ELUCIDATING THE ROLE OF INTEGRIN-EXTRACELLULAR MATRIX PROTEIN INTERACTIONS IN REGULATING OSTEOCLAST ACTIVITY
by
Azza Gramoun
A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Graduate Department of the Faculty of Dentistry University of Toronto
© Copyright by Azza Gramoun 2010
Elucidating the Role of Integrin-Extracellular Matrix Protein Interactions in Regulating Osteoclast Activity
Azza Gramoun
Doctor of Philosophy
Faculty of Dentistry University of Toronto
2010
ABSTRACT
Millions of people around the world suffer from the debilitating effects of inflammatory bone diseases characterized by excessive bone loss due to an increase in osteoclast formation and activity. Osteoclasts are multinucleated cells responsible for bone resorption in health and disease. Arthritic joints also have elevated levels of extracellular matrix proteins affecting the disease progression. The interaction between osteoclasts and the external milieu comprised of extracellular matrix proteins through integrins is essential for modulating the formation and activity of osteoclasts. The focus of this thesis was to elucidate how the interaction between the extracellular matrix proteins and osteoclasts regulates osteoclast formation and activity and the role of v3 in this process. In primary rabbit osteoclast cultures, blocking the integrin v3 using Vitaxin, an anti-human v3 antibody, decreased osteoclast resorption by decreasing osteoclast attachment. Vitaxin’s inhibitory effect on osteoclast attachment was enhanced when osteoclasts were pretreated with M-CSF, a growth factor known to induce an activated conformation of the integrin v3. Using the RAW264.7 cell line, the effects of the matrix proteins fibronectin and vitronectin on osteoclast activity were compared to those of osteopontin. Both fibronectin and vitronectin decreased the number of osteoclasts formed
ii compared to osteopontin. Fibronectin’s effect on osteoclastogenesis was through decreasing pre- osteoclast migration and/or fusion but not through inhibiting their recruitment. In contrast, fibronectin induced resorption through increasing resorptive activity per osteoclast in comparison to vitronectin and osteopontin. These stimulatory effects were accompanied by an increase in the pro-inflammatory cytokines nitric oxide and IL-1 Crosstalk between the signalling pathways of nitric oxide and IL-1was suggested by the ability of the nitric oxide inhibitor to decrease the level of IL-1 which occurred exclusively on fibronectin. Osteoclasts on fibronectin also had a compact morphology with the smallest planar area while vitronectin increased the percentage of osteoclast with migratory morphology and osteopontin induced osteoclast spreading. The increase in compact morphology on fibronectin was associated with a decrease in extracellular pH. Low extracellular pH was found to increase the total time osteoclasts spend in a compact phase. These results show that matrix proteins differentially regulate osteoclast formation, activity and morphology.
iii
“If we knew what we were doing, it wouldn't be called research, would it?”
-Albert Einstein
iv ACKNOWLEDGMENTS
Special thanks to Mom and Dad for giving me the chance to embark on this wonderful adventure seven years ago where not only did I get to learn about science and to explore the world, but where I also learnt the most about life and ultimately myself. Throughout the years, their moral and financial support enabled me to continue even through the toughest of times. I would like to specifically thank Dad who by being the great person and scientist he is, has taught me to love science. Through his dedication to and perseverance in research, I have learnt never to give up in the face of obstacles.
I would like to thank Dr. Morris Manolson for being not only a supervisor but also a mentor for me through the course of my PhD in his lab. I will forever remain indebted for the leap of faith he took in accepting me as student without ever having met me in person. I am grateful for all that he has taught me about research and would only hope to be able to put this knowledge to good use.
To my advisory committee members: Dr. Myron Cybulsky, Dr. Johan Heersche and the late Dr. Jaro Sodek, I thank you for all your guidance and constructive criticism that helped shape my research project into something of which I am proud. Special thanks to the late Dr.
Jaro Sodek for the original hypothesis about the effects of extracellular matrix proteins on osteoclasts, and for always dedicating the time to discuss my results with me.
To my wonderful labmates, I thank you for teaching me all that I know technically in cell biology. I appreciate your patience and understanding and I hope I was not at any point a burden on you.
To my beloved friends and family, thank you for believing in me even at the moments when I had ceased to believe in myself. To Suzie Larouche, my roommate, I thank you for all v your help and support. Your usual words of “You will do it Azza” were at times hard to believe yet they eventually came true. To my best friend Rania Nada, you are my rock. I could always find both the encouragement and the nudging I sometimes needed to quit whining and get back on track. The countless conversations, songs of the day, stories and emails we have exchanged over the years are precious and I will always hold them dear to my heart. Special thanks to my trainer Mike Siaflekis. He taught me to endure not only in training, but also in writing.
vi TABLE OF CONTENTS
Abstract ii
Acknowledgments v
Table of Contents vii
Original Contributions by the Author xii
List of Figures and Tables xiv
Abbreviations xvii
1. Introduction 1
1.1 The Structure and Function of Bone 2
1.2 Bone Remodelling 4
1.2.1 Paracrine Regulation of Bone Remodelling by
Pro-resorptive Factors 5
1.3 Osteoclast Differentiation and Its Associated Signalling Pathways 7
1.3.1 M-CSF Induced Signalling Pathways 8
1.3.2 RANKL Induced Signalling Pathways 9
1.3.3 ITAM-associated Receptor Induced Signalling Pathways 12
1.4 Mechanism of Osteoclastic Bone Resorption 13
1.5 Dynamics of Osteoclast Attachment and Morphological Changes 17
1.6 Matrix/Integrin Interactions and Their Effects on Bone Homeostasis 21
1.6.1 Integrin Structure and Function 21
1.6.2 Integrin v3 and Osteoclasts 25
1.6.3 The Molecular Mechanisms Involved in v3 Signalling 29
vii 1.6.4 Extracellular Matrix Proteins 32
1.6.4.1 Osteopontin 32
1.6.4.2 Fibronectin 35
1.6.4.3 Vitronectin 37
1. 7 Rationale and hypothesis 38
2. Effects of Vitaxin®, a Novel Therapeutic in Trial for Metastatic Bone Tumors, on
Osteoclast Functions in vitro 41
2.1 Abstract 42
2.2 Introduction 43
2.3 Materials and Methods 46
2.3.1 Materials 46
2.3.2 Rabbit Osteoclast Isolation 46
2.3.3 Preparation of Devitalized Cortical Bone Slices 47
2.3.4 Attachment Studies 47
2.3.5 Time-Lapse Microscopy 49
2.3.6 Resorption Studies on Bovine Bone Slices 49
2.3.7 Resorption Studies Using the Osteologic Bone Cell Culture System 49
2.3.8 Statistics 50
2.4 Results 51
2.4.1 Vitaxin Inhibits Osteoclast Resorption 51
2.4.2 Vitaxin Decreases the Number of Osteoclasts Attached to Plastic 51
2.4.3 Vitaxin Preferentially Inhibits the Attachment of Small
Osteoclasts (<10 Nuclei) 52
2.4.4 Vitaxin Does not Affect the Resorptive Activity of Attached Osteoclasts 52 viii 2.4.5 Vitaxin Inhibits Attachment but Not Early Stages of
Osteoclast Formation 53
2.4.6 Vitaxin Causes Retraction of Osteoclasts Only in the
Presence of M-CSF 53
2.4.7 Vitaxin's Effect on Attachment is Altered by Factors
Known to Change the Conformation of v3 54
2.5 Discussion 62
3. The Extracellular Matrix Protein Fibronectin Enhances Osteoclast Activity via Nitric
Oxide and Interleukin-1β Mediated Signalling Pathways 66
3.1 Abstract 67
3.2 Introduction 69
3.3 Materials and Methods 72
3.3.1 Materials 72
3.3.2 Immobilizing ECM Proteins on Tissue Culture Plates 73
3.3.3 RAW 264.7-Derived Osteoclast Cultures 73
3.3.4 Splenocyte Derived Osteoclast Cultures 75
3.3.5 Tartrate-Resistant Acid Phosphatase (TRAP) Staining 76
3.3.6 TRAP Activity Assay 76
3.3.7 Cell Viability Assay 76
3.3.8 Secreted TRAP5b Activity Assay 77
3.3.9 Nitrite and Nitrate Measurements 77
3.3.10 Resorption Studies 78
3.3.11 IL-1β ELISA 78
ix 3.3.12 Flow Cytometry Analysis of Integrin Expression 79
3.3.13 Generation of FN Conditional Knockout Mice 79
3.3.13.1 Transgenic Mice 79
3.3.13.2 Histomorphometry 80
3.3.14 Statistics 80
3.4 Results 81
3.4.1 FN Reduces Osteoclast Formation without Affecting RAW
Cell Proliferation or Initial Attachment 81
3.4.2 FN Inhibits Pre-osteoclast Fusion and/or Migration but Not
Pre-osteoclast Recruitment 81
3.4.3 Assessment of Osteoclast Formation in an FN Conditional
Knockout Mouse Model 83
3.4.4 FN Increases Resorption by Increasing Both the Resorptive
Activity per Osteoclast and the Percentage of Resorbing Osteoclasts 84
3.4.5 FN Increases IL-1β in a NO Dependant Manner 85
3.4.6 Blocking v3 and 51 Has Different Effects on Osteoclast Number 85
3.5 Discussion 101
3.6 Conclusions 107
4. Bone Matrix Proteins and Extracellular Acidification; Potential Co-regulators of
Osteoclast Morphology 109
4.1 Abstract 110
4.2 Introduction 111
4.3 Materials and Methods 114
x 4.3.1 Materials 114
4.3.2 RAW 264.7-Derived Osteoclast Cultures 114
4.3.3 Rabbit Osteoclast Isolation 115
4.3.4 Tartrate-Resistant Acid Phosphatase (TRAP) Staining 116
4.3.5 Assessment of Osteoclast Morphological Changes Using
Time-lapse Microscopy 117
4.3.6 Morphometrical Analysis of Changes in Osteoclast’s Morphology 117
4.3.7 Scanning Electron Microscopy 118
4.3.8 Intracellular pH Measurements 118
4.3.9 Statistics 119
4.4 Results 120
4.4.1 Osteoclasts Formed on FN, VN and OPN have Distinct
Morphologies and Planar Area 120
4.4.2 M-CSF Induces Osteoclast Spreading on FN but not on OPN 121
4.4.3 Extracellular pH of Cultures on FN and VN are Lower than
that on OPN 122
4.4.4 Osteoclasts Cycle between Spread and Compact Morphologies
and the Rate of these Changes Depends on Osteoclast Size
and Extracellular pH 123
4.5 Discussion 135
5. Summary and General Discussion 141
6. Future Directions 149
Appendix 154
References 161 xi ORIGINAL CONTRIBUTION BY THE AUTHOR
Publications and submitted manuscripts resulting from this thesis work:
1. Gramoun A*, Shorey S, Bashutski JD, Dixon SJ, Sims SM, Heersche JNM, Manolson MF
(2007) “Effects of Vitaxin®, a novel therapeutic in trial for metastatic bone tumors, on osteoclast functions in vitro”. The Journal of Cellular Biochemistry; 102(2): 341-352.
*Azza Gramoun wrote the manuscript and performed all experiments except: rabbit osteoclast resorption on bone slices (figure 2.3) which was performed by Seema Shorey and time-lapse microscopy of rabbit osteoclasts (figure 2.5) which was performed by Jill Bashutski.
2. Gramoun A*, Azizi N, Sodek J, Heersche JNM, Nakchbandi I, Manolson MF “The extracellular matrix protein fibronectin enhances osteoclast activity via nitric oxide and interleukin-1β mediated signalling pathways”. Submitted to the journal Arthritis Research and
Therapy; February 5 2010, Manuscript ID: 3404146443529669.
*Azza Gramoun wrote the manuscript and performed all experiments except:
Histomorphometrical osteoclast measurements on fibronectin conditional knockout mice (table
3.1) which were performed by Inaam Nakchbandi. Natoosha Azizi assisted in the preliminary experiments conducted to compare the effects of the extracellular matrix proteins.
3. Gramoun A*, Goto T, Nordström T, Rotstein OD, Grinstein S, Heersche JNM, Manolson
MF “Bone matrix proteins and extracellular acidification; potential co-regulators of osteoclast morphology”. Accepted with revisions in the Journal of Cellular Biochemistry; January 19
2010, Manuscript ID JCB-09-0710.
xii *Azza Gramoun wrote the manuscript and performed all experiments except: intracellular pH measurements and changes in rabbit osteoclast morphology under different pH conditions
(figures 4.5 and 4.6) and (tables 4.1, 4.2 and 4.3) which were performed by Tetsuya Goto.
xiii LIST OF FIGURES AND TABLES
Figure 1.1 Osteoclast signalling pathways activated during osteoclastogenesis 11
Figure 1.2 Schematic diagram of a bone-resorbing osteoclast 15
Figure 2.1 Vitaxin decreases osteoclast resorption on osteologic slides 56
Figure 2.2 Vitaxin decreases the attachment of small osteoclasts (OCs)
(<10 nuclei) on plastic 57
Figure 2.3 Vitaxin decreases osteoclast (OC) attachment and resorption on bone but
does not affect resorbed area per osteoclast 58
Figure 2.4 Vitaxin decreases osteoclast (OC) initial attachment but does not affect
osteoclast formation 59
Figure 2.5 Vitaxin decreases osteoclast (OC) planar area only when cultures are
pretreated with M-CSF 60
Figure 2.6 Vitaxin's effect on attachment can be altered by factors affecting
the conformation ofv3 61
Figure 3.1 FN and VN decrease osteoclastogenesis compared to OPN without
affecting initial attachment or proliferation 87
Figure 3.2 VN delays early pre-osteoclast formation while both VN and FN
decrease osteoclast multinucleation 89
Figure 3.3 Matrix proteins have similar effects on the formation of osteoclasts
derived from either splenocytes or RAW cells 90
Figure 3.4 Soluble FN decreases migration/fusion of pre-osteoclasts formed on
physically adsorbed OPN 91
xiv Figure 3.5 FN increases resorptive parameters and NO production 92
Figure 3.6 Osteoclasts on FN coated osteologic discs have more sealing zones 95
Figure 3.7 IL-1β and NO production is increased on FN. Inhibition of IL-1β using
the NO synthase inhibitor L-NMMA suggests that NO is upstream of IL-1β 96
Figure 3.8 Exclusive blockade of v3 in osteoclasts differentiated on FN increases
osteoclast number 97
Figure 3.9 Blocking 5but not v3or CD44, decreases osteoclast number on FN
and its expression is highest on FN 98
Figure 4.1 Osteoclasts differentiated on FN, VN and OPN have different morphologies 126
Figure 4.2 Osteoclast morphology and planar area are modulated by the ECM
proteins FN, VN and OPN 127
Figure 4.3 M-CSF treatment causes osteoclast spreading on FN while
osteoclasts on OPN fail to spread 129
Figure 4.4 The effect of ECM proteins on extracellular pH of culture media 130
Figure 4.5 Phase-contrast and scanning electron micrographs demonstrate the
morphological cycling of an osteoclast at pH 7.0 131
Figure 4.6 Time course of morphological cycling of small and large osteoclasts
at pH 7.5 and pH 7.0 132
Figure 6.1 Osteoclasts and pre-osteoclasts degrade fluorescently labelled FN coating 151
Figure 6.2 Osteoclasts on a high density RGD coated nanopattern
exhibit multiple podosome rings 152
Figure 6.3 Osteoclasts on homogenous RGD coated surfaces exhibit normal
podosome arrangement 153
xv Figure A.1 Osteoclast formation is enhanced by sFN and suppressed by pFN 157
Figure A.2 Pre-osteoclast velocity and polarity are increased on sFN compared
to pFN and cFN 159
Figure A.3 Osteoclasts formed on sFN exhibit an atypical “sealing zone” like
attachment structure while those on cFN contain a typical podosome ring 160
Table 3.1 Histomorphometric osteoclast parameters in the FN conditional knockout
(cKO) Mx mouse line 100
Table 4.1 The Duration of the Compact and Spread Phases of Osteoclasts
at pH 7.0 and 7.5 133
Table 4.3 The Effects of Bafilomycin A1(BFA), Acetazolamide (AZ), DIDS and
Amiloride (Am) on the Duration of Compact and Spread
Phases of Osteoclasts 134
xvi ABBREVIATIONS
A domain von Willebrand factor A domain
ADMIDAS Adjacent to MIDAS
Akt RAC-alpha serine/threonine-protein kinase
Am Amiloride
-MEM -Minimum essential medium
ANOVA One way analysis of variance
AP-1 Activator protein-1
Arp 2/3 Actin related protein 2/3
AZ Acetazolamide
BCECF, AM 2',7'-bis- (2-Carboxyethyl)-5 (6)- carboxyfluorescein,
acetoxymethyl ester
Bcl-2 B-cell lymphoma 2
BFA Bafilomycin A1
BL Basolateral domain
BLNK B-cell linker protein
BMM Bone marrow macrophages
BMU Bone metabolic unit
BS Bone surface
BSA Bovine serum albumin
BSP Bone sialoprotein
CAII Carbonic anhydrase II
xvii CaMK Calcuim/calmodulin-dependent protein kinase c-Cbl Casitas B-lineage lymphoma cFN Cellular fibronectin cKO Knockout
ClC-7 Chloride channel-7
CREB Cyclic-AMP-responsive-element binding protein
CT Control
DAB 3,3'-diaminobenzidine tetrahydrochloride
DAP-12 DNAX activation protein-12
DAPI 4'-6-Diamidino-2-phenylindole
DC-STAMP Dendritic cell specific transmembrane protein
DIDS 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid
DMEM Dulbecco’s modified Eagle’s medium
ECM Extracellular matrix
EIIIA Extra type III repeat A
EIIIB Extra type III repeat B
EM Electron microscopy
ERK Extracellular regulated kinase
FBS Fetal bovine serum
FcR Fc receptor
FN Fibronectin
FRET Fluorescence resonance energy transfer
FSD Functional secretory domain
xviii GLA -carboxy glutamic acid
GRB2 Growth-factor-receptor-bound protein 2
HA hydroxyapatite
HEPES 4-[2-hydroxyethyl] piperazine- N'-[ethanesulfonic acid]
I-domain Insert domain
I-EGF Integrin epidermal growth factor
IIICS V connecting segment
IL-1 Interleukin-1
ITAM Immunoreceptor tyrosine-based activation motif
LIBS Ligand induced binding site
LIMBS Ligand induced metal binding site
L-NMMA L-NG-monomethyl arginine
MAPK Mitogen activated protein kinase
M-CSF Macrophage colony stimulating factor
MIDAS Metal ion dependent adhesion site
MITF Microphthalmia associated transcriptional factor
MMP-9 Matrix metaloprotiease-9
NFATc1 Nuclear factor of activated T cells c1
NFB Nuclear factor B
NO Nitric oxide
O/N Overnight
OC Osteoclast
Oc. N Osteoclast number
xix Oc. S Osteoclast surface
ODF Osteoclast differentiation factor
OPG Osteoprotegrin
OPGL OPG ligand
OPN Osteopontin
OSCAR Osteoclast associated receptor
PBS Phosphate buffered saline pFN Plasma fibronectin
PGE2 Prostaglandin E2 pHi Intracellular pH
PI3K Phosphoinositide 3-Kinase
PIR-A Paired immunoglobulin-like receptor A
PLC- Phospholipase C
PSI Plexin/semaphorin/integrin domain
PTH Parathyroid hormone
PTM Posttransitional modification
PyK2 Proline rich tyrosine kinase
RA Rheumatoid arthritis
RANKL Receptor activator of nuclear factor B ligand
RAW cells RAW264.7 cells
RGD Arg-Gly-Asp
RGDS Arg-Gly-Asp-Ser
RL Ruffled border
xx RT Room temperature
SD Standred deviation
SDGRG Ser- Asp-Gly- Arg-Gly
SEM Standard error of mean sFN Superfibronectin
SH2 domain Src homology 2 domain
SIBLING Small integrin-binding ligands with N-linked glycosylation
SIRPβ Signal regulatory protein β
SL Sealing zone
SLP-76 SH2 domain containing leukocyte protein of 76kDa
Syk Spleen tyrosine kinase
TCP Tissue culture polystyrene
TNF Tumour necrosis factor
TRAF TNF receptor associated factor
TRANCE TNF-related activation-induced ligand
TRAP Tartrate resistant acid phosphatase
TREM-2 Triggering receptor expressed on myeloid cells 2
T-test Student T-test
V-ATPase Vacuolar ATPase
Vitamin D3 1, 25-dihydroxyvitamin D3
VN Vitronectin
Wasp Wiskott-Aldrich syndrome protein
WIP Wasp interacting protein
xxi 1. INTRODUCTION
Inflammatory bone diseases such as rheumatoid arthritis (RA) are prevalent metabolic conditions characterized by progressive bone loss in the affected joints (1). Bone destruction in arthritic joints occurs due to the uncoupling of the two events comprising the ongoing bone remodelling process; bone formation by osteoblasts and bone resorption by osteoclasts. Both processes are synchronized by inter and intracellular signalling events involving calcitropic hormones, cytokines, growth factors and attachment receptors binding to the extracellular matrix (ECM). Due to the abundance of pro-inflammatory cytokines prevalent in the micro- environment in affected joints, the unbalance of bone remodelling is specifically due to an increase in the formation and activity of osteoclasts (2, 3). Despite the recent advances in arthritis treatments controlling different symptoms of the disease such as pain and inflammation, bone loss resulting in permanent joint damage remains a more difficult problem to resolve.
The interaction between osteoclasts and the ECM is essential not only for their attachment and survival but also for their function and it requires the integrin v3 (4). Thus, functionally blocking the interaction between the integrin v3 and its ligands is one of the methods utilized to prevent bone loss (reviewed by (5)). Although ECM proteins play a role in wound healing and tissue repair, their elevation in arthritis is associated with pro-inflammatory cytokine-like properties that amplify joint damage (6-8). In the first part of my thesis I focused on studying the effects of an v3 blocking antibody known as Vitaxin and found that it inhibited osteoclast resorption through impairing their attachment. I was also able to show that integrin activation increases Vitaxin’s inhibitory effects on osteoclast attachment (chapter 2).
Fibronectin (FN) and vitronectin (VN) are two of the bone matrix proteins that have been shown
1 to be elevated in arthritis, yet their effects on osteoclast formation and function were not investigated. I decided to ask if FN and VN differentially regulate osteoclast function and I hypothesized that both proteins promote osteoclast formation and function similar to osteopontin (OPN). I was able to show that both FN and VN decreased osteoclast formation compared to OPN; however, FN stimulated osteoclast resorption and cytokine production
(chapter 3 and 4). In the following sections, I will review the topics relevant to bone, osteoclast function and its interaction with the ECM proteins through integrins.
1.1 The Structure and Function of Bone
Bone is a multifunctional tissue that provides the body with the rigid scaffold essential for shape and support against gravitational forces. As an integral element of the skeletal framework, bones protect vital organs, facilitate motility and serve as the major store for calcium and other minerals; thus contributing to homeostasis. It is estimated that approximately
99% of the body’s calcium content is stored in the skeleton. Through its porous structure, bone simultaneously achieves the lightness needed for locomotion while providing a niche for bone marrow cells and haematopoiesis. As a mineralized connective tissue, bone has a highly complex and intricate structure. It is composed of collagenous and non-collegenous protein meshwork embedded in a hydroxyapatite (HA) matrix. Bone possesses superior elastic properties that are necessary to sustain the constant stresses, allowing it to reversibly deform without reaching its fracture point (9, 10). It is both the structure and composition of this mineralized extracellular matrix that endows the skeletal tissues with the needed strength and rigidity without compromising its weight and flexibility.
Based on the mechanism of bone formation during development, bones of the skeleton can be classified into two major groups; long bones (e.g. tibia and femur) and flat bones (e.g. 2 skull and vertebrae). Endocondoral ossification is the mechanism responsible for long bone formation and it involves the replacement of a cartilage template by mineralized tissues. In contrast, flat bones formation occurs directly by mesenchymal cell condensation at ossification centres; a process known as intramembranous ossification.
The inorganic content of bone, mainly in the form of HA crystals, constitutes up to 50% of the skeleton’s dry weight and gives it the required stiffness. Meanwhile, the resilience and toughness of the skeletal tissues is conferred by collagen type I which accounts for almost 90% of the organic content of bone. In addition to their physical properties, collagen fibres form an interlaced three dimensional meshwork into which HA crystals are deposited, thus protecting
HA, which is the brittle part of the bone matrix.
The remaining 10% of bone’s organic matrix takes the form of a large number of non- collagenous proteins composed of four major classes: glycoproteins (~7%), proteoglycans
(~1%), small integrin-binding ligands with N-linked glycosylation (SIBLING) (0.5%) and γ- carboxy glutamic acid (GLA)-containing proteins. Some of the abundant bone glycoproteins are osteonectin, tetranectin and the Arg, Gly, Asp (RGD) containing glycoproteins FN and VN.
Decorin and biglycan are chondroitin sulphate-containing proteins representing the proteoglycans family in bone tissues. The SIBLING proteins are a large family of RGD containing glycoproteins. Of those, OPN and bone sialoprotein (BSP) are the most relevant bone sialoproteins. Osteocalcin is the most influential type in the GLA-containing protein category.
Because of HA’s ability to physically adsorb serum proteins, a several fold enrichment of albumin and α2-HS-glycoprotein is seen in bone. Finally, other classes of proteins including immunoglobulins, growth factors, cytokines and chemokines synthesized extrinsically and locally can also be found bound to HA. In terms of function, non-collagenous proteins contribute to bone quality mechanically, physically and metabolically. Mechanically, non- 3 collagenous proteins, through functioning as glue-like molecules, protect bone during loading by absorbing and dissipating energy by breaking intrahelical collagen bonds and thus allowing a microscopic increase in fibre length (11, 12). Physically, the majority of the bone non- collagenous family members play a central role in apatite crystal nucleation. Through binding to collagen, these matrix proteins dictate the shape, size and orientation of HA crystals and consequently fibril formation. Therefore, it was not surprising to find that transgenic mice deficient in many of these non-collagenous proteins exhibited bone phenotypes where the quality and quantity of bone were compromised (13-18). To further enhance their functional adaptability, bone matrix proteins and specifically collagen are differentially expressed between long bones and flat bones. The expression patterns in both types of bone yield specific biomechanical properties and render them more suited to their stress profiles (19). In addition to the above mentioned physiomechanical properties, non-collagenous matrix proteins, and specifically those belonging to the RGD-containing glycoprotein family, play an integral role in bone metabolism. The RGD-containing glycoproteins regulate bone homeostasis not only through providing the basis for many attachment related cell functions and mediating integrin signalling transduction, but also through to their more recently discovered cytokine-like properties through other receptors.
1.2 Bone Remodelling
Despite its metabolically static appearance, it is estimated that approximately 10% of the total bone mass in an adult human is replaced per year through a physiological process known as bone remodelling. Bone modelling and remodelling are closely related, yet different processes.
While bone modelling is the process by which bone is formed during growth leading to an increase in the size and shape of bone, bone remodelling is the bone’s unique ability to self 4 repair microdamage. Throughout life, bone is constantly deformed under weight-bearing stresses, causing the accumulation of microfractures and the deterioration of its biomechanical properties, ultimately increasing its risk of fracture. In addition to being the mechanism for microdamage elimination, bone remodelling enables bone to fulfill its functional demands by adapting to the various dynamically changing biomechanical and physiological stimuli. As opposed to bone modelling which is specifically linked to bone formation, the bone remodelling cycle is composed of alternating bone resorption and bone formation cycles. The osteoblast, a cell of mesenchymal origin, is responsible for bone formation, whereas bone resorption is the exclusive function of the osteoclast. Together, the osteoblast and osteoclast form the bone metabolic unit (BMU) (as reviewed by (20)). The osteocyte is the third cell type involved in bone homeostasis. Events such as the microcrack formation, estrogen deficiency and corticosteroid therapy signal osteocytes to undergo apoptosis (21, 22). Osteocyte apoptosis is thought to be the first event triggering osteoclast resorption and consequently bone formation
(23, 24). In a remodelling cycle, the alternating bone resorption and formation cycles are sequential; yet the molecular and cellular processes regulating those two events are not.
1.2.1 Paracrine Regulation of Bone Remodelling by Pro-resorptive Factors
Bone remodelling is a tightly governed process affected by systemic and local factors as well as mechanical loading. The uncoupling of the anabolic and catabolic activities of bone cells as a result of an increase in osteoclast number and/or activity is associated with bone loss and is the prevalent cause for many bone disorders. The key players orchestrating osteoclast formation and activation are the macrophage colony stimulating factor (M-CSF) and the tumour necrosis factor (TNF) family member receptor activator of nuclear factor B (NFB) ligand (RANKL)
5 (25-28). In fact, it is well established that the ratio between RANKL and its decoy receptor osteoprotegrin (OPG) determines the rate of bone turnover and its disruption is an indicator of enhanced osteoclast resorption and bone erosion in arthritis (29-31). Systemic and local osteoclastogenic factors modulate osteoclast function by acting upstream of RANKL and M-
CSF by regulating their expression in osteoblasts, stromal cells and activated T-cells.
Parathyroid hormone (PTH), 1,25-dihydroxyvitamin D3 (vitamin D3), prostaglandin E2 (PGE2) and corticosteroids are some of the important systemic factors (32-34). Additionally, the production of the osteolytic cytokines interleukin-1 (IL-1), IL-6, IL-11, IL-12, IL-17 and IL-23 and growth factors TNF-α, granulocyte-macrophage colony stimulating factor and transforming growth factor in the local bone environment has a stimulatory effect on osteoclasts (35). The mechanisms by which these pro-resorptive factors induce osteoclast activation are complicated and involve both direct and indirect interactions with osteoclasts. The effects of glucocorticoids on osteoclast formation and resorption demonstrate a simultaneous direct and indirect modulation by a pro-resorptive hormone. Glucocorticoids activate osteoclasts by increasing M-
CSF and RANKL and decreasing OPG expression via interacting with osteoblasts (30, 36).
Concurrently, they act directly on osteoclasts attenuating apoptotic signal and promoting their survival (37). In contrast, PTH’s mode of action on osteoclasts is only indirect and is mediated by an increase in RANKL/OPG ratio (31, 38, 39). Similar to glucocorticoids, IL-1 and TNF-α act indirectly through osteoblasts upregulating RANKL (40), but they also synergistically promote osteoclast differentiation and resorption together with RANKL (41, 42). Interestingly, while osteoclast activation by osteoclastogenic factors can be direct and/or indirect, osteoclast formation can exclusively occur in the presence of RANKL and M-CSF.
6 1.3 Osteoclast Differentiation and Its Associated Signalling Pathways
Osteoclasts are multinucleated terminally differentiated cells with the unique ability to dissolve mineralized tissues. They are derived from haematopoietic myeloid precursors and thus they share many phenotypical characteristics with monocytes and macrophages (43).
Osteoclastogenesis is a multistep process in which osteoclast precursors are recruited before they fuse to form the mature osteoclast. Osteoclast differentiation is initiated by the contact between osteoclastogenesis supporting cells (bone marrow stromal cells, osteoblasts and synovial fibroblasts) (44) and osteoclast precursors (CFU-S) inducing their differentiation into pre-osteoclasts (CFU-GM) (as reviewed by (45)). The binding of M-CSF and RANKL expressed by any of these cells to their respective receptors RANK and c-Fms on myeloid precursors is indispensible to the induction of osteoclast differentiation and resorption (46, 47).
Emerging evidence also identified immunoreceptors and other immunoreceptor tyrosine- based activation motif (ITAM) associated receptors as the co-stimulatory partners of M-CSF and
RANKL (48). Subsequent to receptor ligation, the RANKL primed pre-osteoclasts fuse and give rise to a multinucleated cell capable of resorbing bone. The phenotypic markers associated with osteoclast maturation are tartrate resistant acid phosphatase (TRAP), the calcitonin receptor, the integrin subunit 3, the chloride channel ClC-7, the cystine protease cathepsin K, matrix metaloprotiease-9 (MMP-9) and the osteoclast associated receptor (OSCAR) (49-51).
Cell fusion is an integral part of osteoclast formation as multinucleation is a requirement for efficient bone resorption. Several surface molecules have been implicated in osteoclast fusion yet the exact mechanism of that process is not completely understood. Among these molecules that are part of the fusion machinery are the macrophage fusion molecule also known as signal-regulatory protein-α and its ligand CD47 and the transmembrane glycoproteins CD44
7 and CD200 (52, 53). The dendritic cell specific transmembrane protein (DC-STAMP) and the
+ d2 isoform of vacuolar (H ) ATPase (V-ATPase) V0 domain (Atp6v0d2) play an eminent role in facilitating fusion as demonstrated by transgenic mice studies. Mice deficient in DC-STAMP exhibited a total inhibition of pre-osteoclast fusion accompanied by a reduction in bone resorption which resulted in an osteopetrotic phenotype, whereas osteoclast differentiation remained unaffected (54). Similarly, deletion of the Atp6v0d2 led to the abrogation of osteoclast fusion, attenuated their resorptive capacity and rendered them osteopetrotic (55). Both molecules are significantly elevated during osteoclast differentiation and are under the transcriptional regulation of the nuclear factor of activated T cells c1 (NFATc1), the master transcriptional regulator during osteoclastogenesis (56).
1.3.1 M-CSF Induced Signalling Pathways
M-CSF is a membrane bound osteoclastogenic cytokine required during osteoclast differentiation and multinucleation (25). This growth factor is expressed by many cells including endothelial cells (57), however during osteoclastogenesis, cells of mesenchymal/stromal lineage are its major source (44). In addition to its physiological role, M-
CSF is upregulated by TNF- during inflammatory bone loss and by tumour cells in both soluble and bound forms (58-60). M-CSF plays a critical role in the initial steps of pre- osteoclast differentiation from haematopoietic cells and subsequently promotes their proliferation and survival. Its role in osteoclast formation was demonstrated by the osteopetrotic phenotype of the op/op mouse. The op/op mouse harbours a mutation in the Csf1 gene which ablates M-CSF production (26). In addition to its osteopetrotic phenotype, calvarial osteoclasts from the op/op mouse could not support osteoclast formation in vitro (61). The M-CSF receptor,
8 c-Fms is member of the receptor tyrosine kinase super family present on osteoclast precursor cells and its expression is regulated by transcriptional factor PU.1 (62). The M-CSF/c-Fms interaction leads to the autophosphorylation of the receptor and transmits critical downstream signals. The M-CSF-induced activation of growth-factor-receptor-bound protein 2(GRB2) is responsible for the activation of the extracellular regulated kinase (ERK) (63), whereas phosphoinositide 3 kinase (PI3K) acts upstream of Akt (RAC-alpha serine/threonine-protein kinase) (64), both events enhance osteoclast precursors proliferation and survival respectively.
Through activating the microphthalmia-associated transcription factor (MITF), M-CSF further promotes monocyte/macrophage osteoclast precursors survival via stimulating B-cell lymphoma
2 (Bcl-2) (65, 66). Most importantly, M-CSF signalling induces RANK expression by pre- osteoclasts, an event essential for their proper maturation when primed by RANKL (67). In collaboration with the integrin v3, M-CSF also modulates other osteoclast functions and facilitates resorption; such as cytoskeleton reorganization and migration. These effects will be discussed in section 1.6.3.
1.3.2 RANKL Induced Signalling Pathways
The indispensible differentiation factor RANKL is a type II membrane protein and a member of the TNF family. RANKL is also known as TNF-related activation-induced ligand
(TRANCE), OPG ligand (OPGL) and osteoclast differentiation factor (ODF). In addition to being the key regulator of osteoclast development and activation, RANKL is responsible for T- cell mediated dendritic cell activation, lymph node organgenesis and lactating mammary development during pregnancy (68). It was the observation that osteoclast formation in vitro could only be achieved when hematopoietic cells were in contact with stromal/osteoblast lineage
9 cells by Suda’s group that lead to the discovery of this critical factor a year after its soluble inhibitor the decoy receptor OPG was discovered (43, 44). As previously mentioned, multiple hormones and osteolytic cytokines upregulate RANKL’s expression such as vitamin D3, PGE2,
PTH, glucocorticoids, IL-1, IL-6, IL-11 and TNF- in a paracrine manner (34). RANKL deficient mice display severe osteopetrosis accompanied by arrested growth and lack of tooth eruption. These mice are protected from arthritis associated bone loss and can be rescued by recombinant RANKL injection (69, 70). All RANKL induced signalling pathways involve the recruitment of one common adaptor molecule. TNF receptor associated factor (TRAF) 6 (71).
The interaction between RANKL and its receptor RANK causes its trimerization along with its adaptor TRAF6. This is followed by downstream signalling cascades that activate the transcription factor NFB, mitogen-activated protein kinses (MAPKs); p38 and the Jun N- terminal kinase. This results in the activation and association of the two components of the transcription factor activator protein-1 (AP-1); c-Jun and c-Fos (72, 73). These transcription factors collaborate to induce the activation and nuclear translocation of the master switch of osteoclastogenesis NFATc1 which inturn promotes its own transcription, autoamplifying its own expression and results in the activation of a group of osteoclast-related genes and ultimately osteoclast differentiation (74, 75). In parallel, Ca2+ signalling downstream of TRAF6 indirectly causes the phosphorylation of phospholipase C (PLC and is responsible for the c-Fos activation and its recruitment to the AP-1 complex (48, 76). NFATc1 is responsible for the transcriptional regulation of the following osteoclast-specific genes: TRAP, calcitonin receptor,
3-integrin subunit, cathepsin K,OSCAR, DC-STAMP and Atp6v0d2 (49, 51, 56). Together with NFATc1, the transcriptional regulation of these genes is also achieved in collaboration with the transcriptional factors PU.1, MITF and AP-1 (77) as well as the cyclic-AMP-responsive-
10
Figure 1.1 Osteoclast signalling pathways activated during osteoclastogenesis.
Osteoclastogenesis requires the activation of three main signalling pathways downstream of
RANK, c-Fms and ITAM associated immunoreceptors. For more details please refer to section
1.3. 11 element binding protein (CREB) activated by calcium/calmodulin-dependent protein kinase
(CaMK) IV(78).
1.3.3 ITAM-associated Receptor Induced Signalling Pathways
ITAM-coupled receptors are responsible for co-stimulatory signals participating with
RANKL in initiating Ca2+ fluxes essential for osteoclast differentiation (48). ITAM is a conserved motif present in the cytoplasmic domain of certain transmembrane adaptor molecules that can be found coupled with immunoreceptors. These receptors and their associated adaptor proteins are commonly found in cells of hematopoietic lineage including osteoclasts. In myeloid cells, there are at least 20 ITAM-associated immunoreceptors such as OSCAR, triggering receptor expressed on myeloid cells 2 (TREM-2), paired immunoglobulin-like receptor A (PIR-
A) and signal regulatory protein (SIRP). In contrast, there are only two ITAM-containing adaptor proteins expressed by these cells; the DNAX activation protein-12 (DAP12) and Fc receptor (FcR (79). More recently, the tyrosine receptor c-Fms and integrins 3 and 2 were also found to associate with DAP12 and FcR indicating that they play a role in co-stimulatory osteoclast signalling (80, 81). Despite the extensive studies that have delineated the ITAM signalling cascades, the ITAM-associated receptor ligands are yet to be identified, with the exception of FcRs which are known to bind immunoglobulins. The ITAM-associated receptor/ligand binding phosphorylates the ITAM motif of their coupled adaptor molecules
DAP12 and FcR which activates and sequesters the spleen tyrosine kinase (SyK) (76). Syk activation by ITAM adaptor proteins occurs simultaneously through RANKL signalling pathway in collaboration with the immunoreceptors-mediated signalling. Ca2+ dependent signalling activation through the RANKL and ITAM pathways occurs through PLC. PLC
12 activation is a complicated process that requires the formation of a signalling complex composed of the Tec and BtK tyrosine kinases in combination with the adaptor proteins B-cell linker protein (BLNK) and SH2 domain containing leukocyte protein of 76kDa (SLP-76) (82-
85). While BLNK and SLP-76 act downstream of the ITAM-associated receptors,
RANKL/RANK ligation is responsible for the activation of Tec and BtK and is dependent on the c-Src tyrosine kinase (86). This represents another point where the two signalling pathways converge. The activation PLC triggers Ca2+ fluxes that act through the calcium dependent phosphatase calcineurin inducing the dephosphorylation and nuclear translocation of NFATc1
(75). Also, calcium oscillations indirectly mediate NFATc1 in a CaMK/c-Fos dependant mechanism.
1.4 Mechanism of Osteoclastic Bone Resorption
Bone resorption is dependent on the osteoclast’s ability to polarize thereby creating three functional membrane domains (figure 1.2). The domain with the most critical function to bone degradation is the ruffled border (RL); which is a highly convoluted V-ATPase rich membrane found adjacent to the bone surface and directly above the resorption lacuna (87). Opposite to the ruffled border is the functional secretory domain (FSD), where transcytosis of bone degradation products occurs (88, 89). The basolateral domain (BL) is the third osteoclast functional domain and is located lateral to the functional secretory domain. Bone resorption requires tight osteoclast attachment and dynamic cytoskeletal reorganization generating the compact or polarized osteoclast morphology and the three essential functional domains. Although the factors triggering these events are not clearly understood, matrix recognition by the integrin
v3 was shown to play a central role in this process (4, 90). In a resorbing osteoclast, tight
13 attachment is mediated by a complex structure known as the sealing zone (SL) or actin ring
(91). The sealing zone not only facilitates attachment but also creates a tight seal, isolating the bone surface to be resorbed and thus creating the proper environment for mineral and organic components of bone to be removed efficiently (92). Once this environment has been created, the osteoclast’s resorption machinery consisting of H+ pumping V-ATPases and proteolytic enzymes containing lysosomes are sequestered to the ruffled border where the H+ and the enzymes are released creating a resorption lacuna within the sealing zone (93, 94). The cytoplasmic carbonic anhydrase II (CAII) enzyme generates the protons transported by V-
ATPase at the ruffled border (95, 96). During the process of active proton transport by V-
- - + + ATPases, cellular pH homeostasis is maintained through the Cl /HCO3 exchanger and Na /H antiporter while the Cl- channel ClC-7 works in parallel with V-ATPases retaining the cell’s electroneutrality (97-101). The critical role of these enzymes and channels in osteoclast activity is clearly demonstrated in diseases characterized by disruption of their activity. Several human mutations have been reported in V-ATPase, ClC-7 and CA II, that result in a wide range of osteopetrotic phenotypes (102-104). In addition to the hormonal and cytokine regulation of osteoclast activity discussed before, other factors such as osteoclast size (defined by their number of nuclei) and extracellular acidosis were found to promote resorption (105-108).
Patients with Paget’s disease and end-stage renal acidosis have hyperactive large multinucleated osteoclasts (109, 110). Following the demineralization of bone surface, matrix degradation occurs mainly through the proteolytic activity of the cystine protease cathepsin K. Cathepsin K’s optimal acidic pH and its targeted transport in V-ATPase containing vesicles to the ruffled border are evidence that it functions as the major collagenolytic enzyme (94, 111).
14
Figure 1.2 Schematic diagram of a bone-resorbing osteoclast. Actively resorbing osteoclasts are highly polarized cells with three domains; the ruffled border is the most important of all three and is formed by fusion of exocytotic vesicles containing V-ATPases, cathepsin K and CLC-7.
Targeted vesicular trafficking triggered by matrix/ integrin interaction induces the association of these vesicles with microtubules and their subsequent delivery to the ruffled border. Bone degradation products are transcytosed across the osteoclast to be released through the functional secretory domain. CAII, carbonic anhydrase II; RL, ruffled border; BL, basolateral membrane;
FSD, functional secretory domain; SL, sealing zone.
