ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

S E R I E S E D I T O R S

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

PUBLICATIONS EDITOR

Professor Mikko Siponen

University Lecturer Elise Kärkkäinen

Professor Hannu Heusala

Professor Helvi Kyngäs

Senior Researcher Eila Estola

Information officer Tiina Pistokoski

University Lecturer Seppo Eriksson

University Lecturer Seppo Eriksson

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-9317-7 (Paperback)ISBN 978-951-42-9318-4 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1034

ACTA

Annina Sipola

OULU 2009

D 1034

Annina Sipola

EFFECTS OF VASCULAR ENDOTHELIAL GROWTH FACTOR (VEGF-A) AND ENDOSTATIN ON BONE

FACULTY OF MEDICINE,INSTITUTE OF BIOMEDICINE,DEPARTMENT OF ANATOMY AND CELL BIOLOGY,UNIVERSITY OF OULU

A C T A U N I V E R S I T A T I S O U L U E N S I SD M e d i c a 1 0 3 4

ANNINA SIPOLA

EFFECTS OF VASCULAR ENDOTHELIAL GROWTHFACTOR (VEGF-A) AND ENDOSTATIN ON BONE

Academic dissertation to be presented with the assent ofthe Faculty of Medicine of the University of Oulu forpublic defence in Auditorium A101 of the Department ofAnatomy and Cell Biology (Aapistie 7 A), on 4 December2009, at 2 p.m.

OULUN YLIOPISTO, OULU 2009

Copyright © 2009Acta Univ. Oul. D 1034, 2009

Supervised byProfessor Juha Tuukkanen

Reviewed byProfessor Heikki KrögerProfessor George Sándor

ISBN 978-951-42-9317-7 (Paperback)ISBN 978-951-42-9318-4 (PDF)http://herkules.oulu.fi/isbn9789514293184/ISSN 0355-3221 (Printed)ISSN 1796-2234 (Online)http://herkules.oulu.fi/issn03553221/

Cover designRaimo Ahonen

OULU UNIVERSITY PRESSOULU 2009

Sipola, Annina, Effects of vascular endothelial growth factor (VEGF-A) andendostatin on bone. Faculty of Medicine, Institute of Biomedicine, Department of Anatomy and Cell Biology,University of Oulu, P.O.Box 5000, FI-90014 University of Oulu, Finland Acta Univ. Oul. D 1034, 2009Oulu, Finland

AbstractAngiogenesis is essential for the replacement of cartilage by bone during skeletal growth andregeneration. Vascular endothelial growth factor-A (VEGF-A) is a key regulator of angiogenesiswhereas endostatin, a potent inhibitor of endothelial cell proliferation and migration, is a naturalantagonist of VEGF-A. The regulatory roles of these peptides in angiogenesis, bone formation andbone cells were investigated in this study.

In the present work we studied the effects of VEGF-A, delivered with an adenoviral vector, onthe recovery of bone drilling defects in rat femur. Our data confirm the important role of VEGF inbone healing and that adenoviral VEGF gene transfer may modify bone defect healing in a rodentmodel.

We studied the effects of VEGF-A and endostatin on bone resorption activity. It was found thatVEGF-A is a potent stimulator of bone resorption and osteoclast differentiation in vitro andendostatin can antagonize this stimulatory effect when acting directly on bone cells. This suggeststhat endostatin is indeed a regulator of bone resorption, but not a critical one.

In the present study we induced ectopic bone formation in the hamstring muscles of adult mice.The effects of VEGF-A and endostatin in the ectopic bone formation assay were evaluated by thesimultaneous delivery of both peptides with recombinant adenoviral vectors. It was found thatendostatin retards the cartilage phase in endochondral ossification that subsequently reduces boneformation. We conclude that bone growth and healing, which share features with ectopic boneformation, may be regulated by endostatin.

To confirm in vivo effects on bone formation we further investigated the effects of endostatinand VEGF-A on mouse pre-osteoblastic cells in vitro. Finally the effects of endostatin on bonewere studied in transgenic mouse lines overexpressing endostatin, and mice lacking collagenXVIII.

Keywords: bone regeneration, endostatin, osteogenesis, VEGF-A

To my family

6

7

Acknowledgements

This research was carried out in the Department of Anatomy and Cell Biology, at

the University of Oulu.

I would like to express my sincere gratitude to my supervisor, professor Juha

Tuukkanen, DDS, PhD. His guidance, positive attitude and patience have been

the most valuable throughout this project.

I also wish to express my appreciation to Docent Timo Hautala, MD, PhD for

constructive guidance.

I am grateful to Joanna Ilvesaro PhD, for teaching me the basics of lab work

and helping me in so many other ways at the beginning of this study.

I want to thank the former Chair of the Department of Anatomy and Cell

Biology, Professor Hannu Rajaniemi and also the present Chair Petri Lehenkari

for providing me this opportunity to work in the department.

Professor Heikki Kröger, MD, PhD and Professor George Sándor, MD, DDS,

PhD are acknowledged for their critical and valuable comments on this thesis,

and Dr. Deborah Kaska for the careful and expert English revision of this thesis.

I wish to thank all the co-authors: Elli Birr, Pekka Jalovaara, Timo Jämsä,

Ralf Pettersson, Taina Pihlajaniemi, Tatu Tarkka, Lotta Seppinen, Ylermi Soini,

Frej Stenbäck and Seppo Ylä-Herttuala for their contribution to the work.

I would like to warmly thank the whole staff of the Department of Anatomy,

especially the assistants who I worked with during those afternoon hours teaching

group works. I wish to thank Marja Juustila, Taina Poikela and Anna-Maija

Ruonala for their skillfull technical assistance and also Risto Helminen and Matti

Paadar for their help with the a number of computer problems.

I wish to thank all the members of Juha’s group for the opportunity to work

with them, especially my roommate Katri Nelo for all those nice conversations

and help with numerous practical matters, Mari Murtomaa, for pushing me

forward to finishing my thesis with your hilarious style, and Hanna Kokkonen for

sharing the practical arrangements of the dissertation and organization of the

dissertation party. I also warmly thank my dear friend Virpi Muhonen, for being a

priceless help in every work related problem I have had and sharing moments

outside work.

I warmly thank my great friends Antti and Marika, Jussi and Anu, Jone and

Pia for all the enjoyable moments that I have spent in your relaxing company. I

wish to thank my dear friends Heini and Jarno, Hannaleena and Mikko, Marja

and Jani for many unforgettable moments. Your positive energy makes me feel

8

that anything in this world is possible. I wish also to thank my almost like sisters

Suvi, Jonna and Johanna for all the fun times since we were little girls.

I want to express my gratitude to my lovely parents Annikki and Timo for all

the support and love they have given in my life. I owe my loving thanks to my

sister and soul mate Marika and her husband Antti, who is like a brother to me,

and their lovely son Luka for being always there for me. I wish everybody would

have friends like you.

My husband Saku deserves my deepest thanks for his love, support and

amazing never-ending optimism. There is nobody quite like him. Our son Oliver

has taught me what is the meaning of this life. I love you both.

This work was financially supported by the National Graduate School for

Musculo-Skeletal Diseases (TBGS), the Finnish Dental Society Apollonia and the

Emil Aaltonen Foundation.

Oulu, November 2009

9

Abbreviations

α-MEM α-modified minimum essential medium

αVβ3 Vitronectin receptor

ALP Alkaline phosphatase

BMP Bone morphogenetic protein

BMU Bone multicellular unit

BRU Bone remodeling unit

BSA Bovine serum albumin

BSP Bone sialoprotein

CFU Colony-forming unit

DMEM Dulbecco’s modified Eagle’s medium

ECM Extracellular matrix

ERK Extracellular signal-regulated kinase

FBS Fetal bovine serum

FGF Fibroblast growth factor

E Embryonic day

EC Endothelial cell

ENDO Endostatin

ES Endostatin

F-actin Filamentous actin

FBS Fetal bovine serum

FCS Fetal calf serum

FITC Fluorescein isothiocyanate

FSD Functional secretory domain

HSPG Heparan sulfate

IGF Insulin-like growth factor

IL Interleukin

M-CSF Macrophage colony-stimulating factor

MMP Matrix metalloproteinase

µM Micromolar, 10–6 g/mol

NCP Non-collagen protein

MSC Mesenchymal stem cell

NO Nitric oxide

OCN Osteocalcin

OPG Osteoprotegerin

OPN Osteopontin

10

Osx Osterix

PBS Phosphate-buffered saline

PCR Polymerase chain reaction

PDGF Platelet-derived growth factor

PFA Paraformaldehyde

pQCT Peripheral quantitative computed tomography

PTH Parathyroid hormone

RANK Receptor activator of NF-κB

RANKL Receptor activator of NF-κB ligand

RGD Arginine-glycine-aspartic acid

Runx2 Runt-related transcription factor 2

TGF-β Transforming growth factor β

TNF Tumor necrosis factor

TRACP Tartrate-resistant acid phosphatase

SZ Sealing zone

VEGF Vascular endothelial growth factor

WGA Wheat germ agglutinin

1,25(OH)2D3 1,25-dihydroxyvitamin D3

11

List of original publications

This thesis is based on the following articles, which are referred to in the text by

their Roman numerals:

I Tarkka T, Sipola A, Jämsä T, Soini Y, Ylä-Herttuala S, Tuukkanen J & Hautala T (2003) Adenoviral VEGF-A gene transfer induces angiogenesis and promotes bone formation in healing osseous tissue. J Gene Medicine 5: 560–566.

II Sipola A, Nelo K, Hautala T, Ilvesaro J & Tuukkanen J (2006) Endostatin inhibits VEGF-induced osteoclastic bone resorption in vitro. BMC Musculoskelet Disord 13(7): 56.

III Sipola A, Ilvesaro J, Birr E, Jalovaara P, Petterson RF, Stenbäck F, Ylä-Herttuala S, Hautala T & Tuukkanen J (2007) Endostatin inhibits endochondral ossification. J Gene Medicine: 1057–1064

IV Sipola A, Seppinen L, Pihlajaniemi T & Tuukkanen J (2009) Endostatin affects osteoblast behaviour in vitro, but collagenXVIII/endostatin is not essential for skeletal development in vivo. Calcified Tissue International 85(5): 412–420.

12

13

Contents

Abstract Acknowledgements 7 Abbreviations 9 List of original publications 11 Contents 13 1 Introduction 17 2 Review of the literature 19

2.1 Bone ........................................................................................................ 19 2.1.1 Bone formation ............................................................................. 20 2.1.2 Endochondral bone formation ...................................................... 20 2.1.3 Bone matrix .................................................................................. 21

2.2 Bone cells ................................................................................................ 22 2.2.1 Osteoblasts.................................................................................... 22 2.2.2 Differentiation of osteoblasts ........................................................ 23 2.2.3 Wnt-signaling and bone ................................................................ 24 2.2.4 Osteocytes .................................................................................... 26 2.2.5 Osteoclasts .................................................................................... 27 2.2.6 Osteoclast differentiation .............................................................. 27 2.2.7 Osteoclastic bone resorption ......................................................... 28

2.3 Bone remodeling ..................................................................................... 29 2.3.1 Bone remodeling cycle ................................................................. 31

2.4 Vascularization in bone ........................................................................... 33 2.4.1 Formation and structure of blood vessels ..................................... 33 2.4.2 Bone endothelium ......................................................................... 34

2.5 VEGF ...................................................................................................... 35 2.5.1 VEGF receptors ............................................................................ 37 2.5.2 VEGF in endochondral bone formation ....................................... 38 2.5.3 VEGF and chondrocytes ............................................................... 39 2.5.4 VEGF and osteoclasts/chondroclasts ............................................ 40 2.5.5 VEGF and osteoblasts .................................................................. 40

2.6 XVIII collagen and endostatin ................................................................ 41 2.6.1 XVIII collagen .............................................................................. 41 2.6.2 Endostatin ..................................................................................... 42 2.6.3 Collagen XVIII /Endostatin inhibits angiogenesis ....................... 43 2.6.4 Endostatin, bone and cartilage ...................................................... 44

14

2.6.5 Mechanism of endostatin action ................................................... 44 2.7 Gene therapy as a method of Growth Factor Delivery ............................ 45

3 Aims of the study 47 4 Materials and methods 49

4.1 Reagents (I-IV)........................................................................................ 49 4.1.1 Recombinant proteins (II, IV) ...................................................... 49 4.1.2 Recombinant adenoviruses (I, III) ................................................ 49

4.2 Experimental animals in vivo (I, III, IV) ................................................. 49 4.2.1 Bone defect healing model (I) ...................................................... 49 4.2.2 Endochondral ossification animal model (III) .............................. 50 4.2.3 Transgenic mice and knockout mice (IV) ..................................... 50

4.3 Cell Cultures ........................................................................................... 51 4.3.1 Osteoclast isolation and culture (II) .............................................. 51 4.3.2 Osteoclast differentiation assay (II) .............................................. 52 4.3.3 Osteoblast MC3T3 culture (IV) .................................................... 52

4.4 Histology, histomorphometry and immunohistochemistry (I, III,

IV) ........................................................................................................... 53 4.4.1 Haematoxylin-eosin (HE) staining (I, III, IV) .............................. 53 4.4.2 Immunostaining of capillaries by FVIII (I, III) ............................ 54 4.4.3 Collagen XVIII antibodies (IV) .................................................... 54 4.4.4 TRACP staining for identifying osteoclasts (II, III) ..................... 55 4.4.5 Area Measurement of Excavated Pits (II) .................................... 55

4.5 Biological assays ..................................................................................... 55 4.5.1 Determination of cell proliferation (WST-1) (IV) ........................ 55 4.5.2 Alizarin red S staining and quantification of calcium

deposition (IV) ............................................................................. 56 4.6 Radiographical analysis .......................................................................... 56

4.6.1 Radiographic evaluation of bone formation (III) .......................... 56 4.6.2 Peripheral quantitative computed tomography (pQCT) (I,

IV) ................................................................................................ 57 4.6.3 Faxitron X-ray imaging (IV) ........................................................ 57

4.7 Statistical analysis (I-IV) ......................................................................... 57 5 Results 59

5.1 Bone defect healing model (I) ................................................................. 59 5.1.1 Bone mineral content evaluated with peripheral

quantitative computed tomography (pQCT) (I) ............................ 59 5.1.2 Induction of vascularization (I) .................................................... 59

15

5.1.3 Histological evaluation (I) ............................................................ 59 5.2 Effects of VEGF-A and endostatin on osteoclastic bone

resorption and differentiation in vitro (II) ............................................... 62 5.2.1 Bone resorption assay (II) ............................................................ 62 5.2.2 Osteoclast differentiation (II) ....................................................... 62

5.3 Endochondral ossification model (III) .................................................... 62 5.3.1 Radiographic evaluation of bone formation (III) .......................... 62 5.3.2 Induction of vascularization (III) .................................................. 63 5.3.3 Bone histology (III) ...................................................................... 64 5.3.4 The number of cartilage-resorbing chondroclasts (III) ................. 66

5.4 The effects of endostatin on the MC3T3-E1 osteoblastic cell line.......... 66 5.4.1 Osteoblast proliferation in vitro .................................................... 66 5.4.2 Matrix mineralization in vitro ....................................................... 66

5.5 Transgenic mice and knockout mice ....................................................... 67 5.5.1 pQCT and faxitron results ............................................................ 67 5.5.2 The formation of secondary ossification centers .......................... 67 5.5.3 Localization of collagen XVIII..................................................... 68

6 Discussion 71 6.1 Adenoviral VEGF-A gene transfer induces angiogenesis and

promotes bone formation in healing osseous tissues (I) .......................... 71 6.2 Endostatin inhibits VEGF-A induced osteoclastic bone

resorption in vitro (II) .............................................................................. 73 6.3 Endostatin inhibits endochondral ossification (III) ................................. 74 6.4 Endostatin inhibits proliferation of MC3T3-E1 osteoblasts and

together with VEGF-A suppress mineralization (IV) .............................. 77 6.5 The role of collagen XVIII/endostatin on bone development in

vivo in transgenic mice overexpressing endostatin and mice

lacking collagen XVIII ............................................................................ 78 7 Conclusions 81 References 83 Original publications 103

16

17

1 Introduction

Throughout life, bones continuously alter their internal structure by remodeling

unlike other durable structures, such as teeth, tendons, and cartilage. Bone uses

this process to adapt to mechanical stress, repair damaged bone and maintain ion

homeostasis.

During bone growth, development, and remodeling, angiogenesis as well as

osteogenesis are closely associated processes. Physiological bone angiogenesis

occurs primarily during development and fracture healing. The majority of bones

in the vertebrate skeleton are formed by endochondral ossification, and blood

vessel invasion into cartilage is a crucial first step in the process whereby

cartilage is replaced by bone. This vasculature is also required for the transport of

nutrients and precursor cells, such as precursors of chondroclasts and osteoclasts

to the renewing bone tissue. Vascular endothelial growth factor, VEGF, is a key

mediator of physiological and pathophysiological angiogenesis. Bone remodeling

is intimately associated with vascular sinusoids and VEGF has an important role

in mediating interactions between the bone and the vasculature. Bone

endothelium has been found to signal in unique ways to other bone cells,

including bone- resorbing osteoclasts, bone forming osteoblasts, and their

progenitors.

Endostatin, a C-terminal fragment of collagen XVIII that is known to

antagonize VEGF-A, is found in basal membranes and is expressed by human

cartilage and fibrocartilage. Endostatin is capable of neutralizing the endothelial

cell migration, neovascularization, and vascular permeability induced by VEGF.

Resorbing osteoclasts express αv-integrins, which are receptors for endostatin,

thus providing a potential machanism for endostatin to control of osteoclast

function. Angiogenic and antiangiogenic factors play important roles in

ossification. It is hypothesized that endostatin expression by mature cartilage may

suppress osteoclasts while endostatin-deficient ossifying cartilage allows

osteoclastic activity and endochondral ossification.

The aim of this study was to explore the roles of VEGF-A and endostatin in

bone and also in processes that are not directly related to angiogenesis. We

approached these issues by using in vitro cell cultures and in vivo experiments.

We studied the role of VEGF, delivered with a first generation adenoviral vector,

in bone healing of drilling defects in rat femur. We also studied the effects of

VEGF-A and endostatin on bone resorption activity with a rat pit formation assay

and mouse osteoclast differentiation assay. Furthermore, we employed an

18

endochondral ossification model in the mouse hind muscle pouch stimulated by

BMPs, and overexpressed endostatin with or without VEGF-A using recombinant

adenoviral vectors. Finally, we investigated the influences of endostatin and

VEGF-A on mouse pre-osteoblastic cells (MC3T3-E1). The role of endostatin in

bone formation in vivo was also studied on transgenic mouse lines overexpressing

endostatin (ES-tg), and mice lacking collagen XVIII (Col18a1-/-).

19

2 Review of the literature

2.1 Bone

Bone has properties that make it ideal for its functions. It is mechanically strong

having very high tensile and compressive strength, but at the same time being

relatively light and elastic. Bone tissue has three main functions: First, it has a

mechanical function giving shape and rigidity to the body and providing

attachment sites for muscles as simple mechanical lever systems to produce body

movement. Second, bones provide a rigid framework that supports and protects

the soft organs of the body. The skull, for example, protects the brain. Third, it

serves as a metabolic tissue maintaining calcium and phosphate homeostasis.

Bone also provides an environment for hematopoiesis.

Anatomically, bones in the skeleton can be divided into flat bones, such as

skull bones, mandible and scapula and long bones, like humerus, femur and tibia.

In long bones, the shaft, composed mainly of compact bone, is called the

diaphysis and the region between the growth plate and the expanded end of the

bone, the epiphysis. The metaphysis is the junctional region between the growth

plate and the diaphysis. There are two morphologically distinct types of bone.

Cortical (compact) bone forms the outer thick and protective layer whereas

trabecular (cancellous) bone, found inside the bone, has a sponge-like structure

that serves a metabolic function due to its large surface area for remodeling.

(Marks & Odgren 2002)

Based on bone matrix arrangement, bone tissue can be classified as lamellar

or woven. Mature and cortical bone are typically lamellar in that with collagen

fibers are arranged in lamellae. In cortical bone, they are concentrically organized

around a neurovascular canal, termed a Haversian canal, commonly referred to as

osteons. Adjacent osteons are linked together by lateral canals. In trabecular bone

lamellae are arranged differently (parallel to each other) compared to those in the

denser compact bone. Woven bone, observed in the developing or immature bone,

is the type initially laid down during osteogenesis. Bones have an outer fibrous

sheath called the periosteum, and an inner surface, referred to as the endosteum,

which is in direct contact with the marrow. The periosteum covers all bone

surfaces except at the joints and is anchored to the bone by strong, collagenous

fibers called Sharpeys’ fibers that penetrate into the bone tissue. The endosteum is

a membranous sheath that lines the marrow cavity and also envelopes the surface

20

of cancellous bone and lines the blood vessel canals (Volkman’s canals) that run

through the bone. They both contain an abundant supply of blood vessels, nerves,

nerve endings and osteoprogenitor cells. Bone has a rich vascular supply,

receiving 10–20% of the cardiac output. Blood vessels not only supply oxygen

and nutrition to forming and growing bones, but they play an active role in bone

formation. (Marks & Odgren 2002)

2.1.1 Bone formation

Bone formation can occur through two mechanisms, by intramembranous or

endochondral ossification. In either case, bone develops by the replacement of a

preexisting connective tissue. The skeleton is formed by mesenchymal cells,

derived from cranial neural crest, somites and lateral plate mesoderm. These cells

condense and form the future skeletal elements. (Hall & Miyaki 1992, Zelzer &

Olsen 2005) Intramembranous ossification is the process by which a mesenchymal

template is replaced by bone. It takes place mainly during the development of

some flat bones of the skull and facial bones, when mesenchymal precursor cells

differentiate directly into bone-forming osteoblasts and begin to secrete bone

matrix. Endochondral ossification is responsible for the formation of bones of the

extremities and vertebral column. Bone tissue replaces a preexisting hyaline

cartilage, the anlagen of the future bone.

2.1.2 Endochondral bone formation

Endochondral ossification is initiated when mesenchymal progenitor cells

condense and differentiate into chondrocytes to form the calcified cartilaginous

scaffold for new bone formation. Cells in the connective tissue surrounding the

cartilage (perichondrium) differentiate into osteoblasts that deposit a mineralized

bone matrix and give rise to the bone collar, the cortical bone. Capillaries invade

the perichondrium that surrounds the future diaphysis and transform it into the

periosteum.

Hypertrophic chondrocytes in the central region of the cartilage secrete

VEGF to induce the invasion of blood vessels. Vascularization of the calcified

cartilage brings osteoclasts that resorb the cartilage, osteoblasts that deposit bone

matrix and also other mesenchymal cells into the cartilage from the

perichondrium (Carlevaro et al. 2000, Zelzer et al. 2001, Colnot et al. 2004). This

21

invasion results in the formation of a primary ossification center in the midshaft,

and cartilage is eroded and replaced by bone marrow and trabecular bone in the

ossification center. The process of endochondral ossification is seen in figure 1.

Later, secondary ossification centers of ossification appear within the

cartilaginous epiphysis by a mechanism that is similar to the formation of the

primary center (Karaplis 2002). Discrete layers of residual chondrocytes form

growth plates between the epiphyseal and metaphyseal bone centers to support

further postnatal longitudinal bone growth (Maes & Carmeliet 2008).

Endochondral ossification occurs during development as well as in pathological

osteogenesis such as fracture repair and ectopic ossification. The association of

angiogenesis, VEGF and endochondral bone formation will be discussed more in

chapter 2.5.2.

