UNDERSTANDING THE OSTEOIMMUNOLOGICAL ROLE OF … · UNDERSTANDING THE OSTEOIMMUNOLOGICAL ROLE OF ROQUIN IN THE MAINTENANCE OF BONE HOMEOSTASIS BAY SIE LIM BCOM, BSC, BMEDSCI (HONS)

UNDERSTANDING THE OSTEOIMMUNOLOGICAL ROLE OF

ROQUIN IN THE MAINTENANCE OF BONE HOMEOSTASIS

BAY SIE LIM BCOM, BSC, BMEDSCI (HONS)

This thesis is submitted to fulfil the requirements of University of Western Australia for

the degree of

DOCTOR OF PHILOSOPHY

2014

The work presented in this thesis was completed at the University of Western Australia, School of Pathology and Laboratory Medicine, Queen Elizabeth II Medical Centre,

Nedlands, Western Australia

TABLE OF CONTENTS

Table of Contents

Declaration 1

Scientific Abstract 2

Acknowledgements 7

Abstracts and Awards 9

Abbreviations 10

List of Figures 17

Chapter 1: The Biology of Bone Multicellular Unit and Transcriptional

Regulations of Bone Homeostasis 20

1.1. Introduction 21

1.2. Bone cells 22

1.2.1. Osteoclasts 22

1.2.2. Osteoblasts 24

1.2.3. Osteocytes 27

1.2.4. Osteolining cells 29

1.3. Formation and maintenance of bone 31

1.3.1. Osteogenesis 31

1.3.2 Bone modelling and remodelling 32

1.4. Conclusion 36

Chapter 2: Osteoimmunology – Interactions of the Skeletal and Immune

Systems 41


2.2. Osteoclasts and inflammatory bone destruction 42

2.3. Osteoclast and innate immune cell lineages: RANK and TLRs downstream

signalling cascades 43

2.4. Role of inflammatory hematopoietic cytokines as a double-edged sword in

bone modeling and remodeling 46

2.4.1. Tumor necrosis factor (TNFs) 46

2.4.2. Interleukin (IL)-1 48

2.4.3. IL-6 48

2.4.4. IL-8 50

2.4.5. IL-17 51

2.4.6. IL-21 52

2.4.7. Other interleukins 52

2.4.8. Interferons (IFNs) 54

2.4.9. Transforming growth factor (TGF)-β 55

2.5. The orchestration of adaptive immune cells by innate immune cells,

osteoclastogenic RANKL and its antagonist OPG 57

2.6. Delineation of T cell and osteoclastogenic T cell candidates 59

2.7. The immunomodulatory role of bone cells in establishing the haematopoietic

stem cell (HSC) niche 63

2.8. Conclusion 65

Chapter 3: Hypothesis and Aims – Sanroque Mutant Mouse as an

Osteoimmune Mouse Model 68

3.1. General hypothesis and aims 69

3.2. Screening of ENU-induced mutant mouse lines: the sanroque mutation 70

3.2.1. The ubiquitin proteasome system (UPS) and Roquin E3 ligase 70

3.2.2. Sanroque mutant mice exhibits low bone mass phenotype 71

3.2.3. Sanroque mice display a lupus-like autoimmune disease 71

3.3. Specific hypothesis and aims 71

Chapter 4: Materials and Methods 77

4.1. Materials 78

4.1.1. Chemical reagents 78

4.1.2. Commercially purchased kits and molecular products 80

4.1.3. Other products and consumables 82

4.1.4. Tissue culture materials and reagents 83

4.1.5. Cytokines 84

4.1.6. Antibodies 84

4.1.7. Oligonucleotide primers 86

4.1.8. Solutions 89

4.1.9. Equipment 92

4.1.9.1. Centrifugation 94

4.1.10. Software 94

4.2. Animals and bone phenotyping methods 95

4.2.1. The generation of sanroque mutant mouse line 95

4.2.2. Radiographic screening of sanroque mutant mice 96

4.2.3. Microquantitative computed X-ray tomography (MicroCT) analysis of

wildtype and sanroque mutant mice 96

4.2.4. Fluorochrome labelling of wildtype and sanroque mutant mice 97

4.2.5. Histological analysis of wildtype and sanroque mutant mice 97

4.2.5.1. Tissue fixation, decalcification (for hindlimbs), processing, embedding

and sectioning 97

4.2.5.2. Haematoxylin and eosin staining 99

4.2.5.3. TRAP staining 99

4.2.5.4. Goldner Trichome staining on non-decalcified bone samples 100

4.2.5.4.1. Processing, embedding and sectioning of undecalcified bone

samples 100

4.2.5.4.2. Goldner Trichrome staining 101

4.2.5.5. Bone histomorphometric analysis 102

4.2.6. TRAP staining of mouse calvariae 102

4.2.7. Immunofluorescence staining of mouse spleens 103

4.2.7.1. Embedding of sample in Optimal Cutting Temperature (OCT) 103

4.2.7.2. Immunofluorescence staining 103

4.3. Isolation, culture and in vitro studies of primary cell lines 104

4.3.1. Isolation and culture of bone marrow cells 104

4.3.1.1. Generation and maintenance of bone marrow macrophages (BMMs)

104

4.3.1.2. MTS cell proliferation assay 105

4.3.1.3. Osteoclast culture 105

4.3.1.3.1. In vitro osteoclastogenesis assay 106

4.3.1.3.2. In vitro bone resorption assay 106

4.3.1.3.3. In vitro osteoclast cathepsin K release assay using bovine bone

powder 107

4.3.2. Isolation of osteoblastic bone cells from long bones and calvariae 108

4.3.2.1. Osteoblast culture and mineralisation assay 109

4.3.2.2. Osteoblasts and BMMs co-culture experiment 109

4.3.3. Flow cytometry analysis of bone marrow cell niche 110

4.3.3.1. Immunostaining of bone marrow cells 110

4.3.3.2. Flow cytometer and analysis 111

4.3.4. Cryopreservation of cells for long term storage 112

4.4. Molecular biology techniques 112

4.4.1. RNA extraction and quantification 112

4.4.1.1. General handling procedures 112

4.4.1.2. Extraction of total RNA 112

4.4.1.2.1. RNA extraction from bone tissue 112

4.4.1.2.2. RNA extraction from cell culture 113

4.4.1.3. Quantification of RNA concentration 113

4.4.2. Reverse Transcription Polymerise Chain Reaction (RT-PCR) 114

4.4.2.1. Reverse Transcription (RT) 114

4.4.2.2. Polymerase Chain Reaction (PCR) amplification 114

4.4.2.3. DNA agarose gel electrophoresis 115

4.4.3. Quantitative Real-Time PCR (Q-PCR) and Livak’s Equation 116

4.4.4. Protein expression analysis using Sodium Dodecyl Sulfate-PolyAcrylamide

Gel Electrophoresis (SDS-PAGE) and western blotting 117

4.4.4.1. Protein extraction from mouse organs 117

4.4.4.2. Protein extraction from cultured cells 117

4.4.4.3. Quantification of protein concentration 118

4.4.4.4. Separation of proteins by SDS-PAGE 118

4.4.4.5. Protein transfer onto nitrocellulose blotting membrane 120

4.4.4.6. Western blotting 120

4.5. Statistics and data presentation 121

Chapter 5: Bone Phenotypic Characterisation of Sanroque Mutant Mice 122


5.2. Experimental results 124

5.2.1. Roquin gene expression is regulated during osteoblastogenesis and

osteoclastogenesis and is highly expressed in hematopoietic organs 124

5.2.2. Sanroque mutation alters trabecular and cortical bone morphology 124

5.2.3. Sanroque mutant mice have increased osteoclast numbers in vivo 126

5.2.4. Bone mineral apposition rate in sanroque mutant mice is reduced despite

unchanged histological osteoblast parameters 126

5.2.5. RANKL and osteoclast gene markers are highly expressed in sanroque

bone 127

5.3. Discussion 127

Chapter 6: In vitro Investigation on The Role of Roquin in Osteoblast

Biology 141



6.2.1. Reduced bone mineralisation by sanroque mutant COB cultures 143

6.2.2. Increased calcium deposition by mutant sanroque LBOB cultures 143

6.2.3. Alkaline phosphatase activity level is altered throughout osteogenesis in

sanroque LBOB cell culture 144

6.2.4. Mutation of Roquin affects the expression levels of osteoblast gene markers

during differentiation of sanroque LBOBs 145

6.2.5. Mutant LBOB from sanroque mutant mice have reduced capability to

support osteoclastogenesis in vitro 146

6.3. Discussion 147

Chapter 7: The Effect of Sanroque Mutation on Osteoclast Biology 161



7.2.1. Osteoclastogenesis is accelerated in sanroque mutant mice 163

7.2.2. Sanroque mutation elevates the expression of osteoclast marker genes 164

7.2.3. Sanroque mutant osteoclasts demonstrate increased bone lytic activity in

vitro 164

7.2.4. Mutation in Roquin gene alters key RANKL-signalling pathways 165

7.2.5. Ubiquitination of TRAF6 in sanroque BMMs appears unaltered 167

7.2.6. Blood serum from sanroque mutant mice contains proliferative- cytokines

and sanroque BMMs have increased proliferation in response to M-CSF 168

7.3. Discussion 169

Chapter 8: Osteoimmunological Mechanism Behind the Osteopenic

Phenotype of Sanroque Mutant Mice 187



8.2.1. The putative osteoclast progenitor population present within the bone

marrow of sanroque mutant mice is increased. 190

8.2.2. CD4 T cells of sanroque mutant mice expressed less CD70 191

8.2.3. Splenocytes of sanroque mutant mice comprised a higher number of

common myeloid progenitors and follicular helper T cells that express RANKL

192

8.3. Discussion 194

Chapter 9: General Overview and Future Directions 213

9.1. Overview of thesis and general discussion 214

9.2. Future Directions 220

9.3. Conclusion 223

Chapter 10: References 225

Appendix 276

1

DECLARATION

This is to certify that all the work contained herein was performed by myself, except

where indicated otherwise.

_____ ____

Bay Sie LIM

W/Prof. Jiake XU

Dr. Jennifer TICKNER

2

SCIENTIFIC ABSTRACT

The bone organ system is a dynamic mineralized tissue that is continuously being

broken down and rebuilt in a process known as bone remodelling. The remodelling

process is an important metabolic activity involved in the maintenance of the

homeostasis and architecture of bone while performing its versatile functions. The bone

remodelling process is a coupled activity between the catabolic effects of bone-

resorbing osteoclasts and the anabolic effects of bone-forming osteoblasts.

The remodelling cycle begins when there is an external mechanical stimulus, which

initiates accessory cells including osteoblasts to secrete RANKL. RANKL binds onto

the cell surface receptor RANK displayed on osteoclast precursors that originate from

the haematopoietic stem cells. The RANKL-RANK interaction drives

osteoclastogenesis and cell fusion to form large multinucleated osteoclasts. Once

mature, osteoclasts begin to secrete proteolytic enzymes to resorb the bone matrix. The

bone erosion by osteoclasts is a crucial process in remoulding the bone tissue.

To prevent osteoclasts from continuously resorbing bone, osteoblasts, which are derived

from mesenchymal stem cells, also secrete regulatory molecules such as OPG. OPG

acts as a decoy receptor that binds onto RANKL to prevent RANKL-RANK interaction.

Such interactions allow osteoblasts to modulate the bone remodelling cycle, balancing

bone resorption and bone formation. Mature osteoblasts are the primary matrix-

producing cells that build and mineralise bone. As a result, some become entombed

within their own matrix and differentiate into osteocytic cells with long dendritic

processes to maintain cell-cell contact amongst themselves and with bone cells residing

on the surfaces of bone. Osteocytes are generally recognised as the mechanotransducers

of bone tissue, responding to mechanical stimuli and generating signals that play

important roles in regulating the bone remodelling cycle.

The cellular coupling among these bone cells to achieve bone homeostasis is complex

and heavily regulated by local and systemic factors. With the emerging field of

osteoimmunology, many studies have implied that both innate and adaptive immune

systems are major components of bone remodelling, and vice versa. The dialogue

between the skeletal and immune systems is inevitable as stem cells, which give rise to

progenitors that constitute both the bone organ system and immune system, are

3

generated within the bone marrow reservoir.

The co-regulation and intimate relationship between the immune and skeletal system is

highlighted in many inflammatory diseases. Periodontal disease, rheumatoid arthritis,

osteomyelitis and lupus are associated with both local and systemic bone destruction,

which inherently leads to bone fractures and significant bone deformities. However, the

mechanisms by which these pathological conditions arise remain poorly understood.

Beyond these observations, recent studies have also reported that the immune system

might play a role in facilitating bone formation. Osteoimmunological interactions thus

play a significant role in the maintenance of bone homeostasis and are fundamental

mechanisms in bone pathology.

To gain insights into the molecular genetics and mechanisms of osteoimmunology, we

have employed a chemical (ENU) mutagenesis and phenotypic assessment. Through

systematic X-ray screening, we identified a mouse line that displayed reduced bone

mass in comparison to wildtype littermates. The sanroque mutant mouse line carries a

single point (M199R) mutation in Roquin (Rc3h1) gene, which encodes for a RING-

type E3 ubiquitin ligase. E3 ubiquitin ligases, along with E1 and E2 family of enzymes,

are responsible for driving the ubiquitin proteasome system (UPS).

UPS is the process of ubiquitination of molecular products mainly for the purpose of

proteolysis to maintain protein homeostasis. The ubiquitination process marks the

molecular product of interest to be recognised by the proteasome for degradation. UPS

is reported to be involved in transcription factor activation, DNA repair, translational

control, kinase activation, histone activity and endocytosis. The abrogation of the UPS

enzymes, particularly E3s, has been implicated in the pathogenesis of many diseases

such as Parkinson’s, leprosy, osteosarcoma and autoimmune diseases. UPS and E3s are

also implicated in the maintenance of bone homeostasis and in the development of bone

cells.

Roquin gene encoding for Roquin E3 ubiquitin ligase is most abundantly expressed in

haematopoietic organs such as the bone and spleen. Within these organs haematopoietic

stem cells that give rise to common myeloid and lymphoid progenitors express high

levels of Roquin gene. Thus, it is not surprising that the sanroque mutant mice were

discovered to display changes in their immune system resulting in an autoimmune

condition consistent to systemic lupus erythematosus (SLE). The sanroque mutation

4

resulted in the dysregulation of a recently characterised T cell subset, the follicular

helper T cells (TFH), which are crucial in assisting the maturation of B cells within the

germinal centre. The sanroque mutant mice have increased numbers of germinal centres

and developed lymphomegaly and splenomegaly, both of which are key symptoms of

autoimmune disorders.

Q-PCR analysis revealed that Roquin gene expression is regulated throughout

osteoblast and osteoclast differentiation, whereby Roquin expression is elevated

throughout osteoblastogenesis. In contrast, the gene expression of Roquin is repressed

throughout osteoclastogenesis. Such regulation of Roquin expression in bone cells

highlights the potential regulatory role of the gene in bone biology.

The association of Roquin in the maintenance of bone homeostasis was clearly evident

when a low bone mass phenotype was observed in preliminary X-ray screening of

sanroque mice, which was then confirmed by microCT. MicroCT data emphasised the

reduction in both cortical and trabecular bone parameters in sanroque mice, both of

which are hallmarks of osteopenia, as compared to wildtype littermates. The low bone

mass phenotypes persisted in both genders of sanroque mutant mice. Therefore, the

sanroque mutation results in systemic bone loss in both male and female mice.

Histological analysis verified the reduced bone mass phenotype of sanroque mutant

mice as shown in microCT analysis. TRAP-stained long bone sections and whole

calvaria revealed an increase in osteoclast surface in sanroque mutant bone ex vivo. In

keeping with the osteopenic phenotype, in vivo calcein and alizarin red dual labelling

showed a reduction in bone mineral apposition rate in sanroque mutant mice. These

observations indicate that osteoclast and osteoblast coupling in vivo is affected by

sanroque mutation, resulting in dysregulation of bone homeostasis and reduced bone

mass.

To gain insight into the molecular mechanism of the pathogenesis of bone loss in

sanroque mutant mice, RNA isolated from bone tissues was subjected to reverse

transcription and Q-PCR analysis. Results showed elevated RANKL:OPG ratio in the

bone of sanroque mice relative to wildtype mice. Consistently, the expression of

osteoclast gene markers such as TRAP and cathepsin K were elevated in sanroque

mutant mice, whereas the expression of the osteoblast gene marker osteocalcin is

reduced in comparison to wildtype mice. This corresponds with the lower bone mass of

5

sanroque mutant mice, whereby increased RANKL:OPG in favour of RANKL usually

equates with increased osteoclastogenesis and bone resorption.

Additionally, flow cytometry analysis demonstrated that the sanroque mutation resulted

in skewed populations of stem cells and abnormal bifurcation of common progenitors

that give rise to dendritic cells, macrophages and osteoclast precursors. The constitutive

RANKL expression was accompanied by a significant expansion of putative osteoclast

progenitors within the bone marrow of sanroque mice in comparison to that of wildtype

littermates.

To gain insight in the cellular mechanism in the pathogenesis of bone loss in sanroque

mutant mice, flushed bone marrow macrophages (BMMs) were stimulated in vitro with

RANKL and M-CSF over the course of 5 days. Consistent with the observed increase in

osteoclast progenitors and osteoclast surface ex vivo, we also observed enhanced

osteoclastogenesis of BMMs derived from sanroque mutant mice, accompanied by

enhanced RANKL-mediated MAPK signalling.

The function and resorption activity of mature osteoclasts was examined by culturing

BMMs-derived osteoclasts on bovine bone discs. Microscopic analysis showed a

significant increase in the area resorbed by sanroque BMM-derived osteoclasts. This

observation is further supported by the increased cathepsin K released upon stimulation

of mature osteoclasts with bovine bone powder, suggesting an overzealous resorption

activity by sanroque BMM derived osteoclasts.

Previous Q-PCR analysis has also shown the upregulation of Roquin expression

throughout osteoblastogenesis, suggesting that the sanroque mutation could have an

effect on osteoblasts. Consistent with the reduction in MAR in vivo, sanroque calvarial-

osteoblasts formed less mineralised bone nodules as demonstrated by the apparent

reduction in alizarin red staining. This confirms the defective anabolic machinery of

osteoblasts in sanroque mutant mice.

The capability of osteoblasts to support osteoclastogenesis was also elucidated by

performing co-culture experiments between calvaria-derived osteoblasts and BMMs.

Upon stimulation with vitamin D3, sanroque calvaria-derived osteoblasts do not

produce as much osteoclastogenic factors as evidenced by the reduction in the number

of TRAP-positive osteoclasts. This observation suggests that the enhanced numbers of

6

osteoclast progenitors within the bone marrow niche in sanroque mice is not

contributed by osteoblastic production of osteoclastogenic cytokines, but by another

extrinsic factor or cell type.

During chronic immune activation, RANKL and osteoclastogenic factors expressed by

activated immune cells drive pathogenic osteoclast differentiation. Studies have

categorised many of these activated immune cells, such as TH17, and coined them

osteoclastogenic cell subsets. While elucidating the role of Roquin gene in the

pathogenesis of lupus, Vinuesa and colleagues (2005) reported that the sanroque

mutation resulted in elevated RANKL expression in dysregulated TFH that comprised

the immune system of sanroque mutant mice. Thus, the increased number of

dysregulated TFH residing in the germinal centres could possibly contribute to the

increased expression of RANKL in sanroque mutant mice.

To test the aforementioned hypothesis, spleens were cryopreserved, sectioned and

stained with antibodies specific for RANKL and cellular markers of TFH. Confocal

microscopy revealed an obvious increase in staining of RANKL in the sanroque spleen

sections in comparison to the wildtype. Co-localisation of TFH cellular markers and

RANKL was observed particularly within the marginal zones of the spleen from

sanroque mutant mice. To validate the increase in RANKL staining as observed in

confocal microscopy, spleen tissues were collected and subjected to protein isolation.

Western Blot analysis showed an increase in RANKL protein levels in the spleens of

sanroque mutant mice. Collectively, these results indicate that mutation in Roquin gene

resulted in elevated RANKL level in the spleen and that TFH could potentially be

categorised as an osteoclastogenic subset that drive pathogenic osteoclastogenesis.

In summary, sanroque mutant mice displayed reduced trabecular and cortical bone

volume due to an imbalance in bone homeostasis that favours bone resorption. A defect

in osteoblasts contributes to the reduced bone mass as they form less bone in vivo and in

vitro. Although Q-PCR analysis suggests that Roquin may negatively regulate

osteoclastogenesis, Roquin was demonstrated to play a crucial role in regulating the

expression level of RANKL, which in turn regulates osteoclastogenesis. In conclusion,

Roquin is an important regulator of osteoimmunological interaction and could serve as a

potential therapeutic target for the treatment of inflammatory-related systemic bone

loss.

7

ACKNOWLEDGEMENTS

This thesis is dedicated to my Popo and Ah Kong, both of whom while unable to be

with me as I finish this thesis, their memory has accompanied me throughout this

journey, and I hope I have made them proud of me.

My deepest appreciation and sincere gratitude for their guidance and support, is

extended to W/Prof. Jiake Xu and Dr. Jennifer Tickner, you both have been great

mentors throughout my journey in completing this thesis.

I am extremely grateful to W/Prof. Jiake Xu for the opportunity to undertake such an

interesting project. Your positivity, compassion and understanding have provided me

with the motivation to go through challenging times. The enthusiasm that you have for

the research is admirable and I am thankful for the knowledge you have imparted to me

throughout the years.

I wish to thank Dr. Jennifer Tickner, for your patience and guidance throughout the

years, which has enabled me to successfully get through the highs and lows that come

with doing scientific research. You are like the compass that has helped to point me in

the right direction as I navigated through my research.

I would also like to give a special thanks to Dr. Jacky Chim, for teaching me so many of

the vital laboratory techniques that I acquired today. I also appreciate your suggestions

and encouragement during challenging times. I would like to extend my thanks to

A/Prof. Nathan Pavlos for his continuous support and scientific advice. Your knowledge

in the bone field and broad outlook often taught me to think from a different

perspective.

I am also indebted to W/Prof. Wendy Erber and A/Prof. Senta Walton for their

generosity. In spite of having such busy schedules, they have shared their very valuable

time to help me with my project. I am grateful towards Irma Larma for the fun times

during training and also for helping me with the flow cytometry experiments. I would

like to acknowledge Euphemie Landao, Alysia Buckley, A/Prof. Paul Rigby and Tracey

Lee-Pullen as well, for their in-depth training and knowledge in histology, confocal

microscopy and flow cytometry. To Dr. Sarah Rae and Melanie Sultana, thank you for

having me in your lab and your contribution in my research.

8

I am also truly blessed to have the support and assistance from many of my awesome

colleagues. To Benjamin Ng, Dian Astari Teguh, Gaurav Jadhav, Jacob Kenny, Jin Bo,

Lin Zhou, Ryan Ming Li Yang, Vincent Kuek, William Chundawan and Dr. Jasreen

Kular, thank you for your endless encouragement, concern, support, assistance and

above of all, for the fun times, food and laughter that we shared. Your jokes and

friendly banter have helped to brighten up the sometimes-stressful situations I find

myself in. My special thanks and gratitude goes toward Dr. Pei Ying Ng for being such

a wonderful and caring housemate. I have enjoyed your cooking and company, and I

hope that we will be able to continue this friendship when I get to Singapore.

The completion of this thesis would not have been possible without the love, faith and

support from my loved ones. I am truly grateful to Ma and Pa for the unconditional love

and sacrifices that you both have made in order to give me the best in life. Your

immeasurable faith in me has given me inspiration and support. To my precious Bay

Gie and Betty, I must say I am the luckiest sister in the world to be blessed with two

pretty angels as my partners in crime. Thank you for making me smile and being there

for me.

To my dearest Sze Chieh, you filled my heart with so much love and I feel truly blessed

to have you in my life. I look forward to spending the rest of my life with you.

9

ABSTRACTS AND AWARDS

1. Lim B, Chim SM, Landao E, Tickner J, Pavlos NJ , Xu J (2012) “Roquin is a

novel regulator of bone homeostasis.” Sir Charles Gardner Hospital Medical

Research Month Symposium, Perth, Western Australia. (Oral Presentation)


novel regulator of bone homeostasis.” 19th Annual Australian and New Zealand

Orthopaedic Research Society Conference, Sydney, NSW. (Oral Presentation)


novel regulator of bone homeostasis.” 23rd Annual Australia and New Zealand

Bone & Mineral Society Scientific Meeting, Sydney, NSW. (Oral Presentation)


novel regulator of bone homeostasis.” UWA School of Pathology and

Laboratory Medicine Thesis Seminar, Perth, Western Australia. (Oral

Presentation)


novel regulator of bone homeostasis.” Australian Society for Medical Research

Medical Research Week Symposium, Perth, Western Australia. (Oral

Presentation) WA Department of Health Platinum Award


novel regulator of bone homeostasis.” 24rd Annual Australia and New Zealand

Bone & Mineral Society Scientific Meeting, Auckland, NZ (Poster Presentation)


novel regulator of bone homeostasis.” 20th Annual Australian and New Zealand

Orthopaedic Research Society Conference, Adelaide, SA. (Oral Presentation)

PhD Student Award

10

ABBREVIATIONS

AC Adenylyl cyclase

ALP Alkaline phosphatase

AP-1 Activator protein 1

APC Adenomatous polyposis coli protein

APC Allophycocyanin

APCs Antigen presenting cells

APS Ammonium persulfate

Atp6v0d2 d2 isoform of vacuolar (H+) ATPase (v-ATPase) V0 domain

BALP Bone alkaline phosphatase

Bcl B cell lymphoma

BMMs Bone marrow macrophages

BMPR BMP receptor

BMPs Bone morphogenetic proteins

BRU Bone remodelling unit

BSA Bovine serum albumin

BV/TV Bone volume per tissue volume

c-Cbl Casitas B-lineage lymphoma

c-fms Colony-stimulating factor 1

cAMP Cyclic adenosine monophosphate

CatK Cathepsin K

CBFA1 Core-binding factor alpha 1

CD Cluster of differentiation

CIA Collagen-induced arthritis

CMCA Centre for Microscopy, Characterisation and Analysis

COB Calvaria-isolated osteoblast

CT-1 Cardiotrophin-1

Ct.Ar Cross sectional cortical area

Ct.BV Cortical bone volume

Ct.Th Cortical thickness

CTLA-4 Cytotoxic T-lymphocyte protein 4

11

CTR Calcitonin receptor

CTX C-telopeptide of type I collagen

Cx 43 Connexin

CXCL CXC motif ligand

CXCR CXC chemokine receptor

DAG Diacylglycerol

DAMPs Damage-associated molecular patterns

DC Dendritic cell

DC-STAMP Dendritic cell-specific transmembrane protein

DD Death domain

DKK Dickkopf-related protein

DMEM Dulbecco's-Modification of Eagle's Medium

DMSO Dimethyl sulphoxide

DNA Deoxyribonucleic acid

DPX Distyrene plasticizer xylene

Dsh/Dvl Dishevelled

EDTA Ethylenediamine tetraacetic acid

ENU N-ethyl-N-nitrosourea

FBS Foetal bovine serum

FC Flow cytometry

FcRγ Fc-receptor common γ-subunit

FITC Fluorescein isothiocyanate

For Forward

FSC Forward scatter

Fz Frizzled

g Gram

G-CSF Granulocyte colony stimulating factor

GJIC Gap junctional intracellular communication

GM-CSF Granulocyte macrophage colony stimulating factor

Gp Glycoprotein

GSK3 Glycogen synthase kinase-3β

GST Glutatione S-transferase

H&E Haematoxylin and eosin

12

HECT Homologous to E6-AP carboxyl terminus

HSC Haematopoietic stem cell

ICOS Inducible T-cell co-stimulator

ICOSL ICOS ligand

IFNs Interferons

IGFs Insulin like growth factors

IKK IĸB kinase

IL Interleukin

IP3 inositol 1,4,5 triphosphate

IPTG Isopropyl-β-thiogalactopyranoside

IRAK IL-1R-associated kinases

ITAM Immunoreceptor tyrosine based activation motif

JAK Janus kinase

JNK c-Jun N-terminal kinase

kg Kilogram

Kif3a Kinesin family member 3A

kV Kilovolt

LAP Latency-associated peptide

LAT Linker of activation of T cells

LBOB Long bone-isolated osteoblast

LCS Lacunocanalicular system

LDL Low-density lipoprotein

LIF Leukaemia inhibitory factor

LRP LDL receptor-related protein

LTBP Latent TGF-β-binding protein

M-CSF Macrophage colony stimulating factor

M.Ar Marrow area

mA Milliampere

MafB V-maf musculoaponeurotic fibrosarcoma oncogene homolog B

MAPK Mitogen-activated kinases

MAPKs Mitogen activated protein kinases

MAR Mineral apposition rate

Mb Megabases

13

MCP Monocyte chemoattractant protein

MEA Methoxyethyl acetate

mg Milligram

MHC Major histocompatibility complex

MicroCT Microcomputed tomography

MITF Micro-opthalmia associated transcription factor

ml Millilitre

MLO-Y4 Murine long bone osteocyte Y4

mm Millimetre

mM Millimolar

MMA Methyl methacrylate

MMLV-RT Moloney leukaemia virus reverse transcriptase

MMPs Metalloproteinases

mRNA messenger RNA

MSC Mesenchymal stem cell

Mϕ Macrophage

N.Ob Number of osteoblast

N.Ob/BS Number of osteoblast per bone surface

N.OC/BS Number of osteoclast per bone surface

NCBI National Center for Biotechnology Information

NF-ĸB Nuclear factor kappa B

NFAT Nuclear factor of activated T cells

ng Nanogram

NIH National Institutes of Health

NKT Natural killer T cell

Oc.S Osteoclast surface

Oc.S/BS Osteoclast surface per bone surface

OCn Osteocalcin

OCP Commited osteoclast progenitors

OCT Optimal cutting temperature

OPG Osteoprotegerin

OPn Osteopontin

OS Osteoid surface

14

OS/BS Osteoid surface per bone surface

OSM Oncostatin M

Osx Osterix

PAMPs Pathogen-associated molecular patterns

PBS Phosphate buffered saline

PC Polycystin

PE Phycoerythrin

PI3K Phosphoinositide 3-kinase

PINP Procollagen type I N-terminal propeptide

PMSF Phenylmethylsulfonyl fluoride

PPRs Pattern recognition receptors

PTH Parathyroid hormone

Q-PCR Quantitative realtime PCR

RA Rheumatoid arthritis

RANK Receptor activator of NF-ĸB

RANKL RANK ligand

RBC Red blood cell

Rev Reverse

RGD Arginine-glycine-aspartic acid

RING Really interesting new gene

RIPA Radio immunoprecipitation assay buffer

Rn18S 18S ribosomal RNA

RNA Ribonucleic acid

RORγt RAR-related orphan receptor gamma

RT-PCR Reverse transcription polymerase chain reaction

S1P Sphingosine 1-phosphate

SA Spondylarthritis

san sanroque

SBE SMAD-binding element

SDF Stromal cell-derived factor

SDS Sodium dodecyl sulphate

SDS-Page Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

SEM Standard error of the mean

15

Sema4D Semaphorin-4D

SLE Systemic lupus erythematosus

SMADs Sma and Mad related proteins

SMI Structure model index

SOST Sclerostin

SSC Side scatter

STAT Signal transducer and activator of transcription

Syk Spleen tyrosine kinases

TAE Tris-acetate EDTA

TAK1 TGF-β-activated kinase 1

Tb.N Trabecular number

Tb.Sp Trabecular separation

Tb.Th Trabecular thickness

TBS Tris buffered saline

TCF/LEF Transcription factor lymphoid enhancer binding factor 1/T cell-

specific transcription factor

TCR T cell receptor

TEMED Tetramethylethylenediamine

TFH Follicular helper T cell

TGF Transforming growth factor

TH T helper

TLRs Toll-like receptors

TNF Tumor necrosis factor

TNFRSF TNF receptor super family

TRACP-5b Tartrate-resistant acid phosphatase 5b

TRAF TNF-receptor associated factor

TRANCE TNF-related activation-induced cytokine

TRAP Tartrate-resistant acid phosphatase

Tregs Regulatory T cells

TRF T cell-replacing factor

Tt.Ar Total cross sectional area

UPS Ubiquitin proteasome system

UTR Untranslated region

16

UV Ultraviolet

v/v Volume/volume

w/v Weight/volume

WB Western blot

Wnts Wingless type proteins

WP White pulp

WT Wildtype

ZA Zoledronic acid

α-MEM α-Modification of Eagle's Medium

μg Microgram

μl Microlitre

μm Micrometre

μM Micromolar

17

LIST OF FIGURES

Page

Figure 1.1 The different types of bone cells involved in the regulation of bone

microenvironment

38

Figure 1.2 RANK/RANKL signalling pathway during osteoclastogenesis 39

Figure 1.3 The bone remodelling cycle 40

Figure 2.1 Lineage bifurcation model between dendritic cells, macrophages and

osteoclasts

66

Table 2.1 The effect of cytokines in modulating the bone and immune system 67

Figure 3.1 Mapping of missense mutation in the conserved Rc3h1gene 73

Figure 3.2 The basic steps of substrate modification by the ubiquitin

proteasome system

74

Figure 3.3 Mutant sanroque mice appeared normal but exhibited low bone

phenotype

75

Figure 3.4 Mutant sanroque mice displayed lymphomegaly and splenomegaly 76

Figure 5.1 Tissue distribution of Roquin (Rc3h1) mRNA expression 131

Figure 5.2 Expression level of Roquin throughout osteoblast (OB) and

osteoclast (OC) differentiation

132

Figure 5.3 Mutant sanroque mice display reduced bone mass 133

Figure 5.4 Histological analysis showed mutant sanroque mice have reduced

trabecular bone mass

135

Figure 5.5 Increase osteoclast number in trabecular bone of sanroque mutant

mice

136

Figure 5.6 Increased TRAP staining on the surface and in sections of sanroque

mutant calvaria

137

Figure 5.7 Osteoblast number and osteoid surface are unaffected in the

trabecular bone of sanroque mutant mice

138

Figure 5.8 Decreased in vivo bone mineral apposition rate in sanroque mutant

mice

139

Figure 5.9 Sanroque mutant bones expressed higher levels of osteoclast gene

markers and increased RANKL:OPG ratio

140

18

Figure 6.1 Osteoblasts isolated from calvariae (COBs) of sanroque mutant

mice have reduced capability to mineralise in vitro

152

Figure 6.2 Enhanced mineralisation in sanroque mutant osteoblasts isolated

from long bones (LBOBs)

154

Figure 6.3 Mutant LBOBs of sanroque mutant mice displayed atypical alkaline

phosphatase (ALP) activity levels throughout osteogenesis

157

Figure 6.4 Q-PCR analyses of osteoblast gene markers expression levels during

osteogenesis

158

Figure 6.5 Coculture systems between LBOBs and bone marrow macrophages

(BMMs)

159

Figure 7.1 Osteoclast differentiation is increased in sanroque BMM culture 174

Figure 7.2 BMMs derived from sanroque mutant mice have increased capacity

to form osteoclasts at lower doses of RANKL

175

Figure 7.3 The mutation in Roquin alters the expression of osteoclast gene

markers.

173

Figure 7.4 Sanroque mutant osteoclasts resorbed more bone 177

Figure 7.5 Increased cathepsin K released by sanroque mutant osteoclasts 178

Figure 7.6 Increased activation of MAPK signalling cascade during RANKL

stimulation in sanroque osteoclast precursors

179

Figure 7.7 Increased protein levels of transcription factor NFATc1, osteoclast

markers DC-Stamp and D2 subunits during osteoclastogenesis in

sanroque BMMs

181

Figure 7.8 FLAG-tagged TRAF6 immunoprecipitates of HEK293 cells

transfected with wildtype and mutant Roquin (M119R) plasmids

showed no difference in ubiquitination level by immunoblotting

with anti-HA

183

Figure 7.9 No difference in ubiquitination of TRAF6 in sanroque BMMs upon

RANKL stimulation

185

Figure 7.10 Blood serum from sanroque mutant mice contains proliferative-

inducing cytokines and sanroque BMMs exhibit increased cell

proliferation in response to M-CSF

186

Figure 8.1 The bone marrow population of sanroque mutant mice appeared

dysregulated

200

19

Figure 8.2 Increased macrophage and reduced dendritic cell population in

sanroque bone marrow

201

Figure 8.3 Significant expansion of putative osteoclast progenitors in sanroque

bone marrow

202

Figure 8.4 No difference in the percentage of CD4 and CD8 T cell population

between the bone marrow of sanroque mutant mice and wildtype

204

Figure 8.5 Expression of CD70 in T, B and DCs 205

Figure 8.6 Increased follicular helper T cell in sanroque mutant mice 208

Figure 8.7 Localisation of RANKL and TFH in spleen 209

Figure 8.8 RANKL protein and gene expressions are elevated in sanroque

spleens

211

Figure 8.9 Increased common myeloid progenitors in sanroque spleen 212

Figure 9.1 The hypothetical model depicting the possible cellular mechanism

that resulted in systemic bone loss in sanroque mutant mice

224

Supp. Fig. 1 Female mutant sanroque mice display reduced bone mass 277

Supp. Fig. 2 Left shift is observed in peripheral blood smears from sanroque

mutant mice

278

CHAPTER ONE

THE BIOLOGY OF BONE MULTICELLULAR UNIT

AND

TRANSCRIPTIONAL REGULATIONS OF BONE

HOMEOSTASIS

Chapter 1: The Biology of Bone Multicellular Unit and Transcriptional Regulations of Bone Homeostasis

21

1. CHAPTER 1: THE BIOLOGY OF BONE MULTICELLULAR UNIT AND TRANSCRIPTIONAL REGULATIONS OF BONE HOMEOSTASIS

1.1. Introduction

Bone is a specialized connective tissue that, collectively with cartilage, makes up the

skeleton (Meghji, 1992). The calcified tissue is composed of 5 – 10% water, 60 – 70%

inorganic (mineral) components, with organic component comprising the rest of the

tissue. The major component of the mineral phase is hydroxyapatite crystal. The

organic phase is comprised of concentric layers of type I collagen matrix, a variety of

non-collagenous proteins, and bone cells (Feng, 2009; Lee et al., 2001; Morgan et al.,

2008).

The architecture and composition of bone tissue have evolved to dynamically carry out

several major functions (Bab et al., 1994). Bone tissue provides mechanical support and

attachment sites for soft tissues such as muscles for locomotion. Other purposes of bone

include protecting vital organs and housing the brain and the spinal cord (Feng, 2009;

Harada et al., 2003; Meghji, 1992). The bone represents the primary reservoir of

calcium and other mineral ions such as phosphate, sodium and magnesium (Duplomb et

al., 2007; Rubin et al., 1999). Several studies have indicated that bone is a major

component of the immune system as immune cells form in the bone marrow (Arron et

al., 2000b; Carlsten, 2005; Morgan et al., 2008; Takayanagi, 2005; Wein et al., 2005;

Wu et al., 2008b). This is inevitable as it is also the principal site of haematopoiesis,

where it houses pluripotent bone marrow stem cells (Horowitz et al., 1992).

To fulfil a lifelong execution of its versatile functions while maintaining structural

integrity, the bone tissue is continuously being broken down and rebuilt in a process

known as bone remodelling. Bone remodelling is a process coupled between the

catabolic effects of bone resorbing osteoclasts and anabolic effects of bone forming

osteoblasts (Henriksen et al., 2009; Horowitz et al., 1992). Under normal states of bone

homeostasis, an ideal bone remodelling process serves to remove bone mass where the

mechanical loads are low, while forming bone at sites where mechanical stimuli are

transmitted repeatedly. Bone therefore has the capability to maintain itself, depending

on the external mechanical and physiological stimuli from the systemic environment.

The cellular coupling between osteoclastic and osteoblastic cells is complex and is

heavily coordinated by several regulatory systems, both systemic and local, to keep

both remodelling and resorption processes synchronised (Martin et al., 2006; Morgan et

al., 2008). The accentuation of one or the other process eventually leads to bone


22

fragility and/or clinical diseases of the skeleton, such as osteopenia, osteoporosis,

osteopetrosis, periodontitis, arthritis and tumour-induced osteolysis (Hill, 1998; Karsdal

et al., 2007). While these abnormalities are known to associate with an increased risk

for fracture and causing considerable suffering, little is known of the mechanisms

responsible for the dysfunctional bone remodelling that characterises them. This is not

surprising, since we have yet to fully understand the complexity of “normal” bone

remodelling and how the process is so highly coordinated and balanced.

However, the aim to further understand the biochemical and molecular mechanisms of

bone cells to adapt within their mechanical environment is achievable via novel

techniques such as gene ‘knockout’, silencing and transgenic experiments. Molecular

biology and the analysis of the skeletal phenotype of mouse models replicating genetic

mutations in humans has led to a better understanding of the role of factors that govern

bone cell biology, and ultimately, bone homeostasis. Understanding the functions of the

cells within bone, and how they communicate with each other during bone remodelling

is essential to the development of therapeutic options for bone diseases.

1.2. Bone cells

1.2.1. Osteoclasts

Osteoclasts are motile macrophage-like cells that undergo mitosis without cytokinesis

and become multinucleated (Alberts, 2008). The multinucleated osteoclasts serve as

bone-resorbing cells by eroding bone, enabling the tissue to be remodelled during

growth and/or in response to stresses throughout life. Bone resorption is an obligatory

event during bone growth, tooth eruption, fracture healing and the maintenance of blood

calcium level (Vaananen et al., 2000).

Multinucleated osteoclasts are derived from hematopoietic stem cells and are created by

the differentiation of monocyte/macrophage precursor cells at or near the bone surface

(Boyle et al., 2003; Edwards et al., 2011). Hence, osteoclasts share a common pathway

to that of monocyte/macrophages and dendritic cells and the bone resorption process

utilizes similar cellular machinery as phagocytosis (Morgan et al., 2008; Vaananen et

al., 2000).

Depicted in Figure 1.1, the initial step of osteoclastogenesis is the determination of the

stem cell precursor induced by the erythroblast transformation specific (ets)

transcription factor family member PU.1 encoded by Spi-1 gene. At this stage, the cells


23

acquire the colony-stimulating factor 1 (c-fms) receptor that binds to macrophage

colony stimulating factor (M-CSF) to stimulate cell survival and proliferation (Fuller et

al., 1993; Zhang et al., 1994). The initiation of cell proliferation is induced by the

phosphorylation of several kinases, importantly Src, Grb2 and Phosphoinositide 3-

kinase (PI3K), which leads to the activation of cyclin D. M-CSF activates Casitas B-

lineage lymphoma (c-Cbl) for the ubiquitination and degradation of the pro-apoptotic

Bim to ensure the survival of osteoclasts (Akiyama et al., 2003; Blair et al., 2008;

O'Brien, 2009). With the appearance of proto-oncogene c-fos and Receptor Activator of

NF-ĸB (RANK), the progenitors are destined to form osteoclasts (Blair et al., 2008).

Around 1997, several independent laboratories identified the essential osteoclastogenic

factor Receptor Activator of Nuclear Factor Kappa B Ligand (RANKL, also called

TRANCE) (Anderson et al., 1997; Matsuo et al., 2008; Wong et al., 1997a; Wong et

al., 1997b). It is an established fact that RANKL, a trans-membrane glycoprotein and a

member of the Tumor Necrosis Factor (TNF)-α superfamily of cytokines, is essential

for efficient osteoclast differentiation. Numerous cell types such as stromal cells,

osteoblast precursors, mature osteoblasts, osteocytes, chondrocytes and lymphocytes

express RANKL. RANKL has two receptors: RANK, and osteoprotegerin (OPG). The

actions of RANKL are opposed by a soluble TNF-receptor OPG. OPG serves as a

decoy receptor by binding to RANKL and limiting the signalling that induces

osteoclastogenesis (Blair et al., 2008; Miyamoto et al., 2003; O'Brien, 2009; Simonet et

al., 1997; Wada et al., 2006). OPG thus inhibits formation of osteoclasts. OPG-/- mice

exhibit severe osteoporosis due to lack of OPG and excessive osteoclastic bone

resorption (Mizuno et al., 1998).

The binding of RANKL to the extracellular domain of RANK activates TNF receptor-

associated factor (TRAF) adaptor or docking proteins and is considered the key

preliminary step in RANK signalling (Boyle et al., 2003). TRAF-2, -5, and particularly

TRAF-6, are all shown to bind to RANK. TRAF6 then phosphorylates IKK (IĸB

kinase), which ubiquitinates IĸB to liberate the transcription factor NF-ĸB and allow its

translocation to the nucleus for the transcription of osteoclast specific genes. RANK-

RANKL interaction also induces at least other four distinct signalling cascades – c-Jun

N-terminal kinase (JNK), mitogen activated protein kinases (MAPKs) including

extracellular signal-regulated kinase (ERK) and p38, and Src pathways to ensure the

survival and maturation of preosteoclast cells (Boyle et al., 2003; Duplomb et al., 2008;


24

Sundaram et al., 2007; Xu et al., 2009a) (Figure 1.2).

As the preosteoclasts differentiate, PU.1 also interacts with micro-opthalmia associated

transcription factor (MITF), which targets genes that encode essential molecules in

normal osteoclast machinery, such as Tartrate-resistant acid phosphatase (TRAP) and

carbonic anhydrase (Luchin et al., 2000; So et al., 2003). This is followed by

multinucleation brought by cell fusion of the mononuclear preosteoclasts (Miyamoto et

al., 2003). Studies have shown that dendritic cell-specific transmembrane protein (DC-

STAMP) and the d2 isoform of vacuolar (H+) ATPase (v-ATPase) V0 domain

(Atp6v0d2) are essential for the fusion of osteoclasts. The deficiency of DC-STAMP

and Atp6v0d2 did not affect osteoclast differentiation into TRAP positive cells in vivo,

however the maturation into large multinucleated cells was impaired (Kukita et al.,

2004; Lee et al., 2006; Yagi et al., 2005; Yagi et al., 2007).

It is also reported that stimulation of RANKL leads to the activation of

calcineurin/NFAT (nuclear factor of activated T cells) signalling pathway and

upregulation of the transcription factor NFATc1 expression throughout osteoclast

maturation (Figure 1.2). An in vitro study has demonstrated that constitutively active

calcineurin-independent NFATc1 mutant expression is adequate to induce

osteoclastogenesis of RAW264.7 cells (Hirotani et al., 2004). The activation of RANK

initiates the recruitment of spleen tyrosine kinases (Syk) by DAP12 or FcRγ, both of

which are immunoreceptor tyrosine based activation motif (ITAM)-harbouring

adapters. Syk activates phospholipase Cγ to release Ca2+ from intracellular stores. This

event triggers the calmodulin-regulated phosphatase calcineurin to dephosphorylate the

transcription factor NFATc1. NFATc1 undergoes nuclear translocation and, together

with c-fos, magnifies the expression of critical osteoclast genes (Blair et al., 2008;

Mocsai et al., 2004; Sundaram et al., 2007). These findings thus indicated that

calcineurin is an essential downstream effector of RANKL signalling pathway and that

NFATc1 is involve in regulating osteoclastogenesis and cell function in response to

RANKL stimulation.

1.2.2. Osteoblasts

Osteoblasts are the primary cells responsible for bone formation. They originate from

mesenchymal stem cells (Figure 1.1), which have the potential to differentiate into other

musculoskeletal tissues such as cartilage, fat, muscle, ligament and tendon (Pittenger et


25

al., 1999). The commitment of these stem cells to the osteoblast lineage is highly driven

by the growth factors wingless type proteins (Wnts) and bone morphogenetic proteins

(BMPs) (Clarke, 2008; Fakhry et al., 2013; Rosen, 2009).

The significance of the Wnt family of signalling proteins in the maintenance of bone

homeostasis is well documented. Mutations in genes encoding intracellular Wnt

pathway proteins lead to diverse conditions such as osteoporosis pseudoglioma

syndrome (Gong et al., 2001) and high bone mass disorders (Boyden et al., 2002; Little

et al., 2002). The main consequence of Wnt signalling is the activation of β-catenin

dependent transcription. In the absence of Wnt receptor activation, cytoplasmic β-

catenin remains phosphorylated by the axin scaffold proteins, adenomatous polyposis

coli protein (APC), glycogen synthase kinase-3β (GSK3) and casein kinase I.

Phosphorylation of cytoplasmic β-catenin results in its ubiquitination by E3 ubiquitin

ligase β-TrCP, and subsequent proteasomal degradation (Aberle et al., 1997; Su et al.,

2008). In contrast, binding of Wnt proteins to the Frizzled (Fz)/low-density lipoprotein

(LDL) receptor-related protein (LRP) complex at the cell surface activates Dishevelled

(Dsh/Dvl), which inhibits the phosphorylation of β-catenin. This circumvents the

degradation pathway to allow the stabilisation and accumulation of β-catenin in the

cytoplasm and nucleus. Nuclear β-catenin then interacts with the transcription factor

lymphoid enhancer binding factor 1/T cell-specific transcription factor (TCF/LEF) to

mediate gene transcription (Huber et al., 1996).

BMPs are multifunctional growth factors that belong to the transforming growth factor

β (TGF-β) superfamily (Rosen, 2009). They were first identified based on their ability

to initiate ectopic bone formation (Urist, 1965; Wozney et al., 1988) and this unique

feature has allowed for their successful use as therapeutic agents for bone regeneration.

BMPs bind to two types of serine-threonine receptors, BMP receptor (BMPR) type I

and type II. Binding of BMP to BMPR-II leads to the recruitment of BMPR-I to form

an activated heteromeric complex. This complex then phosphorylates and activates

intracellular signal-transducing molecules of the TGF-β superfamily termed Smad

proteins. Receptor-regulated Smads 1, 5 and 8 bind to a co-Smad 4 and then translocate

to the nucleus to initiate transcription (Chen et al., 2012; Fujii et al., 1999). The

survival of mesenchymal stem cells within the bone marrow microenvironment is

maintained by BMP activity (Solmesky et al., 2009; Yang et al., 1999). Endogenously

produced BMPs are crucial survival factors for human MSCs and these findings


26

highlight the involvement of BMP signalling in maintenance of the skeletal stem cell

niche in adult bone.

Once skeletal stem cells are committed to the osteoblast lineage, the helix-loop-helix

proteins Twist and Id ensure the proliferation of preosteoblasts to continue, albeit with

reduced plasticity. The osteoblast precursor cells eventually stop proliferating and begin

secreting type I collagen – the basic building block of bone. These cells also produce

non-collagenous proteins, such as osteocalcin (small calcium binding protein specific to

bone that is carboxylated by vitamin K) and alkaline phosphatase, which are crucial for

mineral deposition. Proteoglycans that modulate collagen fibril formation, and

glycoproteins that regulate mineralization and matrix organization are also produced by

osteoblast precursors. Also several arginine-glycine-aspartate-(RGD)-containing-

glycoproteins that are important in osteoclast matrix interactions and osteoblast

attachment are also secreted by the precursors. These precursors resemble active

osteoblasts, however they lack some of the characteristics including the ability to

mineralise bone (Blair et al., 2008; Kular et al., 2012; Ling et al., 2005; Zittermann,

2001).

Preosteoblasts then mature into mineralization competent cuboidal osteoblasts, marking

the final stage of osteoblast differentiation. Cuboidal osteoblasts have a large eccentric

nucleus with prominent rough endoplasmic reticulum and Golgi areas, all of which are

attributes of typical secretory cells (Lian et al., 2001; Lucht, 1972; Tasat et al., 2007).

They no longer divide and can be distinguished from preosteoblasts by the upregulation

of bone markers such as bone sialoprotein, osteocalcin, BMPR-I, vitamin D3 receptor

and type I collagen (Franz-Odendaal et al., 2006). Besides their role in bone formation,

at this stage the osteoblastic cells are involved in the recruitment and maintenance of

osteoclasts by expressing M-CSF, RANKL and OPG (Simonet et al., 1997; Takahashi

et al., 1991; Udagawa et al., 1999).

Mature osteoblasts have one of the following four fates – (1) they may undergo

apoptosis, those that survive (2) coalesce into the heterogeneous population osteolining

cells or (3) eventually become encased and trapped within the matrix to become

osteocytes, or in some scenarios, (4) transdifferentiation into cells that are capable of

chondrogenesis (Redlich et al., 2004). The proportion of osteoblasts following each

fate varies in all mammals and is not conserved among different types of bone. The age


27

of the mammal may also influence the number of osteoblasts that transform into

osteocytes (Franz-Odendaal et al., 2006). In human cancellous bone, 65% of the

osteoblasts undergo apoptosis and only ~30% transform into osteocytes (Parfitt, 1990).

1.2.3. Osteocytes

Once assumed to be an inert cell, the osteocyte has gained much attention in the recent

years as one of the central responder to various stimuli involved in the maintenance of

skeletal and mineral homeostasis. Their importance is highlighted by the fact that

osteocytes are the most ubiquitous cellular component of bone situated within the bone

matrix. Derived from mature osteoblasts, the prospective cuboidal cells undergo drastic

morphological changes and differentiate into stellate-shaped osteocytes with multiple

dendritic conduits (Dudley et al., 1961; Palumbo, 1986; Palumbo et al., 1990).

Osteocytes form a functional multicellular syncytium by maintaining a complex

network of long filipodial conduits composed of actin filament (Tanaka-Kamioka et al.,

1998). These dendritic processes radiate out of their confined lacunar space via

microscopic canals (canaliculi), to regulate and communicate with neighbouring

osteocytes and with other bone cells and residential cells (Cao et al., 2011; Neve et al.,

2012; Talmage, 1970). The gap junction channel, an intercellular hemichannel

constructed from transmembrane proteins where tips of the cytoplasmic processes meet,

is enriched with connexin 43 (Cx43) (Doty, 1981; Jones et al., 1993; Kamioka et al.,

2007). This gap junction protein allows the diffusion of molecules with a molecular

weight of less than 1kDa and has been reported to play a protective role against

osteocyte apoptosis (Kar et al., 2013; Plotkin et al., 2002). In vitro experiments using

the MLO-Y4 osteocyte-like cell line show that fluid flow significantly increased the

length of the dendritic extensions and redistribution of Cx43 (Cheng et al., 2001).

Furthermore, gap junctional intracellular communication (GJIC) in MLO-Y4 cells is

amplified when they are subjected to low intensity vibration, suggesting that gap

junctions may serve as channels for signals generated by osteocytes in response to

mechanical strains to mediate cellular communication (Uzer et al., 2014).

It has also been reported that osteocytes display protuberances that project from the

body. Osteocytes retain centrosomes which give rise to primary cilia, and the MLO-Y4

cell line has been found to express transcripts for glycoproteins polycystin (PC) 1 and

PC2, and the ciliary proteins Tg737 and Kif3a (Uzbekov et al., 2012; Xiao et al., 2006).


28

These projections of primary cilia on the cell surface are perturbed by bone fluid that

flows through the lacuna-canalicular system. In vitro immunohistochemistry studies

have indicated that the translation of the mechanical stimulus into intracellular

responses by the organelle relies on a molecular mechanism involving membrane-

bound enzyme adenylyl cyclase (AC) 6 and intracellular cyclic adenosine

monophosphate (cAMP) (Kwon et al., 2010; Malone et al., 2007).

Another study revealed that osteocytes in the lacunocanalicular system (LCS) of intact

mouse tibiae displayed intracellular calcium oscillation in response to in situ dynamic

loading. The frequency of calcium spikes correspond to the load magnitude, tissue

strain and sheer stress in the LCS (Jing et al., 2014). These aforementioned cellular

structures of osteocytes and their responses to mechanical strains are some of their

features that lead scientists to hypothesise the stellate cells are the key regulators of

bone mechanotransduction. As mechanosensors of bone, the osteocytic network detects

local mechanical stimulations and in response converts them into biochemical signals

by expressing membrane-bound and releasing soluble factors that coordinate the

function of other bone cells, thus orchestrating the synchronisation of bone cell activity

(Iolascon et al., 2013).

The direct and indirect regulation of osteoblast and osteoclast activity by osteocytes has

been reported. Mechanically stimulated MLO-Y4 cells have been reported to regulate

osteoblast anabolic activity via gap junctions in which osteocytic-osteoblastic cell

contact is essential (Taylor et al., 2007). In vivo studies employing mechanical loading

and unloading experiments in several mouse models with osteoblast/osteocyte-specific

loss of Cx43, cilial Kif3a and AC-6 resulted in altered bone formation (Chung et al.,

2006; Grimston et al., 2008; Lee et al., 2014; Lloyd et al., 2012; Temiyasathit et al.,

2012; Zhang et al., 2011b). Additionally, transgenic osteocyte-ablated mice are resistant

to unloading-induced bone loss, albeit displaying an osteoporotic phenotype

accompanied by osteoblastic dysfunction and microfractures (Tatsumi et al., 2007).

In vivo fracture healing studies revealed osteoblast/osteocyte-specific ablation of Cx43

led to a decrease in GJIC and subsequent reduction in bone mineralisation, recruitment

of osteoclasts and eventually delayed bone formation and healing (Loiselle et al., 2013).

Regions of osteocyte apoptosis due to microdamage have been shown to coincide with

increased local bone turnover and targeted bone resorption (Cardoso et al., 2009; Clark


29

et al., 2005; Gu et al., 2005; Noble et al., 1997; Verborgt et al., 2000). Further research

demonstrated that apoptotic osteocytes regulate the recruitment and differentiation of

osteoclast progenitors through the neighbouring non-apoptotic osteocytes that have

responded to the apoptosis cues (Al-Dujaili et al., 2011; Kennedy et al., 2012). All of

these data support the hypothesis that functional osteocytes are crucial bone

mechanotransducers and osteocytes communicate with other bone cells via GJIC and

other factors in response to physical stimuli.

1.2.4. Osteolining cells

Residing on quiescent (non-remodelling) bone surfaces, osteolining cells form a cellular

layer that separates bone tissue from the marrow. Osteolineage lining cells display a

thin, flat nuclear profile and contain a few cell organelles including mitochondria,

microfilaments, free ribosomes and rough endoplasmic reticulum. The cells are also

reported to display cell processes and are connected with one another via gap junctions

(Miller et al., 1987).

Originally thought to be of endothelial origin, these cells are negative for the expression

of endothelial cell marker and are reported to be positive for osteoblastic markers such

as alkaline phosphatase, osteocalcin and osteonectin (Hauge et al., 2001). Despite the

fact that they express osteoblastic gene markers and are hypothesised to be terminally

differentiated osteoblasts, the origin and the role of osteolining cells are still highly

debatable. While exhibiting a non-mitotic and dormant cell-morphology that is hardly

engaged in the maintenance of bone homeostases (Menton et al., 1984), experimental

evidence has suggested many functions of the osteolining cells, particularly when they

are “re-activated” in response to bone remodelling.

Early SEM studies reported the presence of a fluid compartment between osteolining

cells and bone tissue, which extends into the bone matrix itself and around osteocytes in

the lacunae (Baud, 1968). Osteolining cells were postulated to maintain this space.

Furthermore, it has been documented that osteolining cells possess numerous

protoplasmic extensions that extend into bone and make contact with cell processes of

osteocytes (Talmage, 1969; Talmage, 1970; Talmage et al., 1970), implicating their

possible role in supporting osteocytic mechanotransduction.

Osteolining cells have been considered as potential osteogenic precursor cells especially

in an event that initiates rapid bone formation. An event such as mechanical loading


30

was proposed to increase bone formation by driving osteoprogenitor cells in the bone

marrow to progress through the cell cycle and differentiate into osteoblasts at the bone

surfaces. In rats, the proliferation, differentiation and recruitment of osteoprogenitors to

the bone surfaces require 72 hours post-loading to establish. However, endocortical

osteoblast surface was observed to be significantly increased at 48 hours after loading

stimulus, suggesting that there is an immediate response from existing cells on the bone

surface (Turner et al., 1998). This observation is also supported by Chow et al., who

have documented the morphological evolution of osteolining cells towards osteoblastic

cells after mechanical stimulation (Chow et al., 1998).

A similar immediate increases in bone forming surface is also observed upon

administration of parathyroid hormone (PTH) (Cho et al., 2014; Dempster et al., 2001;

Gunness-Hey et al., 1984; Hamann et al., 2014; Jilka et al., 1999), and since no cell

proliferation and recruitment of osteoblasts was noted on the bone surface, it has also

been proposed that the anabolic effect of PTH is likely to be due to the activation and

differentiation of pre-existing osteolining cells towards osteoblastic cells (Dobnig et al.,

1995; Leaffer et al., 1995; Onyia et al., 1995). Ablation of bone marrow cells using

high dose radiation is reported to be associated with a transient increase in bone

formation due to activation of osteoblastic lining cells. This is accompanied by an

increase in bone resorption leading to an increased bone turnover rate (Turner et al.,

2013).