15 This role is further confirmed by bone scelerosis and pycnodysostosis associated with human mutations causing cathepsin K deficiency (112). While matrix metalloproteases (MMPs) such as
MMP-9 were implicated in bone matrix degradation, this role is not supported by their neutral optimal pH and transient osteopetrosis exhibited by MMP-9 knockout mice (113, 114). MMP-9 was, however, shown to participate in the initiation of bone demineralization via removing the collagenous layer off the bone surface as well as cleaning of lacunae initiating bone formation
(115). It has also been suggested recently that MMP-9’s role varies depending on the origin of osteoclast population (116, 117). Another enzyme secreted by osteoclasts and correlating with their resorptive activity is TRAP. Due to the high levels of the enzyme in osteoclasts, TRAP has been used as a histochemical marker identifying the cells in vivo and in vitro (118). Two TRAP isoforms are present in serum, TRAP5a and TRAP5b. While TRAP5a is the isoform produced by macrophages and dendritic cells, TRAP5b is an osteoclast specific isoform that is proteolytically cleaved and activated by cathepsin K and needs pH 5.8 for optimal activity (119,
120). Although TRAP5b is elevated in patients experiencing excessive bone loss, the exact role of TRAP in bone resorption is not clear (121, 122). The targeted deletion of the TRAP gene in bone has demonstrated that TRAP has a role in bone development and bone resorption. TRAP-/- mice had significant bone phenotypes. These mice had shorter, broader flat and long bones with thicker cortical bone with disorganized growth plate (123). These effects resulted in age progressive osteopetrosis due to a defect in collagen metabolism involving both synthesis and cleavage of collagen (123, 124). The diminished resorptive activity of the TRAP-/- osteoclasts is mainly due to defects in the structure of the ruffled border and intracellular trafficking (125). It is speculated that TRAP’s effects on osteoclast resorption are related to its role in transcytosis and OPN processing (126). After being regarded for a long time as a resorption marker (127,
128), recent evidence indicates that TRAP5b is an osteoclast formation marker rather than a 16 resorption marker and its elevation is associated with an increase in osteoclast number seen in many bone loss diseases (129-131).
1.5 Dynamics of Osteoclast Attachment and Morphological Changes
In osteoclasts, similar to other cells of hematopoietic origin, podosomes are the basic unit of attachment (132, 133). This adhesion structure can also be seen in certain human leukemia cells and other v-Src transformed cells (134, 135). Podosomes partake in cell attachment, migration, matrix degradation and invasion (136-138). Ultrastructurally, a podosome is composed of columnar actin filaments surrounding a small tubular invagination of the plasma membrane perpendicular to the substrate’s surface (139). Numerous focal adhesion and actin polymerization regulatory proteins including Wiskott–Aldrich syndrome protein
(Wasp), actin related protein 2/3 (Arp 2/3), vinculin, paxillin and talin are present within the structure of the podosome (140-143). On a molecular level, podosomes share many of these proteins with focal adhesions, the podosome’s counterpart found in fibroblast-like cells (144). In addition to those common molecules, podosomes contain certain unique actin-binding proteins such as gelsolin, dynamin and cortactin (145, 146). Together these proteins provide a highly dynamic actin cytoskeletal assembly essential for polarization and migration. The arrangement of podosomes into highly organized adhesion complexes depends on two factors: the degree of osteoclast differentiation and matrix composition (147, 148). During early osteoclast differentiation on glass or tissue culture polystyrene (TCP), podosomes are arranged in clusters that are later organized into multiple short lived podosome rings. In mature osteoclasts, a stable peripheral podosome belt is found with an average thickness of 2 μm and inter-podosome distance of 500 ± 140 nm (147, 149). When the podosome belt was examined carefully using 3-
D confocal microscopy and environmental scanning electron microscopy, the F-actin dense 17 podosome core was located inside a less dense actin cloud made of polymerized actin interconnecting branches (147, 149, 150). The centrifugal patterning and growth of podosome clusters into rings which then fuse to form the peripheral podosome belt is controlled by the polymerization of the acetylated microtubules. This was elegantly shown by the nocodazole- induced depolymerisation of microtubules which was followed by disorganization of the podosome belt (151-153). Based on these observations, two distinct actin subdomains were defined in a podosome belt; the podosome or actin core and actin cloud (154, 155). Using the
Wasp interacting protein (WIP) -/- and Src -/- osteoclasts, the existence of these two separate domains was confirmed and the distribution of multiple podosome associated proteins and adhesion receptors was determined. While the podosome core is absent in WIP -/-osteoclasts, no actin cloud is seen in the Src -/- cells (154, 156, 157). It is worth noting that in Src -/- osteoclasts podosome superstructures can be rescued by kinase-dead c-Src expression, indicating that in podosomes c-Src functions as an adaptor molecule and not as a kinase (153). After further examination of the podosome belt, a molecular model was proposed where integrin v3 is central for organizing the actin cloud and for linking the actin cytoskeleton to the extracellular matrix through the adaptor proteins paxillin, vinculin and talin (157-159). To achieve this function, the integrin v3 activates and complexes with c-Src, the proline rich tyrosine kinase
(PyK2) and casitas B-lineage lymphoma (c-Cbl) (146, 160). In the podosome core, CD44 which is a cell surface single pass transmembrane proteoglycan that binds hyaluronan and OPN, plays the main role in actin nucleation. CD44 fulfills this function by directly binding to and activating WASP as well as other actin regulating proteins such as Arp2/3 and cortactin (154,
161). Despite their unique molecular makeup and actin organization, the two actin subdomains play an additive role in osteoclast attachment as indicated by the ability of both WIP -/- and Src
18 -/- osteoclasts to attach (154, 157).
In contrast to the podosome belt seen on glass, osteoclasts on bone exhibit an adhesion superstructure known as the sealing zone. Many studies have demonstrated that sealing zone formation is triggered only by the mineral content of the substrate onto which the osteoclasts are attached and not affected by a substrate’s matrix protein content (147). Using GFP tagged actin expressing osteoclasts and immunofluorescent microscopy, the sealing zone was shown to be composed of a thick continuous central actin band surrounded by an inner and outer vinculin rings (91, 147, 150). Luxenburg et al. have found that the intensity of staining of actin, vinculin and paxillin in the sealing zones is significantly higher than that measured in individual podosomes seen in polarized osteoclasts or in a podosome belt on glass (158). Nonetheless, levels of phosphorylated tyrosine in the sealing zone were significantly less than those measured in individual podosomes and podosome belts. The origin of the sealing zone is a disputed topic.
Although, Saltel et al. earlier reported de-novo sealing zone formation, studies using high- resolution electron microscopy by Luxenburg et al. and Geblinger et al. have shown beyond doubt that the sealing zone has a structure that resembles that of a highly compacted podosome belt (149, 150). In the study by Geblinger and colleagues, scanning electron micrographs clearly reveal that the podosome is the building unit of a sealing zone and that they are connected to by actin fibres (149). The average thickness of the sealing zone is 3-6 μm and the inter-podosome distance is significantly smaller than it is on glass (250 ± 60 nm vs. 500 ± 140 nm respectively)
(149). The average sealing zone thickness and inter-podosome distance on calcite crystals and bone were not significantly different. However, the ruffled border was less pronounced on calcite crystals compared to bone (149). While the molecular composition of the sealing zone is not significantly different from that of the podosome belt, the distribution of these molecules and the effects of their deletion on sealing zone formation are different. In a sealing zone, the 19 adaptor molecules talin and vinculin are found encircling the actin condensation core rather than localizing with it (148). Additionally, c-Src is found in the ruffled border and not the sealing zone. Even though WIP deletion did not affect sealing zone formation and CD44 localization, bone resorption by WIP -/- osteoclasts is impaired (154). Demonstrating c-Src’s critical role in sealing zone formation, Src -/- osteoclasts are devoid of a sealing zone and have fewer podosomes (157).
During an osteoclast’s life span, the cell goes through several resorption cycles before undergoing apoptosis (162). A resorption cycle is a multistep process that is initiated when the osteoclast attaches to the bone surface and undergoes rapid actin repolymerization to become polarized with a compact cytoplasm (163). These changes result in the generation of the sealing zone and ruffled border (91, 164). When the osteoclast is finished resorbing at one site, another series of cytoskeletal rearrangements occur prompting the osteoclast to spread before it migrates to another location and the cycle is repeated. Osteoclast migration involves attachment and formation of lamellipodia on the leading edge and cell detachment on the trailing edge (165,
166). These changes result in the characteristic migrating osteoclast phenotype in which the osteoclast has a dendritic-like morphology with podosome patches at the leading edge where
v3 and F-actin are present but are not localized. Thus, the osteoclast’s resorption cycle corresponds to a morphological cycle with two main morphologies indicative of the function performed by the osteoclast at a certain point. These morphologies are the polarized (compact) and migratory morphologies with occasional transitional spread morphology in between.
Despite the lack of ruffled border in osteoclasts on glass (149), osteoclasts on glass can be seen alternating between the polarized and migratory morphologies similar to those on mineralized surfaces (167). This is further confirmed by our results presented in chapter 3. While only
20 matrix/integrin interaction is responsible for the formation of the sealing zone, RhoA GTPase regulates the organization of the podosomes into a sealing zone or podosome belt (168). While increased activation of Rho activity is needed for sealing zone formation, expression of constitutively active RhoA is not sufficient to initiate sealing zone formation on glass (152,
169). Further proof of the importance of Rho GTPases in sealing zone formation in that when
Rho GTPases are inhibited in polarized osteoclasts, osteoclasts immediately depolarize and spread (148). This is associated with the disappearance of the sealing zone and its replacement by a podosome belt (148). Crosstalk between cytokine and cytoskeletal signalling pathways regulates the sealing zone formation and osteoclast activation. Using a washaway –recovery system, M-CSF, RANKL, IL-1 and TNF were found to directly trigger sealing zone formation
(170).
1.6 Matrix/Integrin Interactions and Their Effects on Bone Homeostasis
1.6.1 Integrin Structure and Function
Integrins are a superfamily of adhesion receptors that act as bidirectional gateways on the cell surface mediating cell to cell and cell to matrix interactions (171). As heterodimeric transmembrane proteins, they are composed of non-covalently bonded and chains. In vertebrates, there are eighteen different subunits and eight different subunits, forming over twenty four distinct integrins. This makes them the largest class of cell adhesion molecules, with a highly diverse structure and function (172, 173). Despite their abundance, integrins have specific non-redundant functions and bind to distinct yet overlapping ligands. Integrins have a wide range of functions in both health and disease. These functions include embryonic development, autoimmune responses, leukocyte trafficking, tumour growth and metastasis,
21 blood clot formation and retraction, mechano-transduction, angiogenesis, bone homeostasis, inflammation and cell survival and apoptosis (173-179). Integrins’ unique roles are evident in the distinct phenotypes resulting from the deletion of different and subunits. Many of these transgenic deletions were embryonically lethal (3, 68v, 8 and some caused severe developmental defects (45v and 8), while others exhibited discrepancies in hemostasis
(IIb2and 3), bone remodelling (3) and angiogenesis (1 and 3) (4, 180-182)(reviewed and listed by (183, 184)).
The structure of integrin is well adapted to meet the demands of the complex and dynamic nature of their functions. Structural and topological information have been generated using X-ray crystal images, nuclear magnetic resonance, fluorescence resonance energy transfer
(FRET), electron microscopy (EM) and site specific mutagenesis. To act as linkers between the cell cytoskeleton and the matrix, integrins possess a single span transmembrane domain, a short cytoplasmic domain (40-70 amino acids) and a large and complex extracellular domain (185).
The extracellular domain of the chain (>940 amino acids) contains four or five domains, in integrins containing the insert (I) domain which is also known as von Willebrand factor A (A) domain (186, 187). These domains are the -propeller (188), the I/A domain, the thigh, the calf-
1 and calf-2 domains. Only half of the integrins contain the I-domain that is inserted in the - propeller and in those integrins, the I-domain is the site of ligand binding (as reviewed by
(189)). The subunit on the other hand is shorter (640 amino acids) and contains eight domains including an I-like domain, hybrid domain, the plexin/semaphorin/integrin (PSI) domain, four repeating integrin epidermal growth factor–like (I-EGF) domains and T domain (190-192).
Both the I and I-like domains present in the and subunits contain a Rossmann fold with metal ion dependent adhesion site (MIDAS) (186). From the crystal structure of integrin v3 it
22 was evident that the integrin can exist in a bent confirmation (192). In this confirmation the most N-terminal fragments of the and chains fold forms what is known as the “headpiece” while the rest of the extracellular domains forms the “tailpiece” (191-194). Most importantly, the I like domain contains the MIDAS which is critical for regulation of ligand binding affinity
(187, 192, 195). Ligand binding occurs through a series of conformational changes in the ligand binding domains induced by the binding of divalent cations and alternations in the MIDAS
(196). In integrins lacking the I domain such as v3, ligand binding occurs in an interface formed by both the propeller and the I-like domain (197). Due to the conformational changes associated with ligand binding, the ectodomain of integrin acquires three conformation states corresponding to its ligand binding (192, 197). Prior to ligand binding, integrins exist in equilibrium between these three activation states; the bent low affinity conformation, the extended conformation with closed headpiece and the extended conformation with open headpiece (198, 199). However, upon ligand binding, the ligand acts as a hatchet locking the integrin in an activated position in a “switchblade” like extended confirmation. While the classical outside-in integrin signal transduction occurs through ligand binding to the ectodomain, inside-out signalling occurs by activation of certain intracellular signalling pathways. The main extracellular factors regulating ligand binding and outside-in integrin activation are type and concentration of divalent cations (200-202). While high concentrations of Mn2+ and low concentrations of Ca2+ synergized with suboptimal Mg2+ are positive regulators inducing integrin activation, high concentrations of Ca2+ negatively regulate integrin’s activity
(203-205). The effects of divalent cation on integrin activation and ligand binding are mediated through MIDAS, ligand induced metal binding site (LIMBS) and adjacent to MIDAS
(ADMIDAS) (205, 206). Mg2+, low concentrations of Ca2+ and high concentrations of Ca2+
23 competed by Mn2+ induce their regulatory effects through MIDAS, LIMBS and ADMIDAS respectively (192, 197). Another extracellular factor triggering superactivation of integrins is a low concentration of an integrin antagonist such as an RGD peptide (207). In contrast to outside-in integrin activation, the mechanism of inside-out signalling transduction relies on conformational changes and separation of the and cytoplasmic domains of integrins (199,
200). Talin, another integrin activator, plays a central role in this process. Activation of intracellular signalling pathways downstream of growth factors results in the activation of talin.
Activated talin binds to the subunit, separating it from the subunit and causing the extension and activation of the extracellular domains, locking the integrin in this confirmation and thereby increases its ligand binding affinity (208). To underscore the specific roles of talin and Src family kinases in integrin activation, it was found that mutations in the cytoplasmic domain of
3 that prevent the binding of talin abrogated inside-out integrin activation. In contrast, mutations in the 3 Src binding domain resulted in inhibition of outside-in integrin activation and the associated cytoskeletal changes (209). Ultimately, integrin activation and bidirectional signalling result in lateral displacement of integrin and in a process known as integrin clustering
(210). Although the specific mechanism is not yet understood, it has been proposed that integrin clustering is required to increase the avidity of the integrin (211). Furthermore, the small
GTPase Rap1 is involved in this process as indicated by the inhibition of IIb3 activation in platelets when the enzyme is deleted (212).
1.6.2 Integrin v3 and Osteoclasts
24 Among the classes of adhesion receptors present in osteoclasts, integrins and most specifically the integrin v3 play an indispensable role not only in osteoclast attachment but also in differentiation and function. Several other integrins were identified in human osteoclasts, among these the 21 collagen/laminin receptor and v1 the fibronectin/vitronectin receptor
(213-215). Bone marrow macrophage derived osteoclast precursors express M1, v5
(another vitronectin receptor) and 41 in vitro (216-218). Integrins v5 and the fibronectin receptor 51 and potentially were also identified in avian osteoclasts (219, 220). Most recently, the integrin 91 was found in osteoclasts bound to the matrix metalloproteinase
ADAM8 (221). Despite the presence of multiple integrins in mature osteoclasts, the integrin
v3 is the predominant osteoclast attachment receptor and is highly enriched in osteoclasts.
The integrin v3 is also known as the vitronectin receptor which is a misnomer since the integrin binds several other matrix proteins such as OPN, FN, BSP, fibrinogen and denatured collagen type-I (222, 223). Similar to other vcontaining integrins, v3’s interaction with matrix proteins is through a common RGD domain. While the highest expression of v3 in vivo is present in osteoclasts, the integrin’s expression is physiologically elevated in the placenta where it was first isolated, and is present at a lower level in megakaryocytes, kidney, endothelial cells and vascular smooth muscle cells (224). The expression of v3 is upregulated during inflammation and bone metastasizing tumours (225, 226). Both aggregation and spreading of pre-fusion osteoclasts are essential for de novo synthesis and proper surface expression of v3 in culture during osteoclastogenesis (227, 228). Interestingly, during osteoclast differentiation the integrins v3 and v5 are reciprocally expressed on osteoclast precursors and mature osteoclasts respectively. v5 is the main integrin in macrophages, however, during osteoclastogenesis its expression progressively decreases and is gradually replaced by v3 25 (218, 229). This suggests the roles the two integrins play in regulating osteoclast formation are opposite and that v5 has an inhibitory effect on osteoclastogenesis. This hypothesis was corroborated when ovariectomized 5 null mice were found to be more prone to bone loss than the control group due to enhanced osteoclast formation and resorption of the 5 -/- osteoclasts in vitro and in vivo (230). In contrast, 3 deletion protects mice against ovariectomy-induced bone loss (231).
Similar to other integrins, v3 exists in two conformations; active and inactive, regulated by outside in and inside out signalling. In its high affinity (active) state, the integrin exposes its ligand induced binding site (LIBS) and this state is the result of either ligand binding extracellularly or growth factor activation intracellularly (232, 233). As previously mentioned,
v3 is found in the actin cloud of podosomes in osteoclasts on glass and on the basal membrane and around the sealing zone on bone. Careful examination of v3 distribution in light of its activation state revealed that the v3 present in podosomes on glass is in an inactive low affinity conformation. Once activated, the integrin translocates from podosomes to the lamellipodia on the leading edge, mediating and promoting osteoclast migration. On bone, the activated form of v3 is present on the ruffled border while the inactive form surrounds the sealing zone (232, 234).
The first evidence of the involvement of v3 in osteoclast activity was provided by the experiments showing that the monoclonal antibody 13C2 inhibited bone resorption (235).
Further investigations identified v3 as the antigen of 13C2 and it was later shown that in osteoclasts it is the major integrin (213, 236). Based on these findings, v3 became a novel target for inhibition of bone loss. Consequently, strategies developed for the prevention of bone loss revolved around interfering with the interaction between the RGD motif and v3 using
26 different methods such as blocking v3 antibodies, RGD peptides and mimetics and disintegrins. Studies have confirmed that bone resorption was inhibited when any of these methods were used in vitro (237, 238). In vivo, the disintegrin echistatin, the anti-rat 3 antibody (mAB F11) and RGD peptidomimetics reduced bone loss and serum calcium levels in hypercalcemic mice fed low calcium diet, a PTH induced bone loss model and in ovariectomized rats respectively (217, 239-241). The mechanism by which v3 inhibition affects bone resorption is still controversial. While in vitro data has shown that RGD peptides and echistatin impair osteoclast attachment and retraction, inhibition of v synthesis using an antisense oligodeoxynucleotide demonstrated that the impairment of osteoclast attachment is associated with induction of apoptosis signalling pathways through reducing the Bcl-2/bax ratio
(242). However, the in vivo findings contradicted the mechanism suggested by the in vitro experiments. In echistatin treated mice, the number of osteoclasts was increased while osteoclasts exhibited a normal morphology and no detachment could be seen (217, 241). Further in vitro investigations revealed that echistatin inhibited bone resorption at concentrations that did not affect osteoclast attachment and that it interfered with M-CSF induced osteoclast migration (243, 244). This proposed mechanism is supported by data showing that v3 activation promotes osteoclast haptotactic migration to OPN (218). Other data implicate specifically the ectodomain of the 3 subunit in this process (245).
Indisputable evidence of v3’s critical and direct role in bone metabolism arises from the transgenic deletion of 3. 3 knockout mice have increased bone mass which progresses into late onset osteopetrosis (4). Despite the 3.5 fold increase in their number, 3 null osteoclasts are dysfunctional due to abnormalities in ruffled border formation. Consistent with their osteopetrotic phenotype, these mice are also hypocalcemic (4). While mature osteoclasts
27 isolated from 3 -/- mice fail to exhibit a sealing zone, those formed in vitro from 3 -/- precursors do. Nonetheless, significantly fewer 3 -/- osteoclasts are formed in vitro and the osteoclasts are not able to form a normal ruffled border and therefore the size and number of resorption pits they form are substantially diminished (4, 246). In addition to those resorption- related defects, osteoclasts from 3 null mice fail to spread on RGD-containing substrates, indicating disruption of the osteoclast’s ability to undergo cytoskeletal reorganization (4). 3 deletion also had an unexpected anti-apoptotic effect on osteoclasts due to lack of caspase-8 signalling that results in cell death (247). Furthermore, the deletion of the cytoplasmic tail of 3 impaired osteoclast’s function in a similar fashion to that seen with the deletion of the full length of 3, indicating that the effects of 3 on osteoclast activity are mediated through its cytoplasmic domain (248). Using a series of 3 point mutations, the residue S752 in 3’s cytoplasmic domain was found to be specifically responsible for regulating osteoclast spreading, sealing zone formation and bone resorption (248). Interestingly, in humans S752P is one of the point mutations identified in some cases of Glanzmann thrombasthenia which is a disease characterized by disruption of hemostasis due to lack of IIb3 activation in platelets.
Conversely, the double Y747F/ Y759F mutation which was shown to inhibit platelet function had no effect on osteoclasts (248). Despite the definitive osteopetrotic phenotype of 3 ablation in mice, osteopetrosis was reported only in one case of Glanzmann mutation in humans. The discrepancy between the two phenotypes related to lack of 3 signalling may be due to the upregulation of 2 in Glanzmann thrombasthenia patients which partially restores bone remodelling (249).
28 1.6.3 The Molecular Mechanisms Involved in v3 Signalling
Three models have been proposed for the signalling cascade downstream of the outside- in v3 activation. These models all involve c-Src activation and recruitment to 3. While the proto-oncogene c-Src is abundantly expressed in osteoclasts, other Src kinase family members
(c-Fyn, c-Yes and c-Lyn) are present at a lower level (250). It was thus not surprising that c-Src deletion results in substantial bone anomalies and that other Src family kinases could not compensate for its absence. Similar to the phenotype seen with 3 ablation, c-Src knockout mice are osteopetrotic despite the increase in osteoclast number (251). c-Src -/- osteoclasts are dysfunctional but have an increased number of podosomes in the absence of c-Src’s regulatory role in podosome recycling and disassembly (160, 252, 253). In the context of v3 signalling, this non-receptor tyrosine kinase functions mainly as an adaptor protein as demonstrated by the ability of kinase deficient Src to partially abolish osteopetrosis when expressed in Src knockout mice (254). The c-Src -/- mouse model was a highly informative molecular tool that assisted in identifying other 3 downstream proteins. When the phosphorylation of PyK2, paxillin and
PLC was compared between c-Src +/+ and c-Src -/- upon cell adhesion, it was evident that all three molecules are part of the integrin mediated signalling pathway as the three molecules were phosphorylated only in the presence of c-Src (253).
Two of the three proposed outside-in signalling models are based on the interaction between c-Src and PyK2 while the third involves c-Src and Syk recruitment (80, 81, 255). In the first model, v3/ligand interaction causes an increase in the intracellular calcium levels which activates PyK2 through phosphorylating it at the Y402 residue (160, 256, 257). Through binding to the Src homology 2 (SH2) domain, PyK2 recruits c-Src to the 3 cytoplasmic tail and prevents its autoinhibition which can occur if c-Src phosphorylated Y527 binds to its own SH2
29 domain. Acting as a substrate for activated c-Src, c-Cbl becomes a part of the 3/PyK2/Src complex after binding to c-Src’s SH3 domain (160, 257, 258). It is thought that c-Cbl is a common signalling molecule between v3 and M-CSF signalling pathways and that its activation by M-CSF is the mechanism through which M-CSF participates in regulating the cytoskeleton (258, 259). Upon its activation and recruitment, c-Cbl acts as a ubiquitin E3 ligase which is responsible for the degradation of the PyK2/Src and M-CSF/c-Fms complexes (260).
In the second model, c-Src is upstream of PyK2 and is responsible for its phosphorylation. c-Src which is constitutively bound to the endodomain of 3 is activated upon ligand binding resulting in PyK2 phosphorylation. Lack of PyK2 phosphorylation in c-Src -/- osteoclasts supports this model (253). Similar to the previous model, PyK2 phosphoY402 and the
SH2 domain of c-Src are crucial for the osteoclast activation. The downstream effector molecules include p130cas, paxillin and PLC (253). Contrary of what would be expected given their central roles in osteoclast activation, Cbl deficient mice are not osteopetrotic but rather osteopenic and PyK2 deletion results only in mild osteopetrosis suggesting the presence of other compensatory molecules (261) (262).
More recently, a novel signalling pathway was illustrated by Teitelbaum’s group. They have shown that the non-receptor tyrosine kinase Syk is involved in the signalling events leading to cytoskeletal reorganization. The ligation of v3 triggers the association of Syk with the 3 cytoplasmic tail and c-Src to form a tri-molecule complex (80). RGD-occupancy of v3 simultaneously phosphorylates both c-Src which in turn activates Syk. The activated Syk binds to the activated ITAM adaptor molecules DAP12 and FcR via Syk’s SH2 domain thus recruiting the ITAM adaptor molecules to the large complex (80). Finally, phosphorylated Syk induces the downstream activation of the guanine nucleotide exchange factor Vav3 (80, 263).
30 As an activator of small GTPases, Vav3 mediates the necessary changes in the osteoclast’s cytoskeleton through inducing the Rho and Rac signaling cascade (246). Additionally, the adaptor molecule SLP-76 serves as a intermediate signalling molecule linking Syk to Vav3 activation (82). In support of their pivotal roles in regulating osteoclast function, Syk and Vav3 ablation results in quantitatively the same osteopetrotic phenotype and with similar osteoclast abnormalities to Src and 3 knockout mice (80, 263).
Cytokines, mainly M-CSF and IL-1, play an important role in modulating v3’s signalling pathways. Independently, M-CSF promotes osteoclast spreading and migration in a phosphoinositide 3-kinases (PI3K) dependent mechanism (264, 265). In macrophages, Rac/Akt activation is responsible for cell migration downstream of PI3K, suggesting that the same mechanism may exist in osteoclasts (266, 267). More importantly, M-CSF induces the spreading of c-Src -/- osteoclasts, demonstrating that the receptor tyrosine kinase does indeed control the actin cytoskeleton through a signalling pathway independent from v3 and c-Src (253). The proposed signalling pathway downstream of M-CSF includes PLC as the effector molecule that potentially activates Rac and Rho GTPases through Vav3 (5, 253). Another novel signalling M-
CSF dependent pathway was elucidated recently where tyrosine phosphorylation of c-Fms transduces the activation and binding of DAP12 and Syk (81). In addition to its specific effects,
M-CSF collaborates with v3 in mediating ligand induced osteoclast activation and differentiation. In 3 -/- mice, osteoclast numbers are unexpectedly increased due to the compensatory elevation of M-CSF which rescues osteoclast differentiation through sustained
ERK/c-Fos activation (268). In vitro, a high dose of M-CSF not only restores the number of osteoclasts formed from 3 -/- macrophages to normal levels but also stimulated osteoclast spreading and sealing zone formation (268). Interestingly, M-CSF was not able to restore bone
31 resorption neither in vivo nor in vitro; indicating that v3 has an indispensable role in bone resorption (4, 268). In a similar fashion, M-CSF is able to induce osteoclast migration and activation in c-Src -/- in a mechanism that involves v3 and causes the activation and recruitment of PLC making the enzyme a link between v3 and M-CSF signalling pathways
(5, 253). Further studies revealed the presence of two v3-containing complexes where either c-Src or c-Fms associate with the integrin (269). The association of v3 and c-Fms in complexes was enhanced in response to M-CSF treatment leading to c-Fms phosphorylation
(269). This complex also contained PyK2, p130cas and c-Cbl, but not paxillin. The multi- molecule signalling complex co-localized with podosomes on glass and with transcytotic vesicles on bone but was not seen in the sealing zone (269). Crosstalk between M-CSF and
v3 can also occur in a different manner, where M-CSF is responsible for inside-out v3 activation and ligation and indirectly triggering the downstream signalling cascade (246). The
S752 residue of 3’s cytoplasmic tail is required for this process (246). In parallel to M-CSF’s collaborative role in v3 signalling, IL-1 was also shown to directly increase sealing zone formation in osteoclasts (270). IL-1 initiation of cytoskeletal rearrangement involves the activation of TRAF6 which associates with the SH3 domain of c-Src followed by PyK2/p130cas complex formation inducing sealing zone formation (271).
1.6.4 Extracellular Matrix Proteins
1.6.4.1 Osteopontin
OPN is an abundant non-collagenous ECM glycoprotein highly expressed in bone (272,
273). Although OPN was initially regarded only as a matrix protein, its wide distribution in multiple tissues and organs including plasma and milk indicates that the protein also functions
32 as a humoral or cytokine-like factor (274, 275). The 300 amino acid protein is an acidic highly phosphorylated sialoprotein with highly conserved sequences across species elucidating its evolutionary fundamental role (276-278). It is transcribed by one gene (4q13-21) and its expression by bone cells is stimulated by pro-resorptive factors such as vitamin D3 and glucocorticoids which activate the transcriptional factors AP1 and Sp1 (279, 280). In addition to being highly phosphorylated, OPN undergoes extensive posttranslational modifications (PTM) that include glycosylation and sulphation (281-283). These PTMs are tissue specific and dictate the effects of OPN on osteoclast/osteoclast attachment and activity and the regulation of hydroxyapatite nucleation (284-288). Several domains are found in OPN including transglutamination domain located most N-terminally, an aspartate domain, the critical RGD domain, a 91/41 domain (162SVVYGLR168), a thrombin cleavage domain, two heparin binding domains and a putative calcium binding domain (289, 290). The tranglutamination domain is important for covalently binding FN and collagen (283). The aspartate domain composed of 10-12 aspartic acid residues, is responsible for the negative charge of OPN mediating its binding to the mineral of bone (291). Cleavage of OPN occurs mainly through thrombin activity; however, both MMP-3 and MMP-7 can also cleave the protein (292). Two functional fragments are created when thrombin cleaves OPN mid length. One fragment (N- terminal fragment) contains the RGD domain which interacts with integrins such as v3 while the C-terminal fragment contains the heparin domains which binds to CD44 (289). OPN cleavage increases the binding affinity of the protein to specific cell types through revealing certain cryptic domains and rendering the RGD domain more accessible (293). In bone, OPN is secreted by both osteoclasts and osteoblasts (279, 294). Through alternative splicing, an intracellular OPN (OPNi) isoform exists which regulates osteoclast migration and fusion
33 through localizing with CD44 and v3 in podosomes and is not perinuclear as is BSP (295,
296). In dendritic cells, the OPNi is transcribed from a site downstream from the 5' start codon which results in an isoform lacking the N terminus required for OPN targeting to secretory vesicles, thus OPNi is retained in the cytoplasm (297). OPN binds to a group of surface receptors including v (1, 3, 5) in an RGD dependent manner and (8, 9,4) 1 through the SVVYGLR motif (290, 298, 299). Finally, heparin binding domains are responsible for the ligation of CD44 isoforms (v6 and v7)/OPN (300).
Because of its multiple effects regulating the functions of many cellular processes in health and disease, OPN is considered a highly versatile ECM protein. The high expression of
OPN by activated T-cells and its pro-inflammatory and anti-inflammatory effects make this
ECM protein a double edged weapon in the progression of inflammation and tissue injury on one hand and tissue repair on the other hand (301). OPN’s pro-inflammatory effects are through promoting the attachment and migration while the anti-inflammatory effects are mediated by inhibiting nitric oxide (NO) expression. In a negative feedback loop, NO induces OPN expression which in turn inhibits NO, ultimately resulting in the inhibition of OPN. Another important role of OPN is in tumour progression (302, 303). High levels of OPN are seen in breast cancer, multiple myeloma and prostate cancer. As a tumour marker, an elevation in
OPN’s plasma level is negatively correlated with the patient’s prognosis (304). During the breast cancer cells’ metastasis to bone, induction of OPN in osteoblasts combined with OPN secreted by tumour cells leads to bone osteolysis (305, 306). In the skeleton, OPN plays an important role in bone mineralization and bone remodelling. In a study by Hunter et al. OPN was found to be the most potent inhibitor of HA formation and its effect was through inhibition of growth of HA crystals rather than inhibition of nucleation (285). The role of OPN in bone
34 remodelling was underscored in the OPN knockout mouse model. Although initially the bone phenotype of OPN -/- mice was overlooked, further investigation demonstrated that these mice exhibited a decrease in bone resorption in metaphyseal trabeculae albeit the increase in osteoclast number (17, 307). There was also an increase in the rigidity of bones. In vitro analysis of osteoclast function revealed that OPN -/- osteoclasts are dysfunctional. These osteoclasts had attenuated CD44 expression which rendered the cells hypomotile. In a different OPN deficient mouse, similar findings were reported (308). Interestingly, when ruffled border formation was compared in the wild type and null osteoclasts, OPN null osteoclasts had a several fold decrease in its length and volume (17, 308). Striking differences in bone phenotype in OPN deleted mice were reported during acute bone remodelling. OPN ablation protects mice against bone loss associated with ovariectomy, PTH treatment and collagen induced arthritis (309-311). OPN is also required for mechanosensing and its deficiency protects mice against osteoporosis in a tail suspension model (312, 313).
1.6.4.2 Fibronectin
FN is another ubiquitous ECM protein that plays an important role in embryonic development (314). The ECM protein is a homodimer composed of two 230-270 kDa peptides joined by a disulphide bond at the C-terminus (315). Similar to other mosaic proteins, FN is composed of repeating modules which make up its different functional domains (315). There are three different types of repeats in an FN molecule; type I, II and III repeats. Disulphide bridges connect type I and II repeats but not type III repeats (316). The critical RGD motif of FN is located in the type III10 and a PHSN synergy motif present in the adjacent type III9 and thus the two repeats are referred to as the cell binding domain. Although there are 17 type III repeats that can be transcribed from a FN gene, only 15 of them are constitutively expressed while the other 35 two are splice variants (317, 318). Two isoforms of FN exist that are generated though alternative splicing at three different sites (319). These isoforms are the plasma and the cellular
FNs. Plasma FN is secreted by hepatocytes and is found in high concentrations in blood while cellular is produced by fibroblasts and osteoblasts as well as numerous other cells and is a matrix protein. The main difference between cellular and plasma FN is the presence of two additional type III repeats known as extra type III A repeat (EIIIA) and extra type III A repeat
(EIIIB) (320). The third alternatively spliced motif is the V connecting segment (IIICS) present in both FN chains of cellular FN and only in one of the two chains in plasma FN. The IIICS motif plays a role in targeting FN to secretory vesicles (321). FN matrix formation is an integrin driven process that requires the interaction of certain FN domains with specific integrins. When
FN is synthesized, it is in a globular inactive form. Integrin binding exerts the force causing conformational changes and extension of the molecule exposing domains that are involved in
FN-FN binding and ultimately fibril formation (322). These conformational changes are initiated by the integrin 51 and fibril formation involves the N-terminal type I repeats (323,
324). Although it was initially thought that the RGD motif is important in FN matrix formation, transgenic mice with an RGE point mutation were able to properly assemble an FN matrix
(325). The type I5 repeat iso-DGR motif was identified in the same group as an alternative FN fibril assembly motif (325). Cell attachment to FN occurs through a large group of integrins recognizing multiple motifs. Although 51 is considered the most critical integrin for RGD and PHSRN (synergy motif) driven FN recognition, the integrins v3, v1, 81 and 31 are also able to bind to the RGD motif (as reviewed in (326)).
In vitro, FN was found to be essential for osteoblast differentiation and for proper collagen matrix formation (327). Osteoblasts treated with an anti-FN antibody have a decreased
36 capacity for mineralization in a bone nodule formation assay (327). Despite these implicating findings, FN’s role in bone homeostasis in vivo is not clear. Additionally, the amount, type and source of FN in bone have not been confirmed. Two recent studies have shed light on FN and its effects on bone remodelling. In the first study, a strong correlation was established between the increase in the plasma levels of the two isoforms of cellular FN; FN EIIIA and oncofetal FN (an
FN isoform glycosylated at the residue 33) and a decrease in osteocalcin, denoting a decrease in bone formation (328). More compelling evidence in line with those findings was generated from comparing two conditional FN knockout mice. The findings of the study by Bentmann et al. demonstrated that the plasma FN is the predominant type in bone and that its deletion decreased bone mineralization and compromised its quality; however, osteoclast and osteoblast functions were not altered (15). In contrast, no significant effects on bone mineral density were detected when cellular FN in osteoblasts was knocked out. Albeit its lack of effect on bone quantity, the deletion of cellular FN did indeed increase osteoblast number and stimulate their ability to lay down osteoid matrix (15).
1.6.4.3 Vitronectin
VN is a plasma protein that acquired its name due its high affinity to glass (vitreous).
This soluble ECM matrix protein that bears resemblance with S-protein is present in high levels in plasma (200-400 μg/ml), yet its expression in the ECM is low. Vitronectin is synthesized by hepatoctyes as a 75 kDa polypeptide but can also be seen as a 65/10 kDa disulfide bonded two- chain form. In addition to its plasma soluble form, VN is also found in a fibrillar form in fibroblast cultures (329, 330). There is no evidence of the presence of any isoforms through alternative splicing or PTMs which is encoded by one gene (331). Although hepatocytes are the main source of VN, its secretion was induced by endotoxins in macrophage cultures (332). 37 Similar to FN and OPN, VN has multiple binding domains; the RGD domain that regulates cell spreading and migration, the somatomedin B region harbouring the plasminogen activator inhibitor, the type I (PAI-1) binding site involved in hemostasis and a heparin binding domain
(333, 334). While the RGD domain is present in the connecting sequence adjacent to the PAI-1 binding domain and is N-terminally located, the heparin binding domain is more C-terminal and found in the middle of hemopexin repeats and contains secondary PAI-1 binding domains (330).
Through binding to VN, PAI-1 is maintained in its active conformation where it inhibits urokinase type plasminogen activator and thrombin, thus preventing the signalling cascades causing clot formation (335). In bone, VN is shown to be widely distributed in the mineralized matrix and its affinity to HA is suggested to be the reason for its presence in the mineralized bone matrix (336). Despite these findings, only the effect of VN on osteoclasts has been studied.
In in vitro studies, VN has been used mainly as a method of increasing osteoclast attachment and study integrin activation related events.
1. 7 Rationale and hypothesis
The integrin v3 is the most abundant osteoclast surface receptor and therefore plays an important role in osteoclast function. Because v3 is essential for bone resorption, several strategies were developed to inhibit its interaction with its RGD motif to be used as treatment for preventing bone loss in arthritis, osteoporosis and osteolytic tumours. One of these strategies lead to the development of the anti-human v antibody Vitaxin® which binds a conformational epitope formed by the v and subunits (both human and rabbit) and blocks the binding of v to ECM proteins. Vitaxin is currently in clinical trials for the treatment of bone loss in metastatic bone tumours. While promising results were seen clinically when
38 Vitaxin was used on cancer patients, the cellular mechanisms of how the antibody mediated its effects are not known. Additionally, the effect of extracellular and intracellular factors regulating v activation state on the binding of blocking antibodies had not been underscored.
Although v3 recognizes a common RGD motif in several ECM proteins in bone, the effect of this interaction is not redundant as indicated by the different phenotypes seen when bone matrix proteins are deleted. The ECM matrix proteins FN and VN are present in the milieu during osteoclast formation; however, their specific effects on osteoclastogenesis have not been studied. Evidence suggests that VN and FN are involved in activating osteoclasts specifically during inflammatory bone loss. The local and systemic increase in VN and FN’s expression in arthritic diseases combined with the reported induction of several cytokines and proteases by these proteins in other cell types made them potential targets to study with respect to osteoclast formation and function.
Hypothesis I: The v3 blocking antibody Vitaxin inhibits osteoclastic bone resorption through impairing osteoclast attachment leading to apoptosis. Factors activating v3 enhance Vitaxin binding to its conformational epitope that is accessible only when the integrin is in its extended form, increasing Vitaxin’s inhibitory effect on bone resorption.
Hypothesis II: ECM matrix proteins differentially regulate osteoclast formation and function through their interaction with v3. Similar to OPN, VN and FN induce osteoclast formation and activation through regulating the expression of pro-resorptive factors.
39 Aim (1): Studying the effect of the v3 blocking antibody Vitaxin on the attachment, differentiation and resorptive activity of pre-fusion and mature rabbit osteoclasts cultures and determining the effects of v3 activating (M-CSF) and inactivating (high Ca2+ concentration) factors on Vitaxin’s inhibition of osteoclast attachment.
Aim (2): Studying the differential effects of the ECM protein FN and VN on osteoclastogenesis compared to those of OPN in the RAW264.7 cell line. The mechanism by which FN and VN modulate the formation, activity and morphology of osteoclasts formed on each of these proteins will be elucidated. The expression of and function of the integrin v3 and pro-resorptive factors by osteoclasts formed on FN and VN will be determined.
40 Chapter 2
Effects of Vitaxin®, a Novel Therapeutic in Trial for Metastatic Bone Tumors, on Osteoclast Functions in vitro
This work was published in the Journal of Cellular Biochemistry; Gramoun A, Shorey S,
Bashutski JD, Dixon SJ, Sims SM, Heersche JNM, Manolson MF “Effects of Vitaxin®, a novel therapeutic in trial for metastatic bone tumors, on osteoclast functions in vitro”. Volume 102(2), pages 341-352, Copyright © 2007 Wiley-Liss, Inc.
41 2.1 ABSTRACT
The integrin v mediates cell-matrix interactions. Vitaxin®, a humanized monoclonal antibody that blocks human and rabbit v integrins, is in clinical trials for metastatic melanoma and prostate cancer. v is the predominant integrin on osteoclasts, the cells responsible for bone resorption in health and disease. Here, we report the first investigation of
Vitaxin's effects on osteoclast activity. Vitaxin (100-300 ng/ml) decreased total resorption by
50%, but did not alter resorptive activity per osteoclast. Vitaxin (300 ng/ml) decreased osteoclast numbers on plastic by 35% after 48 h. Similarly, attachment after 2 h was reduced by
30% when osteoclasts were incubated with Vitaxin (300 ng/ml) for 25 min prior to plating; however, the rate of fusion of osteoclast precursors in Vitaxin-treated and control groups was equal. Using time-lapse microscopy, we evaluated the effect of Vitaxin on osteoclast morphology and found a significant reduction in osteoclast planar area only when cells were pretreated with macrophage colony stimulating factor (M-CSF). Extracellular Ca2+ and M-CSF have opposite effects on v conformation. Elevation of extracellular Ca2+ eliminated the inhibitory effect of Vitaxin on osteoclast attachment. In contrast, the effect of Vitaxin was enhanced in cells pretreated with M-CSF. This action of M-CSF was suppressed by the phosphatidylinositol 3-kinase (PI3-kinase) inhibitor wortmannin, suggesting that M-CSF increases Vitaxin's inhibitory effect by inside-out activation of v. In conclusion, Vitaxin decreases resorption by impairing osteoclast attachment, without affecting osteoclast formation and multinucleation. Our data also show that Vitaxin's inhibitory effects on osteoclasts can be modulated by factors known to alter the conformation of v.
42 2.2 INTRODUCTION
Bone remodeling is a physiological process dependent on the balance between formation and resorption. This balance is determined by both the number and activity of osteoblasts, the bone forming cells, and osteoclasts, the bone resorbing cells (337). Imbalance between formation and resorption leads to pathological bone loss in rheumatoid arthritis (RA) and several other bone diseases. RA is associated with, periarticular and subchondral bone loss, which contributes to joint destruction. There is still a need for RA treatments that are able to effectively relieve pain and block joint destruction (338).
