with TM (purple) and ATM (pink). Throughout, protein backbones are show as cartoons, relevant interacting side chains shown as lines, bound ions shown as spheres, and TM and ATM are shown as sticks.

160

Chapter 4

Conclusion of Dissertation

Fan Meng

Department of Chemistry and Biochemistry, Miami University, Oxford, Ohio 45056

*The author thanks Dr. Michael W. Crowder for his significant contribution to this chapter

161

4.1 Conclusions

Due to the clinical failures of broad-spectrum MMPi, there has emerged debate about whether MMPs could be targets for new therapeutic agents [1, 2]. Research has shown that MMPs regulate and are involved with a myriad of physiological and pathological pathways and that MMPs exhibit highly distinct, cell-specific expression patterns and substrate preferences [3-7]. Off-target inhibition of other MMPs and other Zn(II)-metalloproteinases is likely the main reason for failures in clinical trials [5, 8, 9]. Although certain MMPs clearly promote tumorigenesis in multiple types of cancer, studies have shown that other MMPs may play protective roles in some cells [9, 10]. Therefore, the Tierney and Crowder groups hypothesize that the successful targeting of MMPs to treat cancer or inflammation diseases requires highly-selective and –specific inhibitors. Moreover, the Tierney and Crowder groups postulate that identifying specific and selective inhibitors requires screening potential compounds against multiple MMPs, for example an enzyme from each of the distinct MMP subclasses [11]. Towards this goal, our groups have selected MMP1, MMP3, MMP7, MMP9, MMP12, and MMP16 as suitable representative enzymes from each of the MMP subclasses, and efforts are being made to clone, over-express, purify, and characterize each enzyme. This dissertation focuses on two of these representative MMPs: MMP16, which belongs to the type-I MT-MMPs subclass, and MMP7, which belongs to the matrilysin subclass of secreted MMPs. These two MMPs are expressed in different tissues and associated with different bio-functional events and pathological processes. Therefore, MMP7 and MMP16 are considered as potential drug-targets for distinct diseases. The research in my dissertation can be divided into three major areas: (a) preparation and characterization of the catalytic domains of Zn2-cdMMP7 and Zn2-cdMMP16, (b) preparation and characterization of Co(II)-substituted analogs of cdMMP7 and cdMMP16, and (c) characterization of four inhibitors with cdMMP7 and cdMMP16.

162

Although the catalytic domains of most MMPs are structurally-similar, we discovered that suitable analogs of cdMMP7 and cdMMP16 cannot be obtained using the published protocols used to purify MMP1, MMP3, and MMP12 [12-14]. Therefore, significant effort was made to optimize the preparation of these enzymes. For example in order to obtain large quantities of active cdMMP7 and cdMMP16 from our E. coli-based system, we had to improve the stability of refolding intermediates, and each enzyme required a different strategy (step-wise decrease in urea concentration for cdMMP16 and increased concentration of detergent Brij35 for cdMMP7). We anticipate that the successful purification of cdMMP3, cdMMP9, and cdMMP12 will require optimization of the purification protocol, and we now have a large collection of strategies to accomplish this goal. This dissertation describes our efforts at using pre-steady state kinetics to probe the kinetic mechanisms of these enzymes. There are no known UV-Vis-active substrates that can be used to interrogate the kinetic mechanism of a MMP. Therefore, we used stopped-flow fluorescence studies on cdMMP7 and cdMMP16 (Chapters 2 and 3). Like work previously published by our groups [14], we used the substrate BML-128 to probe the mechanism. In these studies, the “reporter” was a tryptophan, presumably the conserved Trp (Trp159 in cdMMP16 and Trp154 in cdMMP7) that is 10 Å from the catalytic Zn(II), and the binding of substrate and hydrolysis reaction resulted in changes in the intrinsic fluorescence properties of this tryptophan. The resulting fluorescence progress curves were fitted to a kinetic mechanism that shows a rate-limiting hydrolysis step. We also examined the mechanism by using substrate FS-6; however in these experiments, we followed the increase in fluorescence as the substrate was hydrolyzed (Chapter 3), and the “reporter” was one of the hydrolysis products. Interestingly, both cdMMP7 and cdMMP16 exhibited biphasic progress curves, and the progress curves could be adequately simulated by assuming a simple uni-bi hydrolytic mechanism [15]. These studies showed that cdMMP7 exhibited slower substrate binding and hydrolysis rates than cdMMP16, which is consistent with the lower steady-state kcat exhibited by cdMMP7. We attribute the kinetic 163