Osteolining cells are also implicated in the phagocytosis of collagen fibrils and

mineralised bone particles (Takahashi et al., 1986). Osteolining cells are described to

display catabolic activities by secreting matrix metalloproteinases (MMPs) to degrade

bone matrix (Dierkes et al., 2009; Everts et al., 2002). The capability of these

osteolining cells to have both anabolic and catabolic activities indicates that these cells

are a population of different types of cells that serve multiple functions in maintaining

bone homeostasis. Another subpopulation of osteolining cells is the residential

macrophages (Hume et al., 1984). These resident osteal antigen F4/80 macrophages

found intercalated among osteolining cells are termed osteomacs. They are implicated

in supporting bone mineralisation in vitro and the maintenance of bone homeostasis

through regulation of osteoblast maturation and function in vivo (Chang et al., 2008).

Sharing the same common precursor as osteoclasts, osteomacs are TRAP negative and

localised to sites of bone modelling in vivo, where they formed a distinctive canopy


31

structure over mature osteoblasts (Alexander et al., 2011; Chang et al., 2008; Cho et al.,

2014). Osteomacs participate in intramembranous healing and are required for the

deposition of type I collagen and bone mineralisation in tibial injury model (Alexander

et al., 2011). The depletion of these osteal macrophages lead to a reduction of bone

mass and attenuates the anabolic effect of PTH further supporting their positive role in

the maintenance of bone homeostasis (Cho et al., 2014).

1.3. Formation and maintenance of bone

1.3.1. Osteogenesis

Osteogenesis is sculpted by three distinct lineages – the somites, lateral plate mesoderm

and cranial neural crest. The somites and the lateral plate mesoderm give rise to axial

and limb skeleton respectively, whilst the cranial neural crest generates the branchial

arch, craniofacial bones and cartilage. There are two modes of osteogenesis:

intramembranous and endochondral ossification (Gilbert, 2000b). In both processes, the

initial bone formation is made up of disorganised woven collagen fibres, termed

primary bone, which is then replaced by highly organised concentric lamellae to form

architecturally strong secondary bone (Shapiro, 2008; Teti, 2011).

The process of direct condensation of mesenchymal cells that eventually differentiate

into bone-forming osteoblasts to produce flat bones (e.g. skull, scapula and ileum) is

termed intramembranous ossification. The process of intramembranous ossification is

driven by activation of the transcription factor Core-binding factor alpha 1 (CBFA1) by

bone morphogenetic proteins (BMPs) (Ducy et al., 1997; Yamaguchi et al., 2000).

Intramembranous ossification begins with the proliferation and condensation of neural

crest-derived stem cells into compact nodules. Some of these stem cells undergo

angiogenesis to develop capillaries; others undergo osteoblastogenesis to secrete pre-

bone matrix (osteoid) that is made up of collagen-proteoglycan and is able to bind

calcium salts (Gilbert, 2000a; Steele et al., 1988). As the osteoid calcifies, woven bone

spicules emanate out from the region where ossification began. The mesenchymal cells

enveloping the calcified spicules differentiate into a membrane termed periosteum.

Stem cells on the inner surface of the periosteum undergo osteoblastogenesis to secret

more osteoid matrix directly into the condensed mesenchyme, entrapping late-stage

osteoblasts which will evolve into osteocytes. The cycle continues to form many layers

of bone proceeding outward from the nodule centre (Gilbert, 2000a; Teti, 2011).


32

The formation of endochondral bone, which includes the long bones and vertebrae, can

be divided into five stages. Endochondral ossification begins with the condensation and

differentiation of mesenchymal stem cells into a template composed of type II collagen

produced by chondrocytes. These early stages are driven by the transcription factor

SOX-9 (Allen et al., 2014). During the next phase, the chondrocytes proliferate rapidly,

expanding the cartilage template. A subset of the chondrocytes stop dividing and

become hypertrophic. The cartilage then undergoes vascular invasion and

mineralization. During this process, the large hypertrophic chondrocytes undergo

apoptosis, leaving space that will become bone marrow spaces that are loosely

compartmentalised. As the cartilage cells die, osteoprogenitors are recruited and

differentiate into osteoblasts. The osteoblasts begin to secrete osteoid to replace the

partially degraded cartilage. Some cartilage is retained at the ends of the bone at the

surface lining the joints to provide low-friction support (articular cartilage), and in the

growth plates, where further proliferation of chondrocytes allows bones to grow in

length (Blair et al., 2008; Boyce et al., 2008; Gilbert, 2000a).

1.3.2. Bone modelling and remodelling

Osteogenesis and skeletal growth require both bone modelling and remodelling. Bone

modelling is characterized by a process of either bone formation or resorption so as to

“re-shape” pre-existing bone with a net increase in bone mass. The mechanism involves

either activation-formation or activation-resorption which occurs primarily throughout

childhood. Activation is signalled by local tissue strain and involves the recruitment of

progenitor cells that differentiate into mature osteoblasts or osteoclasts. Once the

appropriate cell population is activated, the processes of formation or resorption take

place until sufficient bone mass is added or removed to normalize local strains (Allen et

al., 2014; Teti, 2011).

In contrast, bone remodelling is defined as a process of renewing and maintaining bone

in which osteoblast and osteoclast activities occur sequentially in a coupled manner.

The cycle of activation, resorption of old bone and formation of new bone, occurs

throughout life and is tightly balanced and heavily orchestrated by cell activities (Allen

et al., 2014; Teti, 2011).

The bone remodelling unit and osteoblast-osteoclast communication

The process of bone remodelling is executed by temporary anatomic structures


33

incorporating a cohort of cells known as the bone remodelling unit (BRU) (Figure 1.3)

(Frost, 1973; Parfitt et al., 1996). With the progression in bone research, the BRU no

longer merely consists of osteoclasts, osteoblasts and osteocytes. Other key cells

including osteolining cells, osteomacs and vascular endothelial cells have also been

reported to be associated with the BRU (Andersen et al., 2009; Chang et al., 2008;

Dierkes et al., 2009; Lafage-Proust et al., 2010; Parfitt, 2001; Pettit et al., 2008;

Rasmussen et al., 1973). Bone is a highly vascularised organ and networks of blood

capillaries have been observed at the site of the BRU (Chim et al., 2013).

Vascularisation and angiogenesis are prerequisites for bone formation and remodelling.

They serve multiple purposes including efficient exchange of matrix constituents and

minerals, and the recruitment of bone cell progenitors via blood capillaries to establish

the modelling and remodelling site (Andersen et al., 2009; Kristensen et al., 2013;

McClugage et al., 1973; Shapiro, 2008). The blood supply also allows the BRU to

become accessible to immune cell infiltration and subsequent interaction with bone

cells. It has been demonstrated that there is significant crosstalk between bone and

immune cells, this has led to the development of the field of osteoimmunology, and will

be discussed in detail in Chapter 2. The BRU represents the anatomical basis for

intercellular communication and coupling of bone remodelling. The bone remodelling

cycle occurs in five stages: activation, resorption, reversal, formation and quiescence

(Allen et al., 2014).

Activation

The activation stage commences with the detection of a signal such as mechanical strain

by osteocytes or osteocyte apoptosis, and the recruitment of osteoclast progenitors to

the bone surface. Detecting the mechanical stimuli, osteolining cells form a

“remodelling canopy”, while allowing the interaction between pre-existing osteoblasts

and recruited osteoclast progenitors underneath (Allen et al., 2014; Clarke, 2008;

Raggatt et al., 2010; Rifkin et al., 1979). The signals of M-CSF and RANKL presented

by osteoblasts are necessary for osteoclastogenesis. Other chemokines, such as

monocyte chemoattractant protein (MCP)-1, also known as CCL2, is released by

osteoblasts and is a candidate recruiter of osteoclast precursors. The expression of

MCP-1 receptors on osteoclast precursors is induced by RANKL, suggesting that

osteoblasts are capable of enhancing MCP-1-dependent recruitment through RANKL

(Li et al., 2007a). Another possible chemokine involved in the recruitment of osteoclast


34

precursors is the stromal cell-derived factor (SDF)-1, produced by bone vascular

endothelial and marrow stromal cells, that binds to the chemokine receptor CXCR4

expressed on osteoclast precursors to induce MMP9 expression (Sundaram et al., 2007).

Another form of activation is inflammation, particularly in the context of the pathology

of inflammatory arthritis, whereby inflammatory cytokines secreted by immune cells

are also known to be osteoclastogenic. T cells have been reported to secrete RANKL

and T helper 17 cells have been identified as the osteoclastogenic T cell subset.

Resorption

Upon attachment onto the matrix, differentiated mature osteoclasts create an isolated

microenvironment beneath the cell. The “sealing zone” is where osteoclasts pump in

hydrogen ions to establish an acidic and optimal condition for various collagenolytic

enzymes (Raggatt et al., 2010). One of the most potent collagenolytic cysteine

proteinases that degrades collagen at an acid optimum pH is cathepsin K (Everts et al.,

2005; Morko et al., 2009). Cathepsin K expression is promoted by RANKL (Corisdeo

et al., 2001) and cathepsin K deficiency due to loss of function mutation led to

osteopetrotic bone disease (Gelb et al., 1996). Another collagenase highly expressed in

mature osteoclasts is matrix metalloproteinase-9 (MMP9) (Sundaram et al., 2007).

Osteoclasts also express high levels of lysosomal acid phosphatase TRAP, the enzyme

marker of osteoclasts (Akisaka et al., 1989). TRAP is able to dephosphorylate bone-

associated proteins, and mice deficient for the enzyme exhibit a mild osteopetrotic

phenotype (Andersson et al., 2003; Ek-Rylander et al., 1994; Hayman et al., 1996). The

dissolution of matrix by these enzymes results in the formation of scallop-shaped

resorption pits termed Howship’s lacunae on the surface of trabecular bones and

Haversian canals in cortical bone (Clarke, 2008; Eriksen, 1986; Hattner et al., 1965).

As resorption proceeds, more osteoclasts can be recruited to the remodelling sites to

either support existing osteoclasts or replace those that have undergone apoptosis (Allen

et al., 2014). However, the differentiation of osteoclastic cells is heavily regulated

through OPG expression by osteoblasts to avoid excessive bone resorption (Mizuno et

al., 1998; Morony et al., 1999). To circumvent the antagonist effect of OPG, osteoclasts

have substantial expression of Sema4D, a member of the semaphorin family that

includes both secreted and membrane-associated molecules that interact with plexin and

neuropilins as their primary receptors. Targeted ablation of Sema4D in osteoclasts did


35

not lead to dysfunction of osteoclasts; however, it resulted in an increased trabecular

bone mass due to increased number and activity of osteoblasts, suggesting that Sema4D

is an osteoclast-derived inhibitor of osteoblast differentiation and function (Negishi-

Koga et al., 2011). Notably, T cells also produced Sema4D, demonstrating another

possible local source of factors that inhibit osteoblastogenesis (Bougeret et al., 1992;

Zhang et al., 2013b). Osteoclasts also express Sphingosine 1-phosphate (S1P), which

has both stimulatory and inhibitory effects on osteoblasts depending on the source of

osteoblast precursors and cell differentiation (Ishii et al., 2009; Kikuta et al., 2013;

Pederson et al., 2008).

Reversal

The reversal phase is characterised by the termination of osteoclast-mediated bone

resorption and the initiation of bone formation. Following resorption, osteoclasts

undergo apoptosis or migrate to another surface, leaving the resorption pit filled with

debris of undigested demineralised collagen matrix and a myriad of cells. Osteolining

cells and phagocytic macrophages are implicated in the removal of demineralised debris

and the “reconditioning” of the remodelling site (Everts et al., 2002; Takahashi et al.,

1986). Without the removal of this debris, bone formation by osteoblasts does not

proceed (Allen et al., 2014; Hauge et al., 2001). Among the osteolining cells are the

residential osteal macrophages, which are required to prime the area into an osteogenic

environment for osteoblast function (Alexander et al., 2011; Chang et al., 2008; Cho et

al., 2014). An immediate source of osteoblasts is thought to be present among the

osteolining cells, while osteoblast precursors are recruited from the bone marrow. It has

been proposed that the recruitment of osteoblast precursors could be initiated by bone

matrix-derived factors, such as insulin like growth factors (IGFs) I and II, BMPs and

TGF-β, which were pre-embedded within the matrix by osteoblasts and released during

osteoclast resorption (Shinar et al., 1993; Tang et al., 2009; Xian et al., 2012).

Different hypothetical models of direct osteoclast-osteoblast coupling mechanism have

been proposed. The gp130 family of cytokines have been reported to stimulate

osteoclastogenesis and osteoblast proliferation or differentiation and using mice models,

it was concluded that resorption alone was insufficient to promote coupled bone

formation and that active osteoclasts are the likely source of coupling molecules

(Martin et al., 2005; Sims et al., 2004). Gp130 signalling cytokine, cardiotrophin-1


36

(CT-1), is detected by resorbing osteoclasts and was shown to stimulate osteoblast

differentiation in vitro and bone formation in vivo (Walker et al., 2008). Another

protein family that has been implicated in osteoclast-osteoblast coupling is the

Ephrin/Eph family, which has a capacity for bidirectional signalling. EphrinB2 is

expressed by osteoclasts and in vivo and ex vivo studies of genetically modified mice

suggested that osteoclast-derived ephrinB2 acts through its receptor EphB4, which is

expressed on osteoblasts, to support forward signalling of osteoblast differentiation.

Upon direct contact, the reverse signalling by osteoblast-derived Eph4 suppresses the

formation of osteoclast precursors (Irie et al., 2009; Matsuo et al., 2012; Zhao et al.,

2006).

There is also evidence suggesting that immune cells participate in the regulation of

osteoblast formation and function (Li et al., 2007b). Supporting their possible role in

the reversal phase, B-lymphocytes have also been shown to inhibit osteoclast activity by

expressing OPG and TGF- β in vivo and in vitro (Horowitz et al., 2005; Horowitz et al.,

2004; Li et al., 2007b; Weitzmann, 2013; Weitzmann et al., 2000). Innate immune

cells, such as activated macrophages is also reported to secrete OSM for the

maintenance of osteoblast differentiation (Guihard et al., 2012; Sims et al., 2014).

Formation

Bone formation is a slow phase and can take 3 to 4 months to complete. During the

bone formation stage, mature osteoblasts lay down osteoid, which serves as a template

that eventually becomes mineralised. At the finale of bone remodelling cycle, the newly

formed bone surface is in a state of quiescence while the matrix within the remodelling

unit will continue to mineralise (Allen et al., 2014).

1.4. Conclusion

Although it is important to dissect a complex system by labelling and categorising

different cells whilst identifying their “unique” roles, the skeletal system must be also

be comprehended as an integration of multiple cellular organisations. While the

OPG/RANKL/RANK signalling pathway is essential for the intimate crosstalk between

anabolic osteoblasts and catabolic osteoclasts, and is considered the central axis

controlling bone remodelling, different cell types within the bone remodelling

compartment also play a crucial role in maintaining the crosstalk. The skeletal system

has been shown to be not only regulated by the immune system, but also by the kidney,


37

brain and central nervous system, indicating the complexity of bone homeostasis and

underlying pathology of many bone diseases.

With the many contributors to the regulation of bone homeostasis, future investigation

is essential to understand the complex intercellular communication during bone

remodelling, and how the body compensates in certain pathological conditions, with the

ultimate aim to design interventions that can prevent bone loss while maintaining bone

quality.


38

Figure 1.1 The different types of bone cells involved in the regulation of bone

microenvironment. Initially thought to be static, the bone tissue is dynamically

orchestrated by different types of cells and other systems. Osteoclasts and osteoblastic

cells originate from the hematopoietic and mesenchymal stem cells respectively.

Recently characterised bone cells include the bone lining cells and osteomacs.


39

Figure 1.2 RANK/RANKL signalling pathway during osteoclastogenesis. RANKL

stimulation is crucial in driving the formation and function of osteoclasts. The

interaction between RANKL and RANK initiate the recruitment of TRAF6 to initiate

several signaling cascades, including NF-κB, NFATc1/calcineurin and MAPK

pathways. The differentiation of osteoclasts is also supported M-CSF which induce the

expression of RANK in pre-osteoclasts.


40

Figure 1.3 The bone remodelling cycle. The remodelling unit represents cellular

communication that governs the coupling between bone resorption and formation. This

cycle is supported by other systems, such as the blood vessels, which carries nutrients

and other cell types. Immune system also inevitably interacts with bone cells as their

haematopoiesis is spawned with the marrow cavity. For instance, activated CD4+ T

cells are reported to stimulate the production of RANKL by stromal cells, activated

macrophages secrete OSM for the maintenance of osteoblast differentiation, and B

lymphocytes express OPG and Wnt1b, which favours bone formation in vitro.

CHAPTER TWO

OSTEOIMMUNOLOGY – INTERACTIONS OF THE

SKELETAL AND IMMUNE SYSTEMS

Chapter 2: Osteoimmunology – Interactions of the Skeletal and Immune Systems

42

2. CHAPTER 2: OSTEOIMMUNOLOGY – INTERACTIONS OF THE SKELETAL AND IMMUNE SYSTEMS

2.1. Introduction

The bone organ system is a living mineralized tissue that is continuously being broken

down and rebuilt in a process known as bone remodelling. The bone remodelling

process is a coupled activity between the catabolic effects of bone-resorbing osteoclasts

and the anabolic effects of bone-forming osteoblasts. This coupling activity between

these two bone cells is tightly regulated so as to achieve microarchitecturally strong

bone, whereby the imbalance of this delicate regulation often lead to bone and joint

diseases.

With the emerging field of osteoimmunology, many studies have indicated that both

innate and adaptive immune systems are major components of bone remodelling. The

dialogue between the skeletal and immune systems is inevitable as haematopoietic and

mesenchymal stem cells are housed within the bone marrow, and these gives rise to

progenitors of bone cells and lymphocytes that constitute the immune system. The

intimate relationship between the immune and skeletal system is highlighted in many

inflammatory diseases such as periodontal disease (Kayal, 2013), rheumatoid arthritis

(RA) (Spector et al., 1993), spondylarthritis (SA) (Goldring, 2013), Cherubism (Mukai

et al., 2014), Crohn’s disease (Gordon, 2006), osteomyelitis (Ferguson et al., 2006) and

systemic lupus erythematosus (SLE) (Tang et al., 2013a), all of which are associated

with both local and systemic bone loss with accompanying increased risk of bone

fractures and deformities. Beyond these observations, recent data further demonstrate

that the immune system may have pro-anabolic functions. Osteoimmunological

interactions thus play a significant role in the maintenance of bone homeostasis and are

fundamental mechanisms in bone pathology.

Likewise, it is also important to acknowledge the role of the skeleton in fostering

haematopoiesis and the maturation of immune lineages. In view of this, the overlapping

roles of the essential cytokines, receptors, common signalling transduction and

molecular pathways in the osteoimmune system, and how the dysregulation of these

shared pathways can lead to clinical manifestation of bone diseases are discussed below.

2.2. Osteoclasts and inflammatory bone destruction

One of the most well described osteoimmune diseases is RA, which is characterised by

progressive erosion of bone surrounding inflamed joints (Gough et al., 1994; Spector et

al., 1993). Histology studies of tissue sections from sites of erosion in RA patients and


43

animal models have revealed multinucleated cells that are phenotypically similar to

bone-resorbing osteoclasts in the bone resorption lacunae (Bromley et al., 1984; Leisen

et al., 1988). Mononuclear cells isolated from the synovial fluid obtained from patients

with RA are reported to be able to differentiate into functional resorbing osteoclasts

(Danks et al., 2002). The osteoclastogenic property of synovial fluid collected from RA

patients is also shown to support the formation of numerous TRAP positive cells

capable of lacunar resorption in vitro (Adamopoulos et al., 2006). Consistent with the

aforementioned observations, mice that are unable to form functional osteoclasts are

protected from arthritis and subsequent development of bone erosion despite exhibiting

levels of synovial inflammation comparable to the wildtype littermates (Pettit et al.,

2001; Redlich et al., 2002).

The expression of receptor activator of NF-κB ligand (RANKL), the key differentiation

factor that drives osteoclastogenesis, was detected specifically in the synovium and at

sites of articular bone erosion of patients with RA (Gravallese et al., 2000; Pettit et al.,

2006). The receptor for RANKL is RANK (previously known as the tumour necrosis

factor (TNF) receptor super family member (TNFRSF) 11a), a type I trans-membrane

protein that is highly homologous to CD40 (a co-stimulatory protein expressed on

antigen presenting cells involved in their activation). RANKL-deficient and RANK-

deficient mice models are both osteopetrotic due to impeded osteoclastogenesis, thus

indicating that RANK-RANKL interaction are crucial to initiate osteoclast

differentiation and the maintenance of osteoclast resorptive activity (Dougall et al.,

1999; Fuller et al., 1998; Kong et al., 1999b). The decoy receptor osteoprotegerin (OPG

or TNFRSF11b) prevents the binding of RANKL to RANK and inhibits

osteoclastogenesis. Rescue assays using inhibitors of osteoclastogenesis such as OPG

have been shown to successfully prevent bone erosion in a collagen-induced arthritis

(CIA) model (Romas et al., 2002). This indicates that RANKL and osteoclasts play a

pathological role in the progression of inflammatory arthropathies (Francois et al.,

2006; Neidhart et al., 2009).

2.3. Osteoclast and innate immune cell lineages: RANK and TLRs downstream

signalling cascades

One of the hallmarks of RA pathology is characterised by the perpetual production of

pro-inflammatory cytokines accompanied by massive infiltration of inflammatory


44

immune cells, including T and B cells, into the synovium. Inflammation is considered

as a nonspecific immune response orchestrated by dendritic cells, neutrophils,

monocytes, macrophages, natural killer cells, eosinophils, basophils and mast cells. All

of these effector cells, except for natural killer cells, originate from the haematopoietic

stem cell-derived common myeloid progenitors that also give rise to osteoclast

precursors.

Generation of macrophages and osteoclast precursors from the common myeloid

progenitors is mediated by a lineage specific cytokine termed macrophage-colony-

stimulating factor (M-CSF, also known as CSF-1) and the myeloid master regulator

PU.1. M-CSF directly induces the expression of PU.1, which in turn stimulate the

expression of macrophage colony-stimulating factor 1 receptor (M-CSF1R, also called

c-fms or CD115) (DeKoter et al., 1998; Mossadegh-Keller et al., 2013). M-CSF is

important for the survival and differentiation of osteoclast progenitors. Osteopetrotic

PU.1 deficient mice do not develop macrophage and mature osteoclasts, suggesting that

the transcription factor is crucial in regulating the initial stages of myeloid

differentiation (Tondravi et al., 1997). The induction of PU.1 by M-CSF is restricted by

MafB, which is selectively expressed in mature monocytes and macrophages (Sarrazin

et al., 2009). MafB also negatively regulates osteoclast differentiation (Kim et al.,

2007a).

In the presence of M-CSF, RANK-RANKL interaction induces osteoclastogenesis of

osteoclast precursors and initiates the recruitment of tumour necrosis factor receptor-

associated factors (TRAFs), including TRAF2, 5 and 6, to activate Src kinases and

mitogen-activated kinases (MAPK) (Kobayashi et al., 2001). Simultaneously, RANK

activation results in the phosphorylation of two adaptor proteins associated with distinct

Ig-like receptor, the immunoreceptor tyrosine-based activation motif) in DAP12 and Fc-

receptor common γ-subunit (FcRγ). Ig-like receptor signals are co-stimulatory signals

for RANK during osteoclast differentiation (Kim et al., 2012).

TRAF6 associates with transforming growth factor-β-activated kinase (TAK1) and

activates JUN N-terminal kinase (JNK) and AP-1 (Qi et al., 2014). TRAF6 also induces

nuclear translocation of the transcription factor NF-κB through IκB ubiquitination and

degradation to initiate the transcription of osteoclast specific genes targets (Chen, 2005).

The stimulation of RANKL/RANK interaction also leads to the activation of


45

calcineurin/nuclear factor of activated T cells cytoplasmic 1 (NFATc1) signalling

pathway and along with c-Fos, the nuclear translocation of dephosphorylated NFATc1

amplifies the expression of osteoclast-associated genes (Gohda et al., 2005).

Given the right stimuli, macrophage and myeloid dendritic cells are able to form

osteoclasts in vitro, suggesting that they originate from a subset of common myeloid

progenitor cells. This cell subset can be identified as B220- cKit/CD117+ CD115/c-fms+

CD11blo in the bone marrow niche (Jacome-Galarza et al., 2013; Jacquin et al., 2006).

The differentiation of these common progenitors into macrophages, myeloid dendritic

cells and osteoclast can be considered a dynamic lineage bifurcation (Figure 2.1).

Recent findings have identified CD27 as a marker of the subset that discriminates

osteoclast/dendritic precursors (CD27hi) from more osteoclast-committed progenitors

(CD27lo). The differentiation of B220- cKit/CD117+ CD115/c-fms+ CD11blo CD27hi to

B220- cKit/CD117+ CD115/c-fms+ CD11blo CD27lo is tightly regulated by the

expression of ligand CD70 on activated immune cells, whereby constitutive engagement

of CD27/CD70 favours the differentiation towards dendritic cells (Xiao et al., 2013).

The same common myeloid precursors differentiate into dendritic cells in the presence

of granulocyte-macrophage-colony-stimulating factor (GM-CSF) and interleukin (IL)-4

through the down-regulation of a c-Fos oncoprotein, a member of the AP-1 transcription

factor complex (Grigoriadis et al., 1994; Miyamoto et al., 2001). In vitro experiments

revealed that GM-CSF inhibits osteoclastogenesis and stimulates dendritic cell

development through the suppression of c-Fos. Comparatively, the expression of c-Fos

inhibits dendritic differentiation, thus indicating that early c-Fos expression acts as a

switch in the lineage determination of osteoclast and dendritic cells (Miyamoto et al.,

2001).

Macrophage differentiation is marked by the transcription of various genes including

F4/80, Mac1 (CD-11b), MOMA-2 and major histocompatibility complex (MHC) class

II molecules. The expression of CD11c and also MHC class II indicate dendritic cell

differentiation. MHC class II are a family of glycoproteins present only on antigen-

presenting immune cells. The functional manifestation of the innate immunity by

antigen presenting cells requires the activation of pattern recognition receptors (PPRs)

and is greatly increased by signalling through the TNFR family member CD40 (Caux et

al., 1994). PPRs are activated by (1) pathogen-associated molecular patterns (PAMPs)


46

that are expressed on pathogens, and/or (2) damage-associated molecular patterns

(DAMPs) which are released by apoptotic and/or necrotic cells from surrounding

injured tissues. DAMPs are proposed to activate a subset of PPRs known as Toll-like

receptor (TLRs), which are expressed in both macrophage and dendritic cells, during

sterile inflammation in RA joint (Midwood et al., 2009; Park et al., 2004).

Interestingly, TLR engagement is reported to inhibit the expressions of c-fms and

RANK, thus attenuating osteoclastogenesis and favouring the maturation of APCs (Ji et

al., 2009). The activation of most TLRs leads to the initiation of MyD88-dependent

pathway, which involves the IL-1R-associated kinases (IRAK)-1 and -4, TRAF-6, and

MAPK. It culminates in the activation of AP-1 and NF-κB signalling cascades, which

results in the transcription and production of proteases, chemokines and pro-

inflammatory cytokines such as TNF-α, IL-1, IL-6 and interferons (IFNs), most of

which are capable of expediting osteoclast differentiation (Piccinini et al., 2010).

2.4. Role of inflammatory hematopoietic cytokines as a double-edged sword in

bone modeling and remodeling

Previous studies reveal that the equilibrium between pro- and anti-inflammatory

cytokines determines the fine balance between the osteoclastogenic effects of RANKL

and inhibitory effects of its decoy receptor OPG. Several cytokines involved in the

regulation of inflammation behave as a double-edged sword in modulating bone

remodeling and the progression of inflammatory bone diseases including TNFs,

interleukins, IFNs and TGF-β superfamily members (Table 2.1).

2.4.1. Tumor necrosis factor (TNFs)

Early observations regarding inflammation related bone loss led to the discovery of the

osteoclastogenic potential of tumour necrosis factors (TNFs), which are readily detected

within the RA and sacroiliitis tissue (Bertolini et al., 1986; Di Giovine et al., 1988;

Francois et al., 2006; Thomson et al., 1987). TNF proteins are mainly produced by

macrophages, but also by other lymphoid and non-lymphoid cells. TNFs exert their

biological roles via interaction with their associated membrane receptors, TNF-receptor

type 1 (TNF-R1) and TNF-R2. TNF-R1 is commonly expressed in most tissues and

contains a cytoplasmic interaction domain called death domain (DD), which recruits

other DD-containing proteins and stimulates apoptosis. TNF-R1 is a potent activator of

gene expression via the indirect recruitment of TRAF proteins. Expressed mainly by the


47

immune cells, TNF-R2 also plays a major role in the lymphoid system. TNF-R2 directly

recruits TRAF2 and activates NF-ĸB, JNK, and p38-MAPK signalling pathways to

induce gene expression and also crosstalks with TNF-R1 (Wajant et al., 2003).

TNF-α is reported to stimulate the expression of c-fms, the receptor of M-CSF in

osteoclast precursors, thereby supporting the expansion of the cell population (Yao et

al., 2006). The addition of TNF-α per se fails to induce the differentiation of osteoclast

precursors, however the addition of the cytokine enhances osteoclastogenesis of

preosteoclasts primed with minimal level of RANKL that is insufficient to induce

osteoclast differentiation (Lam et al., 2000). Furthermore, the secretion of TNF-α by

neutrophils and activated macrophages in response to injury promotes the expression of

RANKL by osteoblastic and stromal cells to induce osteoclast differentiation and

resorption (Hofbauer et al., 1999; Thomson et al., 1987). In RA, the increased TNF-α

leads to an elevated expression of Dickkopf-related protein 1 (DKK1), a natural

inhibitor of Wnt protein and Wnt/β-catenin pathway that are crucial to activate

osteoblasts and bone formation (Diarra et al., 2007). The pro-osteoclastogenic activity

of TNFs is supported by the ability of the cytokine to inhibit osteoblast differentiation

and induce apoptosis on osteoblastic cells via Fas-ligand (FasL)/Fas signalling

(Bertolini et al., 1986; Centrella et al., 1988; Gilbert et al., 2000; Jilka et al., 1998;

Kawakami et al., 1997).

FasL, a member of TNF superfamily, interacts with its receptor Fas to mediate caspase-

activated apoptosis. Fas-FasL signalling is important in the elimination of cytotoxic

CD8+ T cells and the regulation of self-reactive T cells (Waring et al., 1999). Both

precursor and mature osteoclasts express Fas and FasL. Treatment of M-CSF/RANKL-

treated osteoclast precursors and mature osteoclasts with Fas-acting antibody increased

osteoclast formation and apoptosis respectively (Wu et al., 2003). Mice models with

spontaneous mutation in the coding for FasL and Fas genes display lymphoproliferative

disorder and autoimmunity (Ramsdell et al., 1994; Roths et al., 1984; Watanabe-

Fukunaga et al., 1992). The bone mass of these mice models are reported to be

contradictory, where by one group detailed a decrease in FasL-deficient mice, while

another found it to be increased (Katavic et al., 2003; Wu et al., 2003). The role of

Fas/FasL system in arthritic joint destruction also remained controversial whereby

different groups reported both anti- and proinflammatory roles in mice with CIA


48

(Ratkay et al., 1993; Tu-Rapp et al., 2004). These findings however reinforced the

notion that general immune disorder directly or indirectly affects normal bone

homeostasis.

2.4.2. Interleukin (IL)-1

The ability of TNFs to stimulate osteoclast formation is mediated by the acute phase

protein IL-1 that is also detected in the synovial fluid (Fontana et al., 1982; Hopkins et

al., 1988; Wei et al., 2005). In vitro, IL-1 is a potent stimulator of bone resorption

(Gowen et al., 1986). IL-1 is implicated in TNF-mediated systemic bone loss in hTNFtg

mouse model with spontaneous arthritis and the abrogation of IL-1 rescued the animal

from severe bone loss (Ness et al., 1991). The osteoclastogenic role of TNF and IL-1 is

further demonstrated with experiments whereby the administration of their respective

inhibitors into mouse models is shown to inhibit osteoclastogenesis and bone resorption

in vivo (Elliott et al., 1994; Kitazawa et al., 1994; Matsuno et al., 2002; Redlich et al.,

2004).

The expression of IL-1 receptor, type I (IL-1RI) is upregulated by RANKL via c-Fos

and NFATc1. The over-expression and c-Fos-mediated induction of IL-1RI receptor in

murine bone marrow macrophages (BMMs) enables IL-1 alone to induce

osteoclastogenesis via the activation of NF-κB, JNK, p38, and ERK. IL-1/IL-1RI

interaction also strongly activates microphtalmia-associated transcription factor (MITF),

which then upregulates osteoclast-specific genes (Kim et al., 2009). IL-1, along with

TNF, is also implicated in the induction of IL-6 gene expression (Zhang et al., 1990).

2.4.3. IL-6

It has been established that the serum level of IL-6 is elevated in RA and SA patients

(Abdel Meguid et al., 2013; Gratacos et al., 1994; Hirano et al., 1988; Karlson et al.,

2009; Sack et al., 1993). IL-6 is a pro-inflammatory cytokine produced by both immune

and non-immune cells that strongly induce local synovial leukocyte accumulation and

antibody production. Fibroblasts isolated from the synovium of RA patients are capable

of inducing IL-6 expression on endothelial cells in vitro, and aid in the adherence of

neutrophils to the endothelium. Thus, IL-6 has been implicated in the shift from acute to

chronic inflammation by facilitating the transition from neutrophil to mononuclear cell

recruitment, which is a hallmark of acute inflammation (Kaplanski et al., 2003).


49

IL-6 exerts its activity via a cell surface receptor protein complex comprising of an IL-6

receptor subunit (IL6R, also referred to as IL-6Rα, gp80 or CD126) and IL-6 signal

transducer glycoprotein (gp) 130 (also denoted as IL-6β or CD130). IL-6R is expressed

mainly in hepatocytes and leukocyte populations (such as monocytes, neutrophils, T

cells and activated B cells), while the latter transmembrane protein molecule gp130 is

expressed ubiquitously (Hibi et al., 1990; Zola et al., 1992). Initially identified as the

signal-transducing component of IL-6R, it is now apparent that gp130 acts as a shared

receptor subunit and signal transducer of other members of the IL-6 family of cytokines,

including oncostatin M (OSM) and leukaemia inhibitory factor (LIF). IL-6, OSM and

LIF have been shown to regulate the development of osteoclasts and bone resorption by

modulating the production of RANKL by osteoblasts and synovial cells, which express

IL-6 receptor (Hashizume et al., 2008; Ishimi et al., 1990; Liu et al., 2005; Palmqvist et

al., 2002). In addition to membrane-bound receptor, a soluble form of IL-6R (sIL-6R) is

found in many body fluids including the synovial fluid (Desgeorges et al., 1997). The

sIL-6R/IL-6 complex is also capable of activating cells via interaction with membrane

bound gp130.

The recognition of IL-6 by its receptor leads to the dimerization of gp130 and triggers

two intracellular signalling cascades: the JAK/STAT pathway, which is mediated by

YxxQ motif of gp130, and MAPK pathway, which is regulated via the cytoplasmic

Y759 residue of gp130 (Abe et al., 2001; Fukada et al., 1996; Heinrich et al., 1998).

F759 mice, which contain a single amino acid substitution in gp130 (Y759F), displayed

enhanced STAT3 activation and develop spontaneous RA-like arthritis (Atsumi et al.,

2002). The translocation of these transcription factors to the nucleus independently

induces the expression of target genes for several physiological processes. For instance,

the JAK/STAT pathway is implicated in controlling immune response and activates

RANKL expression by synovial cells (Hashizume et al., 2008). The MAPK pathway is

reported to induce matrix metalloproteinase (MMP) production by chondrocytes to

enhance proteoglycan degradation in the pathogenesis of osteoarthritis (Hashizume et

al., 2010).

IL-6 is also documented to facilitate the recruitment of osteoclast precursors by

upregulating the expression of sphingosine-1-phosphate (S1P) receptor 2 (S1PR2) on

osteoclast precursors in CIA mouse model. S1P, a bioactive lipid molecule, controls the


50

migration of osteoclast precursors and the administration of IL-6 receptor antibody in

the mouse model significantly inhibits inflammatory bone loss (Tanaka et al., 2014). IL-

6 deficient mice display a reduction in synovial infiltration and are resistant to arthritic

induction (Alonzi et al., 1998; Boe et al., 1999; Nowell et al., 2009; Ohshima et al.,

1998). Equivalently, the suppression of immunoinflammatory cascade by targeting IL-6

has been shown ameliorate human chronic inflammatory arthritis (Schoels et al., 2013).

2.4.4. IL-8

A recent study observed the newly describe member of type III IFN family (IFN-λ) IL-

29 is overexpressed in blood and synovium of RA patients and triggers the production

of both IL-6 and IL-8 by RA synovial fibroblasts (Xu et al., 2013). IL-8 is a chemokine

(chemoattractant cytokines) that is capable of promoting the recruitment, proliferation,

and activation of leukocytes, especially neutrophils (Harada et al., 1994). IL-8

production by various types of somatic cells in response to injury, infection and

inflammation is induced by TNF-α and GM-CSF (Kurt-Jones et al., 2002; Perng et al.,

2012; Yasumoto et al., 1992; Yoshida et al., 1997). Indeed the serum level of IL-8 is

elevated in inflammatory diseases such as RA, osteoarthritis and osteomyelitis

(Fullilove et al., 2000; Kaneko et al., 2000; Troughton et al., 1996).

IL-8 is observed to stimulate RANKL production by osteoblastic stromal cells while

directly enhancing osteoclastogenesis and bone resorption to exacerbate bone

destruction common in metastatic bone disease (Bendre et al., 2002; Bendre et al.,

2005; Bendre et al., 2003; Kamalakar et al., 2014). The effect of IL-8 is associated with

the cell surface expression of receptor of IL-8 (CXCR1) on osteoclasts and their

precursors (Bendre et al., 2003). Sequentially, IL-8 production by osteoclasts is

stimulated by RANKL, suggesting autocrine signalling to amplify osteoclastogenesis

(Bendre et al., 2003; Kopesky et al., 2014; Rothe et al., 1998).

Other chemokines that are implicated in bone remodelling include chemokine (C-X-C)

motif ligand 1 (CXCL1, previously known as growth-regulated oncogene (GRO)-α),

CXCL9 (formerly called monokine induced by gamma interferon (MIG)), C-C motive

ligand 2 (CCL-2, previously referred to as monocyte chemoattractant protein (MCP)-1),

CCL3 (previously known as macrophage inflammatory protein (MIP)-1α), CCL4 (MIP-

1β), CCL5 (also known as regulated on activation, normal T cell expressed and secreted

(RANTES)), all of which are reported to be localised within arthritic bone biopsies


51

(Borzi et al., 1999; Koch et al., 1994; Koch et al., 1995a; Koch et al., 1995b; Konig et

al., 2000; Lisignoli et al., 2002; Lisignoli et al., 1999).

2.4.5. IL-17

IL-17 is a family of proinflammatory cytokines secreted by a specialized subset of

activated CD4 T cells, called T helper type 17 (TH17) (Korn et al., 2009; Moseley et

al., 2003). The receptor of IL-17, IL-17R is ubiquitously expressed, suggesting the

possibility that IL-17 targets a wide variety of cell types (Lee, 2013). IL-17/IL-17R

interaction mediates the recruitment of TRAF6 and activation of NF-κB and MAPK,

leading to the production of proinflammatory cytokines and subsequent myeloid cell

recruitment during inflammation (Awane et al., 1999; Schwandner et al., 2000; Shalom-

Barak et al., 1998).

The osteoclastogenic potential of IL-17, which was abundantly detected in synovial

fluids of RA patients, was reported by Kotake et al. (1999) to stimulate

osteoclastogenesis in an osteoblast-dependent manner (Kotake et al., 1999). The role of

IL-17 in bone remodeling was supported by numerous in vivo and in vitro studies. One

in vitro study revealed that, in combination with TNF-α, IL-17 stimulates bone

resorption (Van bezooijen et al., 1999). The treatment of experimental arthritic mice

with anti-IL-17 antibody dramatically rescued joint inflammation and prevented

cartilage and bone destruction (Koenders et al., 2005; Lubberts et al., 2004).

Most reports suggest that IL-17 stimulate osteoclastogenesis in a co-culture system,

where by IL- 17 induces RANKL expression in osteoclast-supporting cells, such as

osteoblasts (Huang et al., 2009; Koenders et al., 2005; Kotake et al., 1999; Sato et al.,

2006; Zhang et al., 2011a). However, the direct effect of IL-17 on osteoclast precursors

remained controversial. One study demonstrated the ability of IL-17 to directly induce

osteoclastogenesis in human monocytes (Yago et al., 2009). In contrast, IL-17 inhibited

osteoclast formation from RAW264.7 cells (Kitami et al., 2010). A recent study

reported that in the presence of Vitamin D3 (1,25(OH)(2)D(3)), IL-17 inhibits

osteoclastogenesis in mouse osteoblasts-BMM coculture by stimulating the production

of GM-CSF by osteoblast (Balani et al., 2013). The direct effect of IL-17 on

osteoclastogenesis remained controversial, likely due to the fact that multiple factors,

such as the species, culture conditions and source of cells, vary in the experimental

settings involved in the studies above. Nevertheless, these studies highlight the effect of


52

IL-17 in the maintenance of bone homeostasis.

2.4.6. IL-21

IL-21 is a pleiotropic cytokine primarily produced by natural killer T (NKT) cells,

follicular helper T cell (TFH) and TH17 cells (Parrish-Novak et al., 2000; Spolski et al.,

2014). The receptors for IL-21 are IL-21 receptor (IL-21R) 2, IL-21R3 and common

cytokine receptor γ-chain, γc (Asao et al., 2001; Ozaki et al., 2000; Parrish-Novak et

al., 2000). IL-21R is expressed on myeloid cells and other cell types, allowing IL-21 to

exert its effect on other systems (Spolski et al., 2014). IL-21/IL-21R downstream

signalling activates JAK/STAT, MAPK and PI3K pathways (Zeng et al., 2007).

A study implicating IL-21 in bone remodelling and progression of bone destruction in

RA was reported by Kwok et al. (2012). IL-21 level was upregulated in the synovium,

synovial fluid and serum of patients with RA and in serum of mice with CIA. Befitting

to its osteoclastogenic property, IL-21 induced RANKL in CD4 T cells and augments in

vitro osteoclastogenesis (Kwok et al., 2012). Nonetheless, further investigations are

required to support this observation and rescue experiment by administrating IL-21

antibody into CIA animal models may provide an insight in the potency of the

interleukin in exacerbating the progression of inflammatory arthropathies.

2.4.7. Other interleukins

IL-4 and IL-13, produced by TH2, mast cells and basophils, are implicated in

pathological conditions such as asthma and allergy. They share a receptor IL-4Rα to

activate many common signalling pathways, particularly the JAK/STAT pathways

(Jiang et al., 2000; Kelly-Welch et al., 2003). Skeletal analysis of IL-4 and IL-13

knockout mice revealed a reduction in cortical bone mineral content and bone strength,

implicating both IL-4 and IL-13 in the regulation of bone homeostasis (Silfversward et

al., 2007). The depletion of IL-4 and IL-13 does not alter fracture healing in mice,

suggesting that both cytokines may not exert their effects on osteoblastic bone

formation (Silfversward et al., 2008). In vitro investigation demonstrated that IL-4

exerts its function on osteoclast by dose-dependently inhibits RANKL-induced bone

resorption. The anti-osteoclastogenic property of IL-4 affects osteoclast function by

inhibiting NF-κB and Ca2+ signalling, and the formation of actin-ring crucial for

osteoclast migration and resorption (Mangashetti et al., 2005).


53

IL-5 was first reported as a T cell-replacing factor (TRF) that plays a role in eosinophils

and B cell differentiation towards immunoglobulin-secreting cells (Adachi et al., 1998;

Takatsu et al., 1988). IL-5 is a member of the haematopoietic growth factor family of

cytokines and has significant functional homology to GM-CSF and IL-3, thus they share

a common β-chain in their receptors. IL-5 interaction with its receptor IL-5R activates

multiple signalling cascades, including the JAK/STAT pathway (Adachi et al., 1998;

Huang et al., 2013; Takatsu et al., 1988). Mice with T cells that over-expressed IL-5

demonstrated heterotrophic ossification within the spleen. The calcified tissue sections

revealed both woven and lamellar bone with higher growth and mineralisation rate

when comparison to the skeletal bone. This study suggests that IL-5 overexpression

mediates bone modelling by mobilising osteogenic progenitors and/or the inhibition of

recruited osteoclasts (Macias et al., 2001). Miyamoto et al. (2001) further showed that

IL-5 does not inhibit osteoclastogenesis (Miyamoto et al., 2001), suggesting that IL-5

may play a regulatory role on osteoblastic cells instead.

Another interleukin implicated in bone remodelling is IL-10. The expression of IL-10

and its receptor are detected in chondrocytes, synovial fibroblasts and mesenchymal

stem cells (Iannone et al., 2001; Ritchlin et al., 2001; Zheng et al., 2008). IL-10

knockout mice display elevated serum proinflammatory cytokines and develop bone

abnormalities with shortened proliferating zone of the growth plates, lower bone

mineral density with reduced bone formation and mechanical fragility when compared

with wildtype (Cohen et al., 2004; Dresner-Pollak et al., 2004; Jung et al., 2013).

Treatment with IL-10 increased tibial lengths in organ culture and induced chondrocyte

proliferation and hypertrophic differentiation in vitro. Molecular study showed that IL-

10 utilise activated STAT-3, Smad 1/5/8 and MAPK pathways to induce the expression

of bone morphogenetic protein (BMP)-2 and 6, indicating that IL-10 is a pro-osteogenic

cytokine (Jung et al., 2013).

IL-10-producing induced osteoblast-like cells generated by Fujioka et al. (2014) was

shown to strongly inhibit osteoclastogenesis from RANKL-stimulated RAW264.7 cells

(Fujioka et al., 2014). Previous studies have demonstrated IL-10 potently inhibits

osteoclastogenesis, potentially acting upon osteoclast progenitors and by suppressing

the expression of NFATc1, c-Fos and c-Jun (Hong et al., 2000; Mohamed et al., 2007;

Xu et al., 1995).


54

2.4.8. Interferons (IFNs)

Type I IFNs (IFN-α and IFN-β) are secreted by infected cells and activate surveillance

immune cells such as macrophages and dendritic cells. They share the same receptor

and upon the engagement between type I IFNs and IFN-α receptor (IFNAR), canonical

type I IFN signalling activates JAK/STAT pathway to induce antiviral, antimicrobial

and inflammatory programmes (Ivashkiv et al., 2014; Zhang et al., 2008). In vitro

osteoclast activity assay revealed that IFN-α impairs osteoclast differentiation and bone

resorption (Avnet et al., 2007). Optimal induction of IFN-α requires IFN-β (Deonarain

et al., 2000), and RANKL induces the expression of IFN-β in osteoclast progenitors to

function as an auto-regulator to inhibit osteoclastogenesis by interfering with RANKL-

induced expression of c-Fos. This negative-feedback regulatory mechanism for bone

homeostasis was identified in osteopenic mice deficient in IFN-β signalling

(Takayanagi et al., 2002b).

Early in vitro studies have reported the inhibitory actions of type II IFN-γ on bone

resorption (Gowen et al., 1986; Peterlik et al., 1985). IFN-γ inhibits the RANKL/RANK

signalling cascade by accelerating the degradation of recruited TRAF6 via the

ubiquitin/proteasome system (Takayanagi et al., 2000). Mice lacking a functional IFN-γ

receptor exhibit bone destruction by overzealous osteoclasts and are more susceptible to

CIA (Manoury-Schwartz et al., 1997; Takayanagi et al., 2000; Vermeire et al., 1997).

In vivo administration of IFN-γ to tumour-bearing Tax+IFN-γ-/- mice prevented tumour-

associated bone loss and hypercalcemia, further highlighting the critical role of IFN-ɣ in

maintaining bone integrity (Xu et al., 2009b).

In contrast, in vivo administration of IFN-γ into a Cyclosporin A-induced osteopenic

mouse model failed to rescue the bone phenotype and further increased bone resorption

(Mann et al., 1994). IFN-γ treatment of a mouse model and patients with congenital

osteopetrosis showed an opposite outcome, whereby the therapy resulted in an increased

bone turnover and ameliorate the disease (Key et al., 1995; Rodriguiz et al., 1993).

Such opposing observations could be because the anti-osteoclastogenic activity of IFN-

γ appears to mediate its affects during the bifurcation of myeloid progenitors by

complementing the down regulation of RANK and c-fms by TLRs (Ji et al., 2009).

RANKL on the other hand, induces a broad resistance to cellular effects of IFN-γ

(Huang et al., 2003). Such observation is further validated by Gao et al. who reported


55

that direct targeting of IFN-γ on osteoclast precursors blunts osteoclastogenesis, but

stimulates T cell secretion of RANKL and TNF-α to indirectly stimulate osteoclast

formation and promote bone resorption (Gao et al., 2007). Therefore, the fate of

osteoclast precursors relies heavily on the balance of cytokines and is dictated by the

first cytokine they encountered.

2.4.9. Transforming growth factor (TGF)-β

Another cytokine that is abundantly expressed throughout the haematopoietic and bone

system is transforming growth factor-β (TGF-β). TGF-β is secreted by many cell types

and is implicated in many cellular processes such as cellular proliferation and

differentiation, embryonic development, wound healing, haematopoiesis, immune and

inflammatory cell regulation as well as suppression, and also bone development and

remodelling (Centrella et al., 1994). Secreted TGF-β exists in at least three homologous

isoforms: TGF-β1, TGF-β2 and TGF-β3 and their expression levels vary depending on

the tissue (Akhurst et al., 2012; Millan et al., 1991; Schmid et al., 1991). They share a

receptor complex and induce similar signalling pathways to execute different functions

as demonstrated by the distinct and partially overlapping phenotypes between knockout

mice (Dickson et al., 1995; Kulkarni et al., 1993; Millan et al., 1991; Pelton et al.,

1991; Proetzel et al., 1995; Runyan et al., 1992; Sanford et al., 1997; Shull et al., 1992).

TGF-β ligands are synthesised as a large TGF-β complex. The latent complex consists

of mature dimeric TGF-β associated with its latency-associated peptide (LAP) and a

latent TGF-β-binding protein (LTBP). Once activated via proteolytic cleavage, active

TGF-β dimers interact with extracellular domains of TGF-β receptors, which are well-

defined serine-threonine kinases. The specific type II receptors (such as TGF-β receptor

type II (TβRII)) undergo autophosphorylation, which then transphosphorylate type I

receptors (such as TβRI). This event is reported to propagate the specific Sma and Mad

related proteins (SMADs)-dependent canonical pathway and the SMADs-independent

pathways (Derynck et al., 2003).

In the canonical pathway, the phosphorylation of the receptors facilitates the

intracellular SMADs (R-SMADs: SMAD1, SMAD2, SMAD3, SMAD5 and SMAD8)

to form heteromeric complexes with a common mediator SMAD (co-SMAD: SMAD4).

R-SMAD/co-SMAD complexes translocate to the nucleus and preferentially associate

with the genomic SMAD-binding element (SBE). Transcriptional activation by the


56

TGF-β/SMAD pathway is regulated by the antagonistic inhibitory SMADs (I-SMADs:

SMAD 6 and SMAD7), and nuclear proteins SKI and SNO. In the non-SMAD

pathways, the phosphorylated tyrosines of the intracellular receptor domains are capable

of activating Erk through-Ras, Raf, and the downstream MAPK cascades to regulate

gene transcription. The activated TGF-β receptor complex is also able to transmit signal

through other factors, such as TRAF4 or TRAF6, TGF-β-activated kinase (TAK) 1,

phosphoinositide 3 kinase (PI3K)-AKT, JNK and NF-κB (Akhurst et al., 2012; Zhang,

2009).

The role of TGF-β in bone homeostasis is demonstrated by mice lacking TGF-β1 gene,

which exhibit decreased bone mass and elasticity (Geiser et al., 1998). TGF-β1 is

important for the induction of bone mesenchymal stem cells to coordinate bone

formation (Tang et al., 2009). Inhibition of TGF-β receptor signalling in transgenic

mice expressing cytoplasmic truncated TβRII leads to decreased remodelling and aged-

dependent increase in trabecular bone mass with lower levels of osteolysis due to

decreased bone resorption by osteoclasts (Filvaroff et al., 1999). In contrast, the

overexpression of osteoblast-specific TGF-β2 in transgenic mice resulted in an

osteoporotic phenotype due to increased osteoblastic matrix deposition, however this

was coupled with defective bone mineralisation and accompanied by increased

osteoclastic bone resorption (Erlebacher et al., 1996). In humans, TGF-β3 serum levels

were highly correlated with serum osteocalcin levels, suggesting a systemic regulation

of the cytokine in bone homeostasis (Hering et al., 2001).

The osteoimmunoregulatory effects of TGF-β are highlighted in the synovial fluid from

RA patients, which contain high levels of both latent and active TGF-β. Additionally,

arthritic synovial tissues isolated from the same patients are reported to express TGF-β

(Brennan et al., 1990). TGF-β1 is critical in the prevention of autoimmunity, as TGF-

β1-knockout and mutant mice succumb to rapid wasting due to excessive inflammatory

response (Kulkarni et al., 1993; Shull et al., 1992). In arthritic rat model, the systemic

administration of TGF-β1 was described to reduced infiltration of immune cells, thus

preventing the development of polyarthritis (Brandes et al., 1991). On the contrary,

human fibroblast-like synoviocytes isolated from RA and OA patients are reported to

produce higher levels of TNF-α, MIP-1, IL-1 and IL-8, all of which are pro-

inflammatory cytokines, upon TGF-β1 treatment (Cheon et al., 2002). Synovial TGF-β1


57

serum level is associated with the polymorphisms at the TGF-β1 genes, which may

contribute to the progression of joint destruction (Kim et al., 2004).

Interestingly, the severity of RA and OA in human patients correlates with the

upregulation of TGF-β1 (Pohlers et al., 2007). This appears to be consistent with the

anabolic role of TGF-β1, whereby the progression of arthritis is often marked by, but

not limited to, the formation of osteophytes (bone spurs along the joint margins) due to

hyper-modelling at the damaged cartilaginous area. The dual role of TGF-β is further

confounded by other studies which revealed that TGF-β directly acts on RANKL- and

M-CSF-dependent-BMMs to promote osteoclastogenesis via TRAF6/TAK1 pathway

(Fox et al., 2000; Kaneda et al., 2000; Kim et al., 2005; Yasui et al., 2011), whereas

inhibits osteoclastogenesis by stimulating osteoprotegerin and repressing RANKL

production by bone marrow stromal cells and osteoblasts respectively (Quinn et al.,

2001; Takai et al., 1998). These contrasting effects indicate that TGF-β influence

several aspects of osteoblast and osteoclast biology.

2.5. The orchestration of adaptive immune cells by innate immune cells,

osteoclastogenic RANKL and its antagonist OPG

All of the aforementioned cytokines plays a bridging role in the recruitment,

development, expansion and activation of the more specific adaptive immune response.

Correspondingly, B cell-subsets and T cells mediate the adaptive immune system.

Developmentally, B cells originate from the bone marrow, while T cells are generated

in the thymus where circulating lymphoid progenitor cells from the bone marrow are

recruited to acquire cognate receptors and functional reactivity.

The potent stimulator of osteoclastogenesis RANKL, previously known as the TNF

ligand super family member 11 (TNFSF11), and its receptor RANK were first described

to be expressed on antigen-presenting dendritic cells and to play a role in T cell

expansion (Anderson et al., 1997; Bachmann et al., 1999; Wong et al., 1997b). Deletion

of RANKL or RANK resulted in not only an osteopetrotic bone phenotype, but also the

absence of all peripheral and mesenteric lymph nodes (Dougall et al., 1999; Kong et al.,

1999b). The absence of gross lymph nodes in these mouse models is accompanied by

dramatic decreases in B cell development and number. Paradoxically, mouse model

lacking RANK expression on B cell showed normal B cell development with no defects

in antibody production, class switch recombination and somatic hypermutation (Perlot


58

et al., 2012). RANKL-deficient mice also display impaired thymocyte development,

which is not observed in RANK-deficient mice, thus indicating that RANKL regulates

not only osteoclastogenesis, but also early thymocyte development at the stage of pre-T

cell receptor (TCR) expression (Kim et al., 2000; Kong et al., 1999b).

OPG, the decoy receptor of RANKL historically attributed to osteoblasts that inhibits

osteoclastogenesis, is upregulated in B cell and dendritic cell through the activation of

CD40 (Yun et al., 1998). Mature B cells are the source of a large proportion of OPG in

bone marrow, and knockout mice lacking B cells were found to be osteoporotic (Li et

al., 2007b). The role of OPG in the regulation of B cell development and function is

emphasised in OPG-deficient mice, which are also osteoporotic due to overzealous

osteoclastogenesis resulting in a high incidence of fractures (Bucay et al., 1998). The

ability to sustain antigen response is compromised in OPG-deficient mice due to the

perturbed B cell maturation. Functional assays revealed the hyper responsiveness of

pro-B cells in OPG-deficient mice, however with less efficacy in antibody isotype

switch and production, in comparison to those of wild-type control (Yun et al., 2001).

In sum, these studies demonstrate that RANKL/RANK/OPG axis regulates the

development and activation of adaptive immune cells.

A further crosslink between the innate and adaptive immune system is that macrophages

and dendritic cells serve as antigen presenting cells (APCs), which are required to

activate T-dependent responses. After the recognition by TLRs and phagocytosis by

APCs, the antigen is processed by proteolysis and displayed by MHC class II molecules

on the cell surface. The interaction between TCR/CD3 complex with peptide-MHC

complex (pMHC) and co-stimulation of T cell surface CD28 with CD80/CD86 are

required for T cell activation. The continuous co-stimulation of T cell is tightly

regulated by T cell surface molecules such as cytotoxic T-lymphocyte protein 4 (CTLA-

4), which interrupts CD80/CD86-CD28 interactions and provides a feedback signal to

APCs (Walunas et al., 1996). CTLA-4 has been reported to directly inhibit osteoclast

differentiation (Axmann et al., 2008). Such an observation is anticipated since

osteoclast precursors are derived from common progenitors with APCs, and osteoclasts

have been reported to display APC-like features and express cell surface MHC class I

and II, CD80 and CD86 (Li et al., 2010). Interestingly, CD80/CD86 deficient mice

display decreased bone mass due to increased osteoclast numbers. In vitro studies


59

further demonstrated that CD80-/CD86- monocytic-derived osteoclasts are resistant to

physiological inhibition by CTLA-4, indicating that the engagement of CD80/CD86 by

CTLA-4 is important during the inhibition of osteoclastogenesis (Bozec et al., 2014).

The recognised general signalling pathways initiated by the dynamic interaction

between TCR and pMHC have been reviewed comprehensively (Au-Yeung et al., 2009;

Cheng et al., 2011; Salmond et al., 2009; Smith-Garvin et al., 2009). Briefly, proximal

downstream signalling cascade of TCR is mediated by the associated CD3 (γ, δ, and ε)

and ζ chains, which contain conserved motifs of the immune-receptor tyrosine-based

activation motif (ITAM) sequences. The tyrosines within these motifs are

phosphorylated by Src family kinases p56lck (Lck) or p59fyn (Fyn) present at the

cytoplasmic tails of CD4 or CD8, co-receptors that co-engaged with the TCR. The

phosphorylation of ITAM leads to the recruitment and activation of Zap70 tyrosine

kinase. Zap70 proceeds to phosphorylate adaptor proteins Linker of Activation of T

cells (LAT) and SH2 domain-containing leukocyte phosphoprotein of 76kDa (SLP-76),

which results in downstream production of inositol 1,4,5 triphosphate (IP3) and

diacylglycerol (DAG) to perform cell signalling. IP3 releases Ca2+ from the

endoplasmic reticulum and the elevation of intracellular Ca2+ activates calcineurin,

calmodulin and the transcription factor NFAT. DAG initiates two major pathways, Ras

and PKC-θ signalling, which triggers MAPK and NF-κB pathways respectively. The

aforementioned signalling cascades eventually lead to the activation and regulation of

transcription factors in activated T cells.

2.6. Delineation of T cell and osteoclastogenic T cell candidates

T cells are a part of the adaptive immunity that can be subdivided based on the

expression of co-receptors, CD4 helper T cells and CD8 cytotoxic T cells. CD4 T cells

assist the maturation of B cells as well as the activation of CD8 T cell and other immune

cells. Activated CD8 T cells terminate virally infected cells and tumour cells through at

least three mechanisms: granular exocytosis of cytotoxic molecules, Fas/FasL-induced

apoptosis and secretion of TNF-α. Distinctively, the generation and differentiation of

CD4 T cells are driven by the interaction between cytokine-secreting APCs that are

delivered to naïve T cells. Activated CD4 T cells can differentiate and trans-

differentiate into TH1, TH2, TH17, Foxp3+ regulatory T cell (Tregs) and TFH. Other

TH cells in the immune system that are still not well characterised include TH9 and


60

TH22, which secrete IL-9 and IL-22 respectively.

