The role of Rac and Filamin A in osteoclast function and osteoclast heterogeneity

By

Stephanie Goldberg

A thesis submitted in conformity with the requirements for the Degree of Doctor of Philosophy

Faculty of Dentistry

University of Toronto

© Copyright Stephanie Goldberg 2015

ii

The Role of Rac and Filamin A in osteoclast function and osteoclast heterogeneity

Stephanie Goldberg

Degree of Doctor of Philosophy

Faculty of Dentistry

University of Toronto

2015

Abstract

Osteoclasts are multinucleated cells that resorb bone and carry out their function via a dynamic actin

cytoskeleton that is regulated by Rho GTPases. Rho GTPases function as molecular switches that

regulate the actin cytoskeleton in osteoclasts and many other cell types. A Rac GTPase validation

experiment was used to prove that osteoclast function is best characterized through three-dimensional (3-

D) volumetric analysis of resorption pits (Rpits). The success of this method prompted us to measure

Rpits in osteoclasts lacking actin binding protein Filamin A (FLNA), which interacts with Rho GTPases

to regulate osteoclast formation (osteoclastogenesis (OCG)) and function. It was observed using this

technique that FLNA-knock-out (KO) osteoclasts were smaller and produced shallower Rpits compared

to WT osteoclasts. Using an osteoporosis model as well as in vivo techniques measuring bone material,

structural and mechanical properties, it was demonstrated that FLNA knockout (KO) bones were

iii

protected from post-menopausal osteoporosis associated with estrogen depletion. FLNA-KO cortical

bones were also found to be more ductile than WT cortical bones. The discrepancy between bone

behavior in vertebrae vs. femora in FLNA mice triggered the question as to whether osteoclasts behave

differently depending on the skeletal site from which they originated. The osteoporosis model, resorption

pit method and aforementioned in vivo techniques with the exception of bone mechanical properties were

used to determine whether osteoclasts and bone turnover differed between the cranial and appendicular

skeletons. Calvarial bone displayed higher turnover rates compared to long bone and mandibular bone.

The novel in vitro 3-D method, characterization of FLNA in vivo and the assessment of osteoclasts

arising from different bone locations were executed to provide more insight into the development of

pharmacotherapeutics to better manage pathological bone loss.

Acknowledgements

iv

I would like to thank my supervisor Dr. Michael Glogauer for his support and encouragement throughout

this journey. Dr. Glogauer was always compassionate and instilled confidence in me in difficult times. I

would also like to thank my co-supervisor Dr. Marc Grynpas for sharing his expertise and guiding me

through my experimental and writing endeavors. Thank you to all my lab mates for helping me every

step of the way as well as your kindness and friendship throughout my stay in the lab. I would also like

to acknowledge my friends and family, especially my other half Corey Gross for your love and belief in

me. I would not be where I am today without your inspiration and support.

v

Table of Contents

Chapter 1 1

Literature Review 1.1 Motivation and Rationale 1 1.2 Bone Biology 3 1.3 Bone Structure 4 1.4 Bone cells 8 1.5 Bone remodeling 9 1.6 Bone Quality 11 1.7 2-D vs. 3-D assessment of osteoclastic bone resorption 15 1.8 Osteoporosis 17 1.9 Osteoclasts and OCG 21 1.10 Rho (Rac) GTPases 23 1.11 FLNA Discovery 26 1.12 Objectives and Hypotheses 31

Chapter 2 33 A 3D scanning confocal imaging method measures Rpit volume and captures the role of Rac in osteoclast function 33 2.1 Abstract 33 2.2 Introduction 34 2.3 Materials and Methods 36 2.4 Results 40 2.5 Discussion 42 2.6 Conclusions 46 2.7 Figures 47 2.8 Acknowledgements 51 Preface to Chapter 3 52

Chapter 3 53 Deletion of Filamin A protects cortical and trabecular bone from post-menopausal changes in bone microarchitecture 53 3.1 Abstract 53 3.2 Introduction 54 3.3 Materials and Methods 57 3.4 Results 62 3.5 Discussion 66 3.6 Figures 70 3.7 Tables 75 3.7 Acknowledgments 78 Preface to Chapter 4 79

Chapter 4 80 Heterogeneity of osteoclast activity and bone turnover in different skeletal sites 80 4.1 Abstract 80 4.2 Introduction 82 4.3 Materials and Methods 84 4.4 Results 86

vi

4.5 Discussion 88 4.6 Figures 92 4.7 Tables 96

Chapter 5 98 5.1 Summary and Conclusions 98 5.2 Future Directions 103

Chapter 6 105 References 105

vii

List of Tables

3.7.1 Table 1. Micro-CT of aging femora

3.7.2 Table 2. Micro-CT of aging vertebrae

3.7.3 Table 3. Three-point bending of aging femora

3.7.4 Table 4. TRAP histomorphometry of 6-month non-OVX vs. OVX vertebrae

3.7.5 Table 5. Static histomorphometry of 6-month non-OVX vs. OVX vertebrae

3.7.6 Table 6. Dynamic histomorphometry of 6-month non-OVX vs. OVX vertebrae

4.7.1 Table 1. Pre-OVX trabecular architecture derived from undecalcified histomorphometry

4.7.2 Table 2. Post-OVX trabecular architecture derived from undecalcified histomorphometry

4.7.3 Table 3. Summary table

viii

List of Figures

2.7.1 Figure 1. Rpit 3-D visualization

2.7.2 Figure 2. Volume and Surface Area of Rpits

2.7.3 Figure 3. Morphology of Osteoclasts

2.7.4 Figure 4. Model of Rpits

3.6.1 Figure 1. Volume and Surface Area of Osteoclasts and Rpits

3.6.2 Figure 2. Phenotype of 5-MO FLNA-KO bones (Material and Structural Properties)

3.6.3 Figure 3. Phenotype of FLNA-KO bones (Mechanical Properties)

3.6.4 Figure 4. Phenotype of FLNA-KO-OVX bones (Material and Structural Properties)

3.6.5 Figure 5. Phenotype of FLNA-KO-OVX femora (Mechanical Properties)

3.6.6 Figure 6. Phenotype of FLNA-KO-OVX trabecular bone (Material and Structural Properties)

4.6.1 Figure 1. in vitro OCG

4.6.2 Figure 2. in vivo decalcified histology

4.6.3. Figure 3. in vivo undecalcified histomorphometry (bone formation)

4.6.4. Figure 4. in vivo undecalcified histomorphometry (bone formation rate)

4.6.5. Figure 5. in vivo decalcified histology

4.6.6. Figure 6. in vivo undecalcified histomorphometry

4.6.7 Figure 7. in vivo undecalcified histomorphometry

Abbreviations

ix

2-D Two-dimensional

3-D Three-dimensional

α-MEM α-minimal essential medium

ABD Actin binding domain

aBMD Areal bone mineral density

AGE Advanced glycation end products

AP Anterior-Posterior

AP-1 Activator protein 1

B.Ar Bone area

BFR Bone formation rate

BMC Bone mineral content

BMD Bone mineral density

BMP Bone morphogenetic protein

BMU Basic multicellular unit

BSE Back-scattered electron imaging

BV Bone volume

BV/TV Bone volume over tissue volume

CaM kinase II Ca2+/calmodulin-dependent protein kinase II

Cs.Th Cross-sectional thickness

CTSK Cathepsin K

DAPI 4,6,-diamidino-2-phenylindole

DEXA Dual energy x-ray absorptiometry

EDTA Ethylenediaminetetraacetic acid

F-actin Filamentous actin

FBS Fetal bovine serum

FGF Fibroblast growth factor

FLN(A) Filamin (A)

FOV Field of view

G-actin Globular actin

GAP GTP-ase activating protein

GDI Guanine nucleotide dissociation inhibitor

GDP Guanosine diphosphate

GEF Guanine nucleotide exchange factor

GnRH Gonadotropin releasing hormone

GTP Guanosine triphosphate

HCL Hydrochloric acid

IG Immunoglobin

IGF1 Insulin-like growth factor 1

IL Interleukin

JNK c-jun N-terminal kinase

KO Knockout

M-CSF Macrophage colony stimulating factor

MAR Mineral Apposition Rate

Micro-CT Micro-Computed Tomography

MS Mineralized surface

MITF Micro-ophthalmia-associated transcription factor

x

MMP Matrix metalloproteinase

NFATc1 Nuclear factor of activated T cell, cytoplasmic 1

NFκB Nuclear Factor kappa B

OCG Osteoclastogenesis

Oc.N Osteoclast number

Oc.S Osteoclast surface

OPG Osteoprotegerin

OS Osteoid surface

O.Th Osteoid thickness

OV Osteoid volume

OVX Ovariectomized

PAK p21-activated kinase

PBS Phosphate-buffered saline

PCR Polymerase chain reaction

PFA Paraformaldehyde

PLC Phospholipase C

PMMA Poly(methyl methacrylate)

PTH Parathyroid hormone

PVNH Periventricular nodular heterotopia

qRT-PCR Quantitative real-time polymerase chain reaction

RANK Receptor Activator of Nuclear Factor kappa B

RANKL Receptor Activator of Nuclear Factor kappa B ligand

RGD Arg-Gly-Asp

ROCK Rho kinase

ROS Reactive oxygen species

Rpit Resorption pit

SEM Scanning electron microscope

Tb.N Trabecular number

Tb.Sp Trabecular separation

Tb.Th Trabecular thickness

TGF Transforming growth factor

TNF-α Tumor necrosis factor-α

TRAF TNF receptor associated factor

TRAP Tartrate-resistant acid phosphatase

vBMD Volumetric bone mineral density

WASP Wiskott-Aldrich syndrome protein

WIP WASP interacting protein

WT Wild type

  1  

Chapter 1

Literature Review

1.1 Motivation and Rationale Osteoporosis is a bone lytic disease characterized by low bone mass and deterioration of bone tissue.

As a consequence, increased bone fragility and risk of fracture, particularly of the hip, spine, wrist

and shoulder are common. In addition to osteoporosis, rheumatoid arthritis and periodontal disease

are high prevalence bone lytic diseases in which pathological bone loss is a major component of

disease progression [1]. These diseases are associated with considerable morbidity and place heavy

economic burdens on the healthcare system as a result of fractures, joint destruction and tooth loss

respectively. In fact, research shows that fractures from osteoporosis are more common than heart

attacks, strokes and breast cancer combined. At least one in three women and one in five men will

suffer from an osteoporotic fracture during their lifetime. The overall yearly cost to the Canadian

healthcare system of treating osteoporosis and the fractures it causes was over $2.3 billion as of 2010.

This cost includes acute care costs, outpatient care, prescription drugs and indirect costs. This cost

rises to $3.9 billion if a proportion of Canadians are assumed to be living in long-term care facilities

because of osteoporosis [2].

The reduced quality of life for those with osteoporosis is enormous. Osteoporosis can result in

disfigurement, lowered self-esteem, reduction or loss of mobility, and decreased independence.

Currently, bisphosphonates are the treatment of choice for regulating bone resorption in the

aforementioned diseases by targeting bone-resorbing cells (osteoclasts). Unfortunately, significant

side effects following bisphosphonate use include poor bone healing and osteonecrosis of alveolar

bone following fractures and surgery [3]. The observation that bisphosphonates may cause necrosis

of the jaw whereas long bones appear unaffected by this drug, may suggest that osteoclasts from the

jaw may respond differently to bisphosphonates and that this type of bone may thus harbor its own

type of osteoclasts. Such an observation requires further investigation into the role of osteoclasts in

different skeletal sites.

  2  

Bisphosphonates are proposed to target the post-translational modification of small GTPases to affect

osteoclast formation (OCG) [4]. OCG involves multiple fusions of pre-osteoclasts. This unique

process requires a dynamic actin cytoskeleton that is regulated by small GTPases. Fusion events are

exclusive to osteoclasts in the bone marrow and skeletal environments [5]. Thus, osteoclast fusion

would be an ideal target for pharmacotherapy for the aforementioned diseases.

Filamin A (FLNA) is a ubiquitous actin binding protein, first identified as an actin filament cross-

linking protein that regulates the stability of the cortical actin cytoskeleton of motile cells [6, 7].

More recent studies reveal that in addition to its role as a physical filament cross-linking protein,

FLNA also has the potential to function as a signal transduction integrator of small GTPases

signaling to the actin cytoskeleton [8]. Surprisingly, little is known about the role of the FLNA and

the actin cytoskeleton during OCG. For rational design of new drugs to manage pathological bone

loss, the regulatory mechanisms of OCG must be better characterized.

As a result of OCG, mature multinucleated osteoclasts are able to exert their bone resorptive activity

to remove old bone. Rpits produced by mature osteoclasts are most often measured in two-

dimensions via surface area analysis. Unfortunately, surface area and nuclei number analyses are

insufficient in determining resorptive activity in osteoclasts. A 3-D Rpit method has been developed

whereby osteoclast function is more accurately measured. The novel method proves that osteoclast

function is better understood through three-dimensional volumetric measurements of their Rpits.

  3  

1.2 Bone Biology Bone is a rigid organ that constitutes part of the endoskeleton of vertebrates. The endoskeleton serves

many physiological purposes including mineral storage, organ support and protection and production

of red and white blood cells [9]. Bone is a dense connective tissue that differs from other connective

tissues in the body by its rigidity and hardness [9, 10]. Biomechanically, bone is a dynamic,

complex, porous, composite structure that possesses the ability to adapt to changes in its

physiological and mechanical environment [11, 12].

Bone is a composite material consisting of an inorganic mineral phase embedded within an organic

matrix. The inorganic material constitutes 65% of bone and consists of impure hydroxyapatite

Ca10(PO4)6(OH)2. Hydroxyapatite forms crystal structures that may incorporate other constituents

including carbonate, citrate, magnesium, fluoride and strontium. Hydroxyapatite is localized in the

osseous tissue that provides rigidity and strength to the collagen framework that makes up the organic

matrix [9, 13]. The organic matrix constitutes 35% of bone and is comprised of 90% collagen and

10% non-collagenous proteins. Non-collagenous proteins include osteocalcin, osteonectin,

osteopontin and bone sialoprotein. These proteins are postulated to play an important role in size,

orientation and fixation of the hydroxyapatite crystals to the collagen framework. Non-collagenous

proteins are commonly used as bone remodeling markers in blood serum and urine. Type I collagen

is predominant in the organic matrix along with trace amounts of type III and type IV collagen. Type

I collagen is synthesized intracellularly as tropocollagen and then exported forming fibrils. Collagen

fibrils are responsible for providing the structural framework in which hydroxyapatite is inserted [9,

13, 14].

  4  

1.3 Bone Structure

i. Macrostructure Bone is divided morphologically into cortical and trabecular bone. Cortical or compact bone tissue

comprises the outer hard layer of bones. Cortical bone accounts for approximately 80% of the total

bone mass of the adult skeleton and is a non-porous structure that gives bones their smooth, white

and solid appearance. Cortical bone is found predominantly in the diaphysis of long bones and in the

outer shell of cuboid or vertebral bone. Cortical bone functions to support the body, protect organs,

provide levers for movement and store and release chemical elements [15].

The primary functional unit of cortical bone is the osteon. Osteons are organized into Haversian

systems whereby bone forming and resorbing cells are arranged into cylindrical units surrounding a

central vascular canal. The Haversian canals lie parallel to the long axis or anterior-posterior (A-P)

length of bone and form concentric layers of bone during remodeling. Bone structure and

remodeling vary between different animals. Rodents, in particular do not process nor remodel their

cortical bone via Haversian systems. Therefore, animal models may not directly translate into the

human situation. Compared to trabecular bone, the rate of bone remodeling in cortical bone is very

slow [15, 16].

Trabecular bone is the spongy and porous bone found predominantly in the epiphyses and metaphysis

of long bones and the cores of cuboid bone, including the vertebral body. Trabecular bone

constitutes the remaining 20% of the human skeleton and is organized into a 3-D network of

branched out and connected struts. The organization of these struts form along the lines of stress,

demonstrating the adaptive nature of bone to applied loads. Compared to cortical bone, trabecular

bone has a higher surface area to mass ratio resulting in higher turnover rates. Hence, bone turnover

can improve bone properties by increasing trabecular organization. In addition, trabecular bone is

highly vascular, containing red bone marrow where hematopoiesis occurs [12, 15, 16].

ii. Microstructure

Cortical and trabecular bone are comprised of woven and lamellar bone. Woven bone is immature,

newly deposited bone, characterized by haphazard organization of collagen fibers. Woven bone is

  5  

mechanically isotropic; the orientation of the applied force has no bearing on the mechanical

characteristics of the bone. Woven bone serves as a provisional material that is formed quickly

during periods of rapid growth or extensive repair. During the process of bone turnover, woven bone

is eventually resorbed by lamellar bone. Lamellar bone is mature bone that is highly organized and

arranged in concentric sheets termed lamellae. The parallel alignment of collagen fibers in lamellar

bone provides resistance to mechanical stresses in an anisotropic manner; the mechanical behavior

can differ depending on the orientation of the applied force. Furthermore, the greatest strength of

lamellar bone is greatest parallel to the longitudinal axis of collagen fibers [15, 16].

iii. Alveolar Bone

Alveolar bone is specialized bone that forms and protects the sockets that accommodate teeth.

Alveolar bone is especially thick and dense compared to other bone types so that it can provide

adequate support for teeth, along with attachment points for muscles involved in the jaw and gums.

The gums are attached to the ‘alveolar process’, which includes sockets that hold the roots and lower

part of the teeth with each tooth separated from the next by interdental septum. The alveolar process

allows for blood vessels to supply blood to the teeth. Damage to alveolar bone can have serious

consequences, including the risk of loss of teeth and septicemia if the damage is caused by infection

[17].

Alveolar bone is comprised of alveolar bone proper and supporting bone and develops from the

dental follicle. In humans, this type of bone is found in the mandible, or lower part of the jaw, along

with the maxilla, the upper part of the jaw. The ectomesenchymal cells of the dental follicle

differentiate into osteoblasts and lay down the osteoid matrix. Some of these osteoblasts become

embedded in the matrix to become osteocytes. The mandible and maxilla are formed in the second

month of fetal life and form a groove that is opened toward the surface of the oral cavity. Bony septa

grow gradually as tooth germs start to develop [17, 18].

Alveolar bone helps absorb the forces placed upon the tooth by disseminating the force to underlying

tissues. Alveolar bone is composed of the cortical plate, which is the outermost part of the bone. It is

covered in periosteum, varies in thickness and is located on the outside wall of the maxilla and

mandible [18]. The cortical plate consists of Haversian systems (osteons) and interstitial lamellae

and tends to be thicker in the mandible than the maxilla [9, 10]. The alveolar bone proper or lamina

  6  

is an inner, heavily perforated lamellar bone, which forms the alveolar wall. The lamina also

contains osteons similar to the cortical plate but is distinguished by the presence of bundle bone.

Spongiosa lies between the two bony plates and between the lamina cribiformis of adjacent roots or

teeth. The spongiosa consists of delicate trabeculae, between which are fatty marrow filled-spaces.

Periodontal disease affects one or more of the periodontal tissues including alveolar bone, the

periodontal ligament, the cementum and the gingiva [9, 10, 17].

The mandible is the largest and strongest bone of the face, serves for the reception of the lower teeth.

It consists of a curved, horizontal portion, the body, and two perpendicular portions, the rami, which

unite with the ends of the body nearly at right angles. Alveolar bone develops via endochondral

ossification and the mandible develops via intramembranous ossification [9, 10].

iv. Calvarial Bone

The calvarium is the upper part of the neurocranium and covers the cranial cavity containing the

brain. The calvarium is the portion of the skull that includes the braincase but excludes the lower jaw

and facial portion. Calvaria are made up of the superior portions of the frontal bone, occipital bone,

and parietal bones. Most bones of the calvarium consist of internal and external layers of compact

bone, separated by diploe. Diploe is cancellous bone containing red bone marrow and canals, which

contain diploic veins. The internal layer of compact bone is thinner than the outer layer and in some

areas there is only a thin plate of compact bone with no diploe. In the fetus, the calvarium is formed

via intramembranous ossification [21-24].

v. Long Bone

The femur or thighbone is the strongest, longest and most proximal bone of the leg [9, 10, 12]. Long

bones are hard and dense and provide strength, structure and mobility. The femur is comprised of a

diaphysis, the shaft or body, and two epiphysis or extremities that articulate with adjacent bones in

the hip and knee. The femoral body is longer than it is wide with growth plates at both ends

(epiphysis), which has a hard outer surface of compact bone and a spongy inner cancellous bone

containing bone marrow. Both ends of the bone are covered in hyaline cartilage to protect the bone

and aid in shock absorption. The diaphysis is the elongated hollow central portion of the bone

  7  

located between the metaphysis, is made up of compact tissue and encloses the medullary cavity.

The metaphysis is the part of bone between the epiphysis and diaphysis containing the connecting

cartilage enabling the bone to grow. The longitudinal growth of long bones is a result of

endochondral ossification at the epiphyseal plate [9, 10, 12, 13, 15].

In addition to its structural role, long bones also play a crucial role in the production of bone marrow.

Femur bones contain both yellow and red bone marrow, which is vital for the production of blood

cells. Finally, long bones regulate inorganic salt and storage of calcium, phosphate, sodium and

potassium [14, 15].

 

  8  

1.4 Bone cells Bone is composed of three different cell types: osteoblasts, osteoclasts and osteocytes. Osteoblasts of

mesenchymal origin deposit the calcified bone matrix, while osteoclasts of haematopoietic origin are

responsible for bone resorptive activity. Osteocytes are osteoblasts that eventually become

embedded within the cavities of the bone matrix. The aforementioned cell types are responsible for

bone remodelling and regulation of the bony skeleton [25, 26].

Osteoblasts are mono-nucleated bone forming cells that are specialized, terminally differentiated

products of mesenchymal stem cells. Osteoblasts synthesize very dense, cross-linked collagen as

well as several additional specialized proteins including osteocalcin and osteopontin, which comprise

the organic matrix of bone. Calcium and phosphate mineral are then deposited via osteoblasts in a

highly regulated manner into the organic matrix forming the mineralized matrix.

Osteoclasts are multinucleated bone-resorbing cells that arise from the monocyte-macrophage

lineage. Bone resorption is critical in the maintenance and repair of compact skeletal bones in the

mammalian skeleton. Osteoclasts are located on bone surfaces in Howship’s lacunae also known as

Rpits. The lacunae are the end products of the dissolution of bone mineral via tartrate-resistant acid

phosphatase and cathepsin K secretion.

Osteocytes are mature bone cells originating from osteoblasts that migrate into and become trapped

in the bone matrix. Osteocytes have many processes that reach out to connect and communicate with

osteoblasts and other osteocytes. Osteocytes maintain the matrix and are involved in calcium

homeostasis. They are postulated to act as mechano-sensory receptors, regulating the bone's

response to stress and mechanical load [25-27].

  9  

1.5 Bone remodeling Bone remodeling is the process whereby bone is renewed to maintain bone strength and mineral

homeostasis. Remodeling involves continuous removal of discrete packets of old bone, replacement

of these packets with newly synthesized proteinaceous matrix, and subsequent mineralization of the

matrix. The remodeling process resorbs old bone and forms new bone to prevent accumulation of

bone microdamage. Skeletal remodeling requires spatial and temporal orchestration of osteoclast and

osteoblast function. In vivo, this harmonious process is achieved through the formation of transient

structures called basic multi-cellular units (BMU’s) [26, 27].

The BMU is comprised of a central vascular capillary and nerve supply fronted by a cluster of bone

resorbing osteoclasts followed by several matrix-secreting osteoblasts. As the BMU’s travel across

the bone surface, osteoclasts excavate and degrade old bone. Osteoblasts then migrate into the

freshly excavated areas and secrete new matrix. Thus, bone resorption and formation are not separate

entities, but rather highly intertwined and coupled processes. Bone remodeling involves four distinct

phases including the activation, resorption, reversal and formation phases [26-28].

The activation phase is characterized by activating quiescent osteoblasts also known as lining cells.

Activation is accomplished through different inputs, such as a micro-fracture, an alteration of

mechanical loading sensed by the osteocytes or some factors released in the bone microenvironment,

including insulin growth factor-I (IGF-I), tumor necrosis factor-α (TNF-α), parathyroid hormone

(PTH) and interleukin-6 (IL-6). As a consequence, lining cells, increase their own surface expression

of Receptor Activator of Nuclear Factor κB Ligand (RANKL), which in turn interacts with its

receptor Receptor Activator of Nuclear Factor κB (RANK), expressed by pre-osteoclasts.

RANKL/RANK interaction triggers pre-osteoclast fusion and differentiation toward multinucleated

osteoclasts [28].

In the resorption phase, differentiated osteoclasts polarize, adhere to the bone surface and begin to

dissolve bone. Bone dissolution requires acidification of the bone matrix to dissolve the inorganic

component, and release of lysosomal enzymes, such as cathepsin K (CTSK), and of matrix

metalloproteinase 9 (MMP9), to degrade the organic component of bone. In the final stage,

osteoclasts undergo apoptosis in order to avoid excessive bone resorption [26-28].

  10  

The reversal phase involves removal of debris produced during matrix degradation by macrophage-

like cells. In the formation phase, several stored growth factors are released, including bone

morphogenetic proteins (BMPs), fibroblast growth factors (FGFs) and transforming growth factor β

(TGF-β), which are likely responsible for the recruitment of the osteoblasts in the reabsorbed area.

Once recruited, osteoblasts produce the new un-calcified bone matrix (osteoid). Mineralization is

then promoted completing the bone remodeling process [26-28].

An imbalance between resorption and formation can lead to pathological conditions such as

osteoporosis and periodontal disease [29]. Osteoporosis is a common bone disease currently

affecting 1 in 4 females and 1 in 8 males over the age of 50. Osteoporosis is a consequence of an

increased ratio of resorption to formation respectively. Contrarily, osteopetrosis is characterized by

an increased ratio of formation to resorption, respectively. Impaired remodeling in osteopetrosis

results in dense, brittle bones. Both skeletal diseases illustrate through opposite mechanisms, the

importance of homeostatic control of bone remodeling [28-30].

  11  

1.6 Bone Quality Bone quality encompasses the geometric and tissue material factors that contribute to fracture risk.

[31]. Bone is a unique tissue in that it is able to carry loads, resist deformation and yet be flexible in

order to change shape and absorb energy from varying forces exerted on them. For example, bone

undergoes shortening and widening during compression and lengthening and narrowing during

tension. A common misconception in evaluating bone quality is the use of bone mineral density to

assess fracture risk.