Osteoclasts are multinucleated cells formed by fusion of hematopoietic mononuclear precursors (339). They are unique in their ability to degrade both the inorganic and organic components of bone. Essential to osteoclastic bone resorption is the integrin v, although the mechanism of how v affects osteoclastic activity is not fully understood. v is a type I transmembrane glycoprotein receptor consisting of and and subunits that functions as a bi- directional gateway mediating cell-matrix interactions. As the predominant integrin on the osteoclast surface (236), v is postulated to be involved in differentiation, attachment, and resorption (183). v-Mediated attachment to matrix proteins such as osteopontin and vitronectin is through an RGD binding domain (216). Matrix recognition and attachment of osteoclasts via v is responsible for initiation of signaling cascades activating osteoclasts (45,
228). The activated osteoclast is polarized and is characterized by two distinctive structures, the ruffled border and the sealing zone. Together, the sealing zone and the ruffled border create an isolated microenvironment of acidic pH required for mineral dissolution and matrix degradation by cathepsin K and other enzymes.
43 v3 integrin exists in two different conformations, basal or activated (192, 197).
Changes in activation state correlate to ligand binding efficiency (234, 340). Divalent cations such as Mg2+ and Mn2+ induce activation of v integrins and enhance their ligand binding affinity. Conversely, high concentrations of Ca2+ shift the integrin into an inactive conformation, lowering its binding affinity (207). Hepatocyte growth factor and macrophage colony stimulating factor (M-CSF), two growth factors involved in osteoclastogenesis, increase the number of activated receptors through inside-out activation of v (341). It has also been shown that M-CSF induces osteoclast spreading and cytoskeletal reorganization and consequently affects osteoclast attachment and resorptive ability. The effects of M-CSF on osteoclast structure and function are due in part to its ability to increase the affinity of v.
Those effects were found to be mediated through phosphatidylinositol 3-kinase (PI3-kinase) and required the presence of S752 in the 3 cytoplasmic tail (264, 265, 341, 342).
As a key step in bone resorption involves osteoclast attachment via v, numerous efforts to inhibit bone resorption have been initiated using antagonists that mimic the RGD motif and competitively occupy the receptor (343). Alternatively, blocking antibodies have been investigated because of their potential specificity for v. Peptidomimetics, disintegrins, and blocking antibodies reduce resorption in vivo and in vitro, by affecting osteoclast differentiation, migration, and/or attachment (243, 344-348).
In this study, we evaluated the role of v on osteoclast function using the v blocking antibody Vitaxin®. Vitaxin is a humanized monoclonal antibody that binds a conformational epitope formed by the v and subunits (both human and rabbit) (349, 350).
Vitaxin is currently in clinical trials for the treatment of stage IV metastatic melanoma and androgen-independent prostate cancer (351). The outcome of phase II clinical trials showed that
44 Vitaxin increased median survival of metastatic melanoma patients with minimal side effects.
v antagonists, including Vitaxin, have been used to target both primary and metastatic tumors as the integrin is known to be expressed on tumor endothelial cells. The v integrin also affects the establishment and growth of tumors in bone through its role in initiating osteoclastic bone resorption (352-354). Vitaxin was also considered for the treatment of RA due to its antiangiogenic properties (350). To date, there is no data available on how Vitaxin specifically affects osteoclasts and their function.
The focus of this study was to determine the effects of Vitaxin on osteoclastic resorption and the cellular mechanisms underlying these effects. We also studied if Vitaxin had different effects on specific osteoclast populations and how responses to the antibody can be modulated by factors known to affect the conformation of v. Using authentic rabbit osteoclasts, we asked if Vitaxin affected osteoclast attachment and/or formation. Here, we show that Vitaxin reduces resorption by decreasing initial osteoclast attachment and cell spreading. In contrast,
Vitaxin did not affect the resorptive capacity of individual osteoclasts nor did it affect the rate of osteoclast formation. Our data also suggests that the inhibitory effect of Vitaxin on osteoclasts is modulated by conformational changes in v.
45 2.3 MATERIALS AND METHODS
2.3.1 Materials
Vitaxin, an anti-human v antibody, was generously provided by Dr. Su-Yau Mao,
MedImmune, Inc. (Gaithersburg, MD), sheep IgG anti-type I collagen antibody was kindly provided by Dr. J. Sodek (University of Toronto) and biotinylated donkey anti-sheep IgG was obtained from Sigma-Aldrich Ltd. (St. Louis, MO). Recombinant mouse M-CSF was purchased from Calbiochem (EMD BioSciences, Inc. San Diego, CA). -Minimum essential medium (-
MEM) was purchased from Medstores, University of Toronto (Toronto, Ont.), fetal bovine serum (FBS), and antibiotic-antimycotic solution (100X) were obtained from Invitrogen
(Carlsbad, CA). Fast red violet LB salt, naphthol AS-MX, 3,3'-diaminobenzidine tetrahydrochloride (DAB) and wortmannin were obtained from Sigma-Aldrich Ltd. (St. Louis,
MO). Avidin-biotin-peroxidase complex was bought from Dimension Labs (Burlingame, CA).
BioCoatTM OsteologicTM bone cell culture system (16 well), pretreated 6- and 12-well plastic
FalconTM tissue culture plates and 96-well FalconTM microtiter plates were purchased from BD
Biosciences (BD Labwares, Franklin Lakes, NJ).
2.3.2 Rabbit Osteoclast Isolation
Animal protocols were approved by the Animal Care Committees at the University of
Toronto and the University of Western Ontario. Osteoclasts were isolated from the long bones of newly born New Zealand rabbits, as previously described (162, 355). Briefly, bones were cleaned and minced mechanically in 100 mm glass Petri dishes containing 10 ml of -MEM supplemented with 10% FBS and 1% antibiotic-antimycotic solution (supplemented media).
46 Bone fragments were transferred into 50 ml Falcon tubes and cells were resuspended by repeated passage (30 times) through a wide-bore Pasteur pipette. The bone fragments were allowed to briefly settle and the cell suspension was transferred to another tube. An additional 6 ml of supplemented media was then added to the remaining minced bones in the Petri dish and the previous steps were repeated. This technique of cell isolation generates a mixed culture of attached osteoclasts, prefusion osteoclasts, and stromal cells. To ensure that cell counts reflected the number of osteoclasts, only multinucleated cells that stained positive for tartrate-resistant acid phosphatase (TRAP) were counted. To stain for TRAP activity, cultured cells were washed with PBS, and then fixed with 2.5% glutaraldehyde for 5 min. Staining was carried out according to the protocol described in BD Biosciences Technical Bulletin #445.
2.3.3 Preparation of Devitalized Cortical Bone Slices
Cortical bone slices were prepared from bovine long bone obtained from a butcher shop as described previously (116). Briefly, a devitalized block of bovine cortical bone was thawed.
Sections of 120-150 µm thickness were obtained using a Buehler Isomet low speed saw. After polishing these sections, discs of 5 mm diameter were punched out using a cork borer. Slices were then sonicated once in distilled water for 10 min, disinfected by submersion in 70% ethanol overnight, washed in -MEM (3× 10 min) and then incubated overnight at 37°C in supplemented medium.
2.3.4 Attachment Studies
Aliquots (100 µl) of cell suspension were plated in 6-well plastic plates. Osteoclasts were allowed to attach for 1 h (37°C, 5% CO2), after which 1.5 ml of supplemented medium
47 was added. After a further 18-h period, the cultures were washed gently with -MEM using a wide-bore Pasteur pipette to remove non-attached cells and then incubated in supplemented media with or without Vitaxin (30, 100, and 300 ng/ml) for an additional 48 h. Cells were subsequently fixed and stained for TRAP activity.
“Initial attachment” experiments were performed by diluting the initial cell suspension
1:1 with an equal volume of supplemented media with or without Vitaxin at 4°C for 30 min prior to plating. Cultures were incubated for an additional 2, 3.5, 4.5 or 24 h before cells were fixed and stained for TRAP activity. The number of attached osteoclasts was determined by counting the number of TRAP-positive multinucleated cells using a light microscope at 200× magnification.
Experiments to investigate the effects of M-CSF and Ca2+ on Vitaxin-induced inhibition of osteoclast attachment were done by plating 250 µl aliquots of cell suspension in 12-well plastic culture dishes, and incubating for 1 h at 37°C and 5% CO2. Subsequently, 1 ml of supplemented -MEM was added either with or without 2.5 mM Ca2+ or 50 ng/ml M-CSF, followed by an 18-h incubation. Cultures were then washed gently with -MEM to remove non- attached cells and further incubated in the absence or presence of Vitaxin (300 ng/ml) for 48 h.
Cells were stained for TRAP and counted.
Experiments aimed at examining the signaling pathway mediating the effects of M-CSF were done using the PI3-kinase inhibitor, wortmannin. Briefly, osteoclast cultures grown for 48 h were washed with Ca-free Hank's balanced salt solution before they were incubated in PBS +
4% FBS with or without 500 nM wortmannin at 4°C for 1 h. This was followed by incubating the cultures in presence or absence of 50 ng/ml M-CSF and/or 300 ng/ml Vitaxin for an
48 additional 1.5 h at 37°C and 5% CO2 before the experiment was stopped and osteoclasts were fixed.
2.3.5 Time-Lapse Microscopy
Morphology and motility of osteoclasts were assessed using time-lapse video microscopy as previously described (243). Briefly, osteoclasts were either untreated or treated for 30 min with 100 ng/ml of M-CSF. After the pretreatment period, Vitaxin (100-1,000 ng/ml) was added and the effects on osteoclasts were monitored for another 30 min. To quantify responses, the periphery of each cell was outlined and the planar cell area was calculated at 5 min intervals using digital image analysis. Data were normalized as a percentage of the initial area.
2.3.6 Resorption Studies on Bovine Bone Slices
Aliquots (100 µl) of the initially isolated cell suspension were plated on cortical bone slices in
96-well plates. After an 18-h attachment period, the cultures were washed gently with -MEM to remove unattached cells and then incubated for an additional 48 h in the presence or absence of Vitaxin. Cells were then fixed and stained. Immunohistochemical staining for collagen type I was used to identify resorption lacunae as previously described (108).
2.3.7 Resorption Studies Using the Osteologic Bone Cell Culture System
Resorption assays were also performed on 16-well osteologic slides using the same experimental design and culture conditions described for bone slices. Following the 48-h culture period, cells were detached from the surface of osteologic slides according to the manufacturer's
49 recommendation. Briefly, 200 µl of bleach solution (6% NaOCl, 5.2% NaCl) was added to wells before removing media and the slides were agitated for 5 min. Wells were washed 2× with 150
µl dH2O and then examined under the light microscope to ensure complete removal of cells. The contrast between non-resorbed areas and resorption pits was visualized using von Kossa staining. Briefly, 150 µl 3% AgNO3 was added to the wells for 18 min in the dark. Wells were thoroughly washed 5× with 150 µl dH2O, then developed in 0.5% hydroquinone for 3 min.
Wells were washed again 5× with 150 µl dH2O then fixed using 150 µl 2% Na2S2O3·5H2O. The total area of resorption and number of resorbed regions were quantified using an ImagePro® analysis system.
2.3.8 Statistics
Statistics were carried out using SPSS 12.0 for Windows using one way analysis of variance (ANOVA) and Dunnette T3 test. P-values less than 0.05 were considered statistically significant.
50 2.4 RESULTS
2.4.1 Vitaxin Inhibits Osteoclast Resorption
We first evaluated the effect of Vitaxin on osteoclastic resorption using osteologic slides.
Rabbit osteoclasts were left to attach for 18 h then incubated with Vitaxin at 30, 100, and 300 ng/ml for an additional 24 h. Resorption was significantly decreased by ~55% at 100 and 300 ng/ml, whereas 30 ng/ml Vitaxin resulted in a slight increase in resorptive activity (Figure 2.1).
Although the increase at 30 ng/ml was not statistically significant, this observation is consistent with the observation that v antagonists at low concentrations can cause the activation of the integrin (207).
2.4.2 Vitaxin Decreases the Number of Osteoclasts Attached to Plastic
To address whether the decrease in resorption was due to a decrease in osteoclast number, the number of osteoclasts that remained attached after treatment with Vitaxin was examined. Osteoclasts were cultured on plastic for 18 h before introducing Vitaxin (30, 100 and
300 ng/ml) for an additional 48 h. A decrease in the total number of osteoclasts attached was observed with a maximum of ~35% inhibition at 300 ng/ml (Figure 2.2A). The addition of fresh
Vitaxin in the middle of the 48-h incubation made no difference to the results (data not shown) suggesting that there was no proteolytic degradation of the antibody over the 48-h culture period.
Osteoclasts form and enlarge through the process of fusion. To determine whether the decrease in osteoclast number was due to increased fusion of existing multinucleated osteoclasts, the total number of nuclei in the experiment was quantified by hematoxylin staining.
51 The decrease in the total number of nuclei paralleled that seen in the number of osteoclasts
(Figure 2.2B); suggesting that the decrease in osteoclast number is not the result of increased fusion.
2.4.3 Vitaxin Preferentially Inhibits the Attachment of Small Osteoclasts (<10 Nuclei)
Substantial differences exist between large and small osteoclasts, as defined by their number of nuclei, with respect to their resorptive activity and pH regulation (355). We decided to determine whether Vitaxin differentially affected one of the two distinct cell populations.
Figure 2.2C shows that Vitaxin decreased the number of small osteoclasts by an average of 35% whereas its effect on large osteoclasts was negligible. Vitaxin's preferential inhibition of small osteoclasts may reflect the differences in v3 expression noted between large and small osteoclasts (356).
2.4.4 Vitaxin Does not Affect the Resorptive Activity of Attached Osteoclasts
The results described above show that the inhibition of osteoclastic resorption by Vitaxin is explained at least in part by the decrease in the number of attached osteoclasts. We next asked whether Vitaxin might also interfere with the resorptive capacity of individual osteoclasts that remained attached. Similar to our findings with hydroxyapatite-coated slides (Figure 2.1), incubation of osteoclasts on bone slices with Vitaxin (100 ng/ml) for 48 h decreased both the total area of resorption as well as the number of osteoclasts attached by 50% (Figure 2.3A,B).
When the resorptive activity per osteoclast was calculated, we found that treatment with Vitaxin had no significant effect (Figure 2.3C), suggesting that Vitaxin does not affect the resorptive capacity of attached osteoclasts but rather affects osteoclast attachment and/or formation.
52 2.4.5 Vitaxin Inhibits Attachment but Not Early Stages of Osteoclast Formation
Since v3 plays a role in osteoclastogenesis (339), the effect of Vitaxin was determined on early stages of osteoclast formation and multinucleation. To distinguish between actions on osteoclast attachment and formation, we examined the effects of pre-incubation with
Vitaxin on attachment of osteoclasts after 2 h. We also examined the effects of Vitaxin on osteoclast number following incubation in the presence or absence of Vitaxin up to 24 h to assess formation. When the freshly isolated cell suspension was incubated with Vitaxin (300 ng/ml) at 4°C for 30 min prior to plating, osteoclast attachment measured at 2 h was decreased by ~40% (Figure 2.4A). When the cells were cultured for a 3.5, 4.5, and 24 h, we observed a gradual and similar increase in the number of multinucleated osteoclasts in both the control and the Vitaxin-treated groups (Figure 2.4B). Micrographs from this experiment illustrate that neither the number of preosteoclasts nor the size of the aggregates they formed prior to fusing was altered by blocking v (Figure 2.4C). When the percentage increase in osteoclast number was calculated between the 2 and 24 h time points and combined from three different experiments, there was no significant difference between the control and the Vitaxin-treated groups (data not shown). These data are compatible with the view that the major effect of
Vitaxin is on osteoclast attachment, rather than formation and multinucleation.
2.4.6 Vitaxin Causes Retraction of Osteoclasts Only in the Presence of M-CSF
Cell morphology studies were done to assess Vitaxin's effect on osteoclast spreading.
When osteoclasts were treated with Vitaxin (100 ng/ml) alone, no significant changes were observed in their planar area during a 25 min treatment period (Figure 2.5B). In contrast, after treating cultures with 100 ng/ml M-CSF, Vitaxin significantly decreased the osteoclast planar
53 area by 25% within 5 min (Figure 2.5 B,C). Figure 2.5A comprises micrographs taken during the experiment and is representative of the change that occurred in osteoclast morphology after treatment with M-CSF and Vitaxin.
2.4.7 Vitaxin's Effect on Attachment is Altered by Factors Known to Change the
Conformation of v3
Integrins alternate between an inactivated and an activated conformation, with the activated form having a higher ligand affinity (192, 197). Growth factors, divalent cations, and anti-ligand induced binding site antibodies (anti-LIBS) (357) all shift the integrin from one form to the other and therefore affect cell attachment and spreading. It has been shown that M-CSF increases, whereas high concentrations of Ca2+ decreases the number of activated v3 receptors on the plasma membrane of osteoclasts (234, 341). To assess if Vitaxin is dependent on the integrin's conformational state, osteoclasts were incubated with either M-CSF (50 ng/ml) or Ca2+ (2.5 mM Ca2+ total) for 18 h before the addition of 300 ng/ml Vitaxin for a further 48 h of culture. Vitaxin in the presence of M-CSF resulted in a 55 ± 4% decrease in osteoclast number compared to only 33 ± 3% decrease in the Vitaxin only treated group (% change was calculated based on results obtained from three independent experiments). In contrast, in the presence of 2.5 mM Ca2+, there was no inhibition in the Vitaxin-treated group (Figure 2.6A).
These results suggest that Vitaxin preferentially inhibits the activated form of v3. To examine whether M-CSF's effects were mediated through PI3-kinase, we added a PI3-kinase inhibitor, wortmannin, for 1 h prior to stimulating the cultures with M-CSF. In Figure 2.6B, we again show that Vitaxin's inhibitory effect was increased in the presence of M-CSF, but the M-CSF-
54 dependent increase was abolished by wortmannin. The presence of wortmannin alone did not significantly change Vitaxin's inhibitory effect (Figure 2.6B)
55
Figure 2.1 Vitaxin decreases osteoclast resorption on osteologic slides. Rabbit osteoclasts were cultured on osteologic slides for 18 h before adding Vitaxin (30-300 ng/ml) for an additional 24 h.
Non-resorbed areas were visualized using von Kossa staining. The surface area of pits was outlined manually and measured using ImagePro software system. Each data point represents the pooled results from four discs per treatment and is expressed as total area of resorption per well
(means ± SEM). *P-value < 0.05 versus control group. Similar results were obtained in two separate experiments.
56 Figure 2.2 Vitaxin decreases the attachment of small osteoclasts (OCs) (<10 nuclei) on plastic. Rabbit osteoclasts were cultured in 6- well plates for 18 h before adding Vitaxin
(30-300 ng/ml) for an additional 48 h. Cells were (A) stained for TRAP activity or (B) nuclei were stained using hematoxylin. The number of TRAP + osteoclasts attached to plastic and the total number of nuclei in osteoclasts were counted using a light microscope at 100× magnification. C: The total number of TRAP + osteoclasts and the number of TRAP-positive osteoclasts with
≥10 nuclei were counted using a light microscope at 200× magnification. Each data point represents the pooled results from three wells per treatment and is expressed as total number per well. Data are means ± SEM.
Similar results were obtained in three other experiments. *P-value <0.05 versus control
group.
57 Figure 2.3 Vitaxin decreases osteoclast (OC) attachment and resorption on bone but does not affect resorbed area per osteoclast. Rabbit osteoclasts were cultured on bone slices for
18 h before Vitaxin (100 ng/ml) was added for an additional 48 h. The number of TRAP- positive osteoclasts attached to bone slices was counted using a light microscope at
250× magnification (A). Resorption pits were visualized using collagen type I staining. The surface area of pits was outlined and measured using ImagePro software system
(B). and the resorbed area per osteoclast was calculated (C). Each data point represents the pooled results from six bone slices per treatment and is expressed per well. Similar results were obtained in two separate experiments. Data are the means ± SEM. *P- value <0.05 versus control group.
58
Figure 2.4 Vitaxin decreases osteoclast (OC) initial attachment but does not affect osteoclast formation. Rabbit osteoclasts were resuspended in media with or without 300 ng/ml Vitaxin for
30 min at 4°C, then plated in 6-well plates and incubated for an additional 2 h (A), 3.5, 4.5, and
24 h (B) under standard culture conditions in the continued presence or absence of Vitaxin. Cells were stained for TRAP activity and the number of TRAP-positive osteoclasts was counted. Each data point represents the pooled results from three wells per treatment and is expressed per well.
Data are means ± SEM. Similar results were obtained in two other experiments. *P-value < 0.05 versus control group. C: Micrographs show that 300 ng/ml Vitaxin does not affect the number of preosteoclasts or the size of the aggregates they formed over a 24 h culture period in comparison to the control.
59
Figure 2.5 Vitaxin decreases osteoclast (OC) planar area only when cultures are pretreated with M-
CSF. Isolated rabbit osteoclasts were cultured for 24 h before they were incubated for 30 min in the presence or absence of 100 ng/ml of M-CSF. After the pretreatment period, Vitaxin (100 ng/ml) was added and planar cell area was monitored for another 25 min. A: The series of video micrographs shows the response of an M-CSF-treated osteoclast to Vitaxin. Frame i shows an osteoclast before addition of M-CSF (100 ng/ml). Frame ii shows the same osteoclast after addition of M-CSF. Dotted line represents the outline of the cell in Frame i superimposed on Frame ii. Frame iii shows retraction after addition of Vitaxin (100 ng/ml). Dotted line represents the outline of the cell in
Frame ii superimposed on Frame iii. To quantify retraction induced by the antibody, the periphery of each cell was outlined and the planar cell area was calculated at 5 min intervals using digital image analysis and normalized to the initial area (B). Panel C represents the planar area of osteoclasts, 5 min after adding Vitaxin to the cultures. Data are means ± SEM of 12 osteoclasts from three independent experiments. *P < 0.05.
60
Figure 2.6 Vitaxin's effect on attachment can be altered by factors affecting the conformation ofv. A: Rabbit osteoclasts (OCs) were cultured in 12-well plates in the presence and absence of
50 ng/ml M-CSF (activates v) or 2.5 mM Ca2+ (inactivatesv) for 18 h before adding 300 ng/ml Vitaxin for an additional 48 h. B: Osteoclasts were incubated in the presence or absence of
500 nM wortmannin (PI3-kinase inhibitor which blocks M-CSF induced activation) for 1 h at 4°C.
Cultures were subsequently incubated in the presence or absence of 50 ng/ml M-CSF with or without 300 ng/ml Vitaxin for 1.5 h at 37°C 5% CO2. Cells were stained for TRAP and the number of TRAP-positive osteoclasts was counted. Each data point represents the pooled results from four wells per treatment and is expressed as total number of osteoclasts per well. Data are means ± SEM.
Similar results were obtained in two other experiments. a: P-value <0.05 and b: P-value <0.01.
61 2.5 DISCUSSION
Vitaxin, an anti-human vantibody is currently in clinical trials as a treatment for metastatic melanoma and prostate cancer (351, 358). To help understand how Vitaxin prevents metastatic bone loss in cancer patients, the aim of this study was to elucidate the mechanism by which this therapeutic decreases bone resorption. Insights into how osteoclasts interact with the
vantagonist are important for the process of developing new generations of treatment which are more effective and also to determine methods of using the antibody as a cancer imaging and drug delivery agent. Previous studies have shown that RGD mimetics that block the ligand-binding site of integrin valso inhibit bone resorption in vitro and in vivo (243, 344-348). The results reported here show that Vitaxin also inhibits bone resorption.
To study whether the inhibition of resorption was caused by a reduction in the number of resorbing osteoclasts, we examined the effect of Vitaxin on the number of attached osteoclasts. We found a decrease in the number of osteoclasts which paralleled the decrease in resorption.
Since the decrease in the number of osteoclasts could result from fusion of smaller osteoclasts into larger ones, the total number of nuclei in TRAP-positive cells was determined. The decrease in the number of nuclei corresponded to the decrease in the number of osteoclasts, indicating that the reduction in osteoclast number was not due to an effect on fusion. This is consistent with the results of Nakamura et al. (244) who showed that echistatin inhibited resorption through impairment of osteoclast attachment accompanied by a decrease in both actin ring formation and osteoclast spreading. However, in another series of experiments, echistatin in vivo was found to increase the number of morphologically normal osteoclasts but still prevented bone loss in mice with secondary hyperparathyroidism (217). The authors suggested that the reduction in bone
62 turnover resulted from dysfunctional and inefficient osteoclasts, which does not seem to be the case in our experiments. In our resorption studies, Vitaxin was shown not to diminish the ability of osteoclasts to degrade bone.
Our results also indicate that Vitaxin primarily affects small osteoclasts containing <10 nuclei and has no effect on large osteoclasts containing ≥ 10 nuclei. Comparison between osteoclast populations based on their size evolved from the observation that large osteoclasts were prevalent in
RA and Paget's disease. This observation led to the examination of the characteristics of those two groups of osteoclasts. Previous data had shown that large osteoclasts are more actively resorbing than small osteoclasts and both groups use different methods to regulate their cytoplasmic pH (355,
359). Large osteoclasts were also found to express an average of threefold higher levels of v than small osteoclasts (356). Vitaxin's preferential effect on small osteoclasts could be caused by the increased levels of v on the surface of large osteoclasts, making them less susceptible to detachment by Vitaxin. Alternatively, large osteoclasts might be using an v-independent mechanism for attachment. These data also show that Vitaxin does not interfere with osteoclast multinucleation. In contrast, echistatin was found to inhibit both multinucleation and migration in an osteoblast/osteoclast coculture system (244). These different results can be attributed to the different models and compounds used as well as how both osteoclast formation and multinucleation were measured.
Monitoring osteoclast formation from 2 to 24 h after plating revealed that Vitaxin inhibited initial attachment but did not affect the rate of osteoclast formation. The role of the integrin v in osteoclast formation has long been debated. In vitro, bone marrow macrophages (BMMs) derived from mice with the Glanzmann mutation in the subunit, formed fewer osteoclasts with abnormal structure and function (248). Similar results were reported for osteoclasts differentiated from null
63 BMMs (341). In vivo, null mice, as well as those treated with echistatin in a hyperparathyroidism model, all showed increases in the numbers of osteoclasts while resorption was inhibited. This paradox was solved when it was noticed that levels of M-CSF in the sera of those animals was elevated. This was followed by a series of experiments to elucidate the effect of M-CSF on osteoclasts in vitro. High concentrations of M-CSF added to -/- BMM cultures were shown to rescue the formation of osteoclasts. Both the activation of extracellular signal-regulated kinase as well as the expression of the transcription factor c-Fos, which are essential for osteoclastogenesis, were found to be at lower than normal levels and those levels were restored by M-CSF along with osteoclast formation (4, 217). These results infer an overlap between the signaling pathways of c-
Fms and v during osteoclast formation explaining how M-CSF compensates for the absence of
. A similar compensatory mechanism might be occurring in our mixed culture system, and may explain why Vitaxin did not affect osteoclast formation rates.
Conformational changes occurring in the cytoplasmic domain of integrins are a source of their versatility (192, 197, 360). When integrins change their form from bent to flexed, they are said to be in an activated state. This activation state is responsible for modulation of their affinity for ligands and is accompanied by a more stable interaction with the RGD sequence of matrix proteins, which in turn affects attachment and cell spreading (200, 361). Since vactivation is needed for osteoclast activation, polarization, and resorption, we wanted to find out whether modulating v's structure could alter Vitaxin's effect on osteoclast adhesion. Inactivation and activation of v occur through different mechanisms including binding of divalent cations such as Ca2+, Mn2+, and
Mg2+ (362), activating antibodies (234), low antagonist concentrations (207), and growth factors
(232). M-CSF, known to activate v, enhanced Vitaxin's inhibitory effect; whereas elevation of extracellular Ca2+, known to shift integrins to their inactive state, abolished those effects. The PI3-
64 kinase inhibitor wortmannin abolished M-CSF's effects on the action of Vitaxin, suggesting that they are mediated though the PI3-kinase signaling pathway. These findings suggest that Vitaxin might differentially recognize and bind the activated form of v. In support of Vitaxin preferentially binding to activated v is the observation that a reduction of osteoclast planar area with Vitaxin only occurred when the osteoclasts were pretreated with M-CSF. These data are in agreement with studies showing that LM609 reduced osteoclast spreading only in the presence of
M-CSF (342).
Our study is the first to show that Vitaxin decreases bone resorption but does not affect resorption area per osteoclast or their rate of formation. Rather, the decrease in bone resorption was shown to result from an initial reduction in osteoclast attachment. Furthermore, our data suggests that Vitaxin's inhibitory effects are mediated by the interaction with the activated form of vand that the PI3-kinase signaling pathway is involved in this process. These findings are important as they suggest strategies to enhance the therapeutic potentials of Vitaxin as well as other integrin antagonists currently in clinical use.
65 Chapter 3
The Extracellular Matrix Protein Fibronectin Enhances Osteoclast Activity via Nitric Oxide and Interleukin-1β Mediated Signalling Pathways
This work has been submitted to the journal Arthritis Research and Therapy; Gramoun A, Azizi N,
Sodek J, Heersche JNM, Nakchbandi I, Manolson MF “The extracellular matrix protein fibronectin enhances osteoclast activity via nitric oxide and interleukin-1β mediated signalling pathways”.
Submitted February 5 2010, Manuscript ID: 3404146443529669.
66
3.1 ABSTRACT
Introduction: Osteoclasts are bone resorbing multinucleated cells formed by fusion of mononuclear precursors. The extracellular matrix proteins, fibronectin (FN) vitronectin (VN), and osteopontin
(OPN) are implicated in joint destruction and interact with osteoclasts mainly through integrins. Our aim is therefore to define the role of the matrix in affecting osteoclast activity and function and the integrin involved.
Methods: RAW 264.7 (RAW) cells and primary mouse splenocytes were differentiated into osteoclasts on tissue culture polystyrene (TCP) or osteologic™ slides precoated with 0.01-20 µg/ml
FN, VN, and OPN; concentrations shown not to affect cell proliferation.
Results: Using RAW cells we found that at 96 hours, osteoclast number and multinucleation were decreased on VN and FN compared to OPN and TCP. These results were confirmed using mouse splenocyte-derived osteoclasts. When early differentiation was assessed, VN but not FN decreased cytoplasmic tartrate resistant acid phosphatase activity and pre-osteoclast number at 48 hours. OPN had the opposite effect to FN on osteoclast formation. When RAW cells were differentiated on OPN and treated by FN and OPN, osteoclast number only in the FN treated group was 40-60% lower than the control, while the total number of nuclei was unchanged, suggesting that FN delays osteoclast fusion. In contrast to its inhibitory effect on osteoclastogenesis, FN increased resorption by increasing both osteoclast activity and the percentage of resorbing osteoclasts as indicated by an increase in pit number, sealing zone formation and resorption area per osteoclast. This was accompanied by an increase in nitric oxide (NO) levels and interleukin 1β (IL-1β). IL-production was inhibited using the NO-synthase inhibitor L-NG-monomethyl arginine in a dose dependant manner only on FN indicating a FN-specific cross talk between NO and IL- signalling pathways. 67 Functionally blocking the integrin v3 unexpectedly increased osteoclast number on FN. Analysis of adhesion receptors suggested that osteoclast attachment on FN is mediated through 51.
Conclusions: Our findings show that FN upregulates osteoclast activity despite inhibiting osteoclast formation. FN’s pro-resorptive effects are mediated through NO and IL-signalling. Given the abundance of FN in arthritic joints, our results highlight FN as a novel target for inhibiting bone loss in inflammatory diseases.
68 3.2 INTRODUCTION
Rheumatoid arthritis (RA) is a chronic autoimmune disease affecting a large proportion of the world’s population and is more prevalent amongst women (363). Abnormalities in adaptive immune responses associated with RA result in systemic manifestations and multiple organ involvement. However, synovial inflammation and subsequent progressive bone loss in the affected joints are hallmarks of the symptoms dominating RA (1).The skeletal complications of RA consist of subchondral bone erosions and periarticular osteoporosis at the sites of inflammation. Joint destruction is initiated by invasive pannus formation and immune cell (T-cells and macrophages) activation and infiltration. These events lead to the recruitment of the bone resorbing cells; the osteoclasts (2, 3). Osteoclasts are terminally differentiated multinucleated cells formed by fusion of their hematopoietic progenitors. They are characterized by their unique ability to resorb bone through forming a lysosomal like membrane juxtaposed to bone called the ruffled border (87, 364).
Bone loss in inflammatory diseases is the result of the uncoupling of the two events comprising bone remodelling; bone formation by osteoblasts and bone resorption by osteoclasts.
Both processes are synchronized by inter and intracellular signalling involving hormones, growth factors and attachment receptors binding to the extracellular matrix (ECM) (as reviewed by (365)).
In RA, the prevalence of proinflammatory signals disrupts bone homeostasis shifting the balance towards bone resorption (366, 367). An increase in the number, size and activity of osteoclasts in arthritic joints of patients with inflammatory bone diseases and in animal models have been shown in several reports (366, 368-370).
The three ECM proteins fibronectin (FN), vitronectin (VN) and osteopontin (OPN) are elevated in arthritic joints and are implicated in bone loss through their interactions with osteoclasts.
These matrix proteins are adhesive macro-glycoproteins that bind cells and other matrix molecules
69 mediating cell attachment in an integrin dependant mechanism. While FN and VN are abundant in a soluble plasma form, they can also be found in an insoluble matrix fibril form (329).
Physiologically, both plasma and cellular forms of FN exist in connective tissues. Bone contains only plasma VN (371) and both forms of FN with plasma FN being predominant (15). Even though
FN, VN and OPN have multiple binding domains that confer a wide range of biological functions, the Arginine Glycine Asparagine (RGD) motif represents the common integrin binding domain among the three proteins. In addition to their role as adhesive molecules, VN and, to a larger extent,
FN and OPN are now also recognized as cytokines with autocrine proinflammatory properties. This emerging concept is founded on a growing body of evidence showing a marked elevation of VN and
OPN plasma levels during infections, tumour progression and more frequently in inflammatory arthritis (372-375). Furthermore, several of these studies reported a joint specific increase of all three ECM proteins levels during joint inflammation (6-8, 376, 377).This elevation in matrix protein levels was associated with the induction of inflammatory cytokines and proteolytic enzymes promoting joint injury (377, 378) and in turn resulting in the release of more matrix proteins which further contribute to this vicious cycle.
The evidence implicating ECM proteins in bone loss in arthritic joints suggests that they are directly modulating osteoclast formation and activity. While OPN’s stimulatory effect on osteoclasts has been shown extensively (17, 286, 288, 307, 379, 380), studies on FN and VN have been limited to investigating mature osteoclast attachment (90, 222, 381). Here we hypothesize that FN and VN, similar to OPN, have stimulatory effects on both osteoclast formation and resorption. Thus we undertook this study to compare the effects of FN and VN to the well established effects of OPN.
Our data identify FN as a novel promoter of osteoclast activity. Despite its inhibitory effect on osteoclast formation, FN possesses the unique ability to increase osteoclast resorption through increasing the percentage of activated osteoclasts as well as enhancing their resorptive capacity. 70 FN’s effects were mediated via inducing the proinflammatory mediator; nitric oxide (NO) leading to the downstream activation of interleukin-1 (IL-1), which was shown to be a FN specific activation mechanism.
71 3.3 MATERIALS AND METHODS
3.3.1 Materials
Human FN and recombinant human OPN were purchased from Sigma–Aldrich Ltd. (St. Louis, MO) and human VN was purchased from BD Biosciences (BD Labwares, Franklin Lakes, NJ). Bovine
OPN was provided by Dr. J. Sodek (University of Toronto). The RAW 264.7 (RAW) cell line was obtained from American Type Culture Collection (ATCC, Manassas, VA). Dulbecco’s modified
Eagle’s medium (D-MEM), antibiotics and antimycotics (penicillin/streptomycin, fungizone) and fetal bovine serum (FBS) were obtained from Invitrogen (Carlsbad, CA). Fast red violet LB salt, naphthol AS-MX, Phosphatase Substrate and p-nitrophenol and Arg-Gly-Asp-Ser (RGDS) and Ser-
Asp- Gly- Arg-Gly (SDGRG) and 4'-6-Diamidino-2-phenylindole (DAPI) were obtained from
Sigma–Aldrich Ltd. (St. Louis, MO). VNR149, an Armenian hamster anti-rat v3 antibody, was generously provided by Dr. Su-Yau Mao, MedImmune, Inc. (Gaithersburg, MD). Armenian hamsteranti-mouse 3 antibody and goat polyclonal to Armenian hamster IgG FITC were obtained from Abcam (Cambridge, UK) and rat anti-mouse51 was obtained from US Biological
(Swampscott, MA). Rat anti- mouse CD44 was obtained from SouthernBiotech (Birmingham, AL).
Quantikine® Mouse TRAPTM assay (TRACP 5b mouse), was obtained from Immunodiagnostic
Systems IDS (Boldon, Tyne and Wear, UK). L-NG-monomethyl Arginine (L-NMMA) was
® purchased from Biomol International (Plymouth Meeting, PA). CellTiter 96 AQueous Non-
Radioactive Cell Proliferation Assay (MTS) was purchased from Promega (Madison, WI). Mouse
IL-1 beta/IL-1F2 DuoSet ELISA was purchased from R & D Systems (Minneapolis, MN).
BioCoatTM OsteologicTM bone cell culture system (16 well), pretreated 6- and 12-well plastic
FalconTM tissue culture plates and 96-well FalconTM microtiter plates were purchased from BD
72 Biosciences (BD Labwares, Franklin Lakes, NJ). Griess Reagent Kit for Nitrite Determination and rhodamine phalloidin were purchased and from Molecular Probes Inc. (Eugene, OR).
3.3.2 Immobilizing ECM Proteins on Tissue Culture Plates
96-well tissue culture polystyrene (TCP) plates were pre-coated with 100 μl of 10 μg/ml of
FN, VN or OPN dissolved in phosphate buffered saline (PBS) overnight (O/N) at 4 °C. To increase the amount of proteins physically adsorbed onto the TCP plates, the matrix proteins were incubated for an additional 1hour at 37 °C. After aspiration of the matrix proteins, the wells were subsequently blocked using 1% bovine serum albumin (BSA) in PBS for 1hour at 37 °C in a CO2 incubator to minimize nonspecific binding of serum proteins to TCP. Finally the wells were washed 3x with 100
μl PBS and left in PBS until cells were plated to prevent denaturing of the proteins.
3.3.3 RAW 264.7-Derived Osteoclast Cultures
RAW cells were plated on the ECM coated wells and cultured in DMEM supplemented with
10% FBS, 100 μg/ml penicillin/streptomycin and 0.2 μg/ml fungizone and incubated at 37 °C in 5%
CO2. Osteoclasts were generated using 75 ng/ml receptor activator of NFB ligand (RANKL). After
96 hour of incubation, multinucleated osteoclasts were observed. When early differentiation was studied, cultures were stopped at 48 hours. At the end of the experiments, cultures were fixed using
4% formaldehyde for 5-6 min. To assess early differentiation, the number of mononuclear tartrate resistant acid phosphatase positive (TRAP+) cells (pre-osteoclasts) was counted at 48 hours using bright field microscopy. To determine the effect of ECM proteins on osteoclast formation and multinucleation, the number of TRAP+ cells with ≥ 2 nuclei (osteoclasts) and with > 10 nuclei
(large osteoclasts) were counted at 96 hours using bright field microscopy.
73 For assessment of the effects of soluble ECM proteins on osteoclastogenesis, a similar experimental design was used as described above. Briefly, RAW cells were plated in the presence of
75 ng/ml RANKL. Two hours after cell attachment, soluble FN, VN and OPN (10 μg/ml) were added to the cells. The differentiation experiments were carried out and assessed as described above.
Initial attachment was determined by counting nuclei using DAPI as a nuclear stain. Two hours after RAW cells were plated on the adsorbed ECM proteins, the cultures were washed 2x using PBS and fixed using 4% formaldehyde for 5-6 min. The cultures were subsequently stained using DAPI at a final concentration of 30 ng/ml for 3 minutes. Fluorescent micrographs of cultures were captured using a Leica DMIRE2 microscope and Openlab® imaging system and the number of nuclei/field was subsequently counted.
Assessment of the effects of functionally blocking integrins was carried out as follows. After
72 hours of culture in the presence of 75 ng/ml RANKL on 10 μg/ml FN, VN and OPN, RAW cell cultures were incubated with: RGDS (10, 20 µM) or DRGS (10 µM) peptide, anti-v3 (VNR149) antibody or its IgG isotype at 1 µg/ml for an additional 24 hours. The cultures were fixed and stained at 96 hours and the number of TRAP+ osteoclasts (≥ 2 nuclei) was counted.
74 3.3.4 Splenocyte Derived Osteoclast Cultures
Mice were sacrificed by cervical dislocation and their spleens were dissected and each spleen was placed in a sterile 50 ml Falcon tube with 5ml of -MEM. Spleens were crushed separately through a sterile mesh into 2 ml of supplemented media (-MEM + 10% FBS + 1x antibiotic) in a sterile glass dish. After transferring the cell suspension from the dish into a clean 50 ml Falcon tube, the dish was washed 2x using 1 ml supplemented media. All cell suspensions from different washes were pooled into one tube. To increase cell number, the previous steps were repeated with several spleens and cell suspensions from different animals were combined into the same tube. The cell suspension was centrifuged at 200Gs for 5 min after which the supernatant was discarded and the pellet was resuspended in 20 ml FBS per spleen with 5 ml RBC lysis buffer subsequently added per spleen. RBC lysis buffer was prepared in advance as follows: 1. Solution A: 0.16 M NH4Cl was prepared fresh before each experiment. 2. Solution B: 0.17 M Tris pH 7.65. Both solutions were filter sterilized and 90 ml of Solution A were mixed with 10 ml of Solution B immediately prior to using. After 10 min of adding the RBC lysis buffer to the cell suspension, 10 ml of supplemented media were added per spleen. The cell suspension was centrifuged at 200 Gs for 5 min and were washed 3x with 10 ml PBS. Finally the cells were resuspended in 5 ml supplemented media. Cells were plated in a 96-well plate at a density of 10,000 cells/well. The cells were incubated with 100 ng/ml RANKL and 50 ng/ml macrophage colony stimulating factor (M-CSF) for 3 days. On day 3, the media was changed and cells were incubated with 75 ng/ml RANKL and 10 μg/ml FN, VN and
OPN. Cultures were stopped 48 and 72 hours after matrix proteins were added and TRAP staining was used to identify osteoclasts and pre-osteoclasts.
75 3.3.5 Tartrate-Resistant Acid Phosphatase (TRAP) Staining
TRAP staining was carried out according to the protocol described in BD Biosciences
Technical Bulletin #445. Briefly, cell cultures were washed 3x with PBS, fixed with 4% formaldehyde for 5-6 min and incubated in TRAP staining solution (50 mM acetate buffer, 30 mM sodium tartrate, 0.1 mg/ml Naphtol AS-MX phosphate, 0.1% w/v Triton X-100, and 0.3 mg/ml Fast
Red Violet LB stain) for 10 min until the desired staining intensity was reached. The TRAP staining solution was aspirated and the cells were washed 3x with dH2O.
3.3.6 TRAP Activity Assay
TRAP activity assay was used to assess TRAP enzyme levels in RAW cell cultures. The assay was adapted from the Sigma protocol as described by (382). Briefly, after culture media were aspirated, cultures were washed 2x with PBS. Subsequently, cells were lysed using cold citrate lysis buffer (90 mM sodium citrate and 10 mM sodium chloride pH 4.8) containing 0.1% w/v Triton X-
100 by adding 100 μl of the buffer to the 96-well plate. After cells were resuspended in the lysis buffer by pipetting up and down 3-4 times, a 20 μl aliquot of the cell lysate was incubated with the
50 μl of phosphatase substrate (4 mg/ml) and 50 μl 40 mM tartrate acid buffer (40 mM tartrate acid in citrate buffer pH 3.9) in a 96-well plate for 30 min at 37 °C. The reaction was stopped using 80 μl of cold 2N NaOH. Absorbance was measured at 405 nm using a plate reader, and activity calculated from a standard curve generated using p-nitrophenol standards.
3.3.7 Cell Viability Assay
The CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega) was used to measure viability and proliferation of RAW cells on the different matrices according to the manufacturer’s instructions. Briefly, 20 μl of the CellTiter 96 AQueous One Solution Reagent was 76 added to 100 μl culture media for 2h at 37 ° C in a humidified, 5% CO2 incubator and colourimetric changes were measured at 409 nm.