The secretion of IL-12 and IFN-γ by innate immune cells initiates TH1 responses

against intracellular pathogen. These cytokines activate JAK/STAT pathway to induce

the transcription factor T-bet and turn on the IFN-γ gene. The secretion of IFN-γ in turn

activates macrophages and influences B cell maturation, and amplifies their own

responsiveness to IL-12. Correspondingly, parasitic infections favour the development

of TH2-mediated humoral responses. TH2 targeting extracellular pathogen requires IL-4

to activate STAT4. STAT4 promotes the expression of transcription factor GATA-3,

which stimulates the production of IL4, 5 and 13 to inhibit TH1 polarisation and

macrophage activation while stimulating B cells. TH1 and TH2 regulate each other

whereby IFN-γ inhibits TH2 polarisation and concomitantly IL-4 hinders TH1. The

homeostasis between the populations of these two cell types is critical in preventing

pathological consequences.

TH17 is a T helper subset specialised in the protection against extracellular bacterial

and fungal pathogens. The generation of IL-17 to induce local inflammation is governed

TGF-β and IL-6, which signal through STAT3 to upregulate the expression of RAR-

related orphan receptor gamma (RORγt), the master regulator of this lineage (Ivanov et

al., 2006). TGF-β also induces the Tregs-specific transcription factor Foxp3 via the

phosphorylation of SMAD2/3 proteins and the formation of a complex with SMAD4.

However in the presence of IL-6, TGF-β antagonistically inhibits the generation of

Tregs and induces TH17 cells that are highly pro-inflammatory. Tregs are responsible

for maintaining tolerance to self-antigens to prevent autoimmunity and the reciprocal

proportion of TH17 versus Tregs population is clearly demonstrated to be critical for

mediating self-tolerance and preventing tissue inflammation. An imbalance favouring

TH17 differentiation eventually leads to inflammatory disorders and autoimmune

diseases (Bettelli et al., 2006; Mucida et al., 2007). Autoimmune diseases are therefore

the outcome of the imbalance between Treg and TH17 cells, whereby TGF-β is the key

to the delicate balance between immune activation (immunity) and immune non-

responsiveness (tolerance).

CD4 T cells expressing C-X-C chemokine receptor type 5 (CXCR5) are phenotypically

distinct from TH1 and TH2 cells and are named TFH (Breitfeld et al., 2000; Chtanova

et al., 2004; Schaerli et al., 2000). The differentiation of CD4 T cells towards TFH is


61

coordinated by the transcription factor B-cell lymphoma 6 (Bcl6) and is mediated by

STAT3, IL-21, IL-6 and B cell expression of ICOS ligand (ICOSL) (Akiba et al., 2005;

Nurieva et al., 2008; Nurieva et al., 2009; Rasheed et al., 2006; Vogelzang et al., 2008).

TFH express IL-21 to directly regulate B cell proliferation and class switching, and

potentially provide an autocrine action with its concomitant expression of IL-21R

(Chtanova et al., 2004; Nurieva et al., 2008; Rasheed et al., 2006; Vinuesa et al., 2005a;

Vogelzang et al., 2008). The dysregulation of TFH is implicated in many autoimmune

diseases that are partially mediated by autoantibodies, thus highlighting its role in

humoral immunity as professional B cell helper.

Early osteoimmunology studies have shown that activated T cells express RANKL to

directly induce osteoclastogenesis (Horwood et al., 1999; Kong et al., 1999a; Kong et

al., 1999b; Kotake et al., 2001; Weitzmann et al., 2001; Wong et al., 1997b). In this

scenario, TH1 was previously associated with bone destruction, while its antagonist

TH2 was described to attenuate the progression. However, these subsets of activated T

cells also produce cytokines that dampened RANKL-signalling, for example TH1

secrete IFN-γ and TNF, while TH2 produce IL-4, IL-5, IL-10 and IL-13, some of which

are anti-osteoclastogenic (Pappalardo et al., 2013; Takayanagi et al., 2000). Beyond the

TH1/TH2 paradigm, TH17 cells emerged as the osteoclastogenic subset in view of their

pro-inflammatory nature. Indeed the expression of IL-17 uniquely produced by TH17 is

elevated in classic autoimmune diseases such as RA (Kotake et al., 1999; Leipe et al.,

2010; van Hamburg et al., 2011). IL-17 is capable of triggering local inflammation and

the production of inflammatory cytokines. Befitting to its osteoclastogenic descriptions,

TH17 cells express significant amounts of membrane-bound RANKL and do not

produce large amounts of IFN-γ in comparison to TH1 (Sato et al., 2006).

Their role in osteoclastogenesis and the development of arthritis is evident in IL-17-/-

mice, which are protected from CIA (Nakae et al., 2003). Accordingly, local injection

of IL-17 into the knee joint of an arthritis mouse model shows that IL-17

overexpression exacerbates CIA and the neutralisation of IL-17 abates the disease

(Burchill et al., 2003; Lubberts et al., 2002). It is important to note however, that their

osteoclastogenic nature is largely indirect via the influence of IL-17 to stimulate the

production of inflammatory cytokines and RANKL expression on synovial fibroblasts

and mesenchymal cells (Kotake et al., 1999; van Hamburg et al., 2011). Consequently,


62

therapeutic preventions to counteract the differentiation and infiltration of TH17 cells

into the inflammatory joint are of potentially great clinical importance. Clinical trials on

pharmaceutical inhibitors of IL-17 and cytokines favouring TH17 cell differentiation

and expansion such as IL-6, IL-23 and TGF-β, have shown promising results in the

treatment of inflammatory systemic bone loss such as RA and AS (D.Baeten et al.,

2011; Genovese et al., 2010; McInnes et al., 2014).

The neutralisation of IL-17 and IL-6 switched off the generation of TH17 leading to

preferential differentiation to Tregs (Nardelli et al., 2004; Nishihara et al., 2007). On

account of the fact that Tregs attenuate the pro-inflammatory behaviours of TH17, the

elimination of Tregs in vivo is shown to aggravate arthritis while the supplementation of

Tregs decreased the severity of the disease (Frey et al., 2005; Morgan et al., 2005). The

anti-catabolic effects of Tregs on bone appeared to be via their aptitude to directly

inhibit osteoclastogenesis (Kim et al., 2007b; Zaiss et al., 2010). The suppression of

osteoclast formation and activity is supported by the engagement CTLA-4 via cell-cell

contact (Zaiss et al., 2007). Nevertheless, it was recently revealed that under arthritic

conditions, the stability of the transcription factor Foxp3, which is indispensable for the

anti-inflammatory function of Tregs, is compromised under arthritic conditions

(Komatsu et al., 2014). Komatsu and colleagues reported that synovial fibroblast-

derived IL-6 mediates the loss of Foxp3 expression in Tregs leading them to undergo

transdifferentiation into TH17, resulting in accumulation of IL-17 in inflamed joints

(Komatsu et al., 2014). The conversion towards TH17 suggests the plasticity of Tregs

cells in contributing to the pathogenesis of RA and hence, emphasising the importance

of inflammatory cytokines as potent stimulators that drive progression of inflammatory

bone diseases.

Similarly, dysregulation of TFH tipping the balance of the immune system towards

favouring inflammation mediates autoimmune pathologies. Circulating TFH-like CD4+

CXCR5+ T cells were recently documented to be elevated in patients with autoimmune

disorders such as RA and SLE. The increased frequency of TFH correlates with high

levels of IL-21, Bcl6 expression, autoantibody titres and tissue damage (Ma et al., 2012;

Simpson et al., 2010; Terrier et al., 2012). Several mouse models have provided

insights into the role of TFH in autoimmunity. San Roque is an ENU-induced mutant

mouse line, which exhibits a spontaneous deregulation of the germinal centres in the


63

spleen, and generation of circulating autoantibodies associated with an increased in

TFH and IL-21 secretion (Vinuesa et al., 2005a). The single point mutation in the

Roquin-1 (Rc3h1) gene resulted in SLE-like disease and the abrogation of TFH in the

mutant mice ameliorated the pathology, demonstrating that the dysregulation of TFH

cell function mediates the disease (Vinuesa et al., 2005a). Roquin-1 and its paralog

Roquin-2 appear to be crucial in directing the degradation of inducible T-cell

costimulator (ICOS) mRNA (Pratama et al., 2013). ICOS expression on T cells

provides co-stimulation for T cell responses and is critical for TFH cell survival,

germinal centre reactions and generation of B-cell memory (Akiba et al., 2005;

Bossaller et al., 2006; Dong et al., 2001a; Dong et al., 2001b; Mak et al., 2003). Thus,

the disruption of the repressive activity by the mutation in Roquin gene resulted in

increased ICOS expression and accumulation of lymphocytes (Pratama et al., 2013; Yu

et al., 2007).

Interestingly, microarray analysis performed by Vinuesa and colleague revealed that the

TFH isolated from San Roque mice expressed higher level of RANKL in comparison to

WT TFH (Vinuesa et al., 2005a). These observations raised the possibility of TFH to

emerge as a potential osteoclastogenic T cell subset, by directly or indirectly facilitating

osteoclastogenesis, to expedite systemic bone loss in autoimmune disorders.

2.7. The immunomodulatory role of bone cells in establishing the

haematopoietic stem cell (HSC) niche

The bone organ system is heavily integrated with the lymphatic network, harbouring

haematopoietic progenitors with multilineage potential, and is the site of B cell

development and maturation. Within the bone marrow cavity, bone cells share their

microenvironment with HSCs, thus inevitably participating in haematopoiesis and the

modulation of immune cells. Accumulating evidence indicates that cells of the

osteoblast lineage are important components in the maintenance of the HSC niche

(Calvi et al., 2003; Taichman et al., 1994; Taichman et al., 1996; Visnjic et al., 2004).

Taichman et al. (1994) documented the first direct evidence that human osteoblasts

support basal haematopoiesis by constitutive expression of granulocyte colony-

stimulating factor (G-CSF), GM-CSF, IL-6 and TNF-α (Taichman et al., 1994). HSCs

are present in close proximity to endosteal surfaces where osteoblasts and bone lining

cells reside in vivo, and accordingly, osteoblasts are able to support the maintenance of


64

an immature haematopoietic phenotype of bone marrow cells in an in vitro co-culture

system (Taichman et al., 1996). An increase in osteoblastic cells directly boosts the

number of HSCs in vivo via the Notch signalling pathway (Calvi et al., 2003).

Conversely, osteoblast deficiency leads to severe alteration of haematopoiesis, resulting

in the lost of lymphoid, erythroid and myeloid progenitors in transgenic mice with

induced-conditional ablation of the osteoblast lineage (Visnjic et al., 2004).

In addition to their role in HSC niche maintenance, osteoblasts are also documented to

support the microenvironment for sustaining B cells development. In vitro experiment

revealed that bone marrow differentiation towards B cells requires cell-cell contact with

osteoblasts. The selective ablation of osteoblast in transgenic mice severely depleted B

cell lineages, preceding the decline in HSCs (Zhu et al., 2007). Furthermore,

osteoblastic cells express CXCL12 and IL-7, which are crucial factors for B cell

development (Jung et al., 2006; Nagasawa et al., 1996; Zhu et al., 2007). Mice lacking

heterotrimeric G protein α subunit (Gsα), which is required in early osteoblast lineage,

exhibit a B lymphocytopenia due to attenuated IL-7 expression with other mature

hematopoietic lineages being not significantly affected (Wu et al., 2008a).

B cell development in the bone marrow is also modulated by osteoclasts. Osteopetrotic

mice with defective osteoclast bone resorption exhibit a reduction in B cell population.

Zoledronic acid (ZA)-induced osteopetrosis similarly resulted in B lymphocytopenia

within the bone marrow of ZA-treated mice. The aberrant B cell development is the

consequence of decreased expressions of CXCL12 and IL-7 associated with reduced

osteoblastic engagement by inactive osteoclasts (Mansour et al., 2011).

The absence of sclerostin (SOST), a secreted protein expressed by osteocytes, was

recently reported to cause an adverse affects on B cell survival. SOST is well

characterised to induce osteoblast apoptosis and effectively preventing osteoblast

activity and maturation into osteocytes. Mice with the deletions of Sost-coding region

display increased mineralised bone due to increased osteoblast activity, resulting in

reduced bone marrow cavity. Strikingly, the bone marrow of Sost-knockout mice is

depleted of B cells, and this is supplemented by a reduction in CXCL12 (Cain et al.,

2012). Collectively, these observations reinforce the role of bone cells in supporting

normal B cell development, and drastic alteration of the bone microenvironment is

therefore disadvantageous for haematopoiesis.


65

2.8. Conclusion

Inflammatory bone loss highlights the abundance of overlapping molecules and

regulatory pathways shared between the bone organ and immune system. It is now

established that there exists a constant dialogue between these systems in both health

and disease. Within the past four decades, the field of osteoimmunology has made

significant advances in enriching our understanding of the interactions between immune

and bone cells within the bone microenvironment. While the regulation of osteoclast

biology has been widely studied and demonstrated to be a promising target in treatment

of inflammatory bone loss, accumulating evidence shows that to tackle inflammatory

bone diseases requires more than just targeting the RANKL/RANK/OPG pathway. The

use of genetically modified mice will greatly facilitate our understanding of the

molecular basis regulating cell lineage determination and interactions of immune and

bone cells. It is hoped that better understanding leads to the development of strategic

and efficient therapies to prevent and reverse human diseases that affect both systems.


66

Figure 2.1 Lineage bifurcation model between dendritic cells, macrophages and

osteoclasts. Differentiation between macrophages, osteoclasts and dendritic cells from

common myeloid progenitors (Lin- CD117+ CD11b+ CD115+ cells). The differentiation

of macrophages osteoclasts is supported by M-CSF. RANKL is required for

osteoclastogenesis, which is inhibited by GM-CSF through reduction of c-Fos

expression. In contrast, dendritic cell maturation induced by GM-CSF and is inhibited

by the expression of c-Fos or M-CSF.


67

Osteoclast

Osteoblast/

Stromal Cells T cells

IFNs Osteoclastogenesis

Bone resorption

RANKL

TH1 expansion

TH2 expansion

IL-1 Osteoclastogenesis

Resorption

IL-10 Osteoclastogenesis BMP2

IL-17 Resorption RANKL

IL-21 ? ? RANKL

TH17 expansion

IL-4/

IL-13

Osteoclastogenesis

Actin-ring formation

Resorption

= TH2 expansion

IL-5 = Bone formation

IL-6 Recruitment of

osteoclast precursors RANKL

Foxp3

Tregs expansion

TH17 and TFH

expansion

IL-8 Osteoclastogenesis

Resorption RANKL

TGF-β Osteoclastogenesis

Bone formation

OPG

RANKL

Foxp3

Tregs expansion

TNFs c-fms expression

Osteoclastogenesis

RANKL

DKK1

Wnt/β-catenin

Apoptosis

Table 2.1 The effect of cytokines in modulating the bone and immune system.

Cytokines highlighted above are involved in the regulation of immune system and

behave as a double-edged sword to modulate bone remodelling. Whether IL-21 directly

regulates osteoclastogenesis and/or osteoblastogenesis remained to be elucidated.

CHAPTER THREE

HYPOTHESIS AND AIMS

SANROQUE MUTANT MOUSE AS AN OSTEOIMMUNE

MOUSE MODEL

Chapter 3: Hypothesis and Aims – Sanroque Mutant Mouse as an Osteoimmune Model

69

3. CHAPTER 3: HYPOTHESIS AND AIMS – SANROQUE MUTANT MOUSE AS AN OSTEOIMMUNE MOUSE MODEL

3.1. General hypothesis and aims

The bone organ tissue is continuously being broken down and rebuilt in a dynamic

process referred to as bone remodelling. The remodelling process is the coupled activity

between the catabolic effects of bone resorbing osteoclasts and the anabolic effects of

bone forming osteoblasts. The bone remodelling cycle begins when there is an external

stimulus, which initiates osteoblasts to recruit osteoclast precursors by expressing

RANKL. RANKL binds onto the cell surface receptor RANK displayed on osteoclast

precursors to drive osteoclastogenesis and cell fusion, forming multinucleated

osteoclasts. Once osteoclasts mature, they begin to secrete proteolytic enzymes such as

cathepsin K to degrade the bone matrix. To prevent osteoclasts from continuously

resorbing bone, osteoblasts also secrete reversal cues, such as osteoprotegerin that acts

as a decoy receptor by binding onto RANKL. Such cues allow osteoblasts to modulate

the bone remodelling cycle to favour bone formation.

The coupled activity between these two cell types is tightly regulated to achieve a

micro-architecturally strong bone. The dysregulation of the homeostasis within this

bone microenvironment eventually leads to bone diseases. However, the underlying

molecular and cellular mechanisms regulating the state of homeostasis remains poorly

resolved. Understanding this homeostatic mechanism is further complicated by the fact

that the bone microenvironment is supported by, and supports, many other systems and

cell types. Immune cells, which are hematopoietically generated within the marrow

cavity, inevitably infiltrate into the microenvironment and interact with bone cells and

vice versa. The co-regulation and dialogue between these two systems is termed

osteoimmunology, and is highlighted in many inflammatory diseases such as

rheumatoid arthritis, spondylarthritis, periodontal disease, osteomyelitis and lupus.

All of these aforementioned osteoimmune diseases caused significant discomfort and

dilapidating deformities. However, the mechanism(s) by which these conditions occur is

still largely unknown. Therefore, the underlying hypothesis for this project is that there

are factors implicated in regulating the cross talk between the bone organ and immune

system that are yet to be identified. Based on this hypothesis, the aim of this study was

to identify novel factors that play a significant role in the maintenance and regulation of

the osteoimmune system through a chemically induced (ENU) mutagenesis phenotypic

screen.


70

3.2. Screening of ENU-induced mutant mouse lines: the sanroque mutation

Systematic screening of the ENU-mutagenesis library led us to identify a mouse line

named sanroque, which carried a single point M199R mutation in the Roquin (Rc3h1)

gene (Figure 3.1). The T to G substitution in the mutant mice resulted in a non-

conservative Methionine to Arginine codon change in mRNA transcript. The consensus

of the encoded Roquin protein conforms to the structure of RING-type E3 ubiquitin

ligase family, which is one of the enzymatic components that drive the ubiquitin

proteasome system (UPS).

3.2.1. The ubiquitin proteasome system (UPS) and Roquin E3 ligase

The main function of the UPS is the ubiquitination (also referred to as ubiquitylation) of

molecular products for the purpose of proteolysis to maintain protein homeostasis. The

ubiquitination process marks the protein of interest and signals the proteasome to

degrade and recycle the protein. Besides protein degradation, ubiquitination is also

reported to be involved in transcription factor activation, DNA repair, translational

control, kinase activation, histone activity and endocytosis (Pickart et al., 2004). The

UPS is an energy-dependent enzymatic process driven by three enzymes: E1 ubiquitin

activating enzyme, E2 ubiquitin conjugating and E3 ubiquitin ligase (Figure 3.2).

E1, along with ATP, is responsible for the activation of ubiquitin to initiate the

ubiquitination process. E1 enzyme then transfers the activated ubiquitin to E2. E2 acts

as a platform and forms a complex with an E3. E3 recognises both E2 and the protein

that needs to be tagged and catalyses the transfer of activated ubiquitin onto the protein.

E3s are therefore primarily responsible for conferring specificity to ubiquitination and

can be defined by one of the three domains: Homologous to E6-AP Carboxyl Terminus

(HECT), Really Interesting New Gene (RING) and RING-related domains, and A20-

type zinc finger. Roquin E3 ligase is characterised to have the RING domain (Figure

3.1).

The dysfunction of the UPS enzymes, especially E3s, has been implicated in the

dysregulation of biological activities leading to diseases such as, but not limited to,

Parkinson’s, leprosy, osteosarcoma and autoimmune diseases (Ardley et al., 2004;

Cohen, 2014; Severe et al., 2013). UPS and E3s also play a major role in bone

homeostasis and in the development of bone cells (Shu et al., 2013; Zhang et al.,

2013a).


71

3.2.2. Sanroque mutant mice exhibits low bone mass phenotype

While there was no apparent abnormality in the gross body phenotype of the mutant

mice in comparison to the wildtype littermates (Figure 3.3 A), X-ray imaging revealed

that the hindlimbs of sanroque mice have reduced bone mass in comparison to wildtype

hindlimbs. The reduction in bone mass of the sanroque mice was consistently observed

in the vertebrae and tail, indicating systemic bone loss when compared to wildtype

controls (Figure 3.3 B – D).

3.2.3. Sanroque mice display a lupus-like autoimmune disease

Vinuesa and colleagues (2005) were the first to characterise the sanroque mutant mouse

line, which develop high titres of autoantibodies and displayed a severe autoimmune

disease consistent to Systemic Lupus Erythematosus (SLE). The mutation in the Roquin

gene led to the dysregulation of T cell maturation, causing the formation of excessive

numbers of follicular helper T cells (TFH) and germinal centres. Consequently, the

sanroque mutant mouse line exhibited enlarged lymph nodes and spleens (Figure 3.4).

The loss of function mutation disrupted a repressor of ICOS, which is a crucial co-

stimulator receptor for TFH, and resulted in the excessive production of the cytokine

interleukin-21. Roquin gene thus plays a regulatory role in preventing self-reactive T

cells from providing assistance to B cell maturation and production of self-reactive

antibodies.

3.3. Specific hypothesis and aims

In view of the susceptibility of sanroque mice to SLE-like autoimmunity, it is

worthwhile to mention that the effect of SLE disease per se contributes to systemic bone

loss and the deterioration of the patients’ bone mineral density (BMD), microstructure

and strength (Tang et al., 2013a). Consistent with this report, preliminary X-ray

screening revealed the low bone mass phenotype of the sanroque autoimmune mouse

line when compared to their wildtype counterparts. However, there is also a possibility

whereby the autoimmunity is developed due to a compromised bone microenvironment

that consequently resulted in the abnormal development of myeloid cell lineages. Which

of these disease development models are correct remains to be answered.

Interestingly, microarray analysis performed by Vinuesa and colleague revealed that the

TFH isolated from sanroque mice expressed higher level of RANKL in comparison to


72

wildtype TFH (Vinuesa et al., 2005a). These observations raised the possibility of TFH

to emerge as a potential osteoclastogenic T cell subset, by directly or indirectly

facilitating osteoclastogenesis, to expedite systemic bone loss in autoimmune disorders.

Hence we hypothesise that Roquin is an important regulator of bone homeostasis and

sanroque mice emerge as a useful model to gain osteoimmunological insight into

systemic bone loss in autoimmune diseases. The specific aims of this project are:

1. To characterise the bone phenotype of the sanroque mutant mice.

2. To elucidate the pathology of systemic bone loss by understanding the cellular

and molecular mechanisms by which Roquin regulates bone homeostasis.


73

Figure 3.1 Mapping of missense mutation in the conserved Rc3h1gene. (A) The T to

G substitution in roquin mRNA resulted in a non-conservative Methionine to Arginine

codon change (M199R) of the encoded protein. (B) The primary structure of Roquin

illustrating the M199R in the conserved ROQ domain. (C) The RING finger domain of

Roquin aligned perfectly with other RING-type E3 ubiquitin ligases. Figures are

adapted from Vinuesa et al., 2005.

(B)

(C)

(A)

Wildtype

sanroqueENU-induced mutation

Wildtype

sanroque


74

Figure 3.2 The basic steps of substrate modification by the ubiquitin proteasome

system. E1 activates ubiquitin and transfers the activated ubiquitin to E2. E2 serves as a

platform to facilitate the ligation of ubiquitin and substrate by E3. The ubiquitinated

substrate is now marked and recognised by the proteasome for degradation. The

degraded products can be recycled for substrate synthesis.


75

Figure 3.3 Mutant sanroque mice appeared normal but exhibited low bone

phenotype. (A) Phenotypic gross body comparison of 99 day old male wildtype and

sanroque mutant mice. The mutant mice appeared normal in comparison to wildtype

littermates. X-ray images of (B) hindlimbs, (C) vertebrae and (D) tails from 100 day old

male wildtype and sanroque mutant mice highlight the decreased bone mass of

sanroque mice.

(A)

(B)

A)

Wild

type

sa

nroq

ue

Wildtype Wildtype sanroque

Wildtype Wildtype sanroque sanroque(C)

(D)


76

Figure 3.4 Mutant sanroque mice displayed lymphomegaly and splenomegaly. (A)

Post-mortem dissection revealed grotesquely enlarged spleens and (B) lymph nodes of

the mutant mice. Image (B) was taken from Vinuesa et al., 2005.

(A) (B) A)

Wild

type

sa

nroq

ue

sanroque

CHAPTER FOUR

MATERIALS AND METHODS

Chapter 4: Materials and Methods

78

4. CHAPTER 4: MATERIALS AND METHODS

4.1. Materials

4.1.1. Chemical reagents

All chemical reagents involved in all experiments conducted in this thesis were of

analytical grade unless indicated otherwise and were purchased from the following

manufacturers:

Description Manufacturer

1α, 25-dihydroxyvitamin D3 Sigma-Aldrich, USA

2-Ethoxyethanol Sigma-Aldrich, USA

2-Mercaptoethanol 14.3M Sigma-Aldrich, USA

30% Acrylamide/Bis solution Bio-Rad, USA

Acetone Scot Scientific, Australia

Agarose powder Promega Corp., Madison,

Wi., USA

Alizarin Red S Monohydrage MP Biochemicals Inc, USA

Ammonium Persulfate (APS) Sigma-Aldrich, USA

Baxter water Baxter Healthcare Pty Ltd,

Australia

Bovine Serum Albumin Sigma-Aldrich, USA

Bromophenol Blue Sigma-Aldrich, USA

Chloroform ThermoFisher Scientific,

Australia

ClearMount Mounting Solution Invitrogen, Australia

Dexamethasone Sigma-Aldrich, USA

Dimethyl Sulphoxide (DMSO) BDH Laboratory Supplies,

England

Distyrene Plasticizer Xylene (DPX) Sigma-Aldrich, USA

Eosin Y disodium salt Sigma-Aldrich, USA

Ethanol (100%) BDH Laboratory Supplies,

England

Ethidium bromide (2,7-diamino-10-ethyl-9-

phenylphenanthridium bromide;

homidium bromide)

Sigma Chemical Co., USA


79

Ethylenediamine tetraacetic acid (EDTA) BDH Laboratory Supplies,

England

Formaldehyde Sigma-Aldrich, USA

Glacial acetic acid BDH Laboratory Supplies,

England

Glutaraldehyde solution Fisher Scientific, USA

Glycerol BDH Laboratory Supplies,

England

Glycine Sigma-Aldrich, USA

Goat Serum Sigma-Aldrich, USA

Haematoxylin Sigma-Aldrich, USA

Hydrochloric acid BDH Laboratory Supplies,

England

IPTG (Isopropyl-β-thiogalactopyranoside) dioxane free Promega Corp., Madison,

Wi., USA

Isopropanol ThermoFisher Scientific,

Australia

L-ascorbic acid Sigma-Aldrich, USA

Liquid nitrogen (N2) Air Liquid, Australia

Methanol BDH Laboratory Supplies,

England

Methoxyethyl acetate (MEA) Sigma-Aldrich, USA

Methyl methacrylate (MMA) Sigma-Aldrich, USA

Mono-sodium phosphate (NaH2PO4) Sigma-Aldrich, USA

Naphthol AS-MX phosphate Sigma-Aldrich, USA

Paraformaldehyde Sigma-Aldrich, USA

PMSF (Phenylmethylsulfonyl fluoride) Roche Diagnostic,

Germany

Prolong Gold Antifade reagent Invitrogen, Australia

Propan-2-ol (Isopropanol) BDH Laboratory Supplies,

England

Protease Inhibitor Cocktail Roche Diagnostic,

Germany


80

Skim Milk Powder Standard supermarket

brand

Sodium acetate (CH3COONa) Sigma-Aldrich, USA

Sodium chloride (NaCl) BDH Laboratory Supplies,

England

Sodium deoxycholate (C24H39NaO4) BDH Laboratory Supplies,

England

Sodium dodecyl sulphate (SDS) BDH Laboratory Supplies,

England

Sodium hydroxide (NaOH) BDH Laboratory Supplies,

England

Sodium Iodate (NaIO3) Sigma-Aldrich, USA

Sodium Orthovanadate Sigma-Aldrich, USA

Sodium phosphate dibasic (Na2HPO4) Sigma-Aldrich, USA

Sodium phosphate monobasic (NaH2PO4) Sigma-Aldrich, USA

Sodium tartrate dihydrate (C4H4Na2O6·2H2O) AJAX chemicals, Australia

TEMED (N. N, N', N'-tetra-methyl-ethylenediamine) Bio-Rad, USA

Tissue-Tek® OCT compound for cryostat sectioning Sakura Finetek Inc, Usa

Toluidine Blue Sigma-Aldrich, USA

Triton X-100 Sigma-Aldrich, USA

Trizma Base, Tris Sigma-Aldrich, USA

Trizma Hydrochloride Sigma-Aldrich, USA

Tween-20 MP-Biochemicals, France

Xylene Hurst, Scientific, Australia

β-glycerophosphate Sigma-Aldrich, USA

β-mercaptoethanol Sigma Chemical Co., USA

4.1.2. Commercially purchased kits and molecular products


1 kb DNA ladder Promega Corp., Madison,

Wi., USA

100 bp DNA ladder Promega Corp., Madison,


81

Wi., USA

10X Reaction Buffer, 1 ml of 200mM Tris-HCl, pH 8.3,

20mM MgCl2

Sigma-Aldrich, USA

5X First Strand Buffer Invitrogen, USA

7-AAD BD Biosciences, USA

Amplification Grade DNAse I (1 u/μl) Sigma-Aldrich, USA

Anti-Mouse Ig κ/Negative Control (FBS) Compensation

Particles Set (#552843)

BD Biosciences, USA

Anti-Rat and Anti-Hamster Ig κ/Negative Control

Compensation Particles Set (#552845)

BD Biosciences, USA

BD Pharm LyseTM red cell lysis buffer (10X) BD Biosciences, USA

BD Stabilizing Fixative 3X Concentrate (#338036) BD Biosciences, USA

Bio-Rad Protein Assay Bio-Rad, USA

Blue/Orange Loading Dye, 6X Promega Corp., Madison,

Wi., USA

CellTiter 96® AQueous One Solution Cell Proliferation

Assay

Promega Corp., Madison,

Wi., USA

DAPI (4', 6-diamidino-2-phenylindole), dilactate Biotium, USA

DNase I Stop Solution Kit, Amplification Grade Sigma-Aldrich, USA

dNTPs – dATP, dTTP, dCTP and dGTP nucleotides Promega Corp., Madison,

Wi., USA

Goat serum (for immunofluorescence, blocking) Sigma-Aldrich, USA

GoTaq® Polymerase Green Master Mix Promega Corp., Madison,

Wi., USA

GoTaq® Real-Time qPCR Master Mix Promega Corp., Madison,

Wi., USA

LIVE/DEAD® Fixable Aqua Dead Cell Stain Kits

(L34957)

Invitrogen, USA

MMLV Reverse Transcriptase 10000u (200u/μl) Promega Corp., Madison,

Wi., USA

MMLV-RT (Moloney Murine Leukaemia Virus Reverse

Transcriptase)5X Buffer

Promega Corp., Madison,

Wi., USA

Oligo dT Sigma-Aldrich, USA


82

RNAsin® RNAse Inhibitior Promega Corp., Madison,

Wi., USA

RNEasy Mini Kit QIAGEN Gmbh, Australia

TRIZOL® Invitrogen, USA

Western LightningTM Ultra Chemiluminescence Substrate

Kit

PerkinElmer, USA

Zombie UVTM Fixable Viability Kit (#423107) BioLegend, USA

4.1.3. Other products and consumables


Acrodisc Syringe Filters with Supor Membrane, sterile –

0.2 μm, 0.8 μm

Pall Corp, USA

BD 1 mL Syringe Luer-LokTM Tip – sterile BD Biosciences, USA

BD PrecisionGlideTM Needles 25G×11/2" BD Biosciences, USA

Coverslips – round 5 mM, 22×22 mm2 and 22×40 mm2 Knittle Glaser, Germany

Carbon Steel Surgical Blade – sterile Swann Morton LTD, UK

Filter Paper Whatmans International,

UK

Glass Slides – Star Frost Knittle Glaser, Germany

HyBondTM-C 45 Micron Nitrocellulose Membrane Amersham Biosciences,

UK

iCycler iQTM Optical Tape Bio-Rad, USA

iCycler iQTM PCR Plates, 96 wells and 384 wells Bio-Rad, USA

Parafilm Laboratory Film American National

CanTM, USA

Petri Dishes Sarstedt, Germany

PCR Tubes: 0.2 ml thin-walled tubes Unimed Australia Pty Ltd,

Australia

Transfer pipettes Sarstedt, Germany

50 ml centrifuge tubes BD Biosciences, USA

Cling Film Glad, AUS

Diamond Wafering Blade, high concentration (for Buehler, USA


83

isoMet® Low Speed Saw)

Centrifuge Tubes – sterile, 0.5 ml, 1.5 ml and 2.0 ml Sarstedt, Germany

4.1.4. Tissue culture materials and reagents


1α, 25-dihydroxyvitamin D3 Sigma-Aldrich, USA

Biocoat Collagen I cellware – 6 well plate Beckton Dickson Labware,

USA

Cell Scrapers Sarstedt, Germany

Collagenase Type II Sigma-Aldrich, USA

Cryogenic vials (2 ml) Sarstedt, Germany

D-MEM (Dulbecco's-Modification of Eagle's Medium)-

GlutaMaxTM-1

Gibco, Life Technologies,

Australia

Falcon Cell Strainer – 100 μm BD Biosciences, USA

FBS (Foetal Bovine Serum) Gibco, Life Technologies,

Australia

Hanks' Balanced Salt Solution Gibco, Life Technologies,

Australia

L-Glutamax Gibco, Life Technologies,

Australia

Mr. FrostyTM Freezing Container Thermofisher, USA

Penicillin-Streptomycin Gibco, Life Technologies,

Australia

Serological pipettes – 5 ml, 10 ml, 25 ml Sarstedt, Germany

Tissue Culture Flask – 25 cm2, 75 cm2 Nunc, Denmark

Tissue Culture Plate – 6 wells, 12 wells, 24 wells, 96

wells

Sarstedt, Germany

TrypLETM Express Cell Dissociation Reagent (trypsin

replacement)

Gibco, Life Technologies,

Australia

α-MEM (α-Modification of Eagle's Medium) Gibco, Life Technologies,

Australia


84

4.1.5. Cytokines


Murine M-CSF (Macrophage Colony

Stimulating Factor

Generated by Cell Signalling Group,

UWA School of Pathology and Laboratory

Medicine using CMG14-12 cells

conditioned-media as previously described

(Takeshita et al., 2000) (Takeshita et al.,

2000).

Murine RANKL (Receptor Activator of

Nuclear factor Kappa-B Ligand)-GST

Produced by Cell Signalling Group, UWA

School of Pathology and Laboratory

Medicine as previously described (Xu et

al., 2000) (Xu et al., 2000).

4.1.6. Antibodies

Antibody/Probe Manufacturer Dilution

AlexaFluor®700 Rat Anti-Mouse CD8a BioLegend, USA 1 in 200 (FC)

Anti-Mouse IgG Peroxidase Conjugate

(A9917)

Sigma-Aldrich, USA 1 in 5000 (WB

2° Ab)

Anti-Mouse IκB-α (C-21) Santa Cruz

Biotechnology, USA

1 in 1000 (WB

1° Ab)

Anti-Mouse NFaTc1 (#556602) BD Biosciences, USA 1 in 1000 (WB

1° Ab)

Anti-Mouse p38 (#9212) Cell Signalling, USA 1 in 1000 (WB

1° Ab)

Anti-Mouse Phospho-ERK (E-4) Santa Cruz

Biotechnology, USA

1 in 1000 (WB

1° Ab)

Anti-Mouse Phospho-JNK (#92521) Cell Signalling, USA 1 in 1000 (WB

1° Ab)

Anti-Mouse Phospho-p38 (#9211) Cell Signalling, USA 1 in 1000 (WB

1° Ab)

Anti-Mouse SAPK/JNK (#9252) Cell Signalling, USA 1 in 1000 (WB

1° Ab)


85

Anti-Mouse β-actin (JLA20) Calbiochem (EMD

Millipore), USA

1 in 5000 (WB

1° Ab)

Anti-Rabbit IgG Peroxidase Conjugate

(A0545)

Sigma-Aldrich, USA 1 in 5000 (WB

2° Ab)

APC Rat Anti-Mouse CD115, CSF-1R

(#135509)

BioLegend, USA 1 in 200 (FC)

APC Rat Anti-Mouse CD11b (#17-0112) eBioscience, USA 1 in 100 (FC)

APC Rat Anti-Mouse CXCR5 (#560615) BD Biosciences, USA 1 in 100 (IF), 1

in 25 (FC)

APC-H7 Rat Anti-Mouse CD4 (#560181) BD Biosciences, USA 1 in 100 (FC)

Brilliant VioletTM 421 Rat-Anti Mouse

CD117, c-Kit (#562609)

BD Biosciences, USA 1 in 100 (FC)

BUV Rat Anti-Mouse CD45R/B220

(#563793)


BV421 Rat Anti-Mouse CD70 (#562908) BD Biosciences, USA 1 in 100 (FC)

BV510 Rat Anti-Mouse CD11b

(#562950)


BV650 Rat Anti-Mouse CD11b

(#563402)


BV711 Rat Anti-Mouse CD117, c-Kit

(#563160)


FITC Hamster Anti-Mouse CD3ε

(#561827)


FITC Rat Anti-Mouse CD45R/B220

(#553087)


FITC Rat Anti-Mouse CD62L (#553150) BD Biosciences, USA 1 in 100 (FC)

FITC Rat Anti-Mouse Ly-6A/E, Sca-1

(#122505)


Goat Anti-Rabbit FITC conjugate Life Technologies,

USA

1 in 50 (IF)

Goat Anti-Rat DyLight 549 ThermoFisher

Scientific, Australia

1 in 200 (IF)

PE Hamster Anti-Mouse CD279 BD Biosciences, USA 1 in 100 (FC)


86

(#551892)

PE Rat Anti-Mouse CD265, RANK

(#119805)


PE Rat Anti-Mouse CD265, RANK (#12-

6612)

eBioscience, USA 1 in 500 (FC)

PE Rat Anti-Mouse CD279, PD-1

(#109103)


PE-Cy7 Hamster Anti-Mouse CD3ε (#25-

0031)


PE-Cy7 Hamster Anti-Mouse/Rat/Human

CD27 (#124215)


PE-Cy7 Rat Anti-Mouse CD25 (#25-

0251)


PE-Cy7 Rat AntiMouse/Human

CD45R/B220 (#25-0452)


PE-CyTM7 Hamster Anti-Mouse CD3ε

(#552774)


PE-DazzleTM594 Hamster Anti-Mouse

CD11c (#117347)


PerCP-CyTM5.5 Rat Anti-Mouse CD4

(#561115)


PerCP-CyTM5.5 Rat Anti-Mouse CD44

(#560570)


Rabbit Anti-Mouse RANKL Santa Cruz

Biotechnology, USA

1 in 500 (IF)

Rat Anti-Mouse CD4 Life Technologies,

USA

1 in 100 (IF)

V450 Rat Anti-Mouse CD11b (#560455) BD Biosciences, USA 1 in 100 (FC)

V450 Rat Anti-Mouse CD25 (#561257) BD Biosciences, USA 1 in 100 (FC)

4.1.7. Oligonucleotide primers

Oligonucleotide primer sequences for PCR and Real-Time PCR amplifications were

designed using Primer-BLAST (http://www.ncbi.nlm.nih.gov/tools/primer-blast/) which


87

was developed by National Center for Biotechnology Information (NCBI) based on

Primer3 (http://primer3.ut.ee/), and screened through NCBI-BLAST to ensure

specificity (Ye et al., 2012). The primers were purchased from Geneworks, Australia

and were shipped in freeze-dried form. The primers were then reconstituted in sterile

and nuclease-free ddH2O at a final concentration of 100 µM according to

manufacturer’s instructions. Aliquots of resuspended oligos were stored at -20 °C and

placed on ice during usage.

Primers Sequence (5' - 3') Amplicon size (bp)

Tumor necrosis factor receptor

superfamily, member 11a,

NFKB activator RANK

(NM_009399.3)

For: AGC TCA GCA TCC

CTT GCA G

Rev: GGA AGA GCT GCA

GAC CAC AT

178

Tumor necrosis factor (ligand)

superfamily, member 11

Tnfsf11, RANKL

(NM_011613.3)

For: TCT GTT CCT GTA

CTT TCG AGC GCA G

Rev: TGT TGC AGT TCC

TTC TGC ACG GC

198

Tartrate resistant acid

phosphatase 5

TRAP (NM_001102404.1)

For: CAG CAG CCA AGG

AGG ACT AC

Rev: ACA TAG CCC ACA

CCG TTC TC

190

SRY (sex determining region Y)-

box 9

Sox9 (NM_011448.4)

For: AAG CTC TGG AGG

CTG CTG AAC G

Rev: TTC GGC CTC CGC

TTG TCC GT

150

Secreted phosphoprotein,

osteopontin

SPP1, OPn (NM_001204201.1)

For: TCT GAT GAG ACC

GTC ACT GC

Rev: TCT CCT GGC TCT

CTT TGG AA

349

Sclerostin

Sost (NM_024449.5)

For: CAG ACC ATG AAC

CGG GCG GAG

Rev: CAC TGG CCG GAG

CAC ACC AAC

167


88

Rn18s

(NR_003278.3)

For: ACC ATA AAC GAT

GCC GAC T

Rev: TGT CAA TCC TGT

CCG TGT C

216

RING CCCH (C3H) domains 1

Rc3h1 (NM_001024952.2)

For: TAC CAT CGG CTC

CTA CAT TGG

Rev: GTC CAA CCT CTG

CTC TCT CG

115

Osterix, Sp7 transcription factor

Osx (NM_130458.3)

For: TTC CCT CAC TCA

TTT CCT GG

Rev: GTA GGG AGC TGG

GTT AAG GG

348

Osteoprotegerin

OPG (NM_011613.3)

For: GCC GTG CAG AAG

GAA CTG CAA C

Rev: GGT GTG CAA ATG

GCT GGG CCT

127

Osteocalcin

Bglap, OCn (NM_001037939.2)

For: GCG CTC TGT CTC

TCT GAC CT

Rev: ATA GAT GCG TTT

GTA GGC GG

440

Nuclear factor of activated T

cells, cytoplasmic,

calcineurin dependent 1 NFATc1

(NM_001164109.1)

For: CAA CGC CCT GAC

CAC CGA TAG

Rev: GGC TGC CTT CCG

TCT CAT AGT

392

Dentrocyte expressed seven

transmembrane protein

DC-Stamp (NM_001289506.1)

For: TGT GCT TGT GGA

GGA ACC TAA

Rev: GGC CAC AAA GCA

ACA GAC TC

163

Cathepsin K

Ctsk (NM_007802.4)

For: CCA GTG GGA GCT

ATG GAA GA

Rev: AAG TGG TTC ATG

GCC AGT TC

162

Calcitonin receptor For: CGG ACT TTG ACA 186


89

CTR (NM_001042725.1) CAG CAG AA

Rev: CAG CAA TCG ACA

AGG AGT GA

Alkaline phosphatase

Alpl, ALP (NM_001287172.1)

For: AAC TGC TGG CCC

TTG ACC CCT

Rev: TCC TGC CTC CTT

CCA CCA GCA

186

4.1.8. Solutions

Solutions and buffers were prepared using Milli-Q double distilled water (ddH2O). HCl

and NaOH were used to adjust the pH unless stated otherwise.

Solution Composition and preparation

Ammonium

persulphate

10% (w/v) ammonium persulphate dissolved in ddH2O, stored

at 4 °C.

Ascorbic acid 10% (w/v) stock dissolved in 1× PBS and filter-sterilised. A

1:200 working stock was used.

Bromophenol blue 1% (w/v) stock dissolved in ddH2O. Stirred before use to

resuspend any precipitates.

Collagenase type II

solution

2 mg/ml collagenase type II in serum free α-MEM, filter-

sterilised and warmed to 37 °C prior to use.

Decalcifying

solution (EDTA-G)

0.4 mM EDTA, 0.3 mM NaOH and 15% (v/v) glycerol

dissolved in ddH2O. Adjusted to pH 7.3 and stored at 4 °C.

DNA loading buffer 6× stock composition: 0.25% xylene cyanol, 0.25%

bromophenol blue, 30% glycerol.

Stored at 4 °C and dilute 1 in 6 with DNA samples prior to

loading.

dNTPs: dATP,

dCTP, dGTP and

dTTP

Equal volumes of 100 mM stock of each constituent was mixed

to yield a 25 mM combined stock of dNTP, stored at -20 °C.

EDTA 14% (w/v) EDTA, initially adjusted to pH 8.0 to dissolve

EDTA then readjust to pH 7.4.


90

Eosin-Phloxine

staining solution

Eosin stock solution: 1% (w/v) Eosin Y in ddH2O.

Phloxine stock solution: 1% (w/v) Phloxine B in ddH2O.

Working solution (per litre): 100 ml 1% eosin, 100 ml 1%

phloxine B, 780 ml, 95% ethanol and 4 ml glacial acetic acid.

Flow cytometry

wash buffer

1× PBS containing 1% FBS and 0.1% (w/v) sodium azide, store

at 4 °C.

Glutaraldehyde 25% stock diluted with ddH2O to yield 2.5% working solution.

Goldner's trichrome

stain solution A

0.075% (w/v) Ponceau 2R, 0.025% (w/v) acid fuchsin, 0.01%

(w/v) azophloxine and 0.2% (v/v) acetic acid dissolved in

ddH2O.

Goldner's trichrome

stain solution B

2% (w/v) Orange G and 4% (w/v) phosphomolybdic acid

dissolved in ddH2O. The solution was left to stand at room

temperature for a few days to allow the sedimentation of

precipitates and then filtered before use.

Goldner's trichrome

stain solution C

0.2% (w/v) Light green and 0.2% (v/v) acetic acid dissolved in

ddH2O.

Haematoxylin

(Gill's)

Stock solution: 250 ml ethylene glycol, 750 ml ddH2O, 6 g

haematoxylin, 0.6 sodium iodate (NaIO3), 80 g aluminium

sulphate (Al2(SO4)3) and 20 ml glacial acetic acid, stir and leave

to mix for 1 hour.

Dilute 1 in 5 and filter prior to use.

Infiltration reagent

for histology plastic

embedding

10 g plasticiser, 1 g Perkadox, 0.01 g Tinogard® TT

(Scavenger) dissolved in 89 g methyl methacrylate (MMA).

Neutral Buffered

Formaline

10% (v/v) formaldehyde solution, 45 mM Na2HPO4 and 33 mM

NaH2PO4 in ddH2O.

Paraformaldehyde 4% (w/v) Paraformaldehyde in ddH2O heated to 70 °C until

dissolved. Aliquots stored at -20 °C.

Phosphate buffered

saline (PBS)

10× PBS stock solution in ddH2O: 70 mM disodium hydrogen

phosphate (Na2HPO4), 30 mM monosodium phosphate

(NaH2PO4) and 1.3 M sodium chloride (NaCl).

1× PBS: 1 in 10 dilution of 10× PBS stock solution and

calibrated to pH 7.4.


91

PMSF Working stock was prepared by dissolving 17.6 mg/ml in

isopropanol. Aliquots stored at -20 °C.

Radio

Immunoprecipitatio

n Assay Buffer

(RIPA) lysis buffer

50 mM Tris pH 7.5, 150 mM NaCl, 1% Nonidet P-40, 0.1%

SDS and 1% sodium deoxycholate. Adjusted to pH 7.5 and

stored at 4 °C.

Prior to protein isolation from cells or tissue samples, 100

µg/ml PMSF, 1× protease inhibitor, 1 mM sodium

orthovanadate and 500 µg/mL DNase I were promptly added to

the amount of lysis buffer required.

SDS-PAGE

running buffer

10× stock solution in ddH2O: 25 mM Trizma base, 1.92 M

glycine and 1% (v/v) SDS.

1× working solution: 1 in 10 dilution of 10× stock solution in

ddH2O.

SDS-PAGE sample

loading buffer

4× stock solution dissolved in ddH2O: 240 mM Tris-HCl (pH

6.8), 8% (w/v) SDS, 40% (v/v) glycerol, 0.04% (w/v)

bromophenol blue and 5% (v/v) β-mercaptoethanol.

Stored at 4 °C.

SDS-PAGE

separating gel

buffer

1.5 M Tris-HCl (pH 8.8) and 0.4% (w/v) SDS prepared in

ddH2O.

SDS-PAGE

stacking gel buffer

1 M Tris-HCl (pH6.8), 0.4% SDS (w/v) dissolved in ddH2O.

Sodium

deoxycholate

solution (SDS)

10% (w/v) solution dissolved in ddH2O.

Sodium

orthovanadate

(Na3VO4)

0.25 M of Na3VO4·14H2O dissolved in ddH2O and is pH to 10

with HCl, yielding a yellow solution. Boil solution until it turned

colourless and cool on ice until the solution reaches room

temperature. Readjust to pH 10 and repeat boiling/cooling steps a

total of 3-5 times. Stock is stored at -20 °C.

Tris-acetate EDTA

(TAE)

1 L 50× stock solution in ddH2O: 2 M Trizma base, 5.71% (v/v)

glacial acetic acid and 50 mM EDTA.

1× TAE working solution: dilute 1 in 50 in ddH2O prior to use.


92

Tris buffered saline

(TBS)

10× stock solution in ddH2O: 0.5 M Trizma base and 1.5 M

NaCl.

1× TBS: 1 in 10 dilution of stock solution in ddH2O.

TBS-Tween 1× TBS with addition of 0.1% (v/v) Tween-20.

Toluidine blue dye

solution

1% (w/v) toluidine blue dissolved in in ddH2O.

TRAP stain

(complete)

5 mg Naphthol AS-MX dissolved in 250 μl 2-ethoxyethanol with

the addition of 40 mg Fast Red-Violet LB salt dissolved in TRAP

stain solution A.

Aliquots are stored at -20 °C, thawed at 37 °C and filtered before

use.

TRAP stain

solution A

100 mM sodium acetate trihydrate, 50 mM sodium tartrate

dihydrate and 0.22% (v/v) glacial acetic acid. Adjusted to pH 5.0.

Triton X-100/PBS 0.1% (v/v) 100% Triton X-100 in 1× PBS. Filtered and stored at

4 °C.

Weigert's iron

haematoxylin

Solution A: 1% (w/v) haematoxylin and 96% (v/v) ethanol in

ddH2O.

Solution B: 4 ml 30% iron chloride and 1 ml 37% HCl mixed in

ddH2O.

Equal parts of solution A and solution B were mixed to make a

working solution.

Western blot

membrane

stripping buffer

62.5 mM Tris-HCl and 2% (w/v) SDS dissolved in ddH2O and

calibrated to pH 6.7 prior to addition of 100 mM 2-

mercaptoethanol.

Western blot

transfer buffer

25 mM Trizma base, 192 mM glycine and 10% (v/v) methanol in

ddH2O.

β-

glycerophosphate

0.4 M stock dissolved in 1× PBS, filter-sterilised and aliquots

stored at -20 °C.

4.1.9. Equipment


Wide Mini-Sub® Cell GT Bio-Rad, USA


93

Verity 96-well Thermal Cycler Applied Biosystems, USA

Tissue-Tek III Embedding Console Miles Scientific, USA

Tissue Processor – TP1020 Leica Microsystems, Germany

Tissue Cryostat Processor – CM1900 Leica Microsystems, Germany

Thermomixer Comfort Eppendorf AG, Germany

Skyscan 1174 Compact MicroCT Scanner Skyscan, Belgium

Shimadzu Balance ELB300 Walker Scientific, Australia

Senographe DS Digital Mammography Machine GE Healthcare, UK

Rotofix Hettich Zentrifugen HD Scientific Supplies Pty Ltd,

Australia

Rotary Tube Mixer with disc Ratek Instruments, Australia

Removable (UVTP) Gel Tray and Gel Caster Bio-Rad, USA

Ratek Platform Mixer Ratek Instruments, Australia

Power-Pac 300 Bio-Rad, USA

POLARstar OPTIMA Bio-Rad, USA

Olympus CK30 Micrscope Olympus Optical Co Ltd, Japan

Nikon CoolPix S4 Digital Camera Nikon Corp., Japan

Neubauer Improved Cell Counting Chamber Laboroptic Ltd, UK

Nanodrop® ND-1000 UV-Vis

Spectrophotometer V3.3

Thermo Scientific, USA

Movable 7×10 cm2 and 10×10 cm2 UV Perspex

Gel Casting Tray;

8 and 15 well comb for agarose gels

Bio-Rad, USA

Model 680 Microplate Reader Bio-Rad, USA

MiniSpin® and MiniSpin® Plus microcentrifuge Eppendorf AG, Germany

Mini-sub® Cell GT Bio-Rad, USA

Mini Trans-Blot Electrophoretic Transfer Cell Bio-Rad, USA

MilliQ Reagent Water System Millipore Corporation, USA

Microcentrifuge Tubes: 1.5 ml Sarstedt, Germany

Microcentrifuge Tubes: 0.5 ml & 2.0 ml TRACE Biosciences, Australia

Micro One TOMY Centrifuge Quantum Scientific, Australia

Life Technologies ViiA™ 7 Real-Time PCR

System

Life Technologies, USA


94

Letiz Labourlux S Microscope Leitz Wetzlar, Germany

ImageQuantTM LAS-4000 Mini Imager GE Healthcare Life Sciences, USA

isoMet® Low Speed Saw Buehler, USA

Hot air oven Memmert, Germany

Epson Perfection 3490 Photo Scanner Seiko Epson Corporation, Japan

Eppendorf MasterCycle® Personal Thermal

Cycler

Brinkmann, Canada

Eclipse TE2000S Fluorescent Microscope Nikon Instruments Inc., USA

Clemco FCS Fume Cupboard Clemco Industristies Corp., USA

Chiltern Vortex Mixer MT17 Scot Scientific, Australia

Centrifuge 5415D Eppendorf AG, Germany

Biophotometer Plus Eppendorf AG, Germany

Biological Safety Cabinet Class II Gelman Sciences, USA

Automatic CO2 incubator Forma Scientific, USA

4.1.9.1. Centrifugation

Micro-centrifugations (0.5 mL, 1.5 mL and 2.0 mL tubes) at room temperature were

carried out in MiniSpin or MiniSpin Plus microcentrifuge. Micro-centrifugations at 4 ˚C

were carried out using the Eppendorf Centrifuge 5415D. PCR (0.2 mL thin-walled

tubes) centrifugations at room temperature were performed using MicroOne

microcentrifuge. Centrifugations of 15 ml and 50 ml centrifuge tubes were performed

using Hettich Rotofix 32 centrifuge.

4.1.10. Software

Description Supplier

BD FACSDiva™ BD Biosciences, USA

Bioquant Osteo 14.1.6 Bioquant Image Analysis Corp,

USA

CT Analyser 1.10.1.0 Skyscan, Belgium

Dataviewer 1.4.3.0 Skyscan, Belgium

FlowJo FlowJo LLC, USA

ImageJ 1.48 National Institutes of Health, USA


95

Microsoft Research Image Composite Editor

(ICE)

Microsoft, USA

NIS-Elements Basic Research Nikon Instruments Inc, USA

NIS-Elements C Nikon Instruments Inc, USA

NRecon Version 1.5.1.14 Skyscan, Belgium

4.2. Animals and bone phenotyping methods

4.2.1. The generation of sanroque mutant mouse line

The sanroque mutant mice on which this study is based on, were generated by the

Australian Phenomics Facility (APF) of the Australian National University (ANU),

Canberra, Australia. The mutant mice were generated using N-ethyl-N-nitrosourea

(ENU)-induced mutagenesis, which resulted in single-base substitutions at an estimated

rate of 1 per 0.5 megabases (Mb) in the germline (Jun et al., 2003; Nelms et al., 2001;

Vinuesa et al., 2005a). Briefly, normal C57BL/6 male mice 8–15 weeks old (generation

0; G0 mice) were treated with super-mutagenic ENU (100 mg/kg) three times 1 week

apart and were mated with wildtype C57BL/6 female mice. Individual first generation

male progeny (G1) were used as founders of inbreeding pedigrees by breeding with

wildtype C57BL/6 females and then with several G2 daughters. Phenotypic-screening is

performed on G3 offspring with a subset of mutations that would reach homozygosity.

The mapping and sequencing of sanroque mice were also performed by APF. For

mapping of recessive mutations an intercross breeding approach was employed whereby

the mutant C57BL/6.sanroque mice were outcrossed to a mapping strain (CBA/Ca)

mice to produce F1 carrier mice. The F1 mice were intercrossed and 25% of the

generated F2 progeny will be homozygous for the recessive mutation. Using MIT

microsatellite panel, the mutation was mapped to a single locus on distal chromosome 1

in F2 mice. The chromosomal location of sanroque is localised at 1-Mb chromosomal

interval containing 11 predicted genes. Resequencing of mRNA and exons in genomic

DNA revealed that, of the 11 annotated transcription units in the sanroque interval, only

one mRNA contained the mutation in sanroque mice relative to the B6 parental

controls. The single T G nucleotide substitution resulting in a non-conservative Met

199 Arg codon change was identified in the mouse homolog 1 Roquin (Rc3h1) gene.

F6 to F8 wildtype and sanroque mutant littermates of the age of 80 to 100 days old were


96

used in this study. When required, the mice were imported from APF and quarantined

for one week at the Biomedical Research Facility, UWA, Western Australia. Once

cleared from quarantine, the mice were shipped to the School of Pathology and

Laboratory Medicine, UWA and animal handling procedures were performed in

accordance with protocols approved by the UWA Animal Ethics and the ANU Animal

Ethics committees.

4.2.2. Radiographic screening of sanroque mutant mice

Radiographic assessment of forelimbs, hindlimbs, spines and tails of wildtype and

sanroque mutant mice was carried out using the Senographe DS digital mammography

system at the Sir Charles Gairdner Hospital, Perth. The samples were imaged using the

following settings:

Sample Voltage (kV) Current (mA)

Forelimbs and hindlimbs 22 12.5

Spines 25 14.0

Tails 25 10.0

4.2.3. Microquantitative computed X-ray tomography (MicroCT) analysis

of wildtype and sanroque mutant mice

The mouse distal femur or proximal tibia was fixed in neutral buffered formalin for 24

hours, washed in PBS, and then stored in 70% ethanol. The fixed tissue was wrapped in

moistened tissue and placed in an X-ray transparent tube to be scanned using a Skyscan

1174 microCT scanner (Skyscan, Belgium) at a voltage of 50 kV and a current of 0.8

mA with a 0.5 mm aluminium filter. The resolution (pixel size) was set to 6.03 µm and

scanning angular rotation was 180˚ at the angular increment of 0.4˚. Datasets of

tomographic sections were reconstructed using the cone beam CT reconstruction

software (NRecon) facilitated by the Feldkamp algorithm and segmented into binary

images using adaptive local thresholding using CT analyser software (CTAn, Skyscan).

Bone volume analysis of the trabecular bone was performed using analysis software

(CTAn, Skyscan). The secondary spongiosa 0.5 mm distal to the proximal growth plate

at a height of 1 mm was selected for analysis within a conforming volume of interest

(excluding cortical bone). Morphometric parameters including trabecular bone volume


97

(BV/TV), trabecular thickness (Tb.Th), trabecular number (Tb.n) and trabecular

separation (Tb.sp) were measured. Assessment of the mid-diaphysis cortical volume

was performed by similar methods in a region 3.3 mm from the growth plate, with a

height of 1 mm. The cortical bone volume (Cortical BV) and cortical thickness were

calculated.

MicroCT analysis was performed in the same settings for each biological pair of all age

group and sexes of mice. For 3D images, the CT files were saved as .p3g 3D models in

CTan (Skyscan) and were generated using CTVol program (Skyscan).

4.2.4. Fluorochrome labelling of wildtype and sanroque mutant mice

Fluorochrome labelling using calcein was performed to assess in vivo bone formation

rates. Three biological pairs of wildtype and sanroque mutant mice received an

intraperitoneal injection of calcein (5 mg/kg) (day 0), followed by a second

intraperitoneal injection of alizarin red (30 mg/kg) 4 days later, and another calcein

injection on day 8. The mice were sacrificed on day 10 and the hind limbs were

dissected out of the mice. The limbs were fixed in 70% ethanol and were exported to

our collaborator, Dr. Jerry Feng, in China to be processed for plastic embedding.