Currently, the most common diagnostic tool used to assess bone mineral density (BMD) is Dual

Energy X-ray Absorptiometry (DEXA) [30-34]. Unfortunately, BMD represents only one of several

parameters that enable bone to resist fracture. BMD measured by DEXA taken in a clinical setting is

a 2-D observation that does not provide a comprehensive analysis of both material and structural

properties that contribute to the intrinsic strength of bone [32, 33]. Intrinsic and extrinsic bone

strength is influenced by bone remodeling which is dependent upon structural (bone geometry and

connectivity) and material properties (mineralization) of bone respectively. The complexity of bone

requires that no single property predict overall bone quality.

i. Material Properties Currently, the major noninvasive method available for the early diagnosis of osteoporosis is the

measurement of areal BMD (aBMD) using DEXA. DEXA measures the amount of bone mineral

projected in a given area (aBMD). Research shows, however, that aBMD does appear to be a good

predictor of bone strength [35-37]. Although aBMD has a high level of prediction of bone strength,

it is still insufficient in assessing overall bone quality.

aBMD is calculated based on a scanned area where the length and width of the bone is known but not

its depth or volume. aBMD is not able to detect surface-specific changes in aging bone, i.e. bone

resorption and formation on the endocortical, intracortical, and trabecular surfaces. For example,

DEXA incorrectly suggests that bone density increases during growth, which is mainly caused by

increased bone size and does not reflect the mineral changes occurring within the periosteal envelope.

DEXA is also unable to distinguish between sex-specific changes in bone. Male aBMD is generally

higher than female aBMD due to a greater periosteal apposition rate and larger bone size. On the

  12  

other hand, volumetric BMD (vBMD) of the whole bone remains constant or increases slightly

during growth and does not differ between sexes [38]. Micro-computed tomography (Micro-CT) is a

more accurate tool to measure BMD as it assesses vBMD via 3-D measurements.

ii. Bone Mechanical Properties Bone fragility refers to bone’s susceptibility to fracture. Fractures occur when an applied load

exceeds the intrinsic and extrinsic strength of bone and/or when bone is unable to absorb the energy

applied by the exerting forces. Evaluation of biomechanical properties including ultimate load,

failure displacement and energy to fail, is crucial in assessing bone quality as these parameters are

directly related to fracture risk [39, 40]. Stiffness, which is not a direct measure of bone fragility, as

well as the aforementioned biomechanical properties, can be evaluated through mechanical testing

where bones are loaded until failure. Since bone is considered an anisotropic material, meaning that

its mechanical properties vary according to the direction of the load, several types of mechanical tests

are performed on various bone sites in order to assess bone quality. The data obtained from

biomechanical testing is used to construct a load-displacement curve from which the mechanical

properties are derived.

Three-point bending of femora is used to test the mechanical properties of cortical bone undergoing

bending until failure. Three-point bending is considered a combined failure mode test since bone is

tested in both tension and compression. Vertebral compression testing is performed on the lumbar

vertebrae and evaluates the mechanical properties of trabecular bone undergoing compression. From

these mechanical tests, important structural and material properties are evaluated in order to

thoroughly assess skeletal integrity and susceptibility to fracture [39]. The load-displacement curve

described previously can be normalized to account for structural changes in bone. The result is a

stress-strain curve from which material properties such as ultimate stress, strain and toughness can be

obtained.

iii. Structural Properties Bone structural properties are measurements of the extrinsic mechanical characteristics of bone.

Structural as well as bone tissue properties are dependent upon the size and shape of bone [41]. An

increase in effectively distributed bone mass is most beneficial for bone strength. Load-displacement

curves obtained through mechanical testing are used to assess bone’s structural properties. Structural

  13  

properties include strength, work to failure, ductility and stiffness [31]. Extrinsic bone strength

(ultimate load) refers to the height of the curve and work to failure (energy to failure) refers to the

area below the curve. Ductility is estimated from the width of the curve (failure displacement) and

stiffness is the slope of the elastic region of the curve and represents the rigidity of bone [31, 41].

Bone geometrical parameters including trabecular architecture and cortical porosity are also used to

evaluate the extrinsic properties of bone. Micro-CT is used to image excised vertebrae and femora in

order to visualize each specimen in 3-D. The structural properties of cortical bone obtained through

micro-CT include cortical thickness, cross-sectional area and moment of inertia. Typically, larger

bones with thicker cross-sectional areas absorb greater loads before fracture [31]. Structural

properties of trabecular bone relate to trabecular architecture: the number, thickness and separation of

individual trabeculae, as well as trabecular connectivity. Predictably, a greater number of thicker

trabeculae with less separation and more connectivity are associated with greater trabecular strength

[42-45].

iv. Material Properties Independent of shape and size, bone’s ability to resist fracture is also reliant on its intrinsic material

properties. Load-displacement curves are converted to stress-strain curves using engineering

formulae and specimen-specific geometrical information obtained through Micro-CT analysis.

Stress-strain curves are used to assess the intrinsic mechanical properties of bone including strength

(ultimate stress), work to failure (toughness), ductility (failure strain), and stiffness (Young’s

modulus) [31, 39].

As mentioned previously, bone is comprised of organic and mineral phases, both of which contribute

to its mechanical properties. The mineral phase provides bone with strength and stiffness.

Consequently, hyper-mineralized levels can lead to reduced bone ductility, or brittleness. The organic

phase of bone plays a greater role in toughness and ductility [46]. Changes in the quality of the

organic phase can also have profound effects on bone fragility. Reduced collagen cross-linking and

disordered arrangement of collagen fibers is correlated with increased brittleness and reduced

mechanical properties [40, 46-48]. Age-related changes to the collagen network are positively

correlated with age-related deterioration of bone. These changes include the formation of advanced

glycation end-products (AGEs), which are non-enzymatic cross-links in collagen. The accumulation

  14  

of AGEs over time result in stiffening of the collagen network, leading to increased skeletal fragility

and fracture risk [11, 48, 49, 50].

v. Bone Remodeling Skeletal diseases are often the result of an imbalance in bone remodeling leading to increased skeletal

fragility [9]. Bone resorption is assessed through the quantification of tartrate-resistant acid

phosphatase (TRAP) positive osteoclasts from thin sections of bone [51]. Bone formation is

evaluated via dynamic histomorphometry, whereby the distance between fluorescent calcein green

mineralization markers are measured [52].

  15  

1.7 2-D vs. 3-D assessment of osteoclastic bone resorption Structurally and mechanically, bone deficit is determined through Rpits formed during resorption.

Rpits are an essential element in bone remodeling and can predict the effect of bone disease and

treatment. Consequently, quantifying Rpits and their effect on bone strength is relatively limited.

[53]

Most studies extrapolate 3-D characteristics of Rpits from 2-D features measured on histological

sections using stereological formulas. This extrapolation, however, is not without flaws since it

assumes unbiased random sampling and isotropy, which are not fulfilled in bone [54]. 2-D widths

are transformed into 3-D thicknesses by using the parallel plate model. Consequently, the

distribution is corrected for missing measurements and intra-observer, inter-observer, inter-method

and sample variations have to be taken into account [55-60].

Histomorphometric analyses use different staining methods that can highlight certain features.

Toluidine blue is used to identify cavities under polarized light by looking at the presence of cut off

collagen fibers (disruption of the lamellar system) at the edge of the Rpit [58, 59, 61]. Polarized light

allows for visualization collagen lamellae orientation along the mineralized bone surface.

Unfortunately, the identification of scalloped surfaces can be subjective [62]. In addition, TRAP

staining can be used to mark active osteoclasts and thus “active” cavities [63]. Von Kossa/van Gieson

staining discriminates osteoid from mineralized bone [58].

The use of 2-D histomorphometry has limitations: it cannot discriminate between an increase in the

number of remodeling events from an increase in the size of each individual event [64] nor can it

accurately measure the full volumetric extent of a Rpit [65]. The development of 3-D methods to

assess BMU’s is essential for studying how alterations in its morphology occur with disease and

treatment [66]. Scanning Electron Microscope (SEM) images give a mediocre indication of the 3-D

nature of Rpits [67]. Currently, no clinical imaging devices are able to detect Rpits because of their

small size compared to image resolution. It has been shown that high-resolution images (at least 1.4

µm or better) are required to consistently identify and measure individual Rpits [68].

The introduction of 3-D imaging techniques including laser scanning confocal microscopy has

opened the door to quantifying Rpit dimensions in an unbiased manner in 3-D space. These data can

  16  

be included in high-resolution computational models and in parametric descriptions of bone, thereby

improving our understanding of their effect on bone competence. Further exploration of this area of

research will disclose relevant information on the mechanical consequences of metabolic bone

diseases such as osteoporosis and can aid in the development of relevant pharmacological and

physical treatments.

  17  

1.8 Osteoporosis Osteoporosis is a metabolic bone lytic disease characterized by low bone mineral density, weakened

trabecular microarchitecture and altered material properties. Loss of bone mass in osteoporotic

subjects is associated with increased risk of fractures in the hip, vertebral bodies and forearms.

Osteoporotic fractures can immobilize patients for prolonged periods and increase the risk of

mortality in the elderly population by up to 20% following the fracture. Osteoporosis related deaths

are a result of chronic illnesses that lead to falls/fractures as well as acute events such as infections

and post-operative complications [29]. Recently, the percentage of the worldwide population over 65

years of age has increased dramatically and is projected to steadily increase over the next 20 years.

As a consequence, by the year 2030, the number of hip fractures is expected to quadruple.

Worldwide, osteoporosis causes more than 8.9 million fractures annually, resulting in an osteoporotic

fracture every 3 seconds [29, 30, 32].

Osteoporosis affects approximately 1.4 million Canadians, with the majority of this population being

postmenopausal women and the elderly. Since age is a major risk factor for osteoporosis,

osteoporotic related disorders are a continual public health concern. In Europe, disability due to

osteoporosis is greater than that caused by cancers (with the exception of lung cancer) and is

comparable or greater than that due to a variety of chronic non-communicable diseases, such as

rheumatoid arthritis, asthma and high blood pressure-related heart disease. Recent estimates on the

combined annual costs of all osteoporotic fractures have been estimated to be over 50 billion dollars

in the US and Europe and over 130 billion dollars worldwide [29, 30, 32-34].

At the cellular level, osteoporotic bone loss occurs due to an imbalance in bone remodeling. This

imbalance can be caused by a number of factors including age, sex steroid deficiencies, excess

glucocorticoids, reduced mechanical loading, multiple myelomatosis and hyperparathyroidism.

Osteoporosis exists in both a primary and secondary form. Post-menopausal (type 1) and age-related

(type 2) osteoporosis are the most common types of primary osteoporosis. Secondary osteoporosis

(type 3) is a consequence of other diseases or conditions that predispose the bone to increased

deterioration. With age, bone formation during each remodeling cycle decreases and the rate of bone

loss begins to overtake the rate of bone formation. Consequently, this age-induced bone loss is a

  18  

result of a shift in differentiation of multipotent mesenchymal stem cells toward adipocytes instead of

osteoblasts [29, 30].

Type 1 or post-menopausal osteoporosis results from decreased ovarian function leading to a

corresponding depletion in estrogen levels. The consequence is an imbalance in bone remodeling

through the expression of cytokines that induce osteoclast differentiation and activation (RANKL,

TNF-α, IL-6, macrophage colony stimulating factor (M-CSF) and prostaglandin E)) while

suppressing the expression of OCG inhibiting agent osteoprotogerin (OPG). Post-menopausal

osteoporosis primarily affects women between the ages of 50 to 75. The dramatic increase in the rate

of bone turnover in conjunction with increased bone resorption is responsible for the acceleration of

post-menopausal bone loss. Over time, the accelerated phase of type 1 osteoporosis merges

asymptomatically with the late phase of slow bone loss, which continues indefinitely with age (type 2

osteoporosis) [30, 71, 72].

Primary type 2 osteoporosis or senile osteoporosis occurs after age 75 and is observed in both

females and males at a ratio of 2:1. Type 2 osteoporosis is associated with decreased bone formation

along with decreased ability of the kidney to produce 1,25(OH)2D3. The vitamin D deficiency results

in decreased calcium absorption, which increases the PTH level and therefore bone resorption. In

type 2 osteoporosis, cortical and trabecular bone is lost, primarily leading to increased risk of hip,

long bone, and vertebral fractures [29, 30, 71, 72].

Type 3 or secondary osteoporosis occurs equally in men and women and at any age. In men, most

cases are due to disease or drug therapy. In 30-45% of affected individuals no cause can be identified

[29]. Secondary osteoporosis accounts for about 40% of the total number of osteoporotic fractures

seen by physicians. Type 3 osteoporosis is associated with a variety of conditions, including

hormonal imbalances (eg, Cushing's syndrome); cancer (multiple myeloma); gastrointestinal

disorders (inflammatory bowel disease causing malabsorption); drug use (corticosteroids, cancer

chemotherapy, anticonvulsants, heparin, barbiturates, valporic acid, gonadotropin-releasing hormone

(GnRH), aluminum-containing antacids); chronic renal failure; hyperthyroidism; hypogonadism in

men; immobilization; osteogenesis imperfecta and related disorders; inflammatory arthritis

(particularly rheumatoid arthritis); and poor nutrition (malnutrition due to eating disorders) [30, 71,

72].

  19  

i. The aged mouse model of osteoporosis

Animal models are necessary to predict fracture risk in humans for the development of novel

pharmacotherapeutics. Since the mouse genome can be easily manipulated and their skeletons are

similar to that of humans, mice have become increasingly popular for investigating the

pathophysiology of osteoporosis. The ovariectomized (OVX) aged mouse is a well-accepted model

for post-menopausal osteoporosis as it mimics many of the clinical features of the estrogen-depleted

situation in humans. OVX mice display increased bone turnover favoring bone resorption. Like

humans, mice exhibit loss of trabecular bone, thinning of cortical bone, and increased cortical

porosity with advancing age [73].

ii. Osteopetrosis

Osteopetrosis is a clinical syndrome characterized by the failure of osteoclasts to resorb bone. As a

consequence, bone modeling and remodeling are impaired. The defect in bone turnover results in

skeletal fragility despite increased bone mass, hematopoietic insufficiency, disturbed tooth eruption,

nerve entrapment syndromes, and growth impairment [74]. Osteopetrosis is characterized by

marbled-like bone with increased density and radiographic loss of distinction between the cortex and

marrow cavity [75, 76]. More severe cases of osteopetrosis display marrow-filled cavities leading to

smaller endochondral bone. As a result, extra-medullary hematopoiesis and hepatosplenomegaly are

likely to ensue due to inadequate hematopoiesis. Patients with less severe cases of osteopetrosis may

experience normal life expectancy, but have brittle bones that are more susceptible to fracture [74,

75].

Impaired osteoclastic bone resorption, specifically any mutation that arrests OCG or inhibits

osteoclast function, can result in osteopetrosis [76]. For instance, transgenic mice over-expressing

the OCG inhibitor OPG lack osteoclasts and are severely osteopetrotic [77]. Similarly, loss of

function mutations in genes involved in the osteoclast specific RANKL-RANK signaling axis such as

c-fos, NF-kB and TRAF6 can also cause osteopetrosis [75, 78]. Another cause of osteopetrosis is

mutations in genes involved in mature osteoclast function including c-src, CTSK, TRAP, the

vitronectin integrin receptor avb3, a3 subunit of vATPase, carbonic anhydrase 2 and the chloride

channel CLC-7 [74-76]. The osteopetrotic phenotype observed in the aforementioned transgenic

mouse models was specifically caused by the inability of osteoclasts to polarize, spread and degrade

  20  

the bone matrix and not by a loss of function in osteoclast precursor populations [75]. The overall

incidence of osteopetrosis is estimated to be 1/100,000-500,000 [76].

 

  21  

1.9 Osteoclasts and OCG Osteoclasts are mature, multinucleated bone resorbing cells with well-developed Golgi apparatus’,

abundant mitochondria, lysosomes, vacuoles and ribosomes [10, 27, 28]. Osteoclasts possess plasma

membrane projections that comprise the ruffled border. The sealing zone is an actin-rich area that

surrounds the ruffled border. Together, these two entities comprise the resorptive machinery of the

osteoclast. The sealing zone mediates the attachment of the osteoclast to the bone matrix in order to

quarantine the microenvironment beneath the osteoclast that will eventually be degraded [27, 28, 78-

83]. Degradation of the mineral and organic phases of bone are accomplished through acidification

and proteolytic degradation respectively. The products of degradation are endocytosed into the

osteoclast, transocytosed across the cell and exocytosed through the basolateral membrane facing the

bone marrow.

OCG is the term used to describe the intricate formation of mature, multinucleated osteoclasts from

hematopoietic osteoclast precursors. Proliferation, fusion and differentiation of precursor cells are

achieved through regulation of the cytokines M-CSF and RANKL. M-CSF promotes the survival

and proliferation of osteoclast precursors thereby increasing the size of the osteoclast committed

precursor pool [84]. Literature reveals that mice lacking functional M-CSF have few mature

osteoclasts and are severely osteopetrotic. Through signaling cascades, M-CSF promotes cell

spreading, migration and actin cytoskeletal rearrangements in mature osteoclasts. RANKL is a

member of the TNF-α family of proteins [84].

In vivo, OCG is supported by direct cell-to-cell contact between RANKL expressed on the surface of

osteoblasts and bone marrow stromal cells, and RANK, its receptor on the surface of osteoclast

precursors. Similar to M-CSF, RANK and RANKL deficient mice with normal macrophage numbers

failed to generate osteoclasts and were severely osteopetrotic [77]. Conversely, mice over-expressing

RANKL displayed hyper-resorptive osteoclasts and suffered from severe osteoporosis. An additional

molecule that negatively regulates OCG is OPG. OPG is produced by osteoblasts and acts as a decoy

receptor for RANKL, thereby moderating net RANKL activity [77, 85]. Simonet et al. demonstrated

that transgenic mice over-expressing OPG exhibited non-lethal osteopetrosis and concurrent defects

in osteoclast differentiation. Thus, modulation of OCG in vivo is determined by the ratio of RANKL

and OPG [85].

  22  

i. Actin cytoskeleton and osteoclasts The actin cytoskeleton is a dynamic structure that provides a scaffold for cell movement and spatial

organization. The actin cytoskeleton is capable of altering the mechanical properties of non-muscle

cells through reverse polymerization of globular monomeric G-actin into filamentous F-actin. Actin

enables mature osteoclasts to adhere to their substratum, migrate, polarize during resorption and

transmigrate through cell layers. Actin is especially important for podosome formation in osteoclasts.

Podosomes are primary adhesive structures, which make up the most prominent component of the

actin cytoskeleton in monocyte-derived cells [86]. Osteoclast adhesion to the substratum is regulated

through actin polymerization and depolymerization within podosomes [87].

ii. Podosomes and actin rings Podosomes are important for cell adhesion and motility in cells from the myeloid lineage including

osteoclasts, macrophages, and dendritic cells. Podosomes are highly dynamic dot-like structures,

composed primarily of F-actin. The F-actin core contains a high concentration of actin-regulating

proteins including cortactin, Wiskott-Aldrich syndrome protein (WASP), WASP interacting protein

(WIP), Arp2/3, gelsolin, and CD44 [15]. The F-actin core is surrounded by scaffolding proteins,

including vinculin, paxillin, talin, kinases (c-Src, Pyk2), integrins (e.g. αvβ3), and Rho GTPases

(Rho, Rac, Cdc42). Actin rings, also known as the sealing zone, mediate strong anchorage of

osteoclasts to bone. Sealing zones are dependent on osteoclast polarization and form a sealed off

microenvironment within which bone resorption occurs. The actin ring measures 4 μm in width and

height, and has an inner and outer lining of vinculin [87]. The actin ring is generated and stabilized

by the dense interconnection of adjacent podosomes rich in F-actin, and acetylated microtubules,

respectively [88]. Furthermore, the actin ring structure is only formed when osteoclasts are attached

to bone or a mineralized matrix (e.g. dentin and hydroxyapatite-embedded collagen).

  23  

1.10 Rho (Rac) GTPases Cytoskeletal remodeling determines cellular structure and shape. The cytoskeleton is composed of

an array of F-actin and its reorganization is crucial for regulating cell shape, motility and adhesion.

Rho GTPases act as intracellular second messengers to transmit extracellular stimuli that are crucial

for actin cytoskeletal remodeling. Rho GTPases make up a distinct family within the superfamily of

Ras-related small GTPases and are present in all eukaryotic cells [89]. These “molecular switches”

cycle between an active GTP-bound state, and an inactive GDP-bound state. The

activation/inactivation of Rho GTPases are mediated via their regulatory proteins: guanine nucleotide

exchange factors (GEFs) that catalyze the exchange of GDP for GTP. GTPase activating proteins

(GAPs) stimulate the intrinsic activity of Rho GTPases, converting GTP to GDP, thereby inactivating

the GTPase. Lastly, guanine nucleotide dissociation inhibitors (GDIs) sequester the GTPases to

prevent spontaneous activation [89]. GTP-bound Rho GTPases interact with downstream effectors to

ultimately affect cellular function.���

i. Rac, Cdc42, and Rho GTPases In mammalian cells, Rac, Cdc42 and Rho are the most extensively studied of the Rho GTPases. Rac

has 3 isoforms: Rac1, Rac2 and Rac3. Rac1 and Rac2 are present in osteoclasts and their precursors,

with Rac1 being expressed at higher levels [82, 89]. Generally, activation of Rac1 induces

lamellapodia formation, which is a broad sheet of cytoplasm extending from the leading edge in the

direction of cell movement. Cdc42 senses the extracellular environment, which induces the

extension of finger-like projections at the front of the moving cell (filopodia) [90,91]. Rho is

required for cell retraction by activating actin:myosin contractile machinery, and is localized at the

rear (uropod) of a migrating cell. In certain cells, Rho is also involved in the formation of stress

fibers and actin bundles that promote cell attachment through focal adhesions [92]. Activation of

Rho GTPases is regulated spatially and temporally in order to coordinate cell shape, adhesion,

motility, and in some cells, phagocytosis.

ii. Actin filament polymerization and cellular migration During migration, an extracellular stimulus (e.g. a chemoattractant) induces actin cytoskeleton

polarization, where actin polymerization at the front of the cells and actin:myosin contraction at the

  24  

rear provides the driving force for movement. Additionally, the formation of both lamellipodia and

filopodia at the leading edge require actin polymerization, which pushes the leading edge membrane

forward.

iii. Spatio-temporal regulation of Rac/Cdc42/Rho activation Recent studies focus on the crosstalk and antagonism between Rac and Rho during chemotaxis since

they are present at the leading and trailing edges of migrating cells respectively. Specifically, Rac1

locally inhibits Rho’s influence at the trailing edge of migrating cells [93]. FilGAP, a newly

identified GAP specific for Rac, is a downstream effector of ROCK, which mediates the inactivation

of Rac by Rho [94]. FilGAP binds FLNA to suppress and mediate the antagonism of Rac by Rho at

the leading edge of protruding cells. Furthermore, biosensors, which are used to study the

localization and activities of Rho GTPases in embryonic fibroblasts, have identified active RhoA,

Rac1 and Cdc42 at the leading edge of cells [95]. Specifically, RhoA activation corresponds with

edge advancement, whereas Cdc42 and Rac1 are activated 2 μm behind the edge with a delay of 40 s,

showing a spatial-temporal regulation of their antagonistic activities.���

iv. Regulation of osteoclast physiology by Rho GTPases In addition to regulating the osteoclast actin cytoskeleton, Rho GTPases are important in osteoclast

formation, function, and survival. Using adenovirus infection methods to introduce constitutively

active or dominant negative Rac1 into murine osteoclasts in vitro, Rac1 was shown to be involved in

mediating osteoclast survival downstream of the M-CSF receptor. Rac1 also plays a role in

osteoclast cell spreading, membrane ruffling and resorptive activity [96]. Using mice with

conditional Rac1, Rac2, and double null mutations, both Rac1 and Rac2 were found to be necessary

for normal osteoclast formation and differentiation. Rac1, however, is considered the key isoform

since the effects of Rac2 deletion show a milder phenotype. Rac1 was essential for pre-osteoclast

chemotaxis towards M-CSF, generation of actin free barbed ends for actin filament elongation after

M-CSF stimulation, RANKL-induced reactive oxygen species (ROS) generation, and cell spreading

[97]. Rac2 null mice, on the other hand, showed a slight increase in trabecular bone mass and

connectivity even though there were increased numbers of osteoclasts in vivo [98]. Rac2-/-

osteoclasts were defective in chemotaxis, resorptive activity and cell spreading. In addition, in the

absence of Rac2, there appeared to be an abnormal accumulation of actin and disruption of the actin

  25  

ring [99]. Consequently, the requirement for Rac2 in osteoclast biology is not universally

recognized. Lee et al. found that only Rac1 and not Rac2 was activated downstream of RANK

leading to activation of NFκB [99].

���

  26  

1.11 FLNA Discovery FLN was discovered accidentally as a by-product in an attempt to isolate a Ca2+-sensitive myosin

from rabbit macrophages [7]. FLN, spectrin, fimbrin and α-actinin are very important actin

crosslinking proteins. Since only one actin binding protein is required per filament of actin to induce

gelation, it was termed an “actin-saving protein” for its ability to form a seemingly highly

concentrated polymerized actin network using smaller quantities of actin [7]. This “actin-saving

protein” was eventually termed “filamin” since antibodies that recognized it decorated actin-rich

stress fibers in chicken fibroblasts [100]. ���

i. FLN structure and actin crosslinking There are three human FLN isoforms, FLNs A, B, and C that share 70% sequence homology. FLNA

is the most abundant and widely expressed isoform [7]. FLNA is the most potent actin-filament-

crosslinking protein whose main role is to crosslink F-actin into high-angle orthogonal networks.

FLNA’s ability to form high-angle actin filament branching distinguishes it from other crosslinking

proteins, which are only able to form parallel actin bundles [7]. FLNA is present uniformly

throughout the actin cytoskeletal network and can be found at T, X, and Y-shaped junctions where

actin filaments branch or overlap [6,8,101]. Under electron microscopy, human FLN appears to have

a V-shaped structure.