3.3.8 Secreted TRAP5b Activity Assay
The mouse TRAPTM Assay was carried out according the manufacturer’s instructions. To first adsorb the primary antibody to the plates, 100 μl of mouse anti-TRAP antibody was added to each well with continuous shaking (950 rpm) for 60 min at room temperature (RT). The wells were subsequently washed 4x using 300 μl of washing buffer and 100 ul of calibration solution and controls were added to each well. Seventy five μl of 0.9% NaCl was mixed with 25 μl of conditioned media samples and was also added to the wells. Release agent (25 μl) was subsequently added to all wells and the samples was incubated in the microplate wells while shaking (950 rpm) for 60 min at RT. After the incubation period, the washing step was repeated as described previously and 100 μl freshly prepared substrate solution was added to each wells. The microplate was covered and incubated for 2 h at 37 °C. The reaction was stopped with 25 ul 2 N NaOH and absorbance was measured at 405 nm.
3.3.9 Nitrite and Nitrate Measurements
Conditioned media from RAW derived osteoclast cultures on FN, VN and OPN were analyzed for nitrite and nitrate levels at 24 and 48 h using Griess reagent (Molecular Probes Inc) as an indirect method of measuring NO levels according to the manufacturer’s protocol. Briefly, Griess reagent (100 μl of sulfanilic acid, 100 μl of N-(1-naphthyl) ethylenediamine dihydrochloride) was incubated with 100 μl of conditioned media in a microplate for 30 minutes at RT. A standard curve was generated using 1 mM sodium nitrite in deionized water. Absorbance was measured at 548 nm.
77 3.3.10 Resorption Studies
Osteoclastic resorption was studied using 16-well osteologic slides (BD Biosciences) using a similar experimental design and culture conditions described above for osteoclast differentiation.
Briefly, ECM matrix proteins (FN, VN and OPN) at 20 μg/ml were physically adsorbed onto the hydroxyapatite (HA) coated slides and RAW cells were plated and differentiated onto the coated discs for 72 hours in the presence of 75 ng/ml RANKL. To determine both resorption and osteoclast number for the same experiment, two slides were used in parallel for each experiment and all procedures were identical for both slides. Following the 72-h culture period, cells on one slide were detached from the surface of osteologic slides by adding 200 µl of bleach solution (6% NaOCl,
5.2% NaCl) to each well before removing media and agitating for 5 min. Wells were washed 2x with 150 µl dH2O and then examined under the light microscope to ensure complete removal of cells. The contrast between non-resorbed areas and resorption pits was visualized using von Kossa staining as described in (383). The perimeters of the resorption pits were traced manually and their area was measured using Leica DMIRE2 microscope and Openlab imaging system. Cells on the second slide were fixed and stained for TRAP activity at the same time point. The number of osteoclasts was determined using bright field microscopy. To assess actin ring formation at the same time, rhodamine phalloidin at a titre 1:50 was added after TRAP staining. Images were acquired using Leica DMIRE2 microscope and Openlab imaging system. The total number of osteoclasts and actin rings were determined.
3.3.11 IL-1β ELISA
Conditioned media samples from RAW derived osteoclast cultures on FN, VN and OPN (10
μg/ml) were analyzed for IL1β levels at 24 and 48 h using the Quantikine
78 Mouse IL-1_/IL-1F2 Immunoassay (R&D Research). The Assay was carried out exactly according to the manufacturer’s instructions.
3.3.12 Flow Cytometry Analysis of Integrin Expression
Surface expression of integrins v3 and 51 in cultures differentiated on FN, VN and
OPN (10 μg/ml) were analyzed using flow cytometry. RAW cells were plated and differentiated on
ECM protein coated 60 mm culture dishes in the presence of 75 ng/ml RANKL. At 72 hours, the cells were washed 2x using PBS –Ca-Mg before they were incubated with 0.06% EDTA for 45 min at 37 °C in a humidified, 5% CO2 incubator. The cells were subsequently scrapped and centrifuged for 5 min at 200 Gs and 4 °C. The cells were resuspended and incubated in a blocking solution (0.5
% BSA in PBS). After a 1 hour incubation period, the cells were incubated with the primary antibodies (hamster anti-mouse 3 and rat anti-mouse 51) or their IgG isotypes at a titre 1:20 for
1 hour on ice. This was followed by a 30 min incubation period with the secondary antibodies on ice. Finally the cells were washed and analyzed using an Epics Altra Beckman Coulter flow cytometer.
3.3.13 Generation of FN Conditional Knockout Mice
3.3.13.1 Transgenic Mice
Transgenic mice carrying loxP-flanked (floxed) FN (FN-fl/fl) have been reported by Sakai et al. (384). Mice possessing a construct of the inducible Mx promoter driving cre recombinase expression were used to delete FN in hepatocytes hematopoietic cells in mice homozygote for the floxed FN gene (15, 384). Induction of Mx was performed as described (384). Success of the
79 deletion was confirmed by measuring the level of plasma FN by ELISA as previously described
(328).
3.3.13.2 Histomorphometry
Tibiae were fixed and embedded in methylmethacrylate. Masson-Goldner staining was performed as follows: sections were deplasticized, treated with hematoxylin (Gill II), acid fuchsin/ponceau xylidine, and phosphomolybdic acid/orange G to stain the cells and osteoid. Light green was used at the end to stain the mineralized matrix. Primary cancellous bone was defined as the 120 μm band below the growth plate. Cancellous bone was defined as the remaining trabecular area that extends down 2 mm. The ASBMR nomenclature was used (385). The following measurements of the proximal tibia were performed using ImageJ: bone surface (BS), osteoclast number (Oc.N), and osteoclast surface (Oc.S).
3.3.14 Statistics
Statistical evaluation was carried out using SPSS 12.0 for Windows using one way analysis of variance (ANOVA) and Dunnette T3 test. P-values less than 0.05 were considered statistically significant.
80 3.4 RESULTS
3.4.1 FN Reduces Osteoclast Formation without Affecting RAW Cell Proliferation or Initial
Attachment
Several extracellular proteins and cytokines contribute to the process of osteoclastogenesis.
The aim of this study was to ask if the matrix proteins FN, VN and OPN affect osteoclast formation.
RAW cells were differentiated on increasing concentrations of FN, VN and OPN (0.1, 1, 10 and 20
µg/ml) physically adsorbed on TCP dishes. After 96 hours, fewer osteoclasts were formed on FN and VN than on OPN and TCP (figure 3.1A). These effects were not due to differences in initial attachment as the number of cells attached to FN, VN, OPN and TCP 2 hours after initial plating were not significantly different (figure 3.1B). However there seemed to a non-significant trend of enhanced attachment on FN and VN compared to TCP. The differences in osteoclast number were also not due to differences in viability or proliferation since all three ECM proteins did not alter the metabolic activity of RAW cells as shown by the MTS assay (figure 3.1C). Lastly, there were also no differences in total protein content up to 96 hours on 10, 20 and 50 µg/ml of adsorbed matrix proteins (data not shown). These data show that both FN and VN reduce osteoclastogenesis without affecting initial attachment or viability.
3.4.2 FN Inhibits Pre-osteoclast Fusion and/or Migration but Not Pre-osteoclast Recruitment
We next asked which step of osteoclastogenesis was affected by the matrix proteins.
Osteoclast formation is composed of two sequential steps. The first step is the recruitment and differentiation of hematopoietic cells that do not stain for TRAP activity (TRAP-) cells into mononuclear cells which stain positive for TRAP activity (TRAP+) known as pre-osteoclasts or
81 prefusion osteoclasts. Subsequently, pre-osteoclasts fuse into mature multinucleated osteoclasts. To assess both steps, early (48 and 72 hours) and late (96 hours) time points were studied. Cytoplasmic
TRAP has been shown to be proportional to the total number of TRAP+ cells regardless of their size when normalized to the number of nuclei (356). At 48 and 72 hours, cytoplasmic TRAP was not different between FN and TCP, indicating that FN did not alter the rate of RAW cell differentiation
(figure 3.2A). Even though, TRAP levels on VN and OPN were reduced at these time points.
Nonetheless, at 96 hours there were no differences between all groups (figure 3.2A). This indicates that the amount of TRAP was not affected by the matrix proteins despite initial differences at early time points and suggests that ECM VN and OPN delay but do not inhibit pre-osteoclast recruitment.
We next determined the number of pre-osteoclasts (mononuclear TRAP+ cells), osteoclasts
(TRAP+ cells with ≥ 2 nuclei) and large osteoclasts (TRAP+ cells with ≥ 10 nuclei) at 48 and 96 hours after induction of differentiation. In agreement with cytoplasmic TRAP levels at 48 hours, VN and OPN decreased pre-osteoclast numbers whereas FN did not affect these numbers compared to
TCP (figure 3.2B). However, the number of osteoclasts and large osteoclasts was significantly lower on FN and VN than on OPN and TCP at 96 hours (figure 3.2C). Due to the large number of TRAP+ mononuclear cells at 96 hours, counting cells was not feasible. Therefore cytoplasmic TRAP levels were used to assess osteoclastogenesis at this time point.
Soluble FN, VN and OPN at 10 µg/ml had the same effects on RAW cell differentiation as the physically adsorbed proteins (figure 3.3A, B vs. figure 3.2B and C). To confirm the results obtained with RAW cells in primary cell cultures, these experiments were repeated using mouse spleen cells. Mouse spleen cells were plated on TCP dishes in the presence M-CSF for 3 days. On day 3, both RANKL and M-CSF were used to differentiate attached cells in the presence of soluble
FN, VN or OPN. FN and VN reduced osteoclast numbers and multinucleation at 48 and 72 hours
82 compared to OPN and TCP (figure 3.3C, D). These results agree with the data obtained from both the physically adsorbed and soluble ECM proteins in RAW cells.
Based on these results, it appears that FN inhibits osteoclast formation by diminishing pre- osteoclast fusion and/or migration. This is in contrast with OPN which increases the number of osteoclasts by enhancing what appears to be the multinucleation of formed osteoclasts. We therefore asked if soluble FN added during differentiation can inhibit the formation of osteoclasts on OPN.
RAW cells were plated on OPN (10 µg/ml) coated dishes and primed with RANKL. At 2 and 48 hours following cell plating, soluble FN and OPN (20 µg/ml) were added, and then cultures were stopped at 96 hours. Addition of FN at 2 and 48 hours resulted in 40-60 % reduction in osteoclast numbers compared to cells differentiated only on adsorbed OPN (control), but did not affect the total number of nuclei at either time points. In contrast, addition of soluble OPN affected neither osteoclast numbers nor total number of nuclei (figure 3.4 A, B). These results show that FN and
OPN have opposite effects on the steps of osteoclastogenesis and that FN can inhibit OPNs stimulatory effects.
3.4.3 Assessment of Osteoclast Formation in an FN Conditional Knockout Mouse Model
Since FN total knockout mice can not be studied since complete deletion of FN is embryonically lethal and causes death at E8.5 due to defects in cell migration and mesoderm formation (386). Therefore, to study the effects of FN on osteoclasts in vivo, conditional knockouts of FN in hepatocytes were examined. Deletion of plasma FN produced in hepatocytes and hematopoietic stem cells (and hence in osteoclasts) using the inducible Mx promoter was chosen as this significantly reduced ECM FN levels (15). The osteoclast histomorphometric parameters
Oc.N/BS and Oc.S/BS (table 3.1) show a trend for higher osteoclast number and larger cell area in the Mx conditional knockout line; however the results are not significant and thus inconclusive (p= 83 0.14 and p=0.10 respectively). Since FN was not completely deleted in osteoblasts, some FN might have been available to osteoclasts. A complete FN knockout in all bone marrow cells might be needed to reach a conclusive answer about FN’s effect on osteoclasts in vivo. Alternatively, a more rapid bone remodeling system, such as the hind limb unloading model or an ovariectomy model might be necessary to produce a phenotype, as was the case with the OPN knockout mice (307, 309,
372).
3.4.4 FN Increases Resorption by Increasing Both the Resorptive Activity per Osteoclast and the Percentage of Resorbing Osteoclasts
We next investigated the effect of ECM proteins on osteoclast activity. At 72 hours, FN increased resorption by 60%, 70% and 87% compared to OPN, VN and uncoated HA controls respectively (figure 3.5A). The same trend was seen when the number of pits was counted (figure
3.5C). While osteoclast number on FN and VN remained ~ 40% diminished compared to OPN and
HA (figure 3.5B), this decrease in osteoclast numbers was accompanied by a significant increase in resorptive activity per osteoclast (figure 3.5D). Secreted TRAP5b in the conditioned medium, another indicator for osteoclast activity, was also elevated on FN (figure 3.5E).
As NO affects osteoclast resorption and is implicated in arthritic joint destruction (387-391), we asked if FN’s stimulatory effect on osteoclast resorption correlated with NO levels. NO was elevated on FN but not on VN or OPN (figure 3.5F), supporting a relationship between NO synthesis and FN’s stimulatory effects on osteoclast activity.
Osteoclast activity was also assessed by evaluating sealing zone formation on ECM proteins.
The sealing zone is an adhesion structure only seen in an actively resorbing osteoclast and can be visualized by staining for actin rings using rhodamine phalloidin. We found a trend for increased sealing zone formation on all ECM proteins but this increase was only significant on FN where 84 multiple actin rings per osteoclast could be seen (figure 3.6A and 3.6B). The increase in sealing zones on FN indicates a higher number of actively resorbing osteoclasts and supports the data on increased resorption by FN.
3.4.5 FN Increases IL-1β in a NO Dependant Manner
Levels of NO and IL-1β were assessed as both NO and IL-1β are pro-inflammatory mediators, affected by ECM proteins, with cross talking signalling pathways (356, 392, 393). NO in conditioned media was elevated on FN at 24 and 48 hours at a level higher than VN and OPN
(figure 3.7A). Only FN caused an increase of IL-1β at 48 hours in the same conditioned media
(figure 3.7B). To test the interaction between NO and IL-1β the NO-synthase inhibitor L-NMMA
(0.1 and 0.5 µM) was added at 24 hours of culture and the cells were incubated with the inhibitor for an additional 24 hours before the levels of NO and IL-1β were measured on FN and TCP. L-NMMA decreased NO levels in a dose dependant fashion on FN and TCP (figure 3.7C) and inhibited IL-1β production on FN but not on TCP (figure 3.7D). This suggests that FN induction of IL-1β is NO dependent.
3.4.6 Blocking v3 and 51 Has Different Effects on Osteoclast Number
The adhesion receptor v3 has a central role in osteoclast attachment and function; blocking v3 compromises osteoclast attachment and decreases their number (383) (reviewed by
(5)). The effects of blocking v3 on osteoclasts pre-formed on FN, VN and OPN was examined using two methods. The first technique involved non specifically blocking all integrins on the surface of osteoclasts differentiated for 72 hours on the ECM proteins using an RGDS peptide (10-
20 µM) or a reverse DGRS peptide (10 µM) as a control with osteoclast numbers determined after a
85 24-hour incubation period. As expected, osteoclast numbers decreased on VN, OPN and TCP.
Interestingly, they were increased on FN in response to RGD blockade in a dose dependant manner
(figure 3.8A). These results were confirmed using the specific v3 blocking antibody VNR149 (1
µg/ml) which produced similar results. VNR149 treatment caused a 65% increase in osteoclast number on FN compared to its IgG isotype control. A 60%, 40% and 20% decrease in osteoclast numbers on VN, OPN, and TCP was seen respectively with VNR149 compared to the IgG control
(figure 3.8B). When surface expression of v3 was measured using flow cytometry, cells on FN were found to have the highest expression (figure 3.8C). The increase in osteoclast number on FN by blocking v3 could be related to the integrin’s upregulation on FN.
In order to determine which integrin modulates interaction between osteoclasts and FN, a series of blocking antibodies was used. In contrast to blocking v3, blocking 51 on FN decreased osteoclast number while no change on TCP was detected (figure 3.9A). The anti-CD44 antibody had no effect on osteoclasts on both FN and TCP. Assessing simultaneous expression of 3 and 51 on cells cultured on FN revealed a ~20% decrease in 3 positive cells (figure 3.9B) and a
5% increase in 51 positive cells (figure 3.9C) on FN compared to TCP. Nevertheless, the expression per cell for both 3 and 51 was elevated on FN (figure 3.9D). These results suggest that attachment of osteoclasts to FN is mediated through 51and demonstrate that FN alters the integrin profile of osteoclasts.
86
Figure 3.1 FN and VN decrease osteoclastogenesis compared to OPN without affecting initial attachment or proliferation. (A) RAW cells were plated on TCP precoated with FN, VN and OPN at
0.1, 1, 10 and 20 µg/ml. RAW cells were differentiated into osteoclasts for 96 hours in the presence of 75 ng/ml RANKL. The cultures were fixed and stained for TRAP activity and the number of
TRAP+ osteoclasts was counted. Each data point represents the pooled results from four wells per treatment and is expressed per well (means ± SD). (B) RAW cells were plated on TCP precoated with 10 µg/ml FN, VN and OPN and cultured in the presence of RANKL (75 ng/ml) for 2 hours.
Cells were stained with DAPI and the total number of nuclei in eight fields per well was counted.
Each data point represents the pooled results from three dishes per treatment and is expressed per field (means ± SD). (C) RAW cells were plated on TCP dishes precoated with FN, VN and OPN at
10 µg/ml. RAW cells were cultured in the presence of RANKL (75 ng/ml) for 24, 48 and 72 hours.
87 Cell proliferation was indirectly measured at 24, 48 and 72 hours using an MTS assay. Each data point represents the pooled results from 6 wells per treatment and is expressed per well (means ±
SD). *P<0.05 and **P<0.01 versus the TCP group. Similar results were obtained in three separate experiments.
88
Figure 3.2 VN delays early pre-osteoclast formation while both VN and FN decrease osteoclast multinucleation. (A) RAW cells were plated on TCP dishes precoated with 10 µg/ml FN, VN and
OPN and cells were differentiated into osteoclasts with 75 ng/ml RANKL. Experiments were stopped at 48, 72 hours and 96 hours and cytoplasmic TRAP activity was measured using an enzymatic TRAP assay. Each data point represents the pooled results from six wells per treatment and is expressed per well (means ± SD). (B, C) RAW cells were plated on TCP dishes precoated with FN, VN and OPN at 10 µg/ml. The cells were differentiated into osteoclasts with 75 ng/ml
RANKL for 48 hours and 96 hours. TRAP+ pre-osteoclasts (mononuclear cells) at 48 hours and osteoclasts at 96 hours were counted. Each data point represents the pooled results from four dishes per treatment and is expressed per well (means ± SD). *, ¥, # P<0.05 versus the respective TCP control group. Similar results were obtained in three separate experiments.
89
Figure 3.3 Matrix proteins have similar effects on the formation of osteoclasts derived from either splenocytes or RAW cells. RAW cells were plated in the presence of 75 ng/ml RANKL for 2 hours before 10 µg/ml of soluble FN, VN and OPN were added to the cultures. RAW cells were differentiated into osteoclasts for 48 (A) and 96 (B) hours after which the number of TRAP+ mononuclear and multinucleated cells were counted. Isolated splenocytes were incubated with 100 ng/ml RANKL and 50 ng/ml MCSF for 3 days. On day 3, the media was changed and cells were incubated with 75 ng/ml RANKL and 10 μg/ml FN, VN and OPN. Cultures were stopped 48 (C) and 72 (D) hours after matrix proteins were added and mononuclear pre-osteoclasts and osteoclasts were counted. Each data point represents the pooled results from four dishes per treatment and is expressed per well (means ± SD). *, ¥, # P<0.05 versus the respective TCP control group. Similar results were obtained in three separate experiments.
90
Figure 3.4 Soluble FN decreases migration/fusion of pre-osteoclasts formed on physically adsorbed
OPN. (A) RAW cells were plated on TCP dishes precoated with 10 µg/ml of OPN in the presence of
75 ng/ml RANKL. Media supplemented with or without 20 µg/ml FN and OPN was added to the cultures at 2 and 48 hours after the cells were plated. At 96 hours the number of TRAP+ osteoclasts was counted. Each data point represents the pooled results from four dishes per treatment and is expressed per well (means ± SD). (B) DAPI was subsequently used as a nuclear stain and the total number of nuclei in eight fields per well was counted using epi-fluorescent microscopy. Each data point represents the pooled results from four dishes per treatment and is expressed per field (means
± SD). *P<0.05 and **P<0.01 versus the control group. Similar results were obtained in three separate experiments.
91
Figure 3.5 FN increases resorptive parameters and NO production. RAW cells were plated on osteologic (hydroxyapatite coated) slides precoated with FN, VN and OPN at 20 µg/ml. RAW cells were differentiated into osteoclasts for 72 hours in the presence of 75 ng/ml RANKL. To determine both resorption and osteoclast number for the same experiment, two slides were used in parallel for each experiment and all experimental procedures for both slides were identical. On one slide the cells were detached and non-resorbed areas were visualized using von Kossa staining. The perimeter of the resorption pits was traced manually and the area was measured using Openlab software system (A). The number of pits was also counted (C). On the second slide the cultures were fixed and stained for TRAP and the number of TRAP+ osteoclasts was counted (B). The resorption area/osteoclast was calculated from the previously obtained measurements (D). The conditioned
92 media from both slides was collected and a TRAP5b capture ELISA and Griess reagent were used to measure levels of the secreted TRAP5b enzyme (E) and NO nitrites (F) respectively. Each data point represents the pooled results from four discs per treatment and is expressed per well (means ±
SD). *P<0.05 and **P<0.01 versus the HA group. Similar results were obtained in four separate experiments.
93 94 Figure 3.6 Osteoclasts on FN coated osteologic discs have more sealing zones. RAW cells were plated on osteologic (hydroxyapatite coated) slides precoated with FN, VN and OPN at 20 µg/ml.
RAW cells were differentiated into osteoclasts for 72 hours in the presence of 75 ng/ml RANKL.
The cultures were fixed, permeabilized and stained for F-actin using rhodamine phalloidin (red) to visualize actin rings and DAPI (blue) for nuclei (A). The number of actin rings and total number of osteoclasts in eight fields per treatment group were counted. The percentage of osteoclasts exhibiting actin rings per field was calculated (B). Each data point represents the pooled results from four discs per treatment and is expressed per well (means ± SEM). *P<0.05 versus the HA group.
Similar results were obtained in three separate experiments.
95
Figure 3.7 IL-1β and NO production is increased on FN. Inhibition of IL-1β using the NO synthase inhibitor L-NMMA suggests that NO is upstream of IL-1β. RAW cells were cultured with 75 ng/ml
RANKL in pre-coated dishes with 10 µg/ml FN, VN and OPN. The media was changed after 24 hours and the NO-synthase inhibitor L-NG-monomethyl Arginine (L-NMMA) was added (0.1, 0.5
µM) to the FN and uncoated controls for an additional 24 hours. NO was measured indirectly using
Griess reagent at 24 and 48 hours (A, C). IL-1β was measured by ELISA at 24 and 48 hours (B,D).
Each data point represents the pooled results from five dishes per treatment and is expressed per well (means ± SD). (A,B) *P<0.05 and **P<0.01 versus TCP control group. (C, D) *P<0.05 and
**P<0.01 versus the respective untreated control. Similar results were obtained in three separate experiments.
96
Figure 3.8 Exclusive blockade of v3 in osteoclasts differentiated on FN increases osteoclast number. The surface expression of v3 in osteoclasts is the highest on FN.
RAW cells were plated on TCP dishes precoated with FN, VN and OPN at 10 µg/ml and differentiated in the presence of 75 ng/ml RANKL. After 72 hours the cells were incubated with the following: RGDS (10, 20 µM) or DRGS (10 µM) peptide (A), anti-v3 (VNR149) antibody or its
IgG isotype control at 1 µg/ml (B). The cultures were fixed and stained for TRAP and the number of
TRAP+ osteoclasts was counted. Alternatively, after 72 hours the cells were detached and immuno- stained using an anti-v3 antibody and subsequently with a FITC-labeled 2nd antibody. Flow cytometry was used to measure the fluorescence intensity (C). Each data point represents the pooled results from four dishes per treatment and is expressed per well (means ± SD). (A) *P<0.05 and
**P<0.01 versus the respective untreated control. (B) *P<0.05 and **P<0.01 versus the respective
IgG control. (C) *P<0.05 versus the TCP control group. Similar results were obtained in three separate experiments. 97
Figure 3.9 Blocking 5but not v3or CD44, decreases osteoclast number on FN and its expression is highest on FN. RAW cells were plated on TCP dishes precoated with FN at 10 µg/ml and differentiated in the presence of 75 ng/ml RANKL. After 72 hours the cells were incubated with the following: anti-v3anti-3v3/3IgG isotype, anti-CD44, and anti-5at 1 µg/ml. The
98 cultures were fixed and stained for TRAP and the number of TRAP+ osteoclasts was counted (A).
*P<0.05 and **P<0.01 versus the respective control. Similar results were obtained in three separate experiments. Alternatively, after 72 hours the cells were detached and immuno-stained with an anti-
3 and antibody and anti-5or their IgG isotype control. Subsequently the cells were incubated with FITC and Cy5-labeled 2nd antibodies. Flow cytometry was used to measure the percentage of
3 (B), 5(C), both3 and 5 (D) positive cells. 3 and 5 mean fluorescence intensity was also determined (E). Each data point represents the pooled results from three different experiments with three dishes per treatment and is expressed per experiment (means ± SD). *P<0.05 and **P<0.01 versus the TCP control.
99
Table 3.1. Histomorphometric osteoclast parameters in the FN conditional knockout (cKO) Mx mouse line.
Parameter CT (n=10) cKO (n=10) p value Oc.S/BS % 2.5 1.2 3.3 0.8 0.10 Oc.N/mm BS 0.38 0.19 0.50 0.16 0.14
100 3.5 DISCUSSION
The prevalence of ECM proteins during inflammatory bone diseases and their effects on these conditions implicate them as osteoclastogenic promoters. Nonetheless, their effect on osteoclastogenesis has not been thoroughly investigated. To delineate the role of matrix proteins in osteoclast formation, we studied the effects of physically adsorbed FN and VN on differentiating
RAW cells and compared these effects with these on OPN. OPN was chosen as a reference for the other two proteins since OPN’s effects on osteoclast function have been extensively studied (17,
165, 307, 379). FN, VN and OPN will each be discussed separately in the following sections.
Physically adsorbed or soluble FN had the most pronounced inhibitory effect on osteoclastogenesis in both RAW and primary splenocytes cultures. This result is in accord with the observation that adhesive matrix proteins, in particular FN, increase cell growth, proliferation and spreading while reducing cell differentiation (394, 395).
To better understand FN’s effects on osteoclastogenesis, we assessed two time points that coincide with the two main steps of osteoclast formation. At 48 hours the commitment of hematopoietic precursors into pre-osteoclasts was assessed while at 72 and 96 hours the fusion of these pre-osteoclasts into multinucleated cells was examined. Pre-osteoclast commitment on FN was similar to that on TCP and higher than that on OPN and VN at 48 hours. At 96 hours, osteoclast number and their multinucleation were diminished while cytoplasmic TRAP levels were unaffected.
These results suggest that equal number of prefusion osteoclasts were present on FN, OPN and TCP but only a small percentage of these committed cells fused to form osteoclasts on FN. These findings led us to hypothesize that FN inhibits pre-osteoclast fusion and/or migration while OPN promotes these processes. To test this, we asked whether FN could inhibit OPN’s stimulatory effect on multinucleation. Treating cells differentiated on physically absorbed OPN with soluble FN 101 resulted in a 60% decrease in osteoclast number without affecting the total number of nuclei. These results further support the hypothesis that FN’s inhibitory effect on osteoclast formation is via reducing pre-osteoclast fusion and/or migration.
To assess FN’s effect on osteoclasts in vivo, histomorphometrical measurements were performed on conditional knockout mice in which FN was deleted in the circulation and hematopoietic cells using the Mx promoter to drive cre expression. These mice have a significant decrease in the amount of FN in the bone matrix (15), while mice in which FN was deleted in osteoblasts had normal amounts of FN in the bone matrix (15). This is surprising at first glance, since osteoblasts lay down the matrix containing FN which later becomes bone after mineralization.
Considering that circulating FN infiltrates the bone matrix, it was not totally surprising to see an effect on FN content in bone when it was deleted. There was a trend towards an increase in osteoclast number and osteoclast surface area between the FN conditional knockouts and their controls; however, the difference between the two groups was not statistically significant. These results point at a possible effect of FN on osteoclast formation and function in vivo that could be partially masked in this mouse model due to the presence of FN originating from osteoblasts. This is however an interesting effect that needs to be further explored in a double FN knockout in both osteoblasts and hematopoietic cells. Additionally, FN deletion needs to be examined in high bone turnover states where osteoclast numbers are markedly elevated, since it might reveal significant differences.
As bone erosion and periarticular osteoporosis responsible for joint destruction in RA require increased activation of osteoclastic resorption, we next examined how FN affects resorption parameters. Despite its inhibitory effect on osteoclast formation, FN increased total resorption, resorption area per osteoclast, pit number, TRAP5b and NO levels. The increase in total area of resorption on FN is the result of the increase in both the number of resorption pits and the area 102 resorbed per osteoclast. The increase in pit number suggests that FN stimulates osteoclast migration on HA.
TRAP5b is used as a bone turnover marker (396); however, it is controversial whether it correlates with osteoclast number or activity (127-131). Here the induction of TRAP5b on FN correlated with the increase in resorptive activity, not the number of mature osteoclasts. We propose that on FN the correlation is between number of the mononuclear pre-osteoclasts which may be contributing to the elevated TRAP5b levels and potentially to bone resorption as well as the increased activity of osteoclasts.
Another important parameter reflecting osteoclast activity is sealing zone formation.
Although all three matrices promoted sealing zone formation, the highest percentage was seen on
FN indicating it has the highest percentage of active osteoclasts. Furthermore, osteoclasts on FN often had more than one sealing zone suggesting that they possessed more efficient resorbing machinery and explaining the increased resorptive activity per osteoclast we reported. Finally, the proinflammatory mediator NO was elevated in conditioned media only on FN. These data indicate that FN is a proinflammatory molecule capable of activating osteoclasts and enhancing their resorptive efficiency and that NO is involved in this process.
One possible mechanism for the induction of osteoclasts on HA could be related to conformational changes of adsorbed FN. FN can be found in a dimeric globular form that can undergo conformational changes to form a polymeric fibrillar network exposing certain cryptic domains in the molecule under specific conditions (397, 398). Although this process of FN activation is mainly cell/integrin mediated, it was also seen in vitro in the absence of cells when FN was adsorbed onto HA (399, 400). However, the role of FN fibrillogenesis and FN’s cryptic motifs in osteoclast function are not known. Another possible mechanism for increased osteoclast activation on FN involves extracellular pH regulation. Extracellular pH of osteoclasts on FN was 103 indeed lower than cultures on TCP, VN and OPN (Gramoun et al., submitted). As low extracellular pH increases osteoclast resorption in vitro, the low pH resulting from FN offers plausible mechanism for the enhancement of osteoclast activity (107, 108, 401, 402).
As both NO and IL-1are key factors implicated in joint destruction, and as IL-1induces
NO in other cells, we asked whether there is a correlation between their pattern of induction. IL-
1was also a prime candidate for investigation as FN regulates the release of IL-1by rat macrophages (403) and has the second strongest stimulatory effects on IL-1after collagen (404).
Furthermore, IL-1was shown to stimulate osteoclast resorption (356). Here we show that NO levels were elevated on FN at both 24 and 48 hours, however, only at 24 hours was it significantly higher than VN, OPN and TCP. IL-1levels, on the other hand, peaked at 48 hours while no changes were detected on any of the other substrates at either time points. The sustained elevation of
NO at 24 and 48 hours followed by the increase in IL-1suggests that the NO signalling cascade is upstream of IL-1This is not consistent with other studies showing that in chondrocytes and synoviocytes, IL1induces NO (405, 406). Furthermore, avian osteoclast like cell cultures also had a marked increase in NO through iNOS when treated by several interleukins and proinflammatory agents (407). These contradictory findings could be reconciled on the basis that both NO and IL-
1are part of a positive feedback loop that auto-amplifies inflammation and induces the production of each other. Importantly, we show that the general NO inhibitor L-NMMA decreased both NO and
IL-1levels in a dose dependent manner only on FN. This corroborates our hypothesis that IL-
1induction is downstream of NO and that this effect is mediated by FN.
An interesting finding of this study was the stimulatory effect of integrin blockers on osteoclast formation on FN. Integrin blockers, such as integrin specific antibodies, peptides and disintegrins, allosterically block the interaction between the integrin v3 and its ligand thus
104 impairing osteoclast attachment and/or migration (reviewed in Nakamura (5)). When osteoclast cultures were treated with anti-v3 antibody or with an RGD peptide, an increase in osteoclast number was only seen on FN while the expected inhibitory effect was seen on VN, OPN and TCP.
To address this unforeseen result, we asked which receptors were involved in the interaction between FN and osteoclasts mediating their attachment using a series of receptor specific antibodies.
Consistent with our previous experiments, an anti-3 antibody increased osteoclast number on FN, whereas osteoclast attachment was diminished in the presence of an anti-51 antibody on FN, but not on TCP. The v3 and 51 are known to interact with the RGD motif on FN while CD44 binds to FN’s heparin-binding domain (408). These data suggest that osteoclast attachment on FN is mediated by the integrin 51 and that this interaction is substrate specific to FN. CD44 did not affect osteoclast number on either FN or TCP. Similar results by Hofmann et al. show that attachment of the pre-osteoclastic leukemia cell line FLG 29.1 on FN was not mediated by v3 and that only the 51 blocking antibody affected cell adhesion. These effects were substrate specific since the anti-51 did not affect cell adhesion on VN (409). Furthermore, Hu et al. showed that rat osteoclast adhesion on FN was inhibited using an anti-1 antibody but not with an anti-3 antibody
(381).
To explain the stimulatory effect of FN we propose a model based on the competition between two major signalling pathways in osteoclast; v3 and the CD44 signalling pathways. FN is a ligand for both v3 and CD44 (408), both present in osteoclasts. CD44 is a surface receptor mediating pre-osteoclast and macrophage fusion (52, 53) while v3 mediates mature osteoclast attachment. We show that osteoclast attachment on FN is mediated through 51 and not v3.
Furthermore, our data suggests that FNs interaction with v3 inhibits osteoclast formation by preventing pre-osteoclast fusion. Therefore, we propose that when the FN/v3 interaction is 105 blocked, stimulatory signals from CD44 promote the fusion of pre-osteoclasts and increases osteoclast formation.
Because of the distinct functions of the integrins v3 and 51 on FN, we quantified their expression levels. Although the percentage of v3 positive cells on FN cultured cells was 20% lower that TCP, v3 and 51 expression per osteoclast were elevated. This agrees with Hofmann et al.’s results showing that a pre-osteoclastic leukemia cell line growing on FN had increased v3 expression (409). Upregulation of v3 could also explain the difference in behaviour of osteoclasts on FN when v3 is blocked. An increase in v3 could increase pre-osteoclast attachment and hence decrease their mobility and subsequent fusion. Blocking this interaction could decrease their attachment promoting motility and pre-osteoclast fusion.
The results from the matrix protein VN indicate that the protein acts mainly as an adhesive molecule and has no cytokine-like effects. This finding does not confirm our initial hypothesis that elevated VN levels in joints and plasma of arthritis patients contributes to joint damage through increasing osteoclast resorption. VN decreased osteoclast number by reducing both pre-osteoclast recruitment and subsequent fusion. VN’s effects on osteoclast activity were in line with its inhibitory effects on osteoclast differentiation in that it did not stimulate total resorption compared to FN or OPN and only slightly increased pit formation and resorbed area/osteoclast. The slight increase in pit formation and resorption per osteoclast correlates with our observation that VN increases the ratio of osteoclasts with a migratory morphology (Gramoun et al., submitted).
The highly phosphorylated matrix protein OPN is one of the most abundant non-collagenous glycoproteins in the matrix. Not only is OPN an adhesive protein regulating cell adhesion and migration, but it is also a signalling molecule with cytokine like properties that regulates immune responses and tumour progression (373, 410). Many of its diverse effects are related to the degree of
106 its post-translational modifications (PTM) specifically phosphorylation, which differs between tissues (287). The role of OPN and its PTMs is influential in bone homeostasis and osteoclast adhesion, migration and resorption (286, 288, 309, 312). However, OPN’s mechanism of regulating these processes remains controversial. Here we used the most phosphorylated form of OPN from bovine milk. In agreement with our initial hypothesis, we demonstrated that OPN is a positive regulator of osteoclastogenesis when coated on TCP. Even though OPN initially delayed pre- osteoclast recruitment, it stimulated osteoclast formation and promoted multinucleation. These data suggest that OPN can increase pre-osteoclast migration. This is compatible with reports from several groups demonstrating that OPN promotes macrophage and osteoclast migration through a CD44-
Rho kinase dependant mechanism that requires OPN’s phosphorylation and TRAP processing activity (165, 288, 411, 412). Furthermore, Chellaiah et al. found that osteoclasts derived from OPN null bone marrow macrophages have impaired motility which could be rescued by exogenous OPN
(17). In contrast, Rajachar at al. showed that OPN -/- bone marrow-derived macrophages formed more osteoclasts on TCP than OPN +/+ macrophages (380), a finding consistent with previous in vitro and in vivo investigations of OPN null mice (17, 307). Although we found no difference in osteoclast number on OPN compared to HA, all resorption parameters were elevated except for sealing zone formation. This increased resorptive capacity in the presence of OPN, with the exception of one study (379), matches the rest of the literature demonstrating that OPN is required for proper bone resorption both in vivo and in vitro (17, 286, 380).
3.6 CONCLUSIONS
Taken together, these results highlight the important contribution of matrix proteins in modulating bone remodelling. FN, VN and OPN exerted different effects on osteoclast formation 107 and resorption by modifying different aspects of cell signalling, cytokine production and receptor expression. We also demonstrate that integrins have multiple functions which are substrate specific.
Given the prevalence of FN in inflamed joints in RA, an important outcome of this study is the stimulatory effect of FN on osteoclast function.
Increased osteoclast activity in RA results in irreparable joint destruction leading to permanent incapacitation and increased risk of fractures and implant failure, increasing the RA’s morbidity and reducing patients’ quality of life (413). This investigation highlights FN as a novel target that could be exploited to modify the progression of bone loss in inflammatory diseases.
108 Chapter 4
Bone Matrix Proteins and Extracellular Acidification; Potential Co-regulators of Osteoclast Morphology
This work has been accepted with revisions in the Journal of Cellular Biochemistry; Gramoun A, Goto T, Nordström T, Rotstein OD, Grinstein S, Heersche JNM, Manolson MF “Bone matrix proteins and extracellular acidification; potential co-regulators of osteoclast morphology”, January 19 2010, Manuscript ID JCB-09-0710.
109 4.1 ABSTRACT
Osteoclasts are signaled by the bone matrix proteins fibronectin (FN), vitronectin (VN) and osteopontin (OPN) via integrins. To perform their resorptive function, osteoclasts cycle through compact (polarized), spread (non-resorbing) and migratory morphologies. Here we investigate the effect of bone matrix on osteoclast morphology and how those effects are mediated using RAW
264.7 cells differentiated into osteoclasts on FN, VN and OPN-coated culture dishes. After 96 hours,
80% of osteoclasts on FN were compact while 25% and 16% on VN were in compact and migratory states respectively. In contrast, OPN induced osteoclast spreading. Furthermore, osteoclasts formed on VN and FN were 2-4fold smaller than those formed on OPN in the 21-30 nuclei/osteoclast group.
These effects were not due to defects in cytoskeletal reorganization of osteoclasts on VN and FN, demonstrated by the ability of these cells to spread in response to 35 ng/ml macrophage colony stimulating factor (M-CSF). Conversely, osteoclasts on OPN failed to spread when induced by M-
CSF. Moreover, the extracellular pH on FN and VN (7.25 and 7.3 respectively) was significantly lower than that on OPN (~7.4). We further investigated the role of extracellular pH and found that at pH 7.5 the duration of an osteoclast’s compact phase was 25.6 min and that of the spread phase was
62.5 min. Reducing the pH to 7.0 increased the frequency of osteoclast cycling by 3-fold. These results show that matrix proteins play a role in regulating osteoclast morphology, possibly via altering extracellular and intracellular pH.
110 4.2 INTRODUCTION
Bone resorption is an essential process required for maintenance of bone integrity.
Osteoclasts, the cells exclusively responsible for bone resorption, are of myeloid lineage and form by fusion of mononucleated precursors. Osteoclasts are dynamic and versatile cells.
During their functional cycle, osteoclasts alternate between migration and resorption and occasionally resting phases (414).
The focus of this study revolves around determining factors affecting osteoclast morphology; thus, we will first review relevant aspects of osteoclast morphology. During an osteoclast functional cycle leading to resorption, the cell undergoes a series of actin cytoskeletal rearrangements resulting in different morphologies. More specifically, an osteoclast was shown to exist in one of three phenotypes; spread (flat), migratory (motile), and polarized (compact).
A migrating osteoclast is characterized by its dynamic membrane rufflings and lamellipodial formation at the leading edge. During migration podosomes, the osteoclast attachment complexes are disrupted at the trailing edge; allowing for a forward sliding motion of the cell (165, 166).
A resorbing osteoclast is polarized baso-apically creating three functional membrane domains facilitating resorption; the ruffled border, the functional secretory domain and the baso- lateral domain. The ruffled border is juxtaposed to the bone surface and is enclosed in a junctional structure known as the sealing zone. Active membrane trafficking occurs at the ruffled border sequestering proton pumps, chloride channels and vesicles containing proteases to the area. The sealing zone is a dense F-actin band surrounded by 2 narrow vinculin rings on both its sides (153, 163). It surrounds the ruffled border and delineates a tightly sealed acidic compartment between the ruffled border and the bone surface. Finally, a spread osteoclast is a 111 flat, non-resorbing cell, does not exhibit an actin ring, and is neither migrating nor resorbing.
This phase can sometimes be seen between two functional cycles.
These alternating events of the resorption cycle occur continuously on bone and require rapid cycles of actin polymerization and depolymerization. Similar phases can also be seen in osteoclasts on tissue culture polystyrene (TCP) and glass but the ability of osteoclasts to form a sealing zone is substrate dependent and is governed entirely by the mineral content of the substrate (148). While a sealing zone is seen exclusively in osteoclasts on a resorbable surface, another actin rich podosome super-structure known as the podosome belt is located paramarginally in a mature osteoclast on glass and TCP (145). Even though the two structures are similar, they are distinctive in their actin organization (149). A podosome belt is made of an
F-actin condensation known as a podosome core, surrounded by branches of finely interconnected fibers called the actin cloud (150). Nonetheless, osteoclasts with podosome belts are still capable degrading of matrix proteins coated on TCP (136, 137).
The extracellular matrix proteins (ECM) are a large family of macromolecules that contribute to bio-mechanical properties of bone. Among the more abundant bone matrix proteins are fibronectin (FN), vitronectin (VN) and osteopontin (OPN), all three of which are
Arg-Gly-Asp (RGD) containing glycoproteins. The RGD driven attachment of osteoclasts to those proteins occurs via surface receptors known as integrins. Cell binding to these ECM proteins is the basis of cell attachment and survival. Cell spreading is also affected by this process which consequently dictates cell shape.
Despite the abundance of ECM proteins within bone, few studies have focused on their effects on osteoclast shape and morphology. Early studies have found that the proportion of compact osteoclasts is greater when mature rat osteoclasts are plated on dentine or type I collagen compared to plating them on glass (167). Saltel et al. have shown that a mineralized 112 matrix is required for the formation of a sealing zone in mature osteoclasts seeded on a substrate. An organic matrix, such as collagen I, was not sufficient for sealing zone formation but allowed for the formation of podosome belts (148). Various aspects of osteoclast attachment to different bone matrix proteins have been explored (222, 381) but there has not yet been any studies on how the matrix affects the morphology of differentiating osteoclasts.