Briefly, the samples were dehydrated at room temperature under vacuum through 3

changes of 95% ethanol for 4 hours each step. This is followed by 4 changes of 100%

ethanol, 4 hours per change. The samples were then immersed in freshly prepared

embedding solution of 84% (v/v) methalmethacrylate, 14% (v/v) dibutylphthalate, 1%

(v/v) polyethylene glycol 400 and 0.7% (w/v) benzoyl peroxide, and vacuumed. The

infiltration was carried out at room temperature for 3 – 4 days. The infiltrated samples

was subsequently incubated at 37 °C for 2 – 3 days until the plastic resin polymerised.

The samples were then sectioned at 5 µm thickness. The calcein labelling was

visualised by confocal microscopy. The distance between the labels at 3 points along

the cortical bone were measured using ImageJ (NIH) software.

4.2.5. Histological analysis of wildtype and sanroque mutant mice

4.2.5.1. Tissue fixation, decalcification (for hindlimbs), processing, embedding

and sectioning

Tissues or organs of interest were dissected out and fixed in 10% NBF on a Rotary

Tube Mixer (Ratek) overnight. The samples were then washed twice with 1× PBS and


98

left on the Rotary Tube Mixer overnight to completely wash the fixative solution. The

samples were then transferred into 70% ethanol and stored at 4 °C until ready for

processing (or decalcification of bone, if required).

For decalcification of bone tissues, all soft tissues were thoroughly removed, and the

tibia and femur were separated gently. The bones were washed once with 1× PBS

before the decalcification process. The decalcification process were carried out by

immersing the tissues in 14% EDTA (pH 7.4) and incubated at 37 °C. Fresh EDTA was

changed daily for 5 days. Once the bone tissues became soft and flexible, the tissues

were washed with 1× PBS and were stored in 70% ethanol until they are ready to be

processed. The decalcified bones were placed in sample cassettes and processed using

the Leica TP1020 tissue processor over 24 hours according to the following steps:

Station Reagent Time (minutes)

1 70% Ethanol 30

2 95% Ethanol 30

3 95% Ethanol 45

4 100% Ethanol 60

5 100% Ethanol 60

6 100% Ethanol 60

7 50% Ethanol/50% Xylene 60

8 Xylene 90

9 Xylene 120

10 Paraffin wax at 60 °C 90



The processed samples were then paraffin-embedded and cut into 5 µm sections by Mr.

Jacob Kenny from The School of Pathology and Laboratory Medicine, UWA. In brief,

the samples were placed in Tissue-Tek thermal console (Miles Scientific) to be paraffin-

embedded and were cooled using Tissue-Tek cryo console (Miles Scientific) for 15 to

20 minutes. Once set, the embedded samples were sectioned using Leica 2035

microtome and the 5 µm sections were floated on a 37 °C water bath and lifted onto a

clean glass slides. The sections were set onto the glass slides by incubating the samples


99

in a hot air oven at 60 °C for at least one hour.

4.2.5.2. Haematoxylin and eosin staining

Once the sections were set, the slides were dewaxed, hydrated, stained and dehydrated

by immersing the samples into the following series of solutions:

Step Solution Time (minutes)

1 Xylene 5

2 Xylene 5

3 100% Ethanol 3

4 100% Ethanol 3

5 95% Ethanol 3

6 80% Ethanol 3

7 70% Ethanol 3

8 Gently running tap water 3

9 Gill's Haematoxylin 5

10 Gently running tap water 3

11 1% Eosin 0.5

12 Scott's water 1

13 70% Ethanol 1

14 80% Ethanol 1

15 95% Ethanol 1

16 100% Ethanol 2

17 100% Ethanol 2

18 Xylene 5

19 Xylene 5

20 Xylene 5

Haematoxylin and eosin stain nucleic acid in blue and protein in shades of pink

respectively. The stained slides were mounted on coverslips with DPX mounting

medium.

4.2.5.3. TRAP staining

TRAP solutions were warmed up to 37 °C and filtered prior to staining. The slides were

dewaxed and hydrated as listed above (steps 1 – 8). Excess liquid was removed from the


100

slide without letting the section dry out. Filtered pre-warmed TRAP solution was

applied and the slides were incubated at 37 °C. The slides were checked every 30

minutes to monitor the development of the stain. After approximately 2 hours, with

observation of magenta-stained osteoclasts, the slides were rinsed in ddH2O before

counterstaining with a quick immersion in Gill’s Haematoxylin solution. The slides

were rinsed in ddH2O to remove excess stain, followed by a 1 minute immersion in

Scott’s tap water. If the counterstain appeared too intense, a 3 second dip into 1% acid

alcohol was performed to wash off the stain, and further adjusted using Scott’s tap water

until the desired contrasting colour was achieved. The sections were washed briefly in

running tap water and were mounted with Clearmount aqueous mounting medium

(Invitrogen). The mounted slides were left to set in a hot air oven at 60 °C for 45

minutes.

4.2.5.4. Goldner Trichome staining on non-decalcified bone samples

4.2.5.4.1. Processing, embedding and sectioning of undecalcified bone

samples

Tibia and femurs were collected and fixed as previously described in section 4.2.5.1.

For plastic embedding, the decalcification step was omitted and the fixed samples

underwent a series of dehydration process in the Leica TP1020 tissue processor

according to the following steps:

Station Reagent Time (minutes)

1 50% Ethanol 120 120

2 70% Ethanol 120

3 80% Ethanol 120

4 96% Ethanol 120

5 100% Ethanol 180

6 100% Ethanol 180

7 Xylene 60

8 Xylene 720

Excess xylene was gently wiped off and the samples were immersed in the infiltration

reagent (described in section 4.1.8) for 3 – 4 days in vacuum condition at 4 °C. The

samples were then transferred into moulds and were embedded in fresh infiltration


101

reagent. The moulds were sealed tightly and polymerisation was catalysed by

incubating the samples at 29 °C for at least 3 days until the samples were fully

polymerised. The embedded samples were sectioned using the Leica RM2255

Automated Microtome at a thickness of 5 µm. The sections were placed directly on the

glass slides and were gently straightened using a soft artist painting brush dipped in

30% ethanol. A layer of plastic sheet was overlaid onto the straightened sections and the

glass slides were stacked onto one another. To facilitate section adherence onto the

glass, the stacked glass slides were clamped and left to set at 60 °C overnight. The

processing, embedding and sectioning of undecalcified bone samples were performed

with the assistance of Miss Euphemie Landao (Centre for Orthopaedic Research, School

of Surgery, UWA).

4.2.5.4.2. Goldner Trichrome staining

The samples were de-plasticised, rehydrated, stained and dehydrated according to the

following protocol:

Step Solution Time

1 Methoxyethyl acetate (MEA) 20 minutes

2 MEA 20 minutes

3 MEA 20 minutes

4 Xylene 10 minutes

5 Xylene 10 minutes

6 100% Ethanol 10 seconds





11 50% Ethanol 10 minutes

12 ddH2O 15 seconds

13 ddH2O 15 seconds

14 Acid alcohol 3 seconds

15 ddH2O 15 seconds

16 Solution A 5 minutes

17 1% Aqueous acetic acid 15 seconds


102

18 Solution B 3 minutes

19 1% Aqueous acetic acid 15 seconds

20 Solution C 5 minutes

21 1% Aqueous acetic acid 5 minutes


Goldner Trichrome stained the mineralised bone green, osteoid (unmineralised bone)

orange/red, nuclei bluish-grey and cartilage purple.

4.2.5.5. Bone histomorphometric analysis

Stained tibia and femur sections were scanned at 40× magnification using ScanScope®

XT System (Quorum Technologies) at the Centre for Microscopy, Characterisation and

Analysis (CMCA), UWA. Bone histomorphometric analysis on mouse tibia or femur

sections was conducted using BioQuant Osteo software (Bioquant Image Analysis).

Trabecular bone (secondary spongiosa) was analysed within the region of interest of 1

mm in height, set to 0.5 mm below the growth plate. Three independent sections were

analysed per bone sample to yield the following parameters:

Parameter Stained sections

Trabecular number (BV/TV) H&E or Goldner Trichrome

Trabecular thickness (Tb.Th) H&E or Goldner Trichrome

Trabecular number (Tb.N) H&E or Goldner Trichrome

Trabecular separation (Tb.Sp) H&E or Goldner Trichrome

Osteoid surface (OS) Goldner Trichrome

Osteoid surface per bone surface (OS/BS) Goldner Trichrome

Number of osteoclast per bone surface (N.Oc/BS) Goldner Trichrome

Number of osteoblast (N.Ob) Goldner Trichrome

Number of osteoblast per bone surface (N.Ob/BS) Goldner Trichrome

Osteoclast surface (Oc.S) TRAP

Osteoclast surface per bone surface (Oc.S/BS) TRAP

4.2.6. TRAP staining of mouse calvariae

Soft tissues surrounding the calvariae dissected from euthanized wildtype and sanroque

mutant mice were thoroughly cleaned using a scalpel. The calvariae were then


103

extensively washed in 1× PBS and fixed as dictated in section 4.2.5.1. Prior to TRAP

staining, the calvariae were decalcified until they became soft and malleable.

Decalcified calvariae were then extensively washed with 1× PBS and briefly dried on a

paper towel before placing individual calvaria in each well of a 12-well plate. The

samples were then incubated in filtered pre-warmed TRAP stain at 37 °C for

approximately 1 hour. Once apparent magenta-staining was observed, the TRAP

solution was removed and the calvariae were once again washed extensively with 1×

PBS and stored at 4 °C. The superficial surface of the calvaria was photographed using

a Nikon Coolpix S9600 digital compact camera.

4.2.7. Immunofluorescence staining of mouse spleens

4.2.7.1. Embedding of sample in Optimal Cutting Temperature (OCT)

Isolated spleen was gently blotted dry on tissue to remove excess blood and was left to

immerse and equilibrate in OCT (Tissue-Tek) media in a mould for 5 – 10 minutes.

During tissue equilibration isopentane was pre-chilled in a liquid nitrogen bath. The

mould containing the equilibrated OCT-immersed spleen was floated on the surface of

the isopentane until the OCT solidified. Frozen blocks containing samples were stored

at -70 °C until they were ready to cut. Prior to cutting, the blocks were placed into

CM1900 Tissue Cryostat Processor (Leica) to equilibrate for at least 30 minutes. The

blocks were cut into 5 µm sections and allowed to adhere onto charged slides. The

slides were left to air-dry at room temperature for at least 30 minutes prior to fixing. If

the slides were not required to be fixed immediately, the samples were stored at -70 °C

with desiccants to remove humidity and when required for staining were subsequently

air-dried at room temperature for 15 minutes prior to fixing. The slides were fixed in

acetone for 15 minutes at - 20 °C. Excess acetone was gently removed and the slides

were left to air-dry for another 15 minutes. The fixed samples were stained

immediately.

4.2.7.2. Immunofluorescence staining

The tissue section on each slide was isolated by drawing the perimeter with a PAP pen

and the slides were rehydrated in 1× PBS with the addition of 1% serum (1× PBS/1%

serum solution) for 10 minutes at room temperature. The serum used for blocking was

from the same species in which the secondary antibody was raised. Samples were then

blocked with 10% serum in 1× PBS for 30 minutes at room temperature. After blocking,


104

the slides were gently rinsed with 1× PBS and incubated for 60 minutes with primary

antibodies diluted in 1× PBS/1% serum solution. The slides were washed with three

changes of 1× PBS for 3 minutes each wash followed by 45 minutes incubation with

secondary antibodies diluted in 1× PBS/1% serum solution. Subsequently, the

antibodies were washed away with three changes of 1× PBS. For nuclear staining, the

samples were incubated with DAPI diluted in 1× PBS for 15 minutes. The samples were

once again washed three times with 1× PBS and excess liquid was thoroughly removed.

Prolong Gold Antifade (Invitrogen) was used to mount the samples with coverslip and

the samples were protected from light until ready to be viewed under fluorescence

microscope.

4.3. Isolation, culture and in vitro studies of primary cell lines

All cells were cultured in plates or flasks as indicated in a humidified water-jacketed

incubator with 5% CO2, 95% air at 37 °C. Cultured cells were handled using aseptic

techniques in UV sterilised biological safety cabinets (Class II).

4.3.1. Isolation and culture of bone marrow cells

Bone marrow cells were isolated from long bones (tibia, femur and humerus) of

euthanized wildtype and sanroque mutant mice. The long bones were dissected out

from the mice and placed in ice-cold complete α-MEM (α-MEM supplemented with

10% FBS, 2 mM L-glutamine, 100 U/ml Penicillin and 100 µg/mL Streptomycin).

Muscle was removed using a sterile scalpel and the epiphyses were cut off to expose the

marrow cavity. Bone marrow cells were flushed from the bone marrow cavity

repeatedly with complete α-MEM (pre-warmed at 37 °C) using a syringe and 23G

needle in a sterile petri dish. Flushed long bones were set-aside in sterile 1× PBS for

osteoblastic bone cell isolation where required (see 4.3.2). The bone marrow cells were

collected and filtered through a 100 µm sterile cell strainer to remove loose bone and

soft tissue fragments. The bone marrow cells were pelleted by centrifugation at 350 g

for 5 minutes at room temperature and resuspended in complete α-MEM.

4.3.1.1. Generation and maintenance of bone marrow macrophages (BMMs)

Bone marrow macrophages (BMMs) were generated by culturing isolated bone marrow

cells in 10 ml of complete α-MEM supplemented with 1 in 20 dilution of murine M-

CSF conditioned media (α-MEM/M-CSF) in 75-cm2 tissue culture flasks. When cells

were 90% confluent, they were washed with 1× PBS and trypsinised using TrypLETM


105

Express for 5 – 10 minutes at 37 °C, occasionally tapping the flask. BMMs adhere

strongly to the plastic, hence scraping was required to detach the cells. Complete α-

MEM was added to deactivate the enzyme and the cell suspension was pelleted by

centrifugation at 350 g for 5 minutes. The cell pellet was then resuspended in α-

MEM/M-CSF and a cell count was performed using a Neubauer counting chamber.

BMMs were then plated into culture dishes at the appropriate cell density for

downstream experiments.

4.3.1.2. MTS cell proliferation assay

Cell proliferation assay on BMMs using CellTiter 96® AQueous One Solution Cell

Proliferation Assay kit (Promega) containing tetrazolium compound [3-(4,5-

dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium,

inner salt; MTS] was performed according to manufacturer’s instructions. Briefly,

6×103 BMMs were seeded in two 96 well plates, one to be measured at basal level and

the other to be measured after treatment. Once the cells adhered and after 24 hours of

treatment, 20 μl of thawed CellTiter 96® AQueous One Solution Reagent is added into

each well of the 96-well plate containing the cell culture in 100 μl medium. Repeat

pipetting was performed to ensure that the solution is well mixed and the plate was

incubated at 37 °C for 1–4 hours in a humidified, 5% CO2 atmosphere. The reaction

was stopped by the addition of 25µl of 10% SDS into each well. The plate was

protected from light at room temperature before recording the absorbance at 490 nm

using a 96 well plate reader.

4.3.1.3. Osteoclast culture

Cell culture plate

Cell density

6-well plate 5×104/2.0 ml/well



For osteoclastogenesis, seeded BMMs were cultured in α-MEM/M-CSF and left to

settle overnight to allow cell attachment onto the plate. The following day, BMMs were

stimulated with 25 ng/ml RANKL-GST and the culture media (α-MEM/M-

CSF/RANKL) was replaced every two days until multinucleated osteoclast-like cells


106

are observed at day 5 – 7 post-stimulation with RANKL. Unstimulated cells served as a

negative control.

4.3.1.3.1. In vitro osteoclastogenesis assay

Osteoclastogenesis assays were carried out using BMMs cultured as described above in

96-well plates. Once stimulated to form multinucleated osteoclasts, the cells were gently

washed once with 1× PBS and fixed with 4% paraformaldehyde for 15 minutes at room

temperature, followed by 3 gentle washes with 1× PBS. The cells were subsequently

stained with filtered TRAP stain and left to incubate at 37 °C for 30 minutes. The TRAP

stained cells were washed with 1× PBS three times and the plates were stored at 4 °C

until they are ready to be imaged and counted.

The osteoclast cultures were visualised using Eclipse TE2000S (Nikon) microscope and

NIS-Elements Basic Research Nikon software. Serial images covering the entire well

were taken at 40X magnification and then stitched together using Image Composite

Editor (ICE) (Microsoft) to form a single image of each well. The total number of

TRAP stained, magenta-coloured osteoclasts with more than three nuclei were tallied

for each well.

4.3.1.3.2. In vitro bone resorption assay

Bone resorption assays were performed using multinucleated osteoclasts seeded onto

bovine bone discs. Bone discs were produced from bovine bone that was cleaned

thoroughly to remove excess tissue and marrow. The bone was sliced using an IsoMet®

low speed saw (Buehler) into 0.70 mm sections. Bone slices were soaked in ddH2O to

soften the bone matrix and then circular 5 mm bone discs were generated with using a 5

mm diameter hole punch. Bone discs were cleaned by sonicating for 5 minutes to a total

of 10 times in MilliQ ddH2O, replacing the water each time. Bone powder generated

from the action of the low speed saw was collected and reserved for cathepsin K release

experiments (4.3.1.2). After sonication, the bone discs were stored in 70% ethanol at 4

°C. When the bone discs were required, they were thoroughly air dried in a sterile

biological safety cabinet (Class II). The dried and sterile bone discs were washed three

times with serum free α-MEM and then incubated for at least 2 hours in α-MEM/M-

CSF at 37 °C. Before placing the bone discs into the 96-well plate for cell seeding, the

bone discs were washed a further three times with α-MEM/M-CSF.


107

To generate mature multinucleated osteoclasts for bone resorption assays, BMMs were

cultured in 6-well collagen-coated plates and stimulated as described previously. Instead

of maintaining the culture until large osteoclasts were formed, cells were dissociated at

approximately day 3 when small osteoclasts began to form. The cells were incubated

with cell dissociation solution (Sigma) for 10 minutes and observed every 5 minutes

under the microscope. Once the cells began to lift off, remaining adherent cells were

gently scraped off and the cell suspension pelleted by centrifugation at 350 g for 5

minutes. The cells were then reconstituted with 1.5 ml of α-MEM/M-CSF/RANKL

culture media and 100 µl aliquots were seeded onto individual bone discs. Aliquots of

cells were seeded in wells with no bone discs to allow visual monitoring of the

differentiation of the cells.

Osteoclasts were left to settle and resorb the bone discs for 72 hours. Cells on bone

discs were gently washed, fixed, stained and imaged as described above. Osteoclast

numbers were counted on each bone discs, and then osteoclasts were gently removed

using cotton buds (soaked in 1% (v/v) bleach) to expose the underlying resorbed bone.

The cleaned bone discs were then stained with toluidine blue for 5 minutes to stain the

resorption pits dark blue, while the non-resorbed bone surface remained pale blue. The

discs were then washed extensively with 1× PBS and serial images of the whole bone

discs were taken and stitched as previously described. The area of resorption was

measured using ImageJ (NIH).

4.3.1.3.3. In vitro osteoclast cathepsin K release assay using bovine bone

powder

Bovine bone powder was collected as described above. The bone powder was collected

in distilled water as a by-product of cutting multiple bone slices. The powder was

pooled and centrifuged at maximum speed for 5 minutes at room temperature. The

supernatant was discarded and the bone powder was reconstituted in 70% ethanol. The

suspension was vigorously vortexed to ensure thorough sterilisation and the bone

powder was again pelleted by centrifugation. The pelleted bone powder was then

collected and spread in a sterile petri dish and left to dry in a sterile class II biological

safety cabinet. The sterile bone powder was collected and placed in a 50 ml centrifuge

tube to be reconstituted and extensively washed with sterile 1× PBS. The suspension

was vigorously mixed to break up clumps of powder and then centrifuged. The


108

supernatant was discarded and the pelleted bone powder was then washed again with

vortexing in 1× PBS and the wash process was repeated 3 times. The bone powder was

reconstituted in sterile α-MEM and left to agitate on a platform shaker for at least 2

hours to ensure that all bone powder clumps were completely dispersed. The bone

powder suspension was subsequently centrifuged and washed a further three times with

sterile α-MEM. The bone powder was subsequently resuspended 50% v/v with sterile α-

MEM and stored at 4 °C until required. When required, the bone powder suspension

was mixed thoroughly, the appropriate amount aliquoted, centrifuged and resuspended

50% v/v in α-MEM/M-CSF/RANKL culture media.

BMMs were seeded in a 6-well plate at the cell density cited above and were maintained

in α-MEM/M-CSF/RANKL culture media. Once large multinucleated osteoclasts were

formed, 100 µl of 50% v/v bone powder suspension in α-MEM/M-CSF/RANKL culture

media was gently and evenly distributed drop-wise into each well to stimulate cathepsin

K enzymatic release. The cells were stimulated at 4, 8 and 12 hours and cultured media

was collected followed by cell protein extraction as described in 4.4.4.2. The cultured

media and cell proteins were separated by SDS-PAGE gel electrophoresis and analysed

by western blotting using the appropriate antibody as explained in sections 4.4.4.3 –

4.4.4.6.

4.3.2. Isolation of osteoblastic bone cells from long bones and calvariae

Primary osteoblast bone cell isolation procedure from long bones and calvariae was

modified from (Bakker et al., 2012). Flushed long bones were collected and pooled

together as described above. To isolate primary osteoblastic bone cells from adult

mouse calvaria, calvaria from euthanized mice were dissected from the skull using

scissors. Sutures and other tissues were removed by gentle scraping the calvaria. Both

flushed long bones and cleaned calvariae undergo similar processing hereafter. Briefly,

samples were washed with 1× PBS and chopped into small fragments (~1 mm2). The

bone fragments were then incubated in collagenase II solution at 37 °C with gentle

shaking for 2 hours and occasional vigorous shaking by hand. An equal amount of

DMEM containing 15% FCS, 2 mM L-glutamine, 100 U/ml Penicillin and 100 µg/mL

Streptomycin (complete DMEM) was added to inhibit further collagenase digestion, and

the bone fragments were rinse three times with complete DMEM. The bone fragments

were then transferred into 25-cm2 tissue culture flasks containing 5 ml complete DMEM


109

at a density of about 20 – 30 fragments per flask. Adult mouse osteoblastic bone cells

were observed migrating from the bone fragments after 3 – 5 days. The culture medium

was replaced and the flasks were gently agitated every 3 days to prevent crowding

around the bone fragments and encourage cell migration. Bone fragments were

maintained for approximately 10 – 14 days until migrating cells reached subconfluency,

upon which cells were trypsinised and used for experiments. Bone fragments were

discarded.

4.3.2.1. Osteoblast culture and mineralisation assay

Cell culture plate Cell density

12-well plate 2.0 ×105/1.0 ml/well

24-well plate 1.0 ×105/0.5 ml/well

48-well plate 0.5×105/0.5 ml/well

For osteoblastogenesis assays, trypsinised primary osteoblastic cells were seeded into

12- well plates and the cells were maintained in complete DMEM until they are

confluent (approximately 7 – 10 days post cell-seeding). Once confluent, the cells were

stimulated with mineralisation media consisting of complete DMEM supplemented with

10 nM dexamethasone, 2 mM β-glycerophosphate and 50 µg/ml ascorbate and medium

was replaced every 3 days. Bone nodules were observed at day 10 – 14. At day 21the

cells were washed once with 1× PBS and fixed with 2.5% glutaraldehyde solution for

10 minutes followed by three washes with 1× PBS. For staining of the calcium within

mineralised nodules using alizarin red S, the fixed cells were post-fixed with 70%

ethanol three times and left to dry completely. An aliquot of 300 µl of 1% alizarin red

solution was placed into each well and left to stain for 2 minutes before destaining with

three washes of 50% ethanol. The plates were left to air dry completely and stored at

room temperature.

To analyse the area of mineralised nodules, the plates were scanned at 600 dpi using an

Epson Perfection 3490 Photo scanner in TIFF format. ImageJ software was used to

measure the mineralised area.

4.3.2.2. Osteoblasts and BMMs co-culture experiment

To investigate the capability of wildtype and sanroque mutant osteoblasts to support


110

osteoclastogenesis, isolated osteoblastic cells were seeded in 12-well plates and allowed

to settle overnight in complete DMEM. The following day, the medium was removed

and BMMs were seeded directly on the osteoblast monolayer and the coculture system

was maintained in α-MEM. The isolated osteoblastic cells were stimulated to produce

RANKL via the supplementation of complete α-MEM with 10 nM 1α, 25

dihydroxyvitamin D3, and half-medium changes with fresh α-MEM/1α, 25

dihydroxyvitamin D3 carried out every two days. Upon the observed presence of

multinucleated cells, the cells were gently washed once with 1× PBS prior to fixation

with 4% paraformaldehyde for 15 minutes at room temperature. The fixed cells were

washed three times with 1× PBS and the osteoclasts formed were TRAP stained and

counted as previously described.

4.3.3. Flow cytometry analysis of bone marrow cell niche

4.3.3.1. Immunostaining of bone marrow cells

Immunostaining of cell surface markers using antibodies was performed based on the

protocol established by BD Bioscience. Briefly, bone marrow cells were extracted by

flushing the cavity of long bones of euthanized mouse with complete α-MEM as

described above. The collected bone marrow cells were pelleted by centrifugation at

350 g for 5 minutes. To minimise background signal, red blood cells (RBCs) lysis was

performed by resuspending the bone marrow cells in 1 ml of 1× BD Pharm LyseTM red

cell lysis buffer incubating for 5 minutes, occasionally tapping the samples gently.

Samples were then diluted with 9 ml of 1× PBS to terminate the RBC lysis and the cells

were pelleted to remove the supernatant. The cells were then resuspended in 1× PBS

and a cell count was conducted. The cells were once again centrifuged and resuspended

in ice-cold 1× PBS at a concentration of 1× 106 cells/ml.

Cell viability staining was then performed (with either LIVE/DEAD® Fixable Aqua

Dead Cell Stain Kits (Invitrogen) or Zombie UVTM Fixable Viability Kit (BioLegend))

according to manufacturer’s recommendations for 30 minutes at room temperature. If

the cells were not fixed, live/dead staining was assessed by the addition of 7-AAD (BD

Bioscience) just before running the samples through the flow cytometry machine.

Following staining with the fixable viability dyes cells were then washed with ice-cold

1× PBS and centrifuged. The following steps were then carried out on ice. The cells

were resuspended in 50 µl ice-cold wash buffer and the cell suspension was gently


111

mixed with 50 µl of diluted antibody master mix in wash buffer and left to incubate for

30 minutes on ice in the dark. After the incubation, the cell suspension was diluted with

500 µl of wash buffer and cells pelleted at 350g for 5 minutes. Cells were washed twice

more by resuspension in 100 µl of wash buffer followed by pelleting at 350g..

If the cells were to be fixed, the pelleted cells were then resuspended in 400 µl of 1×

BD Stabilising Fixative (BD Biosciences) (~106 cells in 0.5 ml) and were stored at 4 °C

until required for flow cytometric analysis the following day (fixed cells were analysed

within 24 hours). If the cells were to be analysed on the same day and were not required

to be fixed, the cells were resuspended in 400 µl of wash buffer after the third washing

step and were stained with 7-AAD as previously mentioned.

Antibodies involved in this study were titrated to appropriate concentrations to

minimise background staining. Single stained controls were performed using

Compensation Particles Set (BD Biosciences) according to manufacturer’s instructions.

Unstained and Fluorescence Minus One (FMO) controls were also prepared using bone

marrow cells and were treated the same way as test samples. The FMO controls were set

up to facilitate identification and gating of cells in the context of data spread due to the

multiple fluorochromes in the flow panel (Maecker et al., 2006). An FMO control

contains all the fluorochromes in the designed panel, except for the one that is being

analysed and hence a 4-colour panel would require 4 separate FMO controls. FMO

setups were particularly important as many of the flow panels designed for this study

involved more than 4 fluorochromes whereby, despite using the appropriate antibody

concentrations, the major source of background staining tends to be fluorescence spill

over.

4.3.3.2. Flow cytometer and analysis

The FACSCalibur, FACSCanto II and LSR Fortessa TM SORP (BD Biosciences) at the

Centre for Microscopy, Characterisation and Analysis (CMCA) of UWA were used for

flow cytometric studies. FACSDiva (BD Biosciences) software was used for application

setup, data acquisition, and data compensation. The live and single mononuclear cells

were gated using Forward Scatter (FSC) and Side Scatter (SSC) plots and at least

100,000 events were recorded. Acquired and compensated data was then imported into

FlowJo software (Three Star) for analysis.


112

4.3.4. Cryopreservation of cells for long term storage

Monolayer cells reaching subconfluency were trypsinised and pelleted as described

above. The cell pellet was then resuspended in 92% (v/v) FBS and 8% (v/v) DMSO and

aliquoted as 1 ml cell suspension per sterile cryovial. As a rule of thumb, a confluent

75-cm2 tissue culture flasks was divided into 4 cryovials. The samples were frozen at -

80 °C in an isopropanol equilibrated Mr. FrostyTM Freezing Container (Thermofisher,

USA), which is designed to achieve an optimal cryopreservation at the cooling rate of

approximately -1 °C/minute. Cryopreserved samples were subsequently transferred to

liquid nitrogen for long-term storage.

4.4. Molecular biology techniques

4.4.1. RNA extraction and quantification

4.4.1.1. General handling procedures

All buffers, PCR-grade nuclease-free water, microfuge tubes, micropipette tips and

equipment were sterilized by autoclaving or filtration prior to extractions. General

aseptic techniques were employed when handling samples and RNAs. RNAs were

stored in sterile Nuclease-free water at -80˚C. Upon removing from the freezer, RNAs

were allowed to thaw on ice. Caution was taken to avoid keeping RNAs on ice for a

long period of time so as to prevent RNA degradation.

4.4.1.2. Extraction of total RNA

4.4.1.2.1. RNA extraction from bone tissue

Total RNA was extracted from the long bones of wildtype and sanroque mutant mice

using TRIzol reagent (Invitrogen) according to manufacturer’s instructions. Briefly,

cleaned bone (with no muscle tissue attached) was snap-frozen using liquid nitrogen

and was immediately homogenised using a mortar and pestle. The snap-freezing and

grinding process was repeated several times until the tissue was completely

homogenised to powder form. An appropriate amount of TRIzol (500 μl/50 mg tissue)

was then added to the powdered tissue and thoroughly mixed by pipetting before

transferring the lysed tissue mixture into a 1.5 ml microfuge tube.

The mixture was spun down at 12000 g for 10 minutes at 4 ˚C to eliminate debris and

the supernatant was transferred to a new tube. Following 5 minutes incubation at room

temperature, chloroform (0.2 ml/1 ml TRIzol) was added to the supernatant. The

mixture was mixed vigorously by hand, incubated for 2 minutes at room temperature


113

and spun down at 12000 g for 15 minutes at 4 ˚C. The transparent upper phase

containing RNA was carefully transferred to a nuclease-free 1.5 microcentrifuge tube

and 100% isopropanol (0.5 ml/1 ml TRIzol) was added. The mixture was gently mixed

and left to incubate at room temperature for 10 minutes. The sample was then

centrifuged at 12000 g for 10 minutes at 4 ˚C. The supernatant was removed to isolate

the RNA pellet and 75% ethanol (1 ml/1ml TRIzol) was added to wash the pellet. The

sample was briefly vortexed, and then centrifuged at 7500 g for 5 minutes at 4 ˚C. The

supernatant was discarded and the RNA pellet was left to air dry for 5 minutes. 30 – 50

μl of nuclease-free water was directly added to resuspend the RNA pellet and the

sample was incubated at 55 ˚C for 10 minutes to thoroughly dissolve the RNA. The

RNA was ready for immediate downstream application, or stored at -80 ˚C until future

use.

4.4.1.2.2. RNA extraction from cell culture

Total RNA was isolated from cells in culture using TRIzol reagent (Invitrogen)

according to manufacturer’s protocol as mentioned above. Briefly, cells cultured in 6 or

12-well plates were washed once with 1× PBS and directly lysed by adding TRIzol (1

ml and 0.5 ml in 6 and 12-well plate respectively). The cells were scraped and mixed by

pipetting repeatedly prior to transferring into nuclease-free 1.5 ml microcentrifuge tubes

and the RNA isolation and wash was carried out as described above. 20 μl of nuclease-

free water was used to resuspend the pelleted RNA isolated from cell cultures.

4.4.1.3. Quantification of RNA concentration

The concentration of RNA was determined spectrophotometrically by Nanodrop® ND-

1000 UV-Vis Spectrophotometer V3.3 located in WAIMR, WA. Each RNA sample was

diluted 1:4 in nuclease-free water and measured against a nuclease-free water blank.

Two microlitres of the blank or the diluted sample were placed on the lower

measurement pedestal of Nanodrop® and were measured at 260nm and 280nm. The

measurements were repeated three times using separate drops of the sample.

The quality of RNA was determined by the ratio of the readings at 260nm and 280nm

(A260/A280), which provides an estimate of the purity of RNA with respect to

contaminants that absorb UV light, such as protein. Pure RNA has an A260/A280 ratio

of 1.5 – 2.0 and partially dissolved RNA has a ratio of <1.6 (Fleige et al., 2006;

Wilfinger et al., 1997). All purified RNA samples measured for experiments in this


114

thesis had A260/A280 ratios higher than 1.8.

4.4.2. Reverse Transcription Polymerise Chain Reaction (RT-PCR)

4.4.2.1. Reverse Transcription (RT)

Reverse transcription was carried out in a 20 μl volume, with 1000 ng RNA extracted

from bone tissue and cell culture as the starting template. All procedures were

completed on ice unless stated otherwise.

Reagent Volume

RT 1*

RNA 1 μg

dNTP (20 mM) 2 ul

Oligo dT (100 μM) 0.25 μl

Nuclease-free water x μl

*Heated at 75 ˚C for 3 minutes before adding RT2.

RT 2

5× annealing buffer 4 μl

Moloney Leukaemia Virus (MMLV)-Reverse

Transcriptase 0.5 μl

Rnase inhibitor 0.5 μl

Total volume 20 μl

The RT1 master mix, placed in a thin-walled, nuclease-free PCR tube, was spun briefly

to collect all reagent/solution at the bottom of the tube prior to heating in a PCR

machine at 75 ˚C for 3 minutes to disrupt secondary structures. The reaction mixtures

were returned on ice to promote primer annealing and this was followed by the addition

of RT2 master mix. The reaction mixture was gently mixed, spun briefly and incubated

at 42 ˚C for 1 hour to facilitate cDNA transcription. Inactivating the enzyme at 92 ˚C for

10 minutes then terminated the process. Synthesized cDNA was stored at -20 ˚C. As a

negative control, the RNA template was replaced with nuclease-free water to check for

genomic contamination.

4.4.2.2. Polymerase Chain Reaction (PCR) amplification

Polymerase chain reaction (PCR), to amplify the house-keeping gene 18S and other


115

gene markers, was carried out to ensure the presence and the quality of cDNA, and the

specificity of the primers before proceeding with real-time PCR. Where possible

primers used in this study were designed to be specific to the sequence of intron-intron

junctions of cDNA to avoid the possibility of amplifying any possible genomic

contaminants in the cDNA stock.

Reagent Volume

cDNA 1 μl

Forward primer (20 μM) 0.5 μl

Reverse primer (20 μM) 0.5 μl

2× GoTaq® Green Master Mix 10 μl

Nuclease-free water 8 μl

Total volume 20 μl

All the reaction mixes were prepared on ice. Each reaction was mixed to distribute the

ingredients evenly and was briefly centrifuged before placing into a thermal cycler.

Both non-template (negative) PCR and negative RT controls were also prepared to

identify possible contamination that might have occurred during both RT and PCR

steps.

The standard PCR cycling profiles comprised of an initialization step carried out at 94

˚C for 5 minutes, followed by denaturation at 94 ˚C for 40 seconds, annealing at 60 ˚C

for 40 seconds (all primers were designed so that the annealing temperature can be

carried out at 60˚C), and elongation at 72 ˚C for 40 seconds. The cycle of denaturation,

annealing and elongation was repeated 30 times prior to the final extension step at 72 ˚C

for 10 minutes, and was terminated and left at 4 ˚C indefinitely. The generated DNA

amplicons were loaded onto an agarose gel for electrophoresis to verify the size of the

fragments or stored at -20 ˚C.

4.4.2.3. DNA agarose gel electrophoresis

Agarose gel electrophoresis was performed using 1.5% (w/v) agarose gels prepared by

melting a agarose powder (1.5 g) in 1× TAE buffer (100 ml) in a microwave oven.

Ethidium bromide or Sybr Safe DNA stain (6 μl) was added to the molten agarose gel

and the mixture swirled gently to ensure uniform distribution. The UVTP gel tray (Bio-


116

Rad) was fitted into the Gel Caster (Bio-Rad) and the well-forming comb(s) were

placed, prior to pouring the gel mixture into the tray.

The agarose was left to solidify at room temperature. The solidified gel, along with the

tray, was placed in a Mini-Sub Cell (Bio-Rad). 1× TAE buffer was poured into the cell

until the gel was completely submerged in the buffer and the comb was gently removed.

DNA ladder (3 μl), and samples and controls (9 μl) were loaded into individual wells

and the electrophoresis was carried out at 90 V for 30 minutes. For the purpose of the

experiments in this thesis, all primers were designed to amplify amplicons that were

below 500bp in size, thus only a 100bp (260 ng) DNA ladder (Promega) was used.

Separated DNA bands were visualised using ImageQuant LAS-4000 (GE) with

fluorescence via transilluminator (312 nm).

4.4.3. Quantitative Real-Time PCR (Q-PCR) and Livak’s Equation

Following primer optimisation using PCR as described above, real-time PCR was

performed using the same cDNA with the addition of the following ingredients and the

reactions were carried out in either 96 or 384-well PCR plate.

Reagent Volume

cDNA 1.5 μl

Forward primer (100 μM) 0.15 μl

Reverse primer (100 μM) 0.15 μl

SYBR Green PCR Master

Mix 7.5 μl

Nuclease-free water 5.7 μl

Total volume 15 μl

The SYBR Green PCR Master Mix (QIAGEN) contained HotStar Taq Plus DNA

Polymerase, QuantiFast SYBR Green PCR buffer, a mix of dNTPs, SYBR Green I dye,

which emits a fluorescence signal upon binding double-stranded DNA, and ROX dye

for the normalization of fluorescent signals. To reduce pipetting error, a master mix

sufficient for three reactions (to yield 3 CT values) and negative control was prepared

for every single primer set. Each master mix was mixed to distribute the ingredients

evenly and briefly centrifuged before distributing into the corresponding wells. The

real-time PCR was carried out using the ViiA™ 7 Real-Time PCR System (Life


117

Technologies) with the following sequence: 95 ˚C for 10 minutes (activation) followed

by 40 cycles of denaturation at 95 ˚C for 15 seconds, annealing step at 60 ˚C for 20

seconds and an extension step at 72 ˚C for 20 seconds with a final elongation phase at

72 ˚C for one minute. Melt-curve analysis was also carried out to identify non-specific

products. Three CT values (threshold cycle) were obtained to deduce the mean CT value,

which was then used in Livak’s calculation as listed below to derive fold change values.

The experiments carried out in this thesis adopted the Livak’s ∆∆CT or comparative CT

method to analyse the relative fold change in gene expression of the samples and with

18S ribosomal RNA (Rn18S) serving as the internal housekeeping gene (Livak et al.,

2001; Pfaffl, 2001). Equation 1 shows the final form of Livak’s equation for a gene of

interest (GI):

Fold change = 2-∆∆CT

where ∆∆CT = ∆CT of sample A – ∆CT of sample B

∆CT = CT of GI – CT of Rn18S

CT = Mean of CT triplicate

4.4.4. Protein expression analysis using Sodium Dodecyl Sulfate-

PolyAcrylamide Gel Electrophoresis (SDS-PAGE) and western

blotting

4.4.4.1. Protein extraction from mouse organs

Approximately 5 mg of tissue from organs of interest was dissected out and rinsed

briefly with ice-cold 1× PBS. The rinsed tissue was placed in a pre-chilled mortar and

100 µl of ice-cold RIPA Lysis Buffer was added. The tissue was homogenised using the

pestle and the resulting lysate was transferred into a pre-chilled 1.5 ml microcentrifuge

tube. The lysate was passed through a 23G needle and syringe repeatedly to further lyse

the cells. The samples were left to incubate on ice for another 20 minutes. Cell debris

and intact cells were pelleted by centrifugation of the lysates at 12,000 g at 4 °C for 20

minutes. The clear supernatant was retained and transferred to fresh 1.5 ml

microcentrifuge tubes and stored at -20 °C (-80 °C for longer storage) until required.

4.4.4.2. Protein extraction from cultured cells

Monolayers of cultured cells in 6-well plates were washed once with ice-cold 1× PBS

and lysed directly by adding 200 µl of ice-cold RIPA Lysis Buffer per well and


118

incubating on ice for 20 minutes. Cell lysis was assisted by scraping the cells with

pipette tips. The resulting lysates were collected and transferred to 1.5 ml

microcentrifuge tubes. Cell debris and intact cells were pelleted by centrifugation as

described above and the clear supernatant was retained and stored at -20 °C (-80 °C for

longer storage) until required.

4.4.4.3. Quantification of protein concentration

Protein concentrations were determined based on the method established by (Bradford,

1976) using the Bio-Rad Protein Assay kit. A bovine serum albumin (BSA) standard

curve was prepared by serially diluting 10 mg/ml BSA stock to generate 6 protein

standards of known concentration (0.05 mg/ml, 0.1 mg/ml, 0.2 mg/ml, 0.3 mg/ml, 0.4

mg/ml and 0.5 mg/ml). Using a 96-well microplate, 10 ul of each of the protein

standards and protein lysates of interest were aliquotted into each well containing 200 µl

of Bio-Rad Protein Assay dye reagent, which was diluted 1 in 5 with ddH2O and filtered

through Whatman filter paper. The samples were mixed gently and incubated at room

temperature for 5 – 10 minutes followed by an absorbance reading at 595 nm using the

Bio-Rad Microplate reader. Each protein standard and sample was assayed in duplicate

and triplicate respectively and RIPA lysis buffer served as the negative control. The

protein concentration was extrapolated from the BSA standard curve.

4.4.4.4. Separation of proteins by SDS-PAGE

Using the protein concentrations readings, an equal amount of protein (20 µg) per

sample was denatured by adding 4× SDS-PAGE sample loading buffer and heating to

99 °C for 5 minutes using the Thermomixer comfort (Eppendorf). After a brief

centrifugation, denatured protein samples were separated by SDS-PAGE using the

Mini-PROTEAN® 3 Cell Electrophoresis System (Bio-Rad). An acrylamide gel with

1.5 mm thickness consisting of a separating gel (10%) and a stacking gel was prepared

as described below:

Reagent Volume

30% Acrylamide/bis

solution 2.5 ml

1.0M Tris-HCl, pH 6.8 1.875 ml

10% SDS 75 μl

MilliQ H2O 2.97 ml


119

10% APS* 25 μl

TEMED* 6 μl

Total volume 7.5 ml

*APS and TEMED were added last to initiate polymerisation.

Once mixed, the separating gel solution was immediately poured into a pre-assembled

glass plate sandwich aligned in a gel caster. The top surface of the separating gel was

overlaid with 20% ethanol and the gel allowed to polymerise at room temperature. The

ethanol was removed by inversion prior to the addition of 4% stacking gel solution,

prepared as follows:

Reagent Volume

30% Acrylamide/bis solution 0.5 ml

1.0M Tris-HCl, pH 6.8 375 ml

10% SDS 30 μl

MilliQ H2O 2.062 ml

10% APS* 30 μl

TEMED* 3 μl

Total volume 3 ml

*APS and TEMED were added last to initiate polymerisation.

The stacking gel solution was poured to overlay the separating gel and a 1.5 mm 15-

well comb was immediately inserted into the solution to create the sample wells. The

gel was left to set at room temperature for approximately 20 minutes after which the

comb was removed. Once the well comb was removed, the wells were gently rinsed

with ddH2O to remove air bubbles. The gel was then assembled into the Mini-

PROTEAN® 3 Cell Electrophoresis System filled with 1× SDS-PAGE running buffer

to cover the inner chamber and the electrode wire in the tank.

20 µl of denatured proteins and a pre-stained protein molecular weight ladder were

loaded into each well and the gel was electrophoresed at 100 V for approximately 2

hours at room temperature until the dye front of the 4× SDS-PAGE sample loading

buffer was observed dispersing from the bottom of the gel. The separated proteins were

then transferred onto a nitrocellulose membrane for detection of proteins using

antibodies specific to the target proteins of interest.


120

4.4.4.5. Protein transfer onto nitrocellulose blotting membrane

Glass plates were disassembled from the electrophoresis apparatus and were gently

separated using a spatula to expose the gel. The stacking gel was removed and discarded

while the separating gel with protein samples was left to equilibrate in western blot

transfer buffer. Pieces of filter paper and a HyBondTM nitrocellulose membrane were

measured and cut to the dimensions of the gel. The membrane, filter paper and scotch

fibre pads were soaked in the transfer buffer.

The Mini Trans-Blot Electrophoretic Transfer Cell cassette was used to arrange the

“protein transfer sandwich” according to manufacturer’s instruction. The sandwich was

prepared while being submerged in the transfer buffer with the following arrangement:

With the negative (black) side of the cassette facing upward, pre-wetted scotch fibre pad

was laid on the positive (clear or white) side of the cassette in the transfer buffer. This

was followed with three sheets of the filter paper, the pre-wetted nitrocellulose

membrane, followed by the separating gel. The sandwich was completed with another 3

pieces of filter paper on top of the gel followed by the fibre pad, thus the gel is facing

facing the black terminal and membrane is facing the red terminal.

The sandwich was gently rolled with a 50 ml tube to roll out any trapped bubbles and

the negative side of the cassette was gently pressed down onto the sandwich. The

cassette was locked firmly with the white latch and the sandwich was inserted into the

module. The apparatus were then placed in the Western Transfer tank filled with chilled

transfer buffer. An ice block was placed in the tank to prevent over-heating and the

transfer was left to run at a constant current of 0.03 A overnight.

4.4.4.6. Western blotting

After the transfer, the sandwich was unclamped, and the pads, filter papers and gel were

removed to expose the protein-containing membrane. The membrane was immediately

rinsed in 1× TBS for 5 minutes three times. To reduce non-specific binding of the

primary antibody, the membrane was then blocked by immersion in 1× TBS-Tween

containing 5% (w/v) skim milk for 1 hour with gentle rocking at room temperature. The

membrane was then washed with 1× TBS-Tween for 5 minutes at room temperature. To

detect protein expression, the membrane was probed with primary antibody specific to

the protein of interest, diluted according to dilutions listed in 4.1.6 in 1× TBS-Tween

with 1% (w/v) skim milk, and the membrane was then incubated with gentle rocking for


121

at least 4 hours (or overnight) at 4 °C.

The diluted antibody was removed and kept at -20 °C for future use. The membrane was

again rinsed three times with 1× TBS-Tween for 5 minutes per wash before it was

incubated with secondary HRP-conjugated antibody (specific to the host in which the

primary antibody was raised) diluted in 1% skim milk/1× TBS-Tween solution for 45

minutes at room temperature. The secondary antibody solution was then discarded and

the membrane was washed twice in TBS-Tween for 5 minutes and a further two times

in 1× TBS for 5 minutes.

To visualize protein expression, the Western Lightning Ultra Chemiluminescence

Substrate (Perkin-Elmer) was used according to the manufacturer’s instruction. The

membrane was allowed to incubate with the detection solution for 2 minutes at room

temperature in the dark. The detection solution was removed and the membrane was

imaged using the ImageQuant LAS-4000 luminescent image analyser (GE).

4.5. Statistics and data presentation

Data charts were generated and statistical analysis were performed using Microsoft®

Office Excel (Microsoft) software. All data presented in this study is expressed as mean

± standard error of the mean (SEM). The results are representative of at least three

independent experiments and biological repeats. For a direct comparison between

wildtype sample and sanroque mutant sample, an unpaired Student’s t-test was

performed. Statistical significance was assumed at p-values <0.05.

CHAPTER FIVE

BONE PHENOTYPIC CHARACTERISATION OF

SANROQUE MUTANT MICE

Chapter 5: Bone Phenotypic Characterisation of Sanroque Mutant Mice

123

5. CHAPTER 5: BONE PHENOTYPIC CHARACTERISATION OF SANROQUE MUTANT MICE

5.1. Introduction

Bone homeostasis is dynamically maintained by the bone remodelling cycle, which is a

complementary effect of bone resorbing osteoclasts and bone forming osteoblasts

(Henriksen et al., 2009). To achieve a healthy and strong bone tissue, the bone

remodelling process is meticulously regulated in order to maintain bone homeostasis.

Dysregulation of this coupling activity between osteoclastic bone resorption and

osteoblastic bone formation leads to bone diseases, whereby overzealous resorption

results in osteolysis and osteoporotic bone, and an imbalance favouring bone formation

results in osteopetrosis (Teitelbaum, 2000).

The coupled process is orchestrated by different cytokines and factors released locally

and systemically by different cell types (Martin et al., 2006; Morgan et al., 2008). With

the emerging field of osteoimmunology, many studies have shown that the immune

system plays a major role in bone homeostasis, and vice versa (Arron et al., 2000a;

Arron et al., 2000b). The co-regulation and dialogue between the bone organ and

immune system is highlighted in many inflammatory bone diseases, many of which are

associated with excessive bone resorption resulting in net bone loss. These conditions

can be very debilitating and place an economic burden on the society. Many of the

therapeutic treatments available are not sufficient to eradicate the diseases, and little is

known to prevent the occurrence of aberrant bone remodelling. This is because little is

understood about the mechanisms that regulate normal physiological bone remodelling,

let alone the mechanisms that tipped the bone homeostasis in these autoimmune

conditions. Hence, it is crucial to focus our research on identifying the mechanisms

regulating osteoimmune homeostasis and thus, discovering novel therapeutic targets to

prevent and treat these inflammatory bone diseases.

An effort to identify novel genes regulating bone remodelling and osteoimmunology

was carried out in collaboration with the Australian Phenomics Facility of the

Australian National University by employing an ENU-induced mutagenesis screening

approach as described in Chapter 2. Bone phenotype screening of ENU-induced mutant

mouse lines using X-ray led us to the mutant mouse line that is studied in this thesis,

the sanroque mutant mice, in which the mutation and the pre-existing autoimmune

condition had been previously mapped and characterised (Vinuesa et al., 2005a).

The sanroque mutation lies in the Roquin (Rc3h1) gene that encodes for an E3 ubiquitin


124

ligase, resulting in a T to G substitution at the ROQ domain. The E3 ubiquitin ligase

family is one of the enzymes responsible in the enzymatic cascades of the ubiquitin

proteasome system (UPS). UPS has been implicated in many physiological conditions,

including the immune system and bone homeostasis, as the system is crucial in the

maintenance of protein homeostasis.

As highlighted in Chapter 3, the hindlimbs, vertebrae and tails of the sanroque mice

screened using X-ray indicate lower bone mass when compared to wildtype littermates.

This chapter reports the quantitative characterisation of bone phenotype of sanroque

mutant mice.

5.2. Experimental results

5.2.1. Roquin gene expression is regulated during osteoblastogenesis and

osteoclastogenesis and is highly expressed in hematopoietic organs

The mRNA expression of Roquin gene encoding for the Roquin E3 ligase is expressed

in different mouse tissues and cell lines and is most abundantly expressed in

haematopoietic organs such as the bone and spleen (Figure 5.1) (Wu et al., 2009). It is

also interesting to note that haematopoietic stem cells and common myeloid

progenitors, which give rise to osteoclast progenitors, also highly express Roquin gene.

Q-PCR analysis of Roquin gene expression throughout osteoblast and osteoclast

differentiation revealed that Roquin expression is up regulated during the course of

osteoblastogenesis (Figure 5.2 A) but is down regulated during the process of

osteoclastogenesis (Figure 5.2 B). The relative differentiation state of the cells was

assessed by the expression of the marker gene osteocalcin for osteoblasts and calcitonin

receptor for osteoclasts. Such regulation of Roquin expression during differentiation

highlights the potential regulatory role of the gene in bone biology.

5.2.2. Sanroque mutation alters trabecular and cortical bone morphology

To investigate the effect of the mutation in Roquin on the maintenance of bone

remodelling in vivo, the bone architecture of the distal femurs of adult male sanroque

mutant and wildtype littermate mice (average of 100 days old) were quantitatively

analysed using Micro Computed Tomography (µCT) to obtain trabecular and cortical

bone parameters (Bouxsein et al., 2010).

The µCT data revealed that there was a significant reduction in trabecular bone volume


125

normalised to tissue volume fraction (BV/TV) in male 100 day old sanroque mutant

mice compared to littermate controls (Figures 5.3 A and B). This was accompanied by a

decrease in trabecular number (Tb.N) and an increase in trabecular separation (Tb.Sp)

(Figures 5.3 C and D). Despite observing no significant change in the trabecular

thickness (Tb.Th), the structure model index (SMI) value is higher in the sanroque

mutant mice, indicating weaker trabecular bone structure in comparison to that of

wildtype mice (Figures 5.3 E and F). The bone phenotype appeared to be largely

consistent between both sexes whereby the changes in trabecular parameters were also

observed in the female sanroque mice; however, no statistically significance change

was observed in SMI in female mice (Supplementary figure 1).

Consistently, the cortical bone volume (Ct.BV) and thickness (Ct.Th) were significantly

reduced in the femurs of the sanroque mutant mice compared to littermate controls

(Figures 5.3 G and H). While there was a trend of reduction in the cross-sectional area

inside the periosteal envelope (Tt.Ar), the cross-sectional cortical bone area (Ct.Ar)

within this region is significantly reduced and thus there is a significant reduction in the

cortical area fraction (Ct.Ar/Tt.Ar) in the sanroque mutant mice compared to the

wildtype mice (Figures 5.3 I, J and K). The marrow area (M.Ar) was also lower in the

mutant femurs, however it did not reach statistical significance (Figure 5.3 L).

Collectively these results indicate that the cortical bone of the sanroque mutant mice is

thinner. This analysis established the decreased bone mass of the sanroque mutant mice.

H&E, Goldner Trichrome and TRAP staining was performed on sections of tibias from

100 day old male sanroque mutant and wildtype mice. Visual examination of the

stained sections showed an obvious reduction of bone mass in the tibia of sanroque

mutant mice (Figures 5.4 A, 5.5 A and 5.6 A). Histomorphometric analysis was

consistent with the μCT results, male sanroque mutant mice showed a significant

reduction in trabecular bone volume normalised to tissue volume (BV/TV) when

compared to wildtype littermate controls (Figure 5.4 B). The reduction in BV/TV

corresponded to the significant decrease in trabecular number (Tb.N) and increase in

trabecular separation (Tb.Sp) in the tibias of sanroque mutant mice (Figures 5.4 C and

D). Again no significance difference in the trabecular thickness (Tb.Dm) of the tibias

was observed (Figure 5.4 E). These data further confirm the osteopenic phenotype of

sanroque mutant mice.


126

5.2.3. Sanroque mutant mice have increased osteoclast numbers in vivo

To determine if the bone loss observed in sanroque mutant mice is associated with

increased osteoclastogenesis, histomorphometry quantification of TRAP-positive

osteoclasts along bone surfaces was performed. Compared to littermate controls,

sanroque mutant mice have significantly more osteoclasts per unit bone surface

(N.Oc/BS) (Figure 5.5 A). There is no significant difference in the length of osteoclast

surface, however as there is reduced bone surface in the mutant mice there is a trend of

increase in osteoclast surface normalised to bone surface (Figures 5.5 B and C). The

increase in TRAP staining is also apparent on the macroscopic view of the calvarial

surface of the sanroque mutant mice (Figure 5.6 A). This relates to an increase in

TRAP-positive osteoclasts observed by visual microscopic examination of calvarial

sections of sanroque mutant mice when compared to wildtype control (Figure 5.6 B).

Histological analysis thus revealed that an abnormal increase in osteoclasts might

contribute to the osteopenic phenotype of sanroque mutant mice.

5.2.4. Bone mineral apposition rate in sanroque mutant mice is reduced

despite unchanged histological osteoblast parameters

Additional histomorphometry studies were performed on Goldner Trichrome sections to

analyse osteoblast parameters to further examine the possible cellular mechanism of

bone loss observed in sanroque mutant mice. Quantitative analysis on Goldner

Trichrome sections revealed a non-significant trend of increase in osteoblast number per

unit bone surface (N.Ob/BS) in the tibias of sanroque mutant mice when compared to

that of wildtype littermates (Figure 5.7 A). Whilst there is no significant difference in

osteoid surface (OS), there is a trend of increase in osteoid surface normalised to bone

surface (OS/BS) in the tibia of sanroque mice (Figures 5.7 B and C). Histological

assessment hence revealed no significant aberration in osteoblast activity in sanroque

mutant mice.

To investigate the anabolic activity of osteoblasts, Fluorescence-calcein and Alizarin-

red double labelling was performed to quantify bone formation in vivo. Despite

observing a trend of increase in osteoblast number, the measurements of interlabel

distance revealed a significant reduction in the mineral apposition rate (MAR) in

sanroque mutant mice when compared to wildtype littermates (Figure 5.8). Therefore,

the osteopenic phenotype of sanroque mutant mice may partially be due to abnormality


127

in osteoblast activity.

5.2.5. RANKL and osteoclast gene markers are highly expressed in

sanroque bone

Total RNA extracted from the bone tissues of 100 days old wildtype and sanroque mice

was reverse-transcribed in order to assess the expression levels of osteoblast and

osteoclast gene markers. Relative mRNA levels of both wildtype and sanroque bone

analysed using Q-PCR showed a significant increase in the expression level of

osteoclast markers TRAP and Cathepsin K (CatK) in the sanroque mutant (Figure 5.9

A). This data is consistent with the increase in TRAP-positive osteoclasts ex vivo

observed in histological analysis of sanroque mutant mice described above. There is no

statistical significance in the difference in expression levels of osteopontin in the

sanroque mutant bone relative to wildtype bone (Figure 5.9 B). However, expression of

the osteoblast gene marker osteocalcin was significantly reduced, which corresponds

with the observed defect in mineralisation in sanroque mutant mice.

To explore the possible molecular mechanism of bone loss in sanroque mutant bone,

the expression levels of two key cytokines of bone remodelling RANKL and OPG were

also analysed. The expression levels of RANKL and OPG were both increased in the

sanroque bone in comparison to wildtype bone (Figure 5.8 C). However, the increase in

RANKL was higher than that in OPG, hence despite the increased OPG expression

level in sanroque mutant bone, the relative RANKL:OPG mRNA ratio remained

significantly higher (Figure 5.8 D). This correlates with the increase of histologically

quantitated osteoclast number and the elevated expression of osteoclast gene markers in

the sanroque mutant mice. This observation suggests that the elevated RANKL

expression may contribute to the osteoclast abnormality and the osteoporotic-like

phenotype of sanroque mutant mice.

5.3. Discussion

By employing genome-wide, random mutagenesis with the ethylating chemical ENU to

produce and screen genetic variants for osteological phenotypes, we have identified the

low bone mass phenotype of the sanroque coisogenic mutant strain that was previously

characterised as an autoimmune mouse model (Vinuesa et al., 2005a). Genetic

sequencing has mapped the single point M199R mutation in the Roquin gene that

resulted in the loss of function of the E3 ubiquitin ligase Roquin in the sanroque mutant


128

mice. To our knowledge the involvement of Roquin gene in bone homeostasis is

unreported to date, thus highlighting the benefits of ENU-mutagenesis phenotypic

screening to mine for novel molecules involved in the regulation of bone homeostasis.

Bio-GPS analysis showed that the Roquin gene is expressed highly in haematopoietic

organs such as the bone marrow and spleen. This suggests that Roquin may play a role

in the maintenance of haematopoiesis. Furthermore, wildtype haematopoietic stem cells

that give rise to common myeloid progenitors, myeloid cells, immune cells and

osteoclast precursors, expressed high level of Roquin gene. Using Q-PCR analysis, it

was revealed that the expression of Roquin gene is suppressed during RANKL-induced

osteoclastogenesis in wildtype cells, as opposed to osteoclast differentiation gene

markers such as calcitonin receptor. Such a trend supports the fact that Roquin is highly

expressed in common myeloid progenitors in which pre-osteoclasts are derived from,

and that Roquin potentially negatively regulates osteoclast differentiation.

By comparison, the gene expression of Roquin is upregulated throughout

osteoblastogenesis in wildtype cells. Given that other upregulated genes, such as

osteocalcin, are known to be involved during the maturation of osteoblasts, it is possible

that Roquin may also play a role in regulating osteoblast function. Notwithstanding the

opposing trends of the gene expression during osteoblast and osteoclast differentiation,

such regulation of Roquin expression indicates the potential role of the gene in bone cell

biology.

The ex vivo bone phenotype characterisation of sanroque mutant mice provides the

firmest definition of function of Roquin in bone homeostasis. MicroCT analysis

highlighted the obvious overall reduction in bone mass of sanroque mutant mice.