Structurally, human FLNA exists as a 280kDa homodimer. Each monomer is composed of 24 β-

pleated sheet immunoglobulin (Ig) repeats, with approximately 96 amino acid residues folded into

antiparallel overlapping domains to create rod-like structures [8]. The N-terminal actin-binding

domain (ABD) consists of a stretch of 275 amino acids. The ABD is composed of two calponin

homology domains found in many actin-filament binding proteins including β-spectrin, dystrophin,

α-actinin, nesprin, plectin, and fimbrin [7, 8]. Two hinge regions between repeats 15-16 and 23- 24

contain calpain cleavage sites, which provide the FLNA homodimer with flexibility. The sequential

rigid rod structures separated by flexible hinge regions impart FLNA the stiffness yet flexibility to

crosslink actin filaments at perpendicular angles [8]. The C-terminal repeat comprises the

dimerization domain of FLNA. This domain is necessary for high avidity F-actin binding and

crosslinking [101]. Using an extensive library of FLNA mutants, Nakamura et al. showed that in

  27  

addition to the ABDs, Ig repeats 9-15 contain a supplementary ABD that augments high avidity F-

actin binding. Consequently, Ig repeats 9-15 cannot cause actin gelation on its own [101].

ii. Functions of FLNA FLNA is an important regulator of the actin cytoskeleton in different cell types. FLNA affects cell

shape, locomotion, cell-cell adhesion and stability of the cortical actin cytoskeleton. Cells lacking

this actin binding protein exhibit circumferential blebbing due to cortical instability and an inability

to withstand internal hydrostatic pressures [6, 102]. FLNA plays a role in mechanoprotection in cells

subjected to shear stress, by transducing extracellular signal from the β1 integrin to stiffen the actin

cytoskeleton to resist strain [103]. FLNA also binds migfilin, an important adaptor protein at sites of

cell-cell and cell-ECM contact, to connect the plasma membrane at sites of adhesion to the actin

cytoskeleton [104].

iii. Cellular migration FLNA is required for normal cortical neuron migration from their native neural crest location to the

cerebral cortex during brain development [105]. A null-mutation in the FLNA gene results in X-

linked periventricular nodular heterotopia (PVNH). PVNH results from the accumulation of

heterotopic neurons in the lateral ventricles due to a failure to undergo radial migration from the

subventricular zone to the neo-cortex. This phenomenon creates foci which causes seizures in

females; null mutations in FLNA are usually embryonic lethal in males. The FLNA-deficient M2 cell

line derived from human malignant melanoma exhibits defective migratory ability that is restored

upon FLNA transgene rescue [102]. Depletion of FLNA in motile Dictyostelium amoebae results in a

defect in pseudopod extension leading to impaired locomotion and chemotaxis [106]. Observations

from the PVNH patients, M2 cell line, and Dictyostelium amoebae suggests that FLNA plays a major

role in cellular migration and chemotaxis.

iv. Syndromes resulting from FLNA mutations As a result of a missense mutation, the importance of FLNA in vivo is reflected by a myriad of other

syndromes, including otopalatodigital syndrome, frontometaphyseal dysplasia, and Melnick Needles

syndrome. These syndromes are characterized by a combination of skeletal dysplasia, and anomalies

in the craniofacial, cardiovascular, genitourinary, and intestinal structures [107]. Due to the lack of

  28  

animal models comprising these defects, the mechanisms by which FLNA mutations affect these

structures remain unknown. Using a murine model, Feng et al. provided some insight into the

syndromes caused by human FLNA mutations. They showed that FLNA is important for cell-cell

contacts and adherens junctions during the development of many organs, including blood vessels,

heart, and brain [108].

v. FLNA binds actin network regulators FLNA has more than 20 identified C-terminus binding partners ranging from transmembrane proteins

(including integrins), membrane channels, signaling intermediates, and transcription factors [8]. The

majority of the FLN-interacting proteins bind to FLN between repeats 16 and 24 [108, 112]. The

second-last rod domain binds Rho GTPases, and some of their regulatory cofactors and effectors,

which are important signaling intermediates for actin cytoskeleton remodeling. Specifically, the 23rd

repeat of the FLNA homodimer is a binding site for proteins that play regulatory roles in actin

cytoskeletal remodeling including Rho GTPases Rac, Cdc42, RhoA (constitutive binding sites) [109,

110] the Rac guanine nucleotide exchange factors (GEF) Trio, and Rho GTPase effectors Pak1,

ROCK and FilGAP [8, 111, 112]. Literature shows that some of these effectors function only when

bound to FLNA [112]. In addition to its main role in actin crosslinking, binding of these important

actin regulatory molecules suggests a secondary role for FLNA in mediating signal transduction. The

localization of Rho GTPases within close proximity to their regulatory proteins and effectors via

FLNA likely permits the tight regulation of local actin cytoskeleton remodeling [6, 8].

vi. Regulation of FLN Although there is numerous data on how FLN regulates multiple signaling cascades, its self-

regulation remains largely unexplored. FLNs are phosphorylated by multiple serine/threonine protein

kinases: protein kinase A (PKA), protein kinase C (PKC), Ca2+/calmodulin-dependent protein kinase

II (CaM kinase II), and p90 ribosomal S6 kinase [8]. Phosphorylation of FLNA by PKA in platelets,

increases its resistance to cleavage by calpain [113] and phosphorylation of purified chicken gizzard

FLN by CaM kinase II decreases it actin-binding affinity [114]. The importance of FLN regulation

to the actin cytoskeleton as a whole remains unknown.

  29  

vii. FLNA in literature Leung et al. showed that FLNA was required for osteoclast formation; FLNA-null monocytes had

deficiencies in OCG both in vitro and in vivo. Under normal in vitro plating densities, FLNA-null

osteoclasts were smaller, less numerous, and contained fewer nuclei per osteoclast. Quantification of

osteoclasts in vivo in the distal femoral head revealed similar results, illustrating the in vivo

physiologic relevance of FLNA in OCG. In addition FLNA-null mice were deficient in osteogenic

potential as evidenced by decreased alkaline phosphatase activity suggestive of decreased osteoblast

precursor numbers and/or activity. Serum osteocalcin levels, a serum bone formation marker [115,

116] was also decreased compared to WT mice. Thus FLNA-null mice had a phenotype consistent

with low-turnover osteoporosis, where decreases in BMD, BMC, osteoclast and osteoblast numbers

were observed [115, 117].

Using M-CSF and RANKL as chemo-attractants Leung et al. showed that FLNA was crucial in

monocyte migration. FLNA was shown to be required for stabilization of the orthogonal actin

network in the leading-edge lamellae of migrating cells, whereas migration-defective FLNA-deficient

M2 cells showed cortical instability, resulting in surface blebbing and abnormally long, thin actin

filaments [6, 102, 115]. FLNA was also needed for actin reorganization required for filipodia

formation leading to cell migration [115, 118]. Human FLNA loss-of-function mutations result in

PVNH owing to migratory defects in cortical neurons [105]. Leung et al. found that FLNA-null

neutrophils exhibited defective uropod retraction during neutrophil chemotaxis, which was

responsible for their reduced rate of migration toward formyl-methionyl-leucyl-phenylalanine

(fMLP). When cell densities quadrupled in an attempt to rescue OCG through decreasing

intercellular distances that monocytes must travel before fusing, there was significant recovery of the

formation of large, fully functional FLNA-null osteoclasts. FLNA-null osteoclasts approached WT

levels, indicating that FLNA-null monocytes were competent in fusing regardless of migration [115].

Leung et al. investigated whether decreased migration in FLNA-null monocytes could be a result of a

defect in actin nucleation and subsequently measured the ability of activated WT and FLNA- null

monocytes to generate actin free barbed ends (FBEs). Activated FLNA-null monocytes showed half

the actin polymerization rate of activated WT monocytes, resulting in a twofold reduction in

available actin FBEs. Using monocytes, it was previously shown that decreased actin FBE

generation accompanies decreased migration confirming that FLNA is required for normal actin

  30  

polymerization dynamics during cell migration. FLNA was also shown to be required in the early

stages where dynamic actin remodeling is crucial for pre-osteoclast migration [115].

Activation of Rho GTPases was impaired in FLNA-null monocytes, confirming that FLNA is

necessary in the signal transduction cascade leading to Rac1, Cdc42, and RhoA activation

downstream of the M-CSF receptor. FLNA was shown to coordinate the activation of Rho GTPases

since FLNA-null monocytes migrated as long as Rho GTPases were activated [115].

  31  

1.12 Objectives and Hypotheses Bone is a dynamic tissue that is constantly remodeled via osteoclasts and osteoblasts to maintain

mineral homeostasis. Osteoclasts are bone-resorbing cells that excavate Rpits that are very difficult

to accurately measure. The methods most frequently used to determine osteoclast function are TRAP

staining of osteoclasts in decalcified histological tissue sections and the estimation of pit formation

by osteoclasts on dentin or bone slices using optical light microscopy or scanning electron

microscopy [116, 117]. Typically only the total area of the Rpit is measured and not the volume.

Systemic biomarkers of resorptive function, which measure levels of bone collagen degradation

products or osteoclast enzymes [118, 119] can be assayed in cell culture supernatant, however,

osteoclast formation, morphology, and often number of cells cannot be evaluated. The first objective

was to overcome the limitations of Rpit quantification and develop a new in vitro method to

investigate osteoclast function. The method was validated through three-dimensional (3-D)

characterization of picro Sirius red-stained Rpits produced by WT and Rac-null osteoclasts.

FLNA is an actin binding protein that cross-links cortical actin to allow for efficient migration of pre-

osteoclasts to fuse to form mature multinucleated osteoclasts that are capable of resorbing bone

(OCG). Recent literature examining FLNA in vitro showed that FLNA binds actin filaments and the

loss of FLNA produced migratory deficiencies in pre-osteoclasts. It was postulated that FLNA binds

Rho GTPases and their regulators putting them in close proximity to each other to allow for efficient

migration [115, 119, 120]. The possible role that FLNA plays in osteoclast biochemistry due to its

regulation of the primary structure actin filaments or through its integration of signals that regulate

the actin cytoskeleton prompted the design of experimental procedures to determine whether these

results were mimicked in vivo. The novel in vitro method examining Rpits using 3-D laser scanning

confocal microscopy described in objective 1, was performed on FLNA-KO and WT osteoclasts and

the results demonstrated that FLNA-KO osteoclasts are smaller and produce volumetrically smaller

Rpits. This confirmed that FLNA-KO osteoclasts were indeed compromised in size and resorptive

function. The goal of objective 2 was to determine the role of FLNA in osteoclast generation,

trabecular bone morphology and bone turnover in vivo as well as the role of FLNA in regulating the

mechanical properties of the skeleton. An aging and osteoporosis model was applied to assess how

FLNA-KO bones responded to physiological aging and estrogen depletion.

  32  

Recent studies have raised some new questions about the possible existence of different types of

osteoclasts at different bone sites. The observation that bisphosphonates may cause necrosis of the

jaw whereas long bones appear to not be negatively affected by this drug, suggests that osteoclasts

from the jaw may respond differently to bisphosphonates and that this type of bone may thus harbor

its own type of osteoclasts [4]. Many studies comparing the cranial and appendicular skeletons pay

most attention to TRAP staining of osteoclasts in vitro. All of these studies focus on comparing

mandibles to long bone and calvaria to long bone. In addition, the effect that estrogen depletion has

on these bone sites has never been explored. Therefore we endeavored to determine whether bone

turnover differs in the cranial and appendicular skeletons (mandible vs. calvaria vs. long bone) in

vivo and whether estrogen plays a role in these differences.

Hypothesis 1: 3-D resorption pit imaging in conjunction with a collagen stain will provide a more in

depth understanding of osteoclast function.

Hypothesis 2: Since FLNA regulates the migration phase of OCG it is hypothesized that FLNA-

deficient pre-osteoclasts will not migrate and fuse to form mature, multinucleated osteoclasts. This

will ultimately lead to insufficient bone resorption likely resulting in an osteopetrotic phenotype.

Hypothesis 3: Osteoclasts from different sites will differ with respect to TRAP surface area and

volume and surface area of Rpits. It is expected that bone turnover rates will also differ between the

different skeletal sites both in the presence and absence of estrogen.

  33  

Chapter 2

A 3D scanning confocal imaging method measures Rpit volume and captures the role of Rac in osteoclast function Authors: Goldberg S12, Georgiou J2, Glogauer M1, Grynpas MD2

1. Matrix dynamics group, Faculty of Dentistry- 150 College Street (Fitzgerald building), Toronto, Ontario M5S 3E2 2. Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital- 600 University Avenue, Toronto, Ontario, M5G 1X5 Published in the journal Bone 2012;51(1):145-152

2.1 Abstract Modulation of Rho GTPases Rac1 and Rac2 impacts bone development, remodeling, and disease. In

addition, GTPases are considered treatment targets for dysplastic and erosive bone diseases including

Neurofibromatosis type 1. While it is important to understand the effects of Rac modulation on

osteoclast function, two-dimensional Rpit area measurements fall short in elucidating the volume

aspect of bone resorptive activity. Bone marrow from WT, Rac1 and Rac2 null mice was isolated

from femora. OCG was induced by adding M-CSF and RANKL in culture plates containing dentin

slices and later stained with Picro Sirius Red to image resorption lacunae. Osteoclasts were also

plated on glass cover slips and stained with phalloidin and DAPI to measure their surface area and

nuclei. Volumetric images were collected on a laser-scanning confocal system. Sirius Red confocal

imaging provided an unambiguous, continuous definition of the Rpit boundary compared to reflected

and transmitted light imaging. Rac1- and Rac2-deficient osteoclasts had fewer nuclei in comparison

to WT counterparts. Rac1-deficient osteoclasts showed reduced Rpit volume and surface area.

Lacunae made by Rac2 null osteoclasts had reduced volume but surprisingly surface area was

unaffected. Surface area measures are deceiving since volume was reduced in Rpits made by Rac2

null osteoclasts. Our innovative confocal imaging technique allows us to derive novel conclusions

about Rac1 and Rac2 in osteoclast function. The data and method can be applied to study effects of

genes and drugs including Rho GTPase modulators on osteoclast function and to develop

pharmacotherapeutics to treat bone lytic disorders.

  34  

2.2 Introduction The Rho subfamily of small GTP-binding proteins is implicated in diverse cellular functions. Rho

GTPases act as molecular switches that regulate the actin cytoskeleton to control a multitude of

events such as growth, division, cytoskeletal shape and rearrangement, trafficking and transcription

[see reviews by 121, 122]. Rac GTPases are vital intracellular signaling elements in diverse cell

populations from cells of mesenchymal origin to hematopoietic origin [123-125], the latter of which

include the osteoclast. Osteoclasts are bone resorbing multinucleated cells [reviewed by 25, 7]. Rac

GTPases are implicated in human disease [see review by 126] and considered targets for dysplastic

and erosive bone diseases including Neurofibromatosis type 1 [127].

Rac GTPases Rac1 and Rac2 regulate multiple osteoclast activities including lamellipodia formation,

cell spreading and motility [see review by 128, 129]. Using transgenic mouse models we previously

identified a role for Rac1 in OCG [97, 130]. Additional studies have used various approaches to

modify Rac1 or Rac2 function and reported changes in gross osteoclast resorptive function as well as

in cell shape and motility [96-15]. In particular, we previously noted smaller Rpits formed by Rac1-

and Rac2-null osteoclasts and immortalized pre-osteoclasts by way of toluidine blue staining [11,

16]. While Rac GTPases control many aspects of osteoclast biology, the details of how Rac1 and

Rac2 affect osteoclast resorption at the single cell microscopic level have not been studied.

Bone is a rigid yet dynamic tissue that constantly undergoes destruction and new formation. There

are reliable ways to determine the amount and rate of bone formation by osteoblasts using static and

dynamic histomorphometry. However, it is very difficult to establish the functionality of bone

resorbing osteoclasts in vivo. The methods most frequently used to determine osteoclast function are

TRAP staining of osteoclasts in decalcified histological tissue sections and the estimation of pit

formation by osteoclasts on dentin or bone slices using optical light microscopy or scanning electron

microscopy [116, 117]. Typically only the total area of the Rpit (also known as resorption lacunae)

is measured and not the volume. There are of course systemic biomarkers of resorptive function,

which measure levels of bone collagen degradation products or osteoclast enzymes [118, 119]. Such

markers can be assayed in cell culture supernatant, however osteoclast formation, morphology, and

often number of cells cannot be evaluated.

  35  

To study the Rpit produced by an osteoclast as well as osteoclast morphology and activity including

the effects of Rac GTPase deletion, we have applied 3-D laser-scanning confocal microscopy, a well-

recognized research tool that enables submicron resolution imaging with minimal specimen sample

preparation [131]. Here we introduce an innovative technique in the measurement and quantification

of Rpits using Picro Sirius Red fluorescence in conjunction with laser scanning confocal microscopy.

Picro Sirius Red was applied to dentin slices cultured with osteoclasts, prior to imaging, to stain

collagen in the exposed Rpit. Picro Sirius Red staining of dentin slices is a novel technique for

quantifying osteoclast resorption potential. Activated osteoclasts secrete both protons and

proteinases at their attachment site, resulting in dissolution of bone mineral and degradation of the

matrix exposing collagen. The Picro Sirius Red method of staining is based on the observation that

Sirius red in saturated picric acid selectively binds fibrillar collagens, in particular types V and I.

Collagen is composed of a triple helix of amino acid chains, and stains strongly with acid red dyes

due to the affinity of the cationic protein groups for the anionic reactive groups of the acid dye [132].

Picro Sirius Red staining is favorably visualized by fluorescence microscopy compared to bright-

field microscopy for revealing fine collagen fibrils [133].

To examine the role of small Rho GTPases Rac1 and Rac2 in Rpit formation, we plated osteoclasts

derived from the bone marrow of wild-type, Rac1- and Rac2- deficient mice [134] onto dentin discs

and applied our new procedure for 3-D volume characterization of Rpits. This study, through new

methodology, indicates that it is both necessary and correct to measure Rpits in three dimensions

using a collagen-specific stain. Further, this method allows us to conclude that although Rac1 and

Rac2 are 92% homologous [126, 134], they play non-redundant roles in osteoclast function.

  36  

2.3 Materials and Methods

2.3.1 Animals All procedures were carried out in accordance with the Guide for the Humane Use and Care of

Laboratory Animals and were approved by the University of Toronto Animal Care Committee. We

utilized previously generated and characterized mice in which Rac1 was conditionally deleted in

neutrophils and monocytes by crossing Rac1c/- mice (loxP-flanked Rac1) with mice in which Cre

cDNA has been inserted into the endogenous M lysozyme locus (LysMcre/+) to drive Cre-mediated

deletion of Rac1 in neutrophils and monocytes. Rac1c/-;LysMcre/+ mice were bred with Rac2-/-

mice and the resulting offspring were bred to generate optimal breeding pairs (Rac1c/-

;LysMcre/+;Rac2+/- X Rac1c/-; Rac2+/-). Breeding pairs enabled the generation of mice, which are

null for neutrophil Rac1 (Rac1c/-LysMcre/+Rac2+/+, referred to as Rac1-/-), Rac2 nulls

(Rac1+/+;LysMcre/+;Rac2-/-), Rac1/2-/- double mutants (Rac1c/-;LysMcre/+;Rac2-/-), and control

mice (Rac1c/c;LysMcre/+;Rac2+/+) from the same litters, as previously described. This breeding

strategy allows for the controlling of background variations. Genotyping for Rac1, Rac2, and LysM

alleles was carried out as described previously [125, 134]. Experiments were performed on C57BL/6J

mice, which were maintained in a pathogen-free controlled environment. Animals were sacrificed by

CO2 asphyxiation at four-months of age, following animal care protocol established by the

University of Toronto Animal Care Committee.

2.3.2 Dentin slices We prepared dentin slices from narwhal tusk that was provided by Dr. Morris Manolson (Faculty of

Dentistry at the University of Toronto) legally after being confiscated from illegal sources; dentin

offers several advantages over bone including 1) being denser and less brittle than bone it holds its

composition better during sectioning and staining, 2) is not vascularized nor innervated which may

confound Rpit analysis, and 3) the lack of active remodeling ensures there aren’t any Rpits prior to

experimentation. Regardless, the techniques described herein can be used in other tissues including

bone (data not shown). Narwhal tusk was cut into 300 µm thick slices using a precision diamond

band saw. Each slice was sanded in water for 2 minutes on each side using four different grades of

sandpaper, going from course to smooth to achieve a final thickness of 100 µm. Dentin slices were

hole punched using a standard one hole punch into 5 mm diameter discs. Dentin discs were

  37  

sonicated for 10 minutes in 70% ethanol, 10 minutes in sterile double distilled water and 10 minutes

in α-MEM with 1X antibiotics (32.8 IU/mL penicillin G, 10 µg/mL gentamicin, and 0.05 µg/mL

fungizone). Dentin discs were incubated overnight in α-MEM containing 1X antibiotics at 4ºC; they

were then rinsed with α-MEM and incubated at 37ºC in complete α-MEM for two hours before use.

2.3.3 Isolation of monocyte/osteoclast progenitors and in vitro OCG Femur bones from 4-month-old WT, Rac1 and Rac2 mice were dissected aseptically under a laminar

flow hood. The ends of bones were carefully cut, and bone marrow was flushed out using a needle

and syringe containing α-MEM (Life Technologies). Cell aggregates were broken by repeated

aspiration using the same syringe and a 20-gauge needle. Cells were washed once and resuspended

in α-MEM supplemented with 10% fetal bovine serum (FBS) and 5X antibiotics. To remove stromal

cells, bone marrow cells were cultured overnight (in a humidified incubator at 37 ºC with 5 % CO2)

in tissue culture flasks after which cells in the supernatant were pelleted and resuspended in 10 mL of

α-MEM. OCG was initiated by plating 1x10^6 cells in a 48-well tissue culture plate containing

dentin slices with supplementation by 20 ng/mL M-CSF followed by 30 ng/mL RANKL. Cells were

cultured for 6 days with a change in cell culture media and cytokine supplementation every other day.

On day 6, dentin slices were incubated in 2 mM EDTA/PBS for 10 minutes at 37 ºC to remove any

adherent cells.

2.3.4 Picro Sirius Red staining Cultured dentin discs were washed in three changes of distilled water to remove salt crystals. Dentin

discs were treated with 0.5% phosphate buffer (pH 7.4) for 2 minutes. Discs were then immersed in

papain and incubated for 90 minutes at 37ºC to digest any proteoglycans attached to collagen [135].

Following papain digestion, slices were immersed in 0.1% Picro Sirius Red (in saturated aqueous

picric acid, pH = 1.5) for 60 minutes at room temperature. Dentin discs were then washed with 0.1N

HCl for 2 minutes; the acid environment enhances washout of dye [135]. Any water or acid

remaining on the discs was blotted with damp filter paper to avoid Rpit flooding. Discs were

dehydrated with ethanol, cleared with xylene, and mounted flat on microscope slides in Permount

using a number 1 cover slip. Mounted slides were labeled according to genotype and allowed to dry

for 7 days in a laminar flow hood.

  38  

2.3.5 DAPI and phalloidin staining of osteoclasts OCG was initiated from WT, Rac1- and Rac2-null bone marrow cells and cultured for 6 days on 1

cm diameter glass cover slips. On day 6, glass cover slips (containing adherent osteoclasts) were

rinsed three times with PBS and then subjected to 5 minutes in 0.1% Triton/PBS. Cells were then

incubated in 1% BSA/0.1% Triton/PBS for 20-30 minutes at room temperature. A 1:40 ratio of

phalloidin-Alexa488 to 1%BSA/0.1% Triton/PBS was added to cells and allowed to sit in a dark area

for 30 minutes. The cells were washed with 0.1% Triton/PBS three times and then subjected to

DAPI stain for 10 minutes. Glass cover slips were washed with water and allowed to air dry. Dry

glass cover slips were mounted on slides using Dako Fluorescent Mounting medium and Cytoseal to

seal edges.

2.3.6 Confocal laser scanning microscopy Images were collected on a Nikon C1si laser-scanning confocal system with Ti-E inverted

microscope using EZ-C1 software (Nikon Canada, Mississauga, ON, Canada). To identify pits under

low and high magnification, a dry 10X 0.30 NA lens and oil 60X 1.40 NA lens were employed,

respectively. Samples were exposed to 561 nm laser light and the reflected signal was collected in

one detector, while the Sirius red emission was filtered at 605/60 nm and imaged in another detector

channel. The smallest pinhole setting (3 µm) and constant gain were maintained. Scans were of 1024

xy resolution and averaged twice. Z-stacks were only collected at 60X and a step size of 0.3 µm. A

transmitted light image (non-confocal) was also collected through an oil condenser (1.20 NA). For

imaging osteoclasts growing on glass cover slips, DAPI and Alexa488 was excited by 405 nm and

488 nm laser lines respectively, and corresponding emission were sequentially collected through

450/35 and 525/50 filters in front of separate detectors.

2.3.7 Analysis Volume analysis and viewing was performed using Nikon Imaging Software NIS-Elements AR

v3.22 (Nikon Canada, Mississauga, ON, Canada). Rpit aspect ratio (width:length) at the dentin

surface level was calculated by defining the longest linear distance across a pit (length) and the

perpendicular size (width) at the length midpoint. We defined a Rpit as a single entity when

consisting of one contiguous area at the surface lacking fluorescence; if the site pinched off into two

or more entities at a depth greater than 50% of its total z-range, this was still considered to be one

  39  

distinct pit. If the site pinched off within 50% of its z-range, then we measured the ratio of the

longest linear length at the surface to the length between the centers of the dips, and ratios less than 2

were treated as separate pits. Numerical results were expressed as mean ± SEM. Statistical analysis

was performed using SigmaPlot 11.0 (Systat Software, Inc.) using an ANOVA followed by pairwise

comparisons by Tukey test with the level of significance set at p<0.05.

  40  

2.4 Results

2.4.1 Confocal stacks of Rpits We used several imaging modes in tandem to first characterize the detailed shape of Rpits, and later

studied the effect of Rac1 or Rac2 deletion. Bone marrow cells were isolated, plated on dentin tusk

slices for 6 days and washed off. Picro Sirius Red was applied to label the fibrillar collagen network,

and samples were mounted on slides with cover slips for imaging. Confocal imaging of Picro Sirius

Red fluorescence emission revealed a blanket layer of signal on the dentin surface except at the Rpit

hollowed areas (as well as some linearly arranged grooves created during the cutting and sanding

process). At the Rpit locations, focusing deeper into the specimen (axial z-axis) exposed a circular

type of staining pattern that narrowed in diameter (Figure 1a). A bright-field image obtained

through a non-confocal transmitted light detector gave a high quality en-face perspective and a crude

estimate of the axial limits of the Rpit (Figure 1b, left). Confocal imaging of the reflected excitation

laser beam provided a better reveal of the Rpit shape, however there were gaps in signal continuity

when xz or yz profiles were constructed (Figure 1b, shown in green). However, the Picro Sirius Red

fluorescence signal provided for a wider but unambiguous and continuous definition of the pit

boundary over all z-levels (Figure 1b, shown in red). The simultaneous overlay of the reflected

signal with the fluorescence signal (Figure 1b, Merge) revealed greatest overlap at the inner-most

margin of the fluorescent ring, which represents the first surface of exposed collagen.

2.4.2 Quantification of Rpit volume To quantify Rpit area and volume, we defined fluorescent outlines over all z-sections using semi-

automated tools in NIS Elements software. Many sections had convoluted contours, which forced us

to also manually define edges (Figure 1c). Subsequently, we also performed volume rendering to

visualize 3-D shapes of Rpits using either a standard intensity scale or also with a colour scale to

code for depth (Figure 1d, left). We preferred an inverted and skewed viewing angle since this

yielded a fuller representation of Rpit shape. Superimposing the defined binary areas onto 3D views

provided for an alternative perspective (Figure 1d, right). The Rpit can also be visualized as movies

showing the xy fluorescence sequentially over the various depths or in 3-D rotations (Supplemental

Video 1).

  41  

2.4.3 Deletion of Rac1 or Rac2 in osteoclasts We applied the Picro Sirius Red fluorochrome - confocal laser scanning microscopy technique to

compare osteoclasts and their resultant Rpits created by control, Rac1-deficient, or Rac2 null

osteoclasts applied onto dentin discs (Figure 2a). From each pit site shown on the left side of Figure

2a, separate 3D movies were created (Supplemental Video 1, 2, and 3 for control, Rac1-/-, Rac2-/-,

respectively). The pits formed by Rac2-null osteoclasts had an average surface area that was

statistically equivalent to control Rpits. However, despite no change in surface area, the pit volumes

made by Rac2 null cells were indeed significantly reduced (Figure. 2b-d). In contrast, the average

Rac1-/- Rpit showed a reduced surface area compared to Rac2 null and control Rpit, as well as a

corresponding reduction in volume. Remarkably the Rac1-/- pit volume was equivalent to Rac2-null

pit volume, indicating the Rac1-/- pits have greater volume for a given surface area (Figure 2d).