Electrochemical balance and pH homeostasis are crucial for cell survival. pH regulation is not only essential for osteoclast survival but is also pivotal to optimal resorptive capacity. V-
- - + + ATPases, HCO3 / Cl exchangers, Na /H antiporters, p62/CLIC-5b, ClC-7 and carbonic anhydrase II have all been shown to play an important role in osteoclast pH regulation and resorption (95, 97-101, 104, 415, 416). Evidence indicates that extracellular acidification affects pHi regulation and osteoclast morphology (107).
Here we have investigated for the first time the morphological effects of the matrix proteins FN, VN and OPN on differentiating osteoclasts and the role of extracellular pH as a regulator of osteoclast shape. Our results suggest a relationship between matrix proteins and extracellular pH and raise the possibility that they are co-regulators of osteoclast morphology.
113 4.3 MATERIALS AND METHODS
4.3.1 Materials
Human FN was purchased from Sigma–Aldrich Ltd. (St. Louis, MO) and human VN was purchased from BD Biosciences (BD Labwares, Franklin Lakes, NJ). Bovine OPN was kindly provided by the late Dr. J. Sodek (University of Toronto). Recombinant mouse macrophage colony stimulating factor (M-CSF) was obtained from Calbiochem (EMD
BioSciences, Inc. San Diego, CA). The RAW 264.7 cell line was obtained from American Type
Culture Collection (ATCC, Manassas, VA). 4-[2-hydroxyethyl] piperazine- N'-[ethanesulfonic acid] (HEPES), amiloride, 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS), and acetazolamide were obtained from Sigma–Aldrich Ltd. (St. Louis, MO). Dulbecco’s modified
Eagle’s medium (DMEM) and -Minimum essential medium (-MEM) without bicarbonate containing 25 mM HEPES, antibiotics and antimycotics (penicillin/streptomycin, fungizone) and fetal bovine serum (FBS) were obtained from Invitrogen (Carlsbad, CA). Fast red violet LB salt and naphthol AS-MX, were obtained from Sigma–Aldrich Ltd. (St. Louis, MO). 6- and 12- well plastic FalconTM tissue culture plates were purchased from BD Biosciences (BD Labwares,
Franklin Lakes, NJ). 2',7'-bis- (2-Carboxyethyl)-5 (6)- carboxyfluorescein, acetoxymethyl ester
(BCECF, AM) was bought from Molecular Probes Inc. (Eugene, OR). Bafilomycin A1 was bought from Kamiya Biomedical Co. (Thousand Oaks, CA).
4.3.2 RAW 264.7-Derived Osteoclast Cultures
12-well tissue culture polystyrene (TCP) plates were pre-coated with 100 μl of 10 μg/ml of FN, VN or OPN dissolved in phosphate buffered saline (PBS) over night (O/N) at 4 °C. To
114 increase the amount of proteins physically adsorbed on to the TCP plates, the matrix proteins were incubated for an additional 1h at 37 °C. After the aspiration of the matrix proteins, the wells were subsequently blocked using 1% BSA in PBS for 1h at 37 °C in a CO2 incubator to minimize nonspecific binding of serum proteins to TCP. Finally the wells were washed 3x with
100 μl PBS and left in PBS until cells were plated to prevent denaturing of the proteins.
RAW264.7 (RAW) cells were plated on the ECM coated wells and cultured in DMEM supplemented with 10% FBS, 100 μg/ml penicillin/streptomycin and 0.2 μg/ml fungizone and incubated at 37 °C in 5% CO2. Osteoclasts were generated using 75 ng/ml receptor activator of
NFB ligand (RANKL). After 96 h of incubation, multinucleated osteoclasts were observed.
When cultures were stopped using 4% formaldehyde for 5-6 min, these cells stained positive for tartrate resistant acid phosphatase activity (TRAP). Cell morphology and number were determined using bright field microscopy. The number of osteoclasts with compact, migratory and spread morphologies were count on each of the adsorbed ECM proteins and were plotted as a percentage of the total number of osteoclasts with >3 nuclei. No specific morphologies could be identified in osteoclasts with 2 and 3 nuclei.
Osteoclast spreading on ECM proteins was assessed at 96 h after their incubation with
M-CSF. This was achieved by washing cultures 3x with cold PBS -Ca -Mg. The cells were then incubated with PBS -Ca -Mg for 45 min at 37°C and 5% CO2. The cultures were subsequently incubated for 2.5 h in media without FBS and with or without 35 ng/ml M-CSF. The cultures were then fixed and stained for TRAP activity.
4.3.3 Rabbit Osteoclast Isolation
Animal protocols were approved by the Animal Care Committees at the University of
115 Toronto. Osteoclasts were isolated from the long bones of newly born New Zealand rabbits, as described previously (162, 383). Briefly, bones were cleaned and minced mechanically in 100 mm glass Petri dishes containing 10 ml of -MEM supplemented with 10% FBS and 1% penicillin/streptomycin. Bone fragments were transferred into 50 ml Falcon tubes and cells were resuspended by repeated passage (30 times) through a wide-bore Pasteur pipette. The bone fragments were allowed to briefly settle and the cell suspension was transferred to another tube.
An additional 6 ml of supplemented media was then added to the remaining minced bones in the
Petri dish and the previous steps were repeated. Aliquots (100 ml) of cell suspension were plated in 6-well TCP plates. Osteoclasts were allowed to attach for 1 h (37°C, 5% CO2), after which
1.5 ml of supplemented medium was added. After a further 18-h period, the cultures were washed gently with -MEM using a widebore Pasteur pipette to remove non-attached cells. This technique of cell isolation generates a mixed culture of attached osteoclasts, prefusion osteoclasts, and stromal cells. Only multinucleated cells identified under phase contrast were monitored for morphological changes.
4.3.4 Tartrate-Resistant Acid Phosphatase (TRAP) Staining
TRAP staining was carried out according to the protocol described in BD Biosciences
Technical Bulletin #445. Briefly, cell cultures were washed 3x with PBS, fixed with 4% formaldehyde for 5-6 min and incubated in TRAP staining solution (50 mM acetate buffer, 30 mM sodium tartrate, 0.1 mg/ml Naphtol AS-MX phosphate, 0.1% w/v Triton X-100, and 0.3 mg/ml Fast Red Violet LB stain) for 10 min until the desired staining intensity was reached. The
TRAP staining solution was aspirated and the cells were washed 3x with dH2O.
116 4.3.5 Assessment of Osteoclast Morphological Changes Using Time-lapse Microscopy
After an 18-h incubation period of rabbit osteoclasts from the time of their isolation, the cultures were washed gently with bicarbonate free -MEM supplemented media containing
25mM HEPES, using a widebore Pasteur pipette to remove non-attached cells. The cultures were subsequently incubated with 1.5 ml -MEM supplemented media containing 25mM
HEPES, with the pH adjusted to pH 7.0 or 7.5 for up to 24 h. Media pH was adjusted using 0.5
M solution of either hydrochloric acid or sodium hydroxide.
TCP plates were placed in the incubation chamber of a Nikon Diaphot phase-contrast inverted microscope for time lapse photography using a SONY AVC D5 camera. Some plates were initially incubated at pH 7.5 for 5 to 6 h, then washed once and the medium was changed to a fresh medium of pH 7.0.
To determine the role of V-ATPases, Na+/H+ antiporters, anion exchangers and carbonic anhydrase in regulating changes in osteoclast morphology, osteoclasts were incubated with medium at pH 7.0 for 3 hours after which the different inhibitors were added. We tested the effects of 100 nM bafilomycin A1 (V-ATPase inhibitor); 1 mM amiloride (Na+/H+ antiporter inhibitor); 100 μM DIDS (anion exchanger inhibitor) and 10 μM acetazolamide (carbonic anhydrase inhibitor) by incubating the cells with each of these inhibitors for 3 h at pH 7.0.
4.3.6 Morphometrical Analysis of Changes in Osteoclast’s Morphology
Using a Leica DMIRE2 microscope, micrographs of cultures on different ECM proteins were acquired after the cells were fixed and TRAP stained. The perimeter of TRAP+ osteoclasts was outlined manually and the area was measured using Openlab® imaging system. Osteoclasts were divided into groups based on their size determined by the number of nuclei/osteoclast.
117 Morphometrical analysis of rabbit osteoclast spreading and motility were performed using the method described by Zaidi et al.(417). The spread area, A(t) at each time t was obtained by tracing the cell perimeter and its area was quantified using Zeiss Zidas system. Area was expressed as a percentage of mean area
([Δr]+[Δp])/A(t).
4.3.7 Scanning Electron Microscopy
Cultured cells were washed twice with medium and fixed with 2.5% glutaraldehyde and
3.7% paraformaldehyde in 0.1 M PBS for 2 h. After fixation, the specimens were rinsed with
PBS, dehydrated through a graded ethanol series and critical-point dried after ethanol was substituted with CO2. The specimens were places on an aluminum slab, coated with gold sputter and observed using Hitachi S-2500 operated in the secondary electron mode at 10 kV.
4.3.8 Intracellular pH Measurements
As described previously, rabbit osteoclasts were isolated and incubated with 1.5 ml -
MEM supplemented media containing 25mM HEPES, with the pH adjusted to pH 7.0 or 7.5 for up to 24 h prior to measuring the cytoplasmic pH. Media pH was adjusted using 0.5 M solution of either hydrochloric acid or sodium hydroxide. For microfluorimetric studies, osteoclasts were
118 analyzed essentially as described by (97). Briefly, the cells were plated for 48h on acid washed glass coverslips and placed into a Leiden Cover Slip Dish and maintained at 37°C. They were then loaded with BCECF by incubation with 1 μM of the parent acetoxymethyl ester for 15min at 37°C. The osteoclasts were next washed with -MEM and incubated in the indicated conditions. Single cell fluorescence was monitored using a Nikon TMD-Diaphot microscope attached to an M Series Dual Wavelength Illumination System from Photon Technologies Inc.
Illumination was shuttered on and off for 2 and 20 s, respectively and the photometric data was recorded at a rate of 5 points/s. Mean values for each 2 s illumination period were plotted against time. Calibration of the fluorescence ratio vs. pH was performed using the K+/H+ ionophore nigericin. Cells were equilibrated in K+ medium (140 mM) of varying pH in the presence of 5 μM nigericin and calibration curves were constructed by plotting the extracellular pH (which is assumed to be identical to the internal pH (418) against the corresponding fluorescence ratio. The resulting curve was sigmoidal with an inflection point =7.0 as expected from the reported pKa of BCECF.
4.3.9 STATISTICS
Statistical evaluation was carried out using SPSS 12.0 for Windows using one way analysis of variance (ANOVA) and Dunnette T3 test and Student T-test. P-values less than 0.05 were considered statistically significant.
119 4.4 RESULTS
4.4.1 Osteoclasts Formed on FN, VN and OPN have Distinct Morphologies and Planar
Area
To ask whether differentiating osteoclasts in the presence of a specific matrix protein has an effect on their morphology, RAW cells were cultured for 96 h on TCP coated with physically adsorbed 10 μg/ml FN, VN and OPN. The cultures were subsequently fixed and stained for
TRAP activity and the morphology of the cells was assessed. We observed that osteoclasts formed on each of the different ECM proteins had a distinctive and unique morphology.
Osteoclasts formed on FN had a “rounded up” or “compact” morphology similar to that seen when osteoclasts are polarized on a resorbable matrix (figure 4.1A). In contrast, osteoclasts on
OPN were noticeably more “spread” or “flat” (figure 4.1C). Osteoclasts on VN exhibited either a compact morphology with a condensed cytoplasm similar to that seen on FN or a “motile” or
“migrating” morphology (figure 4.1B). Migrating osteoclasts are characterized by having a
“leading edge” and a “trailing edge” with ruffling in their lamellipodia. Based on these observations we identified three morphologies (figure 4.1d: compact, migratory and spread) which were the prevalent osteoclast morphologies in the cultures.
To quantify this observation, the percentage of osteoclasts conforming to the predetermined morphology criteria was determined and is shown in figure 4.2A. Only osteoclasts where one of the three morphologies could clearly be identified were counted and small osteoclasts (2-5 nuclei) were not included as it was difficult to categorize them under any of the three preset morphologies.
On FN, 80% of the total number of osteoclasts was compact, with less than 7% of osteoclasts spread or migratory. Osteoclasts on VN were found to be 25% compact, 7% spread 120 and 16% migratory. Finally, 36% of the total osteoclasts formed on OPN were spread versus 8% and 2% compact and migratory respectively (figure 4.2A). These results show that FN increases the percentage of compact osteoclasts compared to VN and OPN while osteoclasts on OPN were more spread compared to those on FN and VN. The highest percentage of migratory osteoclasts was seen on VN. It is worth noting that only osteoclasts with > 3 nuclei that were included since no specific morphologies could be identified in osteoclasts with 2 and 3 nuclei. Also, the total number of osteoclasts on the different matrix proteins varied significantly due to an effect of these matrix proteins on osteoclast differentiation and these effects are the focus of another study.
To further quantify differences in morphology, we measured osteoclast planar area. As compact osteoclasts have smaller planar areas than spread osteoclasts, we expected and indeed found that osteoclasts on FN had a consistently smaller planar area than those on OPN (figure
2B), irrespective of osteoclast size (defined by number of nuclei/osteoclasts). By presenting the planar area measurements in 4 groups of osteoclast size, one can also see that the ratio of osteoclast planar area per nucleus changed dramatically for osteoclasts on OPN within the 21-30 nuclei group (figure 4.2B). In contrast, the ratio of planar area remained unchanged between all four size groups for TCP, FN, and VN.
4.4.2 M-CSF Induces Osteoclast Spreading on FN but not on OPN
We hypothesized that the differences seen in osteoclast morphology and area on the
ECM proteins underlie a defect in the activation of the osteoclast machinery responsible for osteoclast spreading. As M-CSF activates cytoskeletal reorganization leading to osteoclast spreading, we asked whether osteoclasts on FN, VN, and OPN respond differentially to M-CSF.
Osteoclasts differentiated for 96 h on physically adsorbed matrix proteins were made to retract 121 using cold PBS -Ca-Mg. When cellular retraction was observed, they were subsequently triggered to spread using 35 ng/ml M-CSF.
We observed that the degree of osteoclast spreading changed with osteoclast size.
Therefore, to facilitate comparison between the different matrix proteins and eliminate differences from osteoclast size, the results were grouped by nuclei per osteoclasts. Osteoclast with more than 30 nuclei did not respond to M-CSF irrespective of matrix, (data not shown) and were excluded. The highest response to M-CSF was seen in osteoclasts containing 11-20 nuclei.
Osteoclasts on FN spread appropriately in response to M-CSF, suggesting that the compact shape of the osteoclasts does not arise from a defect in cytoskeletal reorganization
(figure 4.3). In contrast, osteoclast on OPN did not spread in response to M-CSF. This was anticipated as OPN itself induces osteoclast spreading (figure 4.2). VN had an intermediate response in that there was a non-significant increase in osteoclast area, possible reflecting the mixed morphologies observed in this group.
4.4.3 Extracellular pH of Cultures on FN and VN are Lower than that on OPN
Extracellular pH has previously been shown to affect osteoclast morphology. Nordström et al. found that lowering the extracellular pH from 7.5 to 7.0 for 24 h increased the proportion of compact osteoclasts from 8 ± 2 % to 24.5 ± 6 % (107). In light of their finding, we hypothesized that changes in pH of the cultures on FN, VN and OPN might be responsible for the morphological differences of osteoclasts formed on these ECM proteins. To test this, the extracellular pH of the osteoclast cultures differentiated on these matrix proteins was measured.
Coinciding with Nordström’s findings, the extracellular pH of osteoclasts on FN, where 25% of the osteoclasts assumed a compact morphology, was 7.24 ± 0.01. On VN, where 12% of the cells were compact, the pH was 7.29 ± 0.02 (figure 4.4). On the contrary, the OPN group, where 122 the highest percentage of spread osteoclasts was seen (16%), had the highest extracellular pH of
7.37 ± 0.03. Thus the morphological variations observed between different substrates correlate with the changes one would expect as a result of the pH of the respective culture media.
4.4.4 Osteoclasts Cycle between Spread and Compact Morphologies and the Rate of these
Changes Depends on Osteoclast Size and Extracellular pH
The correlation between extracellular pH and the prevalence of a specific morphology indicates that pH plays an important regulatory role. Hence, to investigate how pH regulates osteoclast morphology, we used time-lapse microscopy of authentic rabbit osteoclasts. Figure
4.5 illustrates the morphological cycle of a rabbit osteoclast and outlines how we define the difference between spread and compact osteoclast morphologies in the subsequent figures.
Figure 4.5a-f is a series of phase-contrast micrographs of an osteoclast cultured at pH 7.0. As seen in these images, an osteoclast is continuously undergoing morphological changes between spread (4.5a), compact 90 min later (4.5e), and then spreading again after another 30 min. This figure demonstrates how osteoclasts spontaneously alternate between compact and spread phases. An osteoclast with its nuclei easily recognizable was considered to be spread. An osteoclast with a rounded basolateral surface reflecting the light, giving it a bright ring appearance and concealing its nuclei, was considered to be compact. Microstructural differences between spread and compact could also be seen using scanning electron microscopy (4.5g-i).
Figure 5g shows osteoclasts in the spread phase as flat and with few surface protrusions. In contrast, osteoclasts in compact phase were globular or dome-shaped with more ruffling of the basolateral surface. They also had numerous filopodial projections at their periphery (figure
4.5h-i).
123 Next, we proceeded to investigate the effect of extracellular pH on the osteoclast morphological cycle; changes in osteoclast area and motility were recorded for 6 h at pH 7.5, after which the media was switched to pH 7.0 for an additional 5-9 hours. Because of the differences in spreading between large and small osteoclasts that were previously noted (figure
3) the measurements in figure 4.6 are also grouped according to osteoclast size (nuclei number/OC). For small osteoclasts, the area descriptor was not affected by lowering the pH to
7.0 yet the motility descriptor increased from 1.11 to 1.34 (figure 4.6 A). Small osteoclasts also significantly increased the frequency of morphology cycling at pH 7.0 compared to pH 7.5.
When the results from small osteoclasts were compared to those from large osteoclasts, significant differences in cycling patterns could be seen. Large osteoclasts were found to be predominantly in a spread morphology at pH 7.5 and cycled between spread and large morphologies only when the pH was dropped to 7.0. This was accompanied by an increase in the spread area during the spread phase.
Table (4.1) summarizes the average duration small osteoclasts remained in spread and compact phases at pH 7.5 and 7.0. The average compact phase duration for a small osteoclast at pH 7.5 was 25.6 min and the duration of a spread phase was 62.5 min. At pH 7.0, the duration of both phases decreased significantly; small osteoclasts stayed in a compact phase for 14.1 min and in a spread phase for only 15.7 min. Thus the number of cycles/ hour of a small osteoclast increased by 3-fold at pH 7.0 compared to that at pH 7.5, and that on average, a small osteoclast stayed compact twice as long at the lower pH (table 4.1).
We next examined the pHi associated with these cycling events of an osteoclast maintained at pH 7.0. As shown in table 2, the average pHi of a compact osteoclast was 7.00.
The pHi of spread osteoclasts on the other hand was 7.34 (table 4.2).
124 To determine the mechanisms involved in this pH regulation, the role of the various enzymes involved in osteoclast pH homeostasis were examined. Because of their established
- - + + role in controlling osteoclast pHi, V-ATPase, HCO3 / Cl exchanger, Na /H antiporter, and carbonic anhydrase were evaluated. Specific inhibitors against these enzymes were added at pH
7.0 and the rate of morphological cycling; osteoclast motility and spreading were recorded. The addition of the V-ATPase specific inhibitor bafilomycin A1 increased the duration of the compact phase 4-fold while that of the spread phase decreased (Table 4.3). Bafilomycin A1 did not affect motility on the basis of time lapse photograph and its effects were completely reversible when cells were transferred to fresh media without the inhibitor (data not shown). The
- - + + inhibition of the HCO3 / Cl exchanger by DIDS and the Na /H antiporter by amiloride had the same effect on the duration of the osteoclast compact phase but their effects on the duration of the spread were less pronounced. Finally, inhibiting carbonic anhydrase by acetazolamide had no effect on the length of either phase.
125
Figure 4.1 Osteoclasts (OCs) differentiated on FN, VN and OPN have different morphologies.
RAW cells were plated on TCP dishes precoated with FN, VN and OPN at 10 µg/ml. RAW cells were differentiated into osteoclasts for 96 h in the presence of 75 ng/ml RANKL. The cultures were fixed and stained for TRAP activity. Micrographs of osteoclasts cultures on FN
(A), VN (B) and OPN (C) were taken using Openlab® Imaging system. Scale bar is 25 μm.
Panel 1D is composed of images of osteoclasts representing the three morphologies that were identified in cultures on FN, VN and OPN: (i) migratory osteoclast, (ii) compact osteoclast and
(iii) spread osteoclast. Similar results were obtained in 3 separate experiments.
126 A
Figure 4.2 Osteoclast (OC) morphology and planar area are modulated by the ECM proteins
FN, VN and OPN. RAW cells were plated on TCP dishes precoated with FN, VN and OPN at
10 µg/ml. RAW cells were differentiated into osteoclasts for 96 h in the presence of 75 ng/ml
RANKL. The cultures were fixed and stained for TRAP activity and the total number of TRAP+ osteoclasts was counted. Subsequently the number of osteoclasts with a condensed cytoplasm
(compact), membrane ruffling (migratory) or flat (spread) morphology was determined and finally the they were plotted as a percentage of osteoclasts with >3 nuclei (A). Each data point represents the pooled results from 4 dishes per treatment and is expressed per well (means ±
SD). *P value <0.05 versus VN and OPN groups, #P value <0.05 versus FN and OPN groups and †P value <0.05 versus FN and VN groups. Osteoclast perimeter was traced manually and the
127 ® planar area was measured using Openlab imaging system. Planar areas measurements were normalized to the number of nuclei/osteoclast and the data was divided into groups based on osteoclast size as defined by its number of nuclei/osteoclast (B). Each data point represents pooled results from a minimum of 40 osteoclasts per treatment group and is expressed as osteoclast area/nucleus (means ± SD). *, #, † and ¥ P value <0.05 versus OPN in each respective size category. Similar results were obtained in 3 separate experiments.
128
Figure 4.3 M-CSF treatment causes osteoclast spreading on FN while osteoclasts (OCs) on
OPN fail to spread. RAW cells were plated on TCP dishes precoated with FN, VN and OPN at
10 µg/ml. RAW cells were differentiated into osteoclasts for 96 h in the presence of 75 ng/ml
RANKL. At 96 h, the cultures were washed 3X with cold PBS -Ca –Mg. The cells were incubated with PBS -Ca -Mg for 45 min at 37°C and 5% CO2. The cultures were subsequently incubated with media –FBS with or without 35 ng/ml M-CSF for 2.5 h. The cultures were fixed and stained for TRAP activity. The perimeter of TRAP+ osteoclasts was outlined manually and the area was measured using Openlab® imaging system. Osteoclasts were divided into groups based on their size determined by the number of nuclei/osteoclast; 2 to 5 nuclei (A) and 11 to 20 nuclei (B). A total of 130 osteoclasts were measured per treatment group. Each data point represents the mean ± SD. Similar results were obtained in 3 different experiments.
129
Figure 4.4 The effect of ECM proteins on extracellular pH of culture media. RAW cells were plated on TCP dishes precoated with FN, VN and OPN at 10 µg/ml. RAW cells were differentiated into osteoclasts for 96 h in the presence of 75 ng/ml RANKL. At 96 h, the conditioned media from the cultures was removed and pH determined. Each data point represents the pooled results from 4 dishes per treatment and is expressed per well (means ±
SD). Similar results were obtained in 3 different experiments.
130
Figure 4.5 Phase-contrast and scanning electron micrographs demonstrate the morphological cycling of an osteoclast at pH 7.0. Rabbit osteoclasts were cultured in 35 mm dishes for 16 h before they were washed with bicarbonate-free α-MEM and then incubated with bicarbonate- free HEPES buffered α-MEM of pH 7.0. Phase-contrast images were acquired at the following intervals: a: time 0, b: 30 min, c: 60 min, d: 75 min, e: 90 min, f: 120 min. The osteoclast depicted can be seen altering its morphology from spread (a) to compact (e) and spread again
(f). The compact phase was defined as the time during which a complete bright ring of reflected light can be seen. Scale bars are 50 μm. Scanning electron micrographs of osteoclasts cultured as described above show an osteoclast in spread (g) and compact phase (h,i). Figure (i) is a higher magnification of the periphery of the osteoclast seen in (h). The compact osteoclast in (i) is dome shaped with numerous ruffling on the basolateral surface (arrows) and dense filopodial projections (arrow heads). Scale bars, 60 μm (g); 15 μm (h); 6 μm (i).
131
Figure 4.6 Time course of morphological cycling of small and large osteoclasts at pH
7.5 and pH 7.0. Rabbit osteoclasts were cultured in 35 mm dishes for 16 h before they were washed with bicarbonate-free α-
MEM and then incubated with bicarbonate- free HEPES buffered α-MEM pH 7.5 for 6 h.
This was followed a media change where the osteoclasts were cultured for an additional 5-
9h at pH 7.0. Phase-contrast images were acquired at equal time intervals and the following parameters were calculated:
A(t)/A0 denoting cell spreading area, and
([ΔP]+ [Δr])/A(t) denoting cell motility. (A) time course of a small osteoclast (maximum diameter = 50 μm, number of nuclei = 5), (B) time course of a large osteoclast (maximum diameter = 150 μm, number of nuclei = 28).
132
Table 4.1. The Duration of the Compact and Spread Phases of Osteoclasts at pH 7.0 and 7.5 Compact (A) Spread (B) (A/B) {60/(A+B)} (min) (min) pH 7.5 25.6 ± 6.57 62.5 ± 14.16 0.41 0.68
pH 7.0 14.1 ± 1.49 15.7 ± 1.95 0.9 2.01
The average duration in min of compact and spread phases of osteoclasts, expressed as mean ± SEM. The ratio of compact to spread phase in time (A/B) and the number of cycles per hour {60/(A+B)}; n=15 in each group.
Table 4.2. Intracellular pH in Compact and Spread Osteoclasts Intracellular pH
Compact phase 7.00 ± 0.15
Spread phase 7.34 ± 0.14
Cells were cultured in medium at pH 7.0 for 3 hours before their intracellular pH was measured. Results are expressed as mean ± SD, where n=20 in each group.
133
Table 4.3. The Effects of Bafilomycin A1(BFA), Acetazolamide (AZ), DIDS and Amiloride (Am) on the Duration of Compact and Spread Phases of Osteoclasts Compact Spread * Cell Size (μm) (min) (min)
Control 13.0 ± 2.1 25.7 ± 8.1 79.2 BAF (100 nM) 58.5 ± 13.5 17. 6 ± 6.7 n=7
Control 26.4 ± 5.5 22.7 ± 8.9 85 Am (1mM) 139.9 ± 24.7 42.2 ±28.6 n= 7
Control 19.8 ± 4.1 26.0 ± 8.4 88.7 DIDS (100 μM) 88.4 ± 21.5 21.4 ± 4.0 n=8
Control 10.1 ± 1.9 22.5 ± 3.7 112.5 AZ (10 μM) 18.7 ± 5.8 13.6 ± 3.4 n=5
Results are the mean ± SEM of each duration. *Cell size is represented as the average of maximum osteoclast diameter measured, where n= number of osteoclasts measured
134 4.5 DISCUSSION
The cycling of an osteoclast through its different morphologies is essential to its ability to perform its function. Many intra and extracellular factors contribute to osteoclast morphology. In this investigation we have examined extracellular factors regulating osteoclast shape.
Our data shows that differentiating osteoclasts on different ECM differentially regulates morphology. FN resulted in a predominantly compact population, VN primarily migratory, and spread osteoclasts where mostly seen on OPN. These results suggest that the ECM proteins regulate the cytoskeleton via altered signalling mediated through their podosomes. In addition to the variations in osteoclast shape, we also noticed differences in osteoclast size as defined by planar area. Osteoclasts on FN and VN were smaller than those on OPN, and this difference was maintained even when the planar area was normalized to number of nuclei. Osteoclasts of different sizes (2-4, 5-10, 11-20, and 21-30) had similar ratios of planar area to number of nuclei with the exception of OPN. This ratio was elevated in osteoclasts with 21-30 nuclei on OPN and was due to a disproportional increase in their spreading. The differences between FN and
VN on one hand and OPN on the other hand may reflect enhancement of cell spreading on OPN, compared to the increased cell polarity and compactness on FN and VN. In contrast with our findings, Hu et al. found that the mature rat osteoclasts that attached to FN and OPN were larger and contained extended lamellaepodia with an average of 6-7 nuclei compared to those which adhered on prothrombin and thrombin (381). These contradictory observations are potentially due to differences in experimental design and cell culture. In their study, mature primary rat osteoclasts were plated onto matrix proteins while in this report we used a mouse cell line and differentiated osteoclasts in the presence of the ECM proteins. 135 To ask whether the different morphologies reflect a defect in the activation of osteoclast spreading, the cells’ ability to respond to M-CSF was assessed. M-CSF induces osteoclast spreading and cytoskeletal reorganization on glass and plastic, while on mineralized substrates,
M-CSF triggers osteoclast activation and resorption (243, 264, 342, 383, 419). The cytokine’s action is mediated via phosphatidylinositol 3-kinase (PI3-kinase) which in turn induces integrin
v3 activation via the non-receptor tyrosine kinase c-src (420). The signaling transduction pathways of both c-FMS (encodes the receptor for M-CSF) and v3 are convergent and involve several common signaling molecules from the c-src family (82, 341, 421). To determine any defects in the M-CSF-v3- actin cytoskeleton axis, we challenged osteoclasts by inducing retraction using cold PBS-Ca-Mg followed by M-CSF to provoke cell spreading. Osteoclasts on
FN and VN did spread appropriately in response to M-CSF demonstrating that their smaller size was not due to improper cytoskeletal reorganization in response to extracellular ECM signalling.
In contrast, osteoclasts on OPN did not spread in response to M-CSF treatment. This may be due to the fact that morphological changes of osteoclasts on OPN are independent of the M-CSF signalling pathway. Alternatively, OPN might be stimulating osteoclast spreading to such an extent that no synergisitc effect can be seen with M-CSF. Teti et al. performed similar experiments on mature rabbit osteoclasts and found that osteoclasts on OPN did spread in response to M-CSF, however it is not clear from their figure whether the changes in area were significant or not (342).
In an attempt to understand the mechanism by which diseases characterized by chronic metabolic acidosis cause bone loss, previous experiments by Nordström et al., have shown that osteoclast morphology is related to the pH of the culture medium. Interestingly, they found that when they mimicked acidosis by lowering extracellular pH to 6.5, a 15% increase in cells with a
136 compact phenotype occurred (107), similar to the increase in the percentage of compact osteoclasts we saw on FN. Consistent with their pH findings; when the extracellular pH in our cultures was measured, we found the pH on FN to be substantially lower than on OPN, thus suggesting that each matrix proteins differentially affects the pH of the culture medium.
Further examination of the effects of low extracellular pH on osteoclast phenotype revealed a morphological cycle of osteoclasts on TCP. Similar to osteoclasts on bone, osteoclasts on TCP alternate between spread and compact phases. On an ultramicroscopic level the two morphologies were also distinct. Spread osteoclasts had a flat shape with a few cellular projections while compact osteoclasts had a domed shape basolateral membrane with numerous microvilli and filopodia. When a timeline of those morphological changes was studied at a physiologic (7.5) and low (7.0) pH, striking differences were noted. Both osteoclast motility and the frequency of the morphological cycle were increased at pH 7.0 compared to pH 7.5, resulting in an increase in the total duration of an osteoclast in a compact phase. This, in part, explains the increase in the percentage of compact osteoclasts at low pH (107) and on FN seen in figure 2A. This also shows that compact osteoclasts on TCP are similar to polarized osteoclasts on bone. In general, these observations are compatible with other observations showing an increase in osteoclast polarity and resorption at a low pH (106-108, 402).
Another similarity between compact osteoclasts on TCP and bone lies in the regulation of their cytoplasmic pH. It is known that actively resorbing osteoclasts dissolve a mineralized tissue by pumping H+ into the resorption lacunae. To do so, polarized non- resorbing osteoclasts need to accumulate protons prior to proton extrusion. We found that similar to a polarized osteoclast on bone, a compact osteoclast on TCP has a lower pHi than that in a spread phase.
- - The process of cytosolic proton accumulation involves V-ATPases, HCO3 / Cl exchangers,
Na+/H+ antiporters, and carbonic anhydrase II. We used specific inhibitors to each of these 137 enzymes to assess their contribution to the regulation of pHi and osteoclast morphology. The
- - duration of the compact phase increased when inhibiting the V-ATPase, the HCO3 / Cl exchanger or the Na+/H+ antiporter, the frequency of the morphological cycle on the other hand decreased. In contrast, carbonic anhydrase II had no effect, possibly because of a suboptimal concentration of acetazolamide (422). Alternatively, a different mechanism of proton production, not related to carbonic anhydrase activity may exist. Thus, we have found that in addition to differences in pHi and its regulation seen between large and small and resorbing and non-resorbing osteoclasts (355, 359), extracellular pH and osteoclast shape can be determining factors of pH regulation.
The link between pHi and FN has been established in studies investigating the role of the
ECM proteins in cell proliferation. The rate of cell proliferation was found to be affected by pHi, which in turn, was dependent on cell spreading. Schwartz et al. showed that conditions that induced fibroblast spreading increased cell proliferation, and that was also accompanied by an increase in pHi (394). Similar effects on pHi were found when capillary endothelial cells were cultured on FN (423). Subsequently, they were able to demonstrate that FN led to 51 clustering, which activated the Na+/H+ anti-porter, in turn driving protons out of the cells and raising pHi (424). Based on these findings, we propose the following model for how FN causes osteoclast activation and compact morphology. This model is based on the fact that RAW cell cultures contain two cell populations. At 48 hours, the majority of cells are undifferentiated and
TRAP + mononuclear cells. By 72 hours, a second population of multinucleated osteoclasts starts to appear. On FN, we found that the mononuclear cells were significantly more spread than on TCP. Similar to Schwartz et al. observations with fibroblast, we propose that the spreading of the mononuclear cells is accompanied by activation of the Na+/H+ antiporter, an
138 elevation of their pHi and acidification of their extracellular milieu. The extracellular acidification would then activate osteoclasts and result in a more compact morphology.
One of the intriguing aspects of osteoclasts physiology is their heterogeneous behaviour based on their size. Studies have shown that large osteoclasts (≥ 10 nuclei) are significantly different in their behaviour and resorptive activity from small osteoclasts (2-5 nuclei) (105). The former were hyperactive and produced more pits that were larger (355, 419). In addition, a higher percentage of large osteoclasts can be seen actively resorbing (355). Large osteoclasts are also found predominantly in pathological bone loss conditions and sites of bone inflammation
(110, 425, 426). Subsequent studies have shown that those two osteoclast populations are inherently different. Large osteoclasts have a higher pHi maintained by V-ATPases (359).
Higher expression of the osteoclast enriched V-ATPase a3 subunit was also noted in large osteoclasts (415). In contrast, small osteoclasts had a lower pHi regulated by V-ATPases and
Na+/H+ antiporters (359). More recently it was shown that large and small osteoclasts also differentially express several receptors and osteoclast phenotypic markers that activate different signalling pathways (356).
Similar to these reports, we found a heterogeneous behaviour of osteoclasts based on their size. Small and large osteoclasts had different responses to M-CSF. Large osteoclasts on
TCP, FN and VN showed a greater spreading response than small osteoclasts. Furthermore, large and small osteoclasts exhibited different cycling patterns. At a physiologic pH, small osteoclasts spontaneously cycled between spread and compact morphologies on average twice/hour frequency while large osteoclasts showed no morphological changes. Upon extracellular acidification, small osteoclasts cycled 4-5 times/hour and the large osteoclasts cycled ~ 0.5 times/ hour.
139 Our findings demonstrate a similarity in the behaviour of osteoclasts on bone and TCP.
Osteoclasts on plastic go through a morphological cycle alternating between a compact phase corresponding to the polarized phase seen in actively resorbing osteoclasts, and a spread phase similar to the resting or non-resorbing osteoclasts. These data support the emerging notion that the sealing zone formed on bone and the podosome belt formed on glass are closely related structures and that the former is formed by contraction of the actin core and actin cloud domains found in the podosome belt (138).
Finally, it was previously reported that extracellular acidosis increases osteoclast polarization and resorption (108, 401) via increasing the H +-pumping activity of the V-ATPase
(107). This is compatible with our findings that low extracellular pH increased the percentage of osteoclasts in a compact phase and the total time an osteoclast spends in that phase. Taken together, these data suggest that compact osteoclasts on TCP have structural and functional similarities with their polarized counterparts on mineralized surfaces.
In summary, we have identified ECM proteins and pH as co-players involved in regulating osteoclast shape and present data indicating that they connected. Our data suggests that in addition to the canonical mechanism controlling osteoclast skeletal organization triggered by the mineral content of a matrix, ECM proteins and extracellular acidification should be considered as contributing factors modulating this process.
140 5. SUMMARY AND GENERAL DISCUSSION
Initially, the focus of my project was to define the cellular and molecular mechanisms involved in Vitaxin’s effects on bone homeostasis. The humanized anti-v3 antibody; Vitaxin was in clinical trails as a treatment for pathological bone loss; however, its effects on the bone resorbing cells were not yet determined (349, 351). Using a mixed culture of prefusion and mature primary rabbit osteoclasts, the effects of Vitaxin on osteoclasts cultured on TCP, bone and HA were assessed. In the literature, two mechanisms of action were suggested for inhibition of bone resorption by other blocking antibodies, disintegrins and RGD containing peptides. The first mechanism involved the inhibition of attachment which lead to attachment dependent cell death known as “anoikis” (242), whereas the second mechanism implicated the inhibition of migration as the mechanism showing that even at concentrations where neither cell detachment nor apoptosis occur, inhibition of bone loss can still be observed (217, 244). I was interested in determining which of these mechanisms occurs when osteoclasts are treated with Vitaxin. Since
v3’s role in osteoclast differentiation is crucial (4, 268), it was important to also confirm if the antibody interfered with the process. Attachment, resorption and differentiation assays were performed on both TCP, bone and HA. Attachment of osteoclast on all three substrates was affected and that was not due to increased fusion nor cell toxicity since the concentrations used did not affect cell viability. Resorption on bone and HA was significantly decreased, however, resorption per osteoclast was not altered indicating that osteoclast resorptive activity was not decreased and that Vitaxin did not prevent ruffled border formation. I obtained more conclusive evidence of the mechanism of action of Vitaxin when the cells were treated with Vitaxin prior to seeding and were monitored for the following 24 hours. Vitaxin inhibited the initial attachment
141 of mature osteoclasts without affecting the rate of osteoclast formation in cultures. I concluded from these results that Vitaxin did indeed decrease bone resorption through affecting osteoclast attachment but not their migration nor their ability to resorb bone through preventing ruffled border formation. In addition to its main effect on attachment, Vitaxin’s effect was found to be selective and depending on osteoclast population. Because different studies by our group and others have shown that large osteoclasts with more than 10 nuclei are hyperactive and implicated them in pathological bone loss (105, 355, 356, 359, 415), I was interested in determining if Vitaxin has a differential effect on small (<10 nuclei) and large (≥10 nuclei) osteoclasts. Surprisingly, Vitaxin only impaired the attachment of small osteoclasts but the number of large osteoclasts was not decreased with increasing concentrations of Vitaxin.
Another finding by our group, showing that large osteoclasts had significantly higher levels of both v and 3 subunits, was a plausible reason for the lack of effect of Vitaxin on large osteoclasts (356). Although these results constitute an obstacle to using Vitaxin in inflammatory conditions, they are helpful in determining the dose required to efficiently treat patients with a prevailing large osteoclast population.
Delineating the cellular mechanism of Vitaxin’s effect on bone resorption was the first step in understanding how Vitaxin regulates osteoclast function. Live cell imaging was next used to analyse the morphological changes occurring when osteoclasts are treated by Vitaxin.
Monitoring morphological changes, we found that Vitaxin caused some cell retraction; however, the most significant cell retraction and decrease in osteoclast planar area was only in the presence of M-CSF. The fact that M-CSF was needed for an optimal effect of Vitaxin on osteoclast spreading lead me to think that like other integrins v3 is activated through outside- in growth factor mediated mechanism and then assumes an extended conformation which has a
142 higher binding affinity towards Vitaxin. To confirm this theory, I repeated the attachment experiments in the presence of either M-CSF or a high concentration of Ca2+ which were the two methods I chose to activate or inactivate v3 respectively (207, 246). I was able to show that pretreating osteoclasts with M-CSF did indeed significantly increase Vitaxin’s inhibitory effect on osteoclast attachment while high Ca2+ levels had the opposite effect. Finally, PI3K was found to act as an intermediate signalling molecule mediating M-CSF’s activation to v3. This important finding sheds light on an aspect that should be considered when Vitaxin and other
v3 blocking antibodies are used for pathological bone loss and particularly when they are used as tumour imaging and drug delivery tools. Factors affecting v3 activation can be exploited to increase the efficacy of integrin blocking antibodies.
In the second part of this thesis the focus of my interest shifted from the integrin v3 to its ligands, the RGD containing bone matrix proteins. I was prompted by the lack of studies on the effects of FN and VN on osteoclast function. While FN’s effects on osteoclasts were overlooked, there are lots of studies showing that FN activates many cells during inflammatory bone loss such as chondrocytes and synoviocytes (427). More importantly, similar to OPN, VN and FN were found to be elevated in arthritic joints (6-8). I was also interested in examining if differentiating osteoclasts on these ECM proteins can alter their response to v3 blocking antibodies. I compared the effects of FN and VN to those of OPN on osteoclast formation, function and morphology using the well characterized RAW derived osteoclasts. I hypothesized that similar to OPN, FN and VN can modulate osteoclast formation and upregulate its function.
And indeed FN, VN and OPN each had a specific effect on osteoclast regulation and the most interesting effects were those of FN. While the number of osteoclasts formed on FN was significantly lower than those on OPN or TCP, those osteoclasts were hyper-resorptive. Careful
143 assessment of FN’s effect on osteoclastogenesis revealed that FN does not interfere with the recruitment of osteoclast precursors but rather inhibits their fusion and/or migration, an effect that was opposite to that seen on OPN. These effects were seen using both RAW and splenocyte derived osteoclasts with soluble and physically adsorbed FN. Confirming the inhibitory effect of
FN on osteoclast formation, adding FN to cultures differentiated on OPN significantly decreased the number of osteoclasts compared to the untreated OPN cultures. Using time-lapse microscopy, I have shown that FN decreases both the polarity and speed of migration of osteoclast precursors confirming that FN’s effect on osteoclast formation is partially the result of a decrease in pre-osteoclast migration (appendix figure 2A and B). This inhibitory effect of
FN was v3 specific and could be reversed by blocking the interaction between the integrin and FN using a specific antibody or RGD peptide, both of which stimulated osteoclast formation while not affecting their attachment. In contrast, osteoclast attachment on FN was 51 dependent. Additionally, osteoclasts formed on FN were predominantly compact in morphology and their planar area was the smallest. Moreover, the extracellular pH of FN cultures was lower than VN and OPN. The effect of low pH on osteoclast morphology was found to be the result of an increase in the frequency of the osteoclast morphological cycle where the cells alternate between a spread and a compact or polarized like morphology resulting in an increase in the time osteoclasts spend in a compact morphology. In contrast, FN’s stimulatory effect on osteoclastic resorption was the result of an increase in the percentage of activated osteoclasts and their increased efficiency to resorb denoted by the presence of multiple actin rings in individual osteoclasts. The stimulatory effect on FN was accompanied by an elevation in NO production. NO and IL-1β can be induced by one another during inflammation and the expression of each was also induced by FN in a variety of cells (393, 428-430). Therefore an
144 increase in their levels could contribute to osteoclast activation on FN. Furthermore, NO is required for IL-1 induced osteoclast activation (431).Thus, I analyzed their production patterns in relation to one another in conditioned media from FN cultures and found that the elevation of
NO on FN was followed by the induction of IL-1β. The use of an NO inhibitor confirmed that
NO acts upstream of IL-1β and that the production of IL-1β was FN specific since the NO inhibitor did not affect IL-1β levels on TCP. On the basis of these findings, I propose a model explaining the mechanism by which FN modulates osteoclast formation and activation. This model is based on the findings of studies examining the effect of FN and RGD peptides on fibroblasts and endothelial cell spreading and its correlation with cytoplasmic pH. These studies demonstrated that cell spreading on RGD peptides is associated with a more basal cytoplasmic pH and that this process also occurs on FN in response to the clustering of the integrin 51 which activates the Na+/H+ exchanger (394, 423, 424). Similar to those findings, RAW cells plated and differentiated on FN were notably more spread than those on VN, OPN and TCP 24-
48 hours after plating. Combining this observation with the lower extracellular pH on FN suggests that similar to fibroblasts, FN increases the cytoplasmic pH of RAW cells and osteoclast precursors through activating the Na+/H+ exchanger resulting in a drop in the extracellular pH as a first step of activation. Acidosis has been shown to increase the number and activity of resorbing osteoclasts both in vitro and in vivo (106, 107, 432-434). In fact, a drop in the pH below 7.4 is required for osteoclastic bone resorption and a decrease as small as 0.1 units causes a 2fold increase in resorption (435). It has been suggested that the effects of acidosis on osteoclasts are mediated through the H+-sensing ion channels such as transient receptor potential vanilloid 1(436). Since low extracellular pH increases the activity of V-
ATPases which is also responsible for the regulation of cytoplasmic and extracellular pH in
145 osteoclast precursors, this in turn contributes further to the decrease of the extracellular pH (107,
437). As a second step, osteoclasts on FN start to form in cultures between 48 and 72 hours.