Fig. 1. A schematic presentation of an endochondral process leading to the formation

of the mouse tibia, where the differentiation of chondrocytes from mesenchymal cells

results in the formation of an avascular cartilage model of the future bone.

ED = embryonic day. Modified from (Zelzer & Olsen 2005).

2.1.3 Bone matrix

Bone tissue consists of mineralized extracellular matrix (ECM) and bone cells,

including bone-forming osteoblasts, bone-lining cells, osteocytes and bone-

resorbing osteoclasts. The ECM can be divided to organic and inorganic

components, which constitutes 35% and inorganic which constitutes 65% of bone

tissue, respectively. Inorganic bone matrix consists mainly of hydroxyapatite. The

22

association of these substances gives bone its hardness and resistance. The

organic matrix is composed primarily of the most abundant protein in the human

body, type I collagen, and several non-collagenous proteins. Type I collagen is the

major protein in bone matrix, representing about 90% of the organic matrix. Type

I collagen is a triple-helical structure, which contains two α1 chains and one α2

chain. These assemble into fibrils, which are then arrenged in layers. The network

of type I collagen fibers provides the structure on which bone mineral is

deposited. The remaining 10–15% of the bone total protein is composed of non-

collagenous proteins (NCPs), which can be divided into cell-attachment proteins,

proteoglycans and gamma-carboxylated proteins.

The interaction of cells with the extracellular matrix is essential for their

anchorage, proliferation, migration and differentiation. Virtually all of the bone

matrix proteins are modified post-translationally to contain either N- or O-linked

oligosaccharides, being glycoproteins. In bone matrix, there are multiple

glycoproteins, including fibronectin, thrombospondin, vitronectin, osteoadherin,

osteopontin (OPN) and bone sialoprotein (BSP) that contain the integrin-binding

amino acid sequence Arg-Gly-Asp (RGD) and may therefore mediate cell

attachment to the bone surface (Robey 2002).

2.2 Bone cells

2.2.1 Osteoblasts

Osteoblasts are bone-forming cells located on the surface of bone or osteoid.

Morphologically, osteoblasts are cuboidal and columnar in shape and have an

extremely well developed rough endoplasmic reticulum, a large nucleus, and an

enlarged Golgi complex due to the high level of protein synthesis (Marks &

Odgren 2002). There are four maturational stages of osteoblast development:

preosteoblast, osteoblast, osteocyte and bone-lining cell (Aubin & Liu 1996).

Osteoblasts are polarized cells: they have an architecture in which different

structures are located at each end, based on the cell’s activity. Osteoblasts are

responsible for synthesizing the organic components of bone matrix, type I

collagen and non-collagenous proteins and regulate the formation of

hydroxyapatite crystals in the newly formed bone. The production of type I

collagen by active osteoblasts occurs after osteoblast precursor proliferation,

followed by alkaline phosphatase (ALP) expression that is needed for

23

mineralization of bone matrix. Several non-collagenous proteins are deposited in

the matrix, like bone sialoprotein (Bsp), osteopontin (OPN), osteocalcin (OCN)

and also proteoglycans and glycoproteins.

2.2.2 Differentiation of osteoblasts

Osteoblasts arise from the same mesenchymal stem cells (MSC) as chondroblasts,

adipocytes, myoblasts and fibroblasts (Bennet et al. 1991, Grigoriadis et al. 1988,

Yamaguchi et al. 1991). The proliferation and differentiation of cells of the

osteoblast lineage occur under the influence of a number of a transcription

factors, growth factors and hormones (Aubin & Triffit 2002).

The transcription factors, Runx2 (runt related transcription factor, also known

as core binding factor 1, CBFA1) and Osterix (Osx), which functions downstream

from Runx2 regulate osteoblast differentiation (Karsenty & Wagner 2002,

Karsenty 2003, Nakashima et al. 2002). In the absence of either Runx2 or osterix,

no osteoblasts are formed. Runx2 null mice have an almost perfectly patterned

skeleton composed of cartilage, but no bones (Komori et al. 1997, Otto et al. 1997). Also chondrocyte differentiation to hypertrophy is diminished, except in

the distal appendicular skeleton (tibia-fibula, radius-ulna), where chondrocytes do

differentiate to hypertrophy. In addition, there is no blood vessel invasion into

hypertrophic cartilage in Runx2 null mice (Inada et al. 1999, Kim et al. 1999,

Zelzer & Olsen 2005). Runx2 can be phosphorylated and activated by the

mitogen-activated protein kinase (MAPK) pathway. This pathway can be

stimulated by a variety of signals, including bone morphogenic proteins (BMPs).

Bone morphogenetic proteins (BMPs), members of a family of secreted growth

factors, provide important tissue-specific signals to preosteoblasts that are

essential for full osteogenic differentiation (Chen et al. 2004, Mundy 2002). The

MAPK pathway can be also stimulated by signals initiated by the extracellular

matrix (ECM), fibroblast growth factor (FGF-2), mechanical loading and

hormones such as parathyroid hormone (PTH) (Franceschi & Xiao 2003). Many studies have highlighted the important role of Wnt signaling in bone

formation. The formation and proliferation of preosteoblast cells requires

signaling through the Wnt-frizzled-Lrp5 (low density lipoprotein receptor-related

protein) β-catenin signaling pathway (He et al. 2004).

A number of factors support the growth and differentiation of

osteoprogenitors and the commitment of mesenchymal stem cells to an osteoblast

lineage. This regulation is a complex phenomenon taking place in multiple points

24

in the maturational sequence. The hormones, growth factors and cytokines

participating in this process may have biphasic or opposite activities at different

times or on different bone compartments (Aubin et al. 2006). FGFs, TGFβ, and

BMPs are potent osteoinductive factors at the early stages of osteoprogenitor

differentiation and proliferation. In addition, PTH, PTHRrP, IGFs,

glucocorticoids, vitamin D, and estrogen are important regulators of osteoblast

differentiation and maturation (Aubin & Triffitt 2002, Aubin et al. 2006, Hill

1998).

2.2.3 Wnt-signaling and bone

Wnts are a family of 19 secreted proteins that participate in the development and

maintenance of many organs and tissues, including bone (Wodarz & Nusse 1998,

Logan & Nusse 2004, Moon et al. 2004). Wnt binds to membrane receptor

complexes including low-density lipoprotein receptor-related protein (LRP)-5/6

as well as frizzled proteins (Logan & Nusse 2004, Moon et al. 2004). The

formation of this ligand-receptor complex initiates a number of signaling

cascades that include the canonical/beta-catenin pathway as well as several

noncanonical pathways. Canonical Wnt signaling has been reported to play a

significant role in the control of bone formation.

Briefly, the canonical pathway is transduced through stabilization of the β-

catenin protein, which is the key mediator of the Wnt pathway. Upon Wnt

activation it cannot be phosphorylated by GSK-3 leading to its cytoplasmic

accumulation and translocation into the nucleus where it triggers gene expression.

The absence of Wnt expression or inhibition of binding to its membrane receptors

leads to degradation of ß-catenin and inactivation of the signaling cascade (Liu et al. 2002).

Wnt signaling regulates bone formation through a number of mechanisms

and through different mechanisms at different stages of life (Krishnan et al. 2006). It controls mesenchymal stem cells and it regulates most aspects of

osteoblast physiology (Reya & Clevers 2005). It dictates osteoblast specification

from osteo-/chondroprogenitors and stimulates osteoblast proliferation. In

addition it enhances osteoblast and osteocyte survival (Khosla et al. 2008).

ß-catenin is required for the early and terminal differentiation proliferation and

function of osteoblasts (Hill et al. 2005, Rodda & McMahon 2006). The Wnt

pathway regulates the expression of certain transcription factors, such as Runx2,

Osterix and dlx5, and strongly stimulates osteoblastogenesis through the

25

interaction of various components (Bennett et al. 2005, Yavropoulou & Yovos

2007).

β-catenin has been shown to be essential for determining whether

mesenchymal progenitors become chondrocytes or osteoblasts. Precursors lacking

ß-catenin develop into chondrocytes (Hill et al. 2005, Day et al. 2005) but the

maturation of chondrocytes requires the presence of Wnt signaling (Church et al. 2002, Chimal-Monroy et al. 2002). Wnt signaling has a role in both embryonic

cartilage development and postnatal endochondral bone formation (Yang 2008).

Wnt signaling regulates osteoclast formation and bone resorption. β-catenin

regulates osteoclastogenesis through effects on the expression of osteoprotegerin

and RANKL (Glass et al. 2005, Holmen et al. 2005).

In clinical cases, mutations have been found in the Wnt genes, receptors and

inhibitors of the Wnt signaling pathway that are associated with changes in bone

formation and turnover. Wnt signaling also plays a role during bone regeneration

following fracture (Yavropoulou & Yovos 2007).

Fig. 2. A schematic presentation of the role of canonical Wnt signaling in skeletal

development. Plus signs indicate positive effects of Wnt; minus signs indicate

inhibitory effects of physiological canonical Wnt signaling. Wnt signaling facilitates

osteoblast specification from mesenchymal progenitors at the expense of

adipogenesis. It cooperates with Runx2 and osterix to maintain and promote

osteoblast maturation. Wnt signaling influences osteoclast maturation by regulating

RANKL levels in osteoblasts (Khosla et al. 2008). Modified from (Liu et al. 2008).

26

2.2.4 Osteocytes

Osteocytes are terminally differentiated bone-forming cells that comprises the

majority, 90–95% of bone cells (Bonewald 2008). When osteoblasts have

completed their matrix-forming function, approximately 50–70% die by

apoptosis, and the remainder are either incorporated into the newly formed

osteoid matrix and become osteocytes or remain on the surface as bone lining

cells (Dempster 2006). Bone lining cells can revert back to osteoblasts under

certain circumstances, for example after stimulation by PTH or mechanical force

(Dobnig & Turner 1995). Osteocytes have a very particular location in bone, they

are spaced throughout the mineralized matrix.

Dendritic-shaped osteocytes are highly branched cells that occupy small spaces

between lamellae, called lacunae. From these cell bodies, long cytoplasmic

processes radiate in all directions. Osteocytes form a unique three-dimensional

organization, that is a metabolically and electrically active network via

cytoplasmic processes that pass through the bone matrix via small channels, the

canaliculi, which course through the lamellae and interconnect neighboring

lacunae and the lining cells on the bone (Nijweide et al. 2002, Knothe Tate et al. 2004).

Several functions of osteocytes and their possible roles in the process of bone

remodeling have been proposed (Noble 2008). Based on the morphological

characteristics of osteocytes, the sensation of mechanical stress loaded onto bone

is suspected to be one of their functions. The flow of extracellular fluid in

response to mechanical forces throughout the canaliculi induces shear stress and

deformation of the cell membranes. It has been suggested that primary cilia

mediate mechanosensing in bone cells (Malone et al. 2007, Xiao et al. 2006). It

has also been hypothesized, that osteocytes have a role in regulating local bone

remodeling. A recent finding, that osteocytes are the source of sclerostin, a

molecule that regulates osteoblast function, supports this hypothesis (Noble

2008). In addition, osteocytes participate in calcium homeostasis (Nijweide et al. 2002).

The life span of osteocytes is probably largely determined by bone turnover:

it is thought that osteocytes are phagocytozed by osteoclasts during bone

resorption (Nijweide et al. 2002).

27

2.2.5 Osteoclasts

Osteoclasts are cells of hematopoietic origin (Suda et al. 1992). They function in

bone resorption during bone growth for the resorption of calcified cartilage and

modeling of growing bone. Resorption is also needed during fracture healing, as

well as tooth eruption.

Osteoclasts are giant (20–100 µm) cells, enableing them to cover a relatively

large matrix area and thus operate more efficiently (Bar-Shavit 2007). They are

multinuclear cells having an average of 4–20 nuclei, but around 100 nuclei are

found in osteoclasts in Paget's disease patients (Roodman & Windle 2005).

Osteoclasts are also characterized by numerous mitochondria that are needed to

provide the energy required for the resorption process (Mundy 1990). Osteoclasts

cycle between resorbing and nonresorbing phases. To resorb bone, they attach to

the bone matrix, their cytoskeleton reorganizes and they assume a highly

polarized morphology containing several different plasma membrane domains

(Baron 1989). The characteristics of osteoclasts also include the expression of

calcitonin receptors, vitronectin receptors and tartrate resistant acid phosphatase

(TRACP) (Takahashi et al. 2002)

2.2.6 Osteoclast differentiation

Osteoclasts are multinucleated bone resorbing cells derived from hematopoietic

cells of the monocyte-macrophage progenitor cell lineage in the bone marrow

(Suda et al. 1992, Takahashi et al. 2002). After proliferating in the bone marrow,

mononuclear preosteoclasts fuse into multinucleated cells. They reach the bone

surface through the blood circulation. Mononuclear precursors of osteoclasts also

circulate in the peripheral blood (Väänänen 2005). Under physiological

conditions, osteoclast formation is supported by cell-to-cell contact by

osteoblast/stromal cells, which express the receptor activator of nuclear factor

kappa B (NF-κB) ligand, RANKL, as a membrane bound factor (Suda et al. 1999). The differentiation of osteoclasts is dependent on the presence of RANKL

and macrophage colony-stimulating factor (M-CSF) that are expressed by

osteoblasts and stromal cells (Yasuda et al. 1998). RANKL together with M-CSF,

induces osteoclast differentiation without the need for supportive cells. RANKL

also increases the bone-resorbing activity of mature osteoclasts and enhances

their survival (Yasuda et al. 1998). M-CSF is crucial for the proliferation and

survival of precursor cells of osteoclasts as well as macrophages. Signals are

28

transmitted through the cFMS, which belongs to the receptor tyrosine kinase

superfamily. The op/op mouse lacks functional M-CSF and has osteoclast-

deficient osteopetrosis (Yoshida et al. 1990).

RANKL and RANK are members of the tumor necrosis factor (TNF) and

TNF receptor superfamilies. RANKL is a type II transmembrane protein found on

the surface of expressing cells and as a proteolytically released soluble form

(Lacey et al. 1998, Wong et al. 1997, Anderson et al. 1997). RANK is a type I

homotrimeric transmembrane protein.

Osteoprotegerin (OPG), also released by osteoblasts, but not exclusively, is a

soluble decoy receptor for RANKL. It blocks RANK-RANKL interaction and

thus protects the skeleton from excessive bone resorption (Simonet et al. 1997,

Lacey et al. 1998). OPG expression is regulated by most of the factors that induce

RANKL expression by osteoblasts. In general, upregulation of RANKL is

associated with downregulation of OPG. A recent finding suggests, that OPG

expression is regulated by Wnt/β-catenin signaling in osteoblasts, the same

pathway that regulates osteoblastic bone formation (Boyce et al. 2005, Glass et al. 2005, Holmen et al. 2005).

Systemic and local factors control the recruitment, differentiation and activity

of osteoclasts, mainly by affecting the RANK-RANKL-OPG system through

stromal cells and osteoblasts. Disorders in this system seem to affect the rate of

bone resorption in a several pathological states (Hofbauer & Heufelder 2001).

Parathyroid hormone (PTH), 1,25 dihydroxyvitamin D3, the biologically active

form of vitamin D3 tumor necrosis factor-α (TNF- α), and interleukins (ILs) have

been implicated as factors that enhance osteoclast formation and resorption

(Alsina et al. 1996). Estrogen, TGF-β, calcitonin, IGF and ILs have been shown

to have inhibitory effects (Roodman 2001). Recently, it was found that estrogen

prevents bone loss via the estrogen receptor α and induction of the Fas ligand

(Nakamura et al. 2007). Bone resorption is also regulated by the immune system,

where T-cell expression of RANKL may contribute to pathological conditions

such as periodontitis and autoimmune arthritis (Katagiri & Takahashi 2002,

Schett et al. 2005).

2.2.7 Osteoclastic bone resorption

After osteoclasts migrate to the resorption site, they initiate a multistep process

called the resorption cycle (Väänänen & Zhao 2002). First, mononuclear

precursors fuse to form a polykaryon and move to the site of resorption. Then,

29

osteoclasts attach tightly to the mineralized bone matrix and generate an isolated

resorption site. The part of osteoclast cell membrane, the sealing zone, seals the

cell onto the bone matrix (Väänänen & Horton 1995). The primary adhesion

mediating structures of osteoclasts are actin-rich structures known as podosomes

(Lakkakorpi et al. 1991, Jurdic et al. 2006). avß3 integrin activation has an

important role in the osteoclast attachment leading to podosome formation

(Väänänen 2005). The sealing zone surrounds the ruffled border, which is a

resorbing organelle. Bone demineralization involves acidification of the isolated

extracellular enviroment. pH ~4.5 is created by fusion of acidic vesicles with the

ruffled border and by an electrogenic proton pump (H+-ATPase) coupled to a Cl-

channel (Blair et al. 1989, Väänänen et al. 1990). In addition to these domains,

osteoclasts also have a basolateral membrane facing the bone marrow and a

fourth functional secretory domain (FSD) which is located in the upper part of the

cell opposite to the ruffled border (Bar-Shavit 2007).

Osteoclasts degrade both the mineral and organic component of bone matrix

in resorption lacunae under ruffled border membranes. The acidic milieu first

dissolve bone mineral (hydroxyapatite), exposing the demineralized organic

component of bone for lysosomal proteases. It seems, that cathepsin K is the main

bone matrix-degrading enzyme (Gowen et al. 1999). Metalloproteinases,

especially MMP9s, are also released into resorption lacuna to degrade collagen

and noncollagen proteins (Everts et al. 1998, Xia et al. 1999).

Bone degradation products are endocytosed, endocytic vesicles are

transported to the functional secretory domain and degradation products are

released at the cell’s antiresorptive surface (Nesbitt & Horton 1997, Salo et al. 1997). TRACP has been shown to be localized in transcytotic vesicles and able to

destroy collagen and other proteins (Halleen et al. 1999). Finally, the osteoclast

detaches from the bone and loses its polarized structure. Then, the osteoclast

either relocates to a new resorption site or undergoes apoptosis (Bar-Shavit 2007).

2.3 Bone remodeling

Bones continually alter their internal structure by remodeling. The remodeling

rate varies in different types of bones: trabecular bone is remodeled at a higher

rate than cortical bone in a healthy adult (Parfitt 2002). At any one time,

approximately 20% of the cancellous bone surface is undergoing remodeling, and

at any one surface location, remodeling will occur on average every 2-years

(Parfitt et al. 1997). The bone remodeling processes involve the resorption of the

30

old bone and the formation of new bone, which are coupled to one another and

are a strictly controlled. Most adult skeletal diseases are due to an excess of

osteoclastic activity, leading to an imbalance in bone remodeling which favors

resorption. Such diseases include osteoporosis, periodontal diseases, rheumatoid

arthritis, multiple myeloma, and metastatic cancers (Boyle et al. 2003).

Osteoporotic bone loss is the result of high bone turnover in which bone

resorption outpaces bone deposition (Rodan & Martin 2000, Teitelbaum 2007)

while osteopetrosis, which is a metabolic bone disease characterized by decreased

bone resorption, that leads to the accumulation of excessive amounts of bone.

Although having higher bone mass, osteopetrotic bones are easier to fracture

(Tuukkanen et al. 2000).

Bone remodeling occurs in discrete locations and involves a group of

different kinds of cells. The basic multicellular unit (BMU), which comprises

osteoblasts and osteoclasts within the bone-remodeling cavity, is located in close

proximity to a blood vessel, as seen in figure 3. In cortical bone, the capillary

grows along the excavated tunnel, and on trabecular surfaces, small capillaries are

frequently seen adjacent to osteoblasts (Ott 2000). Another model proposed for

the relationship of osteoblasts, osteoclasts, and capillaries, involves bone

remodeling compartments (BRCs). A BRC comprises the BMU, as well as the

layer of flat lining or stromal cells positive for osteoblast markers associated with

the capillary (Hauge et al. 2001, Parfitt 2000), as seen in figure 4. They are

connected to the osteocyte network through gap junctions. It has been thought

that BRCs may be structures through which mechanosensory signals from the

osteocyte network are transmitted to osteoclasts and osteoblasts so that they can

change their activity on trabecular surfaces (Eriksen et al. 2007).

A BMU is not a permanent structure. It forms in response to a signal or

stimulus, performs its function and disbands. The skeleton contains millions of

BMUs, all at different stages and they all are geographically and chronologically

separated from other BMUs (Hill 1998). The bone remodeling cycle involves a

complex series of steps that are highly regulated and always follows certain

phases. It starts with activation of osteoclast precursors, and osteoclastic

resorption. Then on the reversal phase it continues with apoptosis of osteoclasts

and smoothing of the erosion cavities by mononuclear cells. Finally,

mesenchymal cells differentiate into functional osteoblasts that lay down matrix

mineralization (Frost 1973, Frost 1986).

31

Fig. 3. Illustration of the basic multicellular (BMU) model of bone resorption in cortical

bone. Modified from (Cackowski & Roodman 2007).

2.3.1 Bone remodeling cycle

Activation is a continuous process in BMUs. When a BMU spreads, new surfaces

undergo activation. Activation involves the recruitment of osteoclast precursors to

the bone and their differentiation and fusion into functional osteoclasts. It is not

clear what attracts the resorptive cells to the remodeling site. It has been

suggested that osteoclast precursors (monocytes) are mobilized by chemotaxis,

and the chemoattractants responsible for this activity are derived from the bone

matrix, or in the case of collagen and osteocalcin, directly from the osteoblasts

that produce them (Malone et al. 1982, Bar-Shavit 2007). MMPs produced by

osteoclast lineages have been found to be important for the migration of precursor

cells (Delaisse et al. 2003).

Heino and colleagues have suggested that healthy osteocytes inhibit

resorption, and when these signals cease, the osteoclast precursors migrate toward

the site (Heino et al. 2002). Alternatively, it has been proposed that apoptotic

osteocytes may secrete regulatory factors that induce osteoclast differentiation

and initiate bone repair (Hedgecock et al. 2007).

During the resorption phase, osteoclasts remove both mineral and organic

components of the bone matrix (Blair et al. 1986). Multinucleated osteoclasts are

active for approximately 12 days (Parfitt et al. 1996), and after the osteoclasts

32

have finished resorption, the phase ends with osteoclast apoptosis followed by a

reversal phase, which lasts approximately 9 days (Reddy 2004).

During the reversal phase osteoblast-like cells remove the remaining organic

matrix from the resorption lacuna and it is populated by osteoblast precursors that

proliferate and differentiate into osteoblasts (Everts et al. 2002, Mulari et al. 2004). During reversal, all-important coupling signals are also sent out to

osteoblasts to replace the bone that has been removed (Olsen 2006).

After that, the osteoblasts begin to synthesize the osteoid and finally 15 days

after the organic matrix is secreted, it begins to mineralize. Osteoblasts secrete

small membrane bound vesicles, called matrix vesicles that establish suitable

conditions for initial mineral deposition by the concentration of calcium and

phosphate ions (Anderson 2003).

Fig. 4. Remodeling cycle. A) Activation of osteoclastic precursors. B) Resorption by

activated multinuclear osteoclasts. C) Reversal, when osteoblast like cells remove the

remaining organic matrix from the bottom of the resorption lacuna before deposting

new collagen fibrils. Osteoblast precursors proliferate and differentiate into

osteoblasts and then migrate into the resorption lacuna. D) Formation and

mineralization. Osteoblasts continue to deposit new bone matrix and mineralize

osteoid until the cavity is filled. E) Resting. Once osteoblasts are embedded in

osteoid, they mature into terminally differentiated osteocytes. Modified from (Matsuo

& Irie 2008).