Another interesting finding was that Rac2-null pits were significantly rounder than control pits, as

seen in the greater aspect ratio measured at the dentin disc surface (Figure 2e).

2.4.4 Osteoclast shape Given that Rpit surface area and volume was modified by deletion of Rac1 and Rac2, we proceeded

to assess osteoclast shape. Osteoclast analysis in cultures was achieved by labeling the F-actin

cytoskeleton with phalloidin-Alexa488 and nuclei using DAPI. Representative confocal images of

fluorescence appear in Figure 3a. The number of nuclei and surface area of osteoclasts were

analyzed in the same fashion as in Rpits. Regarding osteoclast surface area, there was a significant

reduction in Rac1 nulls (Figure 3b). The surface area of the osteoclasts growing in culture was

similar to the Rpit surface area described in Figure 2b. Interestingly, the number of nuclei was

significantly reduced in Rac1 and Rac2 nulls when compared to control osteoclasts (Figure 3b).

  42  

2.5 Discussion Osteoclasts are specialized cells that degrade bone matrix and mineral during normal development

and repair of the skeleton. The Rac family of Rho GTPases regulates actin cytoskeleton dynamics,

cell adhesion and cell movement in osteoclast formation, cell growth, gene transcription and reactive

oxygen species production [126, 127, 134]. We studied the role of Rac1 and Rac2 deletion on

osteoclast morphology and their resorptive capacity through microscopic analysis of Rpit volume.

Through volumetric analysis we demonstrated that osteoclasts lacking Rac2 form Rpits that are

shallow, a feature that is not discernable using planar measures. Rac1 null osteoclasts showed

several defects including size, number of nuclei and Rpit volume. Thus, Rac1 and Rac2 have non-

redundant roles in osteoclast function. The defects in Rpit volume were revealed through confocal

analysis; reflected light imaging provided an estimate of the pit shape, whereas the fluorescent

collagen reporter Picro Sirius Red proved superior in terms of resolving the edges and providing for a

continuous definition of the contour. This new information provides further details on the role of

Rac GTPases on osteoclast resorption function.

2.5.1 Confocal imaging of osteoclasts and fluorescent markers for collagen Our innovative confocal technique in conjunction with Picro Sirius Red fluorescence volumetrically

measured Rpits produced by osteoclasts lacking Rac1 and Rac2 small GTPases. This new imaging

method allowed us to precisely map Rpit shape and volume. We encountered several instances

where surface shapes alone mislead us to believe the site consisted of a single Rpit, particularly in the

mutant samples. Even after careful classification, surface area analysis alone was deceiving in that it

was still not a valid predictor of volume. 3-D confocal analysis of the Rpit is thus essential and it

allowed us to derive novel conclusions about the functionality of osteoclasts.

It is likely possible to use other fluorescent stains or antibody-based probes to image the collagen

substratum. It may also be useful to study the pits without removing the osteoclasts and to use

additional probes to visualize their morphology including orientation and interactions; under our

conditions the pits were less than 60 µm in depth, though it would still be possible to use 2-photon

confocal microscopy for osteoclasts within pit sites that are deeper than 100 µm. We opted to obtain

many high resolution images per Rpit site for finer display and accurate analysis, such that our point-

  43  

scanning imaging system required a long acquisition time (~15 minutes per site) to collect the 2-

times averaged, 1024 xy resolution and oversampled z-sections (0.3 µm steps). Note the collection

process can be accelerated >50-fold by not averaging and sampling at 516 x 516 pixels with 2 µm z-

steps, and while this would sacrifice spatial detail it would have minimal impact on estimates of

shape, area, and volume. Alternatively the collection process could be greatly accelerated if one

were to use a spinning-disc type confocal unit. Besides the use of a confocal system, it is possible to

employ other fluorescence imaging systems to image Rpits, including deconvolution-based as well as

structured illumination systems especially where osteoclasts are removed.

Confocal laser scanning microscopes are now increasingly more popular and are being recognized for

their potential in the precise volumetric measurement of cells. Specifically, the utilization of the

confocal principle provides submicron resolution and is the most important improvement in the

progression of Rpit analysis [117]. Although two-dimensional methods such as scanning electron

microscopy can provide excellent images of a Rpit, the considerable sample preparation and

expensive equipment make for a time consuming and non-practical method for their quantification

[136]. Such techniques employed to quantify Rpits fall short in illustrating numerical solutions for

the derivation of volumes, angles and depth, which provide a deeper insight into the function not just

the adherence and spreading of osteoclasts. Area resorbed is most often measured through two-

dimensional methods of pit resorption and may be misleading because an increase in pitted area does

not necessarily occur with an increase in pit volume. Furthermore, the simple demonstration that a

resorptive event occurred does not necessarily give an idea about osteoclast activity. Two-

dimensional methods present a wide range of biological variability in single cell function in any one

experiment [136, 137]. The aforementioned limitations require that Rpits be measured in three

dimensions.

2.5.2 Effects of osteoclast Rac1 deletion We found Rac1 nulls had reduced osteoclast and Rpit size and volume, which likely accounts for

their reduced resorptive function. Rac1 null osteoclasts have about half the nuclei and are half the

overall size of their wild-type counterparts, which also may contribute to impaired resorption. Rac1

may have additional functions, which would further confound interpretation of some assays. For

instance other studies found that delivery of either a constitutively active or dominant negative form

of Rac1 into macrophage-derived cells caused spreading and retraction, respectively [138]. Injection

  44  

of anti-Rac antibody caused osteoclasts to round up [139] while a knock down approach found that

Rac1 regulates osteoclast survival, apoptosis, motility and spreading [96]. The above studies also

suggested there is reduced osteoclast resorptive activity after measuring area resorbed using pit

formation assays and/or collagen concentration in the media [96, 98]. It is important to note that our

approach herein used a genetic approach to delete Rac1 in osteoclasts and we studied their individual

shape and Rpit volume at submicron resolution to reveal defects in resorption at the single cell level.

Modification of Rac1 activity can swing the balance of bone formation and resorption. For the

resorption process to begin, osteoclasts undergo dynamic changes in their cytoskeleton [82, 140].

Attachment of osteoclasts to vitronectin, osteopontin, and type I collagen proteins containing the

Arg-Gly-Asp (RGD) sequence motif is the first step in inducing the polarization of osteoclasts which

induces cell spreading [31, 141]. Osteoclast spreading depends on the substrate the osteoclast

interacts with and requires integrity of the ανβ3 vitronectin receptor and of the c-src proto-oncogene

[83]. Rac1 null osteoclasts are able to form podosomes and eventually sealing zones to resorb bone

sufficiently but are unable to recognize proteins containing the RGD sequence though vitronectin

receptors to induce osteoclast polarization [84]. In vivo data suggests that Rac1 deletion decreases

bone remodeling, but the effect on bone resorption is dominant resulting in increased bone mass and

bone quality [84]. Thus, insufficient resorption produced by Rac1 null osteoclasts results in an

imbalance between resorption and formation. Osteoblasts are recruited at an ordinary rate while

resorption is limited resulting in an osteopetrotic-like phenotype. This defect is opposite to the actin-

and Rac-dependent bone loss (osteoporosis) that predominates in normal ageing of the skeleton [82].

2.5.3 Rac2 deletion yields shallower Rpits Contrary to osteoclasts lacking Rac1, Rac2-null osteoclasts maintain their morphology in the x-y

planar direction based on their shape in culture as well as the surface area of their Rpits. Although

this may suggest a relatively normal phenotype, Rac2-null osteoclasts are in fact functionally

compromised as seen by their low volume values in comparison to their control counterparts.

Furthermore, previous studies have shown Rac2-null macrophages have reduced F-actin levels and

lack podosomes, which are integrin-based adhesion sites consistent with a possible adhesion defect,

which could impact the formation of a Rpit [124]. Our data suggest that Rac2-null osteoclasts are

able to polarize but are unable to form functional sealing zones to properly resorb bone. Preliminary

in vivo data suggests that through decreased resorptive activity, Rac2-null osteoclasts may negatively

  45  

affect the release of anabolic factors normally produced by activated osteoclasts resulting in an

osteopenic phenotype. Garimella et al have reported the importance of BMP’s expressed in activated

osteoclasts and involved in the initiation of the anabolic phase of bone remodeling [142]. The exact

role of Rac2 in regulation of osteoclast functions remains unclear, however our result in conjunction

with in vivo findings suggest that this small GTPase might play a role in osteoclast-mediated

promotion of bone formation. Rac2 null osteoclasts have no significant differences in the number of

nuclei and surface area in comparison to controls. Importantly, it is misleading to conclude that

osteoclasts lacking Rac2 have normal resorptive capacity because in fact their Rpits had significantly

reduced volume.

2.5.4 Rpit relationship to osteoclast size and nuclei Many factors affect osteoclast resorptive activity, including proliferation, differentiation, survival,

recruitment and growth. Cell plating density, cytokines and growth factors including M-CSF affect

the size and number of osteoclast nuclei as well as their transcriptional activities, which correlate

with resorptive function [140, 142-147]. The surface area of Rpits (Figure 2b) was similar to that of

single osteoclasts grown on cover glass (Figure 3a-b); thus the Rpits were likely a result of single cell

action. However, we found that both Rac1 and Rac2 nulls had fewer nuclei than control osteoclasts,

though Rac2 nulls were only modestly affected; additionally, only Rac1 nulls showed a significantly

reduced cell surface area (as well as decreased Rpit area). Thus, failure of Rac1 null osteoclasts to

develop into large multinucleated cells may affect their resorptive function. In the case of Rac2 nulls

however, the reduced Rpit volume must involve other changes not related to their growth.

  46  

2.6 Conclusions It is not possible to generate an accurate description of osteoclast formation and function without

Rpit volume characterization. Although WT and Rac2 null osteoclasts are morphologically similar,

they are in fact very different in their ability to resorb bone. Osteoclast migration tracks were

observed frequently on Rac2 null dentin slices. Thus Rac2-deficient osteoclasts do move over the

surface of the bone, which they partly resorb, and inhibition of osteoclastic movement is not

responsible for the impaired resorption. This phenomenon is observed in other genotypes but at a

significantly lower rate. A model comparing wild-type osteoclasts to the distinct Rac1 and Rac2 null

phenotype and dysfunction appears in Figure 4. The results of this study show that three-dimensional

characterization of an Rpit is necessary to fully characterize osteoclast function. Our Rac Rpit

validation experiment demonstrates that no suitable estimate of resorption can be complete without

reliable 3-D volume imaging. Finally, this method will allow researchers to rapidly assess the effect

of gene and drug manipulation on osteoclast function through reliable evaluation of Rpits. The

consequences of Rac GTPase dysregulation [128] are likely to affect bone remodeling that occurs

during development and repair as they depend on osteoclast resorption function. Rac GTPases are

also considered targets for dysplastic and erosive bone diseases including Neurofibromatosis type 1

[129]. Further research into the role of Rho GTPases in metabolic bone diseases and regulation of

remodeling will require detailed analyses of osteoclast structure and function.

  47  

2.7 Figures

2.7.1 Figure 1. Rpit 3-D visualization

Figure 1. Rpit 3-D visualization and analysis by fluorescence and confocal microscopy. Confocal microscopy images through a Rpit made by control osteoclasts on an dentin disc and revealed by Picro Sirius Red fluorescence. (a) Montage showing a sampling of single plane images (frames 69-

  48  

170 of 201) collected at 0.3-micron z-steps. (b) Three sets of data each collected by a distinct imaging mode: first is a non-confocal transmitted light image collected by an external detector (left, in gray scale), second is a confocal detection of the reflected excitation laser beam (green), and third is a confocal image of Picro Sirius Red fluorescence (red). The final image set in the row is an overlay of reflected and fluorescence signals. Each image set shows the corresponding z-profile views through the x and y axis point marked by the orange crosshair. Compared to the reflected light images, the fluorescence signal was more distinct and provided a continuous definition of the pit boundary along xy and xz/yz views. (c) Analysis of a single section of Picro Sirius Red fluorescence near the dentin surface (left) using software area-autodetect feature (center, yellow line); for images not optimally defined, manual outlining was performed (right, yellow fill). (d) Volume rendering of the confocal image stack (left) using z-depth colour coding (see scale on left, 40 µm range) revealed the three dimensional shape of the Rpit (left; also see Supplemental Video 1). The binary layers containing the defined areas were superimposed onto the 3-D volume view (image on right, yellow).

2.7.2 Figure 2. Volume and Surface Area of Rpits

Figure 2. Reduced Rpit volumes with distinct shapes created by each of Rac1/- and Rac2/- null osteoclasts. (a) WT, Rac1- or Rac2-null osteoclasts were allowed to act on dentin discs, Picro Sirius Red was applied and confocal image stacks of fluorescence were collected. For each genotype shown on a separate row, there are images from two sample fields. An image taken at a z-depth level near the dentin disc surface is displayed alongside the z-profile through the x and y-axis indicated (orange crosshairs); same magnification for all xy images (see scale bar) whereas z scale for each

  49  

example is included. Note that many of the sites shown from mutant osteoclast samples contain multiple Rpits. Corresponding 3-D volume views using colour to code for depth (blue-green-red over each depth range) are shown in the centre of each row. 3D movies of examples shown on the left side appear in Supplemental Videos 2-4. (b) Bar graph showing average Rpit surface area, which was found to contain group differences greater than expected by chance (ANOVA; p<0.001). Surface area was significantly different from control in Rac1-/- (P<0.004) and also Rac2-/- (p<0.002), but Rac2-/- was not different from control (p=0.98). (c) Average volume per resorptive event showed differences between groups (ANOVA; P<0.001); Rac1-/- and Rac2-/- volume were significantly lower than control (P<0.001 for each), while no difference was reported between Rac1- and Rac2- nulls (p=0.99). (d) Surface area to volume relationship represented by data binning. (e) Rpit aspect ratio by genotype measured at dentin surface and based on maximum diameter to perpendicular diameter (significant difference detected by ANOVA; p<0.05); the average Rac2-/- Rpit aspect ratio was greater than that of controls (p=0.02). Sampled from N=3 mice and n=29, 29, 32 pit sites from control, Rac1-/-, Rac2-/-, respectively. Significant differences from control at p<0.05.

2.7.3 Figure 3. Morphology of Osteoclasts

Figure 3. Altered in situ morphology of Rac1- and Rac2-null osteoclasts. (a) Confocal images of control, Rac1- and Rac2-null osteoclasts grown on glass cover slips and stained with DAPI (shown in green pseudocolour) to reveal nuclei and phalloidin-Alexa488 (shown in orange) to reveal cell shape through definition of the F-actin cytoskeleton. Lower images show xz plane at the level of white dash line locations. (b) Plots of surface area and nuclear counts by genotype (N=3 mice each with n=10, 11, 9 cells for control Rac1-/-, Rac2-/-, respectively). Significant differences from WT at p<0.001.

  50  

2.7.4 Figure 4. Model of Rpits

Figure 4. Model of morphological and functional defects in Rac1 or Rac2 nulls. Illustration depicting WT osteoclast morphology and Rpit formation with that of Rac1- and Rac2-null osteoclasts.

  51  

2.8 Acknowledgements The work was supported by a CIHR operating grant to Dr. Michael Glogauer. We thank Dr. Kevin

Conway of Nikon Canada for assistance on confocal analysis and visualization. We would also like

to thank Dr. Morris Manolson from the Faculty of Dentistry at the University of Toronto for

providing us with narwhal tusk to use for dentin slice production.

  52  

Preface to Chapter 3  Rac GTPases act as intracellular second messengers to transmit extracellular stimuli that are crucial

for actin cytoskeletal remodeling. FilGAP is a newly identified GAP specific for Rac and binds

FLNA to suppress and mediate the antagonism of Rac by Rho at the leading edge of protruding cells.

The 23rd repeat of the FLNA homodimer is a binding site for Rac, which has been thought to

function only when bound to FLNA. The localization of Rac within close proximity to their

regulatory proteins and effectors via FLNA likely permits the tight regulation of local actin

cytoskeleton remodeling [6, 8]. The success of the Rac validation experiment using 3-D imaging of

osteoclast Rpits, prompted the evaluation of Rpits of osteoclasts that lack FLNA. We used this novel

technique on FLNA-KO osteoclasts to further validate our method and to expand the minimal in vitro

research that has been accomplished on FLNA. FLNA-KO osteoclasts were smaller and produced

shallower Rpits, similar to what was observed in Rac1 and Rac2 KO osteoclasts. Since little is

known about the role of FLNA in OCG we wanted to take it a step further and see how FLNA

deletion affects bones in vivo. One in vivo study using DEXA showed an osteopenic phenotype in

male FLNA-KO mice [115]. It was important to further characterize the role of FLNA using more

extensive in vivo techniques. Aging and osteoporosis models were employed to provide further

dimension and insight into the role of FLNA in female mice.

  53  

Chapter 3 Deletion of Filamin A protects cortical and trabecular bone from post-menopausal changes in bone microarchitecture Authors: Goldberg S1,2, Glogauer J1, Grynpas MD1,2, Glogauer M1.

1. Matrix dynamics group, Faculty of Dentistry- 150 College Street (Fitzgerald building), Toronto, Ontario M5S 3E2 2. Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital- 600 University Avenue, Toronto, Ontario, M5G 1X5 Submitted to: Calcified Tissue International License Number: 3641520709036

3.1 Abstract

Objective: To determine the in vivo role of FLNA in osteoclast generation and function, through the

assessment of trabecular bone morphology, bone turnover and the resulting changes in mechanical

properties of the skeleton in mice with targeted deletion of FLNA in granulocytes. Methods: Using

a conditional targeted knockdown of FLNA in osteoclasts we assessed in vitro OCG as well as

osteoclast function. Bone characteristics in vivo including Micro-CT, histomorphometric analyses

and bone mechanical properties were assessed in female mice at 5 months of age and in an aging

protocol (comparing 5-month-old and 11-month-old mice) and an osteoporosis protocol (OVX at 5

months of age and then sacrificed at 6 and 11 months of age). Results: In vitro OCG analysis

revealed that WT osteoclasts are larger and contain more nuclei compared to FLNA-KO osteoclasts.

WT Rpits were larger and deeper compared to FLNA-KO pits. In vivo bone densitometry,

mechanical and histomorphometric analyses revealed a mild osteopenic phenotype in the FLNA null

5-month and aging groups. The WT and FLNA-KO bones did not appear to age differently.

However, the volumetric bone mineral density decrease associated with OVX in WT is absent in

FLNA-KO-OVX groups. The skeleton in the FLNA-KO-OVX group did not differ from the

FLNA-KO group both in mechanical and structural properties as shown by mechanical testing of

femora and vertebrae and histomorphometry of vertebrae. Additionally, FLNA-KO femora are

  54  

tougher and more ductile than WT femora. Conclusion: The result of this study indicates that while

FLNA-KO bones are weaker than WT bones, they do not age differently and are protected from

estrogen-mediated post-menopausal osteoporosis.

  55  

3.2 Introduction Bone is a dynamic tissue that is continually remodelled to provide maximal strength with minimal

mass as determined by the physiological needs of vertebrate organisms [10]. Bone formation and

resorption are metabolic processes accomplished through the precise coordination of osteoblast and

osteoclast activity. Osteoblasts of mesenchymal origin deposit the calcified bone matrix while

osteoclasts of haematopoietic origin resorb bone [27]. Osteoclasts are large multinucleated cells, the

only cells capable of resorbing bone. Osteoclasts are highly polarized cells requiring dynamic

cytoskeletal reorganization in order to adhere, migrate and resorb bone mineral and matrix [28, 148].

Diseases such as osteoporosis caused by excessive bone turnover, can result in fractures that can

cause morbidity, shortened lifespan and weakened skeletal structure [29]. Osteoporosis is

characterized by low bone mass and deterioration of bone architecture [30]. One of the major

consequences of osteoporosis is disturbed bone architecture resulting in fractures, which places a

significant economic burden on the healthcare system [32]. Up-regulation of bone turnover through

mechanisms not clearly defined is a direct result of estrogen depletion leading to post-menopausal

osteoporosis [69]. OVX is a valuable tool used in numerous animal models to mimic estrogen

depletion in post-menopausal women. Estrogen deficiency leads to stimulation of bone resorption

and formation through prolonging and shortening the lifespan of osteoclasts and osteoblasts

respectively. As a consequence, a high-turnover state develops which leads to bone loss and

perforation of trabecular plates [70, 72].

Following stimulation by cytokines RANKL and M-CSF, osteoclasts are formed (OCG) through

mononuclear monocyte fusion derived from hematopoietic progenitors in the bone marrow. In vivo,

OCG is supported by cell-to-cell contact between expressed RANKL on the surface of osteoblasts

and RANKL receptor on the surface of osteoclast precursor cells [27]. Migration is a crucial cellular

event in osteoclast formation and function as it brings cells in close proximity prior to fusion to

enable the formation of mature multinucleated osteoclasts [27, 89]. Cellular migration and membrane

fusion require dynamic actin cytoskeleton reorganization achieved through Rho family small GTPase

activity [89, 97, 149]. Recent literature postulates that the flexible hinge region of actin binding

protein FLNA localizes Rho GTPase signaling intermediates within close proximity to each other to

facilitate efficient actin cytoskeletal remodeling.

  56  

FLNA is a ubiquitous actin binding protein that cross-links cortical F-actin into three-dimensional

orthogonal networks at the leading edge of cells in X, T or Y shaped junctions. FLNA is the most

abundant and widely expressed isoform among FLN’s A, B and C, sharing 70% sequence homology

[6, 7, 8]. FLNA is critical for normal cortical neuron migration from their native neural crest location

to the cerebral cortex during brain development. A null mutation in the FLNA gene results in

defective migration of neurons in the lateral ventricle leading to PVNH. The FLNA-deficient M2 cell

line, derived from human malignant melanoma, also exhibits defective migratory ability that is

restored upon FLNA transgene rescue [105].

Deficiencies in OCG in FLNA-null monocytes were observed in recent in vitro studies and

preliminary in vivo studies. Under normal in vitro plating densities, FLNA-null osteoclasts were

smaller, less numerous, and contained fewer nuclei per osteoclast. Quantification of osteoclasts in

vivo in the distal femoral head of male mice revealed similar results, illustrating the in vivo

physiological relevance of FLNA in OCG. Further findings of reduced OCG in vivo in male mice

suggest a possible skeletal phenotype of osteopetrosis, with or without concomitant defects in bone

remodeling [115]. We endeavored to further characterize WT and FLNA-KO osteoclasts and their

resultant Rpits in vitro. Our findings prompted us to study how the loss of FLNA would impact

osteoclasts and bone in vivo particularly in female mice where OVX has a more severe impact on

fracture risk. We postulated that deletion of FLNA in a transgenic mouse model would result in an

osteopetrotic phenotype where FLNA-KO mice will have fewer active osteoclasts and/or reduced

bone remodeling. Our model aims to assess the in vivo impact of the loss of FLNA on the bony

skeleton and its impact on aging and OVX transgenic mice. This was achieved using 5, 6 and 11

month-old FLNA-WT and FLNA-KO mice (n=15 per group). The 5-month group was used to assess

whether there was a phenotype upon loss of FLNA. The 11-month group was compared to the 5-

month group to study the physiological effects of aging and to determine whether WT mice age

differently from FLNA-KO mice. Mice were OVX at 5 months of age and then sacrificed at 6 and

11 months of age to establish an osteoporosis model for loss of FLNA. In vivo micro-CT, mechanical

testing and histomorphometry were used to characterize the role of FLNA in osteoclast generation,

trabecular bone morphology, bone turnover and the consequences of the lack of FLNA in aging mice

and upon estrogen depletion.

  57  

3.3 Materials and Methods

3.3.1 Animals All procedures described were performed in accordance with the Guide for the Humane Use and Care

of Laboratory Animals and were approved by the University of Toronto Animal Care Committee.

SV129-black 6 mice containing the conditional knockout of the X-linked FLNA gene (FLNA-null),

were generated as described previously [108]. Since a global knockout resulted in lethality, a

conditional knockout strategy was used with loxP sites inserted into introns 2 and 7 of the mouse

FLNA gene (FLNAc/c or male FLNAc/y mice). Deletion of the FLNA gene in granulocytes

(neutrophils and monocytes) was accomplished by breeding these mice with mice expressing cre-

recombinase under control of the granulocyte-specific lysozyme M promoter that is active during

early embryogenesis [125]. Cre-mediated recombination deletes exons 3-7, producing a non-sense

mutation with early FLNA truncation at amino acid 121. To confirm deletion of the FLNA gene, tail

snips were used to prepare DNA for PCR analysis as described previously Littermates with

unsuccessful cre-mediated recombination were used as wild-type control (WT) [125]. To ensure that

Flna was deleted at the protein level, cell lysates from freshly-isolated WT and Flna-null monocytes

were subjected to SDS-PAGE and Western blotting (see below) using these antibodies: rabbit

polyclonal anti-mouse filamin A (A301-135A, Bethyl Laboratories, Inc.; 1:2000) followed by HRP-

conjugated donkey anti-rabbit IgG (NA934V, GE Healthcare; 1:2000) [115].

For the purposes of this study 10 groups of mice containing FLNA-WT and FLNA-KO mice were

used. Each group was comprised of 15 mice for a total of 150 mice to establish statistical

significance and guard against losses. In order to investigate bone characteristics in mice missing

this important protein for bone resorption, 5-month-old WT and FLNA-KO female mice were

compared to each other and for the aging model, to 11-month-old WT and FLNA-KO mice (n=15 per

group). Mice were OVX at 5 months of age and then sacrificed at 6 and 11 months to establish

osteoporosis models for loss of FLNA. The 11-month group was utilized to study the natural

physiological aging model of osteoporosis as well as the extreme case of osteoporosis, which is

induced by OVX. No significant differences were reported at the tissue level for bone resorption and

formation in vivo therefore, the 6-month group was introduced to ensure that the effects due to OVX

were taking place at the tissue level, which takes place 4-6 weeks post-surgery.

  58  

Calcein green is a bone formation marker that is preferentially taken up at the site of active

mineralization of bone [150]. Two and ten days before euthanasia 5, 6, and 11 month-old OVX mice

were injected in their peritoneal cavities with calcein green (30 mg/kg) for bone dynamic

histomorphometric analysis. Animals were sacrificed by CO2 asphyxiation, following animal care

protocol established by University of Toronto Animal Care Committee.

3.3.2 Isolation of monocyte/osteoclast progenitors and in vitro OCG Femur bones from 5 month old WT and FLNA-KO mice were dissected aseptically under a laminar

flow hood. See section 2.3.3.