This is the same time point where osteoclasts and osteoclast precursors would be induced by the decreased extracellular pH to upregulate V-ATPase activity and retain the acidosis in cultures.
Downstream from the activation of H+-sensing ion channels and in response to the increased H+ concentration, the transcription factor NFATc1 is activated by calcineurin and translocated to the nucleus resulting in osteoclast activation and increased resorptive activity (438). My findings suggest that a similar mechanism occurs on FN since osteoclasts on FN have increased compact morphology and their resorptive activity is significantly elevated compared to VN,
OPN and TCP. Concurrently, acidosis acts synergistically with RANKL by activating NFB during osteoclastogenesis (437) which can induce the expression of a multitude of pro- resorptive cytokines such as IL-1 which have both paracrine and autocrine effects on osteoclast resorption. IL-1 has a dual effect which can promote osteoclast activation both directly and indirectly (41, 170). IL-1 was shown to directly stimulate osteoclastic resorption and was recently found to increase the percentage of osteoclasts with a compact morphology (439). IL-1 also increased the activity and expression of V-ATPases in macrophages, which is another indirect method of reactivating osteoclasts in a pH mediated mechanism (440). Despite FN’s stimulatory effects on resorption via a pH dependent mechanism, FN had inhibitory effects on pre-osteoclast migration and fusion. Although I have only directly shown FN’s inhibitory effect on pre-osteoclast migration and only indirectly shown its effect on fusion, I believe that FN does affect both processes. The reason for my hypothesis is that osteoclast formation in 80% confluent cultures of RAW cells was still altered although pre-osteoclasts were found to be in contact with one another as illustrated in figure 4.1. Because the interaction between the RGD
146 motif of FN and v3 inhibits osteoclast formation is the predominant signal from the matrix during osteoclastogenesis. Thus, blocking this signalling pathway allows for the interaction of another FN domain with osteoclasts. That can occur possibly via the surface receptor CD44 which is present in osteoclasts and FN’s heparin binding domain (408). Activation of CD44 downstream signalling in osteoclasts was previously shown to induce osteoclast formation through promoting osteoclast fusion (52), a mechanism which might be occurring when v3 is blocked. Although histomorphometrical analysis of FN -/- osteoclast parameters were not conclusive, the trend of increased osteoclast formation does support our in vitro findings; however, more studies need to be performed to confirm those findings and double knockouts of both isoforms of FN found in bone might be needed to see a significant effect. In contrast to FN effects which supported my initial hypothesis, VN’s effects on osteoclastogenesis and function did not confirm my postulation. VN inhibited osteoclast formation through delaying both the recruitment and the multinucleation of pre-osteoclasts. While resorption on VN was slightly increased, it was not comparable to the stimulatory effect seen on FN. VN increased the number of pits suggesting that the ECM protein might be affecting osteoclast migration. Additionally,
VN increases the number of osteoclasts with a migratory morphology; suggesting that VN activates osteoclast motility and migration which is logical speculation since v3 activation through ligation increases osteoclast migration.
Taken together, the findings of this study have illustrated the important role the integrin
v3 and its ligands play in regulating osteoclasts. These data show that through subtle changes in the osteoclast’s external milieu and the composition of the ECM matrix, osteoclast activity can be suppressed or enhanced. The most intriguing findings of this project were those involving FN and its identification as a potential regulator of osteoclast functions through
147 cytokine like effects. FN can now be considered a new target for the control of bone loss specifically in inflammatory diseases where it is upregulated.
148 6. FUTURE DIRECTIONS
Because this study is the first to underscore FN’s mechanism of regulation of osteoclasts,
I was only able to touch the tip of the iceberg in terms of delineating its effects and mechanism of action. While I was able to answer some questions about FN, this study has generated many more research question that need to be addressed. One of those questions is how FN suppresses pre-osteoclast fusion. Of the molecules of interest due to their elucidated role in osteoclast fusion are DAP12 and FcR and their associated receptors making them a good target to investigate. Determining the signalling pathways downstream of FN binding to osteoclast surface receptors is another important aspect that requires further studies. Using the Signalling
PathwayFinder Array, the different signalling molecules up or down regulated can be identified after which the pathway can be determined.
FN is a complicated yet interesting protein to study. Contributing to its complexity is the fact that it can be alternatively spliced generating several isoforms with different functions in connective tissues. Another important characteristic of FN is its ability to form fibrils when activated by integrin binding; a process that was shown to be important for the formation of a proper collagen matrix formed by osteoblasts. During an internship at the University of
Heidelberg, Germany in the Summer of 2008, I initiated a study comparing the effects of two of
FN’s isoforms. I compared the effects of plasma and cellular isoforms as well as a recombinant
FN protein that can spontaneously undergo fibrillogenesis in vitro (sFN), on osteoclast formation and migratory behaviour (please see appendix). From this study, I found that sFN increases the migration of osteoclast precursors through both increasing their velocity and their polarity and this was associated with an increase in osteoclast formation. Since activation of small GTPases is needed for targeted cell migration, the next question to ask would be how sFN
149 affects the activation and the distribution of Rac1 and RhoA compared to the other forms of FN.
Finally, using GTPase inhibitors and/or siRNA on pre-osteoclasts differentiated on sFN can confirm their involvement in enhancing cell migration and the consequent increase in osteoclast formation.
I also made another interesting observation which is that osteoclast and pre-osteoclasts differentiated on FN degrade the FN coating in cultures as seen in figure 6.1 A, B and C. This observation was made by other groups as well where they showed that matrix proteins are degraded by osteoclasts through the enzymatic properties podosomes possess (136, 137). The enzymatic degradation of FN results in the generation of smaller fragments; these fragments possess different properties from the intact protein and are highly prevalent in arthritic joints
(389, 427). The isolation and identification of the FN fragments generated by osteoclasts in culture is an important step. This would be followed by recreating these fragments and testing their effects on osteoclast function which is another project I would like to pursue.
Finally, during my internship, I also got involved in a side project where I tested the effects of geometrically nanopatterned RGD decorated surfaces which are present at different densities on osteoclast attachment and podosome distribution. Preliminary results from my experiments show that osteoclasts exhibit a unique multiple podosome ring formation after 12 hours of attachment on high density nanopatterns compared to homogenous RGD coating
(figure 6.2 and 6.3). Further experiments need to be conducted to elucidate the effect of RGD density on osteoclast attachment.
150
Figure 6.1 Osteoclasts and pre-osteoclasts degrade fluorescently labelled FN coating.
RAW cells were plated on live cell imaging chambers coated with 10 μg/ml rhodamine labelled bovine pFN and differentiated in presence of 75 ng/ml RANKL. After 72 hours the cells were fixed, permeablized and stained for F-actin using Alexa Fluor 488 phalloidin (green) and immunostained for vinculin using an anti-vinculin antibody and an Alexa 648 2nd antibody (red).
Panel A and B show images of (ii) pre-osteoclasts stained for F-actin (green) on (i, ii) pFN (red).
Panel C shows images of (ii) an osteoclast stained for F-actin (green) and vinculin (red) on (i, ii) pFN (red false coloured to purple).
151
Figure 6.2 Osteoclasts on a high density RGD coated nanopattern exhibit multiple podosome rings. RAW cells were differentiated in the presence of 75 ng/ml RANKL. After 72 hours the osteoclasts were non-enzymatically detached and replated on RGD coated nanopattern. cells were allowed to attach for 24 hours before they were fixed, permeablized and stained for F-actin using Alexa Fluor 488 phalloidin (green) and immunostained for vinculin using an anti-vinculin antibody and an Alexa 648 2nd antibody (red). Panel A and B show images of osteoclasts stained for F-actin (green) (i) and vinculin (red) (ii) and the two channels merged (iii).
152
Figure 6.3 Osteoclasts on homogenous RGD coated surfaces exhibit normal podosome arrangment. RAW cells were differentiated in the presence of 75 ng/ml RANKL. After 72 hours the osteoclasts were non-enzymatically detached and replated on RGD coated gold surface. cells were allowed to attach for 24 hours before they were fixed, permeablized and stained for F-actin using Alexa Fluor 488 phalloidin (green) and immunostained for vinculin using an anti-vinculin antibody and an Alexa 648 2nd antibody (red). Panel A and B show images of osteoclasts stained for F-actin (green) (i) and vinculin (red) (ii) and the two channels merged (iii).
153 APPENDIX
SuperFibronectin Induces Osteoclast Formation through Increasing Pre-osteoclast
Migration
The ECM proteins are a component of the external milieu of cells which provides the essential anchorage for cell survival and enables cells to sense and react to their external stimuli by modulating their behaviour. FN is an abundant pleiotropic matrix protein that is essential for many cell processes and plays a critical role in embryogenesis (386). According to the type of cells producing the ECM protein, FN can be divided into two main types; plasma FN (pFN) synthesized by hepatocytes (which is the type of FN studied in chapters 2 and 3 and is enriched in the plasma and cellular FN (cFN) which is produced by stromal and osteoblasts (398). The main difference between cFN and pFN is the extra domains added via alternative splicing that are found in cFN known as extra type III domain A (EIIIA) and B (EIIIB) (318). In bone, both types of FN are present in the matrix and play differential roles. While cFN deletion affects osteoblast function but not bone structure, deleting pFN which is also the more abundant FN type in bone, decreases bone density and mineralization and affects collagen matrix structure
(15). In connective tissues including bone, both types of FN are present as fibrils formed through cell driven activation of the molecule through mechanical forces applied via integrin binding and the actin cytoskeleton (322). In an attempt to recreate this process in vitro in the absence of cells, a recombinant protein was developed using an FN type III repeat fragment called anastellin also known as super FN (sFN) (441). When added to the soluble dimeric pFN, anastellin binds to pFN and causes the spontaneous formation of disulphide bonds between pFN molecules resulting in pFN multimerization and matrix formation (442).
154 In this study, osteoclast formation was compared on pFN, cFN and sFN using RAW cell derived osteoclasts. Osteoclast formation on sFN was significantly enhanced compared to pFN and cFN at both 48 and 72 hours compared to pFN, cFN and TCP (figure A.1A, B). pFN on the other hand decreased osteoclast formation compared to sFN, cFN and TCP. Because the process of pre-osteoclast migration is an essential step for osteoclast formation, RAW cell migration in the presence of RANKL was assessed for 6 hours after plating using time-lapse microscopy. We chose to focus on two parameters in the recorded movies that pertained to cell migration; cell velocity and polarity determined by their aspect ratio. Proper migration requires both an increase in cell velocity in conjunction with an increase in their polarity which is the cell morphology associated with targeted migration. Migratory cells are more elongated and their plasma membrane can be divided into a leading edge and a trailing edge. An increase in cell polarity is denoted by an increase in its aspect ratio. Both cell velocity and polarity were increased on sFN
(figure A.2 A, B). In contrast, both parameters were decreased on pFN and cFN. A strong correlation can be seen between number of osteoclast formed on each FN type and the migration parameters assessed during the first 6 hours of differentiation. Moreover, podosome formation on sFN and cFN was analysed after the cells were stained for two podosome molecular markers,
F-actin (green) and vinculin (red) (figure A.3A, B). Ring-like structures were observed in osteoclasts on sFN and cFN. However, on sFN the ring podosome structure consisted of a single band of F-actin and outer and inner bands of vinculin (figure A.3A). That structure resembled the sealing zone which is the attachment structure seen only on mineralized substrates. On cFN, the ring structure contained punctate podosomes found in a cloud of vinculin, which is the typical structure of a podosome ring on TCP and ECM protein coated non-mineralized surfaces
(figure A.3B).
155 These results show a differential regulation of osteoclast formation by FNs.
Interestingly, the effects of each FN on osteoclast formation can be predicted by the migratory behaviour of pre-osteoclasts as early as 6 hours after plating. For enhanced osteoclast formation as seen on sFN, both parameters of cell migration (velocity and polarity) need to be elevated.
The structure of podosome rings in osteoclasts formed on the different FNs varies with the type of FN and could mimic the structure of a sealing zone seen on bone and dentine. Finally, sFN plays a unique role in modulating osteoclast behaviour and adhesion structure potentially due to its polymeric matrix formation properties. More experiments are needed to determine its mechanism of action.
156
Figure A.1 Osteoclast formation is enhanced by sFN and suppressed by pFN.
RAW cells were plated on TCP dishes coated with 10 μg/ml sFN, pFN and cFN. RAW cells were differentiated into osteoclasts for 48 (A) and 72 (B) hours in presence of 75 ng/ml
RANKL. The cultures were fixed and stained for TRAP activity and the number of TRAP+ osteoclasts was counted. Each data point represents the pooled results from four wells per treatment and is expressed per well (means ± SD). a P < 0.05 compared to sFN, b P < 0.05 compared to cFN, c P < 0.05 compared to TCP.
157
158
Figure A.2 Pre-osteoclast velocity and polarity are increased on sFN compared to pFN and cFN. RAW cells were plated in live cell imaging chambers coated with 10 μg/ml sFN, pFN and cFN in presence of 75 ng/ml RANKL. Using time-lapse microscopy, images were acquired at 3 minute intervals for 5 fields per treatment group for up to 6 hours. Twenty cells from different movies were analysed using Image J software and the outline of the cells was traced manually in each frame. From the multiple generated measurements, two measurements were assessed, (A) cell polarity which was determined by aspect ratio and (B) velocity which was calculated from
X and Y coordinates of the cells at each time point. The number of events of the occurrence of a certain aspect ratio or velocity was counted and was binned in ranges.
159
Figure A.3 Osteoclasts formed on sFN exhibit an atypical “sealing zone” like attachment structure while those on cFN contain a typical podosome ring. RAW cells were plated on live cell imaging chambers coated with 10 μg/ml sFN and cFN in presence of 75 ng/ml RANKL.
After 72 hours the cells were fixed, permeablized and stained for F-actin using Alexa Fluor 488 phalloidin (green) and immunostained for vinculin using an anti-viculin antibody and a TRTC
2nd antibody (red). Panel A shows images of an osteoclast on sFN; image (i) F-actin, (ii) vinculin and (iii) merged image. Panel B shows images of an osteoclast on cFN; image (i) F- actin, (ii) vinculin and (iii) merged image.
160 REFERENCES
(1) Martin, L. (2004) Rheumatoid arthritis: symptoms, diagnosis, and management. Nurs Times 100, 40-4.
(2) Bendele, A., McComb, J., Gould, T., McAbee, T., Sennello, G., Chlipala, E., and Guy, M. (1999) Animal models of arthritis: relevance to human disease. Toxicol Pathol 27, 134-42.
(3) Romas, E., Gillespie, M. T., and Martin, T. J. (2002) Involvement of receptor activator of NFkappaB ligand and tumor necrosis factor-alpha in bone destruction in rheumatoid arthritis. Bone 30, 340-6.
(4) McHugh, K. P., Hodivala-Dilke, K., Zheng, M. H., Namba, N., Lam, J., Novack, D., Feng, X., Ross, F. P., Hynes, R. O., and Teitelbaum, S. L. (2000) Mice lacking beta3 integrins are osteosclerotic because of dysfunctional osteoclasts. J Clin Invest 105, 433- 40.
(5) Nakamura, I., Duong le, T., Rodan, S. B., and Rodan, G. A. (2007) Involvement of alpha(v)beta3 integrins in osteoclast function. J Bone Miner Metab 25, 337-44.
(6) Carsons, S., Mosesson, M. W., and Diamond, H. S. (1981) Detection and quantitation of fibronectin in synovial fluid from patients with rheumatic disease. Arthritis Rheum 24, 1261-7.
(7) Ohshima, S., Yamaguchi, N., Nishioka, K., Mima, T., Ishii, T., Umeshita-Sasai, M., Kobayashi, H., Shimizu, M., Katada, Y., Wakitani, S., Murata, N., Nomura, S., Matsuno, H., Katayama, R., Kon, S., Inobe, M., Uede, T., Kawase, I., and Saeki, Y. (2002) Enhanced local production of osteopontin in rheumatoid joints. J Rheumatol 29, 2061-7.
(8) Rosenblum, G., and Carsons, S. (1996) Quantitation and distribution of vitronectin in synovial fluid and tissue of patients with rheumatic disease. Clin Exp Rheumatol 14, 31- 6.
(9) Turner, C. H. (2006) Bone strength: current concepts. Ann N Y Acad Sci 1068, 429-46.
(10) Lanyon, L. E., and Baggott, D. G. (1976) Mechanical function as an influence on the structure and form of bone. J Bone Joint Surg Br 58-B, 436-43.
(11) Gupta, H. S., Seto, J., Wagermaier, W., Zaslansky, P., Boesecke, P., and Fratzl, P. (2006) Cooperative deformation of mineral and collagen in bone at the nanoscale. Proc Natl Acad Sci U S A 103, 17741-6.
(12) Fantner, G. E., Hassenkam, T., Kindt, J. H., Weaver, J. C., Birkedal, H., Pechenik, L., Cutroni, J. A., Cidade, G. A., Stucky, G. D., Morse, D. E., and Hansma, P. K. (2005)
161 Sacrificial bonds and hidden length dissipate energy as mineralized fibrils separate during bone fracture. Nat Mater 4, 612-6.
(13) Baht, G. S., Hunter, G. K., and Goldberg, H. A. (2008) Bone sialoprotein-collagen interaction promotes hydroxyapatite nucleation. Matrix Biol 27, 600-8.
(14) Malaval, L., Wade-Gueye, N. M., Boudiffa, M., Fei, J., Zirngibl, R., Chen, F., Laroche, N., Roux, J. P., Burt-Pichat, B., Duboeuf, F., Boivin, G., Jurdic, P., Lafage-Proust, M. H., Amedee, J., Vico, L., Rossant, J., and Aubin, J. E. (2008) Bone sialoprotein plays a functional role in bone formation and osteoclastogenesis. J Exp Med 205, 1145-53.
(15) Bentmann, A., Kawelke, N., Moss, D., Zentgraf, H., Bala, Y., Berger, I., Gasser, J. A., and Nakchbandi, I. A. (2009) Circulating Fibronectin Affects Bone Matrix While Osteoblast Fibronectin Modulates Osteoblast Function. J Bone Miner Res.
(16) Mansergh, F. C., Wells, T., Elford, C., Evans, S. L., Perry, M. J., Evans, M. J., and Evans, B. A. (2007) Osteopenia in Sparc (osteonectin)-deficient mice: characterization of phenotypic determinants of femoral strength and changes in gene expression. Physiol Genomics 32, 64-73.
(17) Chellaiah, M. A., Kizer, N., Biswas, R., Alvarez, U., Strauss-Schoenberger, J., Rifas, L., Rittling, S. R., Denhardt, D. T., and Hruska, K. A. (2003) Osteopontin deficiency produces osteoclast dysfunction due to reduced CD44 surface expression. Mol Biol Cell 14, 173-89.
(18) Xu, T., Bianco, P., Fisher, L. W., Longenecker, G., Smith, E., Goldstein, S., Bonadio, J., Boskey, A., Heegaard, A. M., Sommer, B., Satomura, K., Dominguez, P., Zhao, C., Kulkarni, A. B., Robey, P. G., and Young, M. F. (1998) Targeted disruption of the biglycan gene leads to an osteoporosis-like phenotype in mice. Nat Genet 20, 78-82.
(19) van den Bos, T., Speijer, D., Bank, R. A., Bromme, D., and Everts, V. (2008) Differences in matrix composition between calvaria and long bone in mice suggest differences in biomechanical properties and resorption: Special emphasis on collagen. Bone 43, 459-68.
(20) Seeman, E. (2008) in Principles of Bone Biology (Bilezikian, R., Martin, Ed.) pp 3-28, Elsevier.
(21) Hazenberg, J. G., Freeley, M., Foran, E., Lee, T. C., and Taylor, D. (2006) Microdamage: a cell transducing mechanism based on ruptured osteocyte processes. J Biomech 39, 2096-103.
(22) Lane, N. E., Yao, W., Balooch, M., Nalla, R. K., Balooch, G., Habelitz, S., Kinney, J. H., and Bonewald, L. F. (2006) Glucocorticoid-treated mice have localized changes in trabecular bone material properties and osteocyte lacunar size that are not observed in placebo-treated or estrogen-deficient mice. J Bone Miner Res 21, 466-76.
162 (23) Aguirre, J. I., Plotkin, L. I., Stewart, S. A., Weinstein, R. S., Parfitt, A. M., Manolagas, S. C., and Bellido, T. (2006) Osteocyte apoptosis is induced by weightlessness in mice and precedes osteoclast recruitment and bone loss. J Bone Miner Res 21, 605-15.
(24) Clark, W. D., Smith, E. L., Linn, K. A., Paul-Murphy, J. R., Muir, P., and Cook, M. E. (2005) Osteocyte apoptosis and osteoclast presence in chicken radii 0-4 days following osteotomy. Calcif Tissue Int 77, 327-36.
(25) Yao, G. Q., Sun, B. H., Weir, E. C., and Insogna, K. L. (2002) A role for cell-surface CSF-1 in osteoblast-mediated osteoclastogenesis. Calcif Tissue Int 70, 339-46.
(26) Yoshida, H., Hayashi, S., Kunisada, T., Ogawa, M., Nishikawa, S., Okamura, H., Sudo, T., Shultz, L. D., and Nishikawa, S. (1990) The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345, 442-4.
(27) Wong, B. R., Rho, J., Arron, J., Robinson, E., Orlinick, J., Chao, M., Kalachikov, S., Cayani, E., Bartlett, F. S., 3rd, Frankel, W. N., Lee, S. Y., and Choi, Y. (1997) TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N- terminal kinase in T cells. J Biol Chem 272, 25190-4.
(28) Yasuda, H., Shima, N., Nakagawa, N., Yamaguchi, K., Kinosaki, M., Mochizuki, S., Tomoyasu, A., Yano, K., Goto, M., Murakami, A., Tsuda, E., Morinaga, T., Higashio, K., Udagawa, N., Takahashi, N., and Suda, T. (1998) Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci U S A 95, 3597-602.
(29) Ainola, M., Mandelin, J., Liljestrom, M., Konttinen, Y. T., and Salo, J. (2008) Imbalanced expression of RANKL and osteoprotegerin mRNA in pannus tissue of rheumatoid arthritis. Clin Exp Rheumatol 26, 240-6.
(30) Hofbauer, L. C., Gori, F., Riggs, B. L., Lacey, D. L., Dunstan, C. R., Spelsberg, T. C., and Khosla, S. (1999) Stimulation of osteoprotegerin ligand and inhibition of osteoprotegerin production by glucocorticoids in human osteoblastic lineage cells: potential paracrine mechanisms of glucocorticoid-induced osteoporosis. Endocrinology 140, 4382-9.
(31) Ma, Y. L., Cain, R. L., Halladay, D. L., Yang, X., Zeng, Q., Miles, R. R., Chandrasekhar, S., Martin, T. J., and Onyia, J. E. (2001) Catabolic effects of continuous human PTH (1--38) in vivo is associated with sustained stimulation of RANKL and inhibition of osteoprotegerin and gene-associated bone formation. Endocrinology 142, 4047-54.
(32) Kaji, H., Sugimoto, T., Kanatani, M., Nishiyama, K., and Chihara, K. (1997) Dexamethasone stimulates osteoclast-like cell formation by directly acting on hemopoietic blast cells and enhances osteoclast-like cell formation stimulated by parathyroid hormone and prostaglandin E2. J Bone Miner Res 12, 734-41.
163 (33) Wani, M. R., Fuller, K., Kim, N. S., Choi, Y., and Chambers, T. (1999) Prostaglandin E2 cooperates with TRANCE in osteoclast induction from hemopoietic precursors: synergistic activation of differentiation, cell spreading, and fusion. Endocrinology 140, 1927-35.
(34) Suda, T., Takahashi, N., Udagawa, N., Jimi, E., Gillespie, M. T., and Martin, T. J. (1999) Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. Endocr Rev 20, 345-57.
(35) Takayanagi, H. (2009) Osteoimmunology and the effects of the immune system on bone. Nat Rev Rheumatol 5, 667-76.
(36) Rubin, J., Biskobing, D. M., Jadhav, L., Fan, D., Nanes, M. S., Perkins, S., and Fan, X. (1998) Dexamethasone promotes expression of membrane-bound macrophage colony- stimulating factor in murine osteoblast-like cells. Endocrinology 139, 1006-12.
(37) Jia, D., O'Brien, C. A., Stewart, S. A., Manolagas, S. C., and Weinstein, R. S. (2006) Glucocorticoids act directly on osteoclasts to increase their life span and reduce bone density. Endocrinology 147, 5592-9.
(38) Kanzawa, M., Sugimoto, T., Kanatani, M., and Chihara, K. (2000) Involvement of osteoprotegerin/osteoclastogenesis inhibitory factor in the stimulation of osteoclast formation by parathyroid hormone in mouse bone cells. Eur J Endocrinol 142, 661-4.
(39) Fu, Q., Manolagas, S. C., and O'Brien, C. A. (2006) Parathyroid hormone controls receptor activator of NF-kappaB ligand gene expression via a distant transcriptional enhancer. Mol Cell Biol 26, 6453-68.
(40) Nakashima, T., Kobayashi, Y., Yamasaki, S., Kawakami, A., Eguchi, K., Sasaki, H., and Sakai, H. (2000) Protein expression and functional difference of membrane-bound and soluble receptor activator of NF-kappaB ligand: modulation of the expression by osteotropic factors and cytokines. Biochem Biophys Res Commun 275, 768-75.
(41) Tani-Ishii, N., Tsunoda, A., Teranaka, T., and Umemoto, T. (1999) Autocrine regulation of osteoclast formation and bone resorption by IL-1 alpha and TNF alpha. J Dent Res 78, 1617-23.
(42) Fuller, K., Kirstein, B., and Chambers, T. J. (2006) Murine osteoclast formation and function: differential regulation by humoral agents. Endocrinology 147, 1979-85.
(43) Takahashi, N., Yamana, H., Yoshiki, S., Roodman, G. D., Mundy, G. R., Jones, S. J., Boyde, A., and Suda, T. (1988) Osteoclast-like cell formation and its regulation by osteotropic hormones in mouse bone marrow cultures. Endocrinology 122, 1373-82.
(44) Takahashi, N., Akatsu, T., Udagawa, N., Sasaki, T., Yamaguchi, A., Moseley, J. M., Martin, T. J., and Suda, T. (1988) Osteoblastic cells are involved in osteoclast formation. Endocrinology 123, 2600-2. 164 (45) Boyle, W. J., Simonet, W. S., and Lacey, D. L. (2003) Osteoclast differentiation and activation. Nature 423, 337-42.
(46) Hattersley, G., Owens, J., Flanagan, A. M., and Chambers, T. J. (1991) Macrophage colony stimulating factor (M-CSF) is essential for osteoclast formation in vitro. Biochem Biophys Res Commun 177, 526-31.
(47) Lacey, D. L., Timms, E., Tan, H. L., Kelley, M. J., Dunstan, C. R., Burgess, T., Elliott, R., Colombero, A., Elliott, G., Scully, S., Hsu, H., Sullivan, J., Hawkins, N., Davy, E., Capparelli, C., Eli, A., Qian, Y. X., Kaufman, S., Sarosi, I., Shalhoub, V., Senaldi, G., Guo, J., Delaney, J., and Boyle, W. J. (1998) Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93, 165-76.
(48) Koga, T., Inui, M., Inoue, K., Kim, S., Suematsu, A., Kobayashi, E., Iwata, T., Ohnishi, H., Matozaki, T., Kodama, T., Taniguchi, T., Takayanagi, H., and Takai, T. (2004) Costimulatory signals mediated by the ITAM motif cooperate with RANKL for bone homeostasis. Nature 428, 758-63.
(49) Balkan, W., Martinez, A. F., Fernandez, I., Rodriguez, M. A., Pang, M., and Troen, B. R. (2009) Identification of NFAT binding sites that mediate stimulation of cathepsin K promoter activity by RANK ligand. Gene 446, 90-8.
(50) Takayanagi, H. (2007) Osteoimmunology: shared mechanisms and crosstalk between the immune and bone systems. Nat Rev Immunol 7, 292-304.
(51) Crotti, T. N., Flannery, M., Walsh, N. C., Fleming, J. D., Goldring, S. R., and McHugh, K. P. (2005) NFATc1 directly induces the human beta3 integrin gene in osteoclast differentiation. J Musculoskelet Neuronal Interact 5, 335-7.
(52) Sterling, H., Saginario, C., and Vignery, A. (1998) CD44 occupancy prevents macrophage multinucleation. J Cell Biol 143, 837-47.
(53) Vignery, A. (2000) Osteoclasts and giant cells: macrophage-macrophage fusion mechanism. Int J Exp Pathol 81, 291-304.
(54) Yagi, M., Miyamoto, T., Sawatani, Y., Iwamoto, K., Hosogane, N., Fujita, N., Morita, K., Ninomiya, K., Suzuki, T., Miyamoto, K., Oike, Y., Takeya, M., Toyama, Y., and Suda, T. (2005) DC-STAMP is essential for cell-cell fusion in osteoclasts and foreign body giant cells. J Exp Med 202, 345-51.
(55) Lee, S. H., Rho, J., Jeong, D., Sul, J. Y., Kim, T., Kim, N., Kang, J. S., Miyamoto, T., Suda, T., Lee, S. K., Pignolo, R. J., Koczon-Jaremko, B., Lorenzo, J., and Choi, Y. (2006) v-ATPase V0 subunit d2-deficient mice exhibit impaired osteoclast fusion and increased bone formation. Nat Med 12, 1403-9.
165 (56) Kim, K., Lee, S. H., Ha Kim, J., Choi, Y., and Kim, N. (2008) NFATc1 induces osteoclast fusion via up-regulation of Atp6v0d2 and the dendritic cell-specific transmembrane protein (DC-STAMP). Mol Endocrinol 22, 176-85.
(57) Nakano, K., Okada, Y., Saito, K., Tanikawa, R., Sawamukai, N., Sasaguri, Y., Kohro, T., Wada, Y., Kodama, T., and Tanaka, Y. (2007) Rheumatoid synovial endothelial cells produce macrophage colony-stimulating factor leading to osteoclastogenesis in rheumatoid arthritis. Rheumatology (Oxford) 46, 597-603.
(58) Kitaura, H., Zhou, P., Kim, H. J., Novack, D. V., Ross, F. P., and Teitelbaum, S. L. (2005) M-CSF mediates TNF-induced inflammatory osteolysis. J Clin Invest 115, 3418- 27.
(59) Wei, S., Kitaura, H., Zhou, P., Ross, F. P., and Teitelbaum, S. L. (2005) IL-1 mediates TNF-induced osteoclastogenesis. J Clin Invest 115, 282-90.
(60) Yagiz, K., and Rittling, S. R. (2009) Both cell-surface and secreted CSF-1 expressed by tumor cells metastatic to bone can contribute to osteoclast activation. Exp Cell Res 315, 2442-52.
(61) Takahashi, N., Udagawa, N., Akatsu, T., Tanaka, H., Isogai, Y., and Suda, T. (1991) Deficiency of osteoclasts in osteopetrotic mice is due to a defect in the local microenvironment provided by osteoblastic cells. Endocrinology 128, 1792-6.
(62) Tondravi, M. M., McKercher, S. R., Anderson, K., Erdmann, J. M., Quiroz, M., Maki, R., and Teitelbaum, S. L. (1997) Osteopetrosis in mice lacking haematopoietic transcription factor PU.1. Nature 386, 81-4.
(63) Insogna, K. L., Sahni, M., Grey, A. B., Tanaka, S., Horne, W. C., Neff, L., Mitnick, M., Levy, J. B., and Baron, R. (1997) Colony-stimulating factor-1 induces cytoskeletal reorganization and c-src-dependent tyrosine phosphorylation of selected cellular proteins in rodent osteoclasts. J Clin Invest 100, 2476-85.
(64) Gingery, A., Bradley, E., Shaw, A., and Oursler, M. J. (2003) Phosphatidylinositol 3- kinase coordinately activates the MEK/ERK and AKT/NFkappaB pathways to maintain osteoclast survival. J Cell Biochem 89, 165-79.
(65) McGill, G. G., Horstmann, M., Widlund, H. R., Du, J., Motyckova, G., Nishimura, E. K., Lin, Y. L., Ramaswamy, S., Avery, W., Ding, H. F., Jordan, S. A., Jackson, I. J., Korsmeyer, S. J., Golub, T. R., and Fisher, D. E. (2002) Bcl2 regulation by the melanocyte master regulator Mitf modulates lineage survival and melanoma cell viability. Cell 109, 707-18.
(66) Lagasse, E., and Weissman, I. L. (1997) Enforced expression of Bcl-2 in monocytes rescues macrophages and partially reverses osteopetrosis in op/op mice. Cell 89, 1021- 31.
166 (67) Arai, F., Miyamoto, T., Ohneda, O., Inada, T., Sudo, T., Brasel, K., Miyata, T., Anderson, D. M., and Suda, T. (1999) Commitment and differentiation of osteoclast precursor cells by the sequential expression of c-Fms and receptor activator of nuclear factor kappaB (RANK) receptors. J Exp Med 190, 1741-54.
(68) Nakashima, T., Wada, T., and Penninger, J. M. (2003) RANKL and RANK as novel therapeutic targets for arthritis. Curr Opin Rheumatol 15, 280-7.
(69) Pettit, A. R., Ji, H., von Stechow, D., Muller, R., Goldring, S. R., Choi, Y., Benoist, C., and Gravallese, E. M. (2001) TRANCE/RANKL knockout mice are protected from bone erosion in a serum transfer model of arthritis. Am J Pathol 159, 1689-99.
(70) Kong, Y. Y., Yoshida, H., Sarosi, I., Tan, H. L., Timms, E., Capparelli, C., Morony, S., Oliveira-dos-Santos, A. J., Van, G., Itie, A., Khoo, W., Wakeham, A., Dunstan, C. R., Lacey, D. L., Mak, T. W., Boyle, W. J., and Penninger, J. M. (1999) OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397, 315-23.
(71) Wong, B. R., Josien, R., Lee, S. Y., Vologodskaia, M., Steinman, R. M., and Choi, Y. (1998) The TRAF family of signal transducers mediates NF-kappaB activation by the TRANCE receptor. J Biol Chem 273, 28355-9.
(72) Kobayashi, N., Kadono, Y., Naito, A., Matsumoto, K., Yamamoto, T., Tanaka, S., and Inoue, J. (2001) Segregation of TRAF6-mediated signaling pathways clarifies its role in osteoclastogenesis. Embo J 20, 1271-80.
(73) Ikeda, F., Nishimura, R., Matsubara, T., Tanaka, S., Inoue, J., Reddy, S. V., Hata, K., Yamashita, K., Hiraga, T., Watanabe, T., Kukita, T., Yoshioka, K., Rao, A., and Yoneda, T. (2004) Critical roles of c-Jun signaling in regulation of NFAT family and RANKL- regulated osteoclast differentiation. J Clin Invest 114, 475-84.
(74) Asagiri, M., Sato, K., Usami, T., Ochi, S., Nishina, H., Yoshida, H., Morita, I., Wagner, E. F., Mak, T. W., Serfling, E., and Takayanagi, H. (2005) Autoamplification of NFATc1 expression determines its essential role in bone homeostasis. J Exp Med 202, 1261-9.
(75) Sun, L., Peng, Y., Zaidi, N., Zhu, L. L., Iqbal, J., Yamoah, K., Wang, X., Liu, P., Abe, E., Moonga, B. S., Epstein, S., and Zaidi, M. (2007) Evidence that calcineurin is required for the genesis of bone-resorbing osteoclasts. Am J Physiol Renal Physiol 292, F285-91.
(76) Mao, D., Epple, H., Uthgenannt, B., Novack, D. V., and Faccio, R. (2006) PLCgamma2 regulates osteoclastogenesis via its interaction with ITAM proteins and GAB2. J Clin Invest 116, 2869-79.
(77) Matsumoto, M., Kogawa, M., Wada, S., Takayanagi, H., Tsujimoto, M., Katayama, S., Hisatake, K., and Nogi, Y. (2004) Essential role of p38 mitogen-activated protein kinase 167 in cathepsin K gene expression during osteoclastogenesis through association of NFATc1 and PU.1. J Biol Chem 279, 45969-79.
(78) Sato, K., Suematsu, A., Nakashima, T., Takemoto-Kimura, S., Aoki, K., Morishita, Y., Asahara, H., Ohya, K., Yamaguchi, A., Takai, T., Kodama, T., Chatila, T. A., Bito, H., and Takayanagi, H. (2006) Regulation of osteoclast differentiation and function by the CaMK-CREB pathway. Nat Med 12, 1410-6.
(79) Mocsai, A., Humphrey, M. B., Van Ziffle, J. A., Hu, Y., Burghardt, A., Spusta, S. C., Majumdar, S., Lanier, L. L., Lowell, C. A., and Nakamura, M. C. (2004) The immunomodulatory adapter proteins DAP12 and Fc receptor gamma-chain (FcRgamma) regulate development of functional osteoclasts through the Syk tyrosine kinase. Proc Natl Acad Sci U S A 101, 6158-63.
(80) Zou, W., Kitaura, H., Reeve, J., Long, F., Tybulewicz, V. L., Shattil, S. J., Ginsberg, M. H., Ross, F. P., and Teitelbaum, S. L. (2007) Syk, c-Src, the alphavbeta3 integrin, and ITAM immunoreceptors, in concert, regulate osteoclastic bone resorption. J Cell Biol 176, 877-88.
(81) Zou, W., Reeve, J. L., Liu, Y., Teitelbaum, S. L., and Ross, F. P. (2008) DAP12 couples c-Fms activation to the osteoclast cytoskeleton by recruitment of Syk. Mol Cell 31, 422- 31.
(82) Reeve, J. L., Zou, W., Liu, Y., Maltzman, J. S., Ross, F. P., and Teitelbaum, S. L. (2009) SLP-76 couples Syk to the osteoclast cytoskeleton. J Immunol 183, 1804-12.
(83) Yablonski, D., and Weiss, A. (2001) Mechanisms of signaling by the hematopoietic- specific adaptor proteins, SLP-76 and LAT and their B cell counterpart, BLNK/SLP-65. Adv Immunol 79, 93-128.
(84) Lee, S. H., Kim, T., Jeong, D., Kim, N., and Choi, Y. (2008) The tec family tyrosine kinase Btk Regulates RANKL-induced osteoclast maturation. J Biol Chem 283, 11526- 34.
(85) Shinohara, M., Koga, T., Okamoto, K., Sakaguchi, S., Arai, K., Yasuda, H., Takai, T., Kodama, T., Morio, T., Geha, R. S., Kitamura, D., Kurosaki, T., Ellmeier, W., and Takayanagi, H. (2008) Tyrosine kinases Btk and Tec regulate osteoclast differentiation by linking RANK and ITAM signals. Cell 132, 794-806.
(86) Epple, H., Cremasco, V., Zhang, K., Mao, D., Longmore, G. D., and Faccio, R. (2008) Phospholipase Cgamma2 modulates integrin signaling in the osteoclast by affecting the localization and activation of Src kinase. Mol Cell Biol 28, 3610-22.
(87) Palokangas, H., Mulari, M., and Vaananen, H. K. (1997) Endocytic pathway from the basal plasma membrane to the ruffled border membrane in bone-resorbing osteoclasts. J Cell Sci 110 ( Pt 15), 1767-80.
168 (88) Salo, J., Metsikko, K., Palokangas, H., Lehenkari, P., and Vaananen, H. K. (1996) Bone- resorbing osteoclasts reveal a dynamic division of basal plasma membrane into two different domains. J Cell Sci 109 ( Pt 2), 301-7.
(89) Nesbitt, S. A., and Horton, M. A. (1997) Trafficking of matrix collagens through bone- resorbing osteoclasts. Science 276, 266-9.
(90) Abu-Amer, Y., Ross, F. P., Schlesinger, P., Tondravi, M. M., and Teitelbaum, S. L. (1997) Substrate recognition by osteoclast precursors induces C-src/microtubule association. J Cell Biol 137, 247-58.
(91) Lakkakorpi, P., Tuukkanen, J., Hentunen, T., Jarvelin, K., and Vaananen, K. (1989) Organization of osteoclast microfilaments during the attachment to bone surface in vitro. J Bone Miner Res 4, 817-25.
(92) Vaananen, H. K., and Horton, M. (1995) The osteoclast clear zone is a specialized cell- extracellular matrix adhesion structure. J Cell Sci 108 ( Pt 8), 2729-32.
(93) Bekker, P. J., and Gay, C. V. (1990) Biochemical characterization of an electrogenic vacuolar proton pump in purified chicken osteoclast plasma membrane vesicles. J Bone Miner Res 5, 569-79.
(94) Sahara, T., Itoh, K., Debari, K., and Sasaki, T. (2003) Specific biological functions of vacuolar-type H(+)-ATPase and lysosomal cysteine proteinase, cathepsin K, in osteoclasts. Anat Rec A Discov Mol Cell Evol Biol 270, 152-61.
(95) Asotra, S., Gupta, A. K., Sodek, J., Aubin, J. E., and Heersche, J. N. (1994) Carbonic anhydrase II mRNA expression in individual osteoclasts under "resorbing" and "nonresorbing" conditions. J Bone Miner Res 9, 1115-22.
(96) Laitala, T., and Vaananen, H. K. (1994) Inhibition of bone resorption in vitro by antisense RNA and DNA molecules targeted against carbonic anhydrase II or two subunits of vacuolar H(+)-ATPase. J Clin Invest 93, 2311-8.
(97) Nordstrom, T., Rotstein, O. D., Romanek, R., Asotra, S., Heersche, J. N., Manolson, M. F., Brisseau, G. F., and Grinstein, S. (1995) Regulation of cytoplasmic pH in osteoclasts. Contribution of proton pumps and a proton-selective conductance. J Biol Chem 270, 2203-12.
(98) Bastani, B., Ross, F. P., Kopito, R. R., and Gluck, S. L. (1996) Immunocytochemical localization of vacuolar H+-ATPase and Cl--HCO3- anion exchanger (erythrocyte band- 3 protein) in avian osteoclasts: effect of calcium-deficient diet on polar expression of the H+-ATPase pump. Calcif Tissue Int 58, 332-6.
(99) Gupta, A., Edwards, J. C., and Hruska, K. A. (1996) Cellular distribution and regulation of NHE-1 isoform of the NA-H exchanger in the avian osteoclast. Bone 18, 87-95.
169 (100) Hall, T. J., and Chambers, T. J. (1990) Na+/H+ antiporter is the primary proton transport system used by osteoclasts during bone resorption. J Cell Physiol 142, 420-4.
(101) Schlesinger, P. H., Blair, H. C., Teitelbaum, S. L., and Edwards, J. C. (1997) Characterization of the osteoclast ruffled border chloride channel and its role in bone resorption. J Biol Chem 272, 18636-43.