Trabecular bone parameters such as trabecular bone volume (BV/TV), trabecular

separation (Tb.Sp) and trabecular number (Tb.N) indicate reduce trabecular bone mass

of the sanroque mutant mice. The structure model index (SMI) was developed to

provide an estimation of the microarchitectural deterioration of cancellous bone. An

aging and/or diseased bone is characterised by a conversion from plate elements to rod

elements, in which an ideal plate and rod structure has an SMI value of 0 and 3

respectively (Hildebrand et al., 1997). Consistently, the higher SMI parameter of the

sanroque mice corresponds to the weaker trabecular structure in comparison to that of

wildtype mice. The reduction in trabecular bone mass is accompanied by a decrease in


129

cortical bone mass in the sanroque mutant mice, as highlighted by reduced cortical bone

parameters such as cortical bone volume (Ct.BV), cortical thickness (Ct.Th) and cortical

area fraction (Ct.Ar/Tt.Ar).

Histomorphometric analysis further confirms the reduced bone mass of sanroque

mutant mice and could potentially provide some insights into the possible cellular

mechanisms underlying the osteopenic bone phenotype. The increase in osteoclast

number is evident in the TRAP-stained tibia and calvaria of sanroque mutant mice. This

observation agrees with the decreased bone mass of sanroque mutant mice, as increased

osteoclast numbers are associated with increased bone resorption. The increased

osteoclast number and activity in sanroque mutant bone also correlates with the

increased expression levels of the osteoclast gene markers TRAP and Cathepsin K.

Cathepsin K is a protease expressed predominantly by osteoclasts to support resorption

(Bossard et al., 1996), consequently, increased expression of Cathepsin K could

possibly indicate increased osteoclastic resorption in sanroque mutant mice. Hence,

sanroque mutation resulted in an increase in osteoclast number that potentially

contributed to bone loss.

It was intriguing to note the trend of increase in osteoblast number and osteoid surface

quantitated in Goldner-Trichrome-stained tibia sections of sanroque mutant mice. Such

observation theoretically equates with increased bone formation. Despite such

observation, calcein and Alizarin-red double labelling experiments revealed a reduction

in mineral apposition rate in the sanroque mutant mice, suggesting that the sanroque

mutation resulted in impaired osteoblast matrix deposition and mineralisation in vivo.

These observations implied that sanroque mutation possibly resulted in a condition

similar to osteomalacia, a bone disease characterised by increased amount of osteoid

surface that does not mineralise in a timely manner (Maricic, 2008). Consistently, Q-

PCR analysis also revealed a significant reduction in the expression level of the

osteoblast-specific gene marker osteocalcin, which plays a major role in mineralisation

(Neve et al., 2013), in the bone of sanroque mutant mice.

Collectively, our data suggest that there are abnormalities in both osteoclast and

osteoblast regulation of bone remodelling in the sanroque mutant mice. The bone

homeostasis was disrupted with the balance tipped in favour of bone resorption, and

subsequently increased bone turnover in the sanroque mutant mice. The apparent


130

increase in osteoblast number in sanroque mutant bone could possibly be a

compensatory effect in order for the mutant osteoblasts to cope with the defective

mineralisation and increased osteoclast resorption in vivo. Nonetheless, such assumption

has to be validated by measuring serum bone turnover markers, including biochemical

markers of bone resorption, such as serum C-telopeptide of type I collagen (CTX) and

tartrate-resistant acid phosphatise (TRACP-5b), and also bone formation markers, such

as serum procollagen type I N-terminal propeptide (PINP) and bone alkaline

phosphatase (BALP). It is also important to acknowledge the possibility of the existing

autoimmune condition that may contribute to the pathogenesis of systemic bone loss

observed in sanroque mutant mice.

In conclusion, through the bone phenotypic characterisation of sanroque mutant mice,

Roquin has been identified as a novel regulator of bone homeostasis. The sanroque

mutation appears to alter the physiological roles of bone cells in vivo, leading to

compromised bone microarchitecture and reduced bone mass of sanroque mutant mice.


131

Figure 5.1 Tissue distribution of Roquin (Rc3h1) mRNA expression. High-

throughput gene expression profiling using BioGPS (http://biogps.org) to investigate the

transcription levels of Roquin in mouse-derived tissues, organs and cell lines. The

expression of Roquin gene is highly elevated in hematopoietic organs such as the spleen

and bone marrow, and in haematopoietic stem cells, common myeloid progenitors and

immune cells.


132

Figure 5.2 Expression level of Roquin throughout osteoblast (OB) and osteoclast

(OC) differentiation. (A) Wildtype mouse calvaria-derived osteoblasts were cultured

with mineralising medium for 7, 14, and 21 days to stimulate osteoblastogenesis. (B)

Wildtype mouse derived-BMMs were stimulated with RANKL to drive

osteoclastogenesis for 5 days. RNA isolated from the cells was subjected to RT and Q-

PCR analysis to determine the expression level of Roquin throughout the differentiation

of bone cells. Rn18s was used as the housekeeping gene while osteocalcin and

calcitonin receptor were used as osteoblast and osteoclast marker genes respectively.

The expression levels analysed using Livak’s equation revealed that the expression of

Roquin increased throughout osteoblast differentiation, conversely Roquin expression

decreased throughout osteoclastogenesis. Data are presented as mean ± SEM. * p<0.05;

** p<0.001 comparison against osteoblast day 7 (OB day 7) and osteoclast day 0 (OC

day 0).

(A)

(B)


133

Figure 5.3 Mutant sanroque mice display reduced bone mass. (A) Representative 2D

μCT images and morphometric analysis of (B-E) trabecular and (G-L) cortical bone of

tibias from 100 days-old male wildtype and sanroque mice. Data are collected from 4

Wildtype sanroque (A) (A)

Sagi

ttal s

ectio

n Tr

ansa

xial

sect

ion

(B)

(C)

(D) (E)

Wildtype

sanroque

(F)

(I) (G) (H)

(J) (K) (L)


134

mice and are presented as mean ± SEM. * p<0.05; ** p<0.001; n.s. = no significance in

comparison to wildtype littermates, n = 4 per group.


135

Figure 5.4 Histological analysis showed mutant sanroque mice have reduced

trabecular bone mass. (A) H&E stained sections of 100 day old sanroque and wildtype

decalcified paraffin-embedded tibias. Histomorphometric quantification of (B) BV/TV,

(C) Tb.N, (D) Tb.Sp and (E) Tb.Dm of mice. Data are presented as mean ± SEM. *

p<0.05; ** p<0.001; n.s. = no significance in comparison to wildtype littermates. Data

are representative for 3 mice per group.

(B)

Wildtype sanroque (A)

(C)

(D) (E)

Wildtype sanroque


136

Figure 5.5 Increase osteoclast number in trabecular bone of sanroque mutant mice.

(A) TRAP stained sections of 100 day old sanroque and wildtype non-decalcified

plastic-embedded tibias. Histomorphometric quantification of (B) number of osteoclasts

per bone surface (N.Oc/BS, OC/mm), (C) osteoclast surface normalized to bone surface

(Oc.S/BS, %) and (D) osteoclast surface (mm). Data are presented as mean ± SEM. *

p<0.05; ** p<0.001; n.s. = no significance in comparison to wildtype littermates. Data

are representative for 3 mice per group.

Wildtype sanroque(A)

(B) (C)

(C) sanroque

Wildtype


137

Figure 5.6 Increased TRAP staining on the surface and in sections of sanroque

mutant calvaria. (A) Whole calvariae were fixed and stained for TRAP to identify

osteoclasts on the calvarial surface. (B) Representative images of TRAP stained

sections of 100 day old sanroque and wildtype decalcified paraffin-embedded calvaria.

Original magnifications ×10 (top) and ×20 (bottom).

(A) A)

Wild

type

sa

nroq

ue

Wildtype sanroque

×10

×20

(B)


138

Figure 5.7 Osteoblast number and osteoid surface are unaffected in the trabecular

bone of sanroque mutant mice. (A) Goldner Trichrome stained sections of 100 day old

sanroque and wildtype non-decalcified plastic-embedded tibias. Histomorphometric

quantification of (B) number of osteoblasts per bone surface (N.Ob/BS, OB/mm), (C)

osteoid surface normalized to bone surface (OS/BS, %) and (D) osteoid surface (mm).

Data are presented as mean ± SEM. * p<0.05; ** p<0.001; n.s. = no significance in

comparison to wildtype littermates. Data are representative for 3 mice per group.


(B) (C)

(D) sanroque

Wildtype


139

Figure 5.8 Decreased in vivo bone mineral apposition rate in sanroque mutant

mice. (A) Representative images of calcein and Alizarin-red double labelled tibias of 70

day old male wildtype and sanroque mice. (B) Consistent with the osteoporotic

phenotype, the mineral apposition rate (MAR) in sanroque mutant mice is significantly

reduced. Data are presented as mean ± SEM. * p<0.05; ** p<0.001; n.s. = no

significance in comparison to wildtype littermates, n = 3 per group.

sanroque

Wildtype


(B)


140

Figure 5.9 Sanroque mutant bones expressed higher levels of osteoclast gene

markers and increased RANKL:OPG ratio. Q-PCR analysis on mRNA extracted

from wildtype and sanroque mutant bones revealed (A) elevated expression of

osteoclast gene markers, (B) reduced osteoblast-related osteocalcin expression and (C)

increased expression of RANKL and OPG, both of which are key cytokines of bone

remodelling, in sanroque mutant bone. (D) Despite the increase in OPG mRNA level,

the RANKL:OPG relative mRNA ratio remains significantly higher in mutant bone.

Results are derived from 3 biological repeats and are presented as mean ± SEM. *

p<0.05; ** p<0.001; n.s. = no significance in comparison to wildtype littermates.

(B)

(D)

Wildtype sanroque

(C)

(A)

CHAPTER SIX

IN VITRO INVESTIGATION ON THE ROLE OF ROQUIN IN

OSTEOBLAST BIOLOGY

Chapter 6: In vitro Investigation on The Role of Roquin in Osteoblast Biology

142

6. CHAPTER 6: IN VITRO INVESTIGATION ON THE ROLE OF ROQUIN IN OSTEOBLAST BIOLOGY

6.1. Introduction

The sanroque ENU-induced mutant mouse model was reported to develop an

autoimmune disease consistent to SLE (Vinuesa et al., 2005a; Yu et al., 2007). As

established in the previous chapter, the sanroque mutant mice also display reduced bone

mass consistent with an osteopenic phenotype. The sanroque mutant mice carry a single

point mutation in the Roquin gene, which resulted in the loss of function of an encoded

protein that has a structure similar to an E3 ubiquitin ligase. Q-PCR analysis revealed

that Roquin gene transcription is upregulated throughout osteoblastogenesis, however is

suppressed during osteoclastogenesis.

Histomorphometric characterisation of the bone phenotype of sanroque mutant mice

revealed some insight into the possible cellular mechanisms underlying the reduced

bone mass. While a non-significant trend of increase in osteoblast number was observed

in the sanroque mice ex vivo, this analysis does not provide evidence of the osteoblast

functional defects that may contribute to the low bone mass phenotype. To assess

osteoblast function, double fluorochrome labelling was performed. Quantitative analysis

of the two labels revealed that sanroque mutants have reduced mineral apposition rate,

implying a possible deficiency in bone deposition by sanroque mutant osteoblasts in

vivo.

Consistently, Q-PCR revealed a reduction in osteocalcin gene transcription in the bone

of sanroque mutants. Given that osteocalcin is a terminal marker of osteoblastic

differentiation, this suggests that the sanroque mutation could possibly play a role in

osteoblastic gene expression and bone formation in vivo. In line with these findings, this

chapter investigates the role of Roquin in osteoblast biology through a series of in vitro

experiments.


Based on the aforementioned observations, we hypothesised that the low bone mass

phenotype of sanroque mutant mice was due to a cell-autonomous defect of osteoblasts.

To assess this, in vitro osteoblastogenesis was carried out on cells isolated from two

sources: the long bones and calvariae of wildtype and mutant mice. Due to the nature of

the bone tissues in which these cells come from, whereby long bones and calvariae are

formed via endochondral and intramembranous ossification respectively, we assumed

that the osteoblasts isolated from these two sources were pre-programmed differently to


143

form bone and hence were treated separately rather than being pooled together.

The calvaria isolated osteoblasts (COBs) and long bone isolated osteoblasts (LBOBs)

were cultured in osteogenic media to induce osteoid production and mineralisation.

Since larger numbers of cells are obtainable from the long bones, LBOB cell cultures

were also subjected to RNA and protein extraction for gene expression analysis and

alkaline phosphatase analysis respectively. In addition, the capability of sanroque

mutant LBOBs to support osteoclastogenesis was examined by performing an in vitro

co-culture system between osteoblasts and bone marrow macrophages.

6.2.1. Reduced bone mineralisation by sanroque mutant COB cultures

Osteoblasts isolated from calvariae of 100 days old mice were cultured with osteogenic

medium for 21 days in vitro. Alizarin red staining of mineralisation formed revealed an

obvious reduction in mutant sanroque COB cultures (Figure 6.1 A). Using ImageJ to

examine the percentage area of mineralisation, the analysis confirmed that the bone

formation by sanroque COBs is completely abrogated in comparison to wildtype

control (Figure 6.1 B).

In an attempt to rescue the abated osteoblast function from the calvariae of sanroque

mutant mice, COBs were cultured with osteogenic media in the presence of bone

morphogenetic protein-2 (BMP-2), a strong inducer of osteoblast differentiation. The

addition of BMP2 at a dose of 50 μg/ml significantly elevated the formation of

mineralised nodules by wildtype COBs, emphasising the potency of the cytokine as an

osteoinducer. However, the addition of BMP2 did not improve the abrogated

mineralisation in sanroque mutant COB.

These results supports the reduced mineral apposition rate based on the in vivo double

calcein labelling experiment in the sanroque mutant mice, suggesting that the mutation

in the Roquin gene may result in osteoblast-autonomous defect.

6.2.2. Increased calcium deposition by mutant sanroque LBOB cultures

To determine if the osteopenic phenotype of sanroque mutant long bones is due to cell

autonomous defects in osteoblasts, osteoblasts from long bones of 100 days old mice

were isolated and expanded. LBOBs were cultured in osteogenic medium for 21 days

and alizarin red staining showed more calcium deposition from sanroque mutant

LBOBs than from wildtype controls (Figure 6.2 A). Quantitative analysis of the area of


144

mineralisation revealed a significant increase in LBOB cultures from sanroque mutant

mice when compared to wildtype LBOB (Figure 6.2 C).

The addition of the osteoinducer BMP2 into wildtype LBOB cultures significantly

elevated the area of mineralisation (Figure 6.2 B and C). Interestingly, the addition of

BMP2 into sanroque LBOB culture did not significantly increase the area of

mineralisation, which was unexpected (Figure 6.2 B and C). On a closer inspection, the

areas of mineralisation formed by sanroque LBOBs appeared morphologically different

from those observed in wildtype LBOB cell cultures, which formed a distinctly

organised structure with relatively smooth edges. On the other hand, mutant sanroque

LBOBs formed calcified deposits that seemed irregular, and had rough edges (Figure

6.2 B). However, the morphology of the calcified deposits in vivo requires further

investigation with 3D-ultrastructural analysis using immunohistochemistry, confocal

microscopy and μCT. Collectively, these results show that there is a cell-autonomous

mineralisation defect and suggest an acceleration of osteoblast maturation in the long

bone of sanroque mutant mice.

6.2.3. Alkaline phosphatase activity level is altered throughout osteogenesis

in sanroque LBOB cell culture

During osteoblast differentiation, the matrix maturation phase is characterised by

maximal expression of alkaline phosphatase (ALP). To further evaluate the activity of

mutant osteoblasts of sanroque mutant mice, LBOBs were cultured in osteogenic

medium. Cell cultures were stained for ALP (blue staining) at the end point of day 21 to

reveal an increase in ALP staining in sanroque LBOB culture when compared wildtype

control (Figure 6.3 A). The staining however does not provide quantitative evidence of

ALP activity in sanroque mutant LBOB.

To determine ALP activity protein was collected from LBOB at 0, 7, 14 and 21 days

post osteogenic induction to determine ALP activity using colorimetric assay. As

expected, the ALP activity of wildtype LBOBs cultured in osteogenic medium

progressively increased throughout osteoblast differentiation and is significantly

elevated at day 14 and 21 in comparison to day 0. However, sanroque mutant LBOB

showed no significant changes in ALP activity during osteogenic induction (Figure 6.3

B). When compared against wildtype LBOB on day 0, 7 and 14, sanroque LBOB

displayed a trend of increase, albeit not significant, in ALP activity. At day 21, ALP


145

activity of sanroque LBOB decreased slightly. When ALP activity throughout

osteoblast differentiation by wildtype and sanroque LBOB is compared on respective

days, the difference however did not reach statistical significance, most likely due to the

large error bars.

6.2.4. Mutation of Roquin affects the expression levels of osteoblast gene

markers during differentiation of sanroque LBOBs

To gain some insight into the abnormal osteoblastic activity of sanroque LBOBs, the

expression of osteoblast gene markers was compared between wildtype and sanroque

mutant LBOBs. RNA was isolated from wildtype and sanroque LBOBs at day 0, 7, 14

and 21 of culture in osteogenic medium and reverse transcribed to obtain cDNA. The

expression levels of osteoblast gene markers alkaline phosphatase, osteocalcin,

osteopontin and osterix were monitored by Q-PCR and normalised against the house

keeping gene Rn18s.

By comparing the expression levels of alkaline phosphatase on each day, the mRNA

level of alkaline phosphatase in non-stimulated sanroque LBOBs (day 0) is significantly

lower in comparison to that of wildtype LBOBs (Figure 6.4 A). Seven days post

osteogenic induction resulted in a significant increase in alkaline phosphatase

expression in sanroque LBOB culture. However there is no significant difference

observed in day 14. The significant increase in the mRNA level of alkaline phosphatase

in sanroque LBOB in comparison to that of wildtype is observed on day 21. This

observation suggests a possible role of Roquin in regulating alkaline phosphatase

mRNA levels.

Wildtype LBOBs also show a significant trend of increase in osteocalcin expression

level throughout osteogenic induction (Figure 6.4 B). Comparative analysis of

osteocalcin expression at day 0 showed a significant reduction in non-stimulated

sanroque LBOBs relative to wildtype controls. Similar to alkaline phosphatase

expression level at day 7, the stimulation with osteogenic media resulted in a significant

increase in the expression level of osteocalcin by sanroque LBOB culture in

comparison to wildtype LBOB. However at day 14 and 21 there is no statistical

difference in osteocalcin expression level of both wildtype and sanroque LBOBs. When

compared at day 0, the sanroque LBOB culture showed a significant decreased in

osteopontin expression level. The expression level then increased to draw the same level


146

to wildtype in day 7 and 14. Despite showing no significant difference on day 7 and 14,

the expression level of osteopontin in sanroque mutant LBOBs is significantly higher

on day 21 when compared to wildtype control.

The upregulation of osteoblast gene markers during osteogenesis requires the

transcription factor osterix. The progressive increase in expression levels of alkaline

phosphatase, osteocalcin and osteopontin corresponds to the significant increase of

osterix expression level at day 14 and 21 against respective day 0 in wildtype LBOB

cell culture (Figure 6.4 D). In both wildtype and sanroque LBOB cultures, osteogenic

stimulation induced the expression of osterix as early as day 7. It is interesting to note

that at day 0, osterix expression in non-stimulated sanroque LBOBs is significantly

higher than that in wildtype LBOBs. The comparative analysis of osterix expression

between wildtype and sanroque LBOBs on day 7 and 14 revealed no significant

changes. However at day 21, the level of osterix expression is significantly elevated in

sanroque mutant LBOB culture when compared against wildtype culture.

Collectively, Q-PCR analyses indicate that the expression of osteoblast gene markers in

sanroque mutant LBOB throughout osteogenesis is altered. This could contribute to the

abnormal increase in in vitro mineralisation and altered alkaline phosphatase activity by

sanroque LBOB.

6.2.5. Mutant LBOB from sanroque mutant mice have reduced capability

to support osteoclastogenesis in vitro

As observed in Chapter 5, RANKL gene expression is increased in the long bones of

sanroque mice. As osteoblasts are a major source of RANKL in vivo, an in vitro

coculture system was set up to mimic in vivo osteoblast-osteoclast interactions. LBOBs

grown in monolayers were stimulated with 1,25-dihydroxyvitamin D3 (D3) to induce

RANKL and M-CSF expression. Bone marrow macrophages (BMMs) were then seeded

onto the LBOB monolayer and these factors then stimulate early osteoclast precursors

to differentiate into mature osteoclasts. The experiment was set up in four groups: (1)

wildtype LBOB and wildtype BMMs (WT OB/WT BMM), (2) sanroque LBOB and

sanroque BMM (san OB/san BMM); (3) wildtype LBOB and sanroque BMMs (WT

OB/san BMM), and (4) sanroque LBOB and wildtype BMMs (san OB/WT BMM).

Upon the formation of large multinucleated osteoclasts, the cell cultures were fixed and

stained for tartrate-resistant acid phosphatase (TRAP) activity. TRAP-positive


147

osteoclasts with more than 3 nuclei were then tallied.

Osteoclast counts from WT OB/WT BMM and san OB/WT BMM were compared to

explore the ability of sanroque osteoblasts to support osteoclast differentiation of

presumably “normal” BMMs from wildtype mice and the analysis revealed a significant

decline in osteoclast number in san OB/WT BMM cultures (Figure 6.5). This suggests

that sanroque mutant LBOBs have reduced capability to support osteoclastogenesis in

vitro. In contrast to this result, when sanroque BMM were cultured with wildtype

osteoblasts (WT OB/san BMM) in comparison to wildtype osteoblasts and BMM (WT

OB/WT BMM) a significant increase in osteoclasts was observed, suggesting enhanced

osteoclastogenesis by sanroque BMMs. This trend of increased osteoclastogenesis was

also observed when comparing san OB/ WT BMM with san OB/san BMM. Therefore,

osteoclastogenesis of sanroque BMM is not affected by the diminished ability of

sanroque LBOBs to support osteoclastogenesis. Overall, the assay revealed a

dysregulated cell-cell interaction between sanroque mutant LBOBs and BMMs that

may contribute to the osteopenic phenotype.

6.3. Discussion

In this chapter, the cellular mechanism(s) by which Roquin is involved in the

maintenance of bone homeostasis was investigated by primarily addressing the

differentiation and activity of osteoblasts isolated from sanroque mutant mice. The in

vitro experiments documented here have indicated the abnormal osteoblast activities in

sanroque mutant mice, which may contribute to the osteopenic phenotype.

Analysing their ability to form mineralised nodules in vitro provides an insight in the

maturation and activity of osteoblasts. Due to the nature of the bone organ tissues that

the cells originated from, calvaria isolated osteoblasts (COBs) and long bone isolated

osteoblasts (LBOBs) were treated as two distinct osteoblast populations. This is because

bone tissues develop via two different processes: intramembranous ossification to form

most of the craniofacial skeleton and endochondral ossification that is responsible for

formation of the long bones (Ortega et al., 2004). Endochondral ossification requires the

differentiation of mesenchymal cells into chondrocytes to serve as a template for bone

morphogenesis. This process requires the essential osteogenic factor Indian hedgehog

(Ihh), which is not required during intramembranous ossification where mesenchymal

cells are capable of directly differentiating into bone-forming cells (Chung et al., 2004;


148

St-Jacques et al., 1999). Different signalling pathways direct these two distinct bone

morphogeneses and hence bone cells from these two sources are pre-programmed

differently. Therefore, bone cells have distinct signalling properties that reflect their

origin and this should be considered in the research of bone biology.

The important differences between calvarial and long bone osteoblasts are especially

obvious in sanroque mutant osteoblast cultures. Mutant COBs and LBOBs responded to

osteogenic induction differently, whereby sanroque COBs failed to formed bone

nodules, whereas sanroque LBOBs displayed higher calcific deposits in comparison to

wildtype LBOB cultures (Figure 6.1 A and 6.2 A). Interestingly, osteogenic medium

stimulates both wildtype COBs and LBOBs to form mineralised nodules. The analysis

of mineralisation formed by sanroque COBs was consistent with the reduction of

BV/TV and MAR in the mutant bone as discussed in the previous chapter. Although it

seems contradictory, the increase bone nodule formation by sanroque LBOB does not

necessarily disagree with the μCT, histology and MAR results discussed above. The

accelerated mineralisation in vitro by sanroque LBOB could possibly be cell intrinsic,

which resulted in the abnormal osteoblastic machinery that affect normal calcium and

mineralisation as observed in bone nodules that are morphologically different from that

of wildtype (Figure 6.2 B). Nevertheless, the morphology of the bone nodules has to be

further confirmed with in vitro bone nodule 3D-ultrastructural analysis using a

combination of techniques such as transmission electron microscopy, X-ray microprobe

and fluorescence-confocal scanning laser microscopy (Bhargava et al., 1988; Xynos et

al., 2000).

Accelerated mineralisation is often associated with the production of immature, woven

bone characterised by poor lamellation and fragile bone, often seen in bone diseases

such as osteomalacia (Corsi et al., 2003; Tranquilli Leali et al., 2009). The increased

SMI of sanroque mutant bone that was picked up during μCT analysis indicates a

weaker bone structure. However, this has to be further validated by measuring the

biomechanical properties of the mutant bone. There is also a possibility of increased

mineralization despite the reduction of bone matrix deposited (MAR) in sanroque long

bones. This could be assessed by analysing parameters related to matrix mineralisation,

such as mineralisation lag time and osteocyte number. Hypothetically, defective

mineralisation should show decreased mineralisation lag time with an increase in


149

osteocyte number in vivo (Meury et al., 2010).

The cell autonomous defect of sanroque mutant osteoblasts is further shown by the

response of COBs and LBOBs toward BMP2. In COB cell culture, BMP2 was not able

to rescue to aberrant bone formation by sanroque COBs, whereas the cytokine

significantly elevated the formation of bone by wildtype COBs. The insensitivity

towards the potent osteoinducer BMP2 is also observed in sanroque LBOBs; however,

this could be because osteogenesis of sanroque LBOBs is inherently higher. A potential

reason for this observation is that the BMP2 signalling pathway itself may be altered

due to the sanroque mutation. Several publications have implicated the role of E3

ubiquitin ligases in BMP2 signalling (Zhang et al., 2001; Zhao et al., 2003; Zhu et al.,

1999). Further more, a recent publication has reported that BMP/Smad pathway is being

repressed in osteoblastic cells of SLE patients, leading to impaired osteoblastic

differentiation and may underlie the pathology of osteoporosis in these patients (Tang et

al., 2013b). Thus, the analysis of signalling pathway would be useful in order to

investigate the role of Roquin as an E3 ubiquitin ligase in BMP2 signalling.

Alkaline phosphatase and osteocalcin are both reported to play crucial roles in

mineralisation (Boskey et al., 1998; Orimo, 2010). Consistently, the accelerated

mineralisation by sanroque LBOB is accompanied by a significant increase in the

expression of osteoblast gene markers ALP and OCN within 7 days of osteogenic

induction (Figure 6.4 A and B). Osteopontin expression levels are also increased,

particularly at day 21 post osteogenic induction (Figure 6.4 C). The increase in

expression of transcription factor osterix at day 0 could partially explain the increase in

ALP and OCN in sanroque LBOBs (Figure 6.4 D), and hints at differential in vivo

regulation of osteoblast progenitors in sanroque mice. Q-PCR analysis revealed that

sanroque mutation affects the expression of osteoblast gene markers throughout

osteoblastogenesis, yet a more in-depth gene expression study is still required. The

analysis of other key osteoblast gene markers such as transcription factor Runx2 would

also provide valuable insight into the molecular mechanism regulating

osteoblastogenesis of sanroque mutant osteoblasts. Since sanroque mutant osteoblasts

were insensitive to osteoinduction by BMP2 by, the receptors for BMP2 expression

levels would be worth analysing. It would also be interesting to analyse the expression

of collagen I in order to further elucidate the molecular mechanism through which


150

Roquin regulates mineralisation.

The coculture experiments have provided a possible cellular mechanism underlying the

bone loss of sanroque mutant mice. BMMs of sanroque mutant mice showed enhanced

osteoclastogenesis and the result is consistent with the histomorphometric data of

increased TRAP positive osteoclasts in the long bones of sanroque mice. The increase

in osteoclast formation could have been explained by an increase in osteoclastogenic

factor production by sanroque LBOBs. However the co-culture experiments revealed

that sanroque mutant LBOBs have reduced capability to support physiological

osteoclastogenesis, hence indicating that the osteoblasts might not be the contributor of

osteoclastogenic factors in vivo. Since the RANKL:OPG ratio is elevated in long bones

(Chapter 5), this suggests that there might be other cell types that contribute to the

production of osteoclastogenic factors resulting in enhanced osteoclastogenesis in

sanroque mice.

The coculture experiment further revealed that BMMs from sanroque mice have

enhanced capacity to form osteoclasts irrespective of the supporting osteoblast cells. A

possible explanation for the enhanced osteoclastogenesis could be an increase of

putative osteoclast progenitors within the bone marrow of sanroque mutant long bones.

As a result, the pre-priming effect in vivo leads to an increase in osteoclast formation

supported by both wildtype and sanroque LBOBs in vitro. The differences between

activity of osteoblasts and osteoblasts-osteoclast interaction observed in vitro could also

be due to the alteration of cellular composition of the bone marrow in sanroque mutant

mice. Assessing the percentages of stem cells and progenitor cells in the marrow by

flow cytometry would provide an indication of the bone marrow composition. However,

it is also possible that the enhanced osteoclastogenesis is observed due to the existing

autoimmune condition of the mutant mice. As discussed in chapter 2, inflammatory

cytokines have been reported to have both pro- and anti-osteogenic effects on

homeostasis. These inflammatory cytokines indirectly affect the normal signalling

pathways governing the differentiation of bone cells in vivo. When taken out of this

inflammatory milieu, it is expected that they would behave differently in vitro as well.

In summary, this chapter characterises the osteoblast defects of sanroque mutant mice.

Through a series of in vitro experiments, the results have shown the potential role of

Roquin in osteoblast differentiation and function. Osteoblasts of sanroque mutant mice


151

showed abnormal mineralisation and reduced capability to support osteoclastogenesis.

However, the molecular mechanism(s) through which Roquin regulates osteoblast

functions remains to be resolved and should be addressed in future studies.


152


Osteogenicmedia

×40

Osteogenicmedia + BMP2

×40


153

Figure 6.1 Osteoblasts isolated from calvariae (COBs) of sanroque mutant mice

have reduced capability to mineralise in vitro. Osteoblasts isolated from calvariae

were cultured with osteogenic medium in the presence of dexamethasone (10 nM), β-

glycerophosphate (2 mM) and ascorbate (50 μg/ml) for 21 days in vitro. Upon the

apparent formation of mineralised areas, cell cultures were fixed with 1%

Glutaraldehyde and stained with alizarin red. (A) Representative images of wildtype and

sanroque mutant COBs at day 21 post-confluence. The reduction of in vitro

mineralisation by mutant sanroque COBs is clearly evident when compared wildtype

COBs culture. While 50 μg/ml of BMP2 further induced wildtype COBs to mineralise,

the cytokine did not rescue the aberrant bone formation by sanroque COB. (B) The

culture plates were scanned and mineralised area was quantitated using ImageJ to reveal

the significant reduction of mineralised nodule formation by sanroque mutant COBs,

with or without BMP2, in comparison to wildtype COBs. Results are representative of 3

independent biological repeats and are presented as mean ± SEM. * p<0.05; **

p<0.001; n.s. = no significance in comparison to wildtype controls and to osteogenic

medium only cultures.

sanroque

Wildtype (B)


154

(A)

Wildtype sanroque

Osteogenic media

×10

×40


155

Figure 6.2 Enhanced mineralisation in sanroque mutant osteoblasts isolated from

long bones (LBOBs). Primary osteoblasts isolated from long bones of 100 days old

mice were cultured in osteogenic medium for 21 days to induce mineralisation in vitro.

(B)

Wildtype sanroque

Osteogenic media + BMP2

×10

×40

×80

sanroque

Wildtype (C)


156

(A) Representative images of alizarin red-stained mineralisation formed by wildtype

and sanroque mutant LBOBs. (B) The potent osteoinducer, BMP2, significantly

increased the mineralisation efficiency and the area of mineralisation in wildtype LBOB

cell culture. Interestingly, BMP2 did not have any effect on the area mineralised by

sanroque LBOBs. Observation under higher magnifications revealed disorganised

nodules formed by sanroque LBOB. (C) Quantification of the area of mineralisation

from the images in (A) and (B) using ImageJ. Results are representative of 3

independent biological repeats and are presented as mean ± SEM. * p<0.05; **

p<0.001; n.s. = no significance in comparison to wildtype controls and to cell cultures

treated with only osteogenic medium in the presence of dexamethasone (10 nM), β-

glycerophosphate (2 mM) and ascorbate (50 μg/ml).


157

Figure 6.3 Mutant LBOBs of sanroque mutant mice displayed atypical alkaline

phosphatase (ALP) activity levels throughout osteogenesis. (A) Representative

images of primary LBOB cells cultured in osteogenic media for 21 days. Cells were

fixed and stained for ALP to reveal a more intense blue staining in sanroque LBOB cell

culture. (B) Colorimetric analysis of ALP activity revealed an abnormal trend of the

enzymatic activity throughout sanroque LBOB maturation as opposed to the steady

increase of ALP activity observed in wildtype LBOB culture. Results are representative

of 3 independent biological repeats and are presented as mean ± SEM. * p<0.05; **

p<0.001; n.s. = no significance in comparison to wildtype controls and to respective day

0.

sanroque

Wildtype (B)

×80


×40


158

Figure 6.4 Q-PCR analyses of osteoblast gene markers expression levels during

osteogenesis. Primary LBOB cells were cultured in osteogenic medium and RNA was

collected at day 0, 7, 14 and 21. Osteoblast differentiation was evaluated with Q-PCR

analysis of osteoblast marker genes (A) alkaline phosphatase, (B) osteocalcin, (C)

osteopontin and (D) osterix normalized to 18s ribosomal RNA expression. Results are

representative of 3 independent biological repeats and are presented as mean ± SEM. *

p<0.05; ** p<0.001; n.s. = no significance in comparison to wildtype controls.

Wildtype sanroque

(B)

(C) (D)

(A)


159

Figure 6.5 Coculture systems between LBOBs and bone marrow macrophages

(BMMs). (A) Representative images of four coculture groups: (1) wildtype LBOB and

wildtype BMMs (WT OB/WT BMM), (2) sanroque LBOB and sanroque BMM (san

OB/san BMM), (3) wildtype LBOB and sanroque BMMs (WT OB/san BMM) and (4)

sanroque LBOB and wildtype BMMs (san OB/WT BMM). LBOBs were stimulated

with 1,25-dihydroxyvitamin D3 (10 nM) to induce RANKL and M-CSF expression.

san OB/WT BMM WT OB/san BMM

WT OB/san BMM san OB/san BMM

(A)

sanroque LBOB/sanroque BMM

(B)

Wildtype LBOB/sanroque BMM

sanroque LBOB/Wildtype BMM Wildtype LBOB/Wildtype BMM

×10

×10


160

Osteoclasts were identified by staining for tartrate-resistant acid phosphatase (TRAP).

(B) TRAP-positive osteoclasts with more than 3 nuclei were then tallied. Results are

representative of 3 independent biological repeats and are presented as mean ± SEM. *

p<0.05; ** p<0.001; n.s. = no significance when compared to each coculture system.

CHAPTER SEVEN

THE EFFECT OF SANROQUE MUTATION ON

OSTEOCLAST BIOLOGY

Chapter 7: The Effect of Sanroque Mutation on Osteoclast Biology

162

7. CHAPTER 7: THE EFFECT OF SANROQUE MUTATION ON OSTEOCLAST BIOLOGY

7.1. Introduction

As described in the previous chapter, the mutation in Roquin gene results in perturbed

osteoblast activity in vivo and in vitro, which in turn may contribute to the osteopenic

phenotype of the autoimmune sanroque mouse model. However, multiple observations

indicate abnormal osteoclast activity in sanroque mutant mice. Histomorphometry

studies revealed increased osteoclast number and TRAP staining in sanroque bone

tissues. Q-PCR analysis showed elevated expression of osteoclast gene markers TRAP

and Cathepsin K in the bone of sanroque mutants. Furthermore, the co-culture

experiment showed that sanroque bone marrow macrophages (BMMs) inherently

formed more osteoclasts in comparison to wildtype BMMs culture, regardless of

osteoblast source.

At a glance, these observations correlate well with the increased bone loss due to

increased osteoclast functions seen in many patients of autoimmune diseases. However,

the underlying mechanisms of the overzealous osteoclast activities in these pathological

conditions are complex and remain ill defined. Several osteoimmunology studies have

shown that immune cells and osteoclasts often share co-stimulatory signalling

molecules and pathways (Bozec et al., 2014; Franzoso et al., 1997; Zhang et al., 2013a).

Therefore, molecular mechanisms, such as the ubiquitin-proteasome system (UPS), that

mediate abnormal immune cell functions could also be functioning in osteoclasts and

may contribute to the bone loss in these autoimmune patients.

The systemic regulation of protein turnover by UPS underlies a wide variety of

molecular pathways essential for normal cellular functioning (Ang et al., 2009). UPS

mediates intracellular protein degradation by marking the protein of interest with a

chain of small polypeptides called ubiquitin. As described previously, UPS involves the

coordinated interplay between E1 (ubiquitin-activating), E2 (ubiquitin-conjugating) and

E3 (ubiquitin-ligating enzymes) (Pickart et al., 2004). In the context of osteoclast

differentiation, UPS is critical in mediating RANKL/RANK signalling cascades (Ang et

al., 2009). The ubiquitination of tumour necrosis factor receptor-associated factor 6

(TRAF6), a crucial adaptor of the cytoplasmic domain of RANK, is an essential step in

activating the signalling complex mediating osteoclast differentiation (Takayanagi et

al., 2000). Biochemical studies demonstrate that TRAF6 is in fact an E3 ubiquitin

ligase. Upon stimulation, TRAF6 promotes polyubiquitination of itself and of target


163

proteins of the NF-κB signalling pathway (Deng et al., 2000). The ubiquitination and

proteasomal degradation of phosphorylated IκB kinase allows the nuclear translocation

of NF-κB proteins to initiate gene transcription of downstream gene targets (Chen,

2005). NF-κB along with c-Fos, c-Jun, MITF and NFATc1 turn on the expression of

genes such as TRAP, Cathepsin K and DC-Stamp in osteoclasts (Qi et al., 2014).

In addition to TRAF6, an increasing number of E3 ubiquitin ligases have been recently

implicated in osteoclast development and function (Shu et al., 2013; Zhang et al.,

2013a). Roquin is reported to be an E3 ubiquitin ligase (Vinuesa et al., 2005a) and Q-

PCR analysis has shown that the expression of the gene is down regulated throughout

osteoclastogenesis (Chapter 3). Thus, in this chapter the effect of the sanroque

mutation, which resulted in loss of function of the protein Roquin, on osteoclast biology

is examined.


Initial characterisation of the bone phenotype of sanroque mutant mice revealed an

increase in osteoclast number and osteoclast gene expression markers. Based on these

observations, we hypothesised that the mutation in Roquin gene resulted in increased

osteoclast differentiation and function. To investigate this, bone marrow was extracted

from age and sex matched pairs of wildtype and sanroque mutant mice and a series of

in vitro experiments were then carried out to investigate osteoclast differentiation and

function.

7.2.1. Osteoclastogenesis is accelerated in sanroque mutant mice

To evaluate the formation of osteoclasts in sanroque mutant mice, in vitro

osteoclastogenesis assays were conducted. Bone marrow cells flushed from the long

bones of wildtype and sanroque mutant mice were expanded into bone marrow

macrophages (BMMs) in the presence of M-CSF. BMMs were subsequently stimulated

with RANKL and M-CSF over a time-course of 0, 1, 3 and 5 days to stimulate

osteoclastogenesis. The cells were fixed, TRAP stained and the numbers of

multinucleated osteoclasts generated were quantified. A significant increase in the

number of osteoclasts formed by BMM derived from sanroque mutant mice was

evident at days 3 and 5, indicating enhanced osteoclastogenesis in the mutant mice

(Figure 7.1).


164

The sensitivity of the wildtype and sanroque BMM cultures to RANKL was also

assessed by stimulating the cells with 12.5, 25 or 50 ng/ml RANKL for 5 days, while

the concentration of M-CSF was maintained. Quantitative analysis of the TRAP

positive multinucleated osteoclasts showed that RANKL induces a dose-dependent

increase in osteoclast number in both wildtype and sanroque BMM cultures (Figure

7.2). The increased osteoclastogenesis is also apparent in this assay, in which sanroque

mutant BMM generated significantly more osteoclasts in comparison to the number of

wildtype osteoclasts at all RANKL doses. This implies that sanroque mutant osteoclast

precursors are more sensitive to RANKL stimulation, or that there are increased

numbers of osteoclast precursors in the sanroque bone marrow, as they show increased

osteoclast numbers at lower doses of RANKL.

7.2.2. Sanroque mutation elevates the expression of osteoclast marker

genes

Total RNA isolated from RANKL-induced osteoclasts at various time points were

subjected to reverse transcription to yield cDNA. Q-PCR was performed to investigate

if the expression of osteoclast marker genes is altered in sanroque osteoclasts. The

crucial transcription factor NFATc1 involved in osteoclastogenesis is upregulated

during osteoclast differentiation of both wildtype and sanroque BMMs. In comparison

to wildtype osteoclast cultures, sanroque mutant osteoclasts showed a drastic increase

of NFATc1 expression at day 3 and 5 post-RANKL stimulation (Figure 7.3 A).

Concomitantly, the expression DC-Stamp, which encodes for the transmembrane

protein required for cell fusion during osteoclastogenesis, is markedly increased during

osteoclastogenesis of sanroque mutant cultures (Figure 7.3 B). These results suggest

that Roquin may play a regulatory role in the transcription of genes during osteoclast

differentiation upon RANKL stimulation. However, RANK expression levels showed

no significant difference throughout osteoclastogenesis between wildtype and sanroque

mutant cultures, indicating that the expression of RANK was not affected by sanroque

mutation.

7.2.3. Sanroque mutant osteoclasts demonstrate increased bone lytic

activity in vitro

The decreased bone mass of sanroque mutant mice could be associated with increased

osteoclast resorptive activity. To assess the consequence of sanroque mutation on


165

osteoclast function, in vitro bone resorption and secretion of the matrix proteinase

cathepsin K by mature osteoclasts was assessed.

To perform the bone resorption assay, wildtype and sanroque mutant BMMs were

cultured on a collagen-coated plate in the presence of M-CSF and stimulated with

RANKL to induce osteoclastogenesis. Upon the apparent fusion of BMMs to generate

small osteoclasts, cells were detached and counted. Similar numbers of small osteoclasts

were then seeded onto bovine bone discs in the presence of M-CSF and RANKL. After

36 hours the cell cultures on the bone discs were then fixed, TRAP stained and

multinucleated osteoclasts were tallied (Figure 7.4 A and C). Subsequently, the

osteoclasts were gently removed to reveal the “pits” formed by osteoclast resorptive

activity. The bone discs were stained with Toluidine blue and representative

microphotographs of bone resorption are shown (Figure 7.4 B). The darker blue areas of

resorptive pits were quantified and the data is presented as resorptive area per osteoclast

(Figure 7.4 D). In comparison to wildtype osteoclasts, each sanroque mutant osteoclast

demonstrated a significant increase in resorbed area indicating an increase in resorptive

activity.

To further evaluate the bone lytic activity of sanroque mutant osteoclasts, BMMs were

cultured in a plastic plate and stimulated to undergo osteoclastogenesis as described

above. Once multinucleated osteoclasts were observed, bovine bone powder was added

into the cell culture to initiate cathepsin K release into the medium. The supernatant and

cell lysates were collected at 4, 8 and 12 hours post addition of bone powder and were

subjected to western blotting. As shown in Figure 7.5, the addition of bone powder to

wildtype osteoclast cultures resulted in the gradual build up of endogenous cathepsin K

and prompted the release of the enzyme into the supernatant in a timely fashion. In

contrast, such a trend is not observed in sanroque osteoclast cultures. The secretion of

cathepsin K into the medium of sanroque osteoclast cultures was significantly increased

at all time points analysed (Figure 7.5). These results indicate that loss of function of

Roquin positively regulates cathepsin K release and bone resorption by sanroque

mutant osteoclasts. Thus, sanroque mutation causes increased numbers of highly active

osteoclasts to form in vitro from the bone marrow of mutant mice.

7.2.4. Mutation in Roquin gene alters key RANKL-signalling pathways

To investigate the possible mechanisms by which sanroque mutation promotes


166

osteoclastogenesis in vitro, RANKL/RANK signalling cascades were examined by

immunoblot analysis. BMMs from wildtype and sanroque mutant mice were serum

starved for four hours prior to stimulation with RANKL over a short time course of one

hour (Figure 7.6). Long time course of RANKL stimulation throughout

osteoclastogenesis was also performed (Figure 7.7). Total protein was extracted from

cell cultures at the indicated time points and western blots were performed for proteins

involved in major RANKL induced signalling pathways: ERK and p38 (MAPK

signalling), NF-κB signalling (IκB-α) and NFATc1 (Calcineurin-NFAT signalling),

with β-actin serving as an internal loading control.

As expected, RANKL stimulation rapidly induced the phosphorylation of ERK within

the first 30 minutes, followed by a rapid degradation of the protein in both wildtype and

sanroque mutant BMMs (Figure 7.6 A). Interestingly, semi-quantitative analysis of the

blots revealed significantly increased levels of phosphorylated ERK at all time points of

sanroque BMM cell cultures compared to wildtype (Figure 7.6 B). This suggests that

sanroque mutation augments the RANKL-induced ERK signalling pathway.

Endogenous phosphorylated p38 also accumulates within sanroque BMMs at earlier

time points as indicated by the marked increase in the protein level at 10 minutes post

RANKL-stimulation (Figure 7.6 A and B). Although the levels of phosphorylated p38

are comparable to wildtype over the remaining course of RANKL-stimulation, these

data suggest that sanroque mutation enhances RANKL-induced MAPK signalling

pathways.

Densitometry analysis also showed that RANKL-stimulation resulted in the rapid

degradation of phosphorylated IκB-α within the first 30 minutes, however the level of

the phosphorylated protein dramatically increased at 60 minutes. Such a trend is

observed in both wildtype and sanroque mutant cultures, though the level of

phosphorylated IκB-α of sanroque mutant cultures remained higher throughout the

experiment after the initial 10 minutes of stimulation, in comparison to wildtype control

(Figure 7.6 A and B). This indicates that the regulation of IκB-α phosphorylation and

degradation is altered in sanroque pre-osteoclasts.

Western blot analysis showed that protein expression of NFATc1 is upregulated during

osteoclastogenesis in wildtype cell cultures. The level of NFATc1 is sanroque mutant

precursors started off to be significantly lower than that of wildtype. However the


167

expression gradually caught up and by the third day of RANKL stimulation, the level of

NFATc1 in sanroque mutant cell cultures drastically increased in comparison to

wildtype cell cultures (Figure 7.7 A and B). The high levels of endogenous NFATc1 in

sanroque cell cultures persisted throughout until osteoclastogenesis was terminated at

day 4.

DC-Stamp and v-ATPase V0 subunit d2 protein expressions were used as markers of

osteoclast formation (Figure 7.7 A and B). Similar to NFATc1, the protein level of DC-

Stamp in sanroque BMM culture started lower than wildtype at day 0. By day 3, the

spike in NFATc1 coincided with the significant increase in DC-stamp protein

expression during osteoclastogenesis of sanroque BMMs in comparison to that of

wildtype. The expression of v-ATPase V0 subunit d2 protein showed a similar trend to

that of DC-Stamp, however the protein could only be detected after 3 days of RANKL

stimulation. The elevated protein expression levels of NFATc1 and DC-Stamp

throughout sanroque osteoclastogenesis were similar to that of the gene expression

analysed using Q-PCR shown above. Overall, the western blot analysis of long time

course RANKL stimulation supported the enhanced in vitro osteoclastogenesis observed

in sanroque cell cultures.

7.2.5. Ubiquitination of TRAF6 in sanroque BMMs appears unaltered

Data from in vitro investigations thus far have emphasised the increased

osteoclastogenesis in sanroque mutant BMM cultures. The increased osteoclastogenesis

could be due to the altered upstream signalling that contributes to the augmented effect

of RANKL/RANK signalling cascades in sanroque BMM cultures, which then resulted

in increased transcription of osteoclast gene markers. One of the earliest proteins being

recruited upon RANKL-stimulation is TRAF6 (Takayanagi et al., 2000). Bearing in

mind the role of Roquin as an E3 ubiquitin ligase, we set to investigate the

ubiquitination of TRAF6, which is crucial in mediating RANKL/RANK signalling.

As a prelude to understanding the E3 ubiquitin ligase activity of Roquin on TRAF6, we

obtained wildtype and mutant Roquin (M199R) plasmids to conduct in vitro

ubiquitination assay using HEK293 cells (Figure 7.8 A). The GFP-tagged constructs are

transfected in HEK293 cells along with HA-Ubiquitin and FLAG-TRAF6 and the cell

lysates were pulled-down using anti-FLAG and Protein G Sepharose beads (Sigma).

Immunoblotting with anti-HA showed no difference in the ubiquitination of FLAG-


168

tagged TRAF6 immunoprecipitates of HEK293 cells transfected with wildtype and

mutant Roquin constructs (Figure 7.8 B). This preliminary result indicates that Roquin

may not play a role in the regulation of TRAF6.

To confirm this in cells that utilise the RANK/TRAF6 signalling pathway, ubiquitinated

TRAF6 was assessed in BMMs of wildtype and sanroque mice. Firstly, BMMs were

incubated for two hours with MG132, a potent protease inhibitor, to prevent the

degradation of ubiquitinated proteins. RANKL was subsequently added in order to

stimulate the ubiquitination of TRAF6. After two hours, protein lysates were collected

and subjected to immunoprecipitation using anti-ubiquitin antibody and Protein G

Sepharose beads (Sigma) to collect all ubiquitinated proteins. The immunoprecipitates

were then subjected to immunoblotting with TRAF6 antibody.

As shown in Figure 7.9, non-stimulated cell cultures showed relatively low levels of

TRAF6 ubiquitination. Stimulation with RANKL only resulted in increased TRAF6

ubiquitination. A similar trend was observed in cell cultures that are pre-treated with

MG-132 (without RANKL stimulation), whereby the protease inhibitor resulted in the

accumulation of ubiquitinated products, including TRAF6. Interestingly, the addition of

MG132 did not appear to further accumulate the amount of ubiquitinated TRAF6 in

MG132-pretreated cell culture stimulated with RANKL. Semi-quantitative densitometry

analysis further showed that the ubiquitination level of TRAF6 in sanroque BMM

seemed relatively normal to that of wildtype in all conditions. This confirms the

previous observation that Roquin does not play a role in the posttranslational

modification of TRAF6, and that the E3 ubiquitin ligase may be involved with other

factors to mediate the increase of osteoclastogenesis observed in sanroque cell culture.

7.2.6. Blood serum from sanroque mutant mice contains proliferative-

cytokines and sanroque BMMs have increased proliferation in

response to M-CSF

During the early stages of osteoclastogenesis, cytokines such as IL-6 and GM-CSF

promote cell proliferation and survival, thereby increasing the osteoclast precursor pool

for subsequent osteoclast formation (Reddy et al., 1998). Therefore, enhanced

osteoclastogenesis could be due to increased proliferation of osteoclast precursors

within the bone marrow. It has been established that serum cytokine levels of

autoimmune patients differ from healthy controls (Kariuki et al., 2010; Lourenco et al.,


169

2009). Considering the existing autoimmune condition in sanroque mutant mice, we

speculated that this altered serum cytokine profile could mediate proliferation of

BMMs.

To investigate this, blood serum from wildtype and sanroque mutant mice was collected

via cardiac puncture. Similar number of BMMs from wildtype and sanroque mutant

mice were then cultured in medium supplemented with 10% of the collected serum for

24 hours, with or without the addition of M-CSF, and MTS assay was performed

(Figure 7.10). In cultures with no addition of M-CSF, there were no significant

differences in proliferation between wildtype and sanroque BMM. However, the

different serum types did alter proliferation rates in comparison to cells cultured in 10%

FCS for both wildtype and sanroque BMMs. Interestingly, sera from sanroque mice

induced significantly higher rates of proliferation in both wildtype and sanroque BMMs

than FCS and sera from wildtype mice.

As expected, the addition of M-CSF elevates cell proliferation in both wildtype and

sanroque BMMs. In comparison to wildtype BMMs cultured in all sera, sanroque

BMMs display higher proliferative activity in the presence of M-CSF. In contrast to the

non-M-CSF treated groups, both wildtype and sanroque sera had no significant

influence on proliferation. This is probably because the amount of M-CSF added was

excessive to mask the proliferative effect of the sera. The experiment should be

redesigned to use a serial increase of M-CSF dosage in order to further analyse if the

sera from sanroque mutant mice augments M-CSF-induced BMM expansion.

Nevertheless, these findings show that the blood serum from sanroque mutant mice

contains cytokines which induce cell proliferation and that the sanroque mutation

elevates the proliferative activity of sanroque BMMs upon stimulation with M-CSF.

7.3. Discussion

Given that the balanced activities between osteoclasts and osteoblasts are required for

the maintenance of healthy bone, it is important to analyse both cell types in the

sanroque mutant mice. The aim of this chapter was to investigate the effect of sanroque

mutation on the differentiation and activity of osteoclasts in sanroque mutant mice.

Based on a series of in vitro experiments, this chapter has shown that Roquin regulates

osteoclastogenesis and bone resorption.


170

Consistent with the increased osteoclast numbers observed in vivo, in vitro

osteoclastogenesis assay revealed the increase in RANKL-induced osteoclast formation

by sanroque mutant BMMs compared to wildtype. RANKL-induced osteoclastogenesis

is driven by the transcription factor NFATc1 (Takayanagi et al., 2002a). Activation of

NFATc1 in turn induces osteoclast fusion via the upregulation of DC-Stamp and v-

ATPase V0 subunit d2 (Kariuki et al., 2010; Kim et al., 2008; Lee et al., 2006). Indeed

the accelerated osteoclastogenesis observed in sanroque cell cultures was supported by

the increased expressions of NFATc1, DC-Stamp and v-ATPase V0 subunit d2 gene

and protein.

Bone resorption assays further showed that the function of sanroque mutant mice is

increased as each mutant osteoclast resorbed a larger area of bone compared to wildtype

osteoclasts. The bone lytic activity was further assessed by measuring the secretion of

lysosomal protease Cathepsin K that is responsible for bone resorption. As shown in

Figure 7.5, the addition of bone particles dramatically increased Cathepsin K secretion

into the supernatant of sanroque mutant osteoclast cultures, consistent with the elevated

expression of Cathepsin K gene expression in sanroque mutant bone. Collectively, these

results showing the increase osteoclastogenesis and lytic activity partially explain the

decrease bone mass of sanroque mutant mice.

Q-PCR analysis revealed the repression of Roquin gene throughout osteoclastogenesis

(Chapter 3), and this implies that the gene may play a negative regulatory role in

osteoclast formation and function. On the other hand, sanroque mutation positively

regulates osteoclastogenesis and bone resorption. The sanroque mutation resulted in

partial loss of function in Roquin (Vinuesa et al., 2005a; Yu et al., 2007), which equates

to the absence of the functional protein at earlier time point of osteoclastogenesis and

this could partially explained the augmented osteoclastogenesis and overzealous lytic

activity.

In an attempt to understand the molecular mechanism(s) by which Roquin regulates

early signalling pathways regulating osteoclastogenesis, canonical RANKL/RANK

signalling cascades were analysed. Upon binding with RANKL, RANK recruits the

adapter protein TRAF6 to activate six major signalling pathways: NFATc1, NF-κB,

Akt/PKB, JNK, ERK and p38, which then direct osteoclast-specific gene expression

leading to differentiation and function (Boyle et al., 2003; David et al., 2002; Feng,


171

2005; Gohda et al., 2005; Kobayashi et al., 2001; Li et al., 2002; Xing et al., 2002).

RANKL-stimulation resulted in the posttranslational modification of TRAF6, reportedly

via ubiquitination (Lamothe et al., 2007; Takayanagi et al., 2000). Immunoprecipitation

experiments revealed that the ubiquitination level of TRAF6 in RANKL-stimulated

sanroque BMM culture does not differ from that in wildtype, suggesting that Roquin

does not play a role in the ubiquitination of TRAF6.

One of the most elucidated pathways is the activation of the transcription factor NF-κB,

which is rapidly induced upon the recruitment of TRAF6 during RANKL stimulation.

In non-stimulated conditions, NF-κB dimers are bound to inhibitory IκB proteins and

remain inactive in the cytoplasm. In the canonical signalling pathway, RANKL

stimulation induces the ubiquitination and degradation of IκB proteins via

phosphorylation by IκB kinase (IKK) complex, thus releasing the bound NF-κB dimers

to allow the translocation to the nucleus and subsequent gene transcription

(Oeckinghaus et al., 2011). Interestingly, the levels of IκB-α phosphorylation in

RANKL-stimulated sanroque mutant BMMs remained higher. This indicates that

Roquin may not play a regulatory role in NF-κB canonical signalling pathway and the

result seemed contradictory to the increased osteoclastogenesis observed in sanroque

mutant mice. A possible explanation for this is that Roquin may play a regulatory role

in the non-canonical NF-κB pathway. It is important to note that the non-canonical NF-

κB pathway relies on phosphorylation and ubiquitination of p100 processing instead of

degradation of IκB-α to activate RelB/p52 NF-κB complex (Sun, 2011). It would be

beneficial to investigate the expression levels of proteins involved in the non-canonical

pathway.

Other RANKL-signalling pathways could probably compensate for the reduced

canonical NF-κB signalling observed in sanroque cell cultures. Conversely,

immunoblotting assay showed that the downstream NFATc1 signalling is increased as

mentioned above. The phosphorylation of p38 in sanroque mutant BMMs also occurred

at an earlier time point, indicating enhanced p38/MAPK-mediated signals, which are

required for inducing osteoclast differentiation (Li et al., 2002). The ERK signalling

pathway is also elevated as represented by the increased phosphorylation of ERK1/2 in

sanroque BMM culture. Consistent with the observed increased osteoclastogenesis and

bone resorption by sanroque mutant cell cultures, ERK has been reported to be


172

associated with cell survival and positively regulates osteoclast differentiation and bone

resorptive activity (He et al., 2011; Miyazaki et al., 2000). However, other

RANKL/RANK signalling cascades including Akt/PKB and JNK have yet to be

examined. A more in depth analysis is certainly required to identify the exact molecular

pathway through which Roquin regulates RANKL/RANK signalling cascades and

osteoclastogenesis. Attempts at immunoprecipitation with an anti-Roquin antibody in

order to find a potential substrate have been quite challenging, as the only available

antibody in the market to date did not successfully identify the protein on

immunoblotting.

The upregulation of osteoclastogenesis and osteoclast activity could also be initiated by

the increased proliferation of osteoclast precursors (Zhou et al., 2006). As highlighted

by Q-PCR analysis on gene expression throughout osteoclastogenesis, the expression of

Roquin is elevated at day 0 when BMMs were maintained in M-CSF and were not

stimulated with RANKL. This implies that Roquin may have a more prominent role in

regulating the maintenance of BMMs. Bearing in mind that the sanroque mutant mice

exhibit a Lupus-like autoimmune disease, we further postulated that the cytokine levels

in the blood serum are altered and may have an effect on the proliferation of sanroque

BMMs. Intriguingly, culture in the medium supplemented with blood sera collected

from sanroque mutant mice elevated the proliferation rate of both wildtype and

sanroque BMMs. MTS assay further revealed that Roquin deficiency accentuates M-

CSF-induced macrophage proliferation. These data show that sanroque BMMs are

potentially pre-primed to be more sensitive towards M-CSF stimulation, or that they

may have higher number of cells expressing c-fms, the receptor for M-CSF, within the

BMMs population due to the altered cytokines in the blood serum of sanroque mice.

However, such assumption should be further validated using flow cytometry and Q-

PCR analysis to assess the expression level of c-fms in BMMs. The levels of

proliferation-inducing cytokines such as GM-CSF and IL-6, in the blood serum should

also be measured. Investigation of the M-CSF downstream signalling pathway, such as

ERK and AKT, in sanroque mutant BMMs may provide another avenue in determining

the exact molecular pathway in which Roquin is involved.

In summary, this chapter has highlighted the novel role of Roquin in the regulation of

osteoclast formation and function. Through an in vitro approach, the data shown have


173

characterised the osteoclast biology of sanroque mutant mice, which showed enhanced

osteoclastogenesis and bone lytic activity. These in vitro observations agreed with the

increased in osteoclast numbers in vivo. Hence, sanroque mutation augments

osteoclastogenesis and the Roquin gene is a negative regulator of osteoclast formation

and function.