3.3.3 Picro Sirius Red staining See section 2.3.4.

3.3.4 DAPI and phalloidin staining of osteoclasts OCG was initiated from 5-month-old WT and FLNA-KO bone marrow cells (2x10^6/well) and

cultured for 6 days on 1 cm diameter glass cover slips. Cells were then fixed in 4% PFA. See

section 2.3.5

3.3.5 Confocal laser scanning microscopy See section 2.3.6

3.3.6 Analysis See section 2.3.7

3.3.7 Bone densitometry and structural (BMD) analyses Micro-CT is a tool used to create high-resolution (~6 microns) three-dimensional images of bone and

for the calculation of vBMD (g/cm^3) as well as to evaluate changes in trabecular bone

microarchitecture. Micro-CT was performed on right femora and sixth lumbar vertebrae from 5 and

11-month-old mice. Femora and sixth lumbar vertebrae, trimmed to leave only the vertebral body,

were mounted in microtubes and scanned using Sky Scan 1172 Micro-CT scanner (Bruker). Femora

and vertebrae were scanned at 11.6 um and 6.1 um resolution respectively. All images were obtained

at an x-ray voltage of 50 kV and current of 800A with a 0.25 mm aluminum filter to ensure a uniform

  59  

beam. All scans were reconstructed and calibrated with the use of two hydroxyapatite standards that

were provided by the manufacturer. Reconstructed images obtained from scanning were analyzed

using the Sky scan CT-Analyzer software (Version 1.6.1). Femoral geometry and vBMD were

assessed from the analysis of a region of interest created 0.25 mm above and below the midpoint.

Femoral midpoints were measured by use of digital calipers and serve as the points of fracture for

three-point bending tests. The density parameter considered was vBMD, and the structural

parameters analyzed were bone area (B.Ar (mm^2)), cross-sectional thickness (Cs.Th (mm)) and AP

diameter (mm). Vertebral trabecular architecture and vBMD were assessed by creating a region of

interest along the length of the vertebrae but with exclusion of vertebral growth plates. Structural

parameters analyzed for vertebrae included percentage of bone volume (BV/TV (%)), trabecular

thickness (Tb.Th. (mm)), trabecular number (Tb.N.) and trabecular separation (Tb.Sp. (mm)).

3.3.8 Mechanical testing

��� Right femora from 5 and 11 month-old mice were tested in three-point bending to evaluate the

mechanical properties of cortical bones. Sixth lumbar vertebrae were tested in compression to

evaluate the properties of trabecular bones.

Three-point bending and vertebral compression were performed using an Instron 4465 materials

testing machine (Instron Canada Inc.). A pre-load of less than 1 N was applied to establish each

sample’s contact with the upper device. Further load was applied by a 100N cell load at a speed of

1mm/min and load versus time data were collected every 0.1 seconds by Lab View data acquisition

software (National Instruments Corp.; Austin, TX) until sample failure. Digital calipers were used to

calculate femoral midpoints. Femora were positioned between two supports, 6 mm apart, with the

posterior side facing downwards.

Vertebral body height and area were measured for data normalization. Adherent soft tissue that may

was removed from vertebrae allowing for the proximal and distal ends of the vertebra to be as flat as

possible for testing purposes. A fine layer of cyanoacrylate-based adhesive was then applied to a

metal plate to securely adhere the distal vertebrae so that its length was perpendicular to the plate.

Displacement was automatically calculated based on speed and time, and a load-displacement graph

was generated to evaluate bones’ structural mechanical properties such as ultimate load (N), energy

  60  

to failure (mJ) and stiffness (N/mm). Afterwards, data was normalized to the bone cross-sectional

and height for femora and vertebrae respectively. A stress-strain graph was generated to evaluate the

material properties of bones such as ultimate stress (MPa), failure strain (%), toughness (Mpa) and

modulus (MPa).

3.3.9 Static and Dynamic Histomorphometric analysis ��� Fifth lumbar vertebrae from 5, 6 and 11 month-old WT and FLNA-KO mice were isolated and fixed

in 70% ethanol. Samples were dehydrated in ascending concentrations of acetone followed by

ascending ratios of unpolymerized spur resin and acetone. Afterwards, bones were embedded in

blocks of spur resin and left to polymerize in a 60°C oven for 48 hours. Using a semiautomatic

microtome (Leica RM 2265), three 5-micron thick coronal sections were cut from each sample and

placed on gelatinized slides for Goldner’s trichrome staining [150]. One 7-micron thick coronal

section was cut, placed on gelatinized slides and left unstained for dynamic histomorphometric

analysis. Trabecular bone was analyzed using a 25x objective lens connected to a video camera

(Retiga 1300). Serial fields using the Leitz Bioquant morphometry system (Bioquant Nova Prime

version 6.50.10) were analyzed from each sample to determine the following static

histomorphometric structural parameters: trabecular bone volume (TBV (mm^3)), Tb.Th. (mm),

Tb.N. (#), Tb.Sp. (mm); and the formation parameters including osteoid volume (OV (mm^3)),

osteoid surface (OS (mm)) and osteoid thickness (O.Th. (mm)).

Dynamic histomorphometry was performed using fluorescence microscopy to measure the bone

labels generated by calcein-green injected mice prior to euthanasia. All mice in this study were given

two single intravenous injections of calcein green (0.6% calcein green; 30 mg/kg rodent) at 10 and 2

days before animal sacrifice. The single and double calcein-green labels were measured on trabecular

bone to calculate mineralizing surface (MS (mm)), percentage mineralizing surface (%.MS), mineral

apposition rate (MAR (um/day)) and bone formation rate (BFR (um/day)). All parameters are in

accordance with the histomorphometric nomenclature and definition of the American Society of

Bone Mineral Research (ASBMR) [150].

3.3.10 TRAP staining analysis ��� Fourth lumbar vertebrae were isolated from 5, 6 and 11 month-old WT and FLNA-KO mice and

fixed in 10% formalin. Samples were decalcified using EDTA (0.5 M, pH 7.4) at 4C, with daily

  61  

solution changes for 8 weeks. Complete decalcification was confirmed by faxitron imaging.

Decalcified samples were then processed (series of formalin, 70% ethanol, 90% ethanol, 100%

ethanol, 100% xylene, and paraffin) and embedded in bone-specific paraffin. 5-micron thick coronal

sections were cut serially using a Leica Reichert Jung 2030 microtome (Leica Microsystems Canada

Inc., Richmond Hill, Ontario) and mounted on Superfrost Plus (high section adhesion) glass slides.

The Acid Phosphatase Leukocyte kit and protocol (Procedure No. 386, Sigma-Aldrich Canada Ltd.,

Oakville, Ontario) were used to prepare and perform TRAP staining. Slides were incubated in the

TRAP stain at 37°C for 1 hour with periodic shaking. Following incubation, slides were washed and

counterstained with Acid hematoxylin. Slides were cover-slipped using a water-soluble mounting

media (Aqua Perm) and were allowed to dry overnight in a 37°C oven prior to analysis.

Three 5-micron thick coronal sections were cut from each sample and placed on glass slides for

Tartrate-Resistant Acid Phosphatase (TRAP) staining. Osteoclasts selectively express and stain

positive for the TRAP enzyme. The Leitz Bioquant morphomety system was used to quantify the

number of osteoclasts (Oc.N.), osteoclast surface (Oc.S. (mm)), percent osteoclast surface (%.Oc.S.),

number of osteoclasts per bone surface (N.Oc.BS) and number of osteoclasts per osteoclast surface

(N.Oc.OcS).

3.3.11 Statistical analysis For all analyses SPSS (version 22.0) was used. Two-way Analysis of Variance (ANOVA, general

linear model) was used to compare the measured parameters between and within groups. A p-value

of <0.1 was required for a trend between 5-month groups. A p value of <0.05 was required to

consider a significant difference in all other groups. All results are presented as mean ± standard

deviation (SD) in graphs and tables.

  62  

3.4 Results

3.4.1 In vitro OCG- phalloidin/DAPI staining and 3-D laser scanning confocal microscopy  Osteoclast analysis in vitro by labeling the F-actin cytoskeleton with phalloidin-Alexa488 and nuclei

using DAPI revealed a significant reduction in osteoclast surface area and number of nuclei in

FLNA-KO osteoclasts compared to WT osteoclasts (Figure 1a). Using a Picro Sirius Red

fluorochrome 3-D confocal laser scanning microscopy technique to compare osteoclasts and their

resultant resorption pits revealed that WT osteoclasts had increased volume of their Rpits compared

to FLNA-KO Rpits (Figure 1b).

3.4.2 Phenotype of 5-month old FLNA-KO mice at macro, cellular and tissue level Our in vitro results prompted us to evaluate the loss of FLNA in an in vivo mouse model. Micro-CT

was performed on 5-month old WT and FLNA-KO right femora and 6th lumbar vertebrae to assess

cortical and vertebral vBMD and geometry respectively. Micro-CT analysis revealed that mean

vBMD and cross-sectional thickness were significantly reduced in 5-month FLNA-KO femora

compared to WT femora (Figure 2a,b respectively). vBMD and Tb.N were significantly reduced

while Tb.Sp was significantly increased (p<0.1) in FLNA-KO compared to WT vertebrae (Figure

2c,d,e). Three point bending data revealed that ultimate load and stiffness were significantly

decreased in 5-month FLNA-KO femora compared to WT femora (Figure 3a,b). Upon normalization

of the load-displacement data, only ultimate stress was reduced in FLNA-KO femora compared to

WT femora (p<0.1). The data suggests that FLNA-KO femora are weaker than WT femora.

Vertebral compression data shows that post-yield displacement and plastic energy are increased in

FLNA-KO vertebrae compared to WT vertebrae suggesting that WT vertebrae are slightly more

brittle than FLNA-KO vertebrae (p<0.1). No significant differences were observed at the cellular

and tissue level for bone resorption and bone formation rates between WT and FLNA-KO vertebrae

(data not shown).

3.4.3 Phenotype of WT and FLNA-KO aging mice (5-monthà11 months) at macro, cellular and tissue levels Micro-CT was performed on 11-month-old WT and FLNA-KO femora and vertebrae to assess

cortical and vertebral vBMD and geometry, respectively. Results were compared to the 5-month

  63  

groups to determine whether skeletal aging differs between WT and FLNA-KO mice. Mean bone

area of femora was significantly increased in the WT aging group (5-months vs. 11 months) (Table

1). FLNA-KO femora showed significant increases in both vBMD and bone area at 11-months

(Table 1). Tb.Sp was significantly increased, while Tb.N was significantly decreased in both WT

and FLNA-KO vertebrae in 11-month groups (Table 2). Vertebral vBMD was significantly increased

in the 11-month FLNA-KO aging group (Table 2). Three-point bending data of femora revealed that

only normalized parameters including ultimate stress, and modulus were significantly decreased in

both WT and FLNA-KO 11-month groups (Table 3). No significant differences were reported

between 5 and 11 month groups in both the WT and FLNA-KO categories for vertebral compression,

TRAP, static and dynamic histomorphometry. Taken together, the data reveals that there are slight

skeletal aging changes in each of the WT and FLNA-KO groups but that FLNA deletion does not

impact on age related skeletal changes.

3.4.4 Phenotype of WT and FLNA-KO OVX mice at macro, cellular and tissue level Post-OVX, 11-month WT-OVX femora show expected significant decreases in vBMD compared to

the FLNA-WT sham group. No difference was obseved between FLNA-KO sham and FLNA-KO-

OVX femora (Figure 4a). Femoral geometry reveals that there is a significant effect due to OVX in

femora of the WT group for B.Ar and Cs.Th however this effect is not evident in FLNA-KO groups

(Figure 4b). Upon estrogen depletion, vBMD was significantly reduced in WT-OVX vertebrae

compared to WT vertebrae (Figure 4c). FLNA-KO-OVX vertebrae did not differ significantly from

the WT-OVX vertebrae as well as FLNA-KO sham vertebrae (Figure 4c). BV/TV and vertebral

geometry including Tb.Sp. were significantly reduced in WT-OVX vertebrae compared to WT sham

vertebrae (4d,e). These parameters did not differ significantly between the FLNA-KO-OVX and the

FLNA-KO sham vertebrae.

With OVX, ultimate load, fail displacement, energy to fail, and stiffness were significantly decreased

in 11-month WT-OVX femora (Figure 5a). Following normalization of the load-displacement data,

ultimate stress, fail strain and modulus were significantly decreased in WT-OVX femora compared to

the WT non-OVX group (Figure 5). The significant reduction in the structural and material

properties of femora due to OVX in the WT groups was not observed among FLNA-KO groups.

Structurally, FLNA-KO and FLNA-KO-OVX femora require more energy to fracture compared to

WT and WT-OVX femora (Figure 5). When structural properties were normalized to cross-sectional

  64  

area of femora, the failure strain was significantly increased in FLNA-KO and FLNA-KO-OVX mice

compared to WT and WT-OVX mice (Figure 5). These data would suggest that FLNA-KO and

FLNA-KO-OVX femora are more ductile than WT and WT-OVX femora. The more energy

dissipating (or toughening) mechanisms that exist, the more difficult it is to break a material.

As mentioned above, stiffness as well as Young’s modulus is significantly decreased in the WT-

OVX femora compared to the WT femora (Figure 5). FLNA-KO femora and FLNA-KO-OVX

femora remain unchanged from one another and are significantly different from the WT-OVX group.

Together, these data suggest that FLNA-KO cortices are protected from estrogen-mediated post-

menopausal susceptibility to fracture and are more ductile than WT.

Vertebral compression data shows expected significant decreases in the WT-OVX group for ultimate

load and stiffness. FLNA-KO vertebrae are not significantly different from FLNA-KO-OVX

vertebrae (Figure 6a,b respectively). It should be mentioned that for the aforementioned parameters

the FLNA-KO non-OVX group is still lower than the WT non-OVX group but is only significant at

p<0.1 which was the contingency for significance in the 5-month group.

TRAP staining revealed no significant differences in osteoclast parameters between WT, FLNA-KO,

WT-OVX and FLNA-KO-OVX 11-month groups. Since there is a 6-month gap between the 5 and

11-month group, we postulated that we were unable to observe changes due to OVX in the 11-month

group. Therefore a 6-month group was introduced in which mice were sacrificed 5 weeks post-OVX.

The 6-month group revealed an expected significant increase in number of osteoclasts and number of

osteoclasts per bone surface in WT-OVX vertebrae compared to WT vertebrae. No significant

differences were reported among FLNA-KO and FLNA-KO-OVX groups (Table 4).

In the 6-month group, static histomorphometry revealed significant increases in osteoid volume per

bone volume, osteoid surface per bone surface and osteoid width in 6-month-old WT-OVX vertebrae

compared to WT vertebrae (Table 5). Significant differences were not observed between FLNA-KO

and FLNA-KO-OVX vertebrae. The single and double calcein-green labels for dynamic

histomorphometry on un-decalcified histological sections revealed significant increases in the

following parameters: mineral apposition rate, inter-label distance, bone formation rate per bone

surface and bone formation rate per bone volume in WT-OVX trabeculae compared to WT

trabeculae. Significant differences were not observed between FLNA-KO and FLNA-KO-OVX

  65  

trabeculae (Table 6).

  66  

3.5 Discussion The results of this study suggest that FLNA-KO bones show a mild osteoporotic phenotype

compared to WT bones. Skeletal aging-related osteoporosis was not altered by the deletion of

FLNA. However, FLNA-KO bones were protected from estrogen-mediated post-menopausal

changes in bone properties. Due to the long duration of time between the 5 and 11-month group,

OVX-related differences in osteoclast and osteoblast behavior were not visible histologically in the

11-month-old OVX group. Since osteoclast and osteoblast activities must be observed 4-6 weeks

post-OVX, the 6-month group was introduced to observe changes in bone due to OVX at the tissue

level. When OVX is introduced, FLNA-KO bones appear to be protected from estrogen-mediated

post-menopausal osteoporosis. Our observations suggest that estrogen depletion has no significant

effect on bone turnover rates in bones lacking FLNA. Under normal circumstances, estrogen

depletion results in high turnover rates leading to decreases in bone mineral density, bone

microarchitecture and mechanical properties. These parameters remain intact in FLNA-KO due to

low bone resorption and formation rates.

The decrease in bone turnover rate in FLNA-KO bones may be explained by insufficient osteoclast

size and nuclei number due to insufficient OCG as observed through DAPI and phalloidin staining of

osteoclasts. Previous literature on FLNA in vitro, reported deficiencies in OCG in FLNA null

monocytes. Under normal in vitro plating densities, FLNA-null osteoclasts were smaller, less

numerous, and contained fewer nuclei per osteoclast. Quantification of osteoclasts in vivo in the

distal femoral head in male mice revealed similar results, illustrating the in vivo physiological

relevance of FLNA in OCG [115]. These findings prompted us to evaluate what the effect of FLNA

deletion would have in vivo on female mice in conjunction with estrogen depletion. The findings in

this study and the aforementioned in vitro study suggests that FLNA-/- null osteoclasts are smaller

and fewer in vivo, which may influence their resorptive capacity. Consequently, in vivo and in vitro

situations do not always complement each other. Based on previous in vitro findings, the 5-month

FLNA group should have displayed an osteopetrotic phenotype resulting in stronger bones with

higher bone mineral density. On the contrary, 5-month FLNA bones were osteopenic displaying

weaker bones and lower bone mineral density. The FLNA osteoporosis group, on the other hand, did

correlate positively with the in vitro findings.

  67  

Our study and existing literature reports that osteoclast size and number is correlated with osteoclast

function [144]. Rpit analysis correlated positively with osteoclast size and number of nuclei in this

study. In addition, Lees and Heersche showed that an increase in the number of larger osteoclasts

was paralleled with an increase in the size and depth of their Rpits. Additionally, when resorption

was expressed as the amount of bone resorbed per osteoclast nucleus, larger osteoclasts resorbed

more per nucleus, suggesting that large osteoclasts, as a population, are more effective resorbers than

small osteoclasts. Interestingly, when osteoclasts were plated at one-fifth the standard density, the

amount of bone resorbed per osteoclast decreased considerably, indicating that resorptive activity is

also affected by cell density of osteoclasts and/or of other cells present [115, 144]. Our findings

suggest that the smaller FLNA-KO osteoclasts have a reduced capacity for bone resorption in vivo.

This explains why the down-regulation of estrogen has no negative impact on bone microarchitecture

in mice lacking FLNA.

FLNA-KO cortices displayed significant increases in energy to fail and toughness, indicating that

they are tougher and more ductile than WT femora. The increased toughness and ductility indicates

that FLNA-KO cortices may have significant alterations in the biochemical and structural

composition of the mineral and organic components of the bone matrix compared to WT cortices.

This phenomenon was not evident in FLNA-KO trabeculae, however. Takahata et al. observed that

the mechanisms of plastic deformation (ductility) in bone are different between bending (cortical)

and compressive (trabecular) loading. Under tensile and shear stresses, as the bone yields and failure

initiates in the form of micro-cracks through the mineral phase, the collagen matrix can act as a

significant toughening mechanism against failure [151]. Additionally, since trabecular remodeling is

much quicker compared to cortical remodeling it is possible that this phenomenon is slightly masked

in the vertebrae of FLNA-KO bones

In terms of the organic components of the bone matrix, collagen maturity in bone is negatively

correlated with bone formation rates; an increase in collagen maturity reflects lower rates of bone

formation. The mature collagen matrix contains more pylorididine cross-links, which may be

associated with the ductility of bone that was observed in this study. Collagen cross-linking is an

important determinant of bone strength, especially in post-yield mechanical properties [11, 14, 35,

36]. The decrease in bone turnover rate in FLNA-KO bones due to insufficient OCG producing

mature osteoclasts may lead to the presence of more “mature” collagen in the FLNA-KO bones

  68  

resulting in larger plastic zones and tougher bones.

Another explanation for the lack of effect of OVX in FLNA-KO bones may be that FLNA and

estrogen regulate the same pathways. Specifically, estrogen and FLNA are both involved in

regulating IL-1 mediated TNF-induced OCG [70, 152, 153, 154]. IL-1 is a cytokine that mediates

the OCG effect of TNF by enhancing stromal cell expression of RANKL and directly stimulating

differentiation of osteoclast precursors [30,32]. RANKL is a TNF family member that is expressed

on the surface of osteoblasts and is essential for osteoclast differentiation. Binding of RANKL to its

receptor, RANK, activates a cascade of transcription factors that are known to be important for OCG.

RANK, similar to other TNF receptor family members, interacts with TNF receptor-associated

factors (TRAFs), which in turn act as adaptors to downstream signaling pathways. Of the six known

TRAFs, RANK interacts with TRAFs 1, 2, 3, and 5 in a membrane-distal region of the cytoplasmic

tail, and with TRAF6 at a distinct membrane-proximal binding motif. TRAF6 appears to be the most

crucial adapter for RANK signaling during OCG in vivo and in vitro, indicating that RANKL-

induced signaling is predominantly mediated by TRAF6 during OCG. Additionally, TRAF6 seems

to be the primary TRAF protein utilized in IL-1 signaling. Literature reveals that FLNA deficiency

down-regulates NF-κB. In a human melanoma cell line deficient in FLNA, TNF failed to activate

NF-κB. Reintroduction of FLNA into these cells restores the TNF response [70, 152, 153].

In the wild-type situation, estrogen suppresses cytokines including IL-1, IL-6, TNF, GM-CSF, M-

CSF, PGE2 and RANKL [153, 154]. In addition, estrogen prolongs osteoblast lifespan and increases

production of TGF-β [70, 153]. Conversely, estrogen depletion results in an increase in the

aforementioned cytokines, prolonged osteoclast lifespan and an increase in osteoblast apoptosis.

Although increased bone resorption induced by OVX may be explained by the cumulative effects of

these cytokines, IL-1 and TNF specifically play a prominent causal role in bone loss associated with

estrogen deficiency. A study done by Kimble et.al revealed that OVX increased the mononuclear cell

secretion of IL-1, TNF and the stromal cell production of M-CSF. OCG was decreased by in vivo

treatment of donor mice with either estrogen or a combination of the IL-1 inhibitor, IL-1 receptor

antagonist, and the TNF inhibitor [155, 156, 157].

The effect of the subsequent regulation of TNF and IL-1 via FLNA and estrogen indicate that

estrogen depleted FLNA-KO mice have down-regulated TNF and IL-1 expression via the absence of

  69  

FLNA and a simultaneous up-regulation of these cytokines via the absence of estrogen. This in

effect, results in a cancelling out effect whereby estrogen deficiency does not result in post-

menopausal loss of bone architecture and quality in mice lacking FLNA.

Finally, our in vitro results did not correlate positively with the in vivo results. Since there are three

FLN isoforms it is speculated that FLNs B or C may be compensating for the loss of FLNA. This

may explain the observed mild phenotype between WT and FLNA-KO bones and between aging

groups. Further research will need to explore these findings. Conclusively, FLNA-/- null bones

appear to be weaker at 5-months of age compared to their WT counterparts. Skeletal aging, however,

does not appear to differ between WT and FLNA-KO bones. Our most intriguing finding is that

FLNA-KO bones are protected from estrogen-mediated post-menopausal osteoporosis as well as

being more ductile than their WT counterparts.

  70  

3.6 Figures

3.6.1 Figure 1. Volume and Surface Area of Osteoclasts and Rpits  a)

b)

  71  

Figure 1- FLNA-KO osteoclasts are smaller and contain fewer nuclei than FLNA-KO osteoclasts. FLNA-KO resorption pits are smaller and shallower than WT pits. DAPI and phalloidin staining revealed a significant decrease in osteoclast size and number of nuclei in 5-month old FLNA-KO osteoclasts compared to WT osteoclasts (Figure 1a). Picro-Sirius red staining and 3d laser scanning confocal microscopy revealed a significant increase in 5-month-old WT resorption pit surface area and volume compared to FLNA-KO resorption pits (Figure 1b). Stars represent values that are statistically significant (p<0.05). Each box in the boxplot represents the interquartile range. The line in the box represents the median. The lower line of the box represents the 1st quartile and the upper line of the box represents the 3rd quartile. The whiskers represent variability outside the upper and lower quartiles. n=30 resorption pits per group and n=30 cells per group.

3.6.2 Figure 2. Phenotype of 5-MO FLNA-KO bones (Material and Structural Properties)

   

Figure 2- 5-month old FLNA-KO bones have reduced bone mineral density and geometric properties compared to 5-month old WT bones. Mean volumetric bone mineral density and cross-sectional thickness are significantly decreased in FLNA-KO femora (Figure 2a,b). Mean volumetric bone mineral density, bone volume/tissue volume and trabecular number are significantly decreased in FLNA-KO vertebrae (Figure 2c,d,e). Stars represent differences that are statistically significant (p<0.1). n=15 per group.  

 

  72  

3.6.3 Figure 3. Phenotype of FLNA-KO bones (Mechanical Properties)

 

Figure 3- 5-month FLNA-KO femora have reduced mechanical properties compared to 5-month WT femora. Mean ultimate load and stiffness are significantly decreased in FLNA-KO femora (Figure 3a,b). Stars represent differences that are statistically significant (p<0.1).  

3.6.4 Figure 4. Phenotype of FLNA-KO-OVX bones (Material and Structural Properties)

   

   

Figure 4- FLNA-KO femora and vertebrae are protected from post-menopausal changes in material and geometric properties of bone and WT-OVX femora show expected significant decreases in

  73  

vBMD in the 11-month group. No significance was reported between FLNA-KO sham and FLNA-KO-OVX femora. Femoral geometry reveals that there is a significant effect due to OVX in the WT group for B.Ar of femora, however FLNA-KO groups remain unchanged from one another (Figure 4a,b). vBMD was significantly reduced in 11 month-old WT-OVX vertebrae. FLNA-KO-OVX vertebrae did not differ significantly from the WT-OVX group as well as the FLNA-KO sham group (Figure 4c). The same phenomena were observed for vertebral geometry including BV/TV and Tb.Sp. (Figures 4d and e respectively). Stars represent differences that are statistically significant (p<0.05). n=15 per group. Shaded boxes represent FLNA-KO and open boxes represent WT.

3.6.5 Figure 5. Phenotype of FLNA-KO-OVX femora (Mechanical Properties)

Figure 5- FLNA-KO femora are protected from estrogen-mediated post-menopausal changes in bone mechanical properties. FLNA-KO/KO-OVX femora are more ductile than WT/WT-OVX bones. Ultimate load reveals a significant effect due to OVX in the WT group (Figure 5a). Structurally FLNA-KO femora require more energy to fracture and have a higher fail strain compared to WT femora (Figure 5b,c respectively). There is a significant effect due to OVX in the WT groups for both the structural and material properties of femora, which in turn are significantly different than the KO groups. Stiffness as well as young’s modulus is significantly decreased in the WT-OVX group (Figure 5d,e). FLNA-KO groups remain unchanged from one another. Stars represent differences that are statistically significant (p<0.05). n=15 per group. Shaded boxes represent FLNA-KO and open boxes represent WT.

  74  

3.6.6 Figure 6. Phenotype of FLNA-KO-OVX trabecular bone (Material and Structural Properties)  

Figure 6- FLNA-KO vertebrae are protected from estrogen-mediated post-menopausal changes in bone mechanical properties. Vertebral compression data shows expected decreases in the WT-OVX group for ultimate load and stiffness. FLNA-KO groups remain unchanged from one another (Figure 6a,b). Stars represent differences that are statistically significant (p<0.05). n=15 per group. Shaded boxes represent FLNA-KO and open boxes represent WT.