(102) Sly, W. S., Hewett-Emmett, D., Whyte, M. P., Yu, Y. S., and Tashian, R. E. (1983) Carbonic anhydrase II deficiency identified as the primary defect in the autosomal recessive syndrome of osteopetrosis with renal tubular acidosis and cerebral calcification. Proc Natl Acad Sci U S A 80, 2752-6.
(103) Frattini, A., Orchard, P. J., Sobacchi, C., Giliani, S., Abinun, M., Mattsson, J. P., Keeling, D. J., Andersson, A. K., Wallbrandt, P., Zecca, L., Notarangelo, L. D., Vezzoni, P., and Villa, A. (2000) Defects in TCIRG1 subunit of the vacuolar proton pump are responsible for a subset of human autosomal recessive osteopetrosis. Nat Genet 25, 343- 6.
(104) Kornak, U., Kasper, D., Bosl, M. R., Kaiser, E., Schweizer, M., Schulz, A., Friedrich, W., Delling, G., and Jentsch, T. J. (2001) Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and man. Cell 104, 205-15.
(105) Piper, K., Boyde, A., and Jones, S. J. (1992) The relationship between the number of nuclei of an osteoclast and its resorptive capability in vitro. Anat Embryol (Berl) 186, 291-9.
(106) Arnett, T. R., and Dempster, D. W. (1986) Effect of pH on bone resorption by rat osteoclasts in vitro. Endocrinology 119, 119-24.
(107) Nordstrom, T., Shrode, L. D., Rotstein, O. D., Romanek, R., Goto, T., Heersche, J. N., Manolson, M. F., Brisseau, G. F., and Grinstein, S. (1997) Chronic extracellular acidosis induces plasmalemmal vacuolar type H+ ATPase activity in osteoclasts. J Biol Chem 272, 6354-60.
(108) Shibutani, T., and Heersche, J. N. (1993) Effect of medium pH on osteoclast activity and osteoclast formation in cultures of dispersed rabbit osteoclasts. J Bone Miner Res 8, 331- 6.
(109) Reddy, S. V., Singer, F. R., Mallette, L., and Roodman, G. D. (1996) Detection of measles virus nucleocapsid transcripts in circulating blood cells from patients with Paget disease. J Bone Miner Res 11, 1602-7.
(110) Kaye, M., Zucker, S. W., Leclerc, Y. G., Prichard, S., Hodsman, A. B., and Barre, P. E. (1985) Osteoclast enlargement in endstage renal disease. Kidney Int 27, 574-81.
170 (111) Everts, V., Delaisse, J. M., Korper, W., and Beertsen, W. (1998) Cysteine proteinases and matrix metalloproteinases play distinct roles in the subosteoclastic resorption zone. J Bone Miner Res 13, 1420-30.
(112) Gelb, B. D., Shi, G. P., Chapman, H. A., and Desnick, R. J. (1996) Pycnodysostosis, a lysosomal disease caused by cathepsin K deficiency. Science 273, 1236-8.
(113) Vu, T. H., Shipley, J. M., Bergers, G., Berger, J. E., Helms, J. A., Hanahan, D., Shapiro, S. D., Senior, R. M., and Werb, Z. (1998) MMP-9/gelatinase B is a key regulator of growth plate angiogenesis and apoptosis of hypertrophic chondrocytes. Cell 93, 411-22.
(114) Delaisse, J. M., Engsig, M. T., Everts, V., del Carmen Ovejero, M., Ferreras, M., Lund, L., Vu, T. H., Werb, Z., Winding, B., Lochter, A., Karsdal, M. A., Troen, T., Kirkegaard, T., Lenhard, T., Heegaard, A. M., Neff, L., Baron, R., and Foged, N. T. (2000) Proteinases in bone resorption: obvious and less obvious roles. Clin Chim Acta 291, 223- 34.
(115) Everts, V., Delaisse, J. M., Korper, W., Jansen, D. C., Tigchelaar-Gutter, W., Saftig, P., and Beertsen, W. (2002) The bone lining cell: its role in cleaning Howship's lacunae and initiating bone formation. J Bone Miner Res 17, 77-90.
(116) Shorey, S., Heersche, J. N., and Manolson, M. F. (2004) The relative contribution of cysteine proteinases and matrix metalloproteinases to the resorption process in osteoclasts derived from long bone and scapula. Bone 35, 909-17.
(117) Everts, V., Korper, W., Hoeben, K. A., Jansen, I. D., Bromme, D., Cleutjens, K. B., Heeneman, S., Peters, C., Reinheckel, T., Saftig, P., and Beertsen, W. (2006) Osteoclastic bone degradation and the role of different cysteine proteinases and matrix metalloproteinases: differences between calvaria and long bone. J Bone Miner Res 21, 1399-408.
(118) Burstone, M. S. (1959) Histochemical demonstration of acid phosphatase activity in osteoclasts. J Histochem Cytochem 7, 39-41.
(119) Ljusberg, J., Ek-Rylander, B., and Andersson, G. (1999) Tartrate-resistant purple acid phosphatase is synthesized as a latent proenzyme and activated by cysteine proteinases. Biochem J 343 Pt 1, 63-9.
(120) Halleen, J. M., Ylipahkala, H., Alatalo, S. L., Janckila, A. J., Heikkinen, J. E., Suominen, H., Cheng, S., and Vaananen, H. K. (2002) Serum tartrate-resistant acid phosphatase 5b, but not 5a, correlates with other markers of bone turnover and bone mineral density. Calcif Tissue Int 71, 20-5.
(121) Seol, J. W., Lee, H. B., Kim, N. S., and Park, S. Y. (2009) Tartrate-resistant acid phosphatase as a diagnostic factor for arthritis. Int J Mol Med 24, 57-62.
171 (122) Halleen, J. M. (2003) Tartrate-resistant acid phosphatase 5B is a specific and sensitive marker of bone resorption. Anticancer Res 23, 1027-9.
(123) Hayman, A. R., Jones, S. J., Boyde, A., Foster, D., Colledge, W. H., Carlton, M. B., Evans, M. J., and Cox, T. M. (1996) Mice lacking tartrate-resistant acid phosphatase (Acp 5) have disrupted endochondral ossification and mild osteopetrosis. Development 122, 3151-62.
(124) Roberts, H. C., Knott, L., Avery, N. C., Cox, T. M., Evans, M. J., and Hayman, A. R. (2007) Altered collagen in tartrate-resistant acid phosphatase (TRAP)-deficient mice: a role for TRAP in bone collagen metabolism. Calcif Tissue Int 80, 400-10.
(125) Hollberg, K., Hultenby, K., Hayman, A., Cox, T., and Andersson, G. (2002) Osteoclasts from mice deficient in tartrate-resistant acid phosphatase have altered ruffled borders and disturbed intracellular vesicular transport. Exp Cell Res 279, 227-38.
(126) Ek-Rylander, B., Flores, M., Wendel, M., Heinegard, D., and Andersson, G. (1994) Dephosphorylation of osteopontin and bone sialoprotein by osteoclastic tartrate-resistant acid phosphatase. Modulation of osteoclast adhesion in vitro. J Biol Chem 269, 14853-6.
(127) Halleen, J. M., Alatalo, S. L., Janckila, A. J., Woitge, H. W., Seibel, M. J., and Vaananen, H. K. (2001) Serum tartrate-resistant acid phosphatase 5b is a specific and sensitive marker of bone resorption. Clin Chem 47, 597-600.
(128) Janckila, A. J., Takahashi, K., Sun, S. Z., and Yam, L. T. (2001) Tartrate-resistant acid phosphatase isoform 5b as serum marker for osteoclastic activity. Clin Chem 47, 74-80.
(129) Alatalo, S. L., Ivaska, K. K., Waguespack, S. G., Econs, M. J., Vaananen, H. K., and Halleen, J. M. (2004) Osteoclast-derived serum tartrate-resistant acid phosphatase 5b in Albers-Schonberg disease (type II autosomal dominant osteopetrosis). Clin Chem 50, 883-90.
(130) Rissanen, J. P., Suominen, M. I., Peng, Z., and Halleen, J. M. (2008) Secreted tartrate- resistant acid phosphatase 5b is a Marker of osteoclast number in human osteoclast cultures and the rat ovariectomy model. Calcif Tissue Int 82, 108-15.
(131) Rissanen, J. P., Ylipahkala, H., Fagerlund, K. M., Long, C., Vaananen, H. K., and Halleen, J. M. (2009) Improved methods for testing antiresorptive compounds in human osteoclast cultures. J Bone Miner Metab 27, 105-9.
(132) Wolosewick, J. J. (1984) Distribution of actin in migrating leukocytes in vivo. Cell Tissue Res 236, 517-25.
(133) Marchisio, P. C., Cirillo, D., Naldini, L., Primavera, M. V., Teti, A., and Zambonin- Zallone, A. (1984) Cell-substratum interaction of cultured avian osteoclasts is mediated by specific adhesion structures. J Cell Biol 99, 1696-705.
172 (134) Gaidano, G., Bergui, L., Schena, M., Gaboli, M., Cremona, O., Marchisio, P. C., and Caligaris-Cappio, F. (1990) Integrin distribution and cytoskeleton organization in normal and malignant monocytes. Leukemia 4, 682-7.
(135) Tarone, G., Cirillo, D., Giancotti, F. G., Comoglio, P. M., and Marchisio, P. C. (1985) Rous sarcoma virus-transformed fibroblasts adhere primarily at discrete protrusions of the ventral membrane called podosomes. Exp Cell Res 159, 141-57.
(136) Badowski, C., Pawlak, G., Grichine, A., Chabadel, A., Oddou, C., Jurdic, P., Pfaff, M., Albiges-Rizo, C., and Block, M. R. (2008) Paxillin phosphorylation controls invadopodia/podosomes spatiotemporal organization. Mol Biol Cell 19, 633-45.
(137) Desai, B., Ma, T., and Chellaiah, M. A. (2008) Invadopodia and matrix degradation, a new property of prostate cancer cells during migration and invasion. J Biol Chem 283, 13856-66.
(138) Machesky, L., Jurdic, P., and Hinz, B. (2008) Grab, stick, pull and digest: the functional diversity of actin-associated matrix-adhesion structures. Workshop on Invadopodia, Podosomes and Focal Adhesions in Tissue Invasion. EMBO Rep 9, 139-43.
(139) Nitsch, L., Gionti, E., Cancedda, R., and Marchisio, P. C. (1989) The podosomes of Rous sarcoma virus transformed chondrocytes show a peculiar ultrastructural organization. Cell Biol Int Rep 13, 919-26.
(140) Zambonin-Zallone, A., Teti, A., Grano, M., Rubinacci, A., Abbadini, M., Gaboli, M., and Marchisio, P. C. (1989) Immunocytochemical distribution of extracellular matrix receptors in human osteoclasts: a beta 3 integrin is colocalized with vinculin and talin in the podosomes of osteoclastoma giant cells. Exp Cell Res 182, 645-52.
(141) Linder, S., Nelson, D., Weiss, M., and Aepfelbacher, M. (1999) Wiskott-Aldrich syndrome protein regulates podosomes in primary human macrophages. Proc Natl Acad Sci U S A 96, 9648-53.
(142) Linder, S., Higgs, H., Hufner, K., Schwarz, K., Pannicke, U., and Aepfelbacher, M. (2000) The polarization defect of Wiskott-Aldrich syndrome macrophages is linked to dislocalization of the Arp2/3 complex. J Immunol 165, 221-5.
(143) Gallego, M. D., de la Fuente, M. A., Anton, I. M., Snapper, S., Fuhlbrigge, R., and Geha, R. S. (2006) WIP and WASP play complementary roles in T cell homing and chemotaxis to SDF-1alpha. Int Immunol 18, 221-32.
(144) Spinardi, L., and Marchisio, P. C. (2006) Podosomes as smart regulators of cellular adhesion. Eur J Cell Biol 85, 191-4.
(145) Akisaka, T., Yoshida, H., Inoue, S., and Shimizu, K. (2001) Organization of cytoskeletal F-actin, G-actin, and gelsolin in the adhesion structures in cultured osteoclast. J Bone Miner Res 16, 1248-55. 173 (146) Bruzzaniti, A., Neff, L., Sanjay, A., Horne, W. C., De Camilli, P., and Baron, R. (2005) Dynamin forms a Src kinase-sensitive complex with Cbl and regulates podosomes and osteoclast activity. Mol Biol Cell 16, 3301-13.
(147) Destaing, O., Saltel, F., Geminard, J. C., Jurdic, P., and Bard, F. (2003) Podosomes display actin turnover and dynamic self-organization in osteoclasts expressing actin- green fluorescent protein. Mol Biol Cell 14, 407-16.
(148) Saltel, F., Destaing, O., Bard, F., Eichert, D., and Jurdic, P. (2004) Apatite-mediated actin dynamics in resorbing osteoclasts. Mol Biol Cell 15, 5231-41.
(149) Geblinger, D., Geiger, B., and Addadi, L. (2009) Surface-induced regulation of podosome organization and dynamics in cultured osteoclasts. Chembiochem 10, 158-65.
(150) Luxenburg, C., Geblinger, D., Klein, E., Anderson, K., Hanein, D., Geiger, B., and Addadi, L. (2007) The architecture of the adhesive apparatus of cultured osteoclasts: from podosome formation to sealing zone assembly. PLoS One 2, e179.
(151) Linder, S., Hufner, K., Wintergerst, U., and Aepfelbacher, M. (2000) Microtubule- dependent formation of podosomal adhesion structures in primary human macrophages. J Cell Sci 113 Pt 23, 4165-76.
(152) Destaing, O., Saltel, F., Gilquin, B., Chabadel, A., Khochbin, S., Ory, S., and Jurdic, P. (2005) A novel Rho-mDia2-HDAC6 pathway controls podosome patterning through microtubule acetylation in osteoclasts. J Cell Sci 118, 2901-11.
(153) Jurdic, P., Saltel, F., Chabadel, A., and Destaing, O. (2006) Podosome and sealing zone: specificity of the osteoclast model. Eur J Cell Biol 85, 195-202.
(154) Chabadel, A., Banon-Rodriguez, I., Cluet, D., Rudkin, B. B., Wehrle-Haller, B., Genot, E., Jurdic, P., Anton, I. M., and Saltel, F. (2007) CD44 and beta3 integrin organize two functionally distinct actin-based domains in osteoclasts. Mol Biol Cell 18, 4899-910.
(155) Saltel, F., Chabadel, A., Bonnelye, E., and Jurdic, P. (2008) Actin cytoskeletal organisation in osteoclasts: a model to decipher transmigration and matrix degradation. Eur J Cell Biol 87, 459-68.
(156) Chou, H. C., Anton, I. M., Holt, M. R., Curcio, C., Lanzardo, S., Worth, A., Burns, S., Thrasher, A. J., Jones, G. E., and Calle, Y. (2006) WIP regulates the stability and localization of WASP to podosomes in migrating dendritic cells. Curr Biol 16, 2337-44.
(157) Destaing, O., Sanjay, A., Itzstein, C., Horne, W. C., Toomre, D., De Camilli, P., and Baron, R. (2008) The tyrosine kinase activity of c-Src regulates actin dynamics and organization of podosomes in osteoclasts. Mol Biol Cell 19, 394-404.
(158) Luxenburg, C., Addadi, L., and Geiger, B. (2006) The molecular dynamics of osteoclast adhesions. Eur J Cell Biol 85, 203-11. 174 (159) Luxenburg, C., Parsons, J. T., Addadi, L., and Geiger, B. (2006) Involvement of the Src- cortactin pathway in podosome formation and turnover during polarization of cultured osteoclasts. J Cell Sci 119, 4878-88.
(160) Sanjay, A., Houghton, A., Neff, L., DiDomenico, E., Bardelay, C., Antoine, E., Levy, J., Gailit, J., Bowtell, D., Horne, W. C., and Baron, R. (2001) Cbl associates with Pyk2 and Src to regulate Src kinase activity, alpha(v)beta(3) integrin-mediated signaling, cell adhesion, and osteoclast motility. J Cell Biol 152, 181-95.
(161) Bourguignon, L. Y., Peyrollier, K., Gilad, E., and Brightman, A. (2007) Hyaluronan- CD44 interaction with neural Wiskott-Aldrich syndrome protein (N-WASP) promotes actin polymerization and ErbB2 activation leading to beta-catenin nuclear translocation, transcriptional up-regulation, and cell migration in ovarian tumor cells. J Biol Chem 282, 1265-80.
(162) Kanehisa, J., and Heersche, J. N. (1988) Osteoclastic bone resorption: in vitro analysis of the rate of resorption and migration of individual osteoclasts. Bone 9, 73-9.
(163) Lakkakorpi, P. T., and Vaananen, H. K. (1991) Kinetics of the osteoclast cytoskeleton during the resorption cycle in vitro. J Bone Miner Res 6, 817-26.
(164) Lakkakorpi, P. T., Horton, M. A., Helfrich, M. H., Karhukorpi, E. K., and Vaananen, H. K. (1991) Vitronectin receptor has a role in bone resorption but does not mediate tight sealing zone attachment of osteoclasts to the bone surface. J Cell Biol 115, 1179-86.
(165) Chellaiah, M. A., and Hruska, K. A. (2003) The integrin alpha(v)beta(3) and CD44 regulate the actions of osteopontin on osteoclast motility. Calcif Tissue Int 72, 197-205.
(166) Pilkington, M. F., Sims, S. M., and Dixon, S. J. (2001) Transforming growth factor-beta induces osteoclast ruffling and chemotaxis: potential role in osteoclast recruitment. J Bone Miner Res 16, 1237-47.
(167) Arkett, S. A., Dixon, S. J., and Sims, S. M. (1992) Substrate influences rat osteoclast morphology and expression of potassium conductances. J Physiol 458, 633-53.
(168) Zhang, D., Udagawa, N., Nakamura, I., Murakami, H., Saito, S., Yamasaki, K., Shibasaki, Y., Morii, N., Narumiya, S., Takahashi, N., and et al. (1995) The small GTP- binding protein, rho p21, is involved in bone resorption by regulating cytoskeletal organization in osteoclasts. J Cell Sci 108 ( Pt 6), 2285-92.
(169) Ory, S., Munari-Silem, Y., Fort, P., and Jurdic, P. (2000) Rho and Rac exert antagonistic functions on spreading of macrophage-derived multinucleated cells and are not required for actin fiber formation. J Cell Sci 113 ( Pt 7), 1177-88.
(170) Fuller, K., Murphy, C., Kirstein, B., Fox, S. W., and Chambers, T. J. (2002) TNFalpha potently activates osteoclasts, through a direct action independent of and strongly synergistic with RANKL. Endocrinology 143, 1108-18. 175 (171) Tamkun, J. W., DeSimone, D. W., Fonda, D., Patel, R. S., Buck, C., Horwitz, A. F., and Hynes, R. O. (1986) Structure of integrin, a glycoprotein involved in the transmembrane linkage between fibronectin and actin. Cell 46, 271-82.
(172) Hynes, R. O. (1992) Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69, 11-25.
(173) Springer, T. A. (1994) Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76, 301-14.
(174) Anderson, D. C., and Springer, T. A. (1987) Leukocyte adhesion deficiency: an inherited defect in the Mac-1, LFA-1, and p150,95 glycoproteins. Annu Rev Med 38, 175-94.
(175) Juliano, R. L. (1993) The role of beta 1 integrins in tumors. Semin Cancer Biol 4, 277- 83.
(176) Juliano, R. L., and Varner, J. A. (1993) Adhesion molecules in cancer: the role of integrins. Curr Opin Cell Biol 5, 812-8.
(177) Shattil, S. J., Ginsberg, M. H., and Brugge, J. S. (1994) Adhesive signaling in platelets. Curr Opin Cell Biol 6, 695-704.
(178) Giancotti, F. G. (1997) Integrin signaling: specificity and control of cell survival and cell cycle progression. Curr Opin Cell Biol 9, 691-700.
(179) Shyy, J. Y., and Chien, S. (1997) Role of integrins in cellular responses to mechanical stress and adhesion. Curr Opin Cell Biol 9, 707-13.
(180) Bader, B. L., Rayburn, H., Crowley, D., and Hynes, R. O. (1998) Extensive vasculogenesis, angiogenesis, and organogenesis precede lethality in mice lacking all alpha v integrins. Cell 95, 507-19.
(181) McCarty, J. H., Monahan-Earley, R. A., Brown, L. F., Keller, M., Gerhardt, H., Rubin, K., Shani, M., Dvorak, H. F., Wolburg, H., Bader, B. L., Dvorak, A. M., and Hynes, R. O. (2002) Defective associations between blood vessels and brain parenchyma lead to cerebral hemorrhage in mice lacking alphav integrins. Mol Cell Biol 22, 7667-77.
(182) Hodivala-Dilke, K. M., McHugh, K. P., Tsakiris, D. A., Rayburn, H., Crowley, D., Ullman-Cullere, M., Ross, F. P., Coller, B. S., Teitelbaum, S., and Hynes, R. O. (1999) Beta3-integrin-deficient mice are a model for Glanzmann thrombasthenia showing placental defects and reduced survival. J Clin Invest 103, 229-38.
(183) Hynes, R. O. (2002) Integrins: bidirectional, allosteric signaling machines. Cell 110, 673-87.
(184) Hynes, R. O., and Wagner, D. D. (1996) Genetic manipulation of vascular adhesion molecules in mice. J Clin Invest 98, 2193-5. 176 (185) de Pereda, J. M., Wiche, G., and Liddington, R. C. (1999) Crystal structure of a tandem pair of fibronectin type III domains from the cytoplasmic tail of integrin alpha6beta4. Embo J 18, 4087-95.
(186) Humphries, M. J. (2000) Integrin structure. Biochem Soc Trans 28, 311-39.
(187) Shimaoka, M., Takagi, J., and Springer, T. A. (2002) Conformational regulation of integrin structure and function. Annu Rev Biophys Biomol Struct 31, 485-516.
(188) Springer, T. A. (1997) Folding of the N-terminal, ligand-binding region of integrin alpha-subunits into a beta-propeller domain. Proc Natl Acad Sci U S A 94, 65-72.
(189) Springer, T. A., and Wang, J. H. (2004) The three-dimensional structure of integrins and their ligands, and conformational regulation of cell adhesion. Adv Protein Chem 68, 29- 63.
(190) Beglova, N., Blacklow, S. C., Takagi, J., and Springer, T. A. (2002) Cysteine-rich module structure reveals a fulcrum for integrin rearrangement upon activation. Nat Struct Biol 9, 282-7.
(191) Takagi, J., Beglova, N., Yalamanchili, P., Blacklow, S. C., and Springer, T. A. (2001) Definition of EGF-like, closely interacting modules that bear activation epitopes in integrin beta subunits. Proc Natl Acad Sci U S A 98, 11175-80.
(192) Xiong, J. P., Stehle, T., Diefenbach, B., Zhang, R., Dunker, R., Scott, D. L., Joachimiak, A., Goodman, S. L., and Arnaout, M. A. (2001) Crystal structure of the extracellular segment of integrin alpha Vbeta3. Science 294, 339-45.
(193) Du, X., Gu, M., Weisel, J. W., Nagaswami, C., Bennett, J. S., Bowditch, R., and Ginsberg, M. H. (1993) Long range propagation of conformational changes in integrin alpha IIb beta 3. J Biol Chem 268, 23087-92.
(194) Weisel, J. W., Nagaswami, C., Vilaire, G., and Bennett, J. S. (1992) Examination of the platelet membrane glycoprotein IIb-IIIa complex and its interaction with fibrinogen and other ligands by electron microscopy. J Biol Chem 267, 16637-43.
(195) Huang, C., Zang, Q., Takagi, J., and Springer, T. A. (2000) Structural and functional studies with antibodies to the integrin beta 2 subunit. A model for the I-like domain. J Biol Chem 275, 21514-24.
(196) Takagi, J., Petre, B. M., Walz, T., and Springer, T. A. (2002) Global conformational rearrangements in integrin extracellular domains in outside-in and inside-out signaling. Cell 110, 599-11.
(197) Xiong, J. P., Stehle, T., Zhang, R., Joachimiak, A., Frech, M., Goodman, S. L., and Arnaout, M. A. (2002) Crystal structure of the extracellular segment of integrin alpha Vbeta3 in complex with an Arg-Gly-Asp ligand. Science 296, 151-5. 177 (198) Luo, B. H., Springer, T. A., and Takagi, J. (2003) Stabilizing the open conformation of the integrin headpiece with a glycan wedge increases affinity for ligand. Proc Natl Acad Sci U S A 100, 2403-8.
(199) Lu, C., Shimaoka, M., Ferzly, M., Oxvig, C., Takagi, J., and Springer, T. A. (2001) An isolated, surface-expressed I domain of the integrin alphaLbeta2 is sufficient for strong adhesive function when locked in the open conformation with a disulfide bond. Proc Natl Acad Sci U S A 98, 2387-92.
(200) Kim, M., Carman, C. V., and Springer, T. A. (2003) Bidirectional transmembrane signaling by cytoplasmic domain separation in integrins. Science 301, 1720-5.
(201) Lefort, C. T., Hyun, Y. M., Schultz, J. B., Law, F. Y., Waugh, R. E., Knauf, P. A., and Kim, M. (2009) Outside-in signal transmission by conformational changes in integrin Mac-1. J Immunol 183, 6460-8.
(202) Xie, C., Shimaoka, M., Xiao, T., Schwab, P., Klickstein, L. B., and Springer, T. A. (2004) The integrin alpha-subunit leg extends at a Ca2+-dependent epitope in the thigh/genu interface upon activation. Proc Natl Acad Sci U S A 101, 15422-7.
(203) D'Souza, S. E., Haas, T. A., Piotrowicz, R. S., Byers-Ward, V., McGrath, D. E., Soule, H. R., Cierniewski, C., Plow, E. F., and Smith, J. W. (1994) Ligand and cation binding are dual functions of a discrete segment of the integrin beta 3 subunit: cation displacement is involved in ligand binding. Cell 79, 659-67.
(204) Bazzoni, G., Shih, D. T., Buck, C. A., and Hemler, M. E. (1995) Monoclonal antibody 9EG7 defines a novel beta 1 integrin epitope induced by soluble ligand and manganese, but inhibited by calcium. J Biol Chem 270, 25570-7.
(205) Chen, J., Salas, A., and Springer, T. A. (2003) Bistable regulation of integrin adhesiveness by a bipolar metal ion cluster. Nat Struct Biol 10, 995-1001.
(206) Mould, A. P., Barton, S. J., Askari, J. A., Craig, S. E., and Humphries, M. J. (2003) Role of ADMIDAS cation-binding site in ligand recognition by integrin alpha 5 beta 1. J Biol Chem 278, 51622-9.
(207) Legler, D. F., Wiedle, G., Ross, F. P., and Imhof, B. A. (2001) Superactivation of integrin alphavbeta3 by low antagonist concentrations. J Cell Sci 114, 1545-53.
(208) Calderwood, D. A., Yan, B., de Pereda, J. M., Alvarez, B. G., Fujioka, Y., Liddington, R. C., and Ginsberg, M. H. (2002) The phosphotyrosine binding-like domain of talin activates integrins. J Biol Chem 277, 21749-58.
(209) Arias-Salgado, E. G., Lizano, S., Shattil, S. J., and Ginsberg, M. H. (2005) Specification of the direction of adhesive signaling by the integrin beta cytoplasmic domain. J Biol Chem 280, 29699-707.
178 (210) Li, R., Mitra, N., Gratkowski, H., Vilaire, G., Litvinov, R., Nagasami, C., Weisel, J. W., Lear, J. D., DeGrado, W. F., and Bennett, J. S. (2003) Activation of integrin alphaIIbbeta3 by modulation of transmembrane helix associations. Science 300, 795-8.
(211) Hynes, R. O. (2003) Structural biology. Changing partners. Science 300, 755-6.
(212) de Bruyn, K. M., Zwartkruis, F. J., de Rooij, J., Akkerman, J. W., and Bos, J. L. (2003) The small GTPase Rap1 is activated by turbulence and is involved in integrin [alpha]IIb[beta]3-mediated cell adhesion in human megakaryocytes. J Biol Chem 278, 22412-7.
(213) Clover, J., Dodds, R. A., and Gowen, M. (1992) Integrin subunit expression by human osteoblasts and osteoclasts in situ and in culture. J Cell Sci 103 ( Pt 1), 267-71.
(214) Hughes, D. E., Salter, D. M., Dedhar, S., and Simpson, R. (1993) Integrin expression in human bone. J Bone Miner Res 8, 527-33.
(215) Nesbitt, S., Nesbit, A., Helfrich, M., and Horton, M. (1993) Biochemical characterization of human osteoclast integrins. Osteoclasts express alpha v beta 3, alpha 2 beta 1, and alpha v beta 1 integrins. J Biol Chem 268, 16737-45.
(216) Horton, M. A. (1997) The alpha v beta 3 integrin "vitronectin receptor". Int J Biochem Cell Biol 29, 721-5.
(217) Masarachia, P., Yamamoto, M., Leu, C. T., Rodan, G., and Duong, L. (1998) Histomorphometric evidence for echistatin inhibition of bone resorption in mice with secondary hyperparathyroidism. Endocrinology 139, 1401-10.
(218) Inoue, M., Namba, N., Chappel, J., Teitelbaum, S. L., and Ross, F. P. (1998) Granulocyte macrophage-colony stimulating factor reciprocally regulates alphav- associated integrins on murine osteoclast precursors. Mol Endocrinol 12, 1955-62.
(219) Athanasou, N. A., Alvarez, J. I., Ross, F. P., Quinn, J. M., and Teitelbaum, S. L. (1992) Species differences in the immunophenotype of osteoclasts and mononuclear phagocytes. Calcif Tissue Int 50, 427-32.
(220) Sago, K., Teitelbaum, S. L., Venstrom, K., Reichardt, L. F., and Ross, F. P. (1999) The integrin alphavbeta5 is expressed on avian osteoclast precursors and regulated by retinoic acid. J Bone Miner Res 14, 32-8.
(221) Rao, H., Lu, G., Kajiya, H., Garcia-Palacios, V., Kurihara, N., Anderson, J., Patrene, K., Sheppard, D., Blair, H. C., Windle, J. J., Choi, S. J., and Roodman, G. D. (2006) Alpha9beta1: a novel osteoclast integrin that regulates osteoclast formation and function. J Bone Miner Res 21, 1657-65.
179 (222) Flores, M. E., Norgard, M., Heinegard, D., Reinholt, F. P., and Andersson, G. (1992) RGD-directed attachment of isolated rat osteoclasts to osteopontin, bone sialoprotein, and fibronectin. Exp Cell Res 201, 526-30.
(223) Helfrich, M. H., Nesbitt, S. A., Lakkakorpi, P. T., Barnes, M. J., Bodary, S. C., Shankar, G., Mason, W. T., Mendrick, D. L., Vaananen, H. K., and Horton, M. A. (1996) Beta 1 integrins and osteoclast function: involvement in collagen recognition and bone resorption. Bone 19, 317-28.
(224) Horton, M. (1990) Vitronectin receptor: tissue specific expression or adaptation to culture? Int J Exp Pathol 71, 741-59.
(225) Marshall, J. F., and Hart, I. R. (1996) The role of alpha v-integrins in tumour progression and metastasis. Semin Cancer Biol 7, 129-38.
(226) Bakewell, S. J., Nestor, P., Prasad, S., Tomasson, M. H., Dowland, N., Mehrotra, M., Scarborough, R., Kanter, J., Abe, K., Phillips, D., and Weilbaecher, K. N. (2003) Platelet and osteoclast beta3 integrins are critical for bone metastasis. Proc Natl Acad Sci U S A 100, 14205-10.
(227) Boissy, P., Machuca, I., Pfaff, M., Ficheux, D., and Jurdic, P. (1998) Aggregation of mononucleated precursors triggers cell surface expression of alphavbeta3 integrin, essential to formation of osteoclast-like multinucleated cells. J Cell Sci 111 ( Pt 17), 2563-74.
(228) Miyamoto, T., Arai, F., Ohneda, O., Takagi, K., Anderson, D. M., and Suda, T. (2000) An adherent condition is required for formation of multinuclear osteoclasts in the presence of macrophage colony-stimulating factor and receptor activator of nuclear factor kappa B ligand. Blood 96, 4335-43.
(229) Feng, X., Teitelbaum, S. L., Quiroz, M. E., Towler, D. A., and Ross, F. P. (1999) Cloning of the murine beta5 integrin subunit promoter. Identification of a novel sequence mediating granulocyte-macrophage colony-stimulating factor-dependent repression of beta5 integrin gene transcription. J Biol Chem 274, 1366-74.
(230) Lane, N. E., Yao, W., Nakamura, M. C., Humphrey, M. B., Kimmel, D., Huang, X., Sheppard, D., Ross, F. P., and Teitelbaum, S. L. (2005) Mice lacking the integrin beta5 subunit have accelerated osteoclast maturation and increased activity in the estrogen- deficient state. J Bone Miner Res 20, 58-66.
(231) Zhao, H., Kitaura, H., Sands, M. S., Ross, F. P., Teitelbaum, S. L., and Novack, D. V. (2005) Critical role of beta3 integrin in experimental postmenopausal osteoporosis. J Bone Miner Res 20, 2116-23.
(232) Faccio, R., Grano, M., Colucci, S., Villa, A., Giannelli, G., Quaranta, V., and Zallone, A. (2002) Localization and possible role of two different alpha v beta 3 integrin conformations in resting and resorbing osteoclasts. J Cell Sci 115, 2919-29. 180 (233) Butler, B., Williams, M. P., and Blystone, S. D. (2003) Ligand-dependent activation of integrin alpha vbeta 3. J Biol Chem 278, 5264-70.
(234) Faccio, R., Grano, M., Colucci, S., Zallone, A. Z., Quaranta, V., and Pelletier, A. J. (1998) Activation of alphav beta3 integrin on human osteoclast-like cells stimulates adhesion and migration in response to osteopontin. Biochem Biophys Res Commun 249, 522-5.
(235) Horton, M. A., Lewis, D., McNulty, K., Pringle, J. A., and Chambers, T. J. (1985) Monoclonal antibodies to osteoclastomas (giant cell bone tumors): definition of osteoclast-specific cellular antigens. Cancer Res 45, 5663-9.
(236) Davies, J., Warwick, J., Totty, N., Philp, R., Helfrich, M., and Horton, M. (1989) The osteoclast functional antigen, implicated in the regulation of bone resorption, is biochemically related to the vitronectin receptor. J Cell Biol 109, 1817-26.
(237) Sato, M., Sardana, M. K., Grasser, W. A., Garsky, V. M., Murray, J. M., and Gould, R. J. (1990) Echistatin is a potent inhibitor of bone resorption in culture. J Cell Biol 111, 1713-23.
(238) Horton, M. A., Dorey, E. L., Nesbitt, S. A., Samanen, J., Ali, F. E., Stadel, J. M., Nichols, A., Greig, R., and Helfrich, M. H. (1993) Modulation of vitronectin receptor- mediated osteoclast adhesion by Arg-Gly-Asp peptide analogs: a structure-function analysis. J Bone Miner Res 8, 239-47.
(239) Engleman, V. W., Nickols, G. A., Ross, F. P., Horton, M. A., Griggs, D. W., Settle, S. L., Ruminski, P. G., and Teitelbaum, S. L. (1997) A peptidomimetic antagonist of the alpha(v)beta3 integrin inhibits bone resorption in vitro and prevents osteoporosis in vivo. J Clin Invest 99, 2284-92.
(240) Fisher, J. E., Caulfield, M. P., Sato, M., Quartuccio, H. A., Gould, R. J., Garsky, V. M., Rodan, G. A., and Rosenblatt, M. (1993) Inhibition of osteoclastic bone resorption in vivo by echistatin, an "arginyl-glycyl-aspartyl" (RGD)-containing protein. Endocrinology 132, 1411-3.
(241) Yamamoto, M., Fisher, J. E., Gentile, M., Seedor, J. G., Leu, C. T., Rodan, S. B., and Rodan, G. A. (1998) The integrin ligand echistatin prevents bone loss in ovariectomized mice and rats. Endocrinology 139, 1411-9.
(242) Villanova, I., Townsend, P. A., Uhlmann, E., Knolle, J., Peyman, A., Amling, M., Baron, R., Horton, M. A., and Teti, A. (1999) Oligodeoxynucleotide targeted to the alphav gene inhibits alphav integrin synthesis, impairs osteoclast function, and activates intracellular signals to apoptosis. J Bone Miner Res 14, 1867-79.
(243) Nakamura, I., Pilkington, M. F., Lakkakorpi, P. T., Lipfert, L., Sims, S. M., Dixon, S. J., Rodan, G. A., and Duong, L. T. (1999) Role of alpha(v)beta(3) integrin in osteoclast migration and formation of the sealing zone. J Cell Sci 112 ( Pt 22), 3985-93. 181 (244) Nakamura, I., Tanaka, H., Rodan, G. A., and Duong, L. T. (1998) Echistatin inhibits the migration of murine prefusion osteoclasts and the formation of multinucleated osteoclast-like cells. Endocrinology 139, 5182-93.
(245) Filardo, E. J., Deming, S. L., and Cheresh, D. A. (1996) Regulation of cell migration by the integrin beta subunit ectodomain. J Cell Sci 109 ( Pt 6), 1615-22.
(246) Faccio, R., Novack, D. V., Zallone, A., Ross, F. P., and Teitelbaum, S. L. (2003) Dynamic changes in the osteoclast cytoskeleton in response to growth factors and cell attachment are controlled by beta3 integrin. J Cell Biol 162, 499-509.
(247) Zhao, H., Ross, F. P., and Teitelbaum, S. L. (2005) Unoccupied alpha(v)beta3 integrin regulates osteoclast apoptosis by transmitting a positive death signal. Mol Endocrinol 19, 771-80.
(248) Feng, X., Novack, D. V., Faccio, R., Ory, D. S., Aya, K., Boyer, M. I., McHugh, K. P., Ross, F. P., and Teitelbaum, S. L. (2001) A Glanzmann's mutation in beta 3 integrin specifically impairs osteoclast function. J Clin Invest 107, 1137-44.
(249) Horton, M. A., Massey, H. M., Rosenberg, N., Nicholls, B., Seligsohn, U., and Flanagan, A. M. (2003) Upregulation of osteoclast alpha2beta1 integrin compensates for lack of alphavbeta3 vitronectin receptor in Iraqi-Jewish-type Glanzmann thrombasthenia. Br J Haematol 122, 950-7.
(250) Horne, W. C., Neff, L., Chatterjee, D., Lomri, A., Levy, J. B., and Baron, R. (1992) Osteoclasts express high levels of pp60c-src in association with intracellular membranes. J Cell Biol 119, 1003-13.
(251) Soriano, P., Montgomery, C., Geske, R., and Bradley, A. (1991) Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64, 693-702.
(252) Boyce, B. F., Yoneda, T., Lowe, C., Soriano, P., and Mundy, G. R. (1992) Requirement of pp60c-src expression for osteoclasts to form ruffled borders and resorb bone in mice. J Clin Invest 90, 1622-7.
(253) Nakamura, I., Lipfert, L., Rodan, G. A., and Le, T. D. (2001) Convergence of alpha(v)beta(3) integrin- and macrophage colony stimulating factor-mediated signals on phospholipase Cgamma in prefusion osteoclasts. J Cell Biol 152, 361-73.
(254) Schwartzberg, P. L., Xing, L., Hoffmann, O., Lowell, C. A., Garrett, L., Boyce, B. F., and Varmus, H. E. (1997) Rescue of osteoclast function by transgenic expression of kinase-deficient Src in src-/- mutant mice. Genes Dev 11, 2835-44.
(255) Duong, L. T., Lakkakorpi, P. T., Nakamura, I., Machwate, M., Nagy, R. M., and Rodan, G. A. (1998) PYK2 in osteoclasts is an adhesion kinase, localized in the sealing zone, activated by ligation of alpha(v)beta3 integrin, and phosphorylated by src kinase. J Clin Invest 102, 881-92. 182 (256) Lev, S., Moreno, H., Martinez, R., Canoll, P., Peles, E., Musacchio, J. M., Plowman, G. D., Rudy, B., and Schlessinger, J. (1995) Protein tyrosine kinase PYK2 involved in Ca(2+)-induced regulation of ion channel and MAP kinase functions. Nature 376, 737- 45.
(257) Lakkakorpi, P. T., Bett, A. J., Lipfert, L., Rodan, G. A., and Duong le, T. (2003) PYK2 autophosphorylation, but not kinase activity, is necessary for adhesion-induced association with c-Src, osteoclast spreading, and bone resorption. J Biol Chem 278, 11502-12.
(258) Tanaka, S., Amling, M., Neff, L., Peyman, A., Uhlmann, E., Levy, J. B., and Baron, R. (1996) c-Cbl is downstream of c-Src in a signalling pathway necessary for bone resorption. Nature 383, 528-31.
(259) Odai, H., Sasaki, K., Hanazono, Y., Ueno, H., Tanaka, T., Miyagawa, K., Mitani, K., Yazaki, Y., and Hirai, H. (1995) c-Cbl is inducibly tyrosine-phosphorylated by epidermal growth factor stimulation in fibroblasts, and constitutively tyrosine- phosphorylated and associated with v-Src in v-src-transformed fibroblasts. Jpn J Cancer Res 86, 1119-26.
(260) Yokouchi, M., Kondo, T., Sanjay, A., Houghton, A., Yoshimura, A., Komiya, S., Zhang, H., and Baron, R. (2001) Src-catalyzed phosphorylation of c-Cbl leads to the interdependent ubiquitination of both proteins. J Biol Chem 276, 35185-93.
(261) Gil-Henn, H., Destaing, O., Sims, N. A., Aoki, K., Alles, N., Neff, L., Sanjay, A., Bruzzaniti, A., De Camilli, P., Baron, R., and Schlessinger, J. (2007) Defective microtubule-dependent podosome organization in osteoclasts leads to increased bone density in Pyk2(-/-) mice. J Cell Biol 178, 1053-64.
(262) Nakajima, A., Sanjay, A., Chiusaroli, R., Adapala, N. S., Neff, L., Itzsteink, C., Horne, W. C., and Baron, R. (2009) Loss of Cbl-b increases osteoclast bone-resorbing activity and induces osteopenia. J Bone Miner Res 24, 1162-72.
(263) Faccio, R., Teitelbaum, S. L., Fujikawa, K., Chappel, J., Zallone, A., Tybulewicz, V. L., Ross, F. P., and Swat, W. (2005) Vav3 regulates osteoclast function and bone mass. Nat Med 11, 284-90.
(264) Pilkington, M. F., Sims, S. M., and Dixon, S. J. (1998) Wortmannin inhibits spreading and chemotaxis of rat osteoclasts in vitro. J Bone Miner Res 13, 688-94.
(265) Palacio, S., and Felix, R. (2001) The role of phosphoinositide 3-kinase in spreading osteoclasts induced by colony-stimulating factor-1. Eur J Endocrinol 144, 431-40.
(266) Wells, C. M., Walmsley, M., Ooi, S., Tybulewicz, V., and Ridley, A. J. (2004) Rac1- deficient macrophages exhibit defects in cell spreading and membrane ruffling but not migration. J Cell Sci 117, 1259-68.
183 (267) Munugalavadla, V., Borneo, J., Ingram, D. A., and Kapur, R. (2005) p85alpha subunit of class IA PI-3 kinase is crucial for macrophage growth and migration. Blood 106, 103-9.
(268) Faccio, R., Takeshita, S., Zallone, A., Ross, F. P., and Teitelbaum, S. L. (2003) c-Fms and the alphavbeta3 integrin collaborate during osteoclast differentiation. J Clin Invest 111, 749-58.
(269) Elsegood, C. L., Zhuo, Y., Wesolowski, G. A., Hamilton, J. A., Rodan, G. A., and Duong le, T. (2006) M-CSF induces the stable interaction of cFms with alphaVbeta3 integrin in osteoclasts. Int J Biochem Cell Biol 38, 1518-29.