differences to subtle structural differences in the active sites of cdMMP7 and cdMMP16, particularly in the S1’ pockets of the enzymes [8, 16]. Based on our studies, we believe that the activities of certain MMPs could be attenuated by using peptidic substrates with products that exhibit varying leaving group potential. Substrates that interact strongly with the S1’ pocket in one MMP (but not all MMPs) would be expected to exhibit a slow release step. While there are a number of crystal structures of MMPs with and without bound inhibitors [16-20], a common challenge is to obtain crystal structures of MMP-inhibitor complexes. In fact, our efforts, albeit preliminary, were unsuccessful at obtaining suitable crystals of cdMMP16 (F. Meng, S. George, unpublished results). Therefore, we are interested in being able to use spectroscopy to probe MMP-inhibitor interactions, particularly inhibitors that are thought to bind directly to the catalytic Zn(II). Unfortunately, Zn2-cdMMPs can not be interrogated using common (except EXAFS) spectroscopic methods, such as UV-vis, NMR and EPR, due to the presence of the diamagnetic, “spectroscopically-silent” [13] Zn(II) ions. Therefore, we established a goal of preparing Co(II)-substituted analogs of cdMMP7 and cdMMP16. Three major achievements were realized. Firstly, we were able to prepare the first Co(II)-substituted analog of a MT-MMP (cdMMP16), We developed a new method to prepare a homodimetallic analog (Co2-cdMMP16), and we were able to prepare heterodimetallic analogs (ZnCo-cdMMP7 and ZnCo-cdMMP16) with Co(II) in the catalytic sites and Zn(II) in the structural sites. The heterodimetallic analogs now afford us an advantage of studying only the catalytic site upon inhibitor binding when using spectroscopic studies. Secondly, the studies in this dissertation are the first to report kinetic and structural features of an as-isolated MT-cdMMPs and a matrilysin by steady-state, ICP-AES, CD spectroscopy, and fluorescence emission spectroscopy. Thirdly, the work in this dissertation, along with studies previously reported by Yang et al. [14], are the first to probe the metal binding site(s) with spectroscopic methods, like EXAFS, NMR, and UV-vis spectroscopies. These Co(II)-substituted analogs can be used in future inhibitor binding as well as in 164

biophysical studies. For example, previous kinetic studies have suggested that the coordination number of the catalytic site varies from 4 to 6 during catalysis [18, 21-23]. These heterobimetallic analogs of the cdMMPs will allow for rapid-freeze quench (RFQ) spectroscopic studies to be conducted. These same analogs may be tested with stopped-flow UV-Vis studies, and the ligand field transitions of Co(II) can be monitored during catalysis [24]. These future studies will likely reveal detailed mechanistic information and allow for a much clearer understanding of how these enzymes catalyze the hydrolysis of peptides/proteins. When we were testing methods to substitute Zn(II) with Co(II), we discovered that cdMMP16 undergoes auto-proteolysis as the metal ions are removed from the enzyme. This proteolysis appears to be unique to the MT-MMP on which we worked and was not reported in studies on MMP1 [14], MMP3 [12], MMP7 [25], and