33

2.4 Vascularization in bone

Bone is a highly vascularized tissue. The significance of capillary growth in bone

modeling and remodeling is well established in bone development and fracture

healing (Stevens & Williams 1999). The vasculature is required for the transport

of oxygen, nutrients, growth enhancing molecules and precursor cells, such as

precursors of chondroclasts and osteoclasts, for the renewing bone tissue.

The majority of bones in the vertebrate skeleton are formed by endochondral

ossification, and blood vessel invasion into the cartilage is an important

phenomenon in endochondral ossification. In embryonic skeletal tissue,

osteogenesis and angiogenesis are temporally related and a lack of VEGF activity

arrests the bone development (Gerber et al. 1999, Ensig et al. 2000). There are

numerous skeletal pathologies such as osteoporosis, osteopetrosis and

inflammatory bone loss that are related to modifications in blood supply (Brandi

& Collin-Osdoby 2006).

Angiogenesis is thought to depend on a finely tuned balance between

endogenous stimulators and inhibitors (Polverini 1995). Molecules that serve as

positive regulators of angiogenesis include fibroblast growth factor (FGF),

transforming growth factor (TGF)-α, TGF-β, hepatocyte growth factor (HGF),

tumor necrosis factor (TNF)-α, angiogenin, interleukin (IL)-8, and the

angiopoietins and particularly members of the vascular endothelial growth factor

(VEGF) family (Folkman & Shing 1992, Yancopoulos et al. 2000, Ferrara et al. 2003, Dai & Rabie 2007).

2.4.1 Formation and structure of blood vessels

The formation of a vascular network is implemented by both vasculogenesis and

angiogenesis. Vasculogenesis is a process whereby vessels are formed de novo

from endothelial cell (EC) precursors, known as angioblasts (Gonzalez-Crussi

1971). Angiogenesis is defined as the growth of new capillaries from a pre-

existing vascular network (Risau 1997). The blood vessel system is hierarchically

organized to deliver oxygen, soluble factors, and various types of cells to all

tissues in a carefully regulated manner. The inside of a blood vessel is lined with

endothelial cells, adhering to each other and to specific matrix molecules. The

endothelium functions as a barrier that limits the movement of cells and

molecules between the circulation and tissues (Risau 1995). The endothelial layer

is attached to the basement membrane (BM), which is composed of collagen IV,

34

XV, XVIII, laminins, heparan sulfate proteoglycans, nidogen-entactin and other

matrix components (Timpl 1996). Angiogenesis is an invasive process that

requires controlled activity of extracellular proteases and their inhibitors (Liotta et al. 1991). As a result the BM dissolves causing the EC to come in contact with

proteolytic fragments of the constituents of the basement-membrane.

Specialized smooth muscular cells called pericytes envelop the surface of the

vascular tube. The interaction between pericytes and the EC is important for the

maturation, remodeling and maintenance of the vascular system. Gap junctions

provide direct connections between the cytoplasm of pericytes and endothelial

cells, and they enable the exchange of ions and small molecules (Bergers & Song

2005). Pericytes and supporting mural cells sense the external environment

through integrin receptors that promote cell survival, growth, migration and tube

formation (Hall et al. 2006). Integrins have an important role in the angiogenic

activity of endothelial cells because their proliferation is anchorage-dependent.

Inhibitors of various integrins have all demonstrated anti-angiogenic effects. The

endothelium signals between surrounding cells through a host of humoral and

growth factors, cytokines and chemokines, reactive metabolites, and polarized

surface-associated molecules (Cleaver & Melton 2003). Pericytes have VEGF

receptors and are capable of differentiating to chondrocytes and osteoblasts

(Doherty & Canfied 1999). Thus they provide a pool of stem cells in

endochondral ossification and bone healing.

2.4.2 Bone endothelium

Bone endothelium has been found to signal to other bone cells, including

osteoclasts, osteoblasts, and their progenitors. Bone remodeling sites associate

intimately with vascular sinusoids, and histological findings suggest that

osteoblasts and osteoprogenitor cells develop concomitantly with endothelial cells

in blood vessels adjacent to sites where new bone is formed (Dai & Rabie 2007).

Osteoclasts, which are derived from hematopoietic precursors are present in both

the bone marrow and the peripheral circulation (Roodman 1999). They must

migrate through the endothelium to reach future sites of bone resorption. It has

been shown, that endothelial cells can stimulate the formation of osteoclasts by

several mechanisms, including RANKL expression, and also promote bone

resorption in vitro (Formigli et al. 1995, Collin-Osdoby et al. 2001) There is also

some evidence, that osteoclasts stimulate angiogenesis (Tanaka et al. 2007). It has

35

been shown that osteoblast lineage cells are also present in the circulation

(Eghbali-Fatourechi et al. 2005).

Vascular endothelial cells synthesize and secrete many factors known to have

an effect on bone cells, including M-CSF, RANKL, various growth factors,

cytokines, chemokines, prostanoids, free radicals, small peptides, adhesion

molecules, and matrix constituents. Additionally, bone endothelial cells are also

able to respond to bone modulators, including PTH, sex steroids, and

proinflammatory cytokines (Brandi et al. 1993, Brandi & Collin-Osdoby 2006).

Two key molecules that have well defined roles are VEGF and angiopoietins.

The receptors for both these ligands are predominantly expressed on ECs and

they regulate the proliferation and survival of these cells. For example,

angiopoietin-1 (Ang-1), expressed predominantly upon ECs, regulates the

proliferation and survival of endothelial cells. Ang-1 is expressed also in

osteoblasts. It binds to the tyrosine kinase receptor, Tie-2, which is expressed in

endothelial cells and also in hematopoietic stem cells (HSC) (Huang & Bao 2004,

Huang et al. 2007). Tie-2 positive HSC are localized to the bone surface where

they are in contact with Ang-1 expressing osteoblasts. It was recently discovered

that a subset of osteoblasts function as a key component of the HSC niche,

controlling HSC numbers. In addition to osteoblasts, HSC interact with other

stromal cells too, including ECs (Yin & Li 2006).

2.5 VEGF

Vascular endothelial growth factor, VEGF, generally referred to as VEGF-A is an

angiogenic factor that is important for vascular development and maintenance in

all mammalian organs and therefore a key mediator of physiological

angiogenesis, for example during embryogenesis, reproductive functions and

skeletal growth. In addition, VEGF-A has a crucial role in pathophysiological

angiogenesis that is associated with numerous malignant, ischemic, inflammatory,

infectious and immune disorders (Carmeliet 2005). Numerous different strategies

that target VEGF receptor signal transduction pathways have been designed to

inhibit pathological angiogenesis in several malignancies (Ferrara et al. 2003).

Additionally, there is also clinical interest in promoting angiogenesis in

conditions such as stroke and ischemia (Olsson et al. 2006).

VEGF is also called the vascular permeability factor (VPF), because in 1983

it was first discovered as a permeability factor based on its ability to induce

vascular leakage (Senger et al. 1983). In the late 1980s VEGF was discovered as

36

an angiogenic factor that was found to stimulate endothelial cells to proliferate,

migrate, and survive in a serum poor environment (Leung et al. 1989).

The importance of VEGF-A during angiogenesis has been demonstrated in

transgenic mouse gene deletion studies. Mice lacking a single Vegf allele resulted

in embryonic lethality, due to deficient and abnormal vascular development

(Carmeliet et al. 1996). Similar results were obtained with mice carrying

deletions of the VEGF receptor allele VEGFR-1 (Fong et al. 1995) or VEGFR-2

(Shalaby et al. 1995).

The VEGF family members are secreted dimeric glycoproteins, which belong

to a family of proteins that include VEGF-A, VEGF-B, VEGF-C, VEGF-D, and

placenta-like growth factor (PLGF). The predominant form, VEGF-A, is

alternatively spliced into at least five different isoforms in humans. VEGF121,

VEGF 145, VEGF 165, VEGF 189, and VEGF 206, that have different biological

activities such as different abilities to interact with the VEGF co-receptors

heparan sulfate proteoglycans (HSPGs) and neuropilin (Houck et al. 1991,

Poltorak et al. 2000). VEGF164 and VEGF188 show increased binding to heparin

sulfate containing proteoglycans present on the cell surface and extracellular

matrix. The VEGF120 splice form is completely soluble and unable to bind

heparin (Ferrara & Davis-Smyth 1997, Hall et al. 2006). Murine VEGF-As are

shorter than the corresponding human isoforms of 121, 145, 165, 189, and 206

amino acids (termed VEGF120, VEGF164, etc.)(Tamayose et al. 1996).

Proteolytic processing of VEGF-A splice variants affects their ability to interact

with the VEGF co-receptors (Lee et al. 2005).

VEGF165, used in this study, is physiologically the most abundant isoform of

VEGF-A (Tischer et al. 1991). VEGF120 is diffusible, VEGF188 binds to the cell

surface and extracellular matrix and VEGF164 combines these properties.

VEGF165 exists in both a diffusible form and as an extracellular matrix bound

form, which can be released or activated by plasmin, lactate (Kumar et al. 2007),

or matrix metalloproteinases (MMPs) (Lee et al. 2005, Breen 2007). Many major

growth factors including epidermal growth factor, TGF-a, TGF-ß, keratinocyte

growth factor, insulin-like growth factor-1, FGF, and platelet-derived growth

factor upregulate VEGF mRNA expression, suggesting that paracrine or autocrine

release of such factors cooperates with local hypoxia in regulating VEGF release

in the microenviroment (Ferrara et al. 2003).

37

2.5.1 VEGF receptors

The biological functions of VEGF are mediated through the binding to specific

receptors and co-receptors that are primarily expressed in the vascular and lymph-

vessel endothelial cells, but are also located on non-endothelial cell types

including monocytes and macrophages, hematopoietic stem cells, epithelial cells,

fibroblasts, smooth muscle cells, and myogenic precursor cells (satellite cells)

(Olsson et al. 2006, Breen 2007).

VEGF ligands bind to three receptor tyrosine kinases (RTKs) denoted VEGFR-1 (Flt-1), VEGFR-2 (KDR/Flk-1), and VEGFR-3 (Flt-4) as well as to co-

receptors. Co-receptors, including heparan sulfate proteoglycans (HSPGs) and

neuropilins (NRP), lack tyrosine kinase activity but may modify the binding to

tyrosine kinase containing VEGFRs or bind VEGF directly to signal a cellular

response (Breen 2007). All isoforms bind VEGFR1 and VEGFR2, whereas only

VEGF164 has been shown to bind NRP-1 (Soker et al. 1998, Neufeld et al. 2002).

The deletion of a single VEGF allele or receptor gene allele, VEGFR-1 or

VEGFR-2, has been shown to result in fatal vascular defects in the embryo (Fong

et al. 1995, Shalaby et al. 1995, Carmeliet et al. 1996, Ferrara et al. 1996).

VEGFR1 is expressed on many types of endothelial cells, monocytes (Barleon et al. 1996) and dendritic progenitors (Wu et al. 1996). VEGFR1 ligands are

chemoattractive for monocytes and osteoclasts (Ensig et al. 2000, Clauss et al. 1990). VEGFR1 acts as a decoy through its ability to bind VEGF-A and regulate

the level of VEGF available to bind to VEGFR2 and initiate vessel formation

during embryogenesis. A soluble form of VEGFR1 also acts as a natural VEGF

trap since it is expressed through alternative transcription and binds to VEGF-A

to prevent angiogenesis (Maynard et al. 2003, Breen 2007). The expression of

VEGFR2 is almost exclusively limited to vascular endothelial cells and their

progenitors (Asahara et al. 1997). VEGFR2 is associated with the integrin-

dependent migration of endothelial cells, since it can form a complex with

integrin αVß3 and induce of endothelial cell proliferation, migration, and in vivo

angiogenesis (Hutchings et al. 2003, Dai & Rabie 2007). VEGFR3 has been

shown to be important for lymphatic- endothelial cell development and function

(Kaipainen et al. 1995).

38

2.5.2 VEGF in endochondral bone formation

Over 40 years ago Trueta and others concluded that blood vessels have an active

role in bone formation (Trueta & Amato 1960, Trueta & Buhr 1963, Trueta &

Trias 1961). Skeleton formation starts with mesenchymal condensation and the

transcription factor SOX9 has been found to be an important regulator in this

process (Bi et al. 1999). Inactivation of SOX9 results in complete absence of

mesenchymal condensation leading to failure of cartilage formation. Recent

findings provide evidence for the involvement of SOX9 in the regulation of

VEGF expression in condensing mesenchyme, whereas VEGF expression has

been found to regulate limb vasculature via a direct long-range mechanism

(Eshkar-Oren et al. 2009)

Harada and others were first to document the expression of VEGF in bone

tissue and osteoblasts (Harada et al. 1994). Vessel invasion has a key role in

endochondral bone formation and VEGF plays an important role in this

phenomenon. VEGF may mediate interactions between endothelial cells and bone

cells. Endothelial cells, chondrocytes (Carlevaro et al. 2000), and osteoblasts

(Wang et al. 1997) secrete endogenous VEGF that may stimulate chondrocytes

(Carlevaro et al.2000), osteoblasts (Mayr-Wohlwart et al. 2002, Deckers et al. 2000) and osteoclasts (Nakagawa et al. 2000). In endochondral ossification, an

avascular cartilage model is first formed. Immature and proliferating

chondrocytes secrete and express angiogenic inhibitors (Moses et al. 1999).

Chondrocytes in the cartilage template and later in the growth plate first

proliferate and then differentiate into mature hypertrophic cells. They express

high levels of VEGF (Gerber et al. 1999).

VEGF expression in the hypertrophic cartilage causes the upregulation of

VEGFR1 and VEGFR2 expression in the perichondral endothelium (Colnot et al. 2001, Zelzer et al. 2001, Zelzer et al. 2002) and thereupon induces vessels to

invade the hypertrophic cartilage from the perichondrium (Colnot et al. 2004).

Zelzer and Olsen have speculated that high levels of VEGF near hypertrophic

chondrocytes maintain endothelial cells in a sprouting mode. This has been

explained by the fact, that sprouting growth plate capillaries contain no basement

membranes or pericytes but, in the region where bone matrix is being

synthesized, endothelial cells are surrounded by BMs and pericytes (Zelzer &

Olsen 2005).

VEGF upregulation is maintained in the sprouting vessels under the growth

plates during all stages of bone development (Gerber et al. 1999). Concurrent

39

with vascular invasion, the hypertrophic cartilage matrix is degraded by invading

osteoclasts/chondroclasts. Osteoblasts and marrow cells start to populate and

ossify two eventual centers of ossification: the primary ossification center on

midshaft and a secondary center on the epiphysis (Maes & Carmeliet 2008). It has

been shown, that MMP-9 specifically regulates proteolysis of nonmineralized

cartilage and the release of ECM-bound VEGF, which has a direct chemotactic

affect on osteoclasts (Engsig et al. 2000).

A study by Gerber and colleagues showed that blocking the VEGF receptors

inhibited vascular invasion of the cartilage and bone formation and resorption in

juvenile mice. In their study, blood vessel invasion was almost completely

suppressed, concomitant with impaired trabecular bone formation and expansion

of the hypertrophic chondrocyte zone (Gerber et al. 1999). In addition, transgenic

mice having chondrocyte specific inactivation of VEGF-A, show embryonic

lethality, disturbance in the blood vessel development and aberrant endochondral

bone formation (Haigh et al. 2000). The specific contributions of the VEGF

isoforms have been studied by using models in mice expressing only one of the

major isoforms. These studies demonstrate a significant role of VEGF at both an

early and late stage of cartilage vascularization. The matrix-binding isoforms

mediate metaphyseal angiogenesis and thereby regulate both trabecular bone

formation and growth plate morphogenesis during endochondral bone formation.

The soluble isoforms diffuse to the perichondrium and are required for epiphyseal

vascularization and secondary ossification in growing bone (Zelzer et al. 2002,

Maes et al. 2002, Maes et al. 2004).

2.5.3 VEGF and chondrocytes

VEGF is expressed at high levels in hypertrophic chondrocytes and low levels in

maturing chondrocytes in rats (Carlevaro et al. 2000). VEGF-A secretion by

hypertrophic chondrocytes is not only a paracrine process that attracts and

stimulates proliferation of endothelial cells, but is also an autocrine process that is

needed for chondrocyte survival (Dai & Rabie 2007). Pufe and others have shown

that VEGF affects the proliferation of chondrocytes in a dose dependent manner

with the C28/12 immortalized chondrocyte cell line (Pufe et al. 2004a).

It has been suggested, that Runx2 is involved in the regulation of VEGF in

hypertrophic chondrocytes. There is no blood vessel invasion into hypertrophic

cartilage in Runx2 null mice, since the upregulated expression of VEGF is absent

in the hypertrophic cartilage of Runx2-deficient mice (Zelzer 2001). The culture

40

of chondrocytes under hypoxic conditions up-regulates VEGF-A expression via

HIF-1a (Pfander et al. 2003). VEGF expression via HIF-1a is also induced by

mechanical overload of cartilage discs (Pufe et al. 2004b).

2.5.4 VEGF and osteoclasts/chondroclasts

Osteoclasts share a common hematopoietic precursor with monocytes and

macrophages, and, like them, osteoclasts express FLT1 as their main VEGF

receptor (Barleon et al. 1996, Niida et al. 1999). However, the expression of KDR

by cultured osteoclasts has also been reported (Nakagawa et al. 2000, Tombran-

Tink & Barnstable 2004). VEGF induces osteoclast differentiation in combination

with RANK from spleen-or bone marrow-derived precursors in culture and

further, a single injection of recombinant human VEGF-A (rhVEGF-A) into

osteopetrotic op/op mice (lack of MCSF-1) induced osteoclast recruitment and

survival, while stimulating osteoclastic bone resorption (Niida et al. 1999). VEGF

has been shown to stimulate the resorption activity of osteoclasts in vitro

(Nakagawa et al. 2000) and to be a chemoattractant for osteoclasts invading into

developing long bones (Ensig et al. 2000). VEGF and RANK stimulate the

migration of the osteoclasts into hypertrophic cartilage through the extracellular

signal-regulated kinases 1 and 2 (ERK1/2) pathway (Henriksen et al. 2003, Min

et al. 2003). VEGF also enhances dose- and time-dependent osteoclastic bone

resorption in a culture of highly purified rabbit mature osteoclasts partially by

enhancing the survival of the cells (Nakagawa et al. 2000).

2.5.5 VEGF and osteoblasts

VEGF has been implicated in various aspects of osteoblast function. Osteoblasts

have been found to express VEGF and its receptors in vitro with the highest

expression levels being found at the late differentiation stages (Harada et al. 1994, Deckers et al. 2000, Harper et al. 2001).

VEGF is a central molecule in osteoblast-stimulated angiogenesis, having

both autocrine and paracrine mechanisms of action (Grellier et al. 2009). Several

factors with important roles in regulating bone formation also induce the

expression of VEGF by osteoblasts. Prostaglandins E1 and E2, BMP-4, BMP-6,

BMP-7, FGF-2, TGF-b, endothelin-1, IGF-1, and vitamin D3 can all induce

VEGF expression in osteoblasts by activating a variety of signaling pathways, but

41

VEGF expression is inhibited by bone catabolic factors such as glucocorticoids

(Zelzer and Olsen 2005).

VEGF acts as a potent chemoattractant for rat (Nakagawa et al. 2000) and

human (Mayr-Wohlwart et al. 2002) osteoblasts, as well as bone-marrow-derived

mesenchymal progenitor cells (Fiedler et al. 2005). A number of studies indicate,

that VEGF induces the differentiation of osteoblastic cells (Midy & Plouet 1994,

Deckers et al. 2000, Street et al. 2002). Zelzer and coworkers showed that VEGF

has a stimulatory effect on bone formation. During calvaria organ culture,

treatment with VEGF led to an increase in the thickness of parietal bone (Zelzer

et al. 2002).

Furthermore, it has been reported that adenovirus-mediated VEGF gene

transfer induced bone formation by increasing osteoblast number and osteoid

forming activity in vivo (Hiltunen et al. 2003). VEGFR-1 signaling has been

shown to be critical for osteoblast activity during bone formation through

participation in osteoclastogenesis, as well as in the maintenance of a bone

marrow hematopoiesis supportive microenvironment. Mice that were deficient in

flt-1 signaling were shown to have very depleted numbers of osteoclasts and

osteoblasts, and also a deficiency in bone marrow cavity formation (Niida et al. 2005). Recently, also VEGFR3 has been found to be important for osteoblast

differentiation (Orlandini et al. 2006).

2.6 XVIII collagen and endostatin

2.6.1 XVIII collagen

Type XVIII collagen has been localized in basement membrane (BM)

prominently in vascular and epithelial BM throughout the body (Muragi et al. 1995). Collagen XVIII is a heparan sulfate proteoglycan (HSPG) (Halfter et al. 1998). HSPGs are composed of a core protein and heparin sulfate (HS) side

chains. Collagen XVIII belongs to the multiplexin (multiple triple-helix domains

and interruptions) subclass of the non-fibrillar collagen family, having ten

interrupted triple-helical (COL) domains that vary in length, separated by 11 non-

collagenous (NC) regions, flanked by N- and C-terminal non-collagenous

domains (Erickson & Couchman 2000). Collagen XVIII is expressed as three

N-terminal variants. These are transcribed from two different promoters,

promoter 1 being responsible for the production of the short variant, and promoter

42

2 for the two longer variants. The middle variant is generated from the long

variant by alternative splicing (Muragi et al. 1995, Rehn & Pihlajaniemi 1995,

Saarela et al. 1998). The C terminus includes a trimerization region, a hinge

region, which provides protease-sensitive sites, and an anti-angiogenic fragment,

endostatin.

2.6.2 Endostatin

Endostatin is a 20kDa carboxy-terminal fragment of the non-collagenous domain

of the collagen XVIII α1 chain. It was originally identified from murine

hemangioendothelioma cells (EOMA) (O´Reilly et al. 1997).

Structurally, endostatin is characterized by compact globular folding with

surface exposed arginine-rich patches (Hohenester et al. 1998). They are involved

in binding to heparin sulfate, which in turn may be important for endostatin’s

anti-angiogenic function (Kreuger et al. 2002, Sasaki et al. 1999, Dixelius et al. 2003).

Endostatin is liberated from the non-collagenous domain by the action of

various proteases, including matrix metalloproteinases (MMP) Cathepsin L, and

the serine protease elastase (Wen et al. 1999, Felbor et al. 2000). The released

forms can be found in the vessel wall, in plateles and freely circulating.

According to a number of studies, the normal mean circulating endostatin plasma

level is in the range of 10–50ng/ml (Dixelius et al. 2003).

43

Fig. 5. A schematic presentation of the structure of collagen XVIII. Ten collagenous

domains are interrupted and flanked by 11 non-collagenous domains. These are

numbered from the carboxy-to the amino terminus, NC1- through NC-11. Collagen

XVIII is expressed as three N-terminal variants. These are transcribed from two

different promoters. Modified from Dixelius et al. 2003, Bix et al. 2005).

2.6.3 Collagen XVIII /Endostatin inhibits angiogenesis

The effects of endostatin are broad, since in endothelial cells it inhibits

endothelial cell proliferation (O´Reilly et al. 1997), migration (Yamaguchi et al. 1999), proteinase activity (Wickström et al. 2001) and vessel stabilization

(Dixelius et al. 2000). Furthermore, endostatin induces endothelial cell apoptosis

(Dhanabal et al. 1999, Dixelius et al. 2000). Endostatin also affects tumor

angiogenesis indirectly by suppressing the endogenous expression of VEGF

(Hajitou et al. 2002).