(270) Jimi, E., Nakamura, I., Duong, L. T., Ikebe, T., Takahashi, N., Rodan, G. A., and Suda, T. (1999) Interleukin 1 induces multinucleation and bone-resorbing activity of osteoclasts in the absence of osteoblasts/stromal cells. Exp Cell Res 247, 84-93.
(271) Nakamura, I., Kadono, Y., Takayanagi, H., Jimi, E., Miyazaki, T., Oda, H., Nakamura, K., Tanaka, S., Rodan, G. A., and Duong le, T. (2002) IL-1 regulates cytoskeletal organization in osteoclasts via TNF receptor-associated factor 6/c-Src complex. J Immunol 168, 5103-9.
(272) Mark, M. P., Prince, C. W., Oosawa, T., Gay, S., Bronckers, A. L., and Butler, W. T. (1987) Immunohistochemical demonstration of a 44-KD phosphoprotein in developing rat bones. J Histochem Cytochem 35, 707-15.
(273) Sodek, J., Goldberg, H. A., Domenicucci, C., Zhang, Q., Kwon, B., Maeno, M., and Kuwata, F. (1989) Characterization of multiple forms of small collagenous apatite- binding proteins in bone. Connect Tissue Res 20, 233-40.
(274) Rittling, S. R., and Denhardt, D. T. (1999) Osteopontin function in pathology: lessons from osteopontin-deficient mice. Exp Nephrol 7, 103-13.
(275) O'Regan, A., and Berman, J. S. (2000) Osteopontin: a key cytokine in cell-mediated and granulomatous inflammation. Int J Exp Pathol 81, 373-90.
(276) Zhang, Q., Domenicucci, C., Goldberg, H. A., Wrana, J. L., and Sodek, J. (1990) Characterization of fetal porcine bone sialoproteins, secreted phosphoprotein I (SPPI, osteopontin), bone sialoprotein, and a 23-kDa glycoprotein. Demonstration that the 23- kDa glycoprotein is derived from the carboxyl terminus of SPPI. J Biol Chem 265, 7583- 9.
(277) Craig, A. M., Smith, J. H., and Denhardt, D. T. (1989) Osteopontin, a transformation- associated cell adhesion phosphoprotein, is induced by 12-O-tetradecanoylphorbol 13- acetate in mouse epidermis. J Biol Chem 264, 9682-9.
(278) Kiefer, M. C., Bauer, D. M., and Barr, P. J. (1989) The cDNA and derived amino acid sequence for human osteopontin. Nucleic Acids Res 17, 3306.
184 (279) Prince, C. W., and Butler, W. T. (1987) 1,25-Dihydroxyvitamin D3 regulates the biosynthesis of osteopontin, a bone-derived cell attachment protein, in clonal osteoblast- like osteosarcoma cells. Coll Relat Res 7, 305-13.
(280) Crosby, A. H., Lyu, M. S., Lin, K., McBride, O. W., Kerr, J. M., Aplin, H. M., Fisher, L. W., Young, M. F., Kozak, C. A., and Dixon, M. J. (1996) Mapping of the human and mouse bone sialoprotein and osteopontin loci. Mamm Genome 7, 149-51.
(281) Nagata, T., Todescan, R., Goldberg, H. A., Zhang, Q., and Sodek, J. (1989) Sulphation of secreted phosphoprotein I (SPPI, osteopontin) is associated with mineralized tissue formation. Biochem Biophys Res Commun 165, 234-40.
(282) Prince, C. W., Oosawa, T., Butler, W. T., Tomana, M., Bhown, A. S., Bhown, M., and Schrohenloher, R. E. (1987) Isolation, characterization, and biosynthesis of a phosphorylated glycoprotein from rat bone. J Biol Chem 262, 2900-7.
(283) Beninati, S., Senger, D. R., Cordella-Miele, E., Mukherjee, A. B., Chackalaparampil, I., Shanmugam, V., Singh, K., and Mukherjee, B. B. (1994) Osteopontin: its transglutaminase-catalyzed posttranslational modifications and cross-linking to fibronectin. J Biochem 115, 675-82.
(284) Katayama, Y., House, C. M., Udagawa, N., Kazama, J. J., McFarland, R. J., Martin, T. J., and Findlay, D. M. (1998) Casein kinase 2 phosphorylation of recombinant rat osteopontin enhances adhesion of osteoclasts but not osteoblasts. J Cell Physiol 176, 179-87.
(285) Hunter, G. K., Hauschka, P. V., Poole, A. R., Rosenberg, L. C., and Goldberg, H. A. (1996) Nucleation and inhibition of hydroxyapatite formation by mineralized tissue proteins. Biochem J 317 ( Pt 1), 59-64.
(286) Razzouk, S., Brunn, J. C., Qin, C., Tye, C. E., Goldberg, H. A., and Butler, W. T. (2002) Osteopontin posttranslational modifications, possibly phosphorylation, are required for in vitro bone resorption but not osteoclast adhesion. Bone 30, 40-7.
(287) Kazanecki, C. C., Uzwiak, D. J., and Denhardt, D. T. (2007) Control of osteopontin signaling and function by post-translational phosphorylation and protein folding. J Cell Biochem 102, 912-24.
(288) Ek-Rylander, B., and Andersson, G. (2009) Osteoclast migration on phosphorylated osteopontin is regulated by endogenous tartrate-resistant acid phosphatase. Exp Cell Res.
(289) Senger, D. R., Perruzzi, C. A., Papadopoulos-Sergiou, A., and Van de Water, L. (1994) Adhesive properties of osteopontin: regulation by a naturally occurring thrombin- cleavage in close proximity to the GRGDS cell-binding domain. Mol Biol Cell 5, 565- 74.
185 (290) Yokosaki, Y., Matsuura, N., Sasaki, T., Murakami, I., Schneider, H., Higashiyama, S., Saitoh, Y., Yamakido, M., Taooka, Y., and Sheppard, D. (1999) The integrin alpha(9)beta(1) binds to a novel recognition sequence (SVVYGLR) in the thrombin- cleaved amino-terminal fragment of osteopontin. J Biol Chem 274, 36328-34.
(291) Boskey, A. L. (1995) Osteopontin and related phosphorylated sialoproteins: effects on mineralization. Ann N Y Acad Sci 760, 249-56.
(292) Agnihotri, R., Crawford, H. C., Haro, H., Matrisian, L. M., Havrda, M. C., and Liaw, L. (2001) Osteopontin, a novel substrate for matrix metalloproteinase-3 (stromelysin-1) and matrix metalloproteinase-7 (matrilysin). J Biol Chem 276, 28261-7.
(293) Smith, L. L., and Giachelli, C. M. (1998) Structural requirements for alpha 9 beta 1- mediated adhesion and migration to thrombin-cleaved osteopontin. Exp Cell Res 242, 351-60.
(294) Merry, K., Dodds, R., Littlewood, A., and Gowen, M. (1993) Expression of osteopontin mRNA by osteoclasts and osteoblasts in modelling adult human bone. J Cell Sci 104 ( Pt 4), 1013-20.
(295) Suzuki, K., Zhu, B., Rittling, S. R., Denhardt, D. T., Goldberg, H. A., McCulloch, C. A., and Sodek, J. (2002) Colocalization of intracellular osteopontin with CD44 is associated with migration, cell fusion, and resorption in osteoclasts. J Bone Miner Res 17, 1486-97.
(296) Zohar, R., Cheifetz, S., McCulloch, C. A., and Sodek, J. (1998) Analysis of intracellular osteopontin as a marker of osteoblastic cell differentiation and mesenchymal cell migration. Eur J Oral Sci 106 Suppl 1, 401-7.
(297) Shinohara, M. L., Kim, H. J., Kim, J. H., Garcia, V. A., and Cantor, H. (2008) Alternative translation of osteopontin generates intracellular and secreted isoforms that mediate distinct biological activities in dendritic cells. Proc Natl Acad Sci U S A 105, 7235-9.
(298) Denhardt, D. T., and Noda, M. (1998) Osteopontin expression and function: role in bone remodeling. J Cell Biochem Suppl 30-31, 92-102.
(299) Denda, S., Reichardt, L. F., and Muller, U. (1998) Identification of osteopontin as a novel ligand for the integrin alpha8 beta1 and potential roles for this integrin-ligand interaction in kidney morphogenesis. Mol Biol Cell 9, 1425-35.
(300) Katagiri, Y. U., Sleeman, J., Fujii, H., Herrlich, P., Hotta, H., Tanaka, K., Chikuma, S., Yagita, H., Okumura, K., Murakami, M., Saiki, I., Chambers, A. F., and Uede, T. (1999) CD44 variants but not CD44s cooperate with beta1-containing integrins to permit cells to bind to osteopontin independently of arginine-glycine-aspartic acid, thereby stimulating cell motility and chemotaxis. Cancer Res 59, 219-26.
186 (301) Singh, K., DeVouge, M. W., and Mukherjee, B. B. (1990) Physiological properties and differential glycosylation of phosphorylated and nonphosphorylated forms of osteopontin secreted by normal rat kidney cells. J Biol Chem 265, 18696-701.
(302) O'Regan, A. W., Chupp, G. L., Lowry, J. A., Goetschkes, M., Mulligan, N., and Berman, J. S. (1999) Osteopontin is associated with T cells in sarcoid granulomas and has T cell adhesive and cytokine-like properties in vitro. J Immunol 162, 1024-31.
(303) Takahashi, F., Takahashi, K., Maeda, K., Tominaga, S., and Fukuchi, Y. (2000) Osteopontin is induced by nitric oxide in RAW 264.7 cells. IUBMB Life 49, 217-21.
(304) Singhal, H., Bautista, D. S., Tonkin, K. S., O'Malley, F. P., Tuck, A. B., Chambers, A. F., and Harris, J. F. (1997) Elevated plasma osteopontin in metastatic breast cancer associated with increased tumor burden and decreased survival. Clin Cancer Res 3, 605- 11.
(305) Noti, J. D. (2000) Adherence to osteopontin via alphavbeta3 suppresses phorbol ester- mediated apoptosis in MCF-7 breast cancer cells that overexpress protein kinase C- alpha. Int J Oncol 17, 1237-43.
(306) Hullinger, T. G., Taichman, R. S., Linseman, D. A., and Somerman, M. J. (2000) Secretory products from PC-3 and MCF-7 tumor cell lines upregulate osteopontin in MC3T3-E1 cells. J Cell Biochem 78, 607-16.
(307) Rittling, S. R., Matsumoto, H. N., McKee, M. D., Nanci, A., An, X. R., Novick, K. E., Kowalski, A. J., Noda, M., and Denhardt, D. T. (1998) Mice lacking osteopontin show normal development and bone structure but display altered osteoclast formation in vitro. J Bone Miner Res 13, 1101-11.
(308) Franzen, A., Hultenby, K., Reinholt, F. P., Onnerfjord, P., and Heinegard, D. (2008) Altered osteoclast development and function in osteopontin deficient mice. J Orthop Res 26, 721-8.
(309) Yoshitake, H., Rittling, S. R., Denhardt, D. T., and Noda, M. (1999) Osteopontin- deficient mice are resistant to ovariectomy-induced bone resorption. Proc Natl Acad Sci U S A 96, 8156-60.
(310) Yumoto, K., Ishijima, M., Rittling, S. R., Tsuji, K., Tsuchiya, Y., Kon, S., Nifuji, A., Uede, T., Denhardt, D. T., and Noda, M. (2002) Osteopontin deficiency protects joints against destruction in anti-type II collagen antibody-induced arthritis in mice. Proc Natl Acad Sci U S A 99, 4556-61.
(311) Ihara, H., Denhardt, D. T., Furuya, K., Yamashita, T., Muguruma, Y., Tsuji, K., Hruska, K. A., Higashio, K., Enomoto, S., Nifuji, A., Rittling, S. R., and Noda, M. (2001) Parathyroid hormone-induced bone resorption does not occur in the absence of osteopontin. J Biol Chem 276, 13065-71.
187 (312) Ishijima, M., Rittling, S. R., Yamashita, T., Tsuji, K., Kurosawa, H., Nifuji, A., Denhardt, D. T., and Noda, M. (2001) Enhancement of osteoclastic bone resorption and suppression of osteoblastic bone formation in response to reduced mechanical stress do not occur in the absence of osteopontin. J Exp Med 193, 399-404.
(313) Ishijima, M., Tsuji, K., Rittling, S. R., Yamashita, T., Kurosawa, H., Denhardt, D. T., Nifuji, A., Ezura, Y., and Noda, M. (2007) Osteopontin is required for mechanical stress-dependent signals to bone marrow cells. J Endocrinol 193, 235-43.
(314) Hynes, R. O. (1990) in Springer Series in Molecular Biology (Rich, A., Ed.) pp 1–538, Springer, New York.
(315) Skorstengaard, K., Jensen, M. S., Sahl, P., Petersen, T. E., and Magnusson, S. (1986) Complete primary structure of bovine plasma fibronectin. Eur J Biochem 161, 441-53.
(316) Constantine, K. L., Brew, S. A., Ingham, K. C., and Llinas, M. (1992) 1H-n.m.r. studies of the fibronectin 13 kDa collagen-binding fragment. Evidence for autonomous conserved type I and type II domain folds. Biochem J 283 ( Pt 1), 247-54.
(317) Hynes, R. O., Schwarzbauer, J. E., and Tamkun, J. W. (1987) Isolation and analysis of cDNA and genomic clones of fibronectin and its receptor. Methods Enzymol 144, 447- 63.
(318) Schwarzbauer, J. E., Patel, R. S., Fonda, D., and Hynes, R. O. (1987) Multiple sites of alternative splicing of the rat fibronectin gene transcript. Embo J 6, 2573-80.
(319) Schwarzbauer, J. E. (1991) Alternative splicing of fibronectin: three variants, three functions. Bioessays 13, 527-33.
(320) Schwarzbauer, J. E. (1991) Fibronectin: from gene to protein. Curr Opin Cell Biol 3, 786-91.
(321) Wayner, E. A., Garcia-Pardo, A., Humphries, M. J., McDonald, J. A., and Carter, W. G. (1989) Identification and characterization of the T lymphocyte adhesion receptor for an alternative cell attachment domain (CS-1) in plasma fibronectin. J Cell Biol 109, 1321- 30.
(322) Fogerty, F. J., and Mosher, D. F. (1990) Mechanisms for organization of fibronectin matrix. Cell Differ Dev 32, 439-50.
(323) Fogerty, F. J., Akiyama, S. K., Yamada, K. M., and Mosher, D. F. (1990) Inhibition of binding of fibronectin to matrix assembly sites by anti-integrin (alpha 5 beta 1) antibodies. J Cell Biol 111, 699-708.
(324) Schwarzbauer, J. E. (1991) Identification of the fibronectin sequences required for assembly of a fibrillar matrix. J Cell Biol 113, 1463-73.
188 (325) Takahashi, S., Leiss, M., Moser, M., Ohashi, T., Kitao, T., Heckmann, D., Pfeifer, A., Kessler, H., Takagi, J., Erickson, H. P., and Fassler, R. (2007) The RGD motif in fibronectin is essential for development but dispensable for fibril assembly. J Cell Biol 178, 167-78.
(326) Johansson, S., Svineng, G., Wennerberg, K., Armulik, A., and Lohikangas, L. (1997) Fibronectin-integrin interactions. Front Biosci 2, d126-46.
(327) Moursi, A. M., Damsky, C. H., Lull, J., Zimmerman, D., Doty, S. B., Aota, S., and Globus, R. K. (1996) Fibronectin regulates calvarial osteoblast differentiation. J Cell Sci 109 ( Pt 6), 1369-80.
(328) Kawelke, N., Bentmann, A., Hackl, N., Hager, H. D., Feick, P., Geursen, A., Singer, M. V., and Nakchbandi, I. A. (2008) Isoform of fibronectin mediates bone loss in patients with primary biliary cirrhosis by suppressing bone formation. J Bone Miner Res 23, 1278-86.
(329) Hayman, E. G., Pierschbacher, M. D., Ohgren, Y., and Ruoslahti, E. (1983) Serum spreading factor (vitronectin) is present at the cell surface and in tissues. Proc Natl Acad Sci U S A 80, 4003-7.
(330) Barnes, D. W., Reing, J. E., and Amos, B. (1985) Heparin-binding properties of human serum spreading factor. J Biol Chem 260, 9117-22.
(331) Jenne, D., and Stanley, K. K. (1985) Molecular cloning of S-protein, a link between complement, coagulation and cell-substrate adhesion. Embo J 4, 3153-7.
(332) Hetland, G., Pettersen, H. B., Mollnes, T. E., and Johnson, E. (1989) S-protein is synthesized by human monocytes and macrophages in vitro. Scand J Immunol 29, 15-21.
(333) Hayman, E. G., Pierschbacher, M. D., Suzuki, S., and Ruoslahti, E. (1985) Vitronectin-- a major cell attachment-promoting protein in fetal bovine serum. Exp Cell Res 160, 245- 58.
(334) Suzuki, S., Oldberg, A., Hayman, E. G., Pierschbacher, M. D., and Ruoslahti, E. (1985) Complete amino acid sequence of human vitronectin deduced from cDNA. Similarity of cell attachment sites in vitronectin and fibronectin. Embo J 4, 2519-24.
(335) Deng, G., Royle, G., Seiffert, D., and Loskutoff, D. J. (1995) The PAI-1/vitronectin interaction: two cats in a bag? Thromb Haemost 74, 66-70.
(336) Seiffert, D. (1996) Detection of vitronectin in mineralized bone matrix. J Histochem Cytochem 44, 275-80.
(337) Manolagas, S. C. (2000) Birth and death of bone cells: basic regulatory mechanisms and implications for the pathogenesis and treatment of osteoporosis. Endocr Rev 21, 115-37.
189 (338) Mullan, R. H., and Bresnihan, B. (2003) Disease-modifying anti-rheumatic drug therapy and structural damage in early rheumatoid arthritis. Clin Exp Rheumatol 21, S158-64.
(339) Teitelbaum, S. L., and Ross, F. P. (2003) Genetic regulation of osteoclast development and function. Nat Rev Genet 4, 638-49.
(340) Geiger, B., Bershadsky, A., Pankov, R., and Yamada, K. M. (2001) Transmembrane crosstalk between the extracellular matrix--cytoskeleton crosstalk. Nat Rev Mol Cell Biol 2, 793-805.
(341) Faccio, R., Takeshita, S., Colaianni, G., Chappel, J., Zallone, A., Teitelbaum, S. L., and Ross, F. P. (2007) M-CSF regulates the cytoskeleton via recruitment of a multimeric signaling complex to c-Fms Tyr-559/697/721. J Biol Chem 282, 18991-9.
(342) Teti, A., Taranta, A., Migliaccio, S., Degiorgi, A., Santandrea, E., Villanova, I., Faraggiana, T., Chellaiah, M., and Hruska, K. A. (1998) Colony stimulating factor-1- induced osteoclast spreading depends on substrate and requires the vitronectin receptor and the c-src proto-oncogene. J Bone Miner Res 13, 50-8.
(343) van der Pluijm, G., Mouthaan, H., Baas, C., de Groot, H., Papapoulos, S., and Lowik, C. (1994) Integrins and osteoclastic resorption in three bone organ cultures: differential sensitivity to synthetic Arg-Gly-Asp peptides during osteoclast formation. J Bone Miner Res 9, 1021-8.
(344) Chambers, T. J., Fuller, K., Darby, J. A., Pringle, J. A., and Horton, M. A. (1986) Monoclonal antibodies against osteoclasts inhibit bone resorption in vitro. Bone Miner 1, 127-35.
(345) Horton, M. A., Taylor, M. L., Arnett, T. R., and Helfrich, M. H. (1991) Arg-Gly-Asp (RGD) peptides and the anti-vitronectin receptor antibody 23C6 inhibit dentine resorption and cell spreading by osteoclasts. Exp Cell Res 195, 368-75.
(346) Nakamura, I., Gailit, J., and Sasaki, T. (1996) Osteoclast integrin alphaVbeta3 is present in the clear zone and contributes to cellular polarization. Cell Tissue Res 286, 507-15.
(347) Carron, C. P., Meyer, D. M., Engleman, V. W., Rico, J. G., Ruminski, P. G., Ornberg, R. L., Westlin, W. F., and Nickols, G. A. (2000) Peptidomimetic antagonists of alphavbeta3 inhibit bone resorption by inhibiting osteoclast bone resorptive activity, not osteoclast adhesion to bone. J Endocrinol 165, 587-98.
(348) Meissner, R. S., Perkins, J. J., Duong le, T., Hartman, G. D., Hoffman, W. F., Huff, J. R., Ihle, N. C., Leu, C. T., Nagy, R. M., Naylor-Olsen, A., Rodan, G. A., Rodan, S. B., Whitman, D. B., Wesolowski, G. A., and Duggan, M. E. (2002) Nonpeptide alpha(v)beta(3) antagonists. Part 2: constrained glycyl amides derived from the RGD tripeptide. Bioorg Med Chem Lett 12, 25-9.
190 (349) Wu, H., Beuerlein, G., Nie, Y., Smith, H., Lee, B. A., Hensler, M., Huse, W. D., and Watkins, J. D. (1998) Stepwise in vitro affinity maturation of Vitaxin, an alphav beta3- specific humanized mAb. Proc Natl Acad Sci U S A 95, 6037-42.
(350) Wilder, R. L. (2002) Integrin alpha V beta 3 as a target for treatment of rheumatoid arthritis and related rheumatic diseases. Ann Rheum Dis 61 Suppl 2, ii96-9.
(351) Tucker, G. C. (2006) Integrins: molecular targets in cancer therapy. Curr Oncol Rep 8, 96-103.
(352) Gutheil, J. C., Campbell, T. N., Pierce, P. R., Watkins, J. D., Huse, W. D., Bodkin, D. J., and Cheresh, D. A. (2000) Targeted antiangiogenic therapy for cancer using Vitaxin: a humanized monoclonal antibody to the integrin alphavbeta3. Clin Cancer Res 6, 3056- 61.
(353) Patel, S. R., Jenkins, J., Papadopolous, N., Burgess, M. A., Plager, C., Gutterman, J., and Benjamin, R. S. (2001) Pilot study of vitaxin--an angiogenesis inhibitor-in patients with advanced leiomyosarcomas. Cancer 92, 1347-8.
(354) Posey, J. A., Khazaeli, M. B., DelGrosso, A., Saleh, M. N., Lin, C. Y., Huse, W., and LoBuglio, A. F. (2001) A pilot trial of Vitaxin, a humanized anti-vitronectin receptor (anti alpha v beta 3) antibody in patients with metastatic cancer. Cancer Biother Radiopharm 16, 125-32.
(355) Lees, R. L., Sabharwal, V. K., and Heersche, J. N. (2001) Resorptive state and cell size influence intracellular pH regulation in rabbit osteoclasts cultured on collagen- hydroxyapatite films. Bone 28, 187-94.
(356) Trebec, D. P., Chandra, D., Gramoun, A., Li, K., Heersche, J. N., and Manolson, M. F. (2007) Increased expression of activating factors in large osteoclasts could explain their excessive activity in osteolytic diseases. J Cell Biochem 101, 205-20.
(357) Frelinger, A. L., 3rd, Du, X. P., Plow, E. F., and Ginsberg, M. H. (1991) Monoclonal antibodies to ligand-occupied conformers of integrin alpha IIb beta 3 (glycoprotein IIb- IIIa) alter receptor affinity, specificity, and function. J Biol Chem 266, 17106-11.
(358) Mikecz, K. (2000) Vitaxin applied molecular evolution. Curr Opin Investig Drugs 1, 199-203.
(359) Lees, R. L., and Heersche, J. N. (2000) Differences in regulation of pH(i) in large (>/=10 nuclei) and small (
(360) Mould, A. P., Askari, J. A., Barton, S., Kline, A. D., McEwan, P. A., Craig, S. E., and Humphries, M. J. (2002) Integrin activation involves a conformational change in the alpha 1 helix of the beta subunit A-domain. J Biol Chem 277, 19800-5.
191 (361) Pelletier, A. J., Kunicki, T., Ruggeri, Z. M., and Quaranta, V. (1995) The activation state of the integrin alpha IIb beta 3 affects outside-in signals leading to cell spreading and focal adhesion kinase phosphorylation. J Biol Chem 270, 18133-40.
(362) Spillmann, C., Osorio, D., and Waugh, R. (2002) Integrin activation by divalent ions affects neutrophil homotypic adhesion. Ann Biomed Eng 30, 1002-11.
(363) Silman, A. J., and Pearson, J. E. (2002) Epidemiology and genetics of rheumatoid arthritis. Arthritis Res 4 Suppl 3, S265-72.
(364) Baron, R., Neff, L., Louvard, D., and Courtoy, P. J. (1985) Cell-mediated extracellular acidification and bone resorption: evidence for a low pH in resorbing lacunae and localization of a 100-kD lysosomal membrane protein at the osteoclast ruffled border. J Cell Biol 101, 2210-22.
(365) Novack, D. V., and Teitelbaum, S. L. (2008) The osteoclast: friend or foe? Annu Rev Pathol 3, 457-84.
(366) Hoshino, K., Hanyu, T., Arai, K., and Takahashi, H. E. (2001) Mineral density and histomorphometric assessment of bone changes in the proximal tibia early after induction of type II collagen-induced arthritis in growing and mature rats. J Bone Miner Metab 19, 76-83.
(367) Takayanagi, H. (2009) Osteoimmunology and the effects of the immune system on bone. Nat Rev Rheumatol.
(368) Shimizu, S., Shiozawa, S., Shiozawa, K., Imura, S., and Fujita, T. (1985) Quantitative histologic studies on the pathogenesis of periarticular osteoporosis in rheumatoid arthritis. Arthritis Rheum 28, 25-31.
(369) Shibutani, T., Murahashi, Y., Tsukada, E., Iwayama, Y., and Heersche, J. N. (1997) Experimentally induced periodontitis in beagle dogs causes rapid increases in osteoclastic resorption of alveolar bone. J Periodontol 68, 385-91.
(370) Ritchlin, C. T., Haas-Smith, S. A., Li, P., Hicks, D. G., and Schwarz, E. M. (2003) Mechanisms of TNF-alpha- and RANKL-mediated osteoclastogenesis and bone resorption in psoriatic arthritis. J Clin Invest 111, 821-31.
(371) Seiffert, D., Crain, K., Wagner, N. V., and Loskutoff, D. J. (1994) Vitronectin gene expression in vivo. Evidence for extrahepatic synthesis and acute phase regulation. J Biol Chem 269, 19836-42.
(372) Saeki, Y., Mima, T., Ishii, T., Ogata, A., Kobayashi, H., Ohshima, S., Ishida, T., Tabunoki, Y., Kitayama, H., Mizuki, M., Katada, Y., Asaoku, H., Kitano, M., Nishimoto, N., Yoshizaki, K., Maeda, M., Kon, S., Kinoshita, N., Uede, T., and Kawase, I. (2003) Enhanced production of osteopontin in multiple myeloma: clinical and pathogenic implications. Br J Haematol 123, 263-70. 192 (373) Oates, A. J., Barraclough, R., and Rudland, P. S. (1996) The identification of osteopontin as a metastasis-related gene product in a rodent mammary tumour model. Oncogene 13, 97-104.
(374) Patarca, R., Saavedra, R. A., and Cantor, H. (1993) Molecular and cellular basis of genetic resistance to bacterial infection: the role of the early T-lymphocyte activation- 1/osteopontin gene. Crit Rev Immunol 13, 225-46.
(375) Seiffert, D., Geisterfer, M., Gauldie, J., Young, E., and Podor, T. J. (1995) IL-6 stimulates vitronectin gene expression in vivo. J Immunol 155, 3180-5.
(376) Lavietes, B. B., Carsons, S., Diamond, H. S., and Laskin, R. S. (1985) Synthesis, secretion, and deposition of fibronectin in cultured human synovium. Arthritis Rheum 28, 1016-26.
(377) Xu, G., Nie, H., Li, N., Zheng, W., Zhang, D., Feng, G., Ni, L., Xu, R., Hong, J., and Zhang, J. Z. (2005) Role of osteopontin in amplification and perpetuation of rheumatoid synovitis. J Clin Invest 115, 1060-7.
(378) Saito, S., Yamaji, N., Yasunaga, K., Saito, T., Matsumoto, S., Katoh, M., Kobayashi, S., and Masuho, Y. (1999) The fibronectin extra domain A activates matrix metalloproteinase gene expression by an interleukin-1-dependent mechanism. J Biol Chem 274, 30756-63.
(379) Contractor, T., Babiarz, B., Kowalski, A. J., Rittling, S. R., Sorensen, E. S., and Denhardt, D. T. (2005) Osteoclasts resorb protein-free mineral (Osteologic discs) efficiently in the absence of osteopontin. In Vivo 19, 335-41.
(380) Rajachar, R. M., Truong, A. Q., and Giachelli, C. M. (2008) The influence of surface mineral and osteopontin on the formation and function of murine bone marrow-derived osteoclasts. J Mater Sci Mater Med 19, 3279-85.
(381) Hu, Y., Ek-Rylander, B., Karlstrom, E., Wendel, M., and Andersson, G. (2008) Osteoclast size heterogeneity in rat long bones is associated with differences in adhesive ligand specificity. Exp Cell Res 314, 638-50.
(382) Voronov, I., Heersche, J. N., Casper, R. F., Tenenbaum, H. C., and Manolson, M. F. (2005) Inhibition of osteoclast differentiation by polycyclic aryl hydrocarbons is dependent on cell density and RANKL concentration. Biochem Pharmacol 70, 300-7.
(383) Gramoun, A., Shorey, S., Bashutski, J. D., Dixon, S. J., Sims, S. M., Heersche, J. N., and Manolson, M. F. (2007) Effects of Vitaxin, a novel therapeutic in trial for metastatic bone tumors, on osteoclast functions in vitro. J Cell Biochem 102, 341-52.
(384) Sakai, T., Johnson, K. J., Murozono, M., Sakai, K., Magnuson, M. A., Wieloch, T., Cronberg, T., Isshiki, A., Erickson, H. P., and Fassler, R. (2001) Plasma fibronectin
193 supports neuronal survival and reduces brain injury following transient focal cerebral ischemia but is not essential for skin-wound healing and hemostasis. Nat Med 7, 324-30.
(385) Parfitt, A. M., Drezner, M. K., Glorieux, F. H., Kanis, J. A., Malluche, H., Meunier, P. J., Ott, S. M., and Recker, R. R. (1987) Bone histomorphometry: standardization of nomenclature, symbols, and units. Report of the ASBMR Histomorphometry Nomenclature Committee. J Bone Miner Res 2, 595-610.
(386) George, E. L., Georges-Labouesse, E. N., Patel-King, R. S., Rayburn, H., and Hynes, R. O. (1993) Defects in mesoderm, neural tube and vascular development in mouse embryos lacking fibronectin. Development 119, 1079-91.
(387) Miesel, R., Kurpisz, M., and Kroger, H. (1996) Suppression of inflammatory arthritis by simultaneous inhibition of nitric oxide synthase and NADPH oxidase. Free Radic Biol Med 20, 75-81.
(388) McCartney-Francis, N., Allen, J. B., Mizel, D. E., Albina, J. E., Xie, Q. W., Nathan, C. F., and Wahl, S. M. (1993) Suppression of arthritis by an inhibitor of nitric oxide synthase. J Exp Med 178, 749-54.
(389) Yasuda, T., Kakinuma, T., Julovi, S. M., Yoshida, M., Hiramitsu, T., Akiyoshi, M., and Nakamura, T. (2004) COOH-terminal heparin-binding fibronectin fragment induces nitric oxide production in rheumatoid cartilage through CD44. Rheumatology (Oxford) 43, 1116-20.
(390) Yasuda, T., and Nakamura, T. (2007) Inhibition of nuclear factor-kappaB by hyaluronan in rheumatoid chondrocytes stimulated with COOH-terminal heparin-binding fibronectin fragment. Mod Rheumatol 17, 391-7.
(391) Brandi, M. L., Hukkanen, M., Umeda, T., Moradi-Bidhendi, N., Bianchi, S., Gross, S. S., Polak, J. M., and MacIntyre, I. (1995) Bidirectional regulation of osteoclast function by nitric oxide synthase isoforms. Proc Natl Acad Sci U S A 92, 2954-8.
(392) Yasuda, T. (2006) Cartilage destruction by matrix degradation products. Mod Rheumatol 16, 197-205.
(393) Schmidt, D. R., and Kao, W. J. (2007) The interrelated role of fibronectin and interleukin-1 in biomaterial-modulated macrophage function. Biomaterials 28, 371-82.
(394) Schwartz, M. A., Both, G., and Lechene, C. (1989) Effect of cell spreading on cytoplasmic pH in normal and transformed fibroblasts. Proc Natl Acad Sci U S A 86, 4525-9.
(395) Ruoslahti, E., and Pierschbacher, M. D. (1987) New perspectives in cell adhesion: RGD and integrins. Science 238, 491-7.
194 (396) Halleen, J. M., Alatalo, S. L., Suominen, H., Cheng, S., Janckila, A. J., and Vaananen, H. K. (2000) Tartrate-resistant acid phosphatase 5b: a novel serum marker of bone resorption. J Bone Miner Res 15, 1337-45.
(397) Baneyx, G., Baugh, L., and Vogel, V. (2001) Coexisting conformations of fibronectin in cell culture imaged using fluorescence resonance energy transfer. Proc Natl Acad Sci U S A 98, 14464-8.
(398) Mao, Y., and Schwarzbauer, J. E. (2005) Fibronectin fibrillogenesis, a cell-mediated matrix assembly process. Matrix Biol 24, 389-99.
(399) Pellenc, D., Berry, H., and Gallet, O. (2006) Adsorption-induced fibronectin aggregation and fibrillogenesis. J Colloid Interface Sci 298, 132-44.
(400) Dolatshahi-Pirouz, A., Jensen, T., Foss, M., Chevallier, J., and Besenbacher, F. (2009) Enhanced Surface Activation of Fibronectin upon Adsorption on Hydroxyapatite. Langmuir.
(401) Carano, A., Schlesinger, P. H., Athanasou, N. A., Teitelbaum, S. L., and Blair, H. C. (1993) Acid and base effects on avian osteoclast activity. Am J Physiol 264, C694-701.
(402) Arnett, T. R., Boyde, A., Jones, S. J., and Taylor, M. L. (1994) Effects of medium acidification by alteration of carbon dioxide or bicarbonate concentrations on the resorptive activity of rat osteoclasts. J Bone Miner Res 9, 375-9.
(403) Beezhold, D. H., and Lause, D. B. (1987) Stimulation of rat macrophage interleukin 1 secretion by plasma fibronectin. Immunol Invest 16, 437-49.
(404) Schiffer, R., Klein, B., Klosterhalfen, B., and Zwadlo-Klarwasser, G. (1999) The contact of human macrophages with extracellular matrix proteins selectively induces expression of proinflammatory cytokines. Pathobiology 67, 233-5.
(405) Charles, I. G., Palmer, R. M., Hickery, M. S., Bayliss, M. T., Chubb, A. P., Hall, V. S., Moss, D. W., and Moncada, S. (1993) Cloning, characterization, and expression of a cDNA encoding an inducible nitric oxide synthase from the human chondrocyte. Proc Natl Acad Sci U S A 90, 11419-23.
(406) Miyasaka, N., and Hirata, Y. (1997) Nitric oxide and inflammatory arthritides. Life Sci 61, 2073-81.
(407) Sunyer, T., Rothe, L., Jiang, X., Osdoby, P., and Collin-Osdoby, P. (1996) Proinflammatory agents, IL-8 and IL-10, upregulate inducible nitric oxide synthase expression and nitric oxide production in avian osteoclast-like cells. J Cell Biochem 60, 469-83.
(408) Jalkanen, S., and Jalkanen, M. (1992) Lymphocyte CD44 binds the COOH-terminal heparin-binding domain of fibronectin. J Cell Biol 116, 817-25. 195 (409) Hofmann, G., Bernabei, P. A., Crociani, O., Cherubini, A., Guasti, L., Pillozzi, S., Lastraioli, E., Polvani, S., Bartolozzi, B., Solazzo, V., Gragnani, L., Defilippi, P., Rosati, B., Wanke, E., Olivotto, M., and Arcangeli, A. (2001) HERG K+ channels activation during beta(1) integrin-mediated adhesion to fibronectin induces an up-regulation of alpha(v)beta(3) integrin in the preosteoclastic leukemia cell line FLG 29.1. J Biol Chem 276, 4923-31.
(410) Wang, K. X., and Denhardt, D. T. (2008) Osteopontin: role in immune regulation and stress responses. Cytokine Growth Factor Rev 19, 333-45.
(411) Al-Shami, R., Sorensen, E. S., Ek-Rylander, B., Andersson, G., Carson, D. D., and Farach-Carson, M. C. (2005) Phosphorylated osteopontin promotes migration of human choriocarcinoma cells via a p70 S6 kinase-dependent pathway. J Cell Biochem 94, 1218- 33.
(412) Chellaiah, M. A., Biswas, R. S., Rittling, S. R., Denhardt, D. T., and Hruska, K. A. (2003) Rho-dependent Rho kinase activation increases CD44 surface expression and bone resorption in osteoclasts. J Biol Chem 278, 29086-97.
(413) Yasuda, Y., Kaleta, J., and Bromme, D. (2005) The role of cathepsins in osteoporosis and arthritis: rationale for the design of new therapeutics. Adv Drug Deliv Rev 57, 973- 93.
(414) Owens, J., and Chambers, T. J. (1993) Macrophage colony-stimulating factor (M-CSF) induces migration in osteoclasts in vitro. Biochem Biophys Res Commun 195, 1401-7.
(415) Manolson, M. F., Yu, H., Chen, W., Yao, Y., Li, K., Lees, R. L., and Heersche, J. N. (2003) The a3 isoform of the 100-kDa V-ATPase subunit is highly but differentially expressed in large (>or=10 nuclei) and small ( (416) Edwards, J. C., Cohen, C., Xu, W., and Schlesinger, P. H. (2006) c-Src control of chloride channel support for osteoclast HCl transport and bone resorption. J Biol Chem 281, 28011-22. (417) Zaidi, M., Alam, A. S., Shankar, V. S., Bax, B. E., Moonga, B. S., Bevis, P. J., Pazianas, M., and Huang, C. L. (1992) A quantitative description of components of in vitro morphometric change in the rat osteoclast model: relationships with cellular function. Eur Biophys J 21, 349-55. (418) Thomas, J. A., Buchsbaum, R. N., Zimniak, A., and Racker, E. (1979) Intracellular pH measurements in Ehrlich ascites tumor cells utilizing spectroscopic probes generated in situ. Biochemistry 18, 2210-8. (419) Lees, R. L., and Heersche, J. N. (1999) Macrophage colony stimulating factor increases bone resorption in dispersed osteoclast cultures by increasing osteoclast size. J Bone Miner Res 14, 937-45. 196 (420) Nakamura, I., Rodan, G. A., and Duong le, T. (2003) Distinct roles of p130Cas and c- Cbl in adhesion-induced or macrophage colony-stimulating factor-mediated signaling pathways in prefusion osteoclasts. Endocrinology 144, 4739-41. (421) Sakai, H., Chen, Y., Itokawa, T., Yu, K. P., Zhu, M. L., and Insogna, K. (2006) Activated c-Fms recruits Vav and Rac during CSF-1-induced cytoskeletal remodeling and spreading in osteoclasts. Bone 39, 1290-301. (422) Hunter, S. J., Rosen, C. J., and Gay, C. V. (1991) In vitro resorptive activity of isolated chick osteoclasts: effects of carbonic anhydrase inhibition. J Bone Miner Res 6, 61-6. (423) Ingber, D. E., Prusty, D., Frangioni, J. V., Cragoe, E. J., Jr., Lechene, C., and Schwartz, M. A. (1990) Control of intracellular pH and growth by fibronectin in capillary endothelial cells. J Cell Biol 110, 1803-11. (424) Schwartz, M. A., Lechene, C., and Ingber, D. E. (1991) Insoluble fibronectin activates the Na/H antiporter by clustering and immobilizing integrin alpha 5 beta 1, independent of cell shape. Proc Natl Acad Sci U S A 88, 7849-53. (425) Makris, G. P., and Saffar, J. L. (1982) Quantitative relationship between osteoclasts, osteoclast nuclei and the extent of the resorbing surface in hamster periodontal disease. Arch Oral Biol 27, 965-9. (426) Singer, F. R. (1996) Paget's disease of bone possible viral basis. Trends Endocrinol Metab 7, 258-61. (427) Homandberg, G. A. (1999) Potential regulation of cartilage metabolism in osteoarthritis by fibronectin fragments. Front Biosci 4, D713-30. (428) Rabkin, S. W. Nitric oxide and peroxynitrite induce gene expression of interleukin receptors increasing IL-21, IL-7, IL-1 and oncostatin M in cardiomyocytes. Life Sci 86, 45-51. (429) Ralston, S. H., Ho, L. P., Helfrich, M. H., Grabowski, P. S., Johnston, P. W., and Benjamin, N. (1995) Nitric oxide: a cytokine-induced regulator of bone resorption. J Bone Miner Res 10, 1040-9. (430) van der Wees, C. G., Bax, W. H., van der Valk, E. J., and van der Laarse, A. (2006) Integrin stimulation induces calcium signalling in rat cardiomyocytes by a NO- dependent mechanism. Pflugers Arch 451, 588-95. (431) van't Hof, R. J., Armour, K. J., Smith, L. M., Armour, K. E., Wei, X. Q., Liew, F. Y., and Ralston, S. H. (2000) Requirement of the inducible nitric oxide synthase pathway for IL-1-induced osteoclastic bone resorption. Proc Natl Acad Sci U S A 97, 7993-8. 197 (432) Arnett, T. R., and Dempster, D. W. (1987) A comparative study of disaggregated chick and rat osteoclasts in vitro: effects of calcitonin and prostaglandins. Endocrinology 120, 602-8. (433) Assapun, J., Charoenphandhu, N., and Krishnamra, N. (2009) Early acceleration phase and late stationary phase of remodeling imbalance in long bones of male rats exposed to long-standing acidemia: a 10-month longitudinal study using bone histomorphometry. Calcif Tissue Int 85, 1-9. (434) Meghji, S., Morrison, M. S., Henderson, B., and Arnett, T. R. (2001) pH dependence of bone resorption: mouse calvarial osteoclasts are activated by acidosis. Am J Physiol Endocrinol Metab 280, E112-9. (435) Arnett, T. R., and Spowage, M. (1996) Modulation of the resorptive activity of rat osteoclasts by small changes in extracellular pH near the physiological range. Bone 18, 277-9. (436) Jahr, H., van Driel, M., van Osch, G. J., Weinans, H., and van Leeuwen, J. P. (2005) Identification of acid-sensing ion channels in bone. Biochem Biophys Res Commun 337, 349-54. (437) Kim, J. M., Min, S. K., Kim, H., Kang, H. K., Jung, S. Y., Lee, S. H., Choi, Y., Roh, S., Jeong, D., and Min, B. M. (2007) Vacuolar-type H+-ATPase-mediated acidosis promotes in vitro osteoclastogenesis via modulation of cell migration. Int J Mol Med 19, 393-400. (438) Komarova, S. V., Pereverzev, A., Shum, J. W., Sims, S. M., and Dixon, S. J. (2005) Convergent signaling by acidosis and receptor activator of NF-kappaB ligand (RANKL) on the calcium/calcineurin/NFAT pathway in osteoclasts. Proc Natl Acad Sci U S A 102, 2643-8. (439) Trebec-Reynolds, D. P., Voronov, I., Heersche, J. N., and Manolson, M. F. IL-1alpha and IL-1beta have different effects on formation and activity of large osteoclasts. J Cell Biochem. (440) Brisseau, G. F., Grinstein, S., Hackam, D. J., Nordstrom, T., Manolson, M. F., Khine, A. A., and Rotstein, O. D. (1996) Interleukin-1 increases vacuolar-type H+-ATPase activity in murine peritoneal macrophages. J Biol Chem 271, 2005-11. (441) Morla, A., Zhang, Z., and Ruoslahti, E. (1994) Superfibronectin is a functionally distinct form of fibronectin. Nature 367, 193-6. (442) Ohashi, T., and Erickson, H. P. (2005) Domain unfolding plays a role in superfibronectin formation. J Biol Chem 280, 39143-51. 198