174

Figure 7.1 Osteoclast differentiation is increased in sanroque BMM culture. BMMs

isolated from sanroque mutant and wildtype mice were cultured in vitro for 5 days in

the presence of M-CSF and RANKL. Cultures were fixed and the indicated time points

and TRAP stained. Significant increase in the number of sanroque BMMs derived

osteoclast at day 3 and 5 is observed. (A) Representative microscopic images of

osteoclast differentiation at day 0, 1, 3 and 5 showing TRAP staining. (B) Quantitative

analysis of OC differentiation. TRAP positive cells with more than 3 nuclei were

categorised as osteoclasts. Results are representative of three independent biological

repeats and are presented as mean ± SEM. * p<0.05; ** p<0.001; n.s. = no significance

in comparison to wildtype controls. (Osteoclast differentiation assay performed by Dr.

Jasreen Kular)

0 1 3 5

Wild

type

sa

nroq

ue

Days of RANKL stimulation (A)

(B)

Wildtype sanroque

(B)(B)


175

Figure 7.2 BMMs derived from sanroque mutant mice have increased capacity to

form osteoclasts at lower doses of RANKL. BMMs isolated from sanroque mutant

and wildtype mice were cultured with 0, 12.5, 25 and 50 ng/ml RANKL for 5 days with

the concentration of M-CSF being the same between the various RANKL doses.

Cultures were TRAP stained and cells with more than 3 nuclei were tallied. A

significant increase in the number of osteoclasts generated from sanroque BMMs in

comparison to wildtype BMM cultures in all RANKL doses is observed. (A)

Representative microscopic images of osteoclast differentiation at the various

concentrations. (B) Quantitative analysis of OC differentiation. Results are

representative of three independent biological repeats and are presented as mean ±

SEM. * p<0.05; ** p<0.001; n.s. = no significance in comparison to wildtype controls.

(A)

12.5 25 50

Wild

type

sa

nroq

ue

) RANKL dose (ng/ml)

0

(B)

Wildtype sanroque


176

Figure 7.3 The mutation in Roquin alters the expression of osteoclast gene

markers. BMMs isolated from sanroque mutant and wildtype mice were stimulated

with RANKL to induce osteoclastogenesis. RNA was isolated from these cultures on

day 0 (non-stimulated), 3 and 5 and Q-PCR was carried out to determine the expression

levels of (A) NFATc1, (B) DC-Stamp and (C) RANK relative to housekeeping gene

Rn18s. Results are representative of three independent biological repeats and are

presented as mean ± SEM. * p<0.05; ** p<0.001; n.s. = no significance when the value

of mutant samples was compared to that of wildtype control samples, and in comparison

to respective day 0.

sanroque

Wildtype

(B) (A)

(C)


177

Figure 7.4 Sanroque mutant osteoclasts resorbed more bone. BMMs were cultured

with M-CSF and RANKL in collagen-coated plates. Small osteoclasts were trypsinised

and similar number of wildtype and sanroque small osteoclasts were transferred onto

bovine bone discs. Representative microphotographs of (A) TRAP-stained osteoclasts

on bone discs and (B) the pits formed by actively resorbing osteoclasts. (C) TRAP-

stained osteoclasts were tallied and (D) the resorbed area was quantified. Results are

representative of 3 biological repeats and are presented as mean ± SEM. * p<0.05; **

p<0.001; n.s. = no significance in comparison to wildtype controls.

Wildtype sanroque

(D) (C)


(B)


178

Figure 7.5 Increased cathepsin K released by sanroque mutant osteoclasts.

Wildtype and sanroque multinucleated osteoclasts were stimulated with bovine bone

particles. (A) Culture medium (supernatant) was collected at 4, 8 and 12 hours post

stimulation and subjected to western blot analysis for the presence of cathepsin K. Cell

lysates were collected to serve as control. (B) Semi-quantitative densitometry analysis

of cathepsin K released in the supernatant relative to cell lysate. Results are



sanroque

Wildtype (B)

(A) Wildtype sanroque

4 8 12 4 8 12

Supernatant

Cell lysate

CatK

CatK


179

Figure 7.6 Increased activation of MAPK signalling cascade during RANKL

stimulation in sanroque osteoclast precursors. Wildtype and sanroque BMMs were

p-ERK 1/2

Total ERK 1/2

IκB-α

p-p38

Total p38

β-actin

Wildtype

0 5 10 15 30 60

sanroque

0 5 10 15 30 60 RANKL stimulation (min)

sanroque

Wildtype

(B)

(A)


180

stimulated with RANKL over the time course of 0, 5, 10, 20, 30 and 60 minutes.

Proteins were extracted and subjected to immunoblotting. Phosphorylation of ERK is

increased in sanroque BMM cultures throughout the course of RANKL stimulation. (A)

Representative western blot images of protein expression involved in RANKL/RANK

signalling cascades with β-actin probed as loading control. (B) Semi-quantitative

densitometry analysis of the protein expressions relative to β-actin. Results are




181

Figure 7.7 Increased protein levels of transcription factor NFATc1, osteoclasts

markers DC-Stamp and D2 subunits during osteoclastogenesis of sanroque BMMs.

Wildtype and sanroque BMMs were stimulated with RANKL over the course of four

days to induce osteoclastogenesis. Proteins were extracted on each day and were

subjected to western blot analysis. Transcription factor NFATc1, DC-Stamp and D2

protein levels were increased in sanroque BMM cultures at day 3 and 4 post RANKL-

stimulation. (A) Representative western blot images of protein expression of involved in

sanroque

Wildtype

(B)

NFATc1

DC-Stamp

v-ATPase V0 subunit d2

β-actin

Wildtype

0 1 2 3 4

sanroque

0 1 2 3 4 RANKL stimulation (day) (A)


182

signalling pathway mediating osteoclastogenesis and bone resorption. (B) Semi-

quantitative densitometry analysis of the protein expressions relative to β-actin. Results

are representative of 3 biological repeats and are presented as mean ± SEM. * p<0.05;

** p<0.001; n.s. = no significance in comparison to wildtype controls.


183

Figure 7.8 FLAG-tagged TRAF6 immunoprecipitates of HEK293 cells transfected

with wildtype and mutant Roquin (M119R) plasmids showed no difference in

NheI KpnI

5’ LTR (MSCV)

IRES

GFP

Nhe1 Kpn1

NcoI

BamHI 3’ LTR (MSCV)

Unique sites in the MCS

BgIII XhoI SalI

EcoRI Xmn1

Xmn1

Am

pici

llin+

ψ

(A)

Wildtype sanroque

50

150

25

75

37

pcD

NA

3.1

WT

Roq

uin

Kpn1

Mut

ant R

oqui

n (M

199R

)

Construct TRAF6-FLAG

Ubiquitin HA + +

+ +

+ +

+

++ +

+ +

Total extracts IP: FLAG

IB: HA

50

150

25

75

37

IB: TRAF6

(B)

+

pcD

NA

3.1

+

WT

Roq

uin

+

LAG

Mut

ant R

oqui

n (M

199R

)


184

ubiquitination level by immunoblotting with anti-HA. (A) Schematic maps for

plasmid adapted from Yu et al. (2007). IRES, internal ribosome entry sequence; LTR,

long terminal repeat; ψ, packaging signal; MCS, multiple cloning site. (B) HEK293

cells were transfected with the indicated plasmids encoding wildtype or mutant Roquin

in the presence of HA-Ub and FLAG-TRAF6. Cells were lysed and immunoprecipitated

with anti-FLAG. Bound proteins were subjected to SDS-PAGE and immunoblotted first

with anti-HA. The membrane was stripped and reprobed with anti-TRAF6. Results are

representative of 2 biological repeats.


185

Figure 7.9 No difference in ubiquitination of TRAF6 in sanroque BMMs upon

RANKL stimulation. Analysis of ubiquitination of TRAF6 in the presence or absence

of protease inhibitors MG132, and in stimulated and unstimulated BMM. (A)

Immunoprecipitates with anti-ubiquitin was blotted with anti-TRAF6 antibody. RANKL

stimulation resulted in the ubiquitination of TRAF6. (B) Semi-quantitative densitometry

analysis of TRAF6 ubiquitination relative to total extracts showed no significant

difference between wildtype and sanroque mutant BMM. Results are representative of 3

biological repeats and are presented as mean ± SEM. * p<0.05; ** p<0.001; n.s. = no

significance in comparison to wildtype cell cultures.

sanroque

Wildtype (B)

(A)

IP: Ubiquitin

Total extracts

25

37

50

MG132 RANKL

+ +

+ -

-

+-

-+

++

-- +

- -

Wildtype sanroque

IB: TRAF6


186

Figure 7.10 Blood serum from sanroque mutant mice contains proliferative-

inducing cytokines and sanroque BMMs exhibit increased cell proliferation in

response to M-CSF. BMMs from wildtype and sanroque mutant mice were plated and

cultured with medium supplemented with either 10% of foetal calf serum (FCS), blood

serum collected from wildtype (WT sera) or sanroque mutant mice (san sera), in the

absence or presence of M-CSF. Cell proliferation was measured 24 hours after

culturing. Results are representative of 3 biological repeats and are presented as mean ±

SEM. * p<0.05; ** p<0.001; n.s. = no significance in comparison to wildtype cell

cultures.

Wildtype sanroque

CHAPTER EIGHT

OSTEOIMMUNOLOGICAL MECHANISM BEHIND THE

OSTEOPOROTIC PHENOTYPE OF

SANROQUE MUTANT MICE

Chapter 8: Osteoimmunological Mechanism Behind the Osteopenic Phenotype of Sanroque Mutant Mice

188

8. CHAPTER 8: OSTEOIMMUNOLOGICAL MECHANISM BEHIND THE OSTEOPENIC PHENOTYPE OF SANROQUE MUTANT MICE

8.1. Introduction

Inflammation-induced bone loss highlights the abundance of overlapping molecules and

regulatory pathways shared between the bone organ and immune system. Firstly, the

complex dialogue between the bone organ and immune system is unavoidable as stem

cells, which give rise to progenitors to bone cells and immune cells, are generated

within the bone marrow. For instance, with the appropriate stimuli, macrophage and

myeloid dendritic cells are able to form osteoclasts in vitro, suggesting that they

originate from a subset of common myeloid progenitors and that share a developmental

pathway. This subset of common myeloid progenitors can be identified as CD3ε- B220-

cKit/CD117+ CD115/c-fms+ CD11blo cells within the mouse bone marrow niche

(Jacome-Galarza et al., 2013).

Using transgenic mice exhibiting an autoimmune disorder due to constitutive expression

of CD70, Xiao et al. (2013) was able to further discriminate these common myeloid

progenitors into osteoclast/dendritic progenitors (CD27hi) and the more osteoclast-

committed precursors (CD27lo). This dynamic lineage bifurcation of B220-

cKit/CD117+ CD115/c-fms+ CD11blo CD27hi towards B220- cKit/CD117+ CD115/c-

fms+ CD11blo CD27lo is tightly regulated by the expression of the ligand CD70 on

activated immune cells, whereby constitutive engagement of CD27/CD70 is required to

sustain the differentiation towards dendritic cells. As a result, the aforementioned

transgenic mice develop a high trabecular bone mass pathology due to impeded

osteoclastogenesis in a CD27-dependent manner (Xiao et al., 2013).

Secondly, cytokines that play a role in orchestrating immune responses also modulate

bone remodeling and the progression of inflammatory bone diseases. The potent

stimulator of osteoclastogenesis RANKL is expressed by antigen presenting dendritic

cells and is crucial for T cell expansion (Anderson et al., 1997; Bachmann et al., 1999;

Wong et al., 1997b). The deletion of RANKL and RANK resulted in osteopetrotic bone

and dysregulated maturation of immune cells, thus highlighting the dual role of the

RANK/RANKL signalling pathway in bone homeostasis and the immune system

(Dougall et al., 1999; Kim et al., 2000; Kong et al., 1999b; Perlot et al., 2012). Early

osteoimmunology studies have shown that activated T cells are able to express RANKL

to directly induce in vitro osteoclastogenesis (Horwood et al., 1999; Kong et al., 1999a;

Kong et al., 1999b; Kotake et al., 2001; Weitzmann et al., 2001; Wong et al., 1997b). In


189

this scenario, T helper (Th) 17 cells emerged as an osteoclastogenic subset in view of its

pro-inflammatory nature. Conforming to its osteoclastogenic properties, Th17 cells also

express significant amounts of membrane-bound RANKL and play a role in the

progression of bone destruction in rheumatic joints (Sato et al., 2006).

OPG, the decoy receptor for RANKL first characterised in osteoblasts as an inhibitor of

osteoclastogenesis, is also reported to play a role in dendritic cell and B cell function

(Yun et al., 1998). OPG-deficient mice are both osteoporotic due to increased osteoclast

activity and have a compromised immune system due to perturbed B cell maturation

(Bucay et al., 1998; Yun et al., 2001). It also appears that B cells abundantly express

OPG that plays a crucial role in the maintenance of bone homeostasis, as knockout mice

lacking B cells exhibit osteoporosis (Li et al., 2007b).

In line with these aforementioned reports, the previous chapter has documented

observations that suggested the blood serum of sanroque mutant mice displayed

different cytokine compositions to that of wildtype controls. Indeed Vinuesa and

colleagues (2010) have reported that the sanroque mutant mice have elevated IL-21,

which contributes to the increase in the number of follicular helper T cells (TFH) and

the pathogenesis of lupus-like autoimmune disease (Vinuesa et al., 2005a).

Furthermore, in vivo and in vitro osteoclastogenesis is elevated in sanroque mutant mice

in comparison to wildtype littermates, suggesting that the mutation may affect

osteoclastogenesis. Following these observations, we set to investigate the population of

common progenitors present within the bone marrow reservoir and the possible in vivo

cellular mechanism which may contribute to the upregulation of putative osteoclast

progenitors in sanroque mutant mice.


Using flow cytometry, we analysed the profile of several sub-populations of myeloid

cells in the bone marrow of both wildtype and sanroque mutant mice. Briefly, sex and

age-matched wildtype and sanroque mutant mice were euthanized and bone marrow

cells were harvested by flushing the long bone with serum-free α-MEM. Spleens were

also collected and homogenised to obtain single cell suspension. Red blood cells were

lysed prior to immuno-staining with antibodies listed in Chapter 4. The stained single

cell suspensions were then analysed and cell phenotypes were identified with cell

surface markers based on the following gating strategies: non-lymphoids/myeloids (Lin-


190

; CD3e- CD45R/B220- CD11b-), haematopoietic stem cells (HSCs; Lin- CD117(cKit)+

Sca1(Ly6A/E)+), mesenchymal stem cells (MSCs; Lin- CD117+ Sca1-), macrophages

(Mϕ; Lin- CD117+ CD11blo CD115(c-fms)+), dendritic cells (DCs: Lin- CD115-

CD11c+), osteoclast/dendritic progenitors (OC/DCP; Lin- CD117+ CD115+ CD11blo

CD27hi), pre-committed osteoclast precursors (pOCP; Lin- CD117+ CD115+ CD11blo

CD27lo CD27lo), committed osteoclast progenitors (OCP; LM- CD117+ CD11blo

CD115+ RANK+), T cells (CD3ε+ CD4+ and CD3ε+ and CD8+ for T helper and T

cytotoxic cells respectively), follicular helper T cells (TFH; CD3ε+ CD4+ CXCR5+

PD1+) and B cells (CD3ε- CD45R/B220+).

8.2.1. The putative osteoclast progenitor population present within the

bone marrow of sanroque mutant mice is increased.

Using antibodies for pan-lymphocyte markers (CD3ε and CD45R/B220) and leukocyte

marker (CD11b) to gate out lymphocyte and myeloid cells, flow cytometric analysis

revealed the very apparent reduction of a lymphocyte population in quadrant 4 (Q3) in

sanroque bone marrow niche in comparison to wildtype (Figure 8.1 A). Further analysis

using a different antibody panel revealed that the missing population within the mutant

bone marrow is CD45R/B220hi, indicating that the differentiation of B cells in sanroque

mutant mice could possible by compromised (Figure 8.1 B).

To provide a rough overview of the stem cell niche of the bone marrow, CD117 and

Sca-1 antibodies were used to further analyse the non-lymphocyte/myeloid (Lin-)

population. As shown in Figure 8.1 C, the Lin- population of sanroque bone marrow

cells were skewed towards Sca1+ and CD117+, indicating that the sanroque mutation is

affecting early differentiation of myeloid stem cells, and possibly haematopoiesis.

By comparison, monocytic CD11b+ only cell population in Q1 is significantly increased

in the bone marrow harvested from sanroque mutant mice compared to that of wildtype

littermates (Figure 8.1 A). On the other hand, the population of non-

lymphocyte/myeloid (Lin-) population in sanroque bone marrow is comparable to the

wildtype. Since common myeloid progenitor cells are a subset of the monocytic Lin-

CD11blo population, we further dissected the population to analyse macrophage and

dendritic cell populations using the pan-macrophage marker CD115 (or c-fms) and pan-

dendritic cell marker CD11c respectively. The analysis revealed that there is a

significant increased in the percentage of macrophages in sanroque bone marrow when


191

compared to wildtype (Figure 8.2 A). This suggests that the differentiation of common

myeloid progenitors is skewed towards macrophages, and as a result the percentage of

dendritic cells in the mutant bone marrow is significantly reduced (Figure 8.2B).

As reported, osteoclast precursors reside within CD11blo subset of Lin- CD117+ CD115+

bone marrow cells and express RANK. Flow cytometry indicated that this population is

also significantly higher in sanroque mutant bone marrow that wildtype controls (Figure

8.3 A). RANK+ committed osteoclast precursors are reported to arise from pre-

committed osteoclast precursors (Lin- CD117+ CD115+ CD11blo CD27lo), which have

higher osteoclastogenic potential in comparison to OC/DC common precursors (Lin-

CD117+ CD115+ CD11blo CD27hi). Consistently, flow cytometric analysis of CD27

expression levels on gated committed osteoclast precursors Lin- CD117+ CD115+

CD11blo revealed that the rare subpopulation of pre-committed osteoclast precursors

CD27lo is significantly higher in sanroque mutant bone marrow (Figure 8.3 B). The

OC/DC CD27hi common precursor subpopulation of sanroque bone marrow on the

other hand is comparable, albeit with a non-significant reduction, to that of wildtype.

Perhaps increasing the sample size may change the statistical significance of the

reduced OC/DCP Lin- CD117+ CD115+ CD11blo CD27hi observed in mutant bone

marrow.

Collectively, these data indicate that haematopoiesis of sanroque mutant mice is

dysregulated, resulting in a significant expansion of putative osteoclast progenitors

within the bone marrow reservoir.

8.2.2. CD4 T cells of sanroque mutant mice expressed less CD70

It has been reported that sustained CD27 interaction with its ligand CD70 prevents the

differentiation of OC/DCP into osteoclasts. The expression of CD27 is down regulated

from the cell surface upon contact with CD70-bearing cells thus favouring

differentiation towards DC. Based on this report, we hypothesised that the CD27-CD70

interaction is reduced in sanroque mice.

We first examined the population of CD4 and CD8 T cells and flow cytometric analysis

revealed a trend of increase, albeit not significant, in both populations in sanroque

mutant mice (Figure 8.4). Increasing the sample size would improve the statistical

power of the data. Among the T cell populations in the bone marrow of the wildtype


192

control, CD70-bearing cells are mostly CD4+ (Figure 8.5A). Comparative analysis of

CD70 expression between wildtype and sanroque CD4 T cells revealed a significant

down regulation of the ligand for CD27 in mutant CD4 T cells (Figure 8.5 B). The

expression of CD70 in CD8 T cells, B and dendritic cells of sanroque bone marrow is

similar to that the wildtype control (Figure 8.5 C, D and E). Therefore, CD4 T cells are

the major contributor of CD70 in wildtype controls and the down regulation of the

ligand in sanroque mutant mice resulted in a reduction of CD27-CD70 interaction,

leading to the observed decrease in the dendritic cell population and an increase in

osteoclast progenitors in sanroque mutant mice.

8.2.3. Splenocytes of sanroque mutant mice comprised a higher number of

common myeloid progenitors and follicular helper T cells that

express RANKL

The elevated expansion of putative osteoclast progenitors in sanroque mutant mice

could be contributed to by the elevated expression of RANKL in the bone of sanroque

mutant mice, which has been documented in the previous chapter. TFH is a newly

characterised T cell subset that has been implicated in the pathogenesis of

autoimmunity. Preliminary microarray analysis by Vinuesa and colleagues (2005)

showed that sanroque mutant TFH expressed higher levels of RANKL in comparison to

wildtype TFH (Vinuesa et al., 2005a). This prompted us to further postulate that the

mutation in Roquin gene resulted in elevated RANKL in mutant TFH, which then

contributes to the elevated osteoclastogenesis in sanroque mutant mice.

In an attempt to test this hypothesis, we performed Q-PCR on sorted TFH from

wildtype and sanroque mutant spleens. While it was possible to attain a reasonable

number of TFH from sanroque mutant mice, the population of the aforementioned cell

type was so rare (< 1% of CD4 T cells) in the wildtype mice that it was not possible to

isolate sufficient RNA to perform a direct comparison with the expression of RANKL

in mutant TFH (Figure 8.6 A). We also endeavoured to perform cryosectioning and

immunofluorescence staining on non-decalcified bone samples to investigate the

potential localisation of the cell subset that contributed to the abundance of RANKL in

sanroque mutant bone. However, this experiment was unsuccessful due to the lack of

appropriate equipment and facilities in our laboratory.

Since it has been previously shown, and we have confirmed using flow cytometry, that


193

most TFH reside within the germinal centres of the spleens, we performed

cryosectioning and immunofluorescence staining on frozen spleens to assess the source

of RANKL in this organ. Briefly, harvested spleens from sex and age matched mice

were immersed and allowed to equilibrate in O.C.T. compound (Tissue-Tek) before

snap freezing in a beaker full of isopentane in a liquid nitrogen bath until frozen. Frozen

spleens were cryosectioned at a thickness of 5 μm and sections were placed on charged

slides before proceeding to blocking steps crucial for immunofluorescence staining.

Sections were probed with antibodies specific for CD4 and CXCR5 to identify TFH,

RANKL and DAPI and visualised under confocal microscopy. Images were captured

around the white pulp (characterised by circular densely populated area) at ×20 and ×30

magnifications while retaining similar voltages and gain settings throughout wildtype

and mutant samples.

As shown in Figure 8.7, RANKL was mainly expressed by cells localised around the

periphery of the white pulp, i.e. the red pulp area of the spleen, while the expression of

CD4 and CXCR5 was spread out in both areas. Consistent to the observation reported

by Vinuesa et al. (2005), the density of CD4 CXCR5 (markers of TFH) double positive

cells is increased due to the dysregulated expansion of the T cell subset in sanroque

mutant mice. Furthermore, we observed co-localisation of RANKL and CD4 CXCR5

double positive T cells (indicated by white arrow) in sanroque mutant spleen. It is also

interesting to note that under the same image settings, RANKL expression appeared

more intense in the sanroque mutant section in comparison to wildtype. Many of these

RANKL-expressing cells also express CD4. This data suggests that not only sanroque

TFH but also other T cells within the mutant spleen expressed high levels of membrane-

bound RANKL.

This observation is supported by western blotting whereby total spleen of sanroque

mutant mice expressed higher levels of RANKL protein when compared to that of

wildtype control (Figure 8.8 A). Q-PCR analysis further showed that both RANKL and

M-CSF gene expression was elevated in sanroque mutant spleen compared to wildtype

(Figure 8.8 B). Although the expression level of OPG was also significantly higher in

sanroque spleens than wildtype spleen, the RANKL:OPG ratio is still significantly

higher in sanroque mutant mice (Figure 8.8 C). The increase in M-CSF and RANKL

may in turn contribute to the increase in common myeloid progenitors in sanroque


194

mutant spleen as shown in the flow cytometric analysis of splenocytes (Figure 8.9).

However, more biological repeats are required to validate the flow cytometric studies on

the population of osteoclast precursors in mutant splenocytes.

In summary, sanroque mutation leads to increased RANKL expression not only within

the bone but also the spleen, both of which are organs of haematopoiesis. The increase

in M-CSF and RANKL gene expression in sanroque mice in turn contributes to the

elevated number of common myeloid progenitors and accelerated osteoclastogenesis

respectively.

8.3. Discussion

In an attempt to elucidate the pathogenesis of systemic bone loss during autoimmunity,

we investigated the possible osteoimmunological mechanisms that contribute to the

accelerated osteoclastogenesis in sanroque mutant mice. The novel field of

osteoimmunology highlights that the bone marrow reservoir provides a conducive

environment for immune cells to interact with osteoblasts, osteoclasts and their

precursors (Takayanagi, 2007). RANKL is not solely expressed by osteoblasts, but also

by certain activated CD4 T cells, including the proinflammatory Th17 subset (Sato et

al., 2006). Furthermore, the interaction between RANKL-bearing T cells with osteoclast

precursors is reported to drive osteoclast development and induce bone loss during

chronic immune activation (Kong et al., 1999a; Kotake et al., 2005; Kotake et al., 2001;

Pettit et al., 2001; Sato et al., 2006).

Our data now implicate the TFH cells as a novel T cell osteoclastogenic subset that may

contribute to the induction of bone loss by expressing RANKL during pathological

conditions of chronic autoimmune disease. TFH are the class of effector TH cells that

regulate the in vivo development of antigen-specific B cell immunity (Fazilleau et al.,

2009). As highlighted in sanroque mutant mice, the dysregulation of TFH is implicated

in autoimmune diseases that are partially mediated by autoantibodies, thus affirming its

role in humoral immunity as a professional B cell helper (Simpson et al., 2010; Vinuesa

et al., 2005a; Vinuesa et al., 2005b). TFH express CXCR5 and abundant amounts of IL-

21 to directly regulate B cell proliferation and class switching (Akiba et al., 2005;

Nurieva et al., 2008; Rasheed et al., 2006; Terrier et al., 2012; Vinuesa et al., 2005b;

Vogelzang et al., 2008).


195

A recent study has associated IL-21, which is elevated in the serum of sanroque mutant

mice, with the pathogenesis of bone loss in patients with RA and in mice with collagen-

induced arthritis. IL-21 augments osteoclastogenesis by inducing RANKL expression in

CD4 T cells, thus confirming osteoclastogenic activity of the cytokine (Kwok et al.,

2012). As such, aside from TFH we also observed increased co-localisation of RANKL

and CD4 in the spleen of sanroque mutant mice. The increase in RANKL expression

was validated by western blotting and Q-PCR, and this was accompanied with increased

M-CSF expression. This attributed to the elevated expansion of common myeloid

progenitors in the spleen of sanroque mutant mice. The subpopulation of osteoclast

progenitors within the common myeloid progenitors within the spleen has yet to be

determined and should be analysed in the future.

On a side note, it would be crucial to validate the expression of RANKL in TFH using a

more accurate tool such as Q-PCR. Since TFH are activated to proliferate during an

immune response, the rare population of TFH in wildtype mice could be boosted by

vaccination (Ma et al., 2014). In vitro co-culture between TFH and BMMs would also

provide a useful insight into the crosstalk and the possible mechanisms by which the T

cell subset supports osteoclastogenesis.

The crosstalk between immune cells and bone cells in mice is also emphasised by the

CD27 interaction with its ligand CD70. CD27 is a member of TNF-receptor superfamily

of co-stimulators and is constitutively expressed on CD4 and CD8 T cells. Expression

of CD27 is also found on murine haematopoietic stem cells and differentiated

precursors including the earliest lymphocyte precursors as well as common lymphoid

progenitors (Igarashi et al., 2002; Nolte et al., 2005; Serwold et al., 2009; Wiesmann et

al., 2000). The ligand for CD27, CD70, is also expressed on T, B and dendritic cells.

CD27-CD70 signalling plays a crucial role in the regulation of cellular immune

responses, whereby the fine balance between CD27-CD70 interaction is crucial in

preventing autoimmunity (Coquet et al., 2013; Nolte et al., 2009; Wang et al., 2013). In

the context of osteoimmunology, CD70-bearing immune cells infiltrate the bone

marrow to inhibit osteoclastogenesis and favour the differentiation of common myeloid

progenitors towards dendritic cells (Xiao et al., 2013). Our data has suggested that CD4

T cells are a major contributor of CD70-bearing cells in murine bone marrow, which is

significantly reduced in sanroque mice. As a result, the population of dendritic cells in


196

sanroque mice is diminished, and the bifurcation of the common myeloid cells was

driven towards the pre-committed osteoclast precursor population. Consistently,

sanroque mutant bone marrow showed a significant expansion of committed osteoclast

precursors.

The relocation of HSCs into the bone marrow at birth establish the bone organ as the

primary site for haematopoiesis in adult mammals (O'Neill et al., 2011). It is interesting

to note that the non-myeloid/lymphocyte population in sanroque mutant mice appeared

atypical in comparison to wildtype, whereby the cells are skewed towards CD117+ and

Sca-1+ population. This preliminary observation indicated that the haematopoiesis in

sanroque mice might be compromised. The abnormality is also apparent in preliminary

analysis of blood films of sanroque mutant mice with an increased number of immature

cells (left-shift) (Supplementary Figure 2, personal communication with W/Prof. Wendy

Erber); however, this requires further validation using appropriate differential counts to

quantify the different types of leukocytes. Additionally, the stem cell population of

sanroque mice also needs to be further dissected with flow cytometry using appropriate

antibodies to determine lineage negative cells and analyse different subpopulations of

hematopoietic stem cells including long term and short term HSCs. Perhaps the

mutation in the E3 ubiquitin ligase Roquin may affect the stem cell niche of sanroque

mutant mice, and the observations reported in this thesis are the downstream effects of

dysregulated differentiation of myeloid and lymphocytic progenitors. Indeed several E3

ubiquitin ligases have been implicated in fine-tuning haematopoiesis (Rathinam et al.,

2008; Rybak et al., 2009; Thompson et al., 2008; Yokomizo et al., 2008).

As previously mentioned, B cells is implicated in contributing a significant amount of

OPG to regulate bone remodelling (Li et al., 2007b). It is also interesting to note the

diminished population of B cells in sanroque mutant bone marrow. This data implies

that the reduction of OPG levels in the sanroque mutant mice is probably due to the

absence of B cells within the bone marrow reservoir. Taken together, the results in this

chapter demonstrate that sanroque mutant mice exhibit diminished CD27-CD70

interaction to favour the differentiation of common myeloid progenitors towards

osteoclast precursors. The mutation in Roquin gene resulted in the augmented

expression of RANKL not only in bone but also in the spleen of sanroque mutant mice.

Using confocal microscopy, much of the RANKL expression observed is contributed by


197

CD4 T cells, some of which are TFH, in sanroque mutant spleen. The increased

RANKL expression is compounded by the increase in MCSF expression and the

reduction in OPG level, resulting in enhanced osteoclast differentiation and function,

thus contributing to systemic bone loss observed in sanroque mutant mice.


198


CD

11b

CD3ε/CD45R/B220

yp q Single live cells

Wildtype

sanroque


199

(B) Wildtype sanroque Single live cells

CD45R/B220

Wildtype

sanroque


200

Figure 8.1 The bone marrow population of sanroque mutant mice appeared

dysregulated. Flow cytometric analysis of bone marrow cells of mice of the indicated

genotype. A minimum total of 100,000 events were recorded and only live mononuclear

cells were analysed using antibodies indicated at the axes. (A) An overview of gating

strategy to gate monocytic cells and lymphocytes. Note the missing population of

CD11bhi

in quadrant 3 (Q3) of sanroque mutant mice flow profile. (B) Further analysis

revealed that the missing population in sanroque mutant bone marrow is

CD45R/B220hi

. (C) Flow cytometric analysis of Sca1 and CD117 (or c-Kit) expression

on gated CD3ε- CD45R/B220

- CD11b

-. Data is representative of at least three

independent biological repeats. Bar chart represents mean percentage and each dot

represents each biological repeat. * p<0.05; ** p<0.001; n.s. = no significance in

comparison to wildtype controls.

Wildtype

sanroque

(C) Wildtype sanroque LM-

Sca1

CD117

LM


201

Figure 8.2 Increased macrophage and reduced dendritic cell population in

sanroque bone marrow. Flow cytometric analysis of (A) macrophage and (B) dendritic

cell population of bone marrow cells. (C) Representative bar charts of indicated cell

population as percentage of live single cells. Data is representative of at three

independent biological repeats. * p<0.05; ** p<0.001; n.s. = no significance in



CD

11b

CD115

Mϕ

CD11blo

CD115+

Wildtype sanroque yp q

CD

115

CD11c

Lin-

DCs

CD115-

CD11c+

(C)

Wildtype

sanroque

Mϕ DCs

(B)

Lin- CD117

+


202

Figure 8.3 Significant expansion of putative osteoclast progenitors in sanroque

bone marrow. (A) Flushed bone marrow cells were co-stained for RANK and CD115

(A) Wildtype sanroque Lin- CD117

+ CD11b

lo

CD115+

RANK+

CD115

Wildtypesanroque

(B) Wildtype sanroque Lin-CD117

+CD115

+

CD27

CD

11b

Wildtype

sanroque

yp q

CD27lo

CD27hi


203

to identify committed osteoclast precursors, which is increased in sanroque bone

marrow. (A) Pre-committed osteoclast precursors (Lin- CD117

+ CD115

+ CD11b

lo

CD27lo

), which have higher osteoclastogenic potential in comparison to OC/DC

common precursors (Lin- CD117

+ CD115

+ CD11b

lo CD27

hi) is also significantly higher

in sanroque bone marrow. Results are mean percentage of populations from three

independent biological repeats. * p<0.05; ** p<0.001; n.s. = no significance in



204

Figure 8.4 No difference in the percentages of CD4 and CD8 T cell population

between the bone marrow of sanroque mutant mice and wildtype. Flow cytometric

analysis of (A) CD4 and (B) CD8 T cells of wildtype and sanroque bone marrow. (C)

Bar charts represent mean percentage of populations from three independent biological

repeats. * p<0.05; ** p<0.001; n.s. = no significance in comparison to wildtype

controls.

(A) Wildtype sanroque Single live cells

CD

3ε

CD4

yp q g

CD3ε+

CD4+

(B) Wildtype sanroque Single live cells

CD3ε+

CD8+

CD

3ε

CD8

Wildtype

sanroque

(C)


205

(A) Wildtype

CD3εCD4CD8

sanroque

CD3εCD4CD8

CD70

Wildtype

sanroque

(B) Wildtype sanroque CD4+ CD4

CD70

CD70- CD70

+


206

Wildtype

sanroque

(C) Wildtype sanroque CD8+ yp q CD8

CD70- CD70

+

CD70

(D) Wildtype sanroque CD3ε- CD45R/B220

+

CD70

Wildtype

sanroque

CD70- CD70

+


207

Figure 8.5 Expression of CD70 in T, B and DCs. Representative of flow cytometric

analysis of CD70 expression in (A) all T cells, (B) CD4 T cells, (C) CD8 T cells, (D) B

cells, (E) DCs of wildtype and sanroque bone marrow. Results are representative of

three independent biological repeats and are presented as mean percentage. * p<0.05; **


(E) Wildtype sanroque

CD70

CD11c+

CD70-

Wildtype

sanroque

CD70+


208

Figure 8.6 Increased follicular helper T cell in sanroque mutant mice. (A)

Representative flow cytometric analysis of CD4+ CD44+ CXCR5+ PD1+ TFH in

wildtype and sanroque spleens. (B) Average percentage of TFH in CD4 T cells. Results

are representative of three independent biological repeats and are presented as mean

percentage. * p<0.05; ** p<0.001; n.s. = no significance in comparison to wildtype

controls.

(A) Wildtype sanroque CD4+ CD44

+

PD-1

CXCR5

Wildtype

sanroque

(B)


209

Wildtype sanroque

×

×

×

p

×

×

×


210

Figure 8.7 Localisation of RANKL and TFH in spleen. Spleens were cryosectioned,

fixed and processed for immunofluorescence using specific antibodies to RANKL

(FITC), CD4 (DyLight-488) and CXCR5 (APC). Nucleuses were stained with DAPI

and white pulp (WP) is demarcated by the dotted white line. Areas of co-localisation

between CD4 and CXCR5 double positive cells with RANKL is denoted by yellow

colour in overlay with specific area indicated by the white arrow. Images are

representative of three independent biological repeats.

Wildtype sanroque p

DAPI RANKL (FITC) CD4 (DyLight488) CXCR5 (APC)


211

Figure 8.8 RANKL protein and gene expressions are elevated in sanroque spleens.

(A) Cell lysates were collected by homogenising spleen tissue and were subjected to

immunoblotting. Detection of RANKL and β-actin in the spleens of wildtype and

sanroque mutant mice. (B) Total RNA was harvested and subjected to RT-QPCR using

primers specific for OPG, RANKL and M-CSF. Gene expression analysis is expressed

relative to the house keeping gene Rn18S. Results are representative of two biological

repeats and are presented as mean ± SEM. * p<0.05; ** p<0.001; n.s. = no significance

in comparison to wildtype controls.

RANKL

β-actin

Wildtyp

e 1

sanroq

ue 1

Wildtyp

e 2

sanroq

ue 2

RANKL:β-actin 1.012 1.394 0.371 1.484

(B) (C)

Wildtype sanroque


212

Figure 8.9 Increased common myeloid progenitors in sanroque spleen. (A)

Representative flow cytometric analysis of splenocytes of mice of the indicated

genotype. (B) Bar charts representing mean percentage of common myeloid progenitors

in total splenocytes. Data is representative of two independent biological repeats and are

presented as mean percentage of single live cells.


CD

11b

CD45R/B220

CD3ε- CDCD33εε

Wildtype

sanroque

(B)

CHAPTER NINE

GENERAL OVERVIEW AND FUTURE DIRECTIONS

Chapter 9: General Overview and Future Directions

214

9. CHAPTER 9: GENERAL OVERVIEW AND FUTURE DIRECTIONS

9.1. Overview of thesis and general discussion

The homeostasis of the bone organ system is maintained by a process known as bone

remodelling. The bone remodelling process is driven by the coupled activity between

bone-resorbing osteoclasts and bone-forming osteoblasts. The cellular coupling between

osteoclastic and osteoblastic cells is heavily coordinated by systemic and local

regulatory systems to keep both remodelling and resorption processes synchronised

(Martin et al., 2006; Morgan et al., 2008). Many osteoimmunological studies on

inflammatory-related bone diseases have signified the dialogue between the skeletal and

the immune system. This is inevitable as spawned within the bone cavity are the

marrow stem cells. Whereas some mesenchymal stem cells differentiate into

osteoblastic and osteocytic cells, haematopoietic stem cells bifurcate into myeloid

progenitors, which give rise to macrophages, dendritic cells and osteoclasts, and

lymphoid progenitors that develop into lymphocytes and other immune cells (Arai et

al., 1999; Miyamoto et al., 2001; Pittenger et al., 1999). Thus, the development of bone

cells and immune cells occurs within the same microenvironment.

Furthermore, osteoclasts and immune cells share several regulatory molecules such as

cytokines, receptors, transcription factors and signalling molecules (Takayanagi, 2007).

Among these, RANKL demonstrates overlapping roles as the cytokine that is essential

for both osteoclast differentiation/function as well as immune development (Kong et al.,

1999a; Kong et al., 1999b). Under homeostatic conditions, RANKL is expressed by

osteoblasts and osteocytes to regulate osteoclast differentiation and activity (Lacey et

al., 1998; Nakashima et al., 2011). Conversely, immune cells also produce RANKL

during chronic immune activation or autoimmunity, thereby aggravating pathological

bone loss and remodelling (Kong et al., 1999a; Sato et al., 2006; Walsh et al., 2006).

Therefore, osteoimmunological interactions become a key player in the maintenance of

bone homeostasis and in the pathogenesis of systemic bone loss in autoimmune diseases

such as rheumatoid arthritis, spondylarthritis (Goldring, 2013), and systemic lupus

erythematosus (SLE) (Tang et al., 2013a). These disease conditions entail a risk of bone

fractures, causing significant discomfort and deformities. However, the exact

mechanism(s) underlying these pathological conditions remain poorly understood. Thus

understanding the molecular mechanisms involved in regulating the cross talk between

the bone organ and immune system could potentially provide an insight for the

development of novel therapeutic targets for these osteoimmune diseases.


215

In an effort to identify novel factors that play significant roles in the maintenance and

regulation of the osteoimmune system, we systematically screened a chemically induced

(ENU) mutagenesis library for bone phenotypes. Using X-ray imaging, we discovered

the ENU-induced sanroque mutant mice line that exhibits reduced bone mass in the

hindlimbs, vertebrae and tail. Genetic mapping revealed that the mutant mouse line

carried a single point mutation in the Roquin (Rc3h1) gene, resulting in a methionine to

arginine substitution in the ROQ domain of the encoded protein. To the best of our

knowledge, this is the first study to document the novel role of Roquin in bone biology.

The structure of Roquin protein is characterised as a RING-type E3 ubiquitin ligase. An

E3 ubiquitin ligase, along with E1 ubiquitin activating enzyme and E2 ubiquitin

conjugating enzyme, is one of the enzymatic components that drive the ubiquitin

proteasome system (UPS). UPS is an ATP-dependent enzymatic cascade traditionally

described to be involved in the maintenance of protein homeostasis by marking the

protein of interest with a chain of ubiquitin (ubiquitination or ubiquitylation) so as to be

recognised by the proteasome for proteolysis. However, the role of UPS has extended

from just protein degradation, and is reported to be involved in transcription factor

activation, DNA repair, translational control, kinase activation, histone activity and

endocytosis (Pickart et al., 2004).

The sanroque mutant mouse line was first characterised by Vinuesa and colleagues in

2005 to display an autoimmune disease consistent to SLE, with splenomegaly and

lymphomegaly (Vinuesa et al., 2005a). Roquin is a repressor of inducible costimulator

(ICOS), a crucial co-stimulator receptor for TFH. As a result of the loss of function

mutation in Roquin, ICOS levels in sanroque mutant mice are elevated to promote

uncontrolled expansion of TFH, excessive production of IL-21 and high titres of

autoantibodies (Yu et al., 2007). Molecular studies revealed that Roquin represses

ICOS at the post-transcriptional level, whereby the ROQ protein domain binds directly

at the 3’ untranslated region (3’ UTR) of ICOS mRNA (Athanasopoulos et al., 2010;

Glasmacher et al., 2010). Roquin gene thus plays a regulatory role in immune

homeostasis in recognising self from non-self, thereby preventing self-reactive T cells

from providing assistance to B cell maturation and production of self-reactive

antibodies.

This study further extends the immunological role of Roquin to encompass its role in


216

the maintenance of bone homeostasis. MicroCT analysis revealed the reduction in mass

and deterioration in microarchitecture of the bone in sanroque mutant mice.

Histomorphometric analysis further confirmed the osteopenic bone phenotype and

provided some hints into a possible cellular mechanism that leads to the bone loss. The

increase in osteoclast number in vivo is evident in sanroque bone histology and this

correlated with the increased expression levels of osteoclast gene markers TRAP and

Cathepsin K. Unexpectedly, quantification of bone formation rate using double

fluorochrome labelling revealed impaired mineral apposition despite the slight increased

in osteoblast number and osteoid surface in vivo. These observations suggest that the

increased amount of osteoid surface in sanroque mice did not mineralise in a timely

manner, a hallmark of osteomalacia (Maricic, 2008). Consistently, Q-PCR analysis

revealed the significant reduction in osteocalcin expression, which plays a crucial role

during osteoblastic mineralisation (Neve et al., 2013), in the total bone of sanroque

mutant mice.

In an attempt to understand the cellular mechanism through which Roquin regulates

bone homeostasis, we performed a series of in vitro experiments to characterise

osteoblast and osteoclast activities in sanroque mutant mice. In vitro bone nodule

formation analysis was conducted by stimulating calvarial (COBs) and long bone

isolated osteoblasts (LBOBs) with osteogenic medium, which revealed that sanroque

mutant COBs and LBOBs responded to osteogenic induction differently. COBs isolated

from sanroque mutant mice failed to form bone nodules, whereas sanroque LBOBs

displayed a greater area of calcium deposition in comparison to wildtype LBOB culture.

The accelerated activity of sanroque LBOB correlates with the elevated expression of

alkaline phosphatase, osteocalcin, osteopontin and the transcription factor osterix.

However, the augmented calcium deposition by sanroque LBOB generated bone

nodules that appeared morphologically different from that of wildtype. This observation

warrants further investigation

The abnormality of sanroque mutant osteoblast activity is further shown by the

response of COBs and LBOBs toward BMP2, a key activator of bone formation (Rosen,

2009). BMP2 was not able to rescue the aberrant bone formation by sanroque COBs.

The insensitive response towards BMP2 is also observed in sanroque LBOBs. These

data suggest that either the receptor of BMP2 is dysfunctional in mutant osteoblasts, or


217

that the BMP2 signalling pathway may be altered due to the sanroque mutation. In

future studies, it would be useful to investigate the role of Roquin in BMP2 signalling.

The analysis of other key osteoblast gene markers such as the transcription factor Runx2

and collagen I would also provide valuable insight into the molecular mechanism

through which Roquin regulates osteoblastogenesis and mineralisation; respectively.

While Q-PCR analysis of the total bone revealed that RANKL:OPG is elevated in

sanroque mutant mice, co-culture experiments showed that sanroque mutant LBOBs

have reduced capability to support physiological osteoclastogenesis, implying that

mutant osteoblasts may not be the contributor of the elevated RANKL in vivo. This

experiment further indicated the atypical functions of mutant osteoblasts and further

highlighted the role of Roquin in osteoblast biology.

In vitro osteoclastogenesis assay revealed increased RANKL-induced osteoclast

formation in cell cultures derived from sanroque mutant BMMs. The accelerated

osteoclastogenesis observed in sanroque cell cultures was supported by the elevated

gene and protein expression of NFATc1, DC-Stamp, v-ATPase V0 subunit d2. BMMs

from sanroque mutant mice also appeared to be more sensitive towards RANKL

stimulation, as they were able to form osteoclasts at lower doses of RANKL. This

suggests that either the RANKL/RANK signalling pathways are elevated in sanroque

BMM, or that there may be an increase in osteoclast progenitors among BMM, which

are primed to form osteoclasts. In vitro bone resorption assays highlight the overzealous

activity of sanroque mutant osteoclasts, as each mutant multinucleated cell resorbed a

larger area of bone. The increased bone lytic activity is further supported by the

elevated cathepsin K secretion into the supernatant of sanroque mutant osteoclast

cultures when stimulated with bone powder. Consistent with the increased osteoclast

numbers observed in vivo, these in vitro observations showing the increase

osteoclastogenesis and lytic activity may partially attribute to the decreased bone mass

of sanroque mutant mice.

As aforementioned, it can be hypothesised that the increased osteoclastogenesis and

bone lytic activity in sanroque BMM and osteoclast cell culture could be due to the

increase in RANKL/RANK signalling pathway. The RANKL/RANK interaction

resulted in the posttranslational modification of the adapter protein TRAF6, reportedly

via ubiquitination, to activate six major signalling pathways: NFATc1, NF-κB,


218

Akt/PKB, JNK, ERK and p38, which then direct osteoclast-specific gene expression

leading to differentiation and function (Boyle et al., 2003; David et al., 2002; Feng,

2005; Gohda et al., 2005; Kobayashi et al., 2001; Lamothe et al., 2007; Li et al., 2002;

Takayanagi et al., 2000; Xing et al., 2002). Since Roquin was characterised as an E3

ubiquitin ligase, we conducted immunoprecipitation experiments that revealed no

difference in the ubiquitination level of TRAF6 between RANKL-stimulated wildtype

and sanroque BMM culture, suggesting that Roquin does not play a role in the

ubiquitination of TRAF6. Intriguingly, the levels of IκB-α phosphorylation in RANKL-

stimulated sanroque mutant BMMs remained higher, which equates to the inactivation

of NF-κB dimers to prevent the translocation to the nucleus and subsequent gene

transcription (Oeckinghaus et al., 2011). This suggests that Roquin may play a

regulatory role downstream of NF-κB canonical signalling, or that it may be involved in

other RANKL/RANK signalling cascades. Conversely, immunoblotting assay showed

that the downstream NFATc1 signalling is increased in RANKL-stimulated sanroque

mutant BMM, along with the phosphorylation of p38 and ERK1/2. However, other


examined. An in depth analysis of the complete RANKL/RANK signalling pathway

array is certainly required to identify the exact molecular pathway through which

Roquin affects osteoclastogenesis. Since the role of Roquin in regulating mRNA

homeostasis has been reported, it would also be worthwhile to do post-transcriptional

studies using tools such as microarrays and ribonomics.

Keeping in mind that sanroque mutant mice exhibit a Lupus-like autoimmune disease,

we explored the possible osteoimmunological interactions by performing co-culture

between BMM supplemented with mouse blood sera. Interestingly, medium

supplemented with sanroque mutant mouse sera elevates the proliferation of both

wildtype and sanroque BMMs. The levels of proliferation-inducing cytokines such as

GM-CSF and IL-6 in the blood serum should be measured as a part of future studies.

MTS assay further revealed that sanroque BMMs proliferate faster in the presence of

M-CSF. These data show that sanroque BMMs are probably more sensitive towards M-

CSF stimulation, or that they may have higher number of cells expressing M-CFS

receptor, c-fms, within the BMMs population of sanroque mice. Investigation of the M-

CSF downstream signalling pathway, such as ERK and AKT, in sanroque mutant

BMMs may provide the underlying molecular pathway in which Roquin is involved.


219

Nevertheless, a role for Roquin in osteoclasts has been established.

As previously hypothesised, increased putative osteoclast progenitors in sanroque

mutant mice may account for the increased osteoclast differentiation and activity in vivo

and in vitro. Subsequently, flow cytometry analysis revealed the significant expansion

of CD3e/CD45R/B220- CD11b+ cells, reported to be typical of cells undergoing

osteoclastogenesis (Jacquin et al., 2006), in the bone marrow of sanroque mutant mice.

Further characterisation of this population revealed increased populations of pre-

committed osteoclast precursors Lin- CD117+ CD115+ CD11blo CD27lo and the more

committed osteoclast precursors Lin- CD117+ CD11blo CD115 (or c-fms) and RANK

double positive cells (Xiao et al., 2013), suggesting that the bone marrow of sanroque

mutant mice are primed to differentiate into osteoclast progenitors.

The priming effect could first be initiated by the reduction of CD70, the ligand for

CD27 in the sanroque mutant mice. CD27/CD70 interaction is important in fine-tuning

the bifurcation of myeloid progenitors and inhibiting osteoclastogenesis to favour the

differentiation towards dendritic cells (Xiao et al., 2013). Additional flow cytometric

analysis revealed that CD4 T cells are the major CD70-bearing cells in the bone marrow

of wildtype mice, which is diminished in sanroque mutant mice. The reduced

expression of CD70 is compounded by the increased RANKL expression in sanroque

mouse, thus providing a more osteoclastogenic environment for the differentiation of

progenitor cells towards osteoclasts. OPG, the decoy receptor of RANKL, first

characterised to be expressed by osteoblasts to inhibit osteoclastogenesis, is also

reported be abundantly expressed by B cells, as knockout mice lacking B cells exhibit

osteoporosis (Li et al., 2007b). Sanroque mutant bone marrow showed a reduction in

CD45R/B220+ B cells, which may explain the increased RANKL:OPG ratio in

sanroque mutant mice.

Using Q-PCR analysis, elevated RANKL expression was observed in both the total

bone and the spleens of sanroque mutant mice. Western blotting further confirmed the

protein expression in the spleens of the sanroque mutant mice. For the first time, we

showed through confocal analysis that the expression of membrane-bound RANKL co-

localised with CD4 and CXCR5 cell surface markers, both of which are markers for

TFH, implicating TFH as a potential osteoclastogenic T cell subset that drives systemic

bone loss in pathological conditions.


220

9.2. Future Directions

The results from this study have underlined several areas that require further

investigation. Addressing these areas will provide deeper understanding into the

molecular mechanisms by which Roquin regulates osteoimmunological interactions

while further establishing the roles of Roquin in bone biology.

The role of Roquin in osteoimmunology and in the pathology of inflammatory

bone loss

Systemic bone loss is common amongst SLE patients; not only due to prolonged

administration of drugs, but the disease per se contributes to the deterioration of bone

microarchitecture in these patients (Tang et al., 2013a). Exacerbated localised and

systemic inflammation due to elevated pro-inflammatory cytokines results in

decoupling of bone remodelling to favour bone loss in lupus patients (Tang et al.,

2013b). The different cytokenic composition of sanroque blood sera is evident in the

co-culture with BMMs. However, a thorough analysis of serum soluble factors using

sensitive tools such as ELISA or multiplex measurements with a cytometric bead

approach is crucial in determining levels of pro-inflammatory cytokines as well as

RANKL and OPG.

The cross-regulation between the bone system and immune system means that there is

also a possibility whereby the autoimmunity is developed due to dysregulated bone

microenvironment. To determine whether the dysregulation in bone homeostasis

resulted in an autoimmune disease, a refined strategy using conditional gene

inactivation in osteoclasts or osteoblasts would be useful. Relying on DNA

recombinase, two recognition (loxP) sites are inserted to excise the flanked (floxed)

gene, i.e. Roquin, through Cre-mediated recombination expressed under a promoter

specific to the cell type, such as calcitonin receptor and osteopontin, which are specific

to osteoclasts and osteoblasts respectively (Friedel et al., 2011).

An alternative approach is to perform bone marrow transplantation, whereby wildtype

and sanroque mutant mice are irradiated and re-transplanted with sanroque and

wildtype bone marrow respectively. This method would be useful to investigate if the

existing osteopenic and autoimmune phenotypes of sanroque mutant mice could be

rescued by the bone marrow of wildtype mice. If the mutation in Roquin gene resulted

in the alteration of transplanted wildtype bone marrow, this may suggest that the


221

existing bone microenvironment may play a role in the pathogenesis of inflammatory-

related bone loss in sanroque mutant mice.

The collagen-induced arthritis (CIA) model has been extensively used to mimic human

rheumatoid arthritis (RA) in a quest to investigate potential mechanisms and to identify

potential targets for therapy. Performing CIA on wildtype and sanroque mutant mice

may provide insights to the osteoimmunological role of Roquin and perhaps further

highlight the osteoclastogenic potential of TFH in inflammatory bone loss.

Localisation of RANKL-bearing cell in sanroque mutant bone

Q-PCR analysis has revealed the significant increase in RANKL expression in the total

bone of sanroque mutant mice. In physiological conditions, RANKL is expressed by

osteoblasts to recruit osteoclast progenitors and induced osteoclastogenesis (Lacey et

al., 1998). However, in vitro osteoblast and BMM coculture revealed the impeded

ability of the mutant osteoblast to support osteoclastogenesis. Thus, the source of the

osteoclastogenic factor has yet to be determined. Imaging-flow cytometric analysis and

immunohistochemistry would be able to provide the identity of the RANKL-bearing

cell in the bone of sanroque mutant mice.

The effect of mutation in Roquin on stem cell niche of sanroque bone marrow

Preliminary flow cytometric analysis revealed atypical populations of the stem cell

niche in the mutant bone marrow. This indicates that the mutation in Roquin gene may

result in the alteration of haematopoiesis in sanroque mutant mice. This in turn may

contribute to the upregulation of common myeloid progenitors and putative osteoclast

progenitors observed in sanroque mice. In order to further investigate stem cell

populations within the niche, and determine exactly how the stem cell populations are

altered, flow cytometry with more specific antibodies against lineage negative bone

marrow cell populations should be performed. This includes the long-term

reconstituting HSCs (LT-HSC), short-term HSCs (ST-HSC) and multipotent

progenitors (MMP) which give rise to common lymphoid progenitors (CLP), B

cell/macrophage bipotent progenitors and common myeloid progenitors.

The role of Roquin in RANKL and BMP2 signalling pathways in sanroque

mutant mice



222

examined. In depth analysis of the complete RANKL/RANK signalling pathway array

behind the regulation of osteoclast formation and function by Roquin is certainly

required to identify the exact molecular pathway through which Roquin affects

osteoclastogenesis.

A recent study reported that BMP2 induced-bone marrow-derived mesenchymal stem

cells (BMMSCs) from SLE patients showed reduced capacity for osteogenic

differentiation due to repressed BMP/Smad pathway (Tang et al., 2013b). This suggests

that repressed BMP/Smad signalling pathway and impaired osteoblastic differentiation

may participate in the pathology of osteoporosis in SLE patients. Bearing in mind the

lupus like autoimmune disease, the sanroque mice emerge as a useful model to

investigate the molecular mechanisms involved in osteoblastogenesis in SLE patients.

Furthermore, our in vitro investigation indicated the atypical response of sanroque

mutant osteoblasts towards BMP2. Therefore, BMP2 signalling analysis would provide

insight into possible mechanisms behind the regulation of osteoblast differentiation and

activity by Roquin. These experiments are currently on going, with the results showing

altered phosphorylation of Smad 1, 5, 8 in sanroque mutant LBOBs.

Since the role of Roquin as an E3 ubiquitin ligase in regulating mRNA homeostasis has

been reported (Yu et al., 2007), it would also be worthwhile to do post-transcriptional

studies using tools such as microarrays and ribonomics.

Confirming TFH as a potential osteoclastogenic T cell subset

Several osteoimmunology studies have reported the expression of RANKL by T cells to

directly induce in vitro osteoclastogenesis (Horwood et al., 1999; Kong et al., 1999a;

Kong et al., 1999b; Kotake et al., 2001; Weitzmann et al., 2001; Wong et al., 1997b).

Using confocal analysis, we have shown that CD4 and CXCR5 double positive cells

express RANKL in sanroque mutant spleen, indicating that TFH could potentially be a

new member of the osteoclastogenic T cell subset along with TH17. However, such

observation has to be further validated by sorting TFH to analyse the expression of both

RANKL and OPG. Since the proliferation and differentiation of TFH is induced during

an immune response, vaccination maybe required to boost the rare population of TFH in

wildtype mice (Ma et al., 2014). In vitro co-culture between TFH and BMMs would be

able to further establish the potential osteoclastogenic role of the T cell subset. A few

vaccinated wildtype and sanroque mutant spleens may have to be pooled to perform the


223

in vitro co-culture experiment, which can be conducted as described (Kong et al.,

1999a), while TFH would have to be maintained throughout until the apparent

formation of osteoclasts (Nurieva et al., 2008).

Sample size planning

One of the limitations of this study is regarding the relatively small sample size. One of

the major obstacles in carrying out some experiments was due to insufficient supplies of

mutant mice from Australian National University (ANU) in Canberra, which was partly

owing to lack of breeding colonies. Many of the mice used in this experiment were

ordered as soon as we were informed by ANU on the availability of the breeding

colonies.

Due to logistical challenges, the earliest time point that we can acquire the animal is

usually when the mice turned 100 days old. In most experiments, three biological

repeats were used in this study and were sufficient to show significant difference

between wildtype and sanroque mutant mice. However, future research with larger

sample size should revisit the hypotheses of this study and to address areas mentioned

above, so as to provide robust statistical significance. Further bone characterisation of

the mice at different time points will also provide a clearer picture on the onset of

osteoporosis in sanroque mutant mice.

9.3. Conclusion

A hypothetical model illustrating the role of Roquin in osteoimmunological cellular

interaction in the pathogenesis of bone loss in sanroque mice is shown in Figure 9.1.

Collectively, this thesis has established the reduced bone mass of sanroque mutant

mice. The increased osteoclast number and activity, compounded by abnormality of

osteoblast function, favour bone loss in sanroque mutant mice. This is expedited by

expansion of putative osteoclast progenitors and the increased RANKL expression in

sanroque mutant mice. TFH may play a role in the pathogenesis of systemic bone loss

during chronic autoimmunity by expressing osteoclastogenic factors. The precise

molecular mechanism(s) through which Roquin regulates bone homeostasis remains to

be clarified. Nevertheless, our data demonstrates that the Roquin gene is an important

regulator of bone homeostasis via ENU mutagenesis and the sanroque mouse emerges

as a useful model to investigate the molecular genetics and mechanisms

osteoimmunology.


224

Figure 9.1 The hypothetical model depicting the possible cellular mechanism that

resulted in systemic bone loss in sanroque mutant mice. The mutation in Roquin

gene resulted in the augmented expansion of putative osteoclast progenitors in sanroque

bone marrow cavity. The uncoupled activity between osteoclasts and osteoblasts is

driven by the elevated RANKL expression (perhaps by activated CD4 TFH cells) in

sanroque mutant mice. Furthermore, IL-21, which stimulates RANKL expression in T

cell is elevated in sanroque mice. A major contributor of OPG, mature B cells, is also

diminished within the sanroque bone marrow niche. As a result, osteoclastogenesis in

sanroque mutant mice is elevated to form more osteoclasts with increased bone lytic

activity. This effect is compounded by the abnormal osteoblastic function, thus

favouring bone resorption and leading to the osteoporotic bone of sanroque mice.