  75  

3.7 Tables

3.7.1 Table 1. Micro-CT of aging femora  

Parameter 5-month WT 11-month WT 5-month KO 11-month-KO p value WT p value-KO

vBMD (g/cm3)

1.16+/-0.035

1.18+/-0.050

1.13+/-0.036

1.17+/-0.044

0.37

0.01*

Bone Area

(mm2)

1.03+/-0.1

1.18+/-0.119

0.97+/-0.085

1.1+/-0.108

0.002*

0.001*

Cross-sectional-thickness (mm)

0.24+/-0.013

0.25+/-0.019

0.23+/-0.015

0.24+/-0.018

0.6

0.07

Table 1. Mean bone area was significantly increased in WT 11-month old femora. The FLNA-KO group showed significant increases in both vBMD and bone area at 11-months. Stars represent differences that are statistically significant (p<0.05). Table represents mean+/-SD.

3.7.2 Table 2. Micro-CT of aging vertebrae  

Parameters

5-month WT 11 month WT 5-month KO 11-month KO p value WT p value KO

vBMD (g/cm3)

0.23+/-0.033

0.21+/-0.081

0.19+/-0.068

0.15+/-0.068

0.59

0.07

BV/TV (%)

32.46+/-4.617

27.94+/-11.54

28.55+/-9.419

22.69+/-7.211

0.18

0.01*

Tb.Th. (mm)

0.043+/-0.004

0.04+/-0.011

0.04+/-0.012

0.04+/-0.012

0.77

0.92

Tb.Sp (mm)

0.09+/-0.015

0.12+/-0.056

0.10+/-0.0312

0.14+/-0.043

0.02*

0.0001*

Tb.N (#)

7.36+/-0.756

6.02+/-1.276

6.85+/-1.915

5.45+/-1.649

0.002*

0.0001*

Table 2. Trabecular separation was significantly increased, while trabecular number was significantly decreased in both WT and FLNA-KO 11-month old vertebrae. vBMD was significantly increased in 11-month FLNA-KO vertebrae. Stars represent differences that are statistically significant (p<0.05). Table represents mean+/-SD.

  76  

3.7.3 Table 3. Three-point bending of aging femora

Table 3. Three-point bending data of femora revealed that only normalized parameters including ultimate stress, and modulus were significantly decreased in both WT and FLNA-KO 11-month groups. All results are reported as means in tables. Stars represent differences that are statistically significant (p<0.05). Table represents mean+/-SD.

3.7.4 Table 4. TRAP histomorphometry of 6-month non-OVX vs. OVX vertebrae  

Parameters WT WT-OVX KO KO-OVX p value WT p value KO

Oc.S (mm)

1.13+/-0.57

1.89+/-0.47

0.62+/-0.39

0.97+/-0.12

0.02*

0.06

Oc.S/BS (%)

0.12+/-0.085

0.16+/-0.036

0.08+/-0.047

0.1+/-0.05

0.16

0.38

N.Oc. (#)

37.67+/-11.11

56.22+/-15.06

26.58+/-15.83

29.37+/-9.17

0.02*

0.72

N.Oc/BS (#/mm)

3.75+/-2.13

8.33+/-2.92

3.35+/-1.78

4.57+/-2.57

0.006*

0.3

Table 4. There was a significant increase in osteoclast surface number of osteoclasts, and number of osteoclasts per bone surface in the 6-month WT-OVX group. No significant differences were reported among 6-month FLNA-KO groups. Stars represent differences that are statistically

Parameters 5-month WT 11-month WT 5-month KO 11-month KO p value WT p value KO

Ultimate Load (N)

22.22+/3.923

20.76+/-3.871

18.14+/-2.62

18.67+/-5.28

0.31

0.75

Fail Disp. (mm)

0.27+/-0.115

0.24+/-0.105

0.29+/-0.09

0.33+/-0.13

0.75

0.38

Energy-to-fail

(mJ)

3.93+/-2.034

3.13+/-1.350

3.61+/-1.56

4.18+/-2.09

0.44

0.41

Stiffness (N/mm)

194.3+/-24.24

198.0+/-43.04

161.0+/-25.4

165.15+/-36.59

0.95

0.72

Ultimate Stress

(MPa)

131.4+/-21.37

98.51+/-24.35

119.1+/-13.78

93.68+/-19.99

0.001*

0.0001*

Fail Strain

(%)

5.79+/-2.548

5.35+/-2.431

6+/-1.480

7.48+/-3.03

0.93

0.12

Modulus

(MPa)

5504.8+/-984.1

4068.3+/-

1352.0

5168.4+/-985.6

3713.5+/-849.4

0.007*

0.0001*

Toughness

(MPa)

4.92

3.35

4.84

4.71

0.15

0.85

  77  

significant (p<0.05). Table represents mean+/-SD.

3.7.5 Table 5. Static histomorphometry of 6-month non-OVX vs. OVX vertebrae  

Parameters WT WT-OVX KO KO-OVX p value WT p value KO

OS/BV (%)

0.13+/-0.011

0.02+/-0.058

0.02+/-0.02

0.03+/-0.028

0.0001*

0.24

OS (mm)

0.73+/-0.289

0.42+/-0.39

0.41+/-0.27

0.33+/-0.199

0.09

0.44

OS/BS (%)

0.17+/-0.05

0.07+/-0.025

0.08+/-0.039

0.09+/-0.062

0.0001*

0.55

O.Wi (mm)

7.03+/-1.5

4.93+/-1.81

4.76+/-2.43

6.35+/-1.99

0.02*

0.13

Table 5. Static histomorphometry revealed significant decreases in osteoid volume/bone volume, osteoid surface/bone surface and osteoid width in the 6-month WT-OVX group. Significant differences were not observed among 6-month KO groups. Stars represent differences that are statistically significant (p<0.05). Table represents mean+/-SD.  

3.7.6 Table 6. Dynamic histomorphometry of 6-month non-OVX vs. OVX vertebrae  

Parameters WT WT-OVX KO KO-OVX p value WT p value KO

MS/BS

0.33+/-0.093

0.28+/-0.048

0.27+/-0.091

0.28+/-0.066

0.25

0.69

Interlab.Wi

19.12+/-6.84

10.3+/-4.29

12.86+/-7.94

10.33+/-7.24

0.02*

0.5

MAR

2.73+/-0.98

1.47+/-0.61

1.84+/-1.13

1.48+/-1.03

0.02*

0.5

Table 6. The single and double calcein-green labels for dynamic histomorphometry on undecalcified histological sections revealed significant decreases in the following parameters: mineralizing surface/bone surface, inter-label distance and mineral apposition rate in 6-month WT-OVX vertebrae. Significant differences were not observed among 6-month KO groups. All results are reported as means in tables and mean +/-SD in graphs. Stars represent data that is statistically significant (p<0.05). Table represents mean+/-SD.      

  78  

3.7 Acknowledgments The work was supported by a CIHR operating grant to Dr. Michael Glogauer. Stephanie Goldberg

participated in making substantial contributions to conception and design, acquisition of data and

analysis and interpretation of data. Stephanie Goldberg, Marc Grynpas and Michael Glogauer

participated in drafting the manuscript and revising it critically for important intellectual content.

Marc Grynpas and Michael Glogauer approved the final version of the submitted manuscript, and all

three authors agreed to be accountable for all aspects of the work in ensuring that questions related to

the accuracy or integrity of any part of the work were appropriately investigated and resolved. Judah

Glogauer participated in data acquisition.

  79  

Preface to Chapter 4  The results of the FLNA study described in chapter 3 showed that FLNA-KO osteoclasts were

smaller and produced volumetrically inferior Rpits compared to WT osteoclasts. In addition, FLNA-

KO bones were protected from estrogen-mediated post-menopausal osteoporosis. Intriguingly,

FLNA-KO cortical bones were more ductile than WT bones but trabecular bone did not display this

phenomenon. Although literature shows that trabecular bone undergoes higher turnover rates

compared to cortical bone, the mechanism is still unclear. Recent studies have raised some new

questions about the possible existence of different types of osteoclasts at different bone sites. The

novel method described in chapter 2 was used to identify whether osteoclasts from different sites

differed from one another with respect to volume and surface area of their Rpits. In vitro TRAP

staining was also performed on osteoclasts from calvarial, long bone and mandibular bone. The

results revealed that calvarial osteoclasts were larger compared to long bone and mandibular

osteoclasts. Rpit analysis showed significantly lower volume and surface area values in mandibular

bone compared to long and calvarial bone. Since in vitro and in vivo experiments often produce

distinct results, we attempted to quantify bone turnover in vivo using histology and

histomorphometry methods that were used in our FLNA study. Since the osteoporosis model used in

the FLNA study showed different results in different bone sites (long bone vs. vertebral bone), it was

important for us to use this model once again to investigate whether estrogen had an effect on bone

turnover in different bone sites.

  80  

Chapter 4

Heterogeneity of osteoclast activity and bone turnover in different skeletal sites in mice Authors: Goldberg S12, Glogauer M1, Grynpas MD2 1. Matrix dynamics group, Faculty of Dentistry- 150 College Street (Fitzgerald building), Toronto, Ontario M5S 3E2 2. Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital- 600 University Avenue, Toronto, Ontario, M5G 1X5 Submitted to: The Journal of Dental Research

4.1 Abstract Objective: To compare osteoclasts and bone turnover in the cranial and appendicular skeletons of

mice and determine the impact of estrogen depletion. Methods: In vitro osteoclastogenesis (OCG)

was performed on bone marrow derived osteoclasts from calvarial, mandibular and long bone. On

day 6, osteoclasts were stained for TRAP to compare the planar surface area of osteoclasts from the

different bone sites. In vitro quantification of osteoclast resorption pit volume and surface area from

different bone sites was achieved using Picrosirius red staining and confocal microscopy. In vivo

TRAP, static and dynamic histomorphometric analyses were performed on 5-month-old calvarial,

long bone and mandibular trabecular bone to compare bone resorption and formation rates

respectively. Mice were ovariectomized (OVX) at 5 months of age and sacrificed at 6 months of age

to establish an osteoporosis model for differences in osteoclasts arising at different bone sites and to

monitor the changes in bone resorption and formation rates in the different bone sites upon estrogen

depletion. Results: In vitro, TRAP-stained calvarial osteoclasts are larger compared to long bone and

mandibular osteoclasts. Mandibular resorption pits were smaller and had lower volume values

compared to long bone and calvarial bone resorption pits. In vivo analysis showed significantly

higher bone formation rates in calvarial trabecular bone compared to long bone and mandibular

trabecular bone. Turnover was enhanced upon estrogen depletion in calvarial trabecular bone.

Resorption was increased without a corresponding increase in bone formation in the trabecular

metaphysis of long bone. Mandibular trabecular bones do not appear to be affected by OVX.

  81  

Conclusion: The cranial and appendicular skeletons differ from one another in that osteoclasts from

calvarial bone have the highest resorptive capacity which is coupled to bone formation both pre and

post-OVX. Mandibular bones show the lowest turnover rates and are not affected by OVX.

Keywords: OCG, TRAP, estrogen, bone turnover, resorption pits

  82  

4.2 Introduction Osteoclasts are multinucleated bone-resorbing cells that arise from the monocyte-macrophage

lineage. Bone resorption is critical in the maintenance and repair of bones in the mammalian

skeleton. Osteoclasts in cortical bone are located on bone surfaces in Howship’s lacunae also known

as Rpits [80]. The lacunae are the end products of the dissolution of bone mineral via lysosomal

enzymes including carbonic anhydrase, matrix metalloproteinases (MMPs), TRAP and CTSK [80,

158]. Recent studies have raised some new questions about the possible existence of different types

of osteoclasts at different bone sites. It is unclear however, how osteoclasts arise at different bone

locations. It may involve differences in osteoclast precursor populations and/or different priming by

the local bone itself or a combination of the two [159]. Since bone remodeling is a homeostatic

system consisting of bone resorption coupled to bone formation, it could be the result of local

priming by cells of the osteoblast lineage [21, 159]. Another important question is whether disparate

embryonic tissue origins impart variable osteogenic potential. For example, bones of the skull arise

from the neural crest and paraxial mesoderm and undergo intra-membranous ossification. Long

bones on the other hand arise from embryonic mesoderm and undergo both intramembranous and

endochondral ossification [21].

The possible existence of different subsets of osteoclasts arose from differences in osteoclast

activities between the axial skeleton and the head region, as evidenced by lack of tooth eruption in

osteopetrotic rodents. Tooth eruption requires time-limited recruitment of osteoclast precursors, their

local formation and subsequent activation. Diseases affecting osteoclast differentiation, or osteoclast

function are not the only ones that can result in reduced or delayed tooth eruption, but also conditions

in which osteoclast precursor recruitment is insufficient [159]. For example, in RANKL−/− mice,

that lack an essential osteoclast differentiation factor, a general osteopetrotic phenotype is seen [161].

However, when Odgren et al. performed rescue experiments with CD4-driven RANKL they noticed

that the teeth did not erupt in the rescued mice. Their findings indicated continued osteoclastic

under-activity in the jaw while at the same time osteopetrosis in the long bones was resolved,

indicating normalization of osteoclast function. These findings suggest that osteoclasts at different

bone sites respond differently to rescue by RANKL presented by CD4-positive immune cells [161].

Many studies focus on comparing long bone to calvarial bone or long bone to mandibular bone in

  83  

vitro. Recent literature comparing long bone to calvarial bone suggests that calvarial osteoclasts are

larger and express a higher level of TRAP compared to long bones and use MMPs as well as

cathepsin K for resorption. Long bone osteoclasts express lower TRAP levels compared to calvarial

cells and primarily use cathepsin K for bone resorption [22]. In addition, anion exchanger2 (AE2)

proved to be essential for resorption by long bone osteoclasts but less important for calvarial

resorption- they have sodium transporter slc4a4 to compensate for loss of AE2 [23]. Resorptive

capacity has not been compared in these bone sites at both the in vitro and in vivo levels. The

determination of osteoclast size via in vitro TRAP staining is not sufficient enough to prove

increased resorptive capacity. We used an Rpit analysis technique to provide the complete in vitro

picture with respect to osteoclast activity in these bone sites. Further determination of osteoclast

activity is provided through in vivo TRAP histology and static and dynamic histomorphometry. is

invaluble [22, 23, 159, 161]. This is the first study that has attempted to compare osteoclasts and

their resorptive capacity in all three bones at both the cellular and tissue level, in both the absence

and presence of estrogen. The goal of this study is to further assess differences in osteoclasts arising

from different bone locations to eventually develop pharmacotherapeutics for osteoporosis that are

targeted to specific locations where they will be most beneficial.

  84  

4.3 Materials and Methods

4.3.1 Animals All procedures described were performed in accordance with the Guide for the Humane Use and Care

of Laboratory Animals and were approved by the University of Toronto Animal Care Committee.

For the purposes of this study, 6 groups of C57BL/6 mice containing 5-month-old WT- calvarial,

femoral and mandibular bone (groups 1, 2 and 3) as well as 6 month old post-OVX calvarial, femoral

and mandibular bone (groups 4, 5 and 6) were used. Each group consists of 12 mice for a total of 72

mice to establish statistical significance and guard against losses. Mice were OVX at 5 months of

age and then sacrificed at 6 months to establish an osteoporosis model to determine bone turnover

rates at different skeletal sites. In order to investigate whether bone turnover rates differ among

different skeletal sites pre and post-menopause, 5-month-old WT calvariae, long bones and

mandibles were compared to each other and to their 6 month-old OVX counterparts.

Two and ten days before euthanasia, all mice were injected in their peritoneal cavities with calcein

green (30 mg/kg) for bone dynamic histomorphometric analysis [150]. Animals were sacrificed by

C02 asphyxiation, following animal care protocol established by University of Toronto Animal Care

Committee.

4.3.2 In vitro OCG and TRAP staining Femur bones, calvarial bone and mandibular bone from 5-month-old mice were dissected aseptically

under a laminar flow hood. See section 2.3.2 & 2.3.3. On day 6, cells were washed twice with PBS,

fixed with 4% (PFA), and stained for TRAP. The number of TRAP osteoclasts and number of nuclei

within these osteoclasts were counted in 10 random fields of view (FOV).

4.3.3 Pirco Sirius red staining, confocal laser scanning microscopy and analysis

See sections 2.3.4, 2.3.6, 2.3.7

4.3.4 Undecalcified Histomorphometry Calvariae, long bone and mandibles from 5 month old and 6 month-old WT and OVX mice were

isolated and fixed in 70% ethanol. See section 3.3.9.

  85  

4.3.5 Decalcified Histology Calvariae, long bone and mandibles from 5 month old and 6 month-old WT and OVX mice were

fixed in 10% formalin. See section 3.3.10.

For all histomorphometric analyses, scanned images were opened in ImageJ and a region of interest

of area of 70,000um2 was traced within the cortices of each bone to obtain trabecular bone. The

image was then opened in the Leitz Bioquant morphometry system to analyze the area of interest.

Traced images of calvarial, long bone and mandibular bone are shown below.

Calvaria Long Bone Mandible

4.3.6 Statistical analysis For all analyses SPSS (version 22.0) was used. One-way Analysis of Variance (ANOVA, general

linear model) and Post-hoc multiple comparisons LSD test was used to compare the measured

parameters between the three bone sites and between pre and post-OVX groups. A p-value of <0.05

was required to consider a significant difference. All results are presented as mean ± standard

deviation (SD).

 

  86  

4.4 Results

1. Pre-OVX Results

4.4.1 In vitro TRAP staining of calvarial, long and mandibular osteoclasts TRAP-stained calvarial osteoclasts displayed significantly larger surface area compared to long and

mandibular osteoclasts. TRAP stained surface area did not differ significantly between long bone

and mandibular osteoclasts (Figure 1a). Rpit analysis reveals that long bone and calvarial Rpits have

significantly larger surface area and volume values compared to mandibular Rpits (Figure 1b, 1c).

4.4.2 Bone resorption Pre-OVX, mandibular trabecular bone showed significantly lower Oc.S/BS and N.Oc./BS compared

to the trabecular metaphysis in long bone and calvarial trabecular bone (Figures 2a&b respectively).

Calvarial trabecular bone and the trabecular metaphysis in long bone did not differ significantly from

another for both parameters (Figures 2a&b).

4.4.3 Bone formation and bone architecture Pre-OVX, calvarial trabecular bone showed significantly greater OS/BS compared to the trabecular

metaphysis in long bone (Figure 3a). There are no significant differences between the three bone

sites for O.Wi (Figure 3b). BV/TV and Tb.Th. were significantly increased in calvarial trabecular

bone compared to the trabecular metaphysis in long bone and mandibular trabecular bone (Table 1).

Tb.Sp was significantly decreased in calvarial trabecular bone compared to mandibular trabecular

bone and the trabecular metaphysis in long bone (Table 1).

4.4.4 Bone formation rate  Pre-OVX, calvarial trabecular bone showed significant increases in BFR/BS compared to the

trabecular metaphysis in long bone and mandibular trabecular bone (Figure 4a). Calvarial trabecular

bone displayed a significant increase in I.Wi compared to the trabecular metaphysis in long bone

(Figure 4b).

  87  

2. Post-OVX Results

4.4.5 Bone resorption Post-OVX, since mandibular bone was not affected by OVX, calvarial bone and the trabecular

metaphysis of long bone were compared to each other. Calvarial trabecular bone has significantly

higher N.Oc/BS compared to the trabecular metaphysis in long bone (Figures 5a&b).

4.4.6 Bone formation and bone architecture Post-OVX, calvarial trabecular bone shows significant increases in OS/BS and O.Wi compared to the

trabecular metaphysis in long bone (Figure 6a). Calvarial trabecular bone shows significant increases

in OS/BS and O.Wi compared to its non-OVX counterparts (Figure 6a&b). BV/TV and Tb.Th were

significantly increased and Tb.Sp was significantly decreased in calvarial trabecular bone compared

to the trabecular metaphysis in long bone (Table 2). Overall, post-OVX, calvarial trabecular bone

shows higher bone formation rates compared to the trabecular metaphysis in long bone.

Bone formation rate was significantly higher in calvarial trabecular bone compared to the trabecular

metaphysis in long bone for BFR/BS and I.Wi (Figures 7a&b). The trabecular metaphysis of long

bone does not differ significantly from its non-OVX counterpart (Figures 7a&b). Post-OVX,

calvarial trabecular bone differs significantly from its non-OVX counterpart for both parameters

(Figures 7a&b). Overall, post-OVX, calvarial trabecular bone shows significantly higher bone

formation rates compared to long and mandibular trabecular bone.

  88  

4.5 Discussion

The results of this study confirm that the cranial and appendicular skeletons differ with respect to

osteoclast activity and bone turnover rates. In vitro TRAP stained calvarial osteoclasts were larger

compared to long bone and mandibular osteoclasts. Calvarial and long bone displayed higher surface

area and volume values of their Rpits compared to mandibular osteoclasts. In vivo, calvarial bone

displayed higher bone resorption rates compared to mandibular bone and increased bone formation

rates compared to both mandibular and long bone (Table 3).

4.5.1. Bone turnover heterogeneity in calvarial, mandibular and long bone pre-OVX

Based on the literature comparing different bone sites, it is not surprising that the mandible, calvaria

and long bone display distinct turnover rates [23, 163-165]. Existing research on osteoclasts

heterogeneity focuses on TRAP staining of osteoclasts and evaluating the proteins involved in bone

degradation. Everts et al. demonstrated that calvarial osteoclasts use MMPs as well as cysteine

proteinases (e.g., CTSK) for resorption, whereas long bone osteoclasts use primarily cysteine

proteinases. Moreover, as we have shown, TRAP stained calvarial osteoclasts were larger than long

bone osteoclasts. In addition, Everts et al. showed that AE-2, a protein essential for regulating the

intracellular pH of cells during extracellular acidification by osteoclasts proved to be essential for

resorption by long bone osteoclasts but less important for resorption by calvarial osteoclasts [23].

Calvarial osteoclasts—but not long-bone osteoclasts—possess a sodium-dependent bicarbonate

(slc4a4) transporting activity, which might compensate for the absence of AE-2 in calvarial

osteoclasts of AE-2-/- mice [23]. Thus, although calvarial and long bones display different modes of

resorption, there are compensatory molecules such as slc4a4 that may result in similar resorption

rates between the two bone sites. This may explain why bone resorption pre-OVX did not deviate

between calvarial and long bone.

Osteoclast heterogeneity can be seen as an adaptation to different bones. Calvaria and long bones

differ in the mode of ossification and, consequently, in the composition of their matrix, regarding not

only the amount of collagen and noncollagenous protein but also the level of collagen cross-linking

[163, 164]. The differences in the bone matrix are related to the distinct proteolytic pathways

osteoclasts used to degrade bone. Everts et al. showed that flat bones (calvariae) contained more

  89  

soluble collagen, pigment epithelium derived factor and osteoglycin; whereas long bones expressed

more chondrocalcin, thrombospondin, fetuin, secreted phosphoprotein 24, and thrombin [163, 164].

It was therefore suggested by Everts that differences in protein composition of flat and long bones

lead to functional differences in formation, resorption and mechanical properties of these bone types.

In addition, osteoclasts from distinct bone sites seem to respond differently to some cytokines. In

RANKL-deficient mice, the effect of transgenically induced RANKL expression by T and B cells

promoted osteoclast formation in long bones but not in the jaw. Therefore, the failure of RANKL-

expressing T and B cells to activate bone resorption in the jaws is a site-specific phenomenon [23].

A possible explanation for the differences observed between jaw and long bone osteoclasts could be

the skeletal site-specific osteogenic properties of the bone marrow cells in orofacial bones (maxilla

and mandible) compared to other bones based on their different embryological origins. The orofacial

bones are formed from ectomesenchymal cells (originated from the association of neural crest cells

and mesoderm) while other bones develop from mesoderm [165]. The distinct origin may, in part,

explain the differences between jaw and long bone with respect to turnover properties observed in

our study.

4.5.2. Bone turnover heterogeneity in calvarial, mandibular and long bone post-OVX

In vivo, upon estrogen depletion (post-OVX), calvarial trabecular bone showed increased bone

resorption as well as formation compared to the trabecular bone of mandibles and the trabecular

metaphysis of long bone. The trabecular metaphysis of long bone showed increased bone resorption

upon estrogen depletion but to a lesser extent than calvarial trabecular bone. On the other hand, bone

formation rate was static in the trabecular metaphysis of long bone resulting in an osteoporotic

phenotype (Table 3). Mandibular trabecular bone did not respond at all to OVX, which suggests that

mandibular trabecular bone is protected from estrogen-mediated post-menopausal increases in bone

turnover. This is consistent with the work of Mavropoulos et al, who showed that mandibular

alveolar (trabecular) bone in rats is less sensitive than the proximal tibia to estrogen-deficiency-

induced osteoporosis [19]. They found that mastication lessened sensitivity to estrogen-deficiency

possibly due to the mechanical loading of the mandibular bone during mastication, which has been

shown to influence the mandibular bone density and micro-architecture in the growing the adult rat

[19, 20].

  90  

Calvarial bone undergoes the opposite phenomenon whereby bone resorption and formation are

significantly increased upon estrogen depletion (Table 3). Because bone resorption is so high in

calvarial bone, it would be expected that this bone is fragile and weak, consistent with an

osteoporotic phenotype. This is not the case, however, since bone formation rates are equal, if not

higher than bone resorption rates. The coupling mechanism in calvarial bone seems to be its most

unique attribute. In addition, it is postulated that calvarial bone has elevated levels of growth factors

that are important for bone formation and regeneration. Finkelman et al showed that calvarial bone

grafts may have greater survival as donor tissue than bone from other sites [24]. Bone contains

growth factors that may play an important roles in the regulation of bone repair. Finkelman et al

proposed that bone from calvaria might be enriched in one or more growth factors. It was concluded

that the increased concentrations of growth factors IGF-1 and TGF-β in calvarial bone may lead to a

greater capacity for bone repair and graft retention. The skull may harbor elevated levels of these

growth factors in order to protect one of the body’s most important organs: the brain [24].

Bone turnover in the trabecular metaphysis in long bone lies somewhere in between the high bone

turnover rates of calvarial bone and the extremely low bone turnover rates of mandibular trabecular

bone (Table 3). The high bone resorption rate with bone formation rates that are not equally coupled

to one another would typically lead to an osteoporotic phenotype [162]. This is not surprising since

the trabecular metaphysis of long bone is the site of hip fractures most commonly seen in elderly

osteoporotic women. Elevated resorption rates in the trabecular metaphysis of long bone would lead

to low bone mass. Our study confirmed deterioration of bone microarchitecture as evidenced by the

low BV/TV and high trabecular separation values, which can lead to fragility and risk of fracture

(Table 2).