MMP12 [26]. Auto-proteolysis of the catalytic domains of MMP14 and MMP16 had been previously reported in in vivo studies, and the authors of this work hypothesized that auto-proteolysis may be part of a regulation system for these enzymes [27, 28]. In our research, we reveal that the auto-proteolysis can be triggered by de-metallization on the MMP16’s catalytic domain, If de-metallization followed by proteolysis is an actual regulatory pathway for the MT-MMPs, then it would be interesting to study whether mutations at the proteolysis sites are associated with MT-MMP’s evasion from regulatory pathway, and resulting in excess accumulation of MT-MMPs during pathological processes. Our ITC and inhibition studies showed that TM and ATM exhibit higher binding affinities against MMP16 and MMP7 than maltol and AHA. In addition, we found that TM exhibited a stronger binding affinity to MMP7 and MMP16 and that ATM is a weaker binder to both enzymes. Docking studies predict that the lower binding affinity of ATM is due to an altered binding mode, with the bidentate Zn(II) binding of TM changing to monodentate when ATM binds. The change in binding mode of ATM is attributed to a steric effect with the methyl group of ATM and the amino acid residues in the S1’ pocket. A similar argument was invoked by Yang et al. 165

when examining the binding of TM and ATM to cdMMP1 (Yang et al., unpublished results). Future spectroscopic studies using the heterobimetallic analogs of cdMMP7 and cdMMP16 will shed light on the binding of these ZBGs. For example, the electronic structure of Co(II) will change upon ZBG binding, and these changes can be monitored using UV-vis, CD, and CW-EPR spectroscopies [14]. 1H- NMR and EXAFS spectroscopies can also be used to probe how ZBGs bind to the enzymes. These studies, coupled with ITC, inhibition, and docking studies, should provide a better understanding of the relationship between the inhibitory potencies of the ZBGs with binding to the metal centers and the S1’ pockets. Our strategy to identify more specific and selective MMPi involves mechanistic, structural, and inhibition studies on a MMP from each of the distinct MMP subclasses. The Crowder/Tierney groups have now characterized MMP1, MMP7, and MMP16. Currently, studies on MMP9 (gelatinase) and MMP12 (metalloelastases) are on-going. MMP9 and MMP12 are involved in the immune response and inflammation [6]. Specifically, MMP9 appears to be the main switch for angiogenesis, when the enzyme hydrolyzes the ECM and releases growth factors [29, 30]. MMP12 mediates macrophage trafficking by processing bioactive signaling molecules [31]. Dysregulation of MMP9 results in a number of pathological events, such as cancer, abdominal sepsis, autoimmunity, brain edema, and hemorrhage [5].

Dysregulation of MMP12 has been associated with cancer and autoimmunity [5]. Moreover, MMP9 and MMP12 are the most extensively studied MMPs from their representative subclasses, and both enzymes have been characterized crystallographically [17, 19]. Bertini and coworkers have prepared Co(II)-substituted analogs of cdMMP12 [26]. The S1’ pockets in MMP1, MMP7, MMP9, MMP12, and MMP16, which are the enzymes studied by the Crowder/Tierney groups, are vastly different [8, 32] (Figure 4.1), and these enzymes will allow for future inhibitor binding studies to probe the role of this specificity pocket in inhibitor/drug binding. In MMP1, the S1’ pocket is shallow but flexible, while in MMP7, the pocket is shallow and rigid [18, 20]. The S1’ pockets in MMP12 and MMP9 are mid-sized [17, 19], and 166

in MMP16, the pocket is deep and tunnel-like [16]. The use of TM, ATM, maltol,

AHA, and other inhibitors [8, 33] (Table 4.1) will allow for structure-activity relationship (SAR) map of the S1’ pockets in the MMPs. In summary, this dissertation described our efforts in characterizing cdMMP7 and cdMMP16 and in preparing Co(II)-substituted analogs of these enzymes. These studies are part of a much larger overall effort to prepare selective and specific inhibitors against the MMPs. It is hoped that our studies and our approach will lead to improved inhibitors, which can be used to as therapeutics to address MMP-associated diseases.