CHAPTER TEN

REFERENCES

Chapter 10: References

226

10. CHAPTER 10: REFERENCES

Abdel Meguid MH, Hamad YH, Swilam RS, Barakat MS (2013). Relation of interleukin-6 in rheumatoid arthritis patients to systemic bone loss and structural bone damage. Rheumatol Int. 33(3): 697-703.

Abe K, Hirai M, Mizuno K, Higashi N, Sekimoto T, Miki T, et al. (2001). The YXXQ motif in gp 130 is crucial for STAT3 phosphorylation at Ser727 through an H7-sensitive kinase pathway. Oncogene 20(27): 3464-3474.

Aberle H, Bauer A, Stappert J, Kispert A, Kemler R (1997). beta-catenin is a target for the ubiquitin-proteasome pathway. EMBO J 16(13): 3797-3804.

Adachi T, Alam R (1998). The mechanism of IL-5 signal transduction. Am J Physiol 275(3 Pt 1): C623-633.

Adamopoulos IE, Danks L, Itonaga I, Locklin RM, Sabokbar A, Ferguson DJ, et al. (2006). Stimulation of osteoclast formation by inflammatory synovial fluid. Virchows Arch 449(1): 69-77.

Akhurst RJ, Hata A (2012). Targeting the TGFbeta signalling pathway in disease. Nat Rev Drug Discov 11(10): 790-811.

Akiba H, Takeda K, Kojima Y, Usui Y, Harada N, Yamazaki T, et al. (2005). The role of ICOS in the CXCR5+ follicular B helper T cell maintenance in vivo. J Immunol 175(4): 2340-2348.

Akisaka T, Subita GP, Kawaguchi H, Shigenaga Y (1989). Different tartrate sensitivity and pH optimum for two isoenzymes of acid phosphatase in osteoclasts. An electron-microscopic enzyme-cytochemical study. Cell Tissue Res 255(1): 69-76.

Akiyama T, Bouillet P, Miyazaki T, Kadono Y, Chikuda H, Chung UI, et al. (2003). Regulation of osteoclast apoptosis by ubiquitylation of proapoptotic BH3-only Bcl-2 family member Bim. EMBO J 22(24): 6653-6664.

Al-Dujaili SA, Lau E, Al-Dujaili H, Tsang K, Guenther A, You L (2011). Apoptotic osteocytes regulate osteoclast precursor recruitment and differentiation in vitro. J Cell Biochem 112(9): 2412-2423.

Alberts B (2008). Molecular biology of the cell. 5th edn. Garland Science: New York.

Alexander KA, Chang MK, Maylin ER, Kohler T, Muller R, Wu AC, et al. (2011). Osteal macrophages promote in vivo intramembranous bone healing in a mouse tibial injury model. J Bone Miner Res 26(7): 1517-1532.


227

Allen MR, Burr DB (2014). Bone Modeling and Remodeling. In: Burr DB, Allen MR (ed)^(eds). Basic and Applied Bone Biology, edn: Academic Press. p^pp 75-90.

Alonzi T, Fattori E, Lazzaro D, Costa P, Probert L, Kollias G, et al. (1998). Interleukin 6 is required for the development of collagen-induced arthritis. J Exp Med 187(4): 461-468.

Andersen TL, Sondergaard TE, Skorzynska KE, Dagnaes-Hansen F, Plesner TL, Hauge EM, et al. (2009). A physical mechanism for coupling bone resorption and formation in adult human bone. Am J Pathol 174(1): 239-247.

Anderson DM, Maraskovsky E, Billingsley WL, Dougall WC, Tometsko ME, Roux ER, et al. (1997). A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature 390(6656): 175-179.

Andersson G, Ek-Rylander B, Hollberg K, Ljusberg-Sjolander J, Lang P, Norgard M, et al. (2003). TRACP as an osteopontin phosphatase. J Bone Miner Res 18(10): 1912-1915.

Ang E, Pavlos NJ, Rea SL, Qi M, Chai T, Walsh JP, et al. (2009). Proteasome inhibitors impair RANKL-induced NF-kappaB activity in osteoclast-like cells via disruption of p62, TRAF6, CYLD, and IkappaBalpha signaling cascades. J Cell Physiol 220(2): 450-459.

Arai F, Miyamoto T, Ohneda O, Inada T, Sudo T, Brasel K, et al. (1999). Commitment and differentiation of osteoclast precursor cells by the sequential expression of c-Fms and receptor activator of nuclear factor kappaB (RANK) receptors. J Exp Med 190(12): 1741-1754.

Ardley HC, Robinson PA (2004). The role of ubiquitin-protein ligases in neurodegenerative disease. Neuro-degenerative diseases 1(2-3): 71-87.

Arron JR, Choi Y (2000a). Bone versus immune system. Nature 408(6812): 535-536.

Arron JR, Choi Y (2000b). Osteoimmunology: Bone versus immune system. Nature 408(6812): 535-536.

Asao H, Okuyama C, Kumaki S, Ishii N, Tsuchiya S, Foster D, et al. (2001). Cutting edge: the common gamma-chain is an indispensable subunit of the IL-21 receptor complex. J Immunol 167(1): 1-5.


228

Athanasopoulos V, Barker A, Yu D, Tan AH, Srivastava M, Contreras N, et al. (2010). The ROQUIN family of proteins localizes to stress granules via the ROQ domain and binds target mRNAs. FEBS J 277(9): 2109-2127.

Atsumi T, Ishihara K, Kamimura D, Ikushima H, Ohtani T, Hirota S, et al. (2002). A point mutation of Tyr-759 in interleukin 6 family cytokine receptor subunit gp130 causes autoimmune arthritis. J Exp Med 196(7): 979-990.

Au-Yeung BB, Deindl S, Hsu LY, Palacios EH, Levin SE, Kuriyan J, et al. (2009). The structure, regulation, and function of ZAP-70. Immunol Rev 228(1): 41-57.

Avnet S, Cenni E, Perut F, Granchi D, Brandi ML, Giunti A, et al. (2007). Interferon-alpha inhibits in vitro osteoclast differentiation and renal cell carcinoma-induced angiogenesis. Int J Oncol 30(2): 469-476.

Awane M, Andres PG, Li DJ, Reinecker HC (1999). NF-kappa B-inducing kinase is a common mediator of IL-17-, TNF-alpha-, and IL-1 beta-induced chemokine promoter activation in intestinal epithelial cells. J Immunol 162(9): 5337-5344.

Axmann R, Herman S, Zaiss M, Franz S, Polzer K, Zwerina J, et al. (2008). CTLA-4 directly inhibits osteoclast formation. Annals of the rheumatic diseases 67(11): 1603-1609.

Bab IA, Einhorn TA (1994). Polypeptide factors regulating osteogenesis and bone marrow repair. J Cell Biochem 55(3): 358-365.

Bachmann MF, Wong BR, Josien R, Steinman RM, Oxenius A, Choi Y (1999). TRANCE, a tumor necrosis factor family member critical for CD40 ligand-independent T helper cell activation. J Exp Med 189(7): 1025-1031.

Bakker AD, Klein-Nulend J (2012). Osteoblast isolation from murine calvaria and long bones. Methods Mol Biol 816: 19-29.

Balani D, Aeberli D, Hofstetter W, Seitz M (2013). Interleukin-17A stimulates granulocyte-macrophage colony-stimulating factor release by murine osteoblasts in the presence of 1,25-dihydroxyvitamin D(3) and inhibits murine osteoclast development in vitro. Arthritis Rheum 65(2): 436-446.

Baud CA (1968). Submicroscopic structure and functional aspects of the osteocyte. Clin Orthop Relat Res 56: 227-236.


229

Bendre MS, Gaddy-Kurten D, Mon-Foote T, Akel NS, Skinner RA, Nicholas RW, et al. (2002). Expression of interleukin 8 and not parathyroid hormone-related protein by human breast cancer cells correlates with bone metastasis in vivo. Cancer Res 62(19): 5571-5579.

Bendre MS, Margulies AG, Walser B, Akel NS, Bhattacharrya S, Skinner RA, et al. (2005). Tumor-derived interleukin-8 stimulates osteolysis independent of the receptor activator of nuclear factor-kappaB ligand pathway. Cancer Res 65(23): 11001-11009.

Bendre MS, Montague DC, Peery T, Akel NS, Gaddy D, Suva LJ (2003). Interleukin-8 stimulation of osteoclastogenesis and bone resorption is a mechanism for the increased osteolysis of metastatic bone disease. Bone 33(1): 28-37.

Bertolini DR, Nedwin GE, Bringman TS, Smith DD, Mundy GR (1986). Stimulation of bone resorption and inhibition of bone formation in vitro by human tumour necrosis factors. Nature 319(6053): 516-518.

Bettelli E, Carrier Y, Gao W, Korn T, Strom TB, Oukka M, et al. (2006). Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells. Nature 441(7090): 235-238.

Bhargava U, Bar-Lev M, Bellows CG, Aubin JE (1988). Ultrastructural analysis of bone nodules formed in vitro by isolated fetal rat calvaria cells. Bone 9(3): 155-163.

Blair HC, Zaidi M, Huang CL, Sun L (2008). The developmental basis of skeletal cell differentiation and the molecular basis of major skeletal defects. Biol Rev Camb Philos Soc 83(4): 401-415.

Boe A, Baiocchi M, Carbonatto M, Papoian R, Serlupi-Crescenzi O (1999). Interleukin 6 knock-out mice are resistant to antigen-induced experimental arthritis. Cytokine 11(12): 1057-1064.

Borzi RM, Mazzetti I, Macor S, Silvestri T, Bassi A, Cattini L, et al. (1999). Flow cytometric analysis of intracellular chemokines in chondrocytes in vivo: constitutive expression and enhancement in osteoarthritis and rheumatoid arthritis. FEBS lett 455(3): 238-242.

Boskey AL, Gadaleta S, Gundberg C, Doty SB, Ducy P, Karsenty G (1998). Fourier transform infrared microspectroscopic analysis of bones of osteocalcin-deficient mice provides insight into the function of osteocalcin. Bone 23(3): 187-196.


230

Bossaller L, Burger J, Draeger R, Grimbacher B, Knoth R, Plebani A, et al. (2006). ICOS deficiency is associated with a severe reduction of CXCR5+CD4 germinal center Th cells. J Immunol 177(7): 4927-4932.

Bossard MJ, Tomaszek TA, Thompson SK, Amegadzie BY, Hanning CR, Jones C, et al. (1996). Proteolytic activity of human osteoclast cathepsin K. Expression, purification, activation, and substrate identification. J Biol Chem 271(21): 12517-12524.

Bougeret C, Mansur IG, Dastot H, Schmid M, Mahouy G, Bensussan A, et al. (1992). Increased surface expression of a newly identified 150-kDa dimer early after human T lymphocyte activation. J Immunol 148(2): 318-323.

Bouxsein ML, Boyd SK, Christiansen BA, Guldberg RE, Jepsen KJ, Muller R (2010). Guidelines for assessment of bone microstructure in rodents using micro-computed tomography. J Bone Miner Res 25(7): 1468-1486.

Boyce BF, Xing L (2008). Functions of RANKL/RANK/OPG in bone modeling and remodeling. Arch Biochem Biophys 473(2): 139-146.

Boyden LM, Mao J, Belsky J, Mitzner L, Farhi A, Mitnick MA, et al. (2002). High bone density due to a mutation in LDL-receptor-related protein 5. N Engl J Med 346(20): 1513-1521.

Boyle WJ, Simonet WS, Lacey DL (2003). Osteoclast differentiation and activation. Nature 423(6937): 337-342.

Bozec A, Zaiss MM, Kagwiria R, Voll R, Rauh M, Chen Z, et al. (2014). T cell costimulation molecules CD80/86 inhibit osteoclast differentiation by inducing the IDO/tryptophan pathway. Sci Transl Med 6(235): 235ra260.

Bradford MM (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254.

Brandes ME, Allen JB, Ogawa Y, Wahl SM (1991). Transforming growth factor beta 1 suppresses acute and chronic arthritis in experimental animals. J Clin Invest 87(3): 1108-1113.

Breitfeld D, Ohl L, Kremmer E, Ellwart J, Sallusto F, Lipp M, et al. (2000). Follicular B helper T cells express CXC chemokine receptor 5, localize to B cell follicles, and support immunoglobulin production. J Exp Med 192(11): 1545-1552.


231

Brennan FM, Chantry D, Turner M, Foxwell B, Maini R, Feldmann M (1990). Detection of transforming growth factor-beta in rheumatoid arthritis synovial tissue: lack of effect on spontaneous cytokine production in joint cell cultures. Clin Exp Immunol 81(2): 278-285.

Bromley M, Woolley DE (1984). Chondroclasts and osteoclasts at subchondral sites of erosion in the rheumatoid joint. Arthritis Rheum 27(9): 968-975.

Bucay N, Sarosi I, Dunstan CR, Morony S, Tarpley J, Capparelli C, et al. (1998). osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification. Genes Dev 12(9): 1260-1268.

Burchill MA, Nardelli DT, England DM, DeCoster DJ, Christopherson JA, Callister SM, et al. (2003). Inhibition of interleukin-17 prevents the development of arthritis in vaccinated mice challenged with Borrelia burgdorferi. Infect Immun 71(6): 3437-3442.

Cain CJ, Rueda R, McLelland B, Collette NM, Loots GG, Manilay JO (2012). Absence of sclerostin adversely affects B-cell survival. J Bone Miner Res 27(7): 1451-1461.

Calvi LM, Adams GB, Weibrecht KW, Weber JM, Olson DP, Knight MC, et al. (2003). Osteoblastic cells regulate the haematopoietic stem cell niche. Nature 425(6960): 841-846.

Cao L, Moriishi T, Miyazaki T, Iimura T, Hamagaki M, Nakane A, et al. (2011). Comparative morphology of the osteocyte lacunocanalicular system in various vertebrates. J Bone Miner Metab 29(6): 662-670.

Cardoso L, Herman BC, Verborgt O, Laudier D, Majeska RJ, Schaffler MB (2009). Osteocyte apoptosis controls activation of intracortical resorption in response to bone fatigue. J Bone Miner Res 24(4): 597-605.

Carlsten H (2005). Immune responses and bone loss: the estrogen connection. Immunol Rev 208: 194-206.

Caux C, Massacrier C, Vanbervliet B, Dubois B, Van Kooten C, Durand I, et al. (1994). Activation of human dendritic cells through CD40 cross-linking. J Exp Med 180(4): 1263-1272.

Centrella M, Horowitz MC, Wozney JM, McCarthy TL (1994). Transforming growth factor-beta gene family members and bone. Endocr Rev 15(1): 27-39.


232

Centrella M, McCarthy TL, Canalis E (1988). Tumor necrosis factor-alpha inhibits collagen synthesis and alkaline phosphatase activity independently of its effect on deoxyribonucleic acid synthesis in osteoblast-enriched bone cell cultures. Endocrinology 123(3): 1442-1448.

Chang MK, Raggatt LJ, Alexander KA, Kuliwaba JS, Fazzalari NL, Schroder K, et al. (2008). Osteal tissue macrophages are intercalated throughout human and mouse bone lining tissues and regulate osteoblast function in vitro and in vivo. J Immunol 181(2): 1232-1244.

Chen G, Deng C, Li YP (2012). TGF-beta and BMP signaling in osteoblast differentiation and bone formation. Int J Biol Sci 8(2): 272-288.

Chen ZJ (2005). Ubiquitin signalling in the NF-kappaB pathway. Nat Cell Biol 7(8): 758-765.

Cheng B, Zhao S, Luo J, Sprague E, Bonewald LF, Jiang JX (2001). Expression of functional gap junctions and regulation by fluid flow in osteocyte-like MLO-Y4 cells. J Bone Miner Res 16(2): 249-259.

Cheng J, Montecalvo A, Kane LP (2011). Regulation of NF-kappaB induction by TCR/CD28. Immunol Res 50(2-3): 113-117.

Cheon H, Yu SJ, Yoo DH, Chae IJ, Song GG, Sohn J (2002). Increased expression of pro-inflammatory cytokines and metalloproteinase-1 by TGF-beta1 in synovial fibroblasts from rheumatoid arthritis and normal individuals. Clin Exp Immunol 127(3): 547-552.

Chim SM, Tickner J, Chow ST, Kuek V, Guo B, Zhang G, et al. (2013). Angiogenic factors in bone local environment. Cytokine Growth Factor Rev 24(3): 297-310.

Cho SW, Soki FN, Koh AJ, Eber MR, Entezami P, Park SI, et al. (2014). Osteal macrophages support physiologic skeletal remodeling and anabolic actions of parathyroid hormone in bone. Proc Natl Acad Sci U S A 111(4): 1545-1550.

Chow JW, Wilson AJ, Chambers TJ, Fox SW (1998). Mechanical loading stimulates bone formation by reactivation of bone lining cells in 13-week-old rats. J Bone Miner Res 13(11): 1760-1767.

Chtanova T, Tangye SG, Newton R, Frank N, Hodge MR, Rolph MS, et al. (2004). T follicular helper cells express a distinctive transcriptional profile, reflecting their role as non-Th1/Th2 effector cells that provide help for B cells. J Immunol 173(1): 68-78.


233

Chung DJ, Castro CH, Watkins M, Stains JP, Chung MY, Szejnfeld VL, et al. (2006). Low peak bone mass and attenuated anabolic response to parathyroid hormone in mice with an osteoblast-specific deletion of connexin43. J Cell Sci 119(Pt 20): 4187-4198.

Chung UI, Kawaguchi H, Takato T, Nakamura K (2004). Distinct osteogenic mechanisms of bones of distinct origins. J Orthop Sci 9(4): 410-414.

Clark WD, Smith EL, Linn KA, Paul-Murphy JR, Muir P, Cook ME (2005). Osteocyte apoptosis and osteoclast presence in chicken radii 0-4 days following osteotomy. Calcif Tissue Int 77(5): 327-336.

Clarke B (2008). Normal bone anatomy and physiology. Clin J Am Soc Nephrol : 3 Suppl 3: S131-139.

Cohen P (2014). Immune diseases caused by mutations in kinases and components of the ubiquitin system. Nat Immunol 15(6): 521-529.

Cohen SL, Moore AM, Ward WE (2004). Interleukin-10 knockout mouse: a model for studying bone metabolism during intestinal inflammation. Inflamm Bowel Dis 10(5): 557-563.

Coquet JM, Middendorp S, van der Horst G, Kind J, Veraar EA, Xiao Y, et al. (2013). The CD27 and CD70 costimulatory pathway inhibits effector function of T helper 17 cells and attenuates associated autoimmunity. Immunity 38(1): 53-65.

Corisdeo S, Gyda M, Zaidi M, Moonga BS, Troen BR (2001). New insights into the regulation of cathepsin K gene expression by osteoprotegerin ligand. Biochem Biophys Res Commun 285(2): 335-339.

Corsi A, Collins MT, Riminucci M, Howell PG, Boyde A, Robey PG, et al. (2003). Osteomalacic and hyperparathyroid changes in fibrous dysplasia of bone: core biopsy studies and clinical correlations. J Bone Miner Res 18(7): 1235-1246.

D.Baeten, Emery P, Sieper J, Braun J, Heijde Dvd, McInnes I, et al. (2011). The Anti-IL17A Monoclonal Antibody Secukinumab (AIN457) Showed Good Safety and Efficacy in the Treatment of Active Ankylosing Spondylitis. In: EULAR 2011 (Abstract 0174). 25-28 May 2011, London, UK: The Annual European Congress of Rheumatology.

Danks L, Sabokbar A, Gundle R, Athanasou NA (2002). Synovial macrophage-osteoclast differentiation in inflammatory arthritis. Ann Rheum Dis 61(10): 916-921.


234

David JP, Sabapathy K, Hoffmann O, Idarraga MH, Wagner EF (2002). JNK1 modulates osteoclastogenesis through both c-Jun phosphorylation-dependent and -independent mechanisms. J Cell Sci 115(Pt 22): 4317-4325.

DeKoter RP, Walsh JC, Singh H (1998). PU.1 regulates both cytokine-dependent proliferation and differentiation of granulocyte/macrophage progenitors. EMBO J 17(15): 4456-4468.

Dempster DW, Cosman F, Kurland ES, Zhou H, Nieves J, Woelfert L, et al. (2001). Effects of daily treatment with parathyroid hormone on bone microarchitecture and turnover in patients with osteoporosis: a paired biopsy study. J Bone Miner Res 16(10): 1846-1853.

Deng L, Wang C, Spencer E, Yang L, Braun A, You J, et al. (2000). Activation of the IkappaB kinase complex by TRAF6 requires a dimeric ubiquitin-conjugating enzyme complex and a unique polyubiquitin chain. Cell 103(2): 351-361.

Deonarain R, Alcami A, Alexiou M, Dallman MJ, Gewert DR, Porter AC (2000). Impaired antiviral response and alpha/beta interferon induction in mice lacking beta interferon. J Virol 74(7): 3404-3409.

Derynck R, Zhang YE (2003). Smad-dependent and Smad-independent pathways in TGF-beta family signalling. Nature 425(6958): 577-584.

Desgeorges A, Gabay C, Silacci P, Novick D, Roux-Lombard P, Grau G, et al. (1997). Concentrations and origins of soluble interleukin 6 receptor-alpha in serum and synovial fluid. J Rheumatol 24(8): 1510-1516.

Di Giovine FS, Nuki G, Duff GW (1988). Tumour necrosis factor in synovial exudates. Ann Rheum Dis 47(9): 768-772.

Diarra D, Stolina M, Polzer K, Zwerina J, Ominsky MS, Dwyer D, et al. (2007). Dickkopf-1 is a master regulator of joint remodeling. Nat Med 13(2): 156-163.

Dickson MC, Martin JS, Cousins FM, Kulkarni AB, Karlsson S, Akhurst RJ (1995). Defective haematopoiesis and vasculogenesis in transforming growth factor-beta 1 knock out mice. Development 121(6): 1845-1854.

Dierkes C, Kreisel M, Schulz A, Steinmeyer J, Wolff JC, Fink L (2009). Catabolic properties of microdissected human endosteal bone lining cells. Calcif Tissue Int 84(2): 146-155.


235

Dobnig H, Turner RT (1995). Evidence that intermittent treatment with parathyroid hormone increases bone formation in adult rats by activation of bone lining cells. Endocrinology 136(8): 3632-3638.

Dong C, Juedes AE, Temann UA, Shresta S, Allison JP, Ruddle NH, et al. (2001a). ICOS co-stimulatory receptor is essential for T-cell activation and function. Nature 409(6816): 97-101.

Dong C, Temann UA, Flavell RA (2001b). Cutting edge: critical role of inducible costimulator in germinal center reactions. J Immunol 166(6): 3659-3662.

Doty SB (1981). Morphological evidence of gap junctions between bone cells. Calcif Tissue Int 33(5): 509-512.

Dougall WC, Glaccum M, Charrier K, Rohrbach K, Brasel K, De Smedt T, et al. (1999). RANK is essential for osteoclast and lymph node development. Genes Dev 13(18): 2412-2424.

Dresner-Pollak R, Gelb N, Rachmilewitz D, Karmeli F, Weinreb M (2004). Interleukin 10-deficient mice develop osteopenia, decreased bone formation, and mechanical fragility of long bones. Gastroenterology 127(3): 792-801.

Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G (1997). Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89(5): 747-754.

Dudley HR, Spiro D (1961). The Fine Structure of Bone Cells. J Biophys Biochem Cytol 11(3): 627-649.

Duplomb L, Baud'huin M, Charrier C, Berreur M, Trichet V, Blanchard F, et al. (2008). Interleukin-6 inhibits receptor activator of nuclear factor kappaB ligand-induced osteoclastogenesis by diverting cells into the macrophage lineage: key role of Serine727 phosphorylation of signal transducer and activator of transcription 3. Endocrinology 149(7): 3688-3697.

Duplomb L, Dagouassat M, Jourdon P, Heymann D (2007). Concise review: embryonic stem cells: a new tool to study osteoblast and osteoclast differentiation. Stem Cells 25(3): 544-552.

Edwards JR, Mundy GR (2011). Advances in osteoclast biology: old findings and new insights from mouse models. Nat Rev Rheumatol 7(4): 235-243.


236

Ek-Rylander B, Flores M, Wendel M, Heinegard D, Andersson G (1994). Dephosphorylation of osteopontin and bone sialoprotein by osteoclastic tartrate-resistant acid phosphatase. Modulation of osteoclast adhesion in vitro. J Biol Chem 269(21): 14853-14856.

Elliott MJ, Maini RN, Feldmann M, Kalden JR, Antoni C, Smolen JS, et al. (1994). Randomised double-blind comparison of chimeric monoclonal antibody to tumour necrosis factor alpha (cA2) versus placebo in rheumatoid arthritis. Lancet 344(8930): 1105-1110.

Eriksen EF (1986). Normal and pathological remodeling of human trabecular bone: three dimensional reconstruction of the remodeling sequence in normals and in metabolic bone disease. Endocr Rev 7(4): 379-408.

Erlebacher A, Derynck R (1996). Increased expression of TGF-beta 2 in osteoblasts results in an osteoporosis-like phenotype. J Cell Biol 132(1-2): 195-210.

Everts V, Beertsen W (2005). External Lysosomes: The Osteoclast and Its Unique Capacities to Degrade Mineralised Tissues. In: Saftig P. Lysosomes, edn. Georgetown, Tex. New York, N.Y.: Landes Bioscience/Eurekah.com ; Springer Science+Business Media. p^pp 197 p.

Everts V, Delaisse JM, Korper W, Jansen DC, Tigchelaar-Gutter W, Saftig P, et al. (2002). The bone lining cell: its role in cleaning Howship's lacunae and initiating bone formation. J Bone Miner Res 17(1): 77-90.

Fakhry M, Hamade E, Badran B, Buchet R, Magne D (2013). Molecular mechanisms of mesenchymal stem cell differentiation towards osteoblasts. World J Stem Cells 5(4): 136-148.

Fazilleau N, Mark L, McHeyzer-Williams LJ, McHeyzer-Williams MG (2009). Follicular helper T cells: lineage and location. Immunity 30(3): 324-335.

Feng X (2009). Chemical and Biochemical Basis of Cell-Bone Matrix Interaction in Health and Disease. Curr Chem Biol 3(2): 189-196.

Feng X (2005). RANKing intracellular signaling in osteoclasts. IUBMB life 57(6): 389-395.

Ferguson PJ, Bing X, Vasef MA, Ochoa LA, Mahgoub A, Waldschmidt TJ, et al. (2006). A missense mutation in pstpip2 is associated with the murine autoinflammatory disorder chronic multifocal osteomyelitis. Bone 38(1): 41-47.


237

Filvaroff E, Erlebacher A, Ye J, Gitelman SE, Lotz J, Heillman M, et al. (1999). Inhibition of TGF-beta receptor signaling in osteoblasts leads to decreased bone remodeling and increased trabecular bone mass. Development 126(19): 4267-4279.

Fleige S, Pfaffl MW (2006). RNA integrity and the effect on the real-time qRT-PCR performance. Mol Aspects Med 27(2-3): 126-139.

Fontana A, Hengartner H, Weber E, Fehr K, Grob PJ, Cohen G (1982). Interleukin 1 activity in the synovial fluid of patients with rheumatoid arthritis. Rheumatol Int 2(2): 49-53.

Fox SW, Fuller K, Bayley KE, Lean JM, Chambers TJ (2000). TGF-beta 1 and IFN-gamma direct macrophage activation by TNF-alpha to osteoclastic or cytocidal phenotype. J Immunol 165(9): 4957-4963.

Francois RJ, Neure L, Sieper J, Braun J (2006). Immunohistological examination of open sacroiliac biopsies of patients with ankylosing spondylitis: detection of tumour necrosis factor alpha in two patients with early disease and transforming growth factor beta in three more advanced cases. Ann Rheum Dis 65(6): 713-720.

Franz-Odendaal TA, Hall BK, Witten PE (2006). Buried alive: how osteoblasts become osteocytes. Dev Dyn 235(1): 176-190.

Franzoso G, Carlson L, Xing L, Poljak L, Shores EW, Brown KD, et al. (1997). Requirement for NF-kappaB in osteoclast and B-cell development. Genes Dev 11(24): 3482-3496.

Frey O, Petrow PK, Gajda M, Siegmund K, Huehn J, Scheffold A, et al. (2005). The role of regulatory T cells in antigen-induced arthritis: aggravation of arthritis after depletion and amelioration after transfer of CD4+CD25+ T cells. Arthritis Res Ther 7(2): R291-301.

Friedel RH, Wurst W, Wefers B, Kuhn R (2011). Generating conditional knockout mice. Methods Mol Biol 693: 205-231.

Frost HM (1973). Bone remodeling and its relationship to metabolic bone diseases. edn. Thomas: Springfield, Ill.,.

Fujii M, Takeda K, Imamura T, Aoki H, Sampath TK, Enomoto S, et al. (1999). Roles of bone morphogenetic protein type I receptors and Smad proteins in osteoblast and chondroblast differentiation. Mol Biol Cell 10(11): 3801-3813.


238

Fujioka K, Kishida T, Ejima A, Yamamoto K, Fujii W, Murakami K, et al. (2014). Inhibition of osteoclastogenesis by osteoblast-like cells genetically engineered to produce interleukin-10. Biochem Biophys Res Commun 456(3): 785-91.

Fukada T, Hibi M, Yamanaka Y, Takahashi-Tezuka M, Fujitani Y, Yamaguchi T, et al. (1996). Two signals are necessary for cell proliferation induced by a cytokine receptor gp130: involvement of STAT3 in anti-apoptosis. Immunity 5(5): 449-460.

Fuller K, Owens JM, Jagger CJ, Wilson A, Moss R, Chambers TJ (1993). Macrophage colony-stimulating factor stimulates survival and chemotactic behavior in isolated osteoclasts. J Exp Med 178(5): 1733-1744.

Fuller K, Wong B, Fox S, Choi Y, Chambers TJ (1998). TRANCE is necessary and sufficient for osteoblast-mediated activation of bone resorption in osteoclasts. J Exp Med 188(5): 997-1001.

Fullilove S, Jellis J, Hughes SP, Remick DG, Friedland JS (2000). Local and systemic concentrations of tumour necrosis factor-alpha, interleukin-6 and interleukin-8 in bacterial osteomyelitis. Trans R Soc Trop Med Hyg 94(2): 221-224.

Gao Y, Grassi F, Ryan MR, Terauchi M, Page K, Yang X, et al. (2007). IFN-gamma stimulates osteoclast formation and bone loss in vivo via antigen-driven T cell activation. J Clin Invest 117(1): 122-132.

Geiser AG, Zeng QQ, Sato M, Helvering LM, Hirano T, Turner CH (1998). Decreased bone mass and bone elasticity in mice lacking the transforming growth factor-beta1 gene. Bone 23(2): 87-93.

Gelb BD, Shi GP, Chapman HA, Desnick RJ (1996). Pycnodysostosis, a lysosomal disease caused by cathepsin K deficiency. Science 273(5279): 1236-1238.

Genovese MC, Van den Bosch F, Roberson SA, Bojin S, Biagini IM, Ryan P, et al. (2010). LY2439821, a humanized anti-interleukin-17 monoclonal antibody, in the treatment of patients with rheumatoid arthritis: A phase I randomized, double-blind, placebo-controlled, proof-of-concept study. Arthritis Rheum 62(4): 929-939.

Gilbert L, He X, Farmer P, Boden S, Kozlowski M, Rubin J, et al. (2000). Inhibition of osteoblast differentiation by tumor necrosis factor-alpha. Endocrinology 141(11): 3956-3964.


239

Gilbert SF (2000a). Osteogenesis: The development of Bones. In: (ed)^(eds). Development Biology, 6th edn. Sunderland, Mass.: Sinauer Associates. p^pp xviii, 749 p.

Gilbert SF (2000b). The paraxial and intermediate mesoderm. In: (ed)^(eds). Developmental Biology, 6th edn. Sunderland, Mass.: Sinauer Associates. p^pp 465-490.

Glasmacher E, Hoefig KP, Vogel KU, Rath N, Du L, Wolf C, et al. (2010). Roquin binds inducible costimulator mRNA and effectors of mRNA decay to induce microRNA-independent post-transcriptional repression. Nat Immunol 11(8): 725-733.

Gohda J, Akiyama T, Koga T, Takayanagi H, Tanaka S, Inoue J (2005). RANK-mediated amplification of TRAF6 signaling leads to NFATc1 induction during osteoclastogenesis. EMBO J 24(4): 790-799.

Goldring SR (2013). Osteoimmunology and bone homeostasis: relevance to spondyloarthritis. Curr Rheumatol Rep 15(7): 342.

Gong Y, Slee RB, Fukai N, Rawadi G, Roman-Roman S, Reginato AM, et al. (2001). LDL receptor-related protein 5 (LRP5) affects bone accrual and eye development. Cell 107(4): 513-523.

Gordon CM (2006). Bone loss in children with Crohn disease: Evidence of "osteoimmune" alterations. Journal Pediatr 148(4): 429-432.

Gough AK, Lilley J, Eyre S, Holder RL, Emery P (1994). Generalised bone loss in patients with early rheumatoid arthritis. Lancet 344(8914): 23-27.

Gowen M, Mundy GR (1986). Actions of recombinant interleukin 1, interleukin 2, and interferon-gamma on bone resorption in vitro. J Immunol 136(7): 2478-2482.

Gratacos J, Collado A, Filella X, Sanmarti R, Canete J, Llena J, et al. (1994). Serum cytokines (IL-6, TNF-alpha, IL-1 beta and IFN-gamma) in ankylosing spondylitis: a close correlation between serum IL-6 and disease activity and severity. Br J Rheumatol 33(10): 927-931.

Gravallese EM, Manning C, Tsay A, Naito A, Pan C, Amento E, et al. (2000). Synovial tissue in rheumatoid arthritis is a source of osteoclast differentiation factor. Arthritis Rheum 43(2): 250-258.


240

Grigoriadis AE, Wang ZQ, Cecchini MG, Hofstetter W, Felix R, Fleisch HA, et al. (1994). c-Fos: a key regulator of osteoclast-macrophage lineage determination and bone remodeling. Science 266(5184): 443-448.

Grimston SK, Brodt MD, Silva MJ, Civitelli R (2008). Attenuated response to in vivo mechanical loading in mice with conditional osteoblast ablation of the connexin43 gene (Gja1). J Bone Miner Res 23(6): 879-886.

Gu G, Mulari M, Peng Z, Hentunen TA, Vaananen HK (2005). Death of osteocytes turns off the inhibition of osteoclasts and triggers local bone resorption. Biochem Biophys Res Commun 335(4): 1095-1101.

Guihard P, Danger Y, Brounais B, David E, Brion R, Delecrin J, et al. (2012). Induction of osteogenesis in mesenchymal stem cells by activated monocytes/macrophages depends on oncostatin M signaling. Stem Cells 30(4): 762-772.

Gunness-Hey M, Hock JM (1984). Increased trabecular bone mass in rats treated with human synthetic parathyroid hormone. Metab Bone Dis Relat Res 5(4): 177-181.

Hamann C, Picke AK, Campbell GM, Balyura M, Rauner M, Bernhardt R, et al. (2014). Effects of parathyroid hormone on bone mass, bone strength, and bone regeneration in male rats with type 2 diabetes mellitus. Endocrinology 155(4): 1197-1206.

Harada A, Sekido N, Akahoshi T, Wada T, Mukaida N, Matsushima K (1994). Essential involvement of interleukin-8 (IL-8) in acute inflammation. J Leukoc Biol 56(5): 559-564.

Harada S, Rodan GA (2003). Control of osteoblast function and regulation of bone mass. Nature 423(6937): 349-355.

Hashizume M, Hayakawa N, Mihara M (2008). IL-6 trans-signalling directly induces RANKL on fibroblast-like synovial cells and is involved in RANKL induction by TNF-alpha and IL-17. Rheumatology 47(11): 1635-1640.

Hashizume M, Mihara M (2010). High molecular weight hyaluronic acid inhibits IL-6-induced MMP production from human chondrocytes by up-regulating the ERK inhibitor, MKP-1. Biochem Biophys Res Commun 403(2): 184-189.

Hattner R, Epker BN, Frost HM (1965). Suggested sequential mode of control of changes in cell behaviour in adult bone remodelling. Nature 206(983): 489-490.


241

Hauge EM, Qvesel D, Eriksen EF, Mosekilde L, Melsen F (2001). Cancellous bone remodeling occurs in specialized compartments lined by cells expressing osteoblastic markers. J Bone Miner Res 16(9): 1575-1582.

Hayman AR, Jones SJ, Boyde A, Foster D, Colledge WH, Carlton MB, et al. (1996). Mice lacking tartrate-resistant acid phosphatase (Acp 5) have disrupted endochondral ossification and mild osteopetrosis. Development 122(10): 3151-3162.

He Y, Staser K, Rhodes SD, Liu Y, Wu X, Park SJ, et al. (2011). Erk1 positively regulates osteoclast differentiation and bone resorptive activity. PLoS One 6(9): e24780.

Heinrich PC, Behrmann I, Muller-Newen G, Schaper F, Graeve L (1998). Interleukin-6-type cytokine signalling through the gp130/Jak/STAT pathway. Biochem J 334 ( Pt 2): 297-314.

Henriksen K, Neutzsky-Wulff AV, Bonewald LF, Karsdal MA (2009). Local communication on and within bone controls bone remodeling. Bone 44(6): 1026-1033.

Hering S, Isken F, Janott J, Jost C, Pommer A, Muhr G, et al. (2001). Analysis of TGFbeta3 gene expression and protein levels in human bone and serum. Exp Clin Endocrinol Diabetes 109(2): 107-115.

Hibi M, Murakami M, Saito M, Hirano T, Taga T, Kishimoto T (1990). Molecular cloning and expression of an IL-6 signal transducer, gp130. Cell 63(6): 1149-1157.

Hildebrand T, Ruegsegger P (1997). Quantification of Bone Microarchitecture with the Structure Model Index. Comput Methods Biomech Biomed Engin 1(1): 15-23.

Hill PA (1998). Bone remodelling. Br J Orthod 25(2): 101-107.

Hirano T, Matsuda T, Turner M, Miyasaka N, Buchan G, Tang B, et al. (1988). Excessive production of interleukin 6/B cell stimulatory factor-2 in rheumatoid arthritis. Eur J Immunol. 18(11): 1797-1801.

Hirotani H, Tuohy NA, Woo JT, Stern PH, Clipstone NA (2004). The calcineurin/nuclear factor of activated T cells signaling pathway regulates osteoclastogenesis in RAW264.7 cells. J Biol Chem 279(14): 13984-13992.

Hofbauer LC, Lacey DL, Dunstan CR, Spelsberg TC, Riggs BL, Khosla S (1999). Interleukin-1beta and tumor necrosis factor-alpha, but not interleukin-6, stimulate


242

osteoprotegerin ligand gene expression in human osteoblastic cells. Bone 25(3): 255-259.

Hong MH, Williams H, Jin CH, Pike JW (2000). The inhibitory effect of interleukin-10 on mouse osteoclast formation involves novel tyrosine-phosphorylated proteins. J Bone Miner Res 15(5): 911-918.

Hopkins SJ, Humphreys M, Jayson MI (1988). Cytokines in synovial fluid. I. The presence of biologically active and immunoreactive IL-1. Clin Exp Immunol 72(3): 422-427.

Horowitz M, Jilka RL (1992). Colony stimulating factors in bone remodeling. In: Cytokines and Bone Metabolism, Gowen M (ed). Boca Raton, FL: CRC Press.

Horowitz MC, Bothwell AL, Hesslein DG, Pflugh DL, Schatz DG (2005). B cells and osteoblast and osteoclast development. Immunol Rev 208: 141-153.

Horowitz MC, Xi Y, Pflugh DL, Hesslein DG, Schatz DG, Lorenzo JA, et al. (2004). Pax5-deficient mice exhibit early onset osteopenia with increased osteoclast progenitors. J Immunol 173(11): 6583-6591.

Horwood NJ, Kartsogiannis V, Quinn JM, Romas E, Martin TJ, Gillespie MT (1999). Activated T lymphocytes support osteoclast formation in vitro. Biochem Biophys Res Commun 265(1): 144-150.

Huang H, Kim HJ, Chang EJ, Lee ZH, Hwang SJ, Kim HM, et al. (2009). IL-17 stimulates the proliferation and differentiation of human mesenchymal stem cells: implications for bone remodeling. Cell Death Differ 16(10): 1332-1343.

Huang H, Li Y, Qi X (2013). Cytokine signaling in the differentiation of innate effector cells. Jak-Stat 2(1): e23531.

Huang W, O'Keefe RJ, Schwarz EM (2003). Exposure to receptor-activator of NFkappaB ligand renders pre-osteoclasts resistant to IFN-gamma by inducing terminal differentiation. Arthritis Res Ther 5(1): R49-59.

Huber O, Korn R, McLaughlin J, Ohsugi M, Herrmann BG, Kemler R (1996). Nuclear localization of beta-catenin by interaction with transcription factor LEF-1. Mech Dev 59(1): 3-10.


243

Hume DA, Loutit JF, Gordon S (1984). The mononuclear phagocyte system of the mouse defined by immunohistochemical localization of antigen F4/80: macrophages of bone and associated connective tissue. J Cell Sci 66: 189-194.

Iannone F, De Bari C, Dell'Accio F, Covelli M, Cantatore FP, Patella V, et al. (2001). Interleukin-10 and interleukin-10 receptor in human osteoarthritic and healthy chondrocytes. Clin Exp Rheumatol. 19(2): 139-145.

Igarashi H, Gregory SC, Yokota T, Sakaguchi N, Kincade PW (2002). Transcription from the RAG1 locus marks the earliest lymphocyte progenitors in bone marrow. Immunity 17(2): 117-130.

Iolascon G, Resmini G, Tarantino U (2013). Mechanobiology of bone. Aging Clin Exp Res 25 Suppl 1: S3-7.

Irie N, Takada Y, Watanabe Y, Matsuzaki Y, Naruse C, Asano M, et al. (2009). Bidirectional signaling through ephrinA2-EphA2 enhances osteoclastogenesis and suppresses osteoblastogenesis. J Biol Chem 284(21): 14637-14644.

Ishii M, Egen JG, Klauschen F, Meier-Schellersheim M, Saeki Y, Vacher J, et al. (2009). Sphingosine-1-phosphate mobilizes osteoclast precursors and regulates bone homeostasis. Nature 458(7237): 524-528.

Ishimi Y, Miyaura C, Jin CH, Akatsu T, Abe E, Nakamura Y, et al. (1990). IL-6 is produced by osteoblasts and induces bone resorption. J Immunol 145(10): 3297-3303.

Ivanov, II, McKenzie BS, Zhou L, Tadokoro CE, Lepelley A, Lafaille JJ, et al. (2006). The orphan nuclear receptor RORgammat directs the differentiation program of proinflammatory IL-17+ T helper cells. Cell 126(6): 1121-1133.

Ivashkiv LB, Donlin LT (2014). Regulation of type I interferon responses. Nat Rev Immunol 14(1): 36-49.

Jacome-Galarza CE, Lee SK, Lorenzo JA, Aguila HL (2013). Identification, characterization, and isolation of a common progenitor for osteoclasts, macrophages, and dendritic cells from murine bone marrow and periphery. J Bone Miner Res 28(5): 1203-1213.

Jacquin C, Gran DE, Lee SK, Lorenzo JA, Aguila HL (2006). Identification of multiple osteoclast precursor populations in murine bone marrow. J Bone Miner Res 21(1): 67-77.


244

Ji JD, Park-Min KH, Shen Z, Fajardo RJ, Goldring SR, McHugh KP, et al. (2009). Inhibition of RANK expression and osteoclastogenesis by TLRs and IFN-gamma in human osteoclast precursors. J Immunol 183(11): 7223-7233.

Jiang H, Harris MB, Rothman P (2000). IL-4/IL-13 signaling beyond JAK/STAT. J Allergy Clin Immunol 105(6 Pt 1): 1063-1070.

Jilka RL, Weinstein RS, Bellido T, Parfitt AM, Manolagas SC (1998). Osteoblast programmed cell death (apoptosis): modulation by growth factors and cytokines. J Bone Miner Res 13(5): 793-802.

Jilka RL, Weinstein RS, Bellido T, Roberson P, Parfitt AM, Manolagas SC (1999). Increased bone formation by prevention of osteoblast apoptosis with parathyroid hormone. J Clin Invest 104(4): 439-446.

Jing D, Baik AD, Lu XL, Zhou B, Lai X, Wang L, et al. (2014). In situ intracellular calcium oscillations in osteocytes in intact mouse long bones under dynamic mechanical loading. FASEB J 28(4):1582-92.

Jones SJ, Gray C, Sakamaki H, Arora M, Boyde A, Gourdie R, et al. (1993). The incidence and size of gap junctions between the bone cells in rat calvaria. Anat Embryol (Berl) 187(4): 343-352.

Jun JE, Wilson LE, Vinuesa CG, Lesage S, Blery M, Miosge LA, et al. (2003). Identifying the MAGUK protein Carma-1 as a central regulator of humoral immune responses and atopy by genome-wide mouse mutagenesis. Immunity 18(6): 751-762.

Jung Y, Wang J, Schneider A, Sun YX, Koh-Paige AJ, Osman NI, et al. (2006). Regulation of SDF-1 (CXCL12) production by osteoblasts; a possible mechanism for stem cell homing. Bone 38(4): 497-508.

Jung YK, Kim GW, Park HR, Lee EJ, Choi JY, Beier F, et al. (2013). Role of interleukin-10 in endochondral bone formation in mice: anabolic effect via the bone morphogenetic protein/Smad pathway. Arthritis Rheum 65(12): 3153-3164.

Kamalakar A, Bendre MS, Washam CL, Fowler TW, Carver A, Dilley JD, et al. (2014). Circulating interleukin-8 levels explain breast cancer osteolysis in mice and humans. Bone 61: 176-185.

Kamioka H, Ishihara Y, Ris H, Murshid SA, Sugawara Y, Takano-Yamamoto T, et al. (2007). Primary cultures of chick osteocytes retain functional gap junctions between osteocytes and between osteocytes and osteoblasts. Microsc Microanal 13(2): 108-117.


245

Kaneda T, Nojima T, Nakagawa M, Ogasawara A, Kaneko H, Sato T, et al. (2000). Endogenous production of TGF-beta is essential for osteoclastogenesis induced by a combination of receptor activator of NF-kappa B ligand and macrophage-colony-stimulating factor. J Immunol 165(8): 4254-4263.

Kaneko S, Satoh T, Chiba J, Ju C, Inoue K, Kagawa J (2000). Interleukin-6 and interleukin-8 levels in serum and synovial fluid of patients with osteoarthritis. Cytokines Cell Mol Ther 6(2): 71-79.

Kaplanski G, Marin V, Montero-Julian F, Mantovani A, Farnarier C (2003). IL-6: a regulator of the transition from neutrophil to monocyte recruitment during inflammation. Trends Immunol 24(1): 25-29.

Kar R, Riquelme MA, Werner S, Jiang JX (2013). Connexin 43 channels protect osteocytes against oxidative stress-induced cell death. J Bone Miner Res 28(7): 1611-1621.

Kariuki SN, Niewold TB (2010). Genetic regulation of serum cytokines in systemic lupus erythematosus. Transl Res 155(3): 109-117.

Karlson EW, Chibnik LB, Tworoger SS, Lee IM, Buring JE, Shadick NA, et al. (2009). Biomarkers of inflammation and development of rheumatoid arthritis in women from two prospective cohort studies. Arthritis Rheum 60(3): 641-652.

Karsdal MA, Martin TJ, Bollerslev J, Christiansen C, Henriksen K (2007). Are nonresorbing osteoclasts sources of bone anabolic activity? J Bone Miner Res 22(4): 487-494.

Katavic V, Lukic IK, Kovacic N, Grcevic D, Lorenzo JA, Marusic A (2003). Increased bone mass is a part of the generalized lymphoproliferative disorder phenotype in the mouse. J Immunol 170(3): 1540-1547.

Kawakami A, Eguchi K, Matsuoka N, Tsuboi M, Koji T, Urayama S, et al. (1997). Fas and Fas ligand interaction is necessary for human osteoblast apoptosis. J Bone Miner Res 12(10): 1637-1646.

Kayal RA (2013). The role of osteoimmunology in periodontal disease. Biomed Res Int 2013: 639368.

Kelly-Welch AE, Hanson EM, Boothby MR, Keegan AD (2003). Interleukin-4 and interleukin-13 signaling connections maps. Science 300(5625): 1527-1528.


246

Kennedy OD, Herman BC, Laudier DM, Majeska RJ, Sun HB, Schaffler MB (2012). Activation of resorption in fatigue-loaded bone involves both apoptosis and active pro-osteoclastogenic signaling by distinct osteocyte populations. Bone 50(5): 1115-1122.

Key LL, Jr., Rodriguiz RM, Willi SM, Wright NM, Hatcher HC, Eyre DR, et al. (1995). Long-term treatment of osteopetrosis with recombinant human interferon gamma. N Engl J Med 332(24): 1594-1599.

Kikuta J, Kawamura S, Okiji F, Shirazaki M, Sakai S, Saito H, et al. (2013). Sphingosine-1-phosphate-mediated osteoclast precursor monocyte migration is a critical point of control in antibone-resorptive action of active vitamin D. Proc Natl Acad Sci U S A 110(17): 7009-7013.

Kim D, Mebius RE, MacMicking JD, Jung S, Cupedo T, Castellanos Y, et al. (2000). Regulation of peripheral lymph node genesis by the tumor necrosis factor family member TRANCE. J Exp Med 192(10): 1467-1478.

Kim HS, Kim DK, Kim AR, Mun SH, Lee SK, Kim JH, et al. (2012). Fyn positively regulates the activation of DAP12 and FcRgamma-mediated costimulatory signals by RANKL during osteoclastogenesis. Cell Signal 24(6): 1306-1314.

Kim JH, Jin HM, Kim K, Song I, Youn BU, Matsuo K, et al. (2009). The mechanism of osteoclast differentiation induced by IL-1. J Immunol 183(3): 1862-1870.

Kim K, Kim JH, Lee J, Jin HM, Kook H, Kim KK, et al. (2007a). MafB negatively regulates RANKL-mediated osteoclast differentiation. Blood 109(8): 3253-3259.

Kim K, Lee SH, Ha Kim J, Choi Y, Kim N (2008). NFATc1 induces osteoclast fusion via up-regulation of Atp6v0d2 and the dendritic cell-specific transmembrane protein (DC-STAMP). Mol Endocrinol 22(1): 176-185.

Kim N, Kadono Y, Takami M, Lee J, Lee SH, Okada F, et al. (2005). Osteoclast differentiation independent of the TRANCE-RANK-TRAF6 axis. J Exp Med 202(5): 589-595.

Kim SY, Han SW, Kim GW, Lee JM, Kang YM (2004). TGF-beta1 polymorphism determines the progression of joint damage in rheumatoid arthritis. Scand J Rheumatol 33(6): 389-394.

Kim YG, Lee CK, Nah SS, Mun SH, Yoo B, Moon HB (2007b). Human CD4+CD25+ regulatory T cells inhibit the differentiation of osteoclasts from peripheral blood mononuclear cells. Biochem Biophys Res Commun 357(4): 1046-1052.


247

Kitami S, Tanaka H, Kawato T, Tanabe N, Katono-Tani T, Zhang F, et al. (2010). IL-17A suppresses the expression of bone resorption-related proteinases and osteoclast differentiation via IL-17RA or IL-17RC receptors in RAW264.7 cells. Biochimie 92(4): 398-404.

Kitazawa R, Kimble RB, Vannice JL, Kung VT, Pacifici R (1994). Interleukin-1 receptor antagonist and tumor necrosis factor binding protein decrease osteoclast formation and bone resorption in ovariectomized mice. J Clin Invest 94(6): 2397-2406.

Kobayashi N, Kadono Y, Naito A, Matsumoto K, Yamamoto T, Tanaka S, et al. (2001). Segregation of TRAF6-mediated signaling pathways clarifies its role in osteoclastogenesis. EMBO J 20(6): 1271-1280.

Koch AE, Kunkel SL, Harlow LA, Mazarakis DD, Haines GK, Burdick MD, et al. (1994). Macrophage inflammatory protein-1 alpha. A novel chemotactic cytokine for macrophages in rheumatoid arthritis. J Clin Invest 93(3): 921-928.

Koch AE, Kunkel SL, Shah MR, Fu R, Mazarakis DD, Haines GK, et al. (1995a). Macrophage inflammatory protein-1 beta: a C-C chemokine in osteoarthritis. Clin Immunol Immunopathol 77(3): 307-314.

Koch AE, Kunkel SL, Shah MR, Hosaka S, Halloran MM, Haines GK, et al. (1995b). Growth-related gene product alpha. A chemotactic cytokine for neutrophils in rheumatoid arthritis. J Immunol 155(7): 3660-3666.

Koenders MI, Lubberts E, Oppers-Walgreen B, van den Bersselaar L, Helsen MM, Di Padova FE, et al. (2005). Blocking of interleukin-17 during reactivation of experimental arthritis prevents joint inflammation and bone erosion by decreasing RANKL and interleukin-1. Am J Pathol 167(1): 141-149.

Komatsu N, Okamoto K, Sawa S, Nakashima T, Oh-hora M, Kodama T, et al. (2014). Pathogenic conversion of Foxp3+ T cells into TH17 cells in autoimmune arthritis. Nat Med 20(1): 62-68.

Kong YY, Feige U, Sarosi I, Bolon B, Tafuri A, Morony S, et al. (1999a). Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature 402(6759): 304-309.

Kong YY, Yoshida H, Sarosi I, Tan HL, Timms E, Capparelli C, et al. (1999b). OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397(6717): 315-323.


248

Konig A, Krenn V, Toksoy A, Gerhard N, Gillitzer R (2000). Mig, GRO alpha and RANTES messenger RNA expression in lining layer, infiltrates and different leucocyte populations of synovial tissue from patients with rheumatoid arthritis, psoriatic arthritis and osteoarthritis. Virchows Archiv 436(5): 449-458.

Kopesky P, Tiedemann K, Alkekhia D, Zechner C, Millard B, Schoeberl B, et al. (2014). Autocrine signaling is a key regulatory element during osteoclastogenesis. Biol Open 3(8): 767-76.

Korn T, Bettelli E, Oukka M, Kuchroo VK (2009). IL-17 and Th17 Cells. Annu Rev Immunol 27: 485-517.

Kotake S, Nanke Y, Mogi M, Kawamoto M, Furuya T, Yago T, et al. (2005). IFN-gamma-producing human T cells directly induce osteoclastogenesis from human monocytes via the expression of RANKL. Eur J Immunol 35(11): 3353-3363.

Kotake S, Udagawa N, Hakoda M, Mogi M, Yano K, Tsuda E, et al. (2001). Activated human T cells directly induce osteoclastogenesis from human monocytes: possible role of T cells in bone destruction in rheumatoid arthritis patients. Arthritis Rheum 44(5): 1003-1012.

Kotake S, Udagawa N, Takahashi N, Matsuzaki K, Itoh K, Ishiyama S, et al. (1999). IL-17 in synovial fluids from patients with rheumatoid arthritis is a potent stimulator of osteoclastogenesis. J Clin Invest 103(9): 1345-1352.

Kristensen HB, Andersen TL, Marcussen N, Rolighed L, Delaisse JM (2013). Increased presence of capillaries next to remodeling sites in adult human cancellous bone. J Bone Miner Res 28(3): 574-585.

Kukita T, Wada N, Kukita A, Kakimoto T, Sandra F, Toh K, et al. (2004). RANKL-induced DC-STAMP is essential for osteoclastogenesis. J Exp Med 200(7): 941-946.

Kular J, Tickner J, Chim SM, Xu J (2012). An overview of the regulation of bone remodelling at the cellular level. Clin Biochem 45(12): 863-873.

Kulkarni AB, Huh CG, Becker D, Geiser A, Lyght M, Flanders KC, et al. (1993). Transforming growth factor beta 1 null mutation in mice causes excessive inflammatory response and early death. Proc Natl Acad Sci U S A 90(2): 770-774.

Kurt-Jones EA, Mandell L, Whitney C, Padgett A, Gosselin K, Newburger PE, et al. (2002). Role of toll-like receptor 2 (TLR2) in neutrophil activation: GM-CSF enhances TLR2 expression and TLR2-mediated interleukin 8 responses in neutrophils. Blood 100(5): 1860-1868.


249

Kwok SK, Cho ML, Park MK, Oh HJ, Park JS, Her YM, et al. (2012). Interleukin-21 promotes osteoclastogenesis in humans with rheumatoid arthritis and in mice with collagen-induced arthritis. Arthritis Rheum 64(3): 740-751.

Kwon RY, Temiyasathit S, Tummala P, Quah CC, Jacobs CR (2010). Primary cilium-dependent mechanosensing is mediated by adenylyl cyclase 6 and cyclic AMP in bone cells. FASEB J 24(8): 2859-2868.

Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, et al. (1998). Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93(2): 165-176.

Lafage-Proust MH, Prisby R, Roche B, Vico L (2010). Bone vascularization and remodeling. Joint Bone Spine 77(6): 521-524.

Lam J, Takeshita S, Barker JE, Kanagawa O, Ross FP, Teitelbaum SL (2000). TNF-alpha induces osteoclastogenesis by direct stimulation of macrophages exposed to permissive levels of RANK ligand. J Clin Invest 106(12): 1481-1488.

Lamothe B, Webster WK, Gopinathan A, Besse A, Campos AD, Darnay BG (2007). TRAF6 ubiquitin ligase is essential for RANKL signaling and osteoclast differentiation. Biochem Biophys Res Commun 359(4): 1044-1049.

Leaffer D, Sweeney M, Kellerman LA, Avnur Z, Krstenansky JL, Vickery BH, et al. (1995). Modulation of osteogenic cell ultrastructure by RS-23581, an analog of human parathyroid hormone (PTH)-related peptide-(1-34), and bovine PTH-(1-34). Endocrinology 136(8): 3624-3631.

Lee CA, Einhorn TA (2001). The Bone Organ System: Form and Function. In: Marcus R, Feldman D, Kelsey JL (ed)^(eds). Osteoporosis, 2nd edn. San Diego, CA: Academic Press.

Lee KL, Hoey DA, Spasic M, Tang T, Hammond HK, Jacobs CR (2014). Adenylyl cyclase 6 mediates loading-induced bone adaptation in vivo. FASEB J 28(3): 1157-1165.

Lee SH, Rho J, Jeong D, Sul JY, Kim T, Kim N, et al. (2006). v-ATPase V0 subunit d2-deficient mice exhibit impaired osteoclast fusion and increased bone formation. Nat Med 12(12): 1403-1409.

Lee Y (2013). The role of interleukin-17 in bone metabolism and inflammatory skeletal diseases. BMB Rep 46(10): 479-483.


250

Leipe J, Grunke M, Dechant C, Reindl C, Kerzendorf U, Schulze-Koops H, et al. (2010). Role of Th17 cells in human autoimmune arthritis. Arthritis Rheum 62(10): 2876-2885.

Leisen JC, Duncan H, Riddle JM, Pitchford WC (1988). The erosive front: a topographic study of the junction between the pannus and the subchondral plate in the macerated rheumatoid metacarpal head. J Rheumatol 15(1): 17-22.

Li H, Hong S, Qian J, Zheng Y, Yang J, Yi Q (2010). Cross talk between the bone and immune systems: osteoclasts function as antigen-presenting cells and activate CD4+ and CD8+ T cells. Blood 116(2): 210-217.

Li X, Qin L, Bergenstock M, Bevelock LM, Novack DV, Partridge NC (2007a). Parathyroid hormone stimulates osteoblastic expression of MCP-1 to recruit and increase the fusion of pre/osteoclasts. J Biol Chem 282(45): 33098-33106.

Li X, Udagawa N, Itoh K, Suda K, Murase Y, Nishihara T, et al. (2002). p38 MAPK-mediated signals are required for inducing osteoclast differentiation but not for osteoclast function. Endocrinology 143(8): 3105-3113.

Li Y, Toraldo G, Li A, Yang X, Zhang H, Qian WP, et al. (2007b). B cells and T cells are critical for the preservation of bone homeostasis and attainment of peak bone mass in vivo. Blood 109(9): 3839-3848.

Lian JB, Stein GS (2001). Osteoblast Biology. In: Marcus R, Feldman D, Kelsey J (ed)^(eds). Osteoporosis, edn, Vol. 1. Burlington, Mass: Elsevier Acadamic Press.

Ling Y, Rios HF, Myers ER, Lu Y, Feng JQ, Boskey AL (2005). DMP1 depletion decreases bone mineralization in vivo: an FTIR imaging analysis. J Bone Miner Res 20(12): 2169-2177.

Lisignoli G, Toneguzzi S, Grassi F, Piacentini A, Tschon M, Cristino S, et al. (2002). Different chemokines are expressed in human arthritic bone biopsies: IFN-gamma and IL-6 differently modulate IL-8, MCP-1 and rantes production by arthritic osteoblasts. Cytokine 20(5): 231-238.