It is also likely that calvarial osteoclasts and osteoclasts originating from long bone differ in

osteoblast signaling. Several stored growth factors are released upon the dissolution of the bone

matrix, including BMPs, FGFs and TGF β, which are likely responsible for the recruitment of the

osteoblasts in the reabsorbed area [26-28]. As mentioned above, calvarial bone is postulated to

contain higher than normal levels of TGF β and growth factors which may lead to the higher than

normal levels of bone resorption seen in this bone site. Therefore, the results obtained in the present

study provide strong support for the view of osteoclast-bone site heterogeneity, showing that bone

from different skeletal locations (jaw, calvaria and long bone) result in different turnover dynamics

  91  

and possible functional differences between osteoclasts especially upon estrogen depletion.

 

  92  

4.6 Figures

1. Pre-OVX Figures

4.6.1 Figure 1. in vitro OCG

a)

 

Figure 1- In vitro OCG and Rpit analysis- Calvarial osteoclasts had significantly higher in vitro TRAP stained surface area compared to mandibular and long bone (Figure 1a). Calvarial and long bone Rpits have higher surface area and volume values compared to mandibular Rpits (Figure 1b, c respectively) (p<0.05). Each box in the boxplot represents the interquartile range. The line in the box represents the median. The lower line of the box represents the 1st quartile and the upper line of the box represents the 3rd quartile. The whiskers represent variability outside the upper and lower quartiles. n=30 cells, n=210 resorption pits.

  93  

4.6.2 Figure 2. in vivo decalcified histology

Figure 2. In vivo TRAP histology- Mandibular bone shows significant decreases in mean osteoclast surface/bone surface and mean number of osteoclasts/bone surface compared to calvarial and long bone (Figure 2a&b). Stars represent significant differences at p<0.05. n=15 per site.

4.6.3. Figure 3. in vivo undecalcified histomorphometry (bone formation)

Figure 3. In vivo static histomorphometry. Calvarial bone shows a significant increase in mean osteoid surface/bone surface compared to long bone (Figure 3a). Stars represent significance at p<0.05. n=15 per site.

  94  

4.6.4. Figure 4. in vivo undecalcified histomorphometry (bone formation rate)

Figure 4- In vivo dynamic histomorphometry- mean bone formation rate/bone surface and mean inter-label width was significantly increased in calvarial bone (Figures 4a&b). Stars represent significance were p<0.05. n=15 per site.

2. Post-OVX Figures

4.6.5. Figure 5. in vivo decalcified histology

Figure 5. In vivo TRAP histology- Calvarial trabecular bone shows highest rate of bone resorption post-OVX compared to long and mandibular trabecular bone. Long bone and calvarial bone showed significant increases in both osteoclast surface/bone surface and number of osteoclasts/bone surface compared to their non-OVX counterparts (Figure 5a&b). Post-OVX, osteoclast surface/bone surface was significantly increased in long and calvarial bone compared to mandibular bone (a vs. b). Number of osteoclasts/bone surface was significantly increased in long bone compared to mandibular bone (a vs. b) and significantly increased in calvarial bone compared to long and mandibular bone (c vs. a&b). Different letters within graphs as well as stars represent statistical significance at p<0.05. n=15 per group. Shaded boxes represent sites that have been OVX. Open boxes represent sites that

  95  

have not been OVX.

4.6.6. Figure 6. in vivo undecalcified histomorphometry

Figure 6- In vivo static histomorphometry- Calvarial trabecular bone shows significantly higher bone formation rates as well as significantly increased osteoid width post-OVX compared to long and mandibular trabecular bone. Post-OVX, calvarial bone shows significant increases in OS/BS and osteoid width compared to mandibular and long bone (Figure 6a, a vs. b). Calvarial bone show significant increases in OS/BS and osteoid width compared to its non-OVX counterparts (Figure 7a). Different letters and stars represent significant differences at p<0.05.

4.6.7 Figure 7. in vivo undecalcified histomorphometry

Figure 7. In vivo Dynamic Histomorphometry- calvarial trabecular bone shows significantly higher bone formation rates as well as significantly increased inter-label width post-OVX compared to long and mandibular trabecular bone. Calvarial bone showed a significant increase in BFR/BS and interlabel width compared to its non-OVX counterparts (Figure 7a&b). Post-OVX, calvarial bone showed significant increases in bone formation rate/bone surface compared to mandibular and long bone (Figure 7a). Calvarial bone also showed significant increases in inter-label width compared to mandibular bone, which was significantly increased from long bone (Figure 7b, a vs. b vs. c). Different letters and stars represent significant differences at p<0.05.

  96  

4.7 Tables

4.7.1 Table 1. Pre-OVX trabecular architecture derived from undecalcified histomorphometry

Calvarial Bone Mandibular Bone Long Bone BV/TV (%)

0.806+/-0.097

0.527+/-0.104

0.621+/-0.172

Tb.Th (mm)

0.156+/-0.014

0.093+/-0.029

0.091+/-0.038

Tb.Sp (mm)

0.001+/-0.01

0.021+/-0.0104

0.0103+/-0.012

Table 1. Calvarial bone shows significantly higher BV/TV, Tb.Th and significantly decreased Tb.Sp compared to mandibular and long bone. Values in table represent mean+/-SD.

4.7.2 Table 2. Post-OVX trabecular architecture derived from undecalcified histomorphometry

Calvarial Bone Mandibular Bone Long Bone BV/TV (%)

0.408+/-0.035

0.65+/-0.105

0.391+/-0.126

Tb.Th (mm)

0.094+/-0.033

0.111+/-0.009

0.072+/-0.024

Tb.Sp (mm)

0.036+/-0.007

0.012+/-0.100

0.029+/-0.012

Table 2. Calvarial bone has significantly higher BV/TV, Tb.Th. and significantly lower Tb.Sp compared to the trabecular metaphysis in long bone. Values in table represent mean+/SD.

  97  

4.7.3 Table 3. Summary table

Calvarial Bone Mandibular Bone Long Bone Pre-OVX Resorption

++ - ++

Pre-OVX Formation

++ + +

Post-OVX Resorption

+++ - +++

Post-OVX Formation

+++ + +

  98  

Chapter 5  

5.1 Summary and Conclusions

5.1.1 Phenotype of FLNA-KO mice In vivo structural, material and mechanical properties of bone were analyzed in 5-month and 11-

month FLNA-KO mouse bones. Structural and material properties were analyzed in a 6-month

FLNA-KO to observe bone formation and resorption post-OVX. The 5-month FLNA-KO group

displayed a mild osteoporotic phenotype whereby mechanical properties and material properties

were slightly lower than the WT group. This observation is peculiar since it would be expected

based on in vitro data that 5-month FLNA-KO bones would be osteopetrotic. Animals harboring

mutations that affect osteoclast differentiation typically exhibit osteopetrotic phenotypes, generally

with increased trabecular thickness, decreased marrow spaces and increased bone densities. For

example, murine models in which loss-of function or null mutations in genes including, but not

limited to RANK/ RANKL, M-CSF, c-Src, NFATc1, NFκB, and β3 integrin give rise to mice that

exhibit osteopetrosis. In vitro 3-D confocal laser scanning osteoclast, nuclei and Rpit data revealed

that FLNA-KO osteoclasts were smaller with fewer nuclei with low volumetric and surface area

values for their Rpits. Osteoclasts and their Rpits were smaller and shallower compared to WT

osteoclasts and their Rpits since pre-osteoclasts lacking FLNA would be unable to migrate

efficiently to form large, mature multinucleated osteoclasts capable of efficiently resorbing bone.

Therefore, osteoclasts in an FLNA-KO mouse wouldn’t be able to resorb bone efficiently and the

result of this would be an osteopetrotic phenotype with denser and tougher bone.

Since the regulation of bone remodeling to maintain a state of homeostasis is determined by the

balanced activities of bone resorbing osteoclasts and bone producing osteoblasts, a decrease in

concurrent osteoblastic activity with that observed in osteoclast activity may explain the phenotype

of the FLNA-KO mice in the 5-month group. Leung et al. showed that osteoblastic activity was

significantly decreased in FLNA-KO mice as determined by bone formation markers such as serum

osteocalcin and alkaline phosphatase activity of whole bone marrow cells in a mineralizing medium

[115]. Indirect evidence for the inhibition of osteoblast formation and/or function came from the

  99  

increased adipogenesis in these mice. Differentiation from mesenchymal progenitors into

adipocytes is regulated by peroxisome proliferator-activated receptor (PPAR) γ and C/EBPs

whereas differentiation into osteoblasts is regulated by Runx2 and Osterix [115, 169-172]. Since

adipocytes and osteoblasts are produced by the same mesenchymal progenitor cells, inhibition of

the signals that direct osteoblast differentiation can push the equilibrium toward adipocyte

differentiation.

In addition, it is not uncommon that in vitro and in vivo studies do not match up. One of the abiding

weaknesses of in vitro experiments is that in some cases they fail to replicate the precise conditions

of an organism. Because of this, in vitro studies may lead to results that do not correspond to the

circumstances occurring around a living organism.

In contrast, the in vitro 3-D confocal Rpit method correlates with the in vivo situation observed in

post-OVX 11-month FLNA-KO mice. Based on what was observed in this group, the expectation

was that osteoclasts would be smaller with fewer nuclei and therefore unable to efficiently resorb

bone. The 11-month post-OVX FLNA-KO group did not respond to estrogen depletion. Under

normal circumstances estrogen depletion produces a higher osteoclast: osteoblast ratio. This results

in higher rates of bone resorption, which leads to weaker bones and reduced bone

microarchitecture. The WT group showed the typical effects of estrogen depletion while the

FLNA-KO group did not. If osteoclasts were insufficient in their capability to resorb bone then

estrogen depletion would not have as great of an effect on these premature osteoclasts as it would

on fully functional mature osteoclasts.

In addition to the 5-month and osteoporosis model we endeavored to study whether aging had a

different effect on WT and FLNA-KO bones. According to our results, aging did not differ

significantly between the WT and FLNA-KO groups. In the first three decades of life, bone

turnover is coupled tightly to maintain a steady state between bone resorption and bone formation.

After reaching peak bone mass, bone turnover continues at a slower rate as evidenced by a rapid

decline in biochemical measures of bone remodeling with the predominance of bone resorption over

bone formation [166, 167, 168]. Five-month FLNA-KO bones showed slight decreases in bone

mineral density and mechanics compared to WT bones. The slower turnover rates observed in the

FLNA-KO 5-month group was observed 6 months later in the 11-month group. Thus, both groups

  100  

followed their paths linearly with respect to how their bones aged. FLNA is well recognized as a

powerful actin-crosslinking protein, and in many cell types it is required for controlling cell

processes involving the actin cytoskeleton. Structural, sequence, and binding studies have shown

that not only does it bind actin filaments, it is a promiscuous binding partner for over 20 identified

proteins, including Rac, Cdc42, RhoA, and their regulatory GEFs and GAPs [7]. Using a mouse

model in which FLNA was deleted specifically in granulocytes, we show that FLNA is required for

OCG since mice lacking FLNA show decreased turnover rates due to the incomplete formation of

mature multinucleated osteoclasts. Investigations into the crosstalk between osteoblasts and

osteoclasts will shed light on how FLNA mutations in osteoclast precursors signal to osteoblasts to

regulate their function.

5.1.2 Osteoclasts and bone turnover from different skeletal sites  Osteoclasts in calvarial, long bone and mandibular bone were compared to each other via TRAP

staining and Rpit analysis in vitro. TRAP stained calvarial osteoclasts were larger compared to

long and mandibular bone. Long bone and mandibular bone however, did not differ with respect to

osteoclast size. This result coincides with recent literature comparing TRAP in calvarial bone vs.

long bone and long bone vs. mandibular bone. Everts et. al showed that jaw and long bone

precursors differed with respect to the dynamics of OCG; long bone cultures osteoclasts were

formed faster and expressed higher levels of TRAP than in the jaw cultures [160]. However, the

jaw cultures did catch up to the long bone cultures. From day 4 to day 6, a threefold increase in the

number of TRAP+ multinucleated cells took place in the jaw cultures and the total number of

osteoclast-like cells formed from jaw and long bone precursors was no longer different at the end of

the culture period (day 6). However, on day 6, the large cells (with more than 10 nuclei) were more

frequently seen in the jaw than in the long bone cultures. This is in contrast to what was observed

on day 4, when almost no large TRAP+ cells were found in the jaw cultures. Therefore, their data

indicated that the initial formation of osteoclasts from the jaw precursors was slower but they catch

up during the onset of culturing to eventually form larger osteoclasts. When calvarial osteoclasts

were compared to long bone osteoclasts, Everts et. al showed considerable differences in TRAP

activity between calvarial and long bones [160]. Calvarial bones contained a 25-fold higher level of

activity than long bones. Osteoclasts isolated from the two types of bone revealed that calvarial

  101  

osteoclasts expressed higher enzyme activity as well as a higher level of mRNA for the enzyme.

Analysis of TRAP-deficient mice revealed higher levels of nondigested bone matrix components in

and around calvarial osteoclasts than in long bone osteoclasts [22, 159].

Bone turnover, in the presence of estrogen, followed this pattern whereby calvarial bone displayed

higher turnover rates compared to long bone and especially mandibular bone. The larger calvarial

osteoclasts are able to resorb bone more efficiently than long bone and mandibular osteoclasts.

Studies involving rabbit long bone osteoclasts have shown that osteoclast size is associated with

increased resorptive activity or efficiency [173, 174]. Large osteoclasts proved to resorb 2.5 times

more per cell than small osteoclasts, but the amount resorbed per nucleus was the same for the two

categories [174]. In a study investigating rat long bone osteoclasts, the small osteoclasts exhibited a

lower capacity for degradation of bone matrix and a lower expression of MMP-9 and cathepsin K

compared to the large osteoclasts. Long bone displayed higher bone resorption rates and larger and

deeper Rpits compared to mandibular bone. This may be due to the fact  that long bones undergo

faster OCG. Early osteoclast formation in the long bone cultures might be due to the differences in

the cellular composition of the two types of bone marrow. The long bone marrow contained more

osteoclast precursor cells of the myeloid lineage (early blasts, myeloid blasts, and monocytes) than

the jaw bone marrow. In addition, among the myeloid precursors of the long bone marrow, there

was a higher number of myeloid blasts than in the jaw bone marrow. It was recently shown that

myeloid blasts are the cells that differentiate into osteoclasts in a relatively short period of time

[160].

Post-OVX, calvarial osteoclasts showed the highest turnover rates compared to long bone and

mandibular bone. Mandibular bone did not respond to OVX, which indicates that they are

protected from estrogen mediated post-menopausal increases in bone turnover. This observation is

consistent with the work of Mavropoulos et. al, who showed that mandibular alveolar (trabecular)

bone in rats is less sensitive than the proximal tibia to estrogen- deficiency induced osteoporosis

[19]. They found that mastication lessened sensitivity to estrogen-deficiency possibly due to the

mechanical loading of the mandibular bone during mastication, which has been shown to influence

the mandibular bone density and micro- architecture in the growing the adult rat [19]. As

mentioned above, calvarial osteoclasts are larger, express higher levels of TRAP and undergo

  102  

higher turnover rates pre-OVX. Post-OVX, bone turnover rates in calvarial bone is even higher

which may indicate that the “super” calvarial osteoclasts become even more “super” upon estrogen

depletion. This leads to an extremely high resorption rate. Long bones also displayed higher

turnover rates upon estrogen depletion but to a lesser extent than calvarial bone. Compared to long

bone, calvarial bone displayed very high bone formation rate as evidenced by increases in osteoid

width and interlabel-width to name a few. Calvarial bone is postulated to contain elevated levels of

growth factors that are important for bone formation and regeneration. Finkelman et. al showed

that calvarial bone grafts may have greater survival as donor tissue than bone from other sites [24].

Bone contains growth factors that may play an important role in the regulation of bone repair.

Finkelman et. al, proposed that bone from calvaria may be enriched in one or more growth factors.

It was concluded that the increased concentrations of growth factors IGF-1 and TGF-B in calvarial

bone might lead to a greater capacity for bone repair and graft retention [24]. Long bone turnover

rates indicate an osteoporotic phenotype, which is consistent with the large number of hip fractures

observed in our population today. Hip fractures are very common injuries mainly affecting older

women. It is one of the most common reasons for being admitted to a bone (orthopedic) treatment

ward in a hospital. This study provides important evidence for the heterogeneity of osteoclasts as

well as bone turnover in different parts of the skeleton.

  103  

5.2 Future Directions    The resorption pit 3-D confocal method described in chapter 2 demonstrates that no suitable estimate

of resorption can be complete without reliable 3-D volume imaging. It may be of interest to try

different methods of collagen staining and/or drying methods to reduce the amount of time it takes to

reproduce the technique. In addition, comparing the efficacy of this experiment on commercial

dentin vs. ivory slices produced in the lab may be beneficial to generate the most accurate results.

This technique will provide researchers with a useful to tool to better understand how osteoclasts

function in vitro, which will allow for the rapid assessment of the effect of gene and drug

manipulation on osteoclast function.

The phenotype of FLNA-null mice can be further characterized by identifying whether there is a

compensatory mechanism involving FLNA-B and/or FLNA-C upon loss of FLNA. This can be

accomplished by using a Western blot to detect the intrinsic levels of the aforementioned proteins in

the mouse upon loss of FLNA. In addition, the mineral composition of bone can be measured by

establishing the amounts of calcium and phosphate relative to that of collagen. Quantitative

determination of the serum levels of fragments of CTX can be performed using a RatLaps ELISA kit

(Immunodiagnostic Systems Ltd., Boldon, UK). Since FLNA-KO cortices were more ductile than

WT cortices, it would be interesting to see if the ratio of mature collagen: immature collagen mineral

differs between FLNA-KO and WT bones. This can be achieved using IR or Raman spectroscopy

[175]. Furthermore, back-scattered electron (BSE) imaging can be done to deduce compositional

information on the bone sample. BSE indirectly quantifies the differences in contrast levels within a

bone area that corresponds to varying chemical compositions. Regions containing higher atomic

numbers have an increased probability of collisions with electrons. The integrity of electron

collisions is not only related to atomic number, but also to the density of the atomic nuclei. Therefore

an area of bone with a greater density of calcium atomic nuclei will show a higher number of electron

collisions and result in higher contrast levels [13]. The experiments outlined above will be able to

provide more information to determine whether FLNA would be a good target for the development

of pharmacotherapeutics to treat women with post-menopausal osteoporosis.

Identifying the differences in osteoclasts and bone turnover in the cranial and appendicular skeletons

can provide insight into developing drugs that are targeted to different bone sites. For example, the

  104  

approach to bone grafting might be different in the jaw vs. long bone. Micro-CT can provide further

insight into the differences in BMD and microarchitecture in these bone sites. Measuring serum

levels of fragments of CTX as mentioned above, can give us an idea of how fast/how much bone is

being resorbed by each of the bone sites. Biomechanical testing will provide further insight into the

intrinsic strength of these bones to aid in the assessment of fracture risk. Using the Rpit method to

quantify Rpits post-OVX, is a crucial step to further determine how osteoclast activity differs in these

bone sites. Furthermore, investigations into the crosstalk between osteoblasts and osteoclasts will

shed light on how an FLNA mutation and different skeletal osteoclast precursors signal to osteoblasts

to regulate their function. Could it be a molecule that becomes released during bone resorption that

signals to osteoblasts, like TGF-β, or an unidentified protein released from osteoclasts that functions

as a paracrine signal to osteoblasts? Successful elucidation of these pathways may offer insight into

the skeletal defects in patients with FLNA mutations or may clarify the distinction in bone turnover

between the cranial and appendicular skeletons.

 

  105  

Chapter 6

References [1] Shea, JE and Miller, S C. Skeletal function and structure: implications for tissue-targeted therapeutics. Adv Drug Deliv Rev 2005;57(7):945-57

[2] Public Health Agency of Canada (2004)

[3] Rodan, GA, Martin, TJ. Therapeutic approaches to bone disease. Science. 2000;289(5484):1508-14

[4] Markiewicz, MR, Margarone, JE 3rd, Campbell, JH, Aguirre, A. Bisphosphonate-associated osteonecrosis of the jaws: a review of current knowledge. J Am Dent Assoc 2005;136:1669-74

[5] Reszka, AA, Rodan, GA. Nitrogen-containing bisphosphonate mechanism of action. Mini Rev Med Chem 2004;4:711-9

[6] Flanagan LA, Chou J, Falet H, Neujahr R, Hartwig JH, Stossel TP. Filamin A, the Arp2/3 complex, and the morphology and function of cortical actin filaments in human melanoma cells. J Cell Biol. 2001;155:511-517

[7] Popowicz GM, Schleicher M, Noegel AA, Holak TA. Filamins: promiscuous organizers of the cytoskeleton. Trends Biochem Sci. 2006;31:411-419.

[8] Stossel TP, Condeelis J, Cooley L, et al. Filamins as integrators of cell mechanics and signalling. Nat Rev Mol Cell Biol. 2001;2:138-145.

[9] Jee WS. Integrated bone tissue physiology: Anatomy and physiology. Bone Mechanics Handbook, S. Cowin, ed., pp. 1.1-1.68, Boca Raton, Florida: CRC Press LLC, 2001.177

[10] Shipman P, Walker A, and Bichell D. The Human Skeleton, Massachusetts, USA Harvard University Press, 1985.

[11] Viguet-Carrin S, Garnero P, Delmas PD. The role of collagen in bone strength. Osteoporos Int. 2006;17(3):319-36

[12] Bonucci E. Basic Composition and Structure of Bone. Mechanical Testing of Bone and the Bone - Implant Interface, Y. An and R. Draughn, eds., p. 624, Boca Raton, Florida: CRC Press LLC, 2000.

[13] Boskey A. Bone Mineralization. Bone Mechanics Handbook, S. Cowen, ed., pp. 5.1-5.13, Boca Raton, Florida: CRC Press LLC, 2001.

[14] Von der Mark K. Components of the organic extracellular matrix of bone and cartilage: Structure and biosynthesis of collagens. Dynamics of Bone and Cartilage Metabolism, M. Seibel, S. Robins and J. Bilezikian, eds., pp. 3-18: Elsevier Inc., 1999.

  106  

[15] Crolet JM, Aoubiza B, Meunier A. Compact bone: numerical simulation of mechanical characteristics. J Biomech. 1993; 26(6):677-87.

[16] Rho JY, Kuhn-Spearing L, Zioupos P. Mechanical properties and the hierarchical structure of bone. Med Eng Phys. 1998;20(2):92-102.

[17] Cawood, CI, Howell RA. A classification of the edentulous jaws. International Journal of Oral and Maxillofacial Surgery. 1988;17(4):232-36

[18] Lekovic V, Camargo PM, Klokkevold PR, Weinlaender M, Kenney EB, Dimitrijevic B, Nedic M. Preservation of alveolar bone in extraction sockets using bioabsorbable membranes. J Periodontol. 1999;69(9):1044-9.

[19] Mavropoulos A, Kiliaridis S2, Rizzoli R3, Ammann P3. Normal masticatory function partially protects the rat mandibular bone from estrogen-deficiency induced osteoporosis. J Biomech. 2014;47(11):2666-71.

[20] Woda A, Mishellany A, Peyron MA. The regulation of masticatory function and food bolus formation. J Oral Rehabil. 2006;33(11):840-9.

[21] Quarto N, Wan DC, Kwan MD, Panetta NJ, Li S, Longaker MT. Origin matters: differences in embryonic tissue origin and Wnt signaling determine the osteogenic potential and healing capacity of frontal and parietal calvarial bones. J Bone Miner Res. 2010;25(7):1680-94.

[22] Perez-Amodio S, Jansen DC, Schoenmaker T, Vogels IM, Reinheckel T, Hayman AR, Cox TM, Saftig P, Beertsen W, Everts V. Calvarial osteoclasts express a higher level of tartrate-resistant acid phosphatase than long bone osteoclasts and activation does not depend on cathepsin K or L activity. Calcif Tissue Int. 2006;79(4):245-54.

[23] Jansen ID, Mardones P, Lecanda F, de Vries TJ, Recalde S, Hoeben KA, Schoenmaker T, Ravesloot JH, van Borren MM, van Eijden TM, Bronckers AL, Kellokumpu S, Medina JF, Everts V, Oude Elferink RP. Ae2(a,b)-deficient mice exhibit osteopetrosis of long bones but not of calvaria. 2009;23(10):3470-81.

[24] Finkelman RD, Eason AL, Rakijian DR, Tutundzhyan Y, Hardesty RA. Elevated IGF-II and TGF-beta concentrations in human calvarial bone: potential mechanism for increased graft survival and resistance to osteoporosis. Plast Reconstr Surg. 1994;93(4):732-8.

[25] Blair HC, Athanasou NA. Recent advances in osteoclast biology and pathological bone resorption. Histology and histopathology 2004;19: 189-99.

[26] Teitelbaum SL. Bone resorption by osteoclasts. Science 2000;289: 1504-8.

[27] Rucci N. Molecular biology of bone remodeling. Clin Cases Miner Bone Metab. 2008;5(1): 49–56.

[28] Boyle WJ, Simonet WS, Lacey DL. Osteoclast differentiation and activation. Nature. 2003;423:337–342.

  107  

[29] Cooper C. Epidemiology of Osteoporosis. Osteoporosis International. 1999;9: S2- S8

[30] Glaser DL, Kaplan S. Osteoporosis. Definition and clinical presentation. Spine. 1997;22:12S-16S

[31] Turner CH. Determinants of skeletal fragility and bone quality. J Musculoskelet Neuronal Interact. 2002;2(6):527-8

[32] Kanis J. Diagnosis of Osteoporosis and Fracture Risk. Lancet: Osteoporosis III. 2002;359:1929-1936

[33] Cummings SR, Melton LJ. Epidemiology and outcomes of osteoporotic fractures. Lancet. 2002;359(9319):1761-7

[34] Felsenberg D, Boonen S. The bone quality framework: determinants of bone strength and their interrelationships, and implications for osteoporosis management. Clin Ther. 2005;27(1):1-11

[35] Hansson T, Roos B, Nachemson A. The bone mineral content and ultimate compressive strength of lumbar vertebrae. Spine. 1980;5(1):46-55

[36] Leichter I, Margulies JY, Weinreb A, et al. The relationship between bone density, mineral content, and mechanical strength in the femoral neck. Clin Orthop Relat Res. 1982;163(163):272-81

[37] Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. Bmj 1996;312(7041):1254-9

[38] Seeman E. Pathogenesis of bone fragility in women and men. Lancet. 2002;359(9320):1841-50

[39] Turner C, Burr D. Experimental techniques for bone mechanics. Bone Mechanics Handbook, S. Cowin. pp. 7.1-7.35, Boca Raton, Florida: CRC Press LLC, 2001.

[40] Turner CH, Burr DB. Basic biomechanical measurements of bone: a tutorial. Bone. 1993;14(4):595-608

[41] Cowin S, "Mechanics of Materials," Bone Mechanics Handbook, S. Cowin, ed., pp. 6-1.6.24, Boca Raton, Florida: CRC Press LLC, 2001.

[42] Cortet B, Marchandise X. Bone microarchitecture and mechanical resistance. Joint Bone Spine. 2001;68:297-305.