167

4.2 References

1. Vandenbroucke, R. E. and Libert, C. (2014) Is there new hope for therapeutic matrix metalloproteinase inhibition? Nature Rev. Drug Dis. 13, 904-927 2. Fingleton, B. (2008) MMPs as therapeutic targets—Still a viable option? Semin. Cell Dev. Biol. 19, 61-68 3. Yan, C. and Boyd, D. D. (2007) Regulation of matrix metalloproteinase gene expression. J. Cell. Physiol. 211, 19-26 4. Kessenbrock, K., Plaks, V. and Werb, Z. (2010) Matrix metalloproteinases: regulators of the tumor microenvironment. Cell. 141, 52-67 5. Dufour, A. and Overall, C. M. (2013) Missing the target: matrix metalloproteinase antitargets in inflammation and cancer. Trends Pharmacol. Sci. 34, 233-242 6. Rodríguez, D., Morrison, C. J. and Overall, C. M. (2010) Matrix metalloproteinases: what do they not do? New substrates and biological roles identified by murine models and proteomics. Biochim. Biophys. Acta-Mol. Cell Res. 1803, 39-54 7. Itoh, Y. (2015) Membrane-type matrix metalloproteinases: Their functions and regulations. Matrix Biol. 44-46, 207-223 8. Jacobsen, J. A., Jourden, J. L. M., Miller, M. T. and Cohen, S. M. (2010) To bind zinc or not to bind zinc: an examination of innovative approaches to improved metalloproteinase inhibition. Biochim. Biophys. Acta-Mol. Cell Res. 1803, 72-94 9. Cathcart, J., Pulkoski-Gross, A. and Cao, J. (2015) Targeting matrix metalloproteinases in cancer: Bringing new life to old ideas. Genes Dis. 2, 26-34 10. Butler, G. S. and Overall, C. M. (2013) Matrix metalloproteinase processing of signaling molecules to regulate inflammation. Periodontol. 2000. 63, 123-148 11. Nagase, H., Visse, R. and Murphy, G. (2006) Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc. Res. 69, 562-573

168

12. Cha, J., Sørensen, M. V., Ye, Q.-Z. and Auld, D. S. (1998) Selective replacement of the catalytic zinc of the human stromelysin-1 catalytic domain. J. Biol.l Inorg. Chem. 3, 353-359 13. Bertini, I. and Luchinat, C. (1983) Cobalt (II) as a probe of the structure and function of carbonic anhydrase. Acc. Chem. Res. 16, 272-279 14. Yang, H., Makaroff, K., Paz, N., Aitha, M., Crowder, M. W. and Tierney, D. L. (2015) Metal ion dependence of the matrix metalloproteinase-1 mechanism. Biochemistry. 54, 3631-3639 15. Galvez, J., Varon, R., Canovas, F. G. and Carmona, F. G. (1982) IV. Transient phase of the Uni-Bi mechanisms. J. Theor. Biol. 94, 413-420 16. Lang, R., Braun, M., Sounni, N. E., Noël, A., Frankenne, F., Foidart, J.-M., Bode, W. and Maskos, K. (2004) Crystal structure of the catalytic domain of MMP-16/MT3-MMP: characterization of MT-MMP specific features. J. Mol. Biol. 336, 213-225 17. Rowsell, S., Hawtin, P., Minshull, C. A., Jepson, H., Brockbank, S. M., Barratt, D. G., Slater, A. M., McPheat, W. L., Waterson, D. and Henney, A. M. (2002) Crystal structure of human MMP9 in complex with a reverse hydroxamate inhibitor. J. Mol. Biol. 319, 173-181 18. Browner, M. F., Smith, W. W. and Castelhano, A. L. (1995) Matrilysin-inhibitor complexes: common themes among metalloproteases. Biochemistry. 34, 6602-6610 19. Lang, R., Kocourek, A., Braun, M., Tschesche, H., Huber, R., Bode, W. and Maskos, K. (2001) Substrate specificity determinants of human macrophage elastase (MMP-12) based on the 1.1 Å crystal structure. J. Mol. Biol. 312, 731-742 20. Lovejoy, B., Welch, A. R., Carr, S., Luong, C., Broka, C., Hendricks, R. T., Campbell, J. A., Walker, K. A., Martin, R. and Van Wart, H. (1999) Crystal structures of MMP-1 and-13 reveal the structural basis for selectivity of collagenase inhibitors. Nat. Struct. Mol. Biol. 6, 217-221