Collagen XVIII deficient mice survive and do not display obvious vascular

abnormalities or altered tumor growth (Fukai et al. 2002). They suffer from eye

abnormalities, having defects in almost all ocular structures that express collagen

XVIII. These abnormalities are similar to those observed in patients with

Knoblochs syndrome, which is caused by an inactivating mutation in the collagen

α1(XVIII) gene (Fukai et al. 2002).

44

2.6.4 Endostatin, bone and cartilage

Only little is known regarding endostatins role in bone and cartilage. Pufe and

others have shown that endostatin/collagen XVIII is expressed in human cartilage

and fibrocartilage and they suggest it might reflect the importance of endostatin

for the development and maintenance of avascular zones. According to their

hypothesis, endostatin expression by mature cartilage may suppress osteoclasts

while endostatin-deficient ossifying cartilage allows osteoclastic activity and

endochondral ossification (Pufe et al. 2004c). Lau and others found that

endostatin may have a role in the anabolic effect of mechanical stimuli on bone

since endostatin can prevent the effect of fluid shear stress on osteoblasts (Lau et al. 2006). There is evidence that endostatin inhibits Wnt-signaling, acting

possibly at the level of ß-catenin in the signaling pathway (Hanai et al. 2002).

2.6.5 Mechanism of endostatin action

The exact mechanism of endostatin mediated inhibition has not yet been

elucidated. Endostatin binds to a number of cell surface molecules, but the signal

transduction events following binding to these receptors that result in the

inhibition of angiogenesis have so far been poorly characterized (Wickström et al. 2005, Delaney et al. 2006). Endostatin modulates many intracellular pathways.

Endostatin has been shown to be a heparin binding protein (Hohenester et al. 1998, Sasaki et al. 1999). This suggests that endostatin may competitively inhibit

the binding of angiogenic growth factors to heparan sulfate proteoglycans, which

are known to act as co-receptors for a number of cytokines (Hohenester et al. 1998). Dixelius and colleagues showed that heparin binding is needed for the

anti-angiogenic effects of endostatin. Endostatin mutants lacking heparin-binding

ability failed to block growth factor induced angiogenesis, in contrast to wild –

type endostatin in the chicken chorioallantoic membrane assay (Dixelius 2000).

Endostatin binding with low affinity to the heparan sulfate proteoglycans

glypican-1 and glypican-4 or with a high affinity to an as yet unidentified receptor

has been reported (Karumanchi et al. 2001).

It has been shown, that endostatin’s anti-angiogenic effects are mediated at

least in part, by interactions with integrins. Integrins play an essential role in the

angiogenic activity of endothelial cells because their proliferation is anchorage-

dependent. Inhibitors of various integrins have all demonstrated anti-angiogenic

effects (Bix & Iozzo 2005). Inhibitory action by endostatin has been proposed to

45

involve binding to the receptor α5β1 (Sudhakar et al. 2003). According to studies

of Wickström and colleagues, endostatin interacts with endothelial cells α5β1

integrin and perhaps concomitantly, with cell surface, heparan sulfate

proteoglycans and caveolin (Wickström et al. 2002, Wickström et al. 2003),

which leads to complex signaling cascades (Wickström et al. 2002, Bix & Iozzo

2005).

Interestingly, endostatin blocks both the VEGFR1 and VEGFR2 receptors of

VEGF and by that mechanism inhibits VEGF signaling (Kim et al. 2002). Recent

evidence shows that endostatin can also bind to VEGFR3 (Kojima et al. 2008). In

addition, endostatin’s inhibitory action is involved with several signaling

pathways, including Wnt/β-catenin (Hanai et al. 2002) and ERK1/2 signaling

(Schmidt et al. 2006). The anti-angiogenic activity of endostatin also requires the

presence of endothelial cell E-selectin, an endothelial cell-specific membrane

glycoprotein that is important for leukocyte attachment (Yu et al. 2004).

Furthermore, endostatin binds tropomyosin 3, a member of a large family of

actin-interacting proteins (MacDonald et al. 2001).

2.7 Gene therapy as a method of Growth Factor Delivery

VEGF has been used to promote angiogenesis of ischemic muscle tissues and

accelerate fracture healing, whereas endostatin has been found to inhibit the

growth of primary tumors and their metastases in a number of animal models and

human subjects in clinical trials (Spector et al. 2000, Folkman 2006, Keramaris et al. 2008, Uchida & Haas 2009).

The application of exogenous proteins may be limited by their short

biological half-lives and because of difficulties associated with obtaining

sufficient quantities for optimal therapeutic benefit. A single dose of recombinant

protein may not be sufficient to increase or decrease angiogenesis or osteogenesis

(Spector et al. 2000). A better strategy for protein delivery may be gene therapy,

which involves the transfer of genetic information to cells, and can be completed

in vivo or ex vivo. For in- vivo gene therapy, a vector carrying the therapeutic

DNA is directly implanted or injected into the target tissue so that host cells are

transfected and express the protein. For ex vivo therapy, gene transfer can be

delivered through transfection of cultured cells, which are then implanted or

injected into the target tissue. The duration of protein synthesis after gene therapy

depends on the techniques used to deliver the gene to the cell. (Kofron &

Laurencin 2006)

46

Vectors are agents that enhance the entry and expresion of DNA in a target

cell. Ideal vector have a high transfection efficiency, low toxicity, and consistent

gene expression. They may be of viral and nonviral origin. Currently, the most

common types of vectors are viruses that have been genetically altered to carry

non-viral DNA. Viruses have evolved specialized molecular mechanisms to

efficiently transport their genomes inside the cells they infect. Some of the

different types of viruses used as gene therapy vectors are for example,

adenoviruses, retroviruses, adeno-associated viruses and lentiviruses.

Adenoviruses can be produced in high titer and can transfect dividing and non-

dividing cells. Adenoviral DNA does not integrate into the genome and is not

replicated during cell division, with the result that transfected cells are gradually

diluted out of the population. The primary disadvantage of adenoviral vectors is a

potential host immune response, which targets the virus and cells transfected by

the virus. (Lieberman et al. 2002, Franceschi 2005, Kofron & Laurencin 2006)

47

3 Aims of the study

Angiogenesis is a fundamental part of osteogenesis as well as bone remodeling.

Since there is an intimate relationship between the vasculature and bone cells

within the bone remodeling site, the role of the basement membranes of the

vascular endothelium in the recruitment of bone cells was studied. One

component of basement membrane is collagen XVIII, a proteoglycan containing

heparan sulfate, and endostatin is a fragment of collagen XVIII, therefore, the

hypothesis was that endostatin could have a role in the activation of bone cells.

The affects of the major angiogenic factor, VEGF, on bone cells was studied.

The aims of the present study were:

1. to investigate the feasibility of a first-generation adenoviral vector to deliver

VEGF overexpression locally at the site of bone injury in order to alter the

course of bone regeneration in the rat femur.

2. to study the possible role of endostatin and VEGF-A on osteoclasts in vitro

by using a classical resorption pit assay, and to investigate osteoclast

mediated bone resorption and osteoclastogenesis in a differentiation assay

using bone marrow hematopoietic stem cells.

3. to follow the bone formation cascade in vivo in an endochondral ossification

model where ectopic bone formation is stimulated by BMPs in the mouse

hind muscle pouch, in the presence of overexpressed endostatin with or

without VEGF-A using a recombinant adenoviral vector.

4. to further investigate the influences of endostatin on osteoblastic behavior in vitro. The role of endostatin in bone was studied in vivo using a transgenic

mouse line overexpressing endostatin (ES-tg) and knockout mice lacking

collagen XVIII (Col18a1-/-).

48

49

4 Materials and methods

4.1 Reagents (I-IV)

4.1.1 Recombinant proteins (II, IV)

VEGF-A

recombinant human VEGF165 produced in Spodoptera frugiperda

(Calbiochem) was diluted in sterile PBS containing ≥ 0.1% BSA.

Endostatin

recombinant human endostatin polypeptide produced in Pichia pastoris

(Calbiochem) was diluted in sterile PBS.

4.1.2 Recombinant adenoviruses (I, III)

Recombinant E1-E3-deleted adenoviral vectors encoding VEGF164 (AdVEGF-

A) (Breier et al.1992) and nuclear targeted β-galactosidase (AdlacZ) under the

CMV promoter were constructed and produced in 293 human embryonic

kidneycells as previously described (Laitinen et al. 1998). Adenoviruses were

analyzed to be free from helper viruses, lipopolysaccharide and bacteriological

contaminants (Laitinen et al. 1998, Puumalainen et al. 1998).

To create a recombinant adenovirus expressing rat endostatin, the expression

cassette was cloned into the pQBI-AdCMV5-IRES-GFP vector (Quantum

Biotechnologies, Montreal, Quebec, Canada). The vector and the plasmid pJM17

(Microbix Biosystems, Toronto, Ontario, Canada) were then cotransfected into

293 cells, and the recombinant adenovirus was isolated, amplified, and purified

according to standard procedures (Pulkkanen et al. 2002).

4.2 Experimental animals in vivo (I, III, IV)

4.2.1 Bone defect healing model (I)

Male Sprague-Dawley rats (350–410 g) were anesthetized by intraperitoneal

injection of fentanyl citrate 0.315 mg/ml, midazolam 5 mg/ml, and fluanisone 10

mg/ml (1 : 1 : 1). A cylindrical 3-mm osteoperiosteal defect was drilled in distal

50

femurs approximately 5 mm from the epiphyseal plate. The surrounding muscle

was closed around the lesion and 2 × 108 plaque-forming units (pfu) of

adenoviral vector encoding vascular endothelial growth factor (AdVEGF-A) were

injected into the muscle in a 10-µl volume. The contralateral femurs served as

controls and received equal amounts of β-galactosidase adenoviral vector

(AdlacZ). The animals were sacrificed 1, 2, or 4 weeks (at least seven animals at

each time point) following the procedure and the femurs were collected and fixed

in 4% paraformaldehyde in 0.15 M sodium phosphate (pH 7.2) for further

analysis.

4.2.2 Endochondral ossification animal model (III)

Balb/c mice, age 10–12 weeks, were anesthetized by intraperitoneal injection of

ketamine midazolam (0.08 ml/10 g body weight). Ectopic bone formation was

induced by implanting native reindeer (Rangifer tarandus) BMP extract combined

with gelatine gel (5 mg of each) in 0.9% saline and recombinant adenoviral

vectors designed to overexpress endostatin (AdEndostatin), VEGF-A (AdVEGF-

A) ,or nuclear targeted β-galactosidase (AdlacZ). The mixture (100 µl) was

injected (1 ml syringe, 20 G needle) into both thigh muscle pouches in the

bilateral hind legs. The mice used in this study (total of 76) were divided into four

groups. The control group received BMP and gelatin gel only (n = 16) while the

second group received BMP combined with the gelatin gel and 1.2 × 109 pfu of

AdVEGF-A (n = 19). The third and the fourth groups were injected with BMP

combined with gelatin gel and 1.5 × 109 pfu Ad-endostatin only (n = 20) or in

conjugation with 1.2 × 109 pfu AdVEGF-A (n = 20), respectively. In addition,

four animals served as controls and received equal amounts, 1.6 × 109 pfu of

AdlacZ in BMP combined with gelatine gel. The animals were sacrificed in a

chamber with carbon dioxide 1, 2 or 3 weeks after the operation and the hind legs

were collected for further analysis. The animal tests were performed after

approval by the National Laboratory Animal Center. All aspects of animal care

complied with the Animal Welfare Act and the recommendations of the NIH-PHS

Guide for the Care and Use of Laboratory animals.

4.2.3 Transgenic mice and knockout mice (IV)

J4 mice overexpressing monomeric endostatin in the keratinocytes under the

keratin-14 promoter (ES-tg) with age- and sex-matched FVB/N wild type

51

controls, and knock-out mice deficient in collagen XVIII (Col18a1-/-) with age-

and sex-matched C57BL/6J wild type controls were used to study the effects of

collagen XVIII/endostatin on bone development in vivo. The generation of

collagen XVIII knock-out and J4 mice has been described in previous reports

(Fukai et al. 2002, Elamaa et al. 2005). Groups of 6–10 mice containing equal

numbers of males and females were used for the experiments, and the mice were

sacrificed at the age of 14 or 30 days. All the experimentats were approved by the

Animal Care and Use Committee of the University of Oulu, and in the case of the

Col18a1-/- mice also by the State Provincial Office in Oulu.

4.3 Cell Cultures

4.3.1 Osteoclast isolation and culture (II)

The procedure for the isolation and culture of osteoclasts described earlier by

Boyde et al. and by Chambers et al. was modified slightly and has been described

in detail previously (Boyde et al. 1984, Chambers et al. 1984). Briefly,

mechanically harvested osteoclasts from the long bones of 1- or 2-day-old

Sprague-Dawley rat pups were allowed to attach to ultrasonicated bovine cortical

bone slices (0.125 cm2 or 0.5 cm2). After 30 minutes, the nonattached cells were

rinsed away, and the remaining cells on the bone slices were cultured in

Dulbecco’s modified Eagle’s medium (α-MEM) buffered with 20 mM HEPES

and containing 0.84 g of sodium bicarbonate/liter, 2 mM L-glutamine, 100 IU of

penicillin/ml, 100 μg of streptomycin/ml and 7–10% heat-inactivated fetal calf

serum (FCS), at +37°C (5% CO2 and 95% air). The cells were divided into ten

groups: the control group had α-MEM-FCS as their culture medium with no

added substances (later referred to as control), the VEGF-A-treated cells had 100

ng/ml, 50 ng/ml or 10 ng/ml VEGF-A in α-MEM-FCS (later referred to as

VEGF). The endostatin groups had 0.02, 0.2 or 2 µg/ml endostatin in α-MEM-

FCS (later referred to as ENDO) and the last groups had both VEGF-A and

endostatin added into the complete culture media (later referred to as

VEGF+ENDO). The last groups had 100 ng/ml VEGF-A and 2 µg/ml endostatin,

50 ng/ml VEGF-A and 0.2 µg/ml endostatin or 10 ng/ml VEGF-A and 0.02µg/ml

endostatin. The cells were allowed to grow for 48 h, after which the bone slices

were fixed with freshly made refrigerated 3% paraformaldehyde (PFA) and 2%

sucrose in phosphate-buffered saline (PBS) for 10 minutes at room temperature.

52

The data shown were gathered from three independent experiments, each of

which gave identical results.

4.3.2 Osteoclast differentiation assay (II)

The procedure for osteoclast differentiation described earlier by Takahashi et al. was slightly modified (Takahashi et al. 1988). Mouse (C57BL/6, 8–12 weeks)

bone marrow cells were isolated from mouse long bones using a syringe and

transferred to a petri dish. After incubation at +37 °C for 2 hours, non-attached

cells were collected from petri dish and seeded in 24-well plates containing

sonicated cortical bovine bone slices (0.125–0.5 cm2) at a concentration of 1 ×

106 cells/well. Cells on the bone slices were cultured in four groups: control,

VEGF (100ng/ml), ENDO (2µg/ml) and VEGF + ENDO respectively. The

control group was cultured in α-MEM medium (Sigma, UK) with 10% heat-

inactivated fetal calf serum (FCS), 2 mM L-glutamine, 100 IU/ml penicillin, 10

μg/ml streptomycin and 20 mM Hepes, a medium later referred to as α-MEM-

FCS. The cells were cultured in 500 μl α- MEM containing 30 ng/ml RANKL

(Peprotech Ec., UK) and 10 ng/ml M-CSF (R&D Systems). Half of the medium

was replaced every 3rd day. The cells were cultured at +37 °C (5% CO2, 95% air)

for 7 days, after which the cultures were stopped by fixing the cells with 3%

paraformaldehyde (PFA) / 2% sucrose in PBS.

4.3.3 Osteoblast MC3T3 culture (IV)

MC3T3-E1 cells are pre-osteoblasts derived from murine calvarias (American

Type Culture Collection, ATCC, USA). They were maintained in α-MEM

supplemented with 10% FCS, 2nM L-glutamine, 100 IU 7ml penicillin and 10

µg/ml streptomycin (α-MEM-FCS), and passaged every 2–3 days using standard

techniques. After pre-culturing the cells were seeded at a density of 10.000 cells /

cm2 in the presence of α-MEM with 50 µg/ml ascorbic acid and 10 mM β-

glycerophosphate (mineralization medium). The cells were cultured in six groups

and the medium was supplemented as follows: 1) control group (α-MEM-FCS as

a culture medium with no added substances), 2) VEGF-A 100 ng/ml (recombinant

human VEGF165 produced in Spodoptera frugiperda, Calbiochem), 3) endostatin

2 μg/ml (recombinant human endostatin polypeptide produced in Pichia pastoris,

Calbiochem), 4) endostatin 20 μg/ml, 5) VEGF-A + endostatin 2 μg/ml and

6) VEGF-A + endostatin 20 μg/ml. The cells were cultured at +37 °C (5% CO2,

53

95% air) and the medium freshly supplemented with VEGF-A, endostatin or

VEGF-A and endostatin together was replaced after 1, 3, 7 and 10 days of culture.

Each experiment was repeated at least twice.

4.4 Histology, histomorphometry and immunohistochemistry (I, III, IV)

For histological analysis to evaluate bone defect healing, the femurs were fixed in

phosphate-buffered 4% paraformaldehyde and decalcifiedin 10%

ethylenediaminetetraacetic acid (EDTA) (I). For evaluation of evaluating of

ectopic new bone, the collected specimens were fixed in 10% neutral formalin

solution and decalcified in 5% formic acid. After decalcification, the bones were

embedded in paraffin (III). When studying transgenic mouse lines (IV), the bone

specimens were fixed in 70% ethanol, decalcified and embedded in paraffin.

4.4.1 Haematoxylin-eosin (HE) staining (I, III, IV)

The decalcified bones were cut into 5-µm tissue sections and stained with

hematoxylin and eosin. Morphological analysis was performed with a digital

image analysis system MCIDM4 (Imaging Research Inc., St Catharines, Canada)

using a Nikon Optiphot II microscope and a Sony DXC-930P color camera.

Study (I): The histology of the tissue was analyzed in order to demonstrate

differences in healing between the VEGF-treated and control femurs. The

proportion of fibrous reparative tissue was measured histomorphometrically from

the cross sections of the defect area 1, 2 or 4 weeks after the procedure. The

amount of cartilage at the periosteal rim of the defect was also measured 1 and 2

weeks after the procedure.

Study (III): Evaluation of ectopic new bone was based on the proportional area of

remaining cartilage in the ossicle 2 and 3 weeks after the injection determined

with morphometrical analysis. The ossicle was defined as the tissue surrounded

by fibrous capsule and containing a cortical outer bone shell and developing bone

marrow around the remaining endochondral cartilage.

Study (IV): Femur samples were studied for the analysis of secondary ossification

centers in 2 and 4 week old Col18a1-/-, C57BL/6J, ES-tg and FVB/N mice.

54

4.4.2 Immunostaining of capillaries by FVIII (I, III)

The VEGF (I, III) and endostatin (III) effect in bone was demonstrated by

immunostaining of capillaries using a polyclonal rabbit anti-human antibody to

FVIII-related antigen (Dako, Dakopatts, Denmark). The primary antibody was

applied on the slides for 1 h (dilution of 1 : 50) followed by a secondary anti-

rabbit antibody (Dako), after which the ABC-complex was applied. The color was

developed with diaminobenzidine. In study (I) the vascular organization was

evaluated and density was estimated as the number of positively stained blood

vessels per high power field (HPF). In each section, a minimum of five HPFs was

analyzed. In study (III) morphometrical analyses of the factor VIII positive blood

vessels were done from three microscopic fields (digital image analysis system

MCID M4 using a Nikon Optiphot II microscope and a Sony DXC- 930P color

camera) of the highest vessel density persection in a continuous fashion. The

numbers of the positively stained blood vessels were counted as proportional area

per microscopic field.

4.4.3 Collagen XVIII antibodies (IV)

Paraffin-embedded 5 μm sections were dewaxed, washed in PBS and stained with

collagen XVIII antibodies using a tyramide signal amplification (TSA) kit

(PerkinElmer). The following antibodies were used in immunohistochemistry: a

polyclonal rabbit-anti mouse antibody recognizing the N-terminal part of type

XVIII collagen (anti-all moXVIII) and a polyclonal rabbit-anti mouse antibody

recognizing the N-terminal part of the two longer variants of type XVIII collagen

(anti-long moXVIII) (Saarela et al., unpublished data) Heat-induced epitope

retrieval was carried out in citrate buffer (10mM sodium citrate, 0.05% Tween,

pH 6.0) for 10 min in a microwave oven. The slides were then incubated

overnight at + 4 ºC with either anti-all moXVIII (4 μg/ml) or anti-long moXVIII

(10 μg/ml) antibody, diluted in TNB buffer (0.1 M Tris-HCl (pH 7.5), 0.15 M

NaCl, 0.5% Blocking Buffer). This was followed by a 30-minute incubation with

a biotinylated secondary anti-rabbit antibody (Vector Laboratories) at room

temperature. Counterstaining was performed with haematoxylin.

55

4.4.4 TRACP staining for identifying osteoclasts (II, III)

The cells were stained for tartrate-resistant acid phosphatase (TRACP), a

commonly accepted marker of osteoclasts (Minkin, calcify tissue int 1982) using

a Sigma TRACP kit (no. 386-A, Sigma Chemicals, St.Louis, MO) according to

the manufacturer’s instructions. To visualize the nuclei, the cells were incubated

with the DNA-binding fluorochrome Hoechst 33258 (1 mg/ml stock diluted 1:800

in PBS, Sigma Chemical Co, St. Louis, MO) for 10 minutes at room temperature.

Study (II): The numbers of multinucleated TRACP-positive cells on each

bone slice were counted, using a Nikon Eclipse 600 microscope and Nikon Plan

Fluor 20x/0.50 objective with an appropriate filterset.

Study (III): The numbers of the multinucleated TRACP-positive cells in

contact with cartilage or newly formed bone trabeculae were counted from five

microscope fields using a Nikon Eclipse 600 microscope with a 40x/0.50

objective with an appropriate filterset.

4.4.5 Area Measurement of Excavated Pits (II)

Prior to the staining of the pits, total detachment of the cells from the bone slices

was ensured by wiping the cellular surface of the slices with a soft brush. The pits

were stained with peroxidase-conjugated wheat germ agglutinin-lectin (WGA; 20

μg/ml, Sigma Chemical Co., St.Louis, MO) for 20 minutes, washed with PBS,

and incubated for 5 minutes in diaminobenzidine (0.5 mg/ml) +0.03% H2O2.

Resorption pits were visualized as brown-colored areas after staining with WGA.

Morphometric analysis of the resorption pits was performed with an MCID image

analyzer, utilizing M2 software (Imaging Research Inc., Brock University,

Ontario, Canada) to the quantify areas of the resorption lacunae.

4.5 Biological assays

4.5.1 Determination of cell proliferation (WST-1) (IV)

MC3T3-E1 cell proliferation was measured by a WST-1 assay (Roche, Germany),

which quantifies the reduction of the tetrazolium salt WST-1 by mitochondrial

dehydrogenases in viable cells. Analyses were performed 3, 7, 10 and 14 days

after the cells were plated. WST-1 reagent (20µl) and α-MEM with ascorbic and

β-glycerophosphate (200 μl) were added to the cells and the culture incubated at

56

+37°C (5% CO2, 95% air) for 1 hour. The absorbance of the reaction product was

measured with a 1420 Victor2 Multilabel counter (Wallac, Turku, Finland) at a

wavelength of 450 nm.