[43] McCalden RW, McGeough JA, Court-Brown CM. Age-related changes in the compressive strength of cancellous bone. The relative importance of changes in density and trabecular architecture. J Bone Joint Surg Am. 1997;79(3):421-7.

[44] Mosekilde L, Danielsen CC, Gasser J. The effect on vertebral bone mass and strength of long term treatment with antiresorptive agents (estrogen and calcitonin), human parathyroid hormone-(1-38), and combination therapy, assessed in aged ovariectomized rats. Endocrinology. 1994;134(5): 2126-34.

  108  

[45] Thomsen JS, Ebbesen EN, Mosekilde L. Predicting human vertebral bone strength by vertebral static histomorphometry. Bone. 2002;30(3):502-8.

[46] Burr DB. The contribution of the organic matrix to bone's material properties. Bone. 2002;31(1): 8-11

[47] Wang X, Bank RA, TeKoppele JM et al. Effect of collagen denaturation on the toughness of bone. Clin Orthop Relat Res. 2000;371(371):228-39.

[48] Wang X, Shen X, Li X et al. Age-related changes in the collagen network and toughness of bone. Bone. 2002;31(1):1-7.

[49] Tomasek JJ, Meyers SW, Basinger JB et al. Diabetic and age-related enhancement of collagen-linked fluorescence in cortical bones of rats. Life Sci. 1994;55(11):855-61.

[50] Vashishth D, Gibson GJ, Khoury JI et al. Influence of nonenzymatic glycation on biomechanical properties of cortical bone. Bone. 2001;28(2):195-201.

[51] Hayman AR, Bune AJ, Bradley JR et al. Osteoclastic tartrate-resistant acid phosphatase (Acp 5): its localization to dendritic cells and diverse murine tissues. J Histochem Cytochem. 2000;48(2):219-28.

[52] Erben R. Bone-labeling techniques. Handbook of Histology Methods for Bone and Cartilage, Y. M. An, Kylie, ed., pp. 99-117, Totowa, New Jersey: Humana Press, 2003.

[53] Vanderoost J, Harry van Lenthe G. From histology to micro-CT: Measuring and modeling resorption cavities and their relation to bone competence. World J Radiol. 2014;6(9):643-656

[54] Vedi S, Compston J. Bone histomorphometry. Methods Mol Med. 2003;80:283–298

[55] Gundersen HJ. Stereology--or how figures for spatial shape and content are obtained by observation of structures in sections. Microsc Acta. 1980;83:409–426.

[56] Kragstrup J, Gundersen HJG, Melsen F, Mosekilde L. Estimation of the 3-Dimensional Wall Thickness of Completed Remodeling Sites in Iliac Trabecular Bone. Metab Bone Dis Relat Res. 1982;4:113–119.

[57] Eriksen EF, Melsen F, Mosekilde L. Reconstruction of the resorptive site in iliac trabecular bone: a kinetic model for bone resorption in 20 normal individuals. Metab Bone Dis Relat Res. 1984;5:235–242.

[58] Compston JE, Croucher PI. Histomorphometric assessment of trabecular bone remodelling in osteoporosis. Bone Miner. 1991;14:91–102.

[59] Vedi S, Compston J. Bone histomorphometry. Methods Mol Med. 2003;80:283–298.

[60] Arlot ME, Delmas PD, Chappard D, Meunier PJ. Trabecular and endocortical bone remodeling in postmenopausal osteoporosis: comparison with normal postmenopausal women. Osteoporos Int.

  109  

1990;1:41–49.

[61] Eriksen EF, Mosekilde L, Melsen F. Trabecular resorption depth decreases with age: differences between normal males and females. Bone. 1985;6:141-146

[62] Glorieux FH, Travers R, Taylor A, Bowen JR, Rauch F, Norman M, Parfitt AM. Normative data for iliac bone histomorphometry in growing children. Bone. 2000;26:103–109.

[63] Palle S, Chappard D, Vico L, Riffat G, Alexandre C. Evaluation of the osteoclastic population in iliac crest biopsies from 36 normal subjects: a histoenzymologic and histomorphometric study. J Bone Miner Res. 1989;4:501–506.

[64] Parfitt AM. Targeted and nontargeted bone remodeling: relationship to basic multicellular unit origination and progression. Bone. 2002;30:5–7.

[65] Hernandez CJ. How can bone turnover modify bone strength independent of bone mass? Bone. 2008;42:1014–1020.

[66] Allen MR, Erickson AM, Wang X, Burr DB, Martin RB, Hazelwood SJ. Morphological assessment of basic multicellular unit resorption parameters in dogs shows additional mechanisms of bisphosphonate effects on bone. Calcif Tissue Int. 2010;86:67–71

[67] Mosekilde L. Consequences of the remodelling process for vertebral trabecular bone structure: a scanning electron microscopy study (uncoupling of unloaded structures) Bone Miner. 1990;10:13–35.

[68] Tkachenko EV, Slyfield CR, Tomlinson RE, Daggett JR, Wilson DL, Hernandez CJ. Voxel size and measures of individual resorption cavities in three-dimensional images of cancellous bone. Bone. 2009;45:487–492.

[69] Garnero P, Sornay-Rendu E, Chapuy M.C. et al. Increased bone turnover in late postmenopausal women is a major determinant of osteoporosis. J Bone Miner Res. 1996;11:337-49

[70] Pacifici R. Estrogen, cytokines, and pathogenesis of postmenopausal osteoporosis. J Bone Miner Res. 1996;8:1043-51

[71] Manolagas SC. Birth and death of bone cells: basic regulatory mechanisms and implications for the pathogenesis and treatment of osteoporosis. Endocr Rev. 2000;21(2):115-37.

[72] Riggs BL, Khosla S, Melton 3rd LJ. A unitary model for involutional osteoporosis: estrogen deficiency causes both type I and type II osteoporosis in postmenopausal women and contributes to bone loss in aging men. J Bone Miner Res. 1998;13(5):763-73.

[73] Jilka RL. The Relevance of Mouse Models for Investigating Age-Related Bone Loss in Humans. J Gerontol A Biol Sci Med Sc. 2013;68(10):1209-1217.

[74] Iotsova V, Caamaño J, Loy J, Yang Y, Lewin A, Bravo R. Osteopetrosis in mice lacking NF-kappaB1 and NF-kappaB2. Nature Medicine. 1997;3:1285-1289.

  110  

[75] Mak TW et al. TRAF6 deficiency results in osteopetrosis and defective interleukin-1, CD40, and LPS signaling. Genes Dev. 1999;13(8):1015-24.

[76] Bilezikian JP, Raisz LG, Martin TJ. Principles of Bone Biology: Two-Volume Set. 2008

[77] Van der Eerden BCJ, Tudpor K, Jongwattapapisan P, Woudenberg-Vrenken TE, Bindels RJM, Hoenderop JGJ, Van Leeuwen JPTM. Thrombin receptor deficiency leads to osteopetrosis by decreasing the RANKL/OPG ratio. Bone Abstracts. 2013;10(1530):

[78] Naito A1, Azuma S, Tanaka S, Miyazaki T, Takaki S, Takatsu K, Nakao K, Nakamura K, Katsuki M, Yamamoto T, Inoue J. Severe osteopetrosis, defective interleukin-1 signalling and lymph node organogenesis in TRAF6-deficient mice. Genes Cells. 1999;4(6):353-62.

[79] Blair HC, Athanasou NA. Recent advances in osteoclast biology and pathological bone resorption. Histology and histopathology 2004;19: 189-99.

[80] Teitelbaum SL. Bone resorption by osteoclasts. Science 2000;289: 1504-8.

[81] Soysa NS, Alles N, Aoki K, Ohya K. Three-dimensional characterization of osteoclast bone-resorbing activity in the resorption lacunae. Journal of medical and dental sciences 2009;56: 107-12.

[82] Novack DV, Faccio R. Osteoclast motility: putting the brakes on bone resorption. Ageing research reviews 2011;10: 54-61.

[83] Geblinger D, Geiger B, Addadi L. Surface-induced regulation of podosome organization and dynamics in cultured osteoclasts. Chembiochem : a European journal of chemical biology 2009;10: 158-65.

[84] Teti A, Taranta A, Migliaccio S, Degiorgi A, Santandrea E, Villanova I, Faraggiana T, Chellaiah M, Hruska KA. Colony stimulating factor-1-induced osteoclast spreading depends on substrate and requires the vitronectin receptor and the c-src proto-oncogene. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research 1998;13: 50-8.

[85] Simonet WS1, Lacey DL, Dunstan CR, Kelley M, Chang MS, Lüthy R, Nguyen HQ, Wooden S, Bennett L, Boone T, Shimamoto G, DeRose M, Elliott R, Colombero A, Tan HL, Trail G, Sullivan J, Davy E, Bucay N, Renshaw-Gegg L, Hughes TM, Hill D, Pattison W, Campbell P, Sander S, Van G, Tarpley J, Derby P, Lee R, Boyle WJ. Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell. 1997 Apr 18;89(2):309-19.

[86] Akisaka T, Yoshida H, Suzuki R, Takama K. Adhesion structures and their cytoskeleton- membrane interactions at podosomes of osteoclasts in culture. Cell Tissue Res. 2008;331:625-641.

[87] Jurdic P, Saltel F, Chabadel A, Destaing O. Podosome and sealing zone: specificity of the osteoclast model. Eur J Cell Biol. 2006;85:195-202.

[88] Saltel F, Chabadel A, Bonnelye E, Jurdic P. Actin cytoskeletal organisation in osteoclasts: a model to decipher transmigration and matrix degradation. Eur J Cell Biol. 2008;87:459-468.

  111  

[89] Jaffe AB, Hall A. Rho GTPases: biochemistry and biology. Annu Rev Cell Dev Biol. 2005;21:247-269.

[90] Kater SB, Rehder V. The sensory-motor role of growth cone filopodia. Curr Opin Neurobiol. 1995;5:68-74.

[91] Nobes CD, Hall A. Rho, rac, and cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell. 1995;81:53-62.

[92] Ridley AJ, Hall A. The small GTP-binding protein rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell. 1992;70:389-399.

[93] Pestonjamasp KN, Forster C, Sun C, et al. Rac1 links leading edge and uropod events through Rho and myosin activation during chemotaxis. Blood. 2006;108:2814-2820.

[94] Ohta Y, Hartwig JH, Stossel TP. FilGAP, a Rho- and ROCK-regulated GAP for Rac binds filamin A to control actin remodelling. Nat Cell Biol. 2006;8:803-814.

[95] Machacek M, Hodgson L, Welch C, et al. Coordination of Rho GTPase activities during cell protrusion. Nature. 2009;461:99-103.

[96] Fukuda A, Hikita A, Wakeyama H, et al. Regulation of osteoclast apoptosis and motility by small GTPase binding protein Rac1. J Bone Miner Res. 2005;20:2245-2253.

[97] Wang Y, Lebowitz D, Sun C, Thang H, Grynpas MD, Glogauer M. Identifying the relative contributions of Rac1 and Rac2 to osteoclastogenesis. J Bone Miner Res. 2008;23:260-270.

[98] Itokowa T, Zhu ML, Troiano N, Bian J, Kawano T, Insogna K. Osteoclasts lacking Rac2 have defective chemotaxis and resorptive activity. Calcif Tissue Int;88:75-86.

[99] Lee NK, Choi HK, Kim DK, Lee SY. Rac1 GTPase regulates osteoclast differentiation through TRANCE-induced NF-kappa B activation. Mol Cell Biochem. 2006;281:55-61.

[100] Wang K, Ash JF, Singer SJ. Filamin, a new high-molecular-weight protein found in smooth muscle and non-muscle cells. Proc Natl Acad Sci U S A. 1975;72:4483-4486.

[101] Nakamura F, Osborn TM, Hartemink CA, Hartwig JH, Stossel TP. Structural basis of filamin A functions. J Cell Biol. 2007;179:1011-1025.

[102] Cunningham CC, Gorlin JB, Kwiatkowski DJ, et al. Actin-binding protein requirement for cortical stability and efficient locomotion. Science. 1992;255:325-327.

[103] Glogauer M, Arora P, Chou D, Janmey PA, Downey GP, McCulloch CA. The role of actin-binding protein 280 in integrin-dependent mechanoprotection. J Biol Chem. 1998;273:1689-1698.

[104] Wu C. Migfilin and its binding partners: from cell biology to human diseases. J Cell Sci. 2005;118:659-664.

  112  

[105] Sheen VL, Dixon PH, Fox JW, et al. Mutations in the X-linked filamin 1 gene cause periventricular nodular heterotopia in males as well as in females. Hum Mol Genet. 2001;10:1775-1783.

[106] Cox D, Condeelis J, Wessels D, Soll D, Kern H, Knecht DA. Targeted disruption of the ABP-120 gene leads to cells with altered motility. J Cell Biol. 1992;116:943-955.

[107] Robertson SP. Filamin A: phenotypic diversity. Curr Opin Genet Dev. 2005;15:301-307.

[108] Feng Y, Chen MH, Moskowitz IP, et al. Filamin A (FLNA) is required for cell-cell contact in vascular development and cardiac morphogenesis. Proc Natl Acad Sci U S A. 2006;103:19836-19841.

[109] Marti A, Luo Z, Cunningham C, et al. Actin-binding protein-280 binds the stress-activated protein kinase (SAPK) activator SEK-1 and is required for tumor necrosis factor-alpha activation of SAPK in melanoma cells. J Biol Chem. 1997;272:2620-2628.

[110] Ohta Y, Suzuki N, Nakamura S, Hartwig JH, Stossel TP. The small GTPase RalA targetsfilamin to induce filopodia. Proc Natl Acad Sci U S A. 1999;96:2122-2128.

[111] Vadlamudi RK, Li F, Adam L, et al. Filamin is essential in actin cytoskeletal assembly mediated by p21-activated kinase 1. Nat Cell Biol. 2002;4:681-690

[112] Ohta Y, Hartwig JH, Stossel TP. FilGAP, a Rho- and ROCK-regulated GAP for Rac binds filamin A to control actin remodelling. Nat Cell Biol. 2006;8:803-814.

[113] Chen M, Stracher A. In situ phosphorylation of platelet actin-binding protein by cAMP- dependent protein kinase stabilizes it against proteolysis by calpain. J Biol Chem.1989;264:14282-14289.

[114] Ohta Y, Hartwig JH. Actin filament cross-linking by chicken gizzard filamin is regulated by phosphorylation in vitro. Biochemistry. 1995;34:6745-6754.

[115] Leung R, Wang Y, Cuddy K, Sun C, Magalhaes J, Grynpas MD, Glogauer M. Filamin A Regulates Monocyte Migration Through Rho Small GTPases During Osteoclastogenesis. J Bone Miner Res. 2010;25:1077-91

[116] Ballanti P, Minisola S, Pacitti MT, Scarnecchia L, Rosso R, Mazzuoli GF, Bonucci E. Tartrate-resistant acid phosphate activity as osteoclastic marker: sensitivity of cytochemical assessment and serum assay in comparison with standardized osteoclast histomorphometry. Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 1997;7: 39-43.

[117] Boyde A, Jones SJ. REALLY CLOSE UP! SURVEYING SURFACES AT SUB-MICROMETRE RESOLUTION: THE MEASUREMENT OF OSTEOCLASTIC RESORPTION LACUNAE. The Photogrammetric Record 1992;14: 59-84.

[118] Pradella SD, Slone SP, Wu YY, Chao TY, Parthasarathy RN, Yam LT, Janckila AJ.

  113  

Applications and performance of monoclonal antibodies to human tartrate resistant acid phosphatase. Journal of immunological methods 2011;372: 162-70.

[119] Gralow JR, Biermann JS, Farooki A, Fornier MN, Gagel RF, Kumar RN, Shapiro CL, Shields A, Smith MR, Srinivas S, Van Poznak CH. NCCN Task Force Report: Bone Health in Cancer Care. Journal of the National Comprehensive Cancer Network : JNCCN 2009;7 Suppl 3: S1-32; quiz S33-5.

[120] Watts NB. Clinical utility of biochemical markers of bone remodeling. Clin Chem. 1999;45 (8 Pt 2): 1359–68.

[121] Toiviainen-Salo S, Mayranpaa MK, Durie PR, et al. Shwachman-Diamond syndrome is associated with low-turnover osteoporosis. Bone. 2007;41:965–72.

[122] Nishita M, Yoo SK, Nomachi A, et al. Filopodia formation mediated by receptor tyrosine kinase Ror2 is required for Wnt5a-induced cell migration. J Cell Biol. 2006;175:555–62

[123] Heasman SJ, Ridley AJ. Mammalian Rho GTPases: new insights into their functions from in vivo studies. Nature reviews. Molecular cell biology 2008;9: 690-701.

[124] Spiering D, Hodgson L. Dynamics of the Rho-family small GTPases in actin regulation and motility. Cell adhesion & migration 2011;5: 170-80.

[125] Clausen BE, Burkhardt C, Reith W, Renkawitz R, Forster I. Conditional gene targeting in macrophages and granulocytes using LysMcre mice. Transgenic research 1999;8: 265-77.

[126] Guo F, Cancelas JA, Hildeman D, Williams DA, Zheng Y. Rac GTPase isoforms Rac1 and Rac2 play a redundant and crucial role in T-cell development. Blood 2008;112: 1767-75.

[127] Wheeler AP, Wells CM, Smith SD, Vega FM, Henderson RB, Tybulewicz VL, Ridley AJ. Rac1 and Rac2 regulate macrophage morphology but are not essential for migration. Journal of cell science 2006;119: 2749-57.

[128] Pai SY, Kim C, Williams DA. Rac GTPases in human diseases. Disease markers 2010;29: 177-87.

[129] Yan J, Chen S, Zhang Y, Li X, Li Y, Wu X, Yuan J, Robling AG, Kapur R, Chan RJ, Yang FC. Rac1 mediates the osteoclast gains-in-function induced by haploinsufficiency of Nf1. Human molecular genetics 2008;17: 936-48.

[130] Itzstein C, Coxon FP, Rogers MJ. The regulation of osteoclast function and bone resorption by small GTPases. Small GTPases 2011;2: 117-130.

[131] Nesbitt SA, Horton MA. Fluorescence imaging of bone-resorbing osteoclasts by confocal microscopy. Methods in molecular medicine 2003;80: 259-81.

[132] Dolber PC, Spach MS. Conventional and confocal fluorescence microscopy of collagen fibers in the heart. The journal of histochemistry and cytochemistry : official journal of the Histochemistry

  114  

Society 1993;41: 465-9.

[133] Trau H, Dayan D, Hirschberg A, Hiss Y, Bubis JJ, Wolman M. Connective tissue nevi collagens. Study with picrosirius red and polarizing microscopy. The American Journal of dermatopathology 1991;13: 374-7.

[134] Sun CX, Magalhaes MA, Glogauer M. Rac1 and Rac2 differentially regulate actin free barbed end formation downstream of the fMLP receptor. The Journal of cell biology 2007;179: 239-45.

[135] Junqueira LC, Bignolas G, Brentani RR. Picrosirius staining plus polarization microscopy, a specific method for collagen detection in tissue sections. The Histochemical journal 1979;11: 447-55.

[136] Boyde A, Jones SJ. Pitfalls in pit measurement. Calcified tissue international 1991;49: 65-70.

[137] Soysa NS, Alles N, Aoki K, Ohya K. Three-dimensional characterization of osteoclast bone-resorbing activity in the resorption lacunae. Journal of medical and dental sciences 2009;56: 107-12

[138] Ory S, Munari-Silem Y, Fort P, Jurdic P. Rho and Rac exert antagonistic functions on spreading of macrophage-derived multinucleated cells and are not required for actin fiber formation. Journal of cell science 2000;113 ( Pt 7): 1177-88.

[139] Razzouk S, Lieberherr M, Cournot G. Rac-GTPase, osteoclast cytoskeleton and bone resorption. European journal of cell biology 1999;78: 249-55.

[140] Athanasou NA. The osteoclast--what's new? Skeletal radiology 2011;40: 1137-40.

[141] Takahashi N, Ejiri S, Yanagisawa S, Ozawa H. Regulation of osteoclast polarization. Odontology / the Society of the Nippon Dental University 2007;95: 1-9.

[142] Garimella R, Tague SE, Zhang J, Belibi F, Nahar N, Sun BH, Insogna K, Wang J, Anderson HC. Expression and synthesis of bone morphogenetic proteins by osteoclasts: a possible path to anabolic bone remodeling. The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society 2008;56: 569-77.

[143] Kaye M, Zucker SW, Leclerc YG, Prichard S, Hodsman AB, Barre PE. Osteoclast enlargement in endstage renal disease. Kidney international 1985;27: 574-81.

[144] Lees RL, Heersche JN. Macrophage colony stimulating factor increases bone resorption in dispersed osteoclast cultures by increasing osteoclast size. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 1999;14: 937-45.

[145] Makris GP, Saffar JL. Quantitative relationship between osteoclasts, osteoclast nuclei and the extent of the resorbing surface in hamster periodontal disease. Archives of oral biology 1982;27: 965-9.

[146] Piper K, Boyde A, Jones SJ. The relationship between the number of nuclei of an osteoclast and its resorptive capability in vitro. Anatomy and embryology 1992;186: 291-9.

  115  

[147] Boissy P, Saltel F, Bouniol C, Jurdic P, Machuca-Gayet I. Transcriptional activity of nuclei in multinucleated osteoclasts and its modulation by calcitonin. Endocrinology 2002;143: 1913-21.

[148] Saltel F, Destaing O, Bard F, Eichert D, Jurdic P. Apatite-mediated actin dynamics in resorbing osteoclasts. Mol Biol Cell. 2004;15:5231–41.

[149] Ory S, Brazier H, Pawlak G, Blangy A. Rho GTPases in osteoclasts: orchestrators of podosome arrangement. Eur J Cell Biol. 2005;87: 469–77

[150] Parfitt AM, Drezner MK, Glorieux FH et al. Bone histomorphometry: standardization of nomenclature, symbols, and units. Report of the ASBMR179 Histomorphometry Nomenclature Committee. J Bone Miner Res. 1987;2:595- 610

[151] Takahata M, Maher JR, Juneja SC, Inzana J, Xing L, Schwarz EM, Berger AJ, Awad HJ. Mechanisms of bone fragility in a glucocorticoid-treated mouse model of rheumatoid arthritis – implications for insufficiency fracture risk. Arthritis Rheum. 2012;11:3649-3659.

[152] Hayashi K1, Altman A. Filamin A is required for T cell activation mediated by protein kinase C-theta. J Immunol. 2006;177(3):1721-8.

[153] Sasaki A1, Masuda Y, Ohta Y, Ikeda K, Watanabe K. Filamin associates with Smads and regulates transforming growth factor-beta signaling. J Biol Chem. 2001;21:17871-7.

[154] Hahnel R, Gschwendt M. The interaction between protein-kinase-C (PKC) and Estrogens. International journal of oncology. 1995;10: 11-16

[155] Marino M, Galluzzo P, Ascenzi P. Estrogen Signaling Multiple Pathways to Impact Gene Transcription. Curr Genomics. 1996;8: 497–508

[156] Kimble RB, Srivastava S, Ross FP, Matayoshi A, Pacifici R. Estrogen deficiency increases the ability of stromal cells to support murine osteoclastogenesis via an interleukin-1and tumor necrosis factor-mediated stimulation of macrophage colony-stimulating factor production. J Biol Chem. 1996;46:28890-7

[157] Leonardi A, Ellinger-Ziegelbauer H, Franzoso G, Brown K, Siebenlist U. Physical and Functional Interaction of Filamin (Actin-binding Protein-280) and Tumor Necrosis Factor Receptor-associated Factor 2. The Journal of Biological Chemistry. 2000;275: 271-278

[158] Kirstein B, Chambers TJ, Fuller K. Secretion of tartrate-resistant acid phosphatase by osteoclasts correlates with resorptive behavior. J Cell Biochem. 2006;98(5):1085-94.

[159] Everts V, de Vries TJ, Helfrich MH. Osteoclast heterogeneity::Lessons from osteopetrosis and inflammatory conditions. Molecular Basis of Disease 2009; 1792(8):757-765.

[160] de Souza Faloni AP, Schoenmaker T, Azari A, Katchburian E, Cerri PS, de Vries TJ, Everts V. Jaw and Long Bone Marrows Have a Different Osteoclastogenic Potential. Calcified Tissue International. 2011;88(1):63-74.

  116  

[161] Odgren PR, Kim N, MacKay CA, Mason-Savas A, Choi Y, Marks SC Jr. The role of RANKL (TRANCE/TNFSF11), a tumor necrosis factor family member, in skeletal development: effects of gene knockout and transgenic rescue. Connect Tissue Res. 2003;44(1):264-71.

[162] Metcalfe D. The pathophysiology of osteoporotic hip fracture. Mcgill J Med. 2008; 11(1): 51–57.

[163] Bos T, Speijer D, Bank RA, Brömme D, Everts V. Differences in matrix composition between calvaria and long bone in mice suggest differences in biomechanical properties and resorption: special emphasis on collagen. Bone. 2008;43:459–468.

[164] Sodek KL, Tupy JH, Sodek J, Grynpas MD. Relationships between bone protein and mineral in developing porcine long bone and calvaria. Bone. 2000;26:189–198.

[165] Akintoye SO, Lam T, Shi S, Brahim J, Collins MT, Robey PG. Skeletal site-specific characterization of orofacial and iliac crest human bone marrow stromal cells in same individuals. Bone. 2006;38:758–768

[166] Raisz L., Seeman E. Causes of age related bone loss and bone fragility: An alternative view. J Bone Miner Res 2001;16: 1948–1952

[167] Seeman E. From density to structure: growing up and growing old on the surfaces of bone. J Bone Miner Res. 1997;12: 509–521

[168] Demontiero O, Vidal C, Duque G. Aging and bone loss: new insights for the clinician. Ther Adv Musculoskelet Dis. 2012; 4(2): 61–76.

[169] Wu Z, Bucher NL, Farmer SR. Induction of peroxisome proliferator-activated receptor gamma during the conversion of 3T3 fibroblasts into adipocytes is mediated by C/EBPbeta, C/EBPdelta, and glucocorticoids. Mol Cell Biol. 1996;16:4128-4136.

[170] Yeh WC, Cao Z, Classon M, McKnight SL. Cascade regulation of terminal adipocyte differentiation by three members of the C/EBP family of leucine zipper proteins. Genes Dev. 1995;9:168-181.

[171] Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G. Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell. 1997;89:747-754.

[172] Nakashima K, Zhou X, Kunkel G, et al. The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell. 2002;108:17-29.

[173] Lees RL, Heersche JN. Differences in regulation of pH(i) in large (≥10 nuclei) and small (≤5 nuclei) osteoclasts. Am J Physiol Cell Physiol. 2000;279:C751–C761.

[174] Lees RL, Sabharwal VK, Heersche JN. Resorptive state and cell size influence intracellular pH regulation in rabbit osteoclasts cultured on collagen-hydroxyapatite films. Bone. 2001;28:187–194.

  117  

[175] Creemers LB, Jansen DC, van Veen-Reurings A, van den Bos T, Everts V. Microassay for the assessment of low levels of hydroxyproline. BioTechniques 1997;22:656-658