169

21. Bertini, I., Calderone, V., Fragai, M., Luchinat, C., Maletta, M. and Yeo, K. J. (2006) Snapshots of the reaction mechanism of matrix metalloproteinases. Angew. Chem. Int. Ed. 45, 7952-7955 22. Manzetti, S., McCulloch, D. R., Herington, A. C. and van der Spoel, D. (2003) Modeling of enzyme–substrate complexes for the metalloproteases MMP-3, ADAM-9 and ADAM-10. J. Comput.-Aided Mol. Des. 17, 551-565 23. Pelmenschikov, V. and Siegbahn, P. E. (2002) Catalytic mechanism of matrix metalloproteinases: two-layered ONIOM study. Inorg. Chem. 41, 5659-5666 24. Matthews, M. L., Periyannan, G., Hajdin, C., Sidgel, T. K., Bennett, B. and Crowder, M. W. (2006) Probing the reaction mechanism of the D-ala-D-ala dipeptidase, VanX, by using stopped-flow kinetic and rapid-freeze quench EPR studies on the Co (II)-substituted enzyme. J. Am. Chem. Soc. 128, 13050-13051 25. Cha, J., Pedersen, M. V. and Auld, D. S. (1996) Metal and pH dependence of heptapeptide catalysis by human matrilysin. Biochemistry. 35, 15831-15838 26. Bertini, I., Fragai, M., Lee, Y. M., Luchinat, C. and Terni, B. (2004) Paramagnetic metal ions in ligand screening: the Co(II) matrix metalloproteinase 12. Angew. Chem. Int. Ed. 43, 2254-2256 27. Lehti, K., Jouko, L., Valtanen, H. and Keski-Oja, J. (1998) Proteolytic processing of membrane-type-1 matrix metalloproteinase is associated with gelatinase A activation at the cell surface. Biochem. J. 334, 345-353 28. Tatti, O., Arjama, M., Ranki, A., Weiss, S. J., Keski-Oja, J. and Lehti, K. (2011) Membrane-type-3 matrix metalloproteinase (MT3-MMP) functions as a matrix composition-dependent effector of melanoma cell invasion. PLoS One. 6, e28325 29. Bergers, G., Brekken, R., McMahon, G., Vu, T. H., Itoh, T., Tamaki, K., Tanzawa, K., Thorpe, P., Itohara, S. and Werb, Z. (2000) Matrix metalloproteinase-9 triggers the angiogenic switch during carcinogenesis. Nat. Cell Biol. 2, 737-744 30. Mott, J. D. and Werb, Z. (2004) Regulation of matrix biology by matrix metalloproteinases. Curr. Opin. Cell Biol. 16, 558-564

170

31. Vaalamo, M., Kariniemi, A.-L., Shapiro, S. D. and Saarialho-Kere, U. (1999) Enhanced expression of human metalloelastase (MMP-12) in cutaneous granulomas and macrophage migration. J. Invest. Dermatol. 112, 499-505 32. Fabre, B., Ramos, A. and de Pascual-Teresa, B. (2014) Targeting matrix metalloproteinases: Exploring the dynamics of the S1′ pocket in the design of selective, small molecule inhibitors: Miniperspective. J. Med. Chem. 57, 10205-10219 33. Agrawal, A., Johnson, S. L., Jacobsen, J. A., Miller, M. T., Chen, L. H., Pellecchia, M. and Cohen, S. M. (2010) Chelator fragment libraries for targeting metalloproteinases. ChemMedChem. 5, 195-199 34. Agrawal, A., Romero‐Perez, D., Jacobsen, J. A., Villarreal, F. J. and Cohen, S. M. (2008) Zinc ‐ binding groups modulate selective inhibition of mmps. ChemMedChem. 3, 812-820 35. Martin, D. P., Blachly, P. G., Marts, A. R., Woodruff, T. M., de Oliveira, C. s. A., McCammon, J. A., Tierney, D. L. and Cohen, S. M. (2014) ‘Unconventional’coordination chemistry by metal chelating fragments in a metalloprotein active site. J. Am. Chem. Soc. 136, 5400-5406