Lisignoli G, Toneguzzi S, Pozzi C, Piacentini A, Grassi F, Ferruzzi A, et al. (1999). Chemokine expression by subchondral bone marrow stromal cells isolated from osteoarthritis (OA) and rheumatoid arthritis (RA) patients. Clin Exp Immunol 116(2): 371-378.


251

Little RD, Carulli JP, Del Mastro RG, Dupuis J, Osborne M, Folz C, et al. (2002). A mutation in the LDL receptor-related protein 5 gene results in the autosomal dominant high-bone-mass trait. Am J Hum Genet 70(1): 11-19.

Liu XH, Kirschenbaum A, Yao S, Levine AC (2005). Cross-talk between the interleukin-6 and prostaglandin E(2) signaling systems results in enhancement of osteoclastogenesis through effects on the osteoprotegerin/receptor activator of nuclear factor-{kappa}B (RANK) ligand/RANK system. Endocrinology 146(4): 1991-1998.

Livak KJ, Schmittgen TD (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25(4): 402-408.

Lloyd SA, Lewis GS, Zhang Y, Paul EM, Donahue HJ (2012). Connexin 43 deficiency attenuates loss of trabecular bone and prevents suppression of cortical bone formation during unloading. J Bone Miner Res 27(11): 2359-2372.

Loiselle AE, Paul EM, Lewis GS, Donahue HJ (2013). Osteoblast and osteocyte-specific loss of Connexin43 results in delayed bone formation and healing during murine fracture healing. J Orthop Res 31(1): 147-154.

Lourenco EV, La Cava A (2009). Cytokines in systemic lupus erythematosus. Curr Mol Med 9(3): 242-254.

Lubberts E, Joosten LA, van de Loo FA, Schwarzenberger P, Kolls J, van den Berg WB (2002). Overexpression of IL-17 in the knee joint of collagen type II immunized mice promotes collagen arthritis and aggravates joint destruction. Inflamm Res 51(2): 102-104.

Lubberts E, Koenders MI, Oppers-Walgreen B, van den Bersselaar L, Coenen-de Roo CJ, Joosten LA, et al. (2004). Treatment with a neutralizing anti-murine interleukin-17 antibody after the onset of collagen-induced arthritis reduces joint inflammation, cartilage destruction, and bone erosion. Arthritis Rheum 50(2): 650-659.

Luchin A, Purdom G, Murphy K, Clark MY, Angel N, Cassady AI, et al. (2000). The microphthalmia transcription factor regulates expression of the tartrate-resistant acid phosphatase gene during terminal differentiation of osteoclasts. J Bone Miner Res 15(3): 451-460.

Lucht U (1972). Osteoclasts and their relationship to bone as studied by electron microscopy. Z Zellforsch Mikrosk Anat 135(2): 211-228.

Ma CS, Deenick EK (2014). Human T follicular helper (Tfh) cells and disease. Immunol Cell Biol 92(1): 64-71.


252

Ma J, Zhu C, Ma B, Tian J, Baidoo SE, Mao C, et al. (2012). Increased frequency of circulating follicular helper T cells in patients with rheumatoid arthritis. Clin Dev Immunol 2012: 827480.

Macias MP, Fitzpatrick LA, Brenneise I, McGarry MP, Lee JJ, Lee NA (2001). Expression of IL-5 alters bone metabolism and induces ossification of the spleen in transgenic mice. J Clin Invest 107(8): 949-959.

Maecker HT, Trotter J (2006). Flow cytometry controls, instrument setup, and the determination of positivity. Cytometry A 69(9): 1037-1042.

Mak TW, Shahinian A, Yoshinaga SK, Wakeham A, Boucher LM, Pintilie M, et al. (2003). Costimulation through the inducible costimulator ligand is essential for both T helper and B cell functions in T cell-dependent B cell responses. Nat Immunol. 4(8): 765-772.

Malone AM, Anderson CT, Tummala P, Kwon RY, Johnston TR, Stearns T, et al. (2007). Primary cilia mediate mechanosensing in bone cells by a calcium-independent mechanism. Proc Natl Acad Sci U S A 104(33): 13325-13330.

Mangashetti LS, Khapli SM, Wani MR (2005). IL-4 inhibits bone-resorbing activity of mature osteoclasts by affecting NF-kappa B and Ca2+ signaling. J Immunol 175(2): 917-925.

Mann GN, Jacobs TW, Buchinsky FJ, Armstrong EC, Li M, Ke HZ, et al. (1994). Interferon-gamma causes loss of bone volume in vivo and fails to ameliorate cyclosporin A-induced osteopenia. Endocrinology 135(3): 1077-1083.

Manoury-Schwartz B, Chiocchia G, Bessis N, Abehsira-Amar O, Batteux F, Muller S, et al. (1997). High susceptibility to collagen-induced arthritis in mice lacking IFN-gamma receptors. J Immunol 158(11): 5501-5506.

Mansour A, Anginot A, Mancini SJ, Schiff C, Carle GF, Wakkach A, et al. (2011). Osteoclast activity modulates B-cell development in the bone marrow. Cell Res 21(7): 1102-1115.

Maricic M (2008). Osteomalacia. Curr Osteoporos Rep 6(4): 130-133.

Martin TJ, Quinn JM, Gillespie MT, Ng KW, Karsdal MA, Sims NA (2006). Mechanisms involved in skeletal anabolic therapies. Ann N Y Acad Sci 1068: 458-470.


253

Martin TJ, Sims NA (2005). Osteoclast-derived activity in the coupling of bone formation to resorption. Trends Mol Med 11(2): 76-81.

Matsuno H, Yudoh K, Katayama R, Nakazawa F, Uzuki M, Sawai T, et al. (2002). The role of TNF-alpha in the pathogenesis of inflammation and joint destruction in rheumatoid arthritis (RA): a study using a human RA/SCID mouse chimera. Rheumatology 41(3): 329-337.

Matsuo K, Irie N (2008). Osteoclast-osteoblast communication. Arch Biochem Biophys 473(2): 201-209.

Matsuo K, Otaki N (2012). Bone cell interactions through Eph/ephrin: bone modeling, remodeling and associated diseases. Cell Adh Migr 6(2): 148-156.

McClugage SG, Jr., McCuskey RS (1973). Relationship of the microvascular system to bone resorption and growth in situ. Microvasc Res 6(1): 132-134.

McInnes IB, Sieper J, Braun J, Emery P, van der Heijde D, Isaacs JD, et al. (2014). Efficacy and safety of secukinumab, a fully human anti-interleukin-17A monoclonal antibody, in patients with moderate-to-severe psoriatic arthritis: a 24-week, randomised, double-blind, placebo-controlled, phase II proof-of-concept trial. Ann Rheum Dis 73(2): 349-356.

Meghji S (1992). Bone remodelling. Br Dent J 172(6): 235-242.

Menton DN, Simmons DJ, Chang SL, Orr BY (1984). From bone lining cell to osteocyte--an SEM study. The Anatomical record 209(1): 29-39.

Meury T, Akhouayri O, Jafarov T, Mandic V, St-Arnaud R (2010). Nuclear alpha NAC influences bone matrix mineralization and osteoblast maturation in vivo. Mol Cell Biol 30(1): 43-53.

Midwood K, Sacre S, Piccinini AM, Inglis J, Trebaul A, Chan E, et al. (2009). Tenascin-C is an endogenous activator of Toll-like receptor 4 that is essential for maintaining inflammation in arthritic joint disease. Nat Med 15(7): 774-780.

Millan FA, Denhez F, Kondaiah P, Akhurst RJ (1991). Embryonic gene expression patterns of TGF beta 1, beta 2 and beta 3 suggest different developmental functions in vivo. Development 111(1): 131-143.

Miller SC, Jee WS (1987). The bone lining cell: a distinct phenotype? Calcif Tissue Int 41(1): 1-5.


254

Miyamoto T, Ohneda O, Arai F, Iwamoto K, Okada S, Takagi K, et al. (2001). Bifurcation of osteoclasts and dendritic cells from common progenitors. Blood 98(8): 2544-2554.

Miyamoto T, Suda T (2003). Differentiation and function of osteoclasts. Keio J Med 52(1): 1-7.

Miyazaki T, Katagiri H, Kanegae Y, Takayanagi H, Sawada Y, Yamamoto A, et al. (2000). Reciprocal role of ERK and NF-kappaB pathways in survival and activation of osteoclasts. J Cell Biol 148(2): 333-342.

Mizuno A, Amizuka N, Irie K, Murakami A, Fujise N, Kanno T, et al. (1998). Severe osteoporosis in mice lacking osteoclastogenesis inhibitory factor/osteoprotegerin. Biochem Biophys Res Commun 247(3): 610-615.

Mocsai A, Humphrey MB, Van Ziffle JA, Hu Y, Burghardt A, Spusta SC, et al. (2004). The immunomodulatory adapter proteins DAP12 and Fc receptor gamma-chain (FcRgamma) regulate development of functional osteoclasts through the Syk tyrosine kinase. Proc Natl Acad Sci U S A 101(16): 6158-6163.

Mohamed SG, Sugiyama E, Shinoda K, Taki H, Hounoki H, Abdel-Aziz HO, et al. (2007). Interleukin-10 inhibits RANKL-mediated expression of NFATc1 in part via suppression of c-Fos and c-Jun in RAW264.7 cells and mouse bone marrow cells. Bone 41(4): 592-602.

Morgan EF, Barnes GL, Einhorn TA (2008). The Bone Organ System: Form and Function. In: Marcus R, Feldman D, Nelson DA, Rosen CJ (ed)^(eds). Osteoporosis, 3rd edn, Vol. 1. Burlington, Mass.: Academic Press/Elsevier. p^pp 2 v. (xxiv, 1941 p., [1918] p. of plates).

Morgan ME, Flierman R, van Duivenvoorde LM, Witteveen HJ, van Ewijk W, van Laar JM, et al. (2005). Effective treatment of collagen-induced arthritis by adoptive transfer of CD25+ regulatory T cells. Arthritis Rheum 52(7): 2212-2221.

Morko J, Kiviranta R, Mulari MT, Ivaska KK, Vaananen HK, Vuorio E, et al. (2009). Overexpression of cathepsin K accelerates the resorption cycle and osteoblast differentiation in vitro. Bone 44(4): 717-728.

Morony S, Capparelli C, Lee R, Shimamoto G, Boone T, Lacey DL, et al. (1999). A chimeric form of osteoprotegerin inhibits hypercalcemia and bone resorption induced by IL-1beta, TNF-alpha, PTH, PTHrP, and 1, 25(OH)2D3. J Bone Miner Res 14(9): 1478-1485.


255

Moseley TA, Haudenschild DR, Rose L, Reddi AH (2003). Interleukin-17 family and IL-17 receptors. Cytokine Growth Factor Rev 14(2): 155-174.

Mossadegh-Keller N, Sarrazin S, Kandalla PK, Espinosa L, Stanley ER, Nutt SL, et al. (2013). M-CSF instructs myeloid lineage fate in single haematopoietic stem cells. Nature 497(7448): 239-243.

Mucida D, Park Y, Kim G, Turovskaya O, Scott I, Kronenberg M, et al. (2007). Reciprocal TH17 and regulatory T cell differentiation mediated by retinoic acid. Science 317(5835): 256-260.

Mukai T, Ishida S, Ishikawa R, Yoshitaka T, Kittaka M, Gallant R, et al. (2014). SH3BP2 Cherubism Mutation Potentiates TNF-alpha-Induced Osteoclastogenesis Via NFATc1 and TNF-alpha-Mediated Inflammatory Bone Loss. J Bone Miner Res 29(12): 2618-35.

Nagasawa T, Hirota S, Tachibana K, Takakura N, Nishikawa S, Kitamura Y, et al. (1996). Defects of B-cell lymphopoiesis and bone-marrow myelopoiesis in mice lacking the CXC chemokine PBSF/SDF-1. Nature 382(6592): 635-638.

Nakae S, Nambu A, Sudo K, Iwakura Y (2003). Suppression of immune induction of collagen-induced arthritis in IL-17-deficient mice. J Immunol 171(11): 6173-6177.

Nakashima T, Hayashi M, Fukunaga T, Kurata K, Oh-Hora M, Feng JQ, et al. (2011). Evidence for osteocyte regulation of bone homeostasis through RANKL expression. Nat Med 17(10): 1231-1234.

Nardelli DT, Burchill MA, England DM, Torrealba J, Callister SM, Schell RF (2004). Association of CD4+ CD25+ T cells with prevention of severe destructive arthritis in Borrelia burgdorferi-vaccinated and challenged gamma interferon-deficient mice treated with anti-interleukin-17 antibody. Clin Diagn Lab Immunol 11(6): 1075-1084.

Negishi-Koga T, Shinohara M, Komatsu N, Bito H, Kodama T, Friedel RH, et al. (2011). Suppression of bone formation by osteoclastic expression of semaphorin 4D. Nat Med 17(11): 1473-1480.

Neidhart M, Baraliakos X, Seemayer C, Zelder C, Gay RE, Michel BA, et al. (2009). Expression of cathepsin K and matrix metalloproteinase 1 indicate persistent osteodestructive activity in long-standing ankylosing spondylitis. Ann Rheum Dis 68(8): 1334-1339.


256

Nelms KA, Goodnow CC (2001). Genome-wide ENU mutagenesis to reveal immune regulators. Immunity 15(3): 409-418.

Ness TJ, Gebhart GF (1991). Interactions between visceral and cutaneous nociception in the rat. II. Noxious visceral stimuli inhibit cutaneous nociceptive neurons and reflexes. J Neurophysiol 66(1): 29-39.

Neve A, Corrado A, Cantatore FP (2013). Osteocalcin: skeletal and extra-skeletal effects. J Cell Physiol 228(6): 1149-1153.

Neve A, Corrado A, Cantatore FP (2012). Osteocytes: central conductors of bone biology in normal and pathological conditions. Acta Physiol 204(3): 317-330.

Nishihara M, Ogura H, Ueda N, Tsuruoka M, Kitabayashi C, Tsuji F, et al. (2007). IL-6-gp130-STAT3 in T cells directs the development of IL-17+ Th with a minimum effect on that of Treg in the steady state. Int Immunol 19(6): 695-702.

Noble BS, Stevens H, Loveridge N, Reeve J (1997). Identification of apoptotic changes in osteocytes in normal and pathological human bone. Bone 20(3): 273-282.

Nolte MA, Arens R, van Os R, van Oosterwijk M, Hooibrink B, van Lier RA, et al. (2005). Immune activation modulates hematopoiesis through interactions between CD27 and CD70. Nat Immunol 6(4): 412-418.

Nolte MA, van Olffen RW, van Gisbergen KP, van Lier RA (2009). Timing and tuning of CD27-CD70 interactions: the impact of signal strength in setting the balance between adaptive responses and immunopathology. Immunol Rev 229(1): 216-231.

Nowell MA, Williams AS, Carty SA, Scheller J, Hayes AJ, Jones GW, et al. (2009). Therapeutic targeting of IL-6 trans signaling counteracts STAT3 control of experimental inflammatory arthritis. J Immunol 182(1): 613-622.

Nurieva RI, Chung Y, Hwang D, Yang XO, Kang HS, Ma L, et al. (2008). Generation of T follicular helper cells is mediated by interleukin-21 but independent of T helper 1, 2, or 17 cell lineages. Immunity 29(1): 138-149.

Nurieva RI, Chung Y, Martinez GJ, Yang XO, Tanaka S, Matskevitch TD, et al. (2009). Bcl6 mediates the development of T follicular helper cells. Science 325(5943): 1001-1005.

O'Brien CA (2009). Control of RANKL gene expression. Bone 46(4): 911-9.


257

O'Neill HC, Griffiths KL, Periasamy P, Hinton RA, Hey YY, Petvises S, et al. (2011). Spleen as a site for hematopoiesis of a distinct antigen presenting cell type. Stem Cells Int 2011: 954275.

Oeckinghaus A, Hayden MS, Ghosh S (2011). Crosstalk in NF-kappaB signaling pathways. Nat Immunol 12(8): 695-708.

Ohshima S, Saeki Y, Mima T, Sasai M, Nishioka K, Nomura S, et al. (1998). Interleukin 6 plays a key role in the development of antigen-induced arthritis. Proc Natl Acad Sci U S A 95(14): 8222-8226.

Onyia JE, Bidwell J, Herring J, Hulman J, Hock JM (1995). In vivo, human parathyroid hormone fragment (hPTH 1-34) transiently stimulates immediate early response gene expression, but not proliferation, in trabecular bone cells of young rats. Bone 17(5): 479-484.

Orimo H (2010). The mechanism of mineralization and the role of alkaline phosphatase in health and disease. J Nippon Med Sch 77(1): 4-12.

Ortega N, Behonick DJ, Werb Z (2004). Matrix remodeling during endochondral ossification. Trends Cell Biol 14(2): 86-93.

Ozaki K, Kikly K, Michalovich D, Young PR, Leonard WJ (2000). Cloning of a type I cytokine receptor most related to the IL-2 receptor beta chain. Proc Natl Acad Sci U S A 97(21): 11439-11444.

Palmqvist P, Persson E, Conaway HH, Lerner UH (2002). IL-6, leukemia inhibitory factor, and oncostatin M stimulate bone resorption and regulate the expression of receptor activator of NF-kappa B ligand, osteoprotegerin, and receptor activator of NF-kappa B in mouse calvariae. J Immunol 169(6): 3353-3362.

Palumbo C (1986). A three-dimensional ultrastructural study of osteoid-osteocytes in the tibia of chick embryos. Cell Tissue Res 246(1): 125-131.

Palumbo C, Palazzini S, Marotti G (1990). Morphological study of intercellular junctions during osteocyte differentiation. Bone 11(6): 401-406.

Pappalardo A, Thompson K (2013). Activated gammadelta T cells inhibit osteoclast differentiation and resorptive activity in vitro. Clin Exp Immunol 174(2): 281-291.

Parfitt AM (2001). The bone remodeling compartment: a circulatory function for bone lining cells. J Bone Miner Res 16(9): 1583-1585.


258

Parfitt AM (1990). Bone-forming cells in clinical conditons. In: Hall BK (ed)^(eds). Bone, edn, Vol. 1: the osteoblast and osteocyte. Boca Raton, FL.: Telford Press and CRC Press. p^pp 351-429.

Parfitt AM, Mundy GR, Roodman GD, Hughes DE, Boyce BF (1996). A new model for the regulation of bone resorption, with particular reference to the effects of bisphosphonates. J Bone Miner Res 11(2): 150-159.

Park JS, Svetkauskaite D, He Q, Kim JY, Strassheim D, Ishizaka A, et al. (2004). Involvement of toll-like receptors 2 and 4 in cellular activation by high mobility group box 1 protein. J Biol Chem 279(9): 7370-7377.

Parrish-Novak J, Dillon SR, Nelson A, Hammond A, Sprecher C, Gross JA, et al. (2000). Interleukin 21 and its receptor are involved in NK cell expansion and regulation of lymphocyte function. Nature 408(6808): 57-63.

Pederson L, Ruan M, Westendorf JJ, Khosla S, Oursler MJ (2008). Regulation of bone formation by osteoclasts involves Wnt/BMP signaling and the chemokine sphingosine-1-phosphate. Proc Natl Acad Sci U S A 105(52): 20764-20769.

Pelton RW, Saxena B, Jones M, Moses HL, Gold LI (1991). Immunohistochemical localization of TGF beta 1, TGF beta 2, and TGF beta 3 in the mouse embryo: expression patterns suggest multiple roles during embryonic development. J Cell Biol 115(4): 1091-1105.

Perlot T, Penninger JM (2012). Development and function of murine B cells lacking RANK. J Immunol 188(3): 1201-1205.

Perng DW, Yang DM, Hsiao YH, Lo T, Lee OK, Wu MT, et al. (2012). miRNA-146a expression positively regulates tumor necrosis factor-alpha-induced interleukin-8 production in mesenchymal stem cells and differentiated lung epithelial-like cells. Tissue Eng Part A 18(21-22): 2259-2267.

Peterlik M, Hoffmann O, Swetly P, Klaushofer K, Koller K (1985). Recombinant gamma-interferon inhibits prostaglandin-mediated and parathyroid hormone-induced bone resorption in cultured neonatal mouse calvaria. FEBS Lett 185(2): 287-290.

Pettit AR, Chang MK, Hume DA, Raggatt LJ (2008). Osteal macrophages: a new twist on coupling during bone dynamics. Bone 43(6): 976-982.


259

Pettit AR, Ji H, von Stechow D, Muller R, Goldring SR, Choi Y, et al. (2001). TRANCE/RANKL knockout mice are protected from bone erosion in a serum transfer model of arthritis. Am J Pathol 159(5): 1689-1699.

Pettit AR, Walsh NC, Manning C, Goldring SR, Gravallese EM (2006). RANKL protein is expressed at the pannus-bone interface at sites of articular bone erosion in rheumatoid arthritis. Rheumatology 45(9): 1068-1076.

Pfaffl MW (2001). A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29(9): e45.

Piccinini AM, Midwood KS (2010). DAMPening inflammation by modulating TLR signalling. Mediators Inflamm 2010: pii: 672395.

Pickart CM, Eddins MJ (2004). Ubiquitin: structures, functions, mechanisms. BBA Clin 1695(1-3): 55-72.

Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, et al. (1999). Multilineage potential of adult human mesenchymal stem cells. Science 284(5411): 143-147.

Plotkin LI, Manolagas SC, Bellido T (2002). Transduction of cell survival signals by connexin-43 hemichannels. J Biol Chem 277(10): 8648-8657.

Pohlers D, Beyer A, Koczan D, Wilhelm T, Thiesen HJ, Kinne RW (2007). Constitutive upregulation of the transforming growth factor-beta pathway in rheumatoid arthritis synovial fibroblasts. Arthritis Res Ther 9(3): R59.

Pratama A, Ramiscal RR, Silva DG, Das SK, Athanasopoulos V, Fitch J, et al. (2013). Roquin-2 shares functions with its paralog Roquin-1 in the repression of mRNAs controlling T follicular helper cells and systemic inflammation. Immunity 38(4): 669-680.

Proetzel G, Pawlowski SA, Wiles MV, Yin M, Boivin GP, Howles PN, et al. (1995). Transforming growth factor-beta 3 is required for secondary palate fusion. Nat Genet 11(4): 409-414.

Qi B, Cong Q, Li P, Ma G, Guo X, Yeh J, et al. (2014). Ablation of Tak1 in osteoclast progenitor leads to defects in skeletal growth and bone remodeling in mice. Sci Rep 4: 7158.


260

Quinn JM, Itoh K, Udagawa N, Hausler K, Yasuda H, Shima N, et al. (2001). Transforming growth factor beta affects osteoclast differentiation via direct and indirect actions. J Bone Miner Res 16(10): 1787-1794.

Raggatt LJ, Partridge NC (2010). Cellular and molecular mechanisms of bone remodeling. J Biol Chem 285(33): 25103-25108.

Ramsdell F, Seaman MS, Miller RE, Tough TW, Alderson MR, Lynch DH (1994). gld/gld mice are unable to express a functional ligand for Fas. European journal of immunology 24(4): 928-933.

Rasheed AU, Rahn HP, Sallusto F, Lipp M, Muller G (2006). Follicular B helper T cell activity is confined to CXCR5(hi)ICOS(hi) CD4 T cells and is independent of CD57 expression. Eur J Immunol 36(7): 1892-1903.

Rasmussen H, Bordier P (1973). The physiological and cellular basis of metabolic bone disease. edn. Williams & Wilkins Co.: Baltimore,.

Rathinam C, Thien CB, Langdon WY, Gu H, Flavell RA (2008). The E3 ubiquitin ligase c-Cbl restricts development and functions of hematopoietic stem cells. Genes Dev 22(8): 992-997.

Ratkay LG, Zhang L, Tonzetich J, Waterfield JD (1993). Complete Freund's adjuvant induces an earlier and more severe arthritis in MRL-lpr mice. J Immunol 151(9): 5081-5087.

Reddy SV, Roodman GD (1998). Control of osteoclast differentiation. Crit Rev Eukaryot Gene Expr 8(1): 1-17.

Redlich K, Gortz B, Hayer S, Zwerina J, Doerr N, Kostenuik P, et al. (2004). Repair of local bone erosions and reversal of systemic bone loss upon therapy with anti-tumor necrosis factor in combination with osteoprotegerin or parathyroid hormone in tumor necrosis factor-mediated arthritis. Am J Pathol 164(2): 543-555.

Redlich K, Hayer S, Ricci R, David JP, Tohidast-Akrad M, Kollias G, et al. (2002). Osteoclasts are essential for TNF-alpha-mediated joint destruction. J Clin Invest 110(10): 1419-1427.

Rifkin BR, Heijl L (1979). The occurrence of mononuclear cells at sites of osteoclastic bone resorption in experimental periodontitis. J Periodontol 50(12): 636-640.


261

Ritchlin C, Haas-Smith SA (2001). Expression of interleukin 10 mRNA and protein by synovial fibroblastoid cells. J Rheumatol 28(4): 698-705.

Rodriguiz RM, Key LL, Jr., Ries WL (1993). Combination macrophage-colony stimulating factor and interferon-gamma administration ameliorates the osteopetrotic condition in microphthalmic (mi/mi) mice. Pediatr Res 33(4 Pt 1): 384-389.

Romas E, Sims NA, Hards DK, Lindsay M, Quinn JW, Ryan PF, et al. (2002). Osteoprotegerin reduces osteoclast numbers and prevents bone erosion in collagen-induced arthritis. Am J Pathol 161(4): 1419-1427.

Rosen V (2009). BMP2 signaling in bone development and repair. Cytokine Growth Factor Rev 20(5-6): 475-480.

Rothe L, Collin-Osdoby P, Chen Y, Sunyer T, Chaudhary L, Tsay A, et al. (1998). Human osteoclasts and osteoclast-like cells synthesize and release high basal and inflammatory stimulated levels of the potent chemokine interleukin-8. Endocrinology 139(10): 4353-4363.

Roths JB, Murphy ED, Eicher EM (1984). A new mutation, gld, that produces lymphoproliferation and autoimmunity in C3H/HeJ mice. J Exp Med 159(1): 1-20.

Rubin E, Farber JL (1999). Pathology. 3rd edn. Lippincott-Raven: Philadelphia.

Runyan RB, Potts JD, Weeks DL (1992). TGF-beta 3-mediated tissue interaction during embryonic heart development. Mol Reprod Dev 32(2): 152-159.

Rybak A, Fuchs H, Hadian K, Smirnova L, Wulczyn EA, Michel G, et al. (2009). The let-7 target gene mouse lin-41 is a stem cell specific E3 ubiquitin ligase for the miRNA pathway protein Ago2. Nat Cell Biol 11(12): 1411-1420.

Sack U, Kinne RW, Marx T, Heppt P, Bender S, Emmrich F (1993). Interleukin-6 in synovial fluid is closely associated with chronic synovitis in rheumatoid arthritis. Rheumatol Int 13(2): 45-51.

Salmond RJ, Filby A, Qureshi I, Caserta S, Zamoyska R (2009). T-cell receptor proximal signaling via the Src-family kinases, Lck and Fyn, influences T-cell activation, differentiation, and tolerance. Immunol Rev 228(1): 9-22.

Sanford LP, Ormsby I, Gittenberger-de Groot AC, Sariola H, Friedman R, Boivin GP, et al. (1997). TGFbeta2 knockout mice have multiple developmental defects that are


262

non-overlapping with other TGFbeta knockout phenotypes. Development 124(13): 2659-2670.

Sarrazin S, Mossadegh-Keller N, Fukao T, Aziz A, Mourcin F, Vanhille L, et al. (2009). MafB restricts M-CSF-dependent myeloid commitment divisions of hematopoietic stem cells. Cell 138(2): 300-313.

Sato K, Suematsu A, Okamoto K, Yamaguchi A, Morishita Y, Kadono Y, et al. (2006). Th17 functions as an osteoclastogenic helper T cell subset that links T cell activation and bone destruction. J Exp Med 203(12): 2673-2682.

Schaerli P, Willimann K, Lang AB, Lipp M, Loetscher P, Moser B (2000). CXC chemokine receptor 5 expression defines follicular homing T cells with B cell helper function. J Exp Med 192(11): 1553-1562.

Schmid P, Cox D, Bilbe G, Maier R, McMaster GK (1991). Differential expression of TGF beta 1, beta 2 and beta 3 genes during mouse embryogenesis. Development 111(1): 117-130.

Schoels MM, van der Heijde D, Breedveld FC, Burmester GR, Dougados M, Emery P, et al. (2013). Blocking the effects of interleukin-6 in rheumatoid arthritis and other inflammatory rheumatic diseases: systematic literature review and meta-analysis informing a consensus statement. Ann Rheum Dis 72(4): 583-589.

Schwandner R, Yamaguchi K, Cao Z (2000). Requirement of tumor necrosis factor receptor-associated factor (TRAF)6 in interleukin 17 signal transduction. J Exp Med 191(7): 1233-1240.

Serwold T, Ehrlich LI, Weissman IL (2009). Reductive isolation from bone marrow and blood implicates common lymphoid progenitors as the major source of thymopoiesis. Blood 113(4): 807-815.

Severe N, Dieudonne FX, Marie PJ (2013). E3 ubiquitin ligase-mediated regulation of bone formation and tumorigenesis. Cell Death Dis 4: e463.

Shalom-Barak T, Quach J, Lotz M (1998). Interleukin-17-induced gene expression in articular chondrocytes is associated with activation of mitogen-activated protein kinases and NF-kappaB. J Biol Chem 273(42): 27467-27473.

Shapiro F (2008). Bone development and its relation to fracture repair. The role of mesenchymal osteoblasts and surface osteoblasts. Eur Cell Mater 15: 53-76.


263

Shinar DM, Endo N, Halperin D, Rodan GA, Weinreb M (1993). Differential expression of insulin-like growth factor-I (IGF-I) and IGF-II messenger ribonucleic acid in growing rat bone. Endocrinology 132(3): 1158-1167.

Shu L, Zhang H, Boyce BF, Xing L (2013). Ubiquitin E3 ligase Wwp1 negatively regulates osteoblast function by inhibiting osteoblast differentiation and migration. J Bone Miner Res 28(9): 1925-1935.

Shull MM, Ormsby I, Kier AB, Pawlowski S, Diebold RJ, Yin M, et al. (1992). Targeted disruption of the mouse transforming growth factor-beta 1 gene results in multifocal inflammatory disease. Nature 359(6397): 693-699.

Silfversward CJ, Larsson S, Ohlsson C, Frost A, Nilsson O (2007). Reduced cortical bone mass in mice with inactivation of interleukin-4 and interleukin-13. J Orthop Res 25(6): 725-731.

Silfversward CJ, Sisask G, Larsson S, Ohlsson C, Frost A, Ljunggren O, et al. (2008). Bone formation in interleukin-4 and interleukin-13 depleted mice. Acta Orthop 79(3): 410-420.

Simonet WS, Lacey DL, Dunstan CR, Kelley M, Chang MS, Luthy R, et al. (1997). Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89(2): 309-319.

Simpson N, Gatenby PA, Wilson A, Malik S, Fulcher DA, Tangye SG, et al. (2010). Expansion of circulating T cells resembling follicular helper T cells is a fixed phenotype that identifies a subset of severe systemic lupus erythematosus. Arthritis Rheum 62(1): 234-244.

Sims NA, Jenkins BJ, Quinn JM, Nakamura A, Glatt M, Gillespie MT, et al. (2004). Glycoprotein 130 regulates bone turnover and bone size by distinct downstream signaling pathways. J Clin Invest 113(3): 379-389.

Sims NA, Quinn JM (2014). Osteoimmunology: oncostatin M as a pleiotropic regulator of bone formation and resorption in health and disease. Bonekey Rep 3: 527.

Smith-Garvin JE, Koretzky GA, Jordan MS (2009). T cell activation. Annu Rev Immunol 27: 591-619.

So H, Rho J, Jeong D, Park R, Fisher DE, Ostrowski MC, et al. (2003). Microphthalmia transcription factor and PU.1 synergistically induce the leukocyte receptor osteoclast-associated receptor gene expression. J Biol Chem 278(26): 24209-24216.


264

Solmesky LJ, Abekasis M, Bulvik S, Weil M (2009). Bone morphogenetic protein signaling is involved in human mesenchymal stem cell survival in serum-free medium. Stem Cells Dev 18(9): 1283-1292.

Spector TD, Hall GM, McCloskey EV, Kanis JA (1993). Risk of vertebral fracture in women with rheumatoid arthritis. Bmj 306(6877): 558.

Spolski R, Leonard WJ (2014). Interleukin-21: a double-edged sword with therapeutic potential. Nat Rev Drug Discov 13(5): 379-395.

St-Jacques B, Hammerschmidt M, McMahon AP (1999). Indian hedgehog signaling regulates proliferation and differentiation of chondrocytes and is essential for bone formation. Genes Dev 13(16): 2072-2086.

Steele DG, Bramblett CA (1988). The anatomy and biology of the human skeleton. 1st edn. Texas A&M University Press: College Station.

Su Y, Fu C, Ishikawa S, Stella A, Kojima M, Shitoh K, et al. (2008). APC is essential for targeting phosphorylated beta-catenin to the SCFbeta-TrCP ubiquitin ligase. Mol cell 32(5): 652-661.

Sun SC (2011). Non-canonical NF-kappaB signaling pathway. Cell Res 21(1): 71-85.

Sundaram K, Nishimura R, Senn J, Youssef RF, London SD, Reddy SV (2007). RANK ligand signaling modulates the matrix metalloproteinase-9 gene expression during osteoclast differentiation. Exp Cell Res 313(1): 168-178.

Taichman RS, Emerson SG (1994). Human osteoblasts support hematopoiesis through the production of granulocyte colony-stimulating factor. J Exp Med 179(5): 1677-1682.

Taichman RS, Reilly MJ, Emerson SG (1996). Human osteoblasts support human hematopoietic progenitor cells in vitro bone marrow cultures. Blood 87(2): 518-524.

Takahashi N, Udagawa N, Akatsu T, Tanaka H, Isogai Y, Suda T (1991). Deficiency of osteoclasts in osteopetrotic mice is due to a defect in the local microenvironment provided by osteoblastic cells. Endocrinology 128(4): 1792-1796.

Takahashi T, Kurihara N, Takahashi K, Kumegawa M (1986). An ultrastructural study of phagocytosis in bone by osteoblastic cells from fetal mouse calvaria in vitro. Arch Oral Biol 31(10): 703-706.


265

Takai H, Kanematsu M, Yano K, Tsuda E, Higashio K, Ikeda K, et al. (1998). Transforming growth factor-beta stimulates the production of osteoprotegerin/osteoclastogenesis inhibitory factor by bone marrow stromal cells. J Biol Chem 273(42): 27091-27096.

Takatsu K, Tominaga A, Harada N, Mita S, Matsumoto M, Takahashi T, et al. (1988). T cell-replacing factor (TRF)/interleukin 5 (IL-5): molecular and functional properties. Immunol Rev 102: 107-135.

Takayanagi H (2005). Mechanistic insight into osteoclast differentiation in osteoimmunology. J Mol Med 83(3): 170-179.

Takayanagi H (2007). Osteoimmunology: shared mechanisms and crosstalk between the immune and bone systems. Nat Rev Immunol 7(4): 292-304.

Takayanagi H, Kim S, Koga T, Nishina H, Isshiki M, Yoshida H, et al. (2002a). Induction and activation of the transcription factor NFATc1 (NFAT2) integrate RANKL signaling in terminal differentiation of osteoclasts. Dev Cell 3(6): 889-901.

Takayanagi H, Kim S, Matsuo K, Suzuki H, Suzuki T, Sato K, et al. (2002b). RANKL maintains bone homeostasis through c-Fos-dependent induction of interferon-beta. Nature 416(6882): 744-749.

Takayanagi H, Ogasawara K, Hida S, Chiba T, Murata S, Sato K, et al. (2000). T-cell-mediated regulation of osteoclastogenesis by signalling cross-talk between RANKL and IFN-gamma. Nature 408(6812): 600-605.

Takeshita S, Kaji K, Kudo A (2000). Identification and characterization of the new osteoclast progenitor with macrophage phenotypes being able to differentiate into mature osteoclasts. J Bone Miner Res 15(8): 1477-1488.

Talmage RV (1969). Calcium homeostasis--calcium transport--parathyroid action. The effects of parathyroid hormone on the movement of calcium between bone and fluid. Clin Orthop Relat Res 67: 211-223.

Talmage RV (1970). Morphological and physiological considerations in a new concept of calcium transport in bone. Am J Anat 129(4): 467-476.

Talmage RV, Cooper CW, Park HZ (1970). Regulation of calcium transport in bone by parathyroid hormone. Vitam Horm 28: 103-140.


266

Tanaka K, Hashizume M, Mihara M, Yoshida H, Suzuki M, Matsumoto Y (2014). Anti-interleukin-6 receptor antibody prevents systemic bone mass loss via reducing the number of osteoclast precursors in bone marrow in a collagen-induced arthritis model. Clin Exp Immunol 175(2): 172-180.

Tanaka-Kamioka K, Kamioka H, Ris H, Lim SS (1998). Osteocyte shape is dependent on actin filaments and osteocyte processes are unique actin-rich projections. J Bone Miner Res 13(10): 1555-1568.

Tang XL, Griffith JF, Qin L, Hung VW, Kwok AW, Zhu TY, et al. (2013a). SLE disease per se contributes to deterioration in bone mineral density, microstructure and bone strength. Lupus 22(11): 1162-1168.

Tang Y, Wu X, Lei W, Pang L, Wan C, Shi Z, et al. (2009). TGF-beta1-induced migration of bone mesenchymal stem cells couples bone resorption with formation. Nat Med 15(7): 757-765.

Tang Y, Xie H, Chen J, Geng L, Chen H, Li X, et al. (2013b). Activated NF-kappaB in bone marrow mesenchymal stem cells from systemic lupus erythematosus patients inhibits osteogenic differentiation through downregulating Smad signaling. Stem Cells Dev 22(4): 668-678.

Tasat DR, Orona NS, Mandalunis PM, Cabrini RL, Ubios AM (2007). Ultrastructural and metabolic changes in osteoblasts exposed to uranyl nitrate. Arch Toxicol 81(5): 319-326.

Tatsumi S, Ishii K, Amizuka N, Li M, Kobayashi T, Kohno K, et al. (2007). Targeted ablation of osteocytes induces osteoporosis with defective mechanotransduction. Cell Metab 5(6): 464-475.

Taylor AF, Saunders MM, Shingle DL, Cimbala JM, Zhou Z, Donahue HJ (2007). Mechanically stimulated osteocytes regulate osteoblastic activity via gap junctions. Am J Physiol Cell Physiol 292(1): C545-552.

Teitelbaum SL (2000). Bone resorption by osteoclasts. Science 289(5484): 1504-1508.

Temiyasathit S, Tang WJ, Leucht P, Anderson CT, Monica SD, Castillo AB, et al. (2012). Mechanosensing by the primary cilium: deletion of Kif3A reduces bone formation due to loading. PLoS One 7(3): e33368.

Terrier B, Costedoat-Chalumeau N, Garrido M, Geri G, Rosenzwajg M, Musset L, et al. (2012). Interleukin 21 correlates with T cell and B cell subset alterations in systemic lupus erythematosus. J Rheumatol 39(9): 1819-1828.


267

Teti A (2011). Bone development: overview of bone cells and signaling. Curr Osteoporos Rep 9(4): 264-273.

Thompson BJ, Jankovic V, Gao J, Buonamici S, Vest A, Lee JM, et al. (2008). Control of hematopoietic stem cell quiescence by the E3 ubiquitin ligase Fbw7. J Exp Med 205(6): 1395-1408.

Thomson BM, Mundy GR, Chambers TJ (1987). Tumor necrosis factors alpha and beta induce osteoblastic cells to stimulate osteoclastic bone resorption. J Immunol 138(3): 775-779.

Tondravi MM, McKercher SR, Anderson K, Erdmann JM, Quiroz M, Maki R, et al. (1997). Osteopetrosis in mice lacking haematopoietic transcription factor PU.1. Nature 386(6620): 81-84.

Tranquilli Leali P, Doria C, Zachos A, Ruggiu A, Milia F, Barca F (2009). Bone fragility: current reviews and clinical features. Clin Cases Miner Bone Metab 6(2): 109-113.

Troughton PR, Platt R, Bird H, el-Manzalawi E, Bassiouni M, Wright V (1996). Synovial fluid interleukin-8 and neutrophil function in rheumatoid arthritis and seronegative polyarthritis. Br J Rheumatol 35(12): 1244-1251.

Tu-Rapp H, Hammermuller A, Mix E, Kreutzer HJ, Goerlich R, Kohler H, et al. (2004). A proinflammatory role for Fas in joints of mice with collagen-induced arthritis. Arthritis Res Ther 6(5): R404-414.

Turner CH, Owan I, Alvey T, Hulman J, Hock JM (1998). Recruitment and proliferative responses of osteoblasts after mechanical loading in vivo determined using sustained-release bromodeoxyuridine. Bone 22(5): 463-469.

Turner RT, Iwaniec UT, Wong CP, Lindenmaier LB, Wagner LA, Branscum AJ, et al. (2013). Acute exposure to high dose gamma-radiation results in transient activation of bone lining cells. Bone 57(1): 164-173.

Udagawa N, Takahashi N, Jimi E, Matsuzaki K, Tsurukai T, Itoh K, et al. (1999). Osteoblasts/stromal cells stimulate osteoclast activation through expression of osteoclast differentiation factor/RANKL but not macrophage colony-stimulating factor: receptor activator of NF-kappa B ligand. Bone 25(5): 517-523.

Urist MR (1965). Bone: formation by autoinduction. Science 150(3698): 893-899.


268

Uzbekov RE, Maurel DB, Aveline PC, Pallu S, Benhamou CL, Rochefort GY (2012). Centrosome fine ultrastructure of the osteocyte mechanosensitive primary cilium. Microsc Microanal 18(6): 1430-1441.

Uzer G, Pongkitwitoon S, Ian C, Thompson WR, Rubin J, Chan ME, et al. (2014). Gap junctional communication in osteocytes is amplified by low intensity vibrations in vitro. PLoS One 9(3): e90840.

Vaananen HK, Zhao H, Mulari M, Halleen JM (2000). The cell biology of osteoclast function. J Cell Sci 113 ( Pt 3): 377-381.

Van bezooijen RL, Farih-Sips HC, Papapoulos SE, Lowik CW (1999). Interleukin-17: A new bone acting cytokine in vitro. J Bone Miner Res 14(9): 1513-1521.

van Hamburg JP, Asmawidjaja PS, Davelaar N, Mus AM, Colin EM, Hazes JM, et al. (2011). Th17 cells, but not Th1 cells, from patients with early rheumatoid arthritis are potent inducers of matrix metalloproteinases and proinflammatory cytokines upon synovial fibroblast interaction, including autocrine interleukin-17A production. Arthritis Rheum 63(1): 73-83.

Verborgt O, Gibson GJ, Schaffler MB (2000). Loss of osteocyte integrity in association with microdamage and bone remodeling after fatigue in vivo. J Bone Miner Res 15(1): 60-67.

Vermeire K, Heremans H, Vandeputte M, Huang S, Billiau A, Matthys P (1997). Accelerated collagen-induced arthritis in IFN-gamma receptor-deficient mice. J Immunol 158(11): 5507-5513.

Vinuesa CG, Cook MC, Angelucci C, Athanasopoulos V, Rui L, Hill KM, et al. (2005a). A RING-type ubiquitin ligase family member required to repress follicular helper T cells and autoimmunity. Nature 435(7041): 452-458.

Vinuesa CG, Tangye SG, Moser B, Mackay CR (2005b). Follicular B helper T cells in antibody responses and autoimmunity. Nat Rev Immunol 5(11): 853-865.

Visnjic D, Kalajzic Z, Rowe DW, Katavic V, Lorenzo J, Aguila HL (2004). Hematopoiesis is severely altered in mice with an induced osteoblast deficiency. Blood 103(9): 3258-3264.

Vogelzang A, McGuire HM, Yu D, Sprent J, Mackay CR, King C (2008). A fundamental role for interleukin-21 in the generation of T follicular helper cells. Immunity 29(1): 127-137.


269

Wada T, Nakashima T, Hiroshi N, Penninger JM (2006). RANKL-RANK signaling in osteoclastogenesis and bone disease. Trends Mol Med 12(1): 17-25.

Wajant H, Pfizenmaier K, Scheurich P (2003). Tumor necrosis factor signaling. Cell Death Differ 10(1): 45-65.

Walker EC, McGregor NE, Poulton IJ, Pompolo S, Allan EH, Quinn JM, et al. (2008). Cardiotrophin-1 is an osteoclast-derived stimulus of bone formation required for normal bone remodeling. J Bone Miner Res 23(12): 2025-2032.

Walsh MC, Kim N, Kadono Y, Rho J, Lee SY, Lorenzo J, et al. (2006). Osteoimmunology: interplay between the immune system and bone metabolism. Annu Rev Immunol 24: 33-63.

Walunas TL, Bakker CY, Bluestone JA (1996). CTLA-4 ligation blocks CD28-dependent T cell activation. J Exp Med 183(6): 2541-2550.

Wang X, Dong C (2013). The CD70-CD27 axis, a new brake in the T helper 17 cell response. Immunity 38(1): 1-3.

Waring P, Mullbacher A (1999). Cell death induced by the Fas/Fas ligand pathway and its role in pathology. Immunol Cell Biol 77(4): 312-317.

Watanabe-Fukunaga R, Brannan CI, Itoh N, Yonehara S, Copeland NG, Jenkins NA, et al. (1992). The cDNA structure, expression, and chromosomal assignment of the mouse Fas antigen. J Immunol 148(4): 1274-1279.

Wei S, Kitaura H, Zhou P, Ross FP, Teitelbaum SL (2005). IL-1 mediates TNF-induced osteoclastogenesis. J Clin Invest 115(2): 282-290.

Wein MN, Jones DC, Glimcher LH (2005). Turning down the system: counter-regulatory mechanisms in bone and adaptive immunity. Immunol Rev 208: 66-79.

Weitzmann MN (2013). The Role of Inflammatory Cytokines, the RANKL/OPG Axis, and the Immunoskeletal Interface in Physiological Bone Turnover and Osteoporosis. Scientifica 2013: 125705.

Weitzmann MN, Cenci S, Haug J, Brown C, DiPersio J, Pacifici R (2000). B lymphocytes inhibit human osteoclastogenesis by secretion of TGFbeta. J Cell Biochem 78(2): 318-324.


270

Weitzmann MN, Cenci S, Rifas L, Haug J, Dipersio J, Pacifici R (2001). T cell activation induces human osteoclast formation via receptor activator of nuclear factor kappaB ligand-dependent and -independent mechanisms. J Bone Miner Res 16(2): 328-337.

Wiesmann A, Phillips RL, Mojica M, Pierce LJ, Searles AE, Spangrude GJ, et al. (2000). Expression of CD27 on murine hematopoietic stem and progenitor cells. Immunity 12(2): 193-199.

Wilfinger WW, Mackey K, Chomczynski P (1997). Effect of pH and ionic strength on the spectrophotometric assessment of nucleic acid purity. BioTechniques 22(3): 474-476, 478-481.

Wong BR, Josien R, Lee SY, Sauter B, Li HL, Steinman RM, et al. (1997a). TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a dendritic cell-specific survival factor. J Exp Med 186(12): 2075-2080.

Wong BR, Rho J, Arron J, Robinson E, Orlinick J, Chao M, et al. (1997b). TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells. J Biol Chem 272(40): 25190-25194.

Wozney JM, Rosen V, Celeste AJ, Mitsock LM, Whitters MJ, Kriz RW, et al. (1988). Novel regulators of bone formation: molecular clones and activities. Science 242(4885): 1528-1534.

Wu C, Orozco C, Boyer J, Leglise M, Goodale J, Batalov S, et al. (2009). BioGPS: an extensible and customizable portal for querying and organizing gene annotation resources. Genome Biol 10(11): R130.

Wu JY, Purton LE, Rodda SJ, Chen M, Weinstein LS, McMahon AP, et al. (2008a). Osteoblastic regulation of B lymphopoiesis is mediated by Gs{alpha}-dependent signaling pathways. Proc Natl Acad Sci U S A 105(44): 16976-16981.

Wu X, McKenna MA, Feng X, Nagy TR, McDonald JM (2003). Osteoclast apoptosis: the role of Fas in vivo and in vitro. Endocrinology 144(12): 5545-5555.

Wu Y, Humphrey MB, Nakamura MC (2008b). Osteoclasts - the innate immune cells of the bone. Autoimmunity 41(3): 183-194.

Xian L, Wu X, Pang L, Lou M, Rosen CJ, Qiu T, et al. (2012). Matrix IGF-1 maintains bone mass by activation of mTOR in mesenchymal stem cells. Nat Med 18(7): 1095-1101.


271

Xiao Y, Song JY, de Vries TJ, Fatmawati C, Parreira DB, Langenbach GE, et al. (2013). Osteoclast precursors in murine bone marrow express CD27 and are impeded in osteoclast development by CD70 on activated immune cells. Proc Natl Acad Sci U S A 110(30): 12385-12390.

Xiao Z, Zhang S, Mahlios J, Zhou G, Magenheimer BS, Guo D, et al. (2006). Cilia-like structures and polycystin-1 in osteoblasts/osteocytes and associated abnormalities in skeletogenesis and Runx2 expression. J Biol Chem 281(41): 30884-30895.

Xing L, Bushnell TP, Carlson L, Tai Z, Tondravi M, Siebenlist U, et al. (2002). NF-kappaB p50 and p52 expression is not required for RANK-expressing osteoclast progenitor formation but is essential for RANK- and cytokine-mediated osteoclastogenesis. J Bone Miner Res 17(7): 1200-1210.

Xu J, Tan JW, Huang L, Gao XH, Laird R, Liu D, et al. (2000). Cloning, sequencing, and functional characterization of the rat homologue of receptor activator of NF-kappaB ligand. J Bone Miner Res 15(11): 2178-2186.

Xu J, Wu HF, Ang ES, Yip K, Woloszyn M, Zheng MH, et al. (2009a). NF-kappaB modulators in osteolytic bone diseases. Cytokine Growth Factor Rev 20(1): 7-17.

Xu L, Feng X, Tan W, Gu W, Guo D, Zhang M, et al. (2013). IL-29 enhances Toll-like receptor-mediated IL-6 and IL-8 production by the synovial fibroblasts from rheumatoid arthritis patients. Arthritis Res Ther 15(5): R170.

Xu LX, Kukita T, Kukita A, Otsuka T, Niho Y, Iijima T (1995). Interleukin-10 selectively inhibits osteoclastogenesis by inhibiting differentiation of osteoclast progenitors into preosteoclast-like cells in rat bone marrow culture system. J Cell Physiol 165(3): 624-629.

Xu Z, Hurchla MA, Deng H, Uluckan O, Bu F, Berdy A, et al. (2009b). Interferon-gamma targets cancer cells and osteoclasts to prevent tumor-associated bone loss and bone metastases. J Biol Chem 284(7): 4658-4666.

Xynos ID, Hukkanen MV, Batten JJ, Buttery LD, Hench LL, Polak JM (2000). Bioglass 45S5 stimulates osteoblast turnover and enhances bone formation In vitro: implications and applications for bone tissue engineering. Calcif Tissue Int 67(4): 321-329.

Yagi M, Miyamoto T, Sawatani Y, Iwamoto K, Hosogane N, Fujita N, et al. (2005). DC-STAMP is essential for cell-cell fusion in osteoclasts and foreign body giant cells. J Exp Med 202(3): 345-351.


272

Yagi M, Ninomiya K, Fujita N, Suzuki T, Iwasaki R, Morita K, et al. (2007). Induction of DC-STAMP by alternative activation and downstream signaling mechanisms. J Bone Miner Res 22(7): 992-1001.

Yago T, Nanke Y, Ichikawa N, Kobashigawa T, Mogi M, Kamatani N, et al. (2009). IL-17 induces osteoclastogenesis from human monocytes alone in the absence of osteoblasts, which is potently inhibited by anti-TNF-alpha antibody: a novel mechanism of osteoclastogenesis by IL-17. J Cell Biochem 108(4): 947-955.

Yamaguchi A, Komori T, Suda T (2000). Regulation of osteoblast differentiation mediated by bone morphogenetic proteins, hedgehogs, and Cbfa1. Endocr Rev 21(4): 393-411.

Yang X, Castilla LH, Xu X, Li C, Gotay J, Weinstein M, et al. (1999). Angiogenesis defects and mesenchymal apoptosis in mice lacking SMAD5. Development 126(8): 1571-1580.

Yao Z, Li P, Zhang Q, Schwarz EM, Keng P, Arbini A, et al. (2006). Tumor necrosis factor-alpha increases circulating osteoclast precursor numbers by promoting their proliferation and differentiation in the bone marrow through up-regulation of c-Fms expression. J Biol Chem 281(17): 11846-11855.

Yasui T, Kadono Y, Nakamura M, Oshima Y, Matsumoto T, Masuda H, et al. (2011). Regulation of RANKL-induced osteoclastogenesis by TGF-beta through molecular interaction between Smad3 and Traf6. J Bone Miner Res 26(7): 1447-1456.

Yasumoto K, Okamoto S, Mukaida N, Murakami S, Mai M, Matsushima K (1992). Tumor necrosis factor alpha and interferon gamma synergistically induce interleukin 8 production in a human gastric cancer cell line through acting concurrently on AP-1 and NF-kB-like binding sites of the interleukin 8 gene. J Biol Chem 267(31): 22506-22511.

Ye J, Coulouris G, Zaretskaya I, Cutcutache I, Rozen S, Madden TL (2012). Primer-BLAST: a tool to design target-specific primers for polymerase chain reaction. BMC bioinformatics 13: 134.

Yokomizo T, Dzierzak E (2008). Fine-tuning of hematopoietic stem cell homeostasis: novel role for ubiquitin ligase. Genes Dev 22(8): 960-963.

Yoshida S, Ono M, Shono T, Izumi H, Ishibashi T, Suzuki H, et al. (1997). Involvement of interleukin-8, vascular endothelial growth factor, and basic fibroblast growth factor in tumor necrosis factor alpha-dependent angiogenesis. Mol Cell Biol 17(7): 4015-4023.


273

Yu D, Tan AH, Hu X, Athanasopoulos V, Simpson N, Silva DG, et al. (2007). Roquin represses autoimmunity by limiting inducible T-cell co-stimulator messenger RNA. Nature 450(7167): 299-303.

Yun TJ, Chaudhary PM, Shu GL, Frazer JK, Ewings MK, Schwartz SM, et al. (1998). OPG/FDCR-1, a TNF receptor family member, is expressed in lymphoid cells and is up-regulated by ligating CD40. J Immunol 161(11): 6113-6121.

Yun TJ, Tallquist MD, Aicher A, Rafferty KL, Marshall AJ, Moon JJ, et al. (2001). Osteoprotegerin, a crucial regulator of bone metabolism, also regulates B cell development and function. J Immunol 166(3): 1482-1491.

Zaiss MM, Axmann R, Zwerina J, Polzer K, Guckel E, Skapenko A, et al. (2007). Treg cells suppress osteoclast formation: a new link between the immune system and bone. Arthritis Rheum 56(12): 4104-4112.

Zaiss MM, Sarter K, Hess A, Engelke K, Bohm C, Nimmerjahn F, et al. (2010). Increased bone density and resistance to ovariectomy-induced bone loss in FoxP3-transgenic mice based on impaired osteoclast differentiation. Arthritis Rheum 62(8): 2328-2338.

Zeng R, Spolski R, Casas E, Zhu W, Levy DE, Leonard WJ (2007). The molecular basis of IL-21-mediated proliferation. Blood 109(10): 4135-4142.

Zhang DE, Hetherington CJ, Chen HM, Tenen DG (1994). The macrophage transcription factor PU.1 directs tissue-specific expression of the macrophage colony-stimulating factor receptor. Mol Cell Biol 14(1): 373-381.

Zhang F, Tanaka H, Kawato T, Kitami S, Nakai K, Motohashi M, et al. (2011a). Interleukin-17A induces cathepsin K and MMP-9 expression in osteoclasts via celecoxib-blocked prostaglandin E2 in osteoblasts. Biochimie 93(2): 296-305.

Zhang H, Wu C, Matesic LE, Li X, Wang Z, Boyce BF, et al. (2013a). Ubiquitin E3 ligase Itch negatively regulates osteoclast formation by promoting deubiquitination of tumor necrosis factor (TNF) receptor-associated factor 6. J Biol Chem 288(31): 22359-22368.

Zhang SY, Boisson-Dupuis S, Chapgier A, Yang K, Bustamante J, Puel A, et al. (2008). Inborn errors of interferon (IFN)-mediated immunity in humans: insights into the respective roles of IFN-alpha/beta, IFN-gamma, and IFN-lambda in host defense. Immunol Rev 226: 29-40.


274

Zhang Y, Chang C, Gehling DJ, Hemmati-Brivanlou A, Derynck R (2001). Regulation of Smad degradation and activity by Smurf2, an E3 ubiquitin ligase. Proc Natl Acad Sci U S A 98(3): 974-979.

Zhang Y, Liu B, Ma Y, Jin B (2013b). Sema 4D/CD100-plexin B is a multifunctional counter-receptor. Cell Mol Immunol 10(2): 97-98.

Zhang Y, Paul EM, Sathyendra V, Davison A, Sharkey N, Bronson S, et al. (2011b). Enhanced osteoclastic resorption and responsiveness to mechanical load in gap junction deficient bone. PLoS One 6(8): e23516.

Zhang YE (2009). Non-Smad pathways in TGF-beta signaling. Cell Res 19(1): 128-139.

Zhang YH, Lin JX, Vilcek J (1990). Interleukin-6 induction by tumor necrosis factor and interleukin-1 in human fibroblasts involves activation of a nuclear factor binding to a kappa B-like sequence. Mol Cell Biol 10(7): 3818-3823.

Zhao C, Irie N, Takada Y, Shimoda K, Miyamoto T, Nishiwaki T, et al. (2006). Bidirectional ephrinB2-EphB4 signaling controls bone homeostasis. Cell Metab 4(2): 111-121.

Zhao M, Qiao M, Oyajobi BO, Mundy GR, Chen D (2003). E3 ubiquitin ligase Smurf1 mediates core-binding factor alpha1/Runx2 degradation and plays a specific role in osteoblast differentiation. J Biol Chem 278(30): 27939-27944.

Zheng ZH, Li XY, Ding J, Jia JF, Zhu P (2008). Allogeneic mesenchymal stem cell and mesenchymal stem cell-differentiated chondrocyte suppress the responses of type II collagen-reactive T cells in rheumatoid arthritis. Rheumatology 47(1): 22-30.

Zhou P, Kitaura H, Teitelbaum SL, Krystal G, Ross FP, Takeshita S (2006). SHIP1 negatively regulates proliferation of osteoclast precursors via Akt-dependent alterations in D-type cyclins and p27. J Immunol 177(12): 8777-8784.

Zhu H, Kavsak P, Abdollah S, Wrana JL, Thomsen GH (1999). A SMAD ubiquitin ligase targets the BMP pathway and affects embryonic pattern formation. Nature 400(6745): 687-693.

Zhu J, Garrett R, Jung Y, Zhang Y, Kim N, Wang J, et al. (2007). Osteoblasts support B-lymphocyte commitment and differentiation from hematopoietic stem cells. Blood 109(9): 3706-3712.


275

Zittermann A (2001). Effects of vitamin K on calcium and bone metabolism. Curr Opin Clin Nutr Metab Care 4(6): 483-487.

Zola H, Flego L (1992). Expression of interleukin-6 receptor on blood lymphocytes without in vitro activation. Immunology 76(2): 338-340.

APPENDIX

Appendix

277

Supplementary Figure 1 Female mutant sanroque mice display reduced bone

mass. MicroCT morphometric analysis of tibias from 83 ± 13 days-old female wildtype

and sanroque mice. Data is presented as mean ± SEM. * p<0.005; ** p<0.001; n.s. = no

significance in comparison to wildtype littermates.

sanroque

Wildtype

Appendix

278

Supplementary Figure 2 Left shift is observed in peripheral blood from sanroque

mutant mice. Geimsa-stained peripheral blood smears from wildtype and sanroque

mutant mice revealed increase number of immature leukocyte in sanroque mice

(personal communication with W/Prof. Wendy Erber of UWA School of Pathology and

Laboratory Medicine).

Wildtype sanroque

×20 ×20

×40 ×40

- END -

Documents

UNDERSTANDING THE OSTEOIMMUNOLOGICAL ROLE OF … · UNDERSTANDING THE OSTEOIMMUNOLOGICAL ROLE OF ROQUIN IN THE MAINTENANCE OF BONE HOMEOSTASIS BAY SIE LIM BCOM, BSC, BMEDSCI (HONS)