4.5.2 Alizarin red S staining and quantification of calcium deposition (IV)

The mineralization of matrix and nodules was determined using Alizarin red S

staining. Briefly, MC3T3-E1 cell cultures were fixed at culture day 14 with 3%

PFA/ 2% sucrose in PBS for 10 minutes at room temperature, washed with PBS

and stained for 30 min with a 1% solution of Alizarin red S (Amresco, USA) pH

4.2 at room temperature with rotation. The cultures were washed three times with

deionized water and once with 70% ethanol before the addition of 100 nM

cetylpyridinium chloride (Sigma) for 1 hour to solubilize and release calcium-

bound Alizarin red into the solution. The solution was collected and the Alizarin

red S concentration was determined by absorbance measurements at 570 nm

using an Alizarin red S standard curve in the same solution.

4.6 Radiographical analysis

4.6.1 Radiographic evaluation of bone formation (III)

Standard lateral position radiographs (100 mA, 20 kV, 0.08 s/exp; Mamex de

Maq, Soredex, Orion, Finland) were taken from all the hind legs of the mice.

Calibration of the equipment for the measurement of optical density was

performed using an aluminum wedge. New bone formation was evaluated as the

area (in mm2) of calcified tissue visible in the radiographs with a digital image

analysis system MCID M4 (Imaging Research Inc., St Catharines, Canada). The

radiographs were digitized on a light table with a ccd camera (Dage MTI 72E,

Dage- MTI, Inc., Michigan City, IN, USA) using a Micro Nikkor 55mm objective

(Nikon, Tokyo, Japan).

57

4.6.2 Peripheral quantitative computed tomography (pQCT) (I, IV)

Bones were scanned with a Stratec XCT 960 A peripheral quantitative computed

tomography (pQCT) system with software version 5.21 (Norland Stratec

Medizintechnic GmbH, Birkenfeld, Germany).

Study (I): The bone mineral content of the femurs treated with AdVEGFAor

AdlacZ was analyzed at 1, 2, and 4 weeks after the operation. Distal femurs were

scanned for three to five consecutive cross-sectional images with slice distances

of 0.5 mm. Bones were scanned with a voxel size of 0.148 × 0.148 × 1.25 mm3. A

slice representing the central area of the defect was chosen for analysis, and the

total bone mineral content (BMC, mg/mm) was recorded.

Study (IV): To evaluate to effects of endostatin in vivo, a voxel size of 0.148

× 0.148 × 1.25 mm3 (transgenic mice overexpressing endostatin and FVB/N

control mice) and 0.092 × 0.092 × 1.25 mm3 (mice deficient in collagen XVIII

(Col18a1-/-) and C57BL/6J control mice) were used for the measurements. The

scout view of the pQCT system determined the middle point of each bone, from

which one cross-sectional slice was scanned and the total mineral content and

density measured.

4.6.3 Faxitron X-ray imaging (IV)

Radiographs of the dissected bones were taken in anteroposterior (AP) and lateral

projection on a Faxitron Specimen Radiography System (Model MX-20,

Wheeling, IL, USA) with a high-resolution algorithm. The imaging parameters

were identical for all the specimens and the following settings were used: 6 s, 30

kV for the 30 day old mice and 20 kV for the 14 day old mice, pixel matrix

1024 × 1054; resolution 200 pixels mm-1. The images were transferred to a

personal computer for analysis. The lengths and areas of tibias and femurs were

measured from the radiographs. As the weight of the mice varied from litter to

litter in both mutant and control mice, we normalized the bone density data by the

length and area of the mouse bones measured from the Faxitron images.

4.7 Statistical analysis (I-IV)

The values are the mean ± SEM. Significance was calculated using one-way

ANOVA followed by Student’s t-test with the statistical package Origin 6.0 or

7.5. (Microcal Software Inc., Northampton, MA)

58

59

5 Results

5.1 Bone defect healing model (I)

5.1.1 Bone mineral content evaluated with peripheral quantitative computed tomography (pQCT) (I)

The bone mineral content (BMC) at the injury site was analyzed by peripheral

quantitative computed tomography (pQCT) revealing a significant increase in

BMC in the VEGF-treated femurs (13.9 mg/mm vs.12.1 mg/mm, difference

14.9%, p < 0.001) 2 weeks after the operation. At the 1-week and 4 week time

points, there was no statistically significant difference between the two sides.

5.1.2 Induction of vascularization (I)

The number of capillaries was significantly increased at the 1-week time point

only (5.5 in control group vs. 7.5 in VEGF group, 36.4% difference, p < 0.01),

after which the difference was no longer significant.

5.1.3 Histological evaluation (I)

The histology of the tissue sections was analyzed in order to demonstrate

differences in healing between the VEGF-treated and the control femurs. One

week after the procedure (Figures 6A and 6B), both sides still had large

hematomas with surrounding seams of mesenchymal stem cells. The

morphological difference in the amount of non-ossified callus is most obviously

seen at the 2-week time point (Figures 6C and 6D), indicating amplified

ossification in the VEGF-treated femurs. The proportion of fibrous tissue

representing residual defect was significantly smaller in the VEGF group 2 weeks

after the operation (19.5% in VEGF group vs. 31.0% in control group, 59.0%

difference, p < 0.05). At the 4-week time point (Figures 6E and 6F), the defect

was difficult to localize due to almost complete healing and little difference was

seen between the VEGF overexpression and the control sides.

60

Fig. 6. Example of histology of bone regeneration in the drilling defects in the VEGF-

treated (right panel) and the control femurs (left panel) 1 (A, B), 2 (C, D), and 4 weeks

(E, F) after the operation. HE-stained decalcified sections. Scale bar: 1mm.

Endochondral ossification took place at the periosteal rim of the drilling defect

whereas direct ossification occurred at the central part of the healing metaphyseal

defect. The amount of hyaline cartilage was high at the early phases of healing,

61

but the VEGF treatment stimulated rapid replacement of the cartilage with mature

bone. At 2 weeks, the amount of cartilage was significantly lower in the VEGF

group compared with the control (0.015 mm2 vs. 0.065 mm2, 3.3-fold difference,

p < 0.01). At 4 weeks, no cartilage was detected in the defect area. The data

support the hypothesis that VEGF overexpression significantly enhanced the

endochondral ossification (Figure 7).

Fig. 7. Examples of cartilage at the periosteal defect side. The control defect 1 week

after the operation (A) had a lower amount of cartilage compared with the VEGF-

treated femur (B). At 2 weeks, the amount of cartilage was significantly higher in the

control (C) compared with the VEGF group (D). Hematoxylin-eosin staining of paraffin-

embedded tissue. Scale bar 50 µm.

62

5.2 Effects of VEGF-A and endostatin on osteoclastic bone resorption and differentiation in vitro (II)

5.2.1 Bone resorption assay (II)

The number of osteoclasts in each sample group was counted by analyzing

multinuclear TRACP positive cells in the resorption assay. No significant

differences in the number osteoclasts were found between the different groups.

The total resorbed area was significantly increased in response to VEGF-A

treatment and the effect was dose dependent. Endostatin alone had no inhibitory

effect on the resorbed area when compared to the basal level of osteoclastic bone

resorption. When both endostatin and VEGF-A were supplemented, the

stimulatory effect by VEGF-A on bone resorption was limited to the basal level.

The resorption activity was even below basal level with the highest

concentrations (p < 0.05 in ctrl vs. VEGF100 and endostatin). Thus, endostatin

could block the VEGF-A-induced stimulatory effect on osteoclastic bone

resorption.

5.2.2 Osteoclast differentiation (II)

VEGF-A was added to the medium containing M-CSF and RANKL in order to

achieve the maximal differentiation stimulus. Endostatin inhibited the maximal

osteoclastogenesis alone or in combination with VEGF.

5.3 Endochondral ossification model (III)

5.3.1 Radiographic evaluation of bone formation (III)

The effect of the recombinant adenoviral endostatin gene transfer with or without

VEGF-A in bone formation was first evaluated from radiographs. X-ray images

of the dissected legs displayed significant bone formation in all treatment groups

at the 2- and 3-week time points. Two weeks after the injection, there were no

significant differences between the study groups. At the 3-week time point,

however, we could see retardation of bone formation in radiographs of the

endostatin-treated animals compared to the VEGFA- treated and the control

63

groups. In addition, the retardation was clearly visible in the animals that received

a combination of VEGF-A and endostatin.

5.3.2 Induction of vascularization (III)

The samples were processed for immunohistochemistry in order to demonstrate

vascular structures positive for the FVIII-related antigen. At the 1-week time

point, previously existing vessels (Figures 8A and 8G) were easily identified due

to the well-defined endothelium. At all time points, endothelial staining at the

bone formation site was increased in the VEGF-A group compared to the control.

After 2 and 3 weeks, endostatin blunted the VEGF-A effect. Endostatin alone,

however, did not decrease the basal neovascularization (Figure 8). At the 2- and

3-week time points, staining for the factor VIII-related antigen indicated intense

and partially premature vascularization in the VEGF-A-stimulated ossification.

The new vessels were morphologically more mature in the endostatin group

compared to the VEGF-A stimulated angiogenesis.

64

Fig. 8. Bones were decalcified, cut into tissue sections, and immunostained for FVIII-

related antigen. The figure shows examples of BMP control (A–C), VEGF-A (D–F),

endostatin (G–I) and VEGF-A+endostatin (J–L) treated ossicles 1 week (A, D, G, J), 2

weeks (B, E, H, K) and 3 weeks (C, F, I, L) after the injection. Pre-existing vessels were

nicely stained all over the induction site. Examples of these old vessels are seen in (A)

and (C). Scale bar: 100 μm.

5.3.3 Bone histology (III)

The histology of the tissue sections was analyzed in order to demonstrate

differences in endochondral bone development. At the 2-week time point, there

was no difference in the proportional ossification of the cartilage phase.

Significant differences were seen in the area of cartilage remnants of the bone

nodules in different test groups 3 weeks after the injection. We could demonstrate

that endostatin inhibited cartilage resorption compared to control and VEGF-A-

65

treated animals. The proportional area of bone was lower in the endostatin group

compared to the control (Figure 9).

Fig. 9. (A) Example of new bone formation induced by BMP injection. Ectopic new

bone was evaluated as a proportional area of bone tissue/osseous nodule 2 and 3

weeks after the injection. (Arrow indicates the cortical shell of the ossicle and ∗

indicates cartilage). (B) At the 3-week time point endostatin inhibited the proportional

area of bone tissue. At 1 or 2 weeks, no difference in proportional area of bone

tissue/ossicle was seen. ∗∗p < 0.01. Scale bar: 200 μm.

66

5.3.4 The number of cartilage-resorbing chondroclasts (III)

Cartilage resorption was studied by identifying the number of TRACP-positive

chondroclasts/osteoclasts. VEGF-A alone stimulated osteoclasts as demonstrated

by the increased number of TRACP-positive cells 3 weeks after the operation.

Endostatin gene transfer decreased the amount of TRACP-positive cells

compared to the VEGF-A group at the 2- and 3-week time points. Endostatin also

decreased the VEGF-A stimulated osteoclastogenesis at the 2-week time point.

5.4 The effects of endostatin on the MC3T3-E1 osteoblastic cell line

5.4.1 Osteoblast proliferation in vitro

MC3T3-E1 cell proliferation was measured by a WST-1 assay. On days 3–10

there were no statistical differences between the groups. On day 14 of culture

endostatin 20 μg/ml decreased osteoblast proliferation compared to the basal

control level (p < 0.05) and VEGF-A group (p < 0.05). Endostatin 2 μg/ml also

decreased osteoblast proliferation, but not significantly. Endostatin 2 μg/ml

together with VEGF-A decreased osteoblast proliferation compared to the basal

level (p < 0.001), VEGF-A (p < 0.001) and endostatin 2 μg/ml (p < 0.05).

Interestingly, VEGF-A together with the higher dose of endostatin, 20 μg/ml did

not significantly affect proliferation.

5.4.2 Matrix mineralization in vitro

The immature osteoblast cell line, MC3T3-E1, up-regulates osteoblast-specific

genes and produces mineralized nodules undergoing the developmental stages

associated with differentiating osteoblasts. The process of mineralization is

initiated around day 10. A low dose of endostatin 2 μg/ml together with VEGF-A

decreased the mineralisation compared either to the basal control level, VEGF-A

or endostatin alone (p < 0.05). Endostatin 20 μg/ml together with VEGF-A does

not significantly affect mineralization when compared to the basal control level,

but if compared to treatment with VEGF-A alone, a decrease in matrix

mineralization can be observed (p < 0.05).

67

5.5 Transgenic mice and knockout mice

5.5.1 pQCT and faxitron results

The bone mineral content or density, as analyzed with pQCT, was not altered in

ES-tg or Col18a1-/- mice as compared to their FVB/N or C57BL/6J controls.

Moreover, there were no significant differences in bone sizes between the

Col18a1-/- mice and their controls. The lengths of tibias and femurs were

measured from radiographs. In the ES-tg mice the lengths of femurs and tibias

were smaller at the age of 14 days, compared to FVB/N controls (p < 0.01). This

delay of growth was transient, as no difference was seen at 30 days of age.

5.5.2 The formation of secondary ossification centers

In the bones of 14-day-old Col18a1-/- mice we detected a delay in the formation

of secondary ossification centers. In all C57BL/6J mice the secondary ossification

centers were already clearly visible in the epiphyseal areas of femurs. However,

in Col18a1-/- mice the secondary ossification centers were either missing, or very

small (Figure 10). At the age of one month, no clear differences could be

observed between the knock-out and control mice: in both, the ossification of the

epiphyses had occurred. The growth plate was still visible in both mouse lines at

this age. In ES-tg mice we could see no difference in the ossification of the

epiphyseal area as compared to their FVB/N controls.

68

Fig. 10. In 2-week-old C57BL/6J control mouse femurs (A) the formation of secondary

ossification centers is clearly visible in the epiphyseal areas, whereas in the Col18a1-

/- mutants (B) the secondary ossification centers are either very small, or, as in the

case of some samples, totally absent at this timepoint. In 2-week-old FVB/N (C) and

ES-tg (D) mice the secondary ossification centers are well developed, and no obvious

differences can be detected between these two mouse lines.

5.5.3 Localization of collagen XVIII

Immunohistochemistry was used to study the localization of collagen XVIII in

bone tissue. Collagen XVIII was localised in the blood vessels of bone marrow:

staining was seen with the antibody against all three variants (anti-all moXVIII).

The short variant of collagen XVIII was responsible for this immunosignal, as the

antibody that stains the two longer variants of collagen XVIII (anti-long

moXVIII) did not stain the vessels (Figure 11). Staining with collagen XVIII

antibodies also gave a signal in the epiphyseal cartilage, surrounding the

chondrocytes, but this signal proved to be unspecific as similar staining could be

seen in cartilage of mice lacking collagen XVIII.

69

Fig. 11. Collagen XVIII was localized in the central vessels and some smaller vessels

of the bone marrow (A and B) and vessels of the periosteum (C) and at the skeletal

muscle attachment site (D). As expected, the collagen XVIII KO used as negative

controls showed no staining (E and F). The samples are from 14-day-old mice.

70

71

6 Discussion

6.1 Adenoviral VEGF-A gene transfer induces angiogenesis and

promotes bone formation in healing osseous tissues (I)

Angiogenesis has a crucial role in bone healing. VEGF is expressed within the

first weeks of bone healing (Uchia et al. 2003) and it may enhance osteoblast

differentiation (Deckers et al. 2000). Pufe and colleagues have shown that VEGF

is expressed in fracture callus at the beginning of fracture healing after which

expression decreases by day 5 (Pufe et al. 2002). We studied the effect of VEGF-

A on bone healing and the feasibility of first –generation adenoviral vector to

deliver VEGF overexpression at the bone injury site.

The exogenous VEGF protein is an unstable, short-lived protein in vivo and

moreover, costly to produce (Keramaris et al. 2008). The use of gene therapy for

bone regeneration enables delivery of VEGF at physiological levels using natural

cellular mechanisms. In order to achieve fracture healing, the required therapeutic

level of VEGF must be maintained for a period of time. A first-generation

adenoviral vector typically produces expression that lasts for 1–2 weeks in vivo,

after which the virus may be eliminated by the immune response (Cao et al. 2002,

Ylä-Herttuala et al. 2000, Okubo et al. 2001). The rodent model is useful for

studying bone healing properties, since rodent bone healing is a very rapid

process. Bone regeneration occurs within 4 weeks even in the absence of

interventions (Uusitalo et al. 2001).

In this study, the defects were healed finally in both groups, but the healing

process was accelerated in the VEGF treated group. In order to monitor bone

recovery, we first measured the bone mineral content (BMC) using peripheral

quantitative computed tomography (pQCT). A number of studies have used

pQCT in mice and rats to observe cortical bone geometry along with cortical and

trabecular bone mineral density (BMD). This imaging method has been used to

monitor such changes after pharmacologic or mechanical interventions and to

monitor fracture healing (Breen et al. 1998, Armamento-Villareal et al. 2005,

Gasser et al. 2008, McCann et al. 2008). Locally VEGF-A overexpression

increased the bone mineral content measured with (pQCT) 2 weeks after the

operation. At the 1-week time point, there was as yet no statistically significant

difference between the two sides and at the 4-week time point, the mineral

72

content of both VEGF-treated and control bones was decreased due to the

reorganization of the callus at the injury location. Endochondral ossification took place at the periosteal rim of the drilling

defect whereas intramembranous ossification occurred at the central part of the

healing metaphyseal defect. The importance of angiogenesis and VEGF on

endochondral ossification has been pointed out in the study of Gerber and

colleagues (Gerber et al. 1999). Increased vasculature stimulates

intramembranous ossification as well (Collin-Osdoby 1994, Zelzer et al. 2002,

Horner et al. 1999).

Factor VIII expression is normally restricted to vascular endothelial cells and

therefore Factor VIII immunostaining is considered to be a reliable method to

detect blood vessels in dense connective tissue. In the present study, the number

of FVIII-related capillaries was significantly increased at the 1-week time point

providing increased blood supply to the healing bone and accelerated healing of

the defect. At 2 or 4 weeks, the difference could not be seen anymore. This could

be due to VEGF production at the early phases of bone healing and viral

expression may shut down soon after the virus delivery. Furthermore, skeletal

muscle is not an ideal target tissue for the recombinant adenoviral vector due to

the low number of viral receptors, the quiescence of myoblasts in mature muscle,

and physical barriers (Cao et al. 2002, Nalbantoglu et al. 2001). Our results thus

support previous results that increased vasculature also stimulates

intramembranous ossification (Collin-Osdoby 1994, Zelzer et al. 2002). The

capillaries were more disorganized in the VEGF-A stimulated condition

compared with the control. The induction of disorganized vascularization

indicates that the viral gene transfer resuted in increased VEGF-A activity. It

could be possible to adjust VEGF dosing with other growth factors to reduce this

phenomenon.

In this study, a first-generation adenoviral vector produced VEGF

concentrations with biological effects on bone metabolism and was able to

modify the histological features of bone healing. The morphological difference in

the amount of non-ossified callus was most obviously seen at the 2-week time

point, where the proportion of fibrous reparative tissue was more efficiently

cleared and replaced with bone in the VEGF-A group, indicating amplified

ossification in the VEGF-A treated femurs. We also measured the amount of

hyaline cartilage at the periosteal rim, and at the early phases of healing the

amount of cartilage was high. However, VEGF treatment stimulated rapid

replacement of the cartilage with mature bone. At the 4-weeks time point the

73

defect was difficult to localize due to almost complete healing, and only little

differences were seen between groups.

6.2 Endostatin inhibits VEGF-A induced osteoclastic bone resorption in vitro (II)

The intimate contact of the bone remodeling site with vascular sinusoids indicates

interplay between the vascular endothelium and the bone cells. A continuous

supply of bone-degrading osteoclasts is needed during bone development.

Osteoclasts are derived from hematopoietic precursors present in both the bone

marrow and the peripheral circulation (Roodman 1999). Osteoclast precursors

need to adhere to and migrate through endothelium to reach future sites of bone

resorption. It has been reported that VEGF is potentially a monocyte

chemoattractant (Clauss et al. 1996). Monocytes have been shown to express

VEGFR1, but not VEGFR2 (Barleon et al. 1996). Soluble Flt-1 blocks invasion

of osteoclasts into the marrow cavity of developing long bones in vivo (Niida et al. 1999). Mature osteoclasts express VEGFR1 and VEGFR2 on the cell surface

(Gerber et al. 1999, Niida et al. 1999, Nakagawa et al. 2000). Endostatin is a

known antagonist of VEGF-mediated endothelial cell migration and proliferation

(Yamaguchi et al. 1999, Taddei et al. 1999). Interestingly, the recruitment and

invasion of osteoclast precursors can be disturbed by endostatin in the metatarsal

model, in which fusion and migration of osteoclast precursors is required for

osteoclastic bone resorption (Henriksen et al. 2003).

Only little is known about the effects of endostatin on other cells and thus we

investigated the possibility that VEGF and endostatin may control

osteoclastogenesis and osteoclastic bone resorption in vitro. Endostatin has been

shown to bind αv- and α5-integrins (Rehn et al. 2001), and resorbing osteoclasts

express at least αv-integrins (Chuntharapai et al. 1993, Duong et al. 2000).

Therefore it is possibile that endostatin could control osteoclast function.

We used a classical resorption pit assay and we analysed the number of

osteoclasts by counting the multinuclear TRACP-positive cells. However

significant differences between the different groups were not detected. Next we

measured osteoclastic bone resorption. In response to VEGF-A treatment the total

resorbed area was significantly increased and the response was dose dependent.

Endostatin alone had no inhibitory effect on the resorbed area when compared to

the basal level of osteoclastic bone resorption, but when both endostatin and

VEGF-A were present, the stimulatory effect by VEGF-A was reduced to the

74

basal level. Our finding concerning the increased resorption achieved by VEGF is

consistent with the result of Nakagawa and colleagues (Nakagawa et al. 2000).

However, they demonstrated that VEGF-A enhanced the survival of pure rabbit

osteoclasts, although in our resorption pit assay containing osteoblasts and

osteoclasts, the number of TRACP-positive cells remained unchanged. The data

show that the increased total resorbed area in our model is not due to enhanced

survival rate. The observed effect is more likely a result of a direct VEGF-A

stimulation of osteoclastic bone resorption.

Since the basal resorption was not affected by endostatin in the pit assay, we

investigated wheter endostatin could inhibit the differentiation of osteoclasts.

Bone marrow hematopoietic cells differentiate into osteoclasts in the presence of

RANKL and M-CSF. When VEGF-A was added to the medium containing

M-CSF and RANKL, endostatin inhibited osteoclastogenesis compared to the

VEGF induced maximal stimulus. Interestingly, we could see the inhibitory effect

also in the absence of VEGF-A.

It seems, however, that the regulatory effect of endostatin is not critical since

endostatin alone cannot modify the basal bone resorption. In the study of

Henriksen and colleagues, endostatin completely inhibited VEGF mediated

osteoclast invasion but had no effect on RANKL or M-CSF induced migration. In

cultures of purified osteoclasts both RANKL and VEGF induced phosphorylation

of ERK1/2 MAP kinase (Henriksen et al. 2003). It is possible that endostatin may

not only limit the migration of osteoclast precursors, it may also inhibit VEGF-A-

induced osteoclast activation. Pufe and colleagues have detected a similar

influence of endostatin on hyaline articular cartilage. Endostatin alone had no

effect on the basal levels of the phosphorylation of ERK1/2, whereas co-

incubation with endostatin blocked VEGF-induced ERK1/2 phosphorylation in

immortalized human chondrocytes (C28/I2) (Pufe et al. 2004c, Pufe et al. 2004d).

6.3 Endostatin inhibits endochondral ossification (III)

Angiogenesis is essential for the replacement of cartilage by bone during skeletal

growth and regeneration as well as on ectopic ossification. The function of

endostatins during the angiogenesis process of bone formation is quite unknown.

In our previous studies we observed that VEGF-A stimulates osteoclasts and

osteoclastogenesis thus regulating the duration of the cartilage phase in this

process. In addition, endostatin inhibited osteoclastogenesis and osteoclast

75

function in vitro indicating that endostatin may coordinate angiogenesis and

ossification in vivo.

It is hypothesized that endostatin expression by mature cartilage may

suppress osteoclasts while endostatin-deficient ossifying cartilage allows

osteoclastic activity and endochondral ossification (Pufe et al. 2004c). Moreover,

it has been suggested that endostatin has a homeostatic function in cartilage

metabolism (Feng et al. 2005). Recently, endostatin has been found to repress the

development and volume of tissue grafts in mice with rheumatoid arthritis, which

is characterized by an inflammatory erosive synovitis, where invasion of

inflammatory cells into the joint synovium is accompanied by excessive

angiogenesis causing destruction of the cartilage and fibrous tissue (Matsuno et al. 2002, Yin et al. 2002).

Since the role of endostatin in the presence of existing stem cells in vivo

requires further exploration, we employed an endochondral ossification model in

the mouse hind muscle pouch stimulated by BMPs. Endostatin was overexpressed

with or without VEGF-A using a recombinant adenoviral vector. BMPs have been

used to heal bone defects in experimental animal tests, but also in clinical

settings, mainly in orthopedic and oral surgery (Pekkarinen et al. 2005).

The muscle pouch model of mouse has been a standard method for studying

the activity of BMP since its discovery and has been shown to be an appropriate

animal model (Sampath & Reddi 1981). In this study, we used injectable BMP

implants, which require only a minimally invasive technique. Previously, they

have been investigated in many animal experiments with promising results

(Forslund et al. 1998, Bax et al. 1999, Muschik et al. 2000, Blokhuis et al. 2001,

Li et al. 2002). The native reindeer BMP extract we used (Pekkarinen et al. 2003)

contains a cocktail of purified proteins consisting of multiple BMPs and other

non-BMP proteins combined with gelatin gel (Aldinger et al. 1991, Iwata et al. 2002, Jortikka et al. 1993a, Jortikka et al. 1993b). Although individual rhBMPs

are able to induce ectopic new bone formation, purified BMP extract from human

matrix has been demonstrated to produce new bone more efficiently than rhBMP-

2 (Bessho et al. 1999). This suggests that the effect of native BMP extract

represents synergistic activity between different BMP proteins. We observed strong ectopic bone formation in all our study groups that was

induced with BMP extract combined with gelatin gel in conjugation with the

viruses. First we evaluated the new bone formation from the X-ray images taken

from the dissected legs at 2- and 3-week time points. At the 1-week time point,

the mineralization of cartilage had not yet started. At 2 weeks, the difference

76

between the study groups was not significant, but we observed less bone

formation in endostatin-treated animals at the 3 weeks time point. Contrary to our

previous findingwith the bone drilling defect model, VEGF overexpression did

not enhanced bone formation. This may indicate that the surrounding tissue may

have regulatory properties that control ossification.

Peng and others have described a synergistic enhancement of VEGF and

BMP-4 or BMP-2 on bone formation (Peng et al. 2002, Peng et al. 2005).

However, VEGF alone did not promote bone regeneration when using a model

where the healing of cortical defects in the parietal bones of mice was studied.

They also reported that the beneficial effect of VEGF on bone healing elicited by

BMP-4 depends critically on the ratio of VEGF to BMP4, with an improper ratio

leading to unfavorable effects on bone healing, while disruption of the BMP-

2/VEGF ratio has less negative effects on bone formation (Peng et al. 2002). Furthermore, Kakudo and others reported an increased amount of VEGF

induced ectopic bone formation seen in X-rays. They implanted human

recombinant BMP-2 in rat calf muscle using a disc implant (Kakudo et al. 2006).

Thus it is possible that species differences and the mode of application of the

BMP and VEGF/BMP ratio influence bone formation. Moreover, Murphy and

colleagues suggested that the increased vascularization achieved with VEGF

enhances mineralized tissue generation, but not necessarily osteoid formation

when using VEGF with the bio-mineralized scaffold compared to mineralized

scaffold alone (Murphy et al. 2004).

In this study, we could not see increased endochondral bone formation caused

by VEGF-A overexpression in the mesenchyme of the muscular space indicating

that the amount of BMP determines the quantity of bone to be formed at the

ectopic site. VEGF-A however, stimulated the number of cartilage resorbing

osteoclasts and the number of FVIII-related antigen-positive blood vessels thus

accelerating the endochondral phase. Endostatin overexpression alone or in

combination with VEGF-A decreased the number of TRACP-positive cells

compared to animals with VEGF-A overexpression at 2- and 3-week time points

suggesting that cartilage resorption by chondroclasts is decreased in response to

endostatin even in the presence of VEGF-A. These results are in line with our

previous results showing that endostatin inhibits VEGF-A-stimulated osteoclastic

bone resorption in vitro independent of VEGF-A. In addition, as expected,

endostatin prevented the formation of vessels and thus inhibition of vascular

development. Endostatin could reduce the amount of BMP-induced bone

77

formation, since cartilage resorption and osteoblast function are dependent on

blood vessel formation.

6.4 Endostatin inhibits proliferation of MC3T3-E1 osteoblasts and together with VEGF-A suppress mineralization (IV)

We investigated how endostatin affects the behaviour of osteoblasts. A number of

studies have pointed out the significant role of canonical Wnt signaling in the

control of osteoblastogenesis and bone formation (Yavropoulou & Yovos 2007)

and recently it was reported that endostatin is a potent inhibitor of Wnt signaling

(Hanai et al. 2002). Thus, endostatin may have a role in the anabolic effect of

bone. Mechanical loading generates interstitial fluid flow that exerts a shear stress

at surfaces of osteoblasts and osteocytes. The canonical Wnt pathway is one of

the pathways that is up-regulated in response to the fluid shear stress, thus

translating fluid shear signals into biological effects in bone cells. Endostatin can

prevent the effect of fluid shear stress on osteoblasts completely (Lau et al. 2006). In the present study we used MC3T3-E1 cells, which behave as immature,

committed osteoblastic cells and go on to differentiate in response to intracellular

and extracellular cues. We measured MC3T3-E1 cell proliferation using a WST-1

assay and the mineralization of matrix and bone nodules was determined using

Alizarin red S staining. VEGF alone had no effect on MC3T3-E1 cell

proliferation, nor a significant effect on mineralization, which is consistent with

previous findings (Kitagawa et al. 2005). Proliferative effects of VEGF, at least in

endothelial cells, are mediated by Flk-1/KDR (Guo et al. 1995, Quinn et al. 1993), that is not expressed by MC3T3-E1 cells, which could explain why VEGF

did not have a proliferative effect on these cells. Kitagawa and others suggested

that since VEGF has earlier been shown to induce markers of osteoblast

differentiation, e.g. alkaline phosphatase and osteocalcin expression, but not to

increase mineralization, other factors are needed in addition to VEGF for the

induction of mineralization (Kitagawa et al. 2005). However, treatment of cells

with endostatin alone decreased the proliferation of osteoblasts. Endostatin 20

μg/ml decreased osteoblast proliferation on day 14 of culture, and endostatin 2

μg/ml also slightly decreased osteoblast proliferation but this difference was not

statistically significant. The addition of endostatin alone did not affect

mineralization either. Interestingly, again we could see the highest inhibitory effect of endostatin

when it was used together with VEGF. A low dose of endostatin (2 μg/ml)

78

together with VEGF decreased osteoblast proliferation compared to the basal

level, VEGF-A, and endostatin 2 μg/ml groups. Likewise, matrix mineralization

was inhibited when a low dose endostatin (2 μg/ml) added together with VEGF-A

was used compared to control, VEGF-A, or endostatin groups. With a higher

concentration of endostatin (20 μg/ml) combined together with VEGF-A, this

inhibitory effect could not seen. It is not known why the inhibitory effect of

endostatin is highest when it is combined with VEGF. The concentration of

endostatin may be crucial for obtaining a certain effect. A recent report indicates

that the doses of endostatin required for meaningful effects in vitro and in animal

models vary over a wide scale (Sun et al. 2009). It has also been shown, that

endostatin’s effects on angiogenesis and tumor growth are biphasic, operating

over a U-shaped curve (Celik et al. 2005). Endostatin has been shown to block the interaction of VEGF with both Flt-1

and Flk-1/KDR receptors (Kim et al. 2002). Recently endostatin has been found

to bind to VEGFR3 (Kojima et al. 2008), which has a role in osteoblast

differentiation (Orlandini et al. 2006). Furthermore, endostatin has been shown to

affect the VEGF-mediated signaling pathway (Kim et al. 2002) and also ERK1/2

signaling (Schmidt et al. 2006). Since MC3T3-E1 cells do not express Flk-

1/KDR receptors, endostatin can only interact with the Flt-1 or VEGFR3, and

VEGF-A only with the Flt-1 receptor and therefore more endostatin peptides may

be available for binding to VEGFR3.

6.5 The role of collagen XVIII/endostatin on bone development in vivo in transgenic mice overexpressing endostatin and mice

lacking collagen XVIII

Our previous studies and those of others, have suggested that endostatin seems to

have a number of effects on bone and therefore, we wanted to further explore its

role in bone development in vivo using transgenic mice overexpressing endostatin

(ES-tg) and mice lacking collagen XVIII (Col18a1-/-). However, it has been

reported that collagen XVIII/endostatin does not have a role as a critical negative

regulator of angiogenesis during development and postnatal growth, since a lack

of collagen XVIII and endostatin in mice and humans does not increase

angiogenesis in major organs (Marneros & Olsen 2005).

In this study, we detected a delay in the formation of secondary ossification

centers in the bones of 14-day-old Col18a1-/- mice. At the age of one month the

ossification of the epiphyses had already occurred and clear differences were not

79

observed. Delayed ossification was also not detected at embryonic stages E15.5

and E18.5. In ES-tg mice no difference in the ossification of the epiphyseal area

could not be seen. Moreover, the bone mineral content or density, as analyzed

with pQCT was not altered and there were no significant differences in bone sizes

between the Col18a1-/- mice and their controls.

The possible involvement of endostatin in Wnt signaling may explain the

delay in the formation of the ossification centers in Col18a1-/- mice. The longest

variant of collagen XVIII contains a Frizzled-module, which has been shown to

be able to inhibit Wnt/beta-catenin signaling by binding to Wnt-molecules and

hence inhibiting their binding to cell-membrane receptors. The lack of the longest

variant could result in altered Wnt signalling (Quelard et al. 2008). In mice that

constitutively overexpress active β-catenin in cartilage, growth plates were totally

disorganized, and the maturation of chondrocytes and endochondral ossification

were inhibited. The activation of Wnt/β-catenin in mature chondrocytes

stimulates hypertrophy, matrix mineralization, and the expression of VEGF,

MMP-13, and several other MMPs among others (Tamamura et al. 2005). It has

been found, that the phenotype of collagen XVIII deficient mice is similar to mice

expressing the VEGF120 and VEGF188 isoforms only. They have been shown to

have differences in solubility and matrix binding properties and also in their

ability to bind to heparin and heparan sulfate proteoglycans. Since collagen XVIII

is a heparan sulfate proteoglycans it has proposed that collagen XVIII may be

involved in VEGF function and is required for the binding of to its receptor

(Hurskainen et al. 2005).

In ES-tg mice we observed a delay in the growth of bone length: femurs and

tibias were smaller at the age of 14 days. This difference could not be seen in 30-

day-old mice suggesting that this delay was temporary. Endostatin overexpression

takes place in the keratinocytes of ES-tg mice, but they show a seven-fold

increase in the concentration of endostatin in their circulation. It is possible that

the endostatin concentration in the bones is not high enough to cause any longer

lasting or permanent defects, but it only leads to a slight delay in the growth rate.

We used immunohistochemistry to determine the localization of collagen XVIII

in bone tissue. Collagen XVIII was detected surrounding the vessel structures in

bone marrow and periosteum, and at the skeletal muscle attachment site by the

antibody against all three variants. The short variant of collagen XVIII was shown

to be responsible for this immunosignal. Pufe and colleagues reported that

collagen XVIII/endostatin was expressed in human cartilage and fibrocartilage

(Pufe et al. 2004c). In present study, there was a signal in the cartilage, but this

80

proved to be unspecific as staining was also seen in the mice lacking collagen

XVIII. However, this can result from unspecific attachment of our primary

antibodies to the matrix surrounding the chondrocytes and thus it is still possible

that collagen XVIII is expressed in mouse cartilage.

Although mild delays in both mutant mouse lines were detected in the bone

development process, collagen XVIII/endostatin does not seem to have a crucial

role in skeletal development. Endostatin is one of the several regulators of

angiogenesis and it is likely that other factors may compensate for the loss of

collagen/endostatin in most tissues.

81

7 Conclusions

Our results confirmed the important role of VEGF in bone healing. VEGF

overexpression was delivered locally to the bone injury site using adenoviral

VEGF-A gene transfer, which was able to modify the histological features of

bone healing in a rodent model. The endochondral phase was completed earlier,

and also bone mineral content was enhanced in animals overexpressing VEGF.

The data from pit assay studies showed that VEGF-A induces bone resorption

and endostatin can suppress this VEGF-A induced resorption to the basal level. In

addition, osteoclastogenesis can be inhibited by endostatin. However, our findings

suggest that the regulatory effect of endostatin is not critical, since endostatin

alone does not modify the basal bone resorption.

The results from the endochondral ossification model, where the mouse hind

muscle pouch was stimulated by BMPs and endostatin was overexpressed with or

without VEGF-A using a recombinant adenoviral vector, showed that endostatin

retards the cartilage phase and subsequently reduces bone formation.

Endostatin has minor effects on the behavior of osteoblasts in cell culture

conditions. These effects are enhanced when cells are treated together with

VEGF.

Finally, we investigated the role of endostatin on bone development in vivo

using transgenic mice overexpressing endostatin and mice lacking collagen

XVIII. Mild delays in both mutant mouse lines were detected in the bone

development process. The formation of secondary ossification centers in the

bones of mice lacking collagen XVIII was delayed, as well as the growth of bone

lenght in ES-tg mice. On the basis of these results, we conclude that collagen

XVIII/endostatin does not seem to have crucial role in skeletal development.

As an overall conclusion, VEGF-A overexpression accelerates bone defect

healing. Endostatin retards bone formation and seems to inhibit VEGF-A

stimulated effects on bone cells. However, collagen XVIII/endostatin does not

seem to have a crucial role in skeletal development.

82

83

References

Aldinger G, Herr G, Kusswetter W, Reis HJ, Thielmann FW & Holz U (1991) Bone morphogenetic protein: a review. Int Orthop 15: 169–177.

Alsina M, Guise TA & Roodman GD (1996) Cytokine regulation of bone cell differentiation. Vitam Horm 52: 63–98.

Anderson DM, Maraskovsky E, Billingsley WL, Dougall WC, Tometsko ME, Roux ER, Teepe MC, DuBose RF, Cosman D & Galibert L (1997) A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature 390(6656): 175–179.

Anderson HC (2003) Matrix vesicles and calcification. Curr Rheumatol Rep: 5(3): 222–226.

Armamento-Villareal R, Sheikh S, Nawaz A, Napoli N, Mueller C, Halstead LR, Brodt MD, Silva MJ, Galbiati E, Caruso PL, Civelli M & Civitelli R (2005) A new selective estrogen receptor modulator, CHF 4227.01, preserves bone mass and microarchitecture in ovariectomized rats. J Bone Miner Res: 20: 2178–2188.

Asahara T, Murohara T, Sullivan A, Silver M, van der Zee R, Li T, Witzenbichler B, Schatteman G & Isner JM (1997) Isolation of putative progenitor endothelial cells for angiogenesis. Science 275(5302): 964–967.

Aubin JE, Lian JB & Stein GS (2006) Bone Formation: Maturation and Functional Activities of Osteoblast Lineage Cells. In Lian JB & Goldring SR (eds) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. 6th ed. Washington, American Society for Bone and Mineral Research: 20–29.

Aubin JE & Liu F (1996) The osteoblast lineage. In: Bilezikian JP, Raisz LG & Rodan GA (eds) Principles of Bone Biology. San Diego CA, Academic Press: 51–67.

Aubin JE & Triffitt JT (2002) Mesenchymal stem cells and osteoblast differentiation. In Bilezikian JP, Raisz LG & Rodan GA (eds) Principles of bone biology. 2ed. San Diego CA, Academic Press: 59–81.

Barleon B, Sozzani S, Zhou D, Weich HA, Mantovani A & Marmé D (1996) Migration of human monocytes in response to vascular endothelial growth factor (VEGF) is mediated via the VEGF receptor flt-1. Blood 87(8): 3336–3343.

Baron R (1989) Polarity and membrane transport in osteoclasts. Connect Tissue Res 20: 109–120.

Bar-Shavit Z (2007) The osteoclast: a multinucleated, hematopoietic-origin, bone-resorbing osteoimmune cell. J Cell Biochem 102(5): 1130–1139.

Bax BE, Wozney JM & Ashhurst DE (1999) Bone morphogenetic protein-2 increases the rate of callus formation after fracture of the rabbittibia. Calcif Tissue Int 65: 83–89.

Bennett CN, Longo KA, Wright WS, Suva LJ, Lane TF, Hankenson KD & MacDougald OA (2005) Regulation of osteoblastogenesis and bone mass by Wnt10b. Proc Natl Acad Sci USA 102(9): 3324–33249.

Bennett JH, Joyner CJ, Triffitt JT & Owen ME (1991) Adipocytic cells cultured from marrow have osteogenic potential. J Cell Sci 99: 131–139.

84

Bergers G & Song S (2005) The role of pericytes in blood-vessel formation and maintenance Neuro Oncol 7(4): 452–464.

Bessho K, Kusumoto K, Fujimura K, Konishi Y, Ogawa Y, Tani Y & Iizuka T (1999) Comparison of recombinant and purified human bone morphogenetic protein. Br J Oral Maxillofac Surg 37: 2–5.

Bi W, Deng JM, Zhang Z, Behringer RR, Crombrugghe B (1999) Sox9 is required for cartilage formation. Nat. Genet. 22:85–89.

Bix G & Iozzo RV (2005) Matrix revolutions: tails of basement-membrane components with angiostatic functions. Trends in cell biology 15(1): 52–60.

Blair HC, Kahn AJ, Crouch EC, Jeffrey JJ & Teitelbaum SL (1986) Isolated osteoclasts resorb the organic and inorganic components of bone. J Cell Biol 102: 1164–1172.

Blair HC, Teitelbaum SL, Ghiselli R & Gluck S (1989) Osteoclastic bone resorption by a polarized vacuolar proton pump. Science 245(4920): 855–857.

Blokhuis TJ, den Boer FC, Bramer JA, Jenner JMG T, Bakker FC, Patka P & Haarman HJ T M (2001) Biomechanical and histological aspects of fracture healing, stimulated with osteogenic protein-1. Biomaterials 22: 725–730.

Bluteau G, Julien M, Magne D, Mallein-Gerin F, Weiss P, Daculsi G & Guicheux J (2007) VEGF and VEGF receptors are differentially expressed in chondrocytes. Bone 40(3): 568–576.

Bonewald LF (2008) Osteocytes. In Rosen V (ed) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. 7th ed. Washington, American Society for Bone and Mineral Research: 22–27.

Boyce BF, Xing L & Chen D (2005) Osteoprotegerin, the bone protector, is a surprising target for beta-catenin signaling. Cell Metab 2(6): 344–345.

Boyce BF & Xing L (2007) The RANKL/RANK/OPG pathway. Curr Osteoporos Rep 5(3): 98–104.

Boyde A, Ali NN & Jones SJ (1984) Resorption of dentine by isolated osteclasts in vitro. Brit Dent J 156: 216–220.

Boyle WJ, Simonet S & Lacey DL (2003) Review article Osteoclast differentiation and activation. Nature 423: 337–342.

Brandi ML & Collin-Osdoby P (2006) Vascular Biology and the Skeleton. J Bone Miner Res 21(2): 183–192.

Brandi ML, Crescioli C, Tanini A, Frediani U, Agnusdei D & Gennari C (1993) Bone endothelial cells as estrogen targets. Calcif Tissue Int 53(5): 312–317.

Breen EC (2007) VEGF in Biological Control. Journal of Cellular Biochemistry 102: 1358–1367.

Breen SA, Loveday BE, Millest AJ & Waterton JC (1998) Stimulation and inhibition of bone formation: Use of peripheral quantitative computed tomography in the mouse in vivo. Lab Anim 32: 467–476.

Breier G, Albrecht U, Sterrer S & Risau W (1992) Expression of vascular endothelial growth-factor during embryonic angiogenesis and endothelial-cell differentiation. Development 114: 521–532.

85

Cackowski FC & Roodman DG (2007) Perspective on the Osteoclast An Angiogenic Cell? Annals of the New York Academy of Sciences Volume 1117 Issue Skeletal Biology and Medicine, Part B: Disease Mechanisms and Therapeutic Challenges: 12–25

Cao B, Mytinger JR & Huard J (2002) Adenovirusmediated gene transfer to skeletal muscle. Microsc Res Tech 58: 45–51.

Carlevaro MF, Cermelli S, Cancedda R & Descalzi Cancedda F (2000) Vascular endothelial growth factor (VEGF) in cartilage neovascularization and chondrocyte differentiation: auto-paracrine role during endochondral bone formation. Journal of Cell Science 113(1): 59–69.

Carmeliet P , Ferreira V, Breier G, Pollefeyt S, Kieckens L, Gertsenstein M, Fahrig M, Vandenhoeck A, Harpal K, Eberhardt C, Declercq C, Pawling J, Moons L, Collen D, Risau W & Nagy A (1996) Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 380: 435–439.

Carmeliet P (2005) VEGF as a key mediator of angiogenesis in cancer. Oncology 69(3): 4–10.

Celik I, Surucu O, Dietz C, Heymach JV, Force J, Hoschele I, Becker CM, Folkman J & Kisker O (2005) Therapeutic efficacy of endostatin exhibits a biphasic dose-response curve. Cancer Res 65: 11044–11050.

Chambers TJ, Revell PA, Fuller K & Athanasou NA (1984) Resorption of bone by isolated rabbit osteoclasts. J Cell Sci 66: 383–399.

Chen D, Zhao M & Mundy GR (2004) Bone Morphogenetic Proteins. Growth Factors 22(4): 233–241.

Chimal-Monroy J, Montero JA, Gañan Y, Macias D, Garcia-Porrero JA & Hurle JM (2002) Comparative analysis of the expression and regulation of Wnt5a, Fz4, and Frzb1 during digit formation and in micromass cultures. Dev Dyn 224(3): 314–320.

Chuntharapai A, Bodary S, Horton M & Kim KJ (1993) Blocking monoclonal antibodies to alpha V beta 3 integrin: a unique epitope of alpha V beta 3 integrin is present on human osteoclasts. Exp Cell Res 205: 345–352.

Church V, Nohno T, Linker C, Marcelle C & Francis-West P (2002) Wnt regulation of chondrocyte differentiation. J Cell Sci 115: 4809–4818.

Clauss M, Gerlach M, Gerlach H, Brett J, Wang F, Familletti PC, Pan YC, Olander JV, Connolly DT & Stern D (1990) Vascular permeability factor: a tumor-derived polypeptide that induces endothelial cell and monocyte procoagulant activity, and promotes monocyte migration. J Exp Med 172(6): 1535–1545.

Clauss M, Weich H, Breier G, Knies U, Rockl W, Waltenberg J & Risau W (1996) The vascular endotheial growth factor receptor Flt-1 mediates biological activities. Implications for a functional role of placenta growth factor in monocyte activatio and chemotaxis. J boil chem 271: 17629–17634.

Cleaver O & Melton DA (2003) Endothelial signaling during development 1. Nat Med 9(6): 661–668.

Collin-Osdoby P (1994) Role of vascular endothelial cells in bone biology. J Cell Biochem 55: 304–309.

86

Collin-Osdoby P, Rothe L, Anderson F, Nelson M, Maloney W & Osdoby P (2001) Receptor activator of NF-kB lligand and osteoprotegerin expression by human microvascular endothelial cells, regulation by inflammatory cytokines, and role on human osteoclastogenesis. J Biol Chem 276: 20659–20672.

Colnot CI & Helms JA (2001) A molecular analysis of matrix remodeling and angiogenesis during long bone development. Mech Dev 100(2): 245–250.

Colnot CI, Lu C, Hu D & Helms JA (2004) Distinguishing the contributions of the perichondrium, cartilage, and vascular endothelium to skeletal development. Dev Biol 269(1): 55–69.

Dai J & Rabie AB (2007) VEGF: an essential mediator of both angiogenesis and endochondral ossification. J Dent Res 86(10): 937–950.

Day TF, Guo X, Garrett-Beal L & Yang Y (2005) Wnt/beta-catenin signaling in mesenchymal progenitors controls osteoblast and chondrocyte differentiation during vertebrate skeletogenesis. Dev Cell 8(5): 739–750.

Deckers MML, Karperien M, Bent van der C, Yamashita T, Papapoulos SE & Löwik CW (2000) Expression of vascular endothelial growth factors and their receptors during osteoblast differentiation. Endocrinology 141: 1667–1674.

Delaissé JM, Andersen TL, Engsig MT, Henriksen K, Troen T & Blavier L (2003) Matrix metalloproteinases (MMP) and cathepsin K contribute differently to osteoclastic activities. Microsc Res Tech 61(6): 504–513.

Delaney CE, Weagent BT & Addison CL (2006) The inhibitory effects of endostatin on endothelial cells are modulated by extracellular matrix. Exp Cell Res 312(13): 2476–2489.

Dempster DW (2006) Anatomy and Functions of the Adult Skeleton. In 2006 American Society for Bone and Mineral Research. In Lian JB & Goldring SR (eds) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. 6th ed. Washington, American Society for Bone and Mineral Research: 7–11.

Dhanabal M, Ramchandran R, Waterman MJF, Lu H, Knebelmann B, Segal M & Sukhatme VP (1999) Endostatin induces endothelial cell apoptosis. J Biol Chem 274: 11721–11726.

Dixelius J, Cross MJ, Matsumoto T & Claesson-Welsh L (2003) Endostatin action and intracellular signaling: beta-catenin as a potential target? Cancer Letters 196(1): 1–12.

Dixelius J, Larsson H, Sasaki T, Holmqvist K, Lu L, Engström A, Timpl R, Welsh M (2000) Claesson-Welsh L. Endostatin-induced tyrosine kinase signaling through the Shb adaptor protein regulates endothelial cell apoptosis. Blood 95(11): 3403–3411.

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: 3632–3638.

Doherty MJ & Canfield AE (1999) Gene expression during vascular pericyte differentiation. Crit Rev Eukaryot Gene Expr 9: 1–17.

Duong LT, Lakkakorpi P, Nakamura I & Rodan GA (2000) Integrins and signaling in osteoclast function. Matrix Biol 19: 97–105.

87

Eghbali-Fatourechi GZ, Lamsam J, Fraser D, Nagel D, Riggs, BL & Khosla S (2005) Circulating osteoblast-lineage cells in humans. N Engl J Med 352: 1959–1966.

Elamaa H, Sormunen R, Rehn M, Soininen R & Pihlajaniemi T (2005) Endostatin overexpression specifically in the lens and skin leads to cataract and ultrastructural alterations in basement membranes. Am J Pathol 166(1): 221–229.

Engsig MT, Chen QJ, Vu TH, Pedersen AC, Therkidsen B, Lund LR, Henriksen K, Lenhard T, Foged NT, Werb Z & Delaissé JM (2000) Matrix metalloproteinase 9 and vascular endothelial growth factor are essential for osteoclast recruitment into developing long bones. J Cell Biol 151: 879–889.

Erickson AC & Couchman JR (2000) Still more complexity in mammalian basement membranes. J Histochem Cytochem 48(10): 1291–1306.

Eriksen EF, Eghbali-Fatourechi GZ & Khosla S (2007) Remodeling and vascular spaces in Bone. J Bone Miner Res 22: 1–6.

Eshkar-Oren I, Viukov SV, Salameh S, Krief S, Oh C, Akiyama H, Gerber HP, Ferrara N & Zelzer E (2009) The forming limb skeleton serves a signaling center for limb vasculature patterning via regulation of Vegf. Development 136: 1263–1272.

Everts V, Delaissé JM, Korper W & Beertsen W (1998) Cysteine proteinases and matrix metalloproteinases play distinct roles in the subosteoclastic resorption zone. J Bone Miner Res 13(9): 1420–1430.

Everts V, Delaissé JM, Korper W, Jansen DC, Tigchelaar-Gutter W, Saftig P & Beertsen W (2002) The bone lining cell: its role in cleaning Howship's lacunae and initiating bone formation. J Bone Miner Res 17(1): 77–90.

Felbor U, Dreier L, Bryant RA, Ploegh HL, Olsen BR & Mothes W (2000) Secreted cathepsin L generates endostatin from collagen XVIII. EMBO J 19(6): 1187–1194.

Feng Y, Wu YP, Zhu XD, Zhang YH & Ma QJ (2005) Endostatin promotes the anabolic program of rabbit chondrocyte. Cell Res 15: 201–206.

Ferrara N, Carver-Moore K, Chen H, Dowd M, Lu L, O'Shea KS, Powell-Braxton L, Hillan KJ & Moore MW (1996) Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 380(6573): 439–442.

Ferrara N & Davis-Smythe T (1997) The biology of vascular endothelial growth factor. Endocr Rev 18: 4–25.

Ferrara N, Gerber HP & LeCouter J (2003) The biology of VEGF and its receptors. Nat Med 9(6): 669–676.

Fiedler J, Leucht F, Waltenberger J, Dehio C & Brenner RE (2005). VEGF-A and PlGF-1 stimulate chemotactic migration of human mesenchymal progenitor cells. Biochem Biophys Res Commun 334: 561–568.

Folkman J (2006) Antiangiogenesis in cancer therapy- endostatin and its mechanism of action. Exp Cell Res 312: 594–607.

Folkman J & Shing Y (1992) Angiogenesis. J Biol Chem 267(16): 10931–10934. Fong GH, Rossant J, Gertsenstein M & Breitman ML (1995) Role of the Flt-1 receptor

tyrosine kinase in regulating the assembly of vascular endothelium. Nature 376(6535): 66–70.

88

Formigli L, Orlandini S, Benvenuti S, Masi L, Pinto A, Gattei V, Bernabei P, Robey P, Collin-osdoby P, Brandi M (1995) In vitro structural and functional relationships between preosteoclastic and bone endothelial cells: A juxtacrine model for migration and adhesion of osteoclast precursors. J Cell Physiol 162: 199–212.

Forslund C & Aspenberg P (1998) OP-1 has more effect than mechanical signals in the control of tissue differentiation in healing rat tendons. Acta Orthop Scand 69: 622–626.

Franceschi RT (2005) Biological approaches to bone regeneration by gene therapy. J Dent Res 84(12): 1093–1103.

Franceschi RT & Xiao G (2003) Regulation of the osteoblast-specific transcription factor, Runx2: responsiveness to multiple signal transduction pathways. J Cell Biochem 88(3): 446–454.

Frost HM (1973) Bone remodeling and its relationship to metabolic bone disease. Springfield IL, Charles C Thomas.

Frost HM (1986) Intermediary organization of the skeleton. Boca Raton FL, CRC Press. Fukai N, Eklund L, Marneros AG, Oh SP, Keene DR, Tamarkin L, Niemelä M, Ilves M, Li

E, Pihlajaniemi T & Olsen BR (2002) Lack of collagen XVIII/endostatin results in eye abnormalities. EMBO J 21(7): 1535–1544.

Gasser JA, Ingold P, Venturiere A, Shen V & Green JR (2008) Longterm protective effects of zoledronic acid on cancellous and cortical bone in the ovariectomized rat. J Bone Miner Res 23: 544–551.

Gerber HP, Vu TH, Ryan AM, Kowalski J, Werb Z & Ferrara N (1999) VEGF couples hypertrophic cartilage remodeling, ossification and angiogenesis during endochondral bone formation. Nat Med 5(6): 623–628.

Glass DA 2nd, Bialek P, Ahn JD, Starbuck M, Patel MS, Clevers H, Taketo MM, Long F, McMahon AP, Lang RA & Karsenty G (2005) Canonical Wnt signaling in differentiated osteoblasts controls osteoclast differentiation. Dev Cell 8(5): 751–764.

Gonzales-Crussi F (1971) Vasculogenesis in thc chick embryo: an ultrastruc tural study. Am J Anat 130: 441–460.

Gowen M, Lazner F, Dodds R, Kapadia R, Feild J, Tavaria M, Bertoncello I, Drake F, Zavarselk S, Tellis I, Hertzog P, Debouck C & Kola I (1999) Cathepsin K knockout mice develop osteopetrosis due to a deficit in matrix degradation but not demineralization. J Bone Miner Res 14(10): 1654–1663.

Grellier M, Ferreira-Tojais N, Bourget C, Bareille R, Guillemot F, Amedee J (2009) Role of vascular endothelial growth factor in the communication between human osteoprogenitors end endothelial cells. J Cellular Bioch 106: 390–398.

Grigoriadis AE, Heersche JN & Aubin JE (1988) Differentiation of muscle, fat, cartilage, and bone from progenitor cells present in a bone-derived clonal cell population: effect of dexamethasone. J Cell Biol 106: 2139–2151.

Guo D, Jia Q, Song HY, Warren RS & Donner DB (1995) Vascular endothelial cell growth factor promotes tyrosine phosphorylation of mediators of signal transduction that contain SH2 domains. Association with endothelial cell proliferation. J Biol Chem 270: 6729–6733.

89

Haigh JJ, Gerber HP, Ferrara N & Wagner EF (2000) Conditional inactivation of VEGF-A in areas of collagen2a1 expression results in embryonic lethality in the heterozygous state. Development 127(7): 1445–1453.

Hajitou A, Grignet C, Devy L, Berndt S, Blacher S, Deroanne CF, Bajou K, Fong T, Chiang Y, Foidart JM & Noël A (2002) The antitumoral effect of endostatin and angiostatin is associated with a down-regulation of vascular endothelial growth factor expression in tumor cells. FASEB J 16(13): 1802–1804.

Halfter W, Dong S, Schurer B & Cole GJ (1998) Collagen XVIII is a basement membrane heparan sulfate proteoglycan. J Biol Chem 273(39): 25404–25412.

Hall AP, Westwood FR & Wadsworth PF (2006) Review of the Effects of Anti-Angiogenic Compounds on the Epiphyseal Growth Plate. Toxicologic Pathology 34: 131–147.

Hall BK & Miyake T (1992) The membranous skeleton: the role of cell condensations in vertebrate skeletogenesis. Anat Embryol 186(2): 107–124.

Halleen JM, Räisänen S, Salo JJ, Reddy SV, Roodman GD, Hentunen TA, Lehenkari PP, Kaija H, Vihko P & Väänänen HK (1999) Intracellular fragmentation of bone resorption products by reactive oxygen species generated by osteoclastic tartrate-resistant acid phosphatase. The Journal of biological chemistry 274(33): 22907–22910.

Hanai J, Gloy J, Karumanchi SA, Kale S, Tang J, Hu G, Chan B, Ramchandran R, Jha V, Sukhatme VP & Sokol S (2002) Endostatin is a potential inhibitor of Wnt signaling. Cell Biol 158(3): 529–539.

Harada S, Nagy JA, Sullivan KA, Thomas KA, Endo N & Rodan G (1994) Induction of vascular endothelial growth factor expression by prostaglandin E2 and E1 in osteoblasts. J Clin Invest 93: 2490–2496.

Harper J, Gerstenfeld LC & Klagsbrun M (2001) Neuropilin-1 expression in osteogenic cells: down-regulation during differentiation of osteoblasts into osteocytes. J Cell Biochem 81: 82–92.

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.

He X, Semenov M, Tamai K & Zeng X (2004) LDL receptor-related proteins 5 and 6 in Wnt/beta-catenin signaling: arrows point the way. Development 131(8): 1663–1677.

Hedgecock NL, Hadi T, Chen AA, Curtiss SB, Martin RB & Hazelwood SJ (2007) Quantitative regional associations between remodeling, modeling, and osteocyte apoptosis and density in rabbit tibial midshafts. Bone 40(3): 627–637.

Heino TJ, Hentunen TA & Väänänen HK (2002) Osteocytes inhibit osteoclastic bone resorption through transforming growth factor-beta: enhancement by estrogen. J Cell Biochem 85(1): 185–197.

Henriksen K, Karsdal M, Delaisse JM & Engsig MT (2003) RANKL and vascular endothelial growth factor (VEGF) induce osteoclast chemotaxis through an ERK1/2-dependent mechanism. Biol Chem 278(49): 48745–48753.

Hill PA (1998) Bone remodeling. British Journal of Orthodontics 25: 101–107.

90

Hill TP, Später D, Taketo MM, Birchmeier W & Hartmann C (2005) Canonical Wnt/beta-catenin signaling prevents osteoblasts from differentiating into chondrocytes. Dev Cell 8(5): 727–738.

Hiltunen MO, Ruuskanen M, Huuskonen J, Mahonen AJ, Ahonen M, Rutanen J, Kosma VM, Mahonen A, Kroger H & Yla-Herttuala S (2003). Adenovirus-mediated VEGF-A gene transfer induces bone formation in vivo. FASEB J 17: 1147–1149.

Hofbauer LC & Heufelder AE (2001) Role of receptor activator of nuclear factor-kappaB ligand and osteoprotegerin in bone cell biology. J Mol Med 79(5–6): 243–253.

Hohenester E, Sasaki T, Olsen BR & Timpl R (1998) Crystal structure of the angiogenesis inhibitor endostatin at 1.5 A resolution. EMBO J 17(6): 1656–1664.

Holmen SL, Zylstra CR, Mukherjee A, Sigler RE, Faugere MC, Bouxsein ML, Deng L, Clemens TL & Williams BO (2005) Essential role of beta-catenin in postnatal bone acquisition. J Biol Chem 280(22): 21162–21168.

Horner A, Bishop NJ, Bord S, Beeton C, Kelsall AW, Coleman N & Compston JE (1999) Immunolocalisation of vascular endothelial growth factor (VEGF) in human neonatal growth plate cartilage. J Anat 194: 519–524.

Houck KA, Ferrara N, Winer J, Cachianes G, Li B & Leung DW (1991) The vascular endothelial growth factor family: identification of a fourth molecular species and characterization of alternative splicing of RNA. Mol Endocrinol 5: 1806–1814.

Huang X, Cho S & Spangrude GJ (2007) Hematopoietic stem cells: generation and self renewal. Cell Death Differ 14: 1851–1859.

Huang Z & Bao SD (2004) Roles of main pro-and anti-angiogenic factors in tumor angiogenesis. World J Gastroenterol 10: 463–470.

Hurskainen M, Eklund L, Hagg PO, Fruttiger M, Sormunen R, Ilves M & Pihlajaniemi T (2005) Abnormal maturation of the retinal vasculature in type XVIII collagen/endostatin deficient mice and changes in retinal glial cells due to lack of collagen types XV and XVIII. FASEB J 19: 1564–1566.

Hutchings H, Ortega N & Plouet J (2003) Extracellular matrix-bound vascular endothelial growth factor promotes endothelial cell adhesion, migration, and survival through integrin ligation. FASEB J 17: 1520–1522.

Inada M, Yasui T, Nomura S, Miyake S, Deguchi K, Himeno M, Sato M, Yamagiwa H, Kimura T, Yasui N, Ochi T, Endo N, Kitamura Y, Kishimoto T & Komori T (1999) Maturational disturbance of chondrocytes in Cbfa1-deficient mice. Dev Dyn 214(4): 279–290.

Iwata H, Sakano S, Itoh T & Bauer TW (2002) Demineralized bone matrix and native bone morphogenetic protein in orthopaedic surgery. Clin Orthop Relat Res 395: 99–109.

Jortikka L, Marttinen A & Lindholm TS (1993a) Purification of monocomponent bovine bonemorphogenetic protein in a watersoluble form. Ann Chir Gynaecol Suppl 207: 25–30.

Jortikka L, Marttinen A & Lindholm TS (1993b) High yield of osteoinductivity can be derived from demineralized bone matrix using collagenase digestion. Ann Chir Gynaecol Suppl 207: 31–35.

91

Jurdic P, Saltel F, Chabadel A & Destaing O (2006) Podosome and sealing zone: specificity of the osteoclast model. Eur J Cell Biol 85(3–4): 195–202.

Kaipainen A, Korhonen J, Mustonen T, van Hinsbergh VW, Fang GH, Dumont D, Breitman M & Alitalo K (1995) Expression of the fms-like tyrosine kinase 4 gene becomes restricted to lymphatic endothelium during development. Proc Natl Acad Sci USA (8): 3566–3570.

Kakudo N, Kusumoto K, Wang YB, Iguchi Y & Ogawa Y (2006) Immunolocalization of vascular endothelial growth factor on intramuscular ectopic osteoinduction by bone morphogenetic protein-2. Life Sci 79(19): 1847–55.

Karaplis AC (2002) Embryonic Development of Bone and the Molecular Regulation of Intramembranous and Endochondral Bone Formation. In Bilezikian JP, Raisz LG & Rodan GA (eds) Principles of Bone Biology, 2.ed. San Diego CA, Academic Press: 33–58.

Karsenty G (2003) The complexities of skeletal biology. Nature 423(6937): 316–318. Karsenty G & Wagner EF (2002) Reaching a genetic and molecular understanding of

skeletal development. Dev Cell 2(4): 384–406. Karumanchi SA, Jha V, Ramchandran R, Karihaloo A, Tsiokas L, Chan B, Dhanabal M,

Hanai JI, Venkataraman G, Shriver Z, Keiser N, Kalluri R, Zeng H, Mukhopadhyay D, Chen RL, Lander AD, Hagihara K, Yamaguchi Y, Sasisekharan R, Cantley L & Sukhatme VP (2001) Cell surface glypicans are low-affinity endostatin receptors. Mol Cell 7(4): 811–822.

Katagiri T & Takahashi N (2002) Regulatory mechanisms of osteoblast and osteoclast differentiation. Oral Dis 8(3): 147–159.

Keramaris NC, Calori GM, Nikolaou VS, Schemitsch EH & Giannoudis PV (2008) Fracture vascularity and bone healing: A systematic review of the role of VEGF. Injury 39: 45–57.

Khosla S, Westendorf JJ & Oursler MJ (2008) Building bone to reverse osteoporosis and repair fractures. J Clin Invest 118: 421–428.

Kim IS, Otto F, Zabel B & Mundlos S (1999) Regulation of chondrocyte differentiation by Cbfa1. Mech Dev 80(2): 159–170.

Kim YM, Hwang S, Kim YM, Pyun BJ, Kim TY, Lee ST, Gho YS & Kwon YG (2002) Endostatin blocks vascular endothelial growth factor-mediated signaling via direct interaction with KDR/Flk-1. J Biol Chem 277(31): 27872–27879.

Kitagawa Y, Dai J, Zhang J, Keller JM, Nor J, Yao Z & Keller ET (2005) Vascular endothelial growth factor contributes to prostate cancer-mediated osteoblastic activity. Cancer Res 65: 10921–10929.

Knothe Tate ML, Adamson JR, Bauer TW (2004) The osteocyte. Int J Biochem Cell Biol 36(1): 1–8.

Kofron MD & Laurencin CT (2006) Bone tissue engineering by gene delivery. Adv Drug Deliv Rev 58(4): 555–576.

Kojima T, Azar DT & Chang JH (2008) Neostatin-7 regulates bFGF-induced corneal lymphangiogenesis. FEBS Lett. 582: 2515–2520.

92

Komori T, Yagi H, Nomura S, Yamaguchi A, Sasaki K, Deguchi K, Shimizu Y, Bronson RT, Gao YH, Inada M, Sato M, Okamoto R, Kitamura Y, Yoshiki S & Kishimoto T (1997) Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89(5): 677–680.

Kreuger J, Matsumoto T, Vanwildemeersch M, Sasaki T, Timpl R, Claesson-Welsh L, Spillmann D & Lindahl U (2002) Role of heparan sulfate domain organization in endostatin inhibition of endothelial cell function. EMBO J 21(23): 6303–6311.

Krishnan V, Bryant H & MacDougald O (2006) Regulation of bone mass by Wnt signaling. J Clin Invest 116(5): 1202–1209.

Kumar VB, Viji RI, Kiran MS & Sudhakaran PR (2007) Endothelial cell response to lactate. J Cell Physiol 211: 477–485.

Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, Elliott R, Colombero A, Elliott G, Scully S, Hsu H, Sullivan J, Hawkins N, Davy E, Capparelli C, Eli A, Qian YX, Kaufman S, Sarosi I, Shalhoub V, Senaldi G, Guo J, Delaney J & Boyle WJ (1998) Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93(2): 165–176.

Laitinen M, Makinen K, Manninen H, Matsi P, Kossila M, Agrawal RS, Pakkanen T, Luoma JS, Viita H, Hartikainen J, Alhava E, Laakso M & Yla-Herttuala S (1998) Adenovirusmediated gene transfer to lower limb artery of patientswith chronic critical leg ischemia. Hum Gene Ther 9: 1481–1486.

Lakkakorpi PT, Horton MA, Helfrich MH, Karhukorpi EK & Väänänen HK (1991) Vitronectin receptor has a role in bone resorption but does not mediate tight sealing zone attachment of osteoclast to the bone surface. J Cell Biol 115: 1179–1186.

Lau KH, Kapur S, Kesavan C & Baylink DJ (2006) Up-regulation of the Wnt, estrogen receptor, insulin-like growth factor-I, and bone morphogenetic protein pathways in C57BL/6J osteoblasts as opposed to C3H/HeJ osteoblasts in part contributes to the differential anabolic response to fluid shear. J Biol Chem 281(14): 9576–9588.

Lee S, Jilani SM, Nikolova GV, Carpizo D & Iruela-Arispe ML (2005) Processing of VEGF-A by matrix metalloproteinases regulates bioavailability and vascular pattering in tumors. J Cell Biol 169(4): 681–691.

Leung DW, Cachianes G, Kuang WJ, Goeddel DV & Ferrara N (1989) Vascular endothelial factor is a secreted angiogenic mitogen. Science 246: 1306–1309.

Li G, Bouxsein ML, Luppen C, Li XJ, Wood M, Seeherman HJ, Wozney JM & Simpson H (2002) Bone consolidation is enhanced by rhBMP-2 in a rabbit model of distraction osteogenesis. J Orthop Res 20: 779–788.

Lieberman JR, Daluiski A & Einhorn TA (2002) The role of growth factors in the repair of bone. J Bone Joint Surg Am 84: 1032–1044.

Liotta LA, Steeg PS & Stetler-Stevenson WG (1991) Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation. Cell 64(2): 327–336.

Liu C, Li Y, Semenov M, Han C, Baeg GH, Tan Y, Zhang Z, Lin X & He X (2002) Control of beta-catenin phosphorylation/degradation by a dual-kinase mechanism. Cell 2002 108(6): 837–847.

93

Liu F, Kohlmeier S & Wang C (2008) Wnt signaling and skeletal development. Cell Signal 20: 999–1009.

Logan CY & Nusse R (2004) The Wnt signaling pathway in development and disease. Annu Rev Cell Dev Biol 20: 781–810.

MacDonald NJ, Shivers WY, Narum DL, Plum SM, Wingard JN, Fuhrmann SR, Liang H, Holland-Linn J, Chen DH & Sim BK (2001) Endostatin binds tropomyosin. A potential modulator of the antitumor activity of endostatin. J Biol Chem 276(27): 25190–25196.

Maes C & Carmeliet G (2008) Vascular and nonvascular roles of VEGF in bone development. In Ruhrberg C (ed) VEGF in Development. Austin TX, Landes Bioscience: 79–90.

Maes C, Carmeliet P, Moermans K, Stockmans I, Smets N, Collen D, Bouillon R & Carmeliet G (2002) Impaired angiogenesis and endochondral bone formation in mice lacking the vascular endothelial growth factor isoforms VEGF164 and VEGF188. Mech Dev 111(1–2): 61–73.

Maes C, Stockmans I, Moermans K, Van Looveren R, Smets N, Carmeliet P, Bouillon R & Carmeliet G (2004) Soluble VEGF isoforms are essential for establishing epiphyseal vascularization and regulating chondrocyte development and survival. J Clin Invest 113(2): 188–199.

Malone AM, Anderson CT, Tummala P, Kwon RY, Johnston TR, Stearns T & Jacobs CR (2007) Primary cilia mediate mechanosensing in bone cells by a calcium-independent mechanism. Proc Natl Acad Sci USA 104: 13325–13330.

Malone JD, Teitelbaum SL, Griffin GL, Senior RM & Kahn AJ (1982) Recruitment of osteoclast precursors by purified bone matrix constituents. The Journal of Cell Biology 92: 227–230.

Marks SC & Odgren PR (2002) Structure and development of the skeleton. In Bilezikian JP, Raisz LG, & Rodan GA (eds) Principles of bone biology. 2ed. San Diego CA, Academic Press: 3–15.

Marneros AG & Olsen BR (2005) Physiological role of collagen XVIII and endostatin. Faseb J 19: 716–728.

Matsuno H, Yodoh K, Uzuki M, Nakazawa T, Sawai N, Yamaguchi N, Olsen BR & Kimura T (2002) Treatment with the angiogenesis inhibitor endostatin: a novel therapy in rheumatoid arthritis. J Rheumatol 29: 890–895.

Matsuo K & Irie N (2008) Osteoclast-osteoblast communication. Arch Biochem Biophys 473(2): 201–209.

Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, Libermann TA, Morgan JP, Sellke FW, Stillman IE, Epstein FH, Sukhatme VP & Karumanchi SA (2003) Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 111(5): 649–658.

Mayr-Wohlfart U, Waltenberger J, Hausser H, Kessler S, Günther KP, Dehio C, Puhl W & Brenner RE (2002) Vascular endothelial growth factor stimulates chemotactic migration of primary human osteoblasts. Bone 30(3): 472–477.

94

McCann RM, Colleary G, Geddis C, Clarke SA, Jordan GR, Dickson GR & Marsh D (2008) Effect of osteoporosis on bone mineral density and fracture repair in a rat femoral fracture model. J Orthop Res 26: 384–393.

Midy V & Plouet J (1994) Vasculotropin/vascular endothelial growth factor induces differentiation in cultured osteoblasts. Biochem Biophys Res Commun 199: 380–386.

Min JK, Kim YM, Kim YM, Kim EC, Gho YS, Kang IJ, Lee Sy, Kong YY & Kwon YG (2003) Vascular endothelial growth factor up-regulates expression of receptor activator of NF-kappa B (RANK) in endothelial cells. Concomitant increase of angiogenic responses to RANK ligand. J Biol Chem 278: 39548–39557.

Moon RT, Kohn AD, De Ferrari GV & Kaykas A (2004) WNT and beta-catenin signalling: diseases and therapies. Nat Rev Genet 5(9): 691–701.

Moses MA, Wiederschain D, Wu I, Fernandez CA, Ghazizadeh V, Lane WS, Flynn E, Sytkowski A, Tao T & Langer R (1999) Troponin I is present in human cartilage and inhibits angiogenesis. Proc Natl Acad Sci USA 96: 2645–2650.

Mulari MT, Qu Q, Härkönen PL & Väänänen HK (2004) Osteoblast-like cells complete osteoclastic bone resorption and form new mineralized bone matrix in vitro. Calcif Tissue Int 75(3): 253–261.

Mundy GR (1990) Bone resorbing cells. In: Favus MJ (ed) Primer on the metabolic bone disease and disorders of mineral metabolism. Kelseyville, American Society of Bone and Mineral Research: 18–22.

Mundy GR (2002) Directions of drug discovery in osteoporosis. Annual Review of Medicine 53: 337–354.

Muragaki Y, Timmons S, Griffith CM, Oh SP, Fadel B, Quertermous T & Olsen BR (1995) Mouse Col18a1 is expressed in a tissue-specific manner as three alternative variants and is localized in basement membrane zones. Proc Natl Acad Sci USA 92(19): 8763–8767.

Murphy WL, Simmons CA, Kaigler D & Mooney DJ (2004) Bone regeneration via a mineral substrate and induced angiogenesis. J Dent Res 83: 204–210.

Muschik M, Schlenzka D, Ritsila V, Tennstedt C & Lewandrowski KU (2000) Experimental anterior spine fusion using bovine bone morphogenetic protein: a study in rabbits. J Orthop Sci 5: 165–170.

Nakagawa M, Kaneda T, Arakawa T, Morita S, Sato T, Yomada T, Hanada K, Kumegawa M & Hakeda Y (2000) Vascular endothelial growth factor (VEGF) directly enhances osteoclastic bone resorption and survival of mature osteoclasts. FEBS Lett 473(2): 161–164.

Nakamura T, Imai Y, Matsumoto T, Sato S, Takeuchi K, Igarashi K, Harada Y, Azuma Y, Krust A, Yamamoto Y, Nishina H, Takeda S, Takayanagi H, Metzger D, Kanno J, Takaoka K, Martin TJ, Chambon P & Kato S (2007) Estrogen prevents bone loss via estrogen receptor alpha and induction of Fas ligand in osteoclasts. Cell 130(5): 811–823.

Nakashima K, Zhou X, Kunkel G, Zhang Z, Deng JM, Behringer RR & de Crombrugghe B (2002) The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell 108(1): 17–29.

95

Nalbantoglu J, Larochelle N, Wolf E, Karpati G, Lochmuller H & Holland PC (2001) Muscle-specific overexpression of the adenovirus primary receptor CAR overcomes low efficiency of gene transfer to mature skeletal muscle. J Virol 75: 4276–4282.

Nesbitt SA & Horton MA (1997) Trafficking of matrix collagens through bone-resorbing osteoclasts. Science 276(5310): 266–269.

Neufeld G, Kessler O & Herzog Y (2002) The interaction of Neuropilin-1 and Neuropilin-2 with tyrosine-kinase receptors for VEGF. Adv Exp Med Biol 515: 81–90.

Niida S, Kaku M, Amano H, Yoshida H, Kataoka H, Nishikawa S, Tanne K, Maeda N, Nishikawa S & Kodama H (1999) Vascular endothelial growth factor can substitute for macrophage colony-stimulating factor in the support of osteoclastic bone resorption. J Exp Med 190(2): 293–298.

Niida S, Kondo T, Hiratsuka S, Hayashi S, Amizuka N, Noda T, Ikeda K & Shibuya M (2005) VEGF receptor 1 signaling is essential for osteoclast development and bone marrow formation in colony-stimulating factor 1-deficient mice. Proc Natl Acad Sci USA 102(39): 14016–14021.

Nijweide PJ, Burger EH & Klein-Nulend J (2002) The Osteocyte. In Bilezikian JP, Raisz LG, & Rodan GA (eds) Principles of bone biology. 2 ed. San Diego CA, Academic Press: 93–108.

Noble BS (2008) The osteocyte lineage. Arch Biochem Biophys 473:106–111. Okubo Y, Bessho K, Fujimura K, Kaihara S, Iizuka T & Miyatake SI (2001) The time

course study of osteoinduction by bone morphogenetic protein-2 via adenoviralvector. Life Sci 70: 325–336.

Olsen BR (2006) Bone Embryology. In Lian JB & Goldring SR (eds) Primer on yhe Metabolic Bone Diseases and Disorders of Mineral Metabolism. Washington, American Society for Bone and Mineral Research: 2–6.

Olsson AK, Dimberg A, Kreuger J & Claesson-Welsh L (2006) VEGF receptor signalling- in control of vascular function. Nat Rev Mol Cell Biol 7: 359–371.

O'Reilly MS, Boehm T, Shing Y, Fukai N, Vasios G, Lane WS, Flynn E, Birkhead JR, Olsen BR & Folkman J (1997) Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell 88: 277–285.

Orlandini M, Spreafico A, Bardelli M, Rocchigiani M, Salameh A, Nucciotti S, Capperucci C, Frediani B & Oliviero S (2006) Vascular endothelial growth factor-D activates VEGFR-3 expressed in osteoblasts inducing their differentiation. J Biol Chem 281: 17961–17967.

Otto F, Thornell AP, Crompton T, Denzel A, Gilmour KC, Rosewell IR, Stamp GW, Beddington RS, Mundlos S, Olsen BR, Selby PB & Owen MJ (1997) Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 89(5): 765–771.

Ott SM (2002) Histomorphometric analysis of bone remodeling. In Bilezikian JP, Raisz LG, & Rodan GA (eds) Principles of bone biology. 2 ed. San Diego CA, Academic Press: 303–320.

Parfitt AM (2002) High bone turnover is intrinsically harmful: two paths to a similar conclusion. The Parfitt view. J Bone Miner Res 17(8): 1558–1559.

96

Parfitt AM (2000) The mechanism of coupling: a role for the vasculature. Bone 26(4): 319–323.

Parfitt AM, Han ZH, Palnitkar S, Rao DS, Shih MS & Nelson D (1997) Effects of ethnicity and age or menopause on osteoblast function, bone mineralization, and osteoid accumulation in iliac bone. J Bone Miner Res 12(11): 1864–1873.

Partfitt 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: 150–159.

Peng H, Wright V, Usas A, Gearhart B, Shen HC, Cummins J & Huard J (2002) Synergistic enhancement of bone formation and healing by stem cell-expressed VEGF and bone morphogenetic protein-4. J Clin Invest 110: 751–759.

Peng H, Usas A, Olshanski A, Ho AM, Gearhart B, Cooper GM & Huard J (2005) VEGF improves whereas sFlt1 inhibits, BMP2-induced bone formation and bone healing through modulation of angiogenesis. J. Bone Miner Res 20: 2017–2027.

Pekkarinen T, Lindholm TS, Hietala O & Jalovaara P (2005) The effect of different mineral frames on ectopic bone formation in mouse hind leg muscles induced by native reindeer bone morphogenetic protein. Arch Orthop Trauma Surg 125: 10–15.

Pfander D, Cramer T, Schipani E & Johnson RS (2003) HIF-1alpha controls extracellular matrix synthesis by epiphyseal chondrocytes. J Cell Sci 116(9): 1819–1826.

Poltorak Z, Cohen T & Neufeld G (2000) The VEGF splice variants: properties, receptors, and usage for the treatment of ischemic diseases. Herz 25(2): 126–129.

Polverini PJ (1995) The pathophysiology of angiogenesis. Crit Rev Oral Biol Med 6(3): 230–47.

Pufe T, Harde V, Petersen W, Goldring MB, Tillmann B & Mentlein R (2004a) Vascular endothelial growth factor (VEGF) induces matrix metalloproteinase expression in immortalized chondrocytes. J Pathol 202(3): 367–374.

Pufe T, Lemke A, Kurz B, Petersen W, Tillman B, Grodzinsky AJ, Mentlein R (2004b) Mechanical overload induces VEGF in cartilage discs via hypoxia-inducible factor. Am J Pathol 164(1): 185–192.

Pufe T, Mentlein R, Tsokos M, Steven P, Varoga D, Goldring MB, Tillmann BN & Paulsen FP (2004d) VEGF expression in adult permanent thyroid cartilage: implications for lack of cartilage ossification. Bone 35: 543–552.

Pufe T, Petersen WJ, Miosge N, Goldring MB, Mentlein R, Varoga DJ & Tillmann BN (2004c) Endostatin/collagen XVIII--an inhibitor of angiogenesis--is expressed in cartilage and fibrocartilage. Matrix Biol 23(5): 267–276.

Pufe T, Wildemann B, Petersen W, Mentlein R, Raschke M & Schmidmaier G (2002) Quantitative measurement of the splice variants 120 and 164 of the angiogenic peptide vascular endothelial growth factor in the time flow of fracture healing: a study in the rat. Cell Tissue Res 309: 387–392.

Pulkkanen KJ, Laukkanen JM, Fuxe J, Kettunen MI, Rehn M, Kannasto JM, Parkkinen JJ, Kauppinen RA, Pettersson RF & Yla-Herttuala S (2002) The combination of HSV-tk and endostatin gene therapy eradicates orthotopic human renal cell carcinomas in nude mice. Cancer Gene Ther 9: 908–916.

97

Puumalainen AM, Vapalahti M, Agrawal RS, Kossila M, Laukkanen J, Lehtolainen P, Viita H, Paljarvi L, Vanninen R & Yla-Herttuala S (1998) Betagalactosidase gene transfer to human malignant glioma in vivo using replication-deficient retroviruses and adenoviruses. Hum Gene Ther 9: 1769–1774.

Quélard D, Lavergne E, Hendaoui I, Elamaa H, Tiirola U, Heljasvaara R, Pihlajaniemi T, Clément B, Musso O (2008) A cryptic frizzled module in cell surface collagen 18 inhibits Wnt/beta-catenin signaling. PLoS One 2;3(4):e1878.

Quinn TP, Peters KG, De VC, Ferrara N & Williams LT (1993) Fetal liver kinase 1 is a receptor for vascular endothelial growth factor and is selectively expressed in vascular endothelium. Proc Natl Acad Sci USA 90: 7533–7537.

Reddy SV (2004) Regulatory mechanisms operative in osteoclasts. Crit Rev Eukaryot Gene Expr 14(4): 255–270.

Rehn M & Pihlajaniemi T (1995) Identification of three N-terminal ends of type XVIII collagen chains and tissue-specific differences in the expression of the corresponding transcripts. The longest form contains a novel motif homologous to rat and Drosophila frizzled proteins. J Biol Chem 270: 4705–4711.

Rehn M, Veikkola T, Kukk-Valdre E, Nakamura H, Ilmonen M, Lombardo C, Pihlajaniemi T, Alitalo K & Vuori K (2001) Interaction of endostatin with integrins implicated in angiogenesis. Proc Natl Acad Sci USA 98: 1024–1029.

Reya T & Clevers H (2005) Wnt signalling in stem cells and cancer. Nature 434(7035): 843–850.

Risau W (1995) Differentiation of endothelium. FASEB J 9: 926–933. Risau W (1997) Mechanisms of angiogenesis. Nature 386: 671–674 . Robey PG (2002) Bone matrix proteoglycans and glycoproteins. In: Bilezikian JP, Raisz

LG & Rodan GA (eds) Principles of bone biology. 2 ed. San Diego CA, Academic Press: 225–237.

Rodan GA & Martin TJ (2000) Therapeutic approaches to bone diseases. Science 289(5484): 1508–1514.

Rodda SJ & McMahon AP (2006) Distinct roles for Hedgehog and canonical Wnt signaling in specification, differentiation and maintenance of osteoblast progenitors. Development 133(16): 3231–3244.

Roodman GD (1999) Cell biology of the osteoclast. Exp Hematol 27: 1229–1241. Roodman GD (2001) Biology of osteoclast activation in cancer. J Clin Oncol 19(15):

3562–3571. Roodman GD & Windle JJ (2005) Paget disease of bone. J Clin Invest 115: 200–208. Saarela J, Ylikarppa R, Rehn M, Purmonen S & Pihlajaniemi T (1998) Complete primary

structure of two variant forms of human type XVIII collagen and tissue-specific differences in the expression of the corresponding transcripts. Matrix Biol 16: 319–328.

Salo J, Lehenkari P, Mulari M, Metsikkö K & Väänänen HK (1997) Removal of osteoclast bone resorption products by transcytosis. Science 276(5310): 270–273.

98

Sampath TK & Reddi AH (1981) Dissociative extraction and reconstitution of extracellular matrix components involved in local bone differentiation. Proc Natl Acad Sci USA 78(12): 7599–7603.

Sasaki T, Larsson H, Kreuger J, Salmivirta M, Claesson-Welsh L, Lindahl U, Hohenester E & Timpl R (1999) Structural basis and potential role of heparin/heparan sulfate binding to the angiogenesis inhibitor endostatin. EMBO J 18(22): 6240–6248.

Schett G, Hayer S, Zwerina J, Redlich K & Smolen JS (2005) Mechanisms of Disease: the link between RANKL and arthritic bone disease. Nat Clin Pract Rheumatol 1(1): 47–54.

Schmidt A, Wenzel D, Thorey I, Sasaki T, Hescheler J, Timpl R, Addicks K, Werner S, Fleischmann BK & Bloch W (2006) Endostatin influences endothelial morphology via the activated ERK1/2-kinase endothelial morphology and signal transduction. Microvasc Res 71: 152–162.

Senger DR, Galli SJ, Dvorak AM, Perruzzi CA & Harvey VS (1983) Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid. Science 219: 983–985.

Shalaby F, Rossant J, Yamaguchi TP, Gertsenstein M, Wu XF, Breitman ML & Schuh AC (1995) Failure of blood-island formation and vascuogenesis in Flk-1-deficient mice. Nature 376(6535): 62–66.

Simonet WS, Lacey DL, Dunstan CR, Kelley M, Chang MS, Luthy R, Nguyen HQ, Wooden S, Bennett L, Boone T, Shimamoto G, DeRose M, Elliot R, Colombero A, Tan HL, Trail G, Sullivan J, Davy E, Bucay N, Renshaw-Gegg L, Hughes TM, Hill D, Pattison W, Campbell P, Sander S, Van G, Tarpley J, Derby P, Lee R & Boyle WJ (1997) Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89(2): 309–319.

Soker S, Takashima S, Miao HQ, Neufeld G & Klagsbrun M (1998) Neuropilin-1 is expressed by endothelial and tumor cells as an isoform-specific receptor for vascular endothelial growth factor. Cell 92(6): 735–745.

Spector JA, Mehrara BJ, Luchs JS, Greenwald JA, Fagenholz PJ, Saadeh PB, Steinbrech DS & Longater MT (2000) Expression of adenovirally delivered gene products in healing osseous tissues. Ann Plast Surg 44(5): 522–528.

Stevens DA & Williams GR (1999) Hormone regulation of chondrocyte differentiation and endochondral bone formation. Mol Cell Endocrinol 151: 195–204.

Street J, Bao M, deGuzman L, Bunting S, Peale FV, Ferrara N, Steinmetz H, Hoeffel J, Cleland JL, Daugherty A, Bruggen van N, Redmond HP, Carano RA & Filvaroff EH (2002) Vascular endothelial growth factor stimulates bone repair by promoting angiogenesis and bone turnover. Proc Natl Acad Sci USA 99: 9656–9661.

Suda T, Takahashi N & Martin TJ (1992) Modulation of osteoclast differentiation. Endocr Rev 13: 66–80.

Suda T, Takahashi N, Udagawa N, Jimi E, Gillespie MT & Martin TJ (1999) Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. Endocr Rev 20: 345–357.

99

Sudhakar A, Sugimoto H, Yang C, Lively J, Zeisberg M & Kalluri R (2003) Human tumstatin and human endostatin exhibit distinct antiangiogenic activities mediated by alpha v beta 3 and alpha 5 beta 1 integrins. Proc Natl Acad Sci USA 100(8): 4766–4771.

Sun XD, Jeng L, Bolliet C, Olsen BR & Spector M (2009) Non-viral endostatin plasmid transfection of mesenchymal stem cells via collagen scaffolds. Biomaterials 30(6): 1222–1231.

Taddei L, Chiarugi P, Brogelli L, Cirri P, Magnelli L, Raugei G, Ziche M, Granger H J, Chiarugi V & Ramponi G (1999) Inhibitory effect of full-length human endostatin on in vitro angiogenesis Biochem. Biophys Res Commun 263: 340–345.

Takahashi N, Yamana H, Yoshiki S, Roodman GD, Mundy GR, Jones SJ, Boyde A & Suda T (1988) Osteoclast-like cell formation and its regulation by osteotropic hormones in mouse bone marrow cultures. Endocrinology 122: 1373–1382.

Takahashi N, Udagawa N, Takami M & Suda T (2002) Cells of bone. Osteoclast generation. In Bilezikian JP, Raisz LG & Rodan GA (eds) Principles of bone biology. San Diego CA, Academic Press: 109–126.

Tanaka Y, Abe M, Hiasa M, Oda A, Amou H, Nakano A, Takeuchi K, Kitazoe K, Kido S, Inoue D, Moriyama K, Hashimoto T, Ozaki S, Matsumoto T (2007) Myeloma cell-osteoclast interaction enhances angiogenesis together with bone resorption: a role for vascular endothelial cell growth factor and osteopontin. Clin Cancer Res. 13: 816–823.

Tamamura Y, Otani T, Kanatani N, Koyama E, Kitagaki J, Komori T, Yamada Y, Costantini F, Wakisaka S, Pacifici M, Iwamoto M & Enomoto-Iwamoto M (2005) Developmental regulation of Wnt/beta-catenin signals is required for growth plate assembly, cartilage integrity, and endochondral ossification. J Biol Chem 280: 19185–19195.

Tamayose K, Hira, Y & Shimada T (1996) A new strategy for large-scale preparation of high-titer recombinant adeno-associated virus vectors by using packaging cell line. Hum Gene Ther 7: 507–513.

Teitelbaum SL (2007) Osteoclasts: what do they do and how do they do it? Am J Pathol 170(2): 427–435.

Timpl R (1996) Macromolecular organization of basement membranes. Curr Opin Cell Biol 8(5): 618–624.

Tischer E, Mitchell R, Hartman T, Silva M, Gospodarowicz D, Fiddes JC & Abraham JA (1991) The human gene for vascular endothelial growth factor. Multiple protein forms are encoded through alternative exon splicing. J Biol Chem 266: 11947–11954.

Tombran-Tink J & Barnstable CJ (2004) Osteoblasts and osteoclasts express PEDF, VEGF-A isoforms, and VEGF receptors: possible mediators of angiogenesis and matrix remodeling in the bone. Biochem Biophys Res