171

4.3 Table and figure

a b Table 4.1: Select inhibitors of MMPs with reported IC50 values

AHA Maltol THM ATM TPMA

O S S S

OH OH OH OH

O O O O H N HO O

MMP1: IC50 = 41,600 M MMP1: IC50 = 4,200 M MMP1: IC50 = 400 M MMP1: IC50 = 41 M MMP3: IC50 = 120 M

MMP2: IC50 = 15,000 M MMP2: IC50 = 2,600 M MMP2: IC50 = 140 M MMP16: IC50 = 43 M

MMP3: IC50 = 25,000 M MMP3: IC50 = 5,700 M MMP3: IC50 = 210 M

MMP16: IC50 = 10,300 M

MMP16: IC50 = 41,000 M MMP16: IC50 = 30 M

Thiohopo Compound 9 94E8 Hopo

N N N OH OH HO HO O S S S N O

MMP1: IC50 = 490M MMP3: IC50 = 137M MMP2: IC50 = 13M MMP1: IC50 = 5,960M

MMP2: IC50 = 100M MMP3: IC50 = 33M MMP2: IC50 = 5,600M

MMP3: IC50 = 30M MMP9: IC50 = 17M MMP3: IC50 = 1600M

94E12 94G5 94H2

172

HO HO HO

S N S N S N F

CF3

MMP2: IC50 = 33M MMP2: IC50 = 1M MMP2: IC50 = >50M

MMP3: IC50 = >50M MMP3: IC50 = 4M MMP3: IC50 = >50M

MMP9: IC50 = 33M MMP9: IC50 = 2M MMP9: IC50 = 30M

1,2-HOPO-2 PY-2 AM-2

O O O

N OH OH H N H H OH N N O O O O O

MMP1: IC50 = >50M MMP1: IC50 = >50M MMP1: IC50 = >50M

MMP2: IC50 = 0.92M MMP2: IC50 = 4.4M MMP2: IC50 = 9.3M

MMP3: IC50 = 0.56M MMP3: IC50 = 0.077M MMP3: IC50 = 0.24M

MMP7: IC50 = >50M MMP7: IC50 = >50M MMP7: IC50 = >50M

MMP8: IC50 = 0.086M MMP8: IC50 = 0.25M MMP8: IC50 = 0.064M

MMP9: IC50 = 27.1M MMP9: IC50 = 32.3M MMP9: IC50 = >50M

MMP12: IC50 = 0.018M MMP12: IC50 = 0.085M MMP12: IC50 = 0.022M

MMP13: IC50 = 4.1M MMP13: IC50 = 6.6M MMP13: IC50 = 20.6M

173

a Structure of ZBGs were obtained from Seth M. Cohen’s reviews on chelator fragements libraries(CFL) [33] and MMP ZBGs libraries [8] b IC50 values of ZBGs were summarized from Seth M. Cohen’s reported work [8, 33-35] on MMPs and F.Meng et al and H. Yang et al’ s unreported works

174

Figure 4.1: Crystal structures of (A) cdMMP1 (yellow, PDB#966C), (B) cdMMP7 (green, PDB#1MMP), (C) cdMMP9(pink, PDB#4H1Q), (D) cdMMP12 (blue,PDB#2POJ), and (E) cdMMP16 (cyan, PDB#1RM8). The S1’ pockets of each MMPs are highlighted with a magenta mesh surface, and surrounding amino acid side chains are labeled with same color. (Inset) Table of amino acid sequences of the S1’ pockets in each MMP [16-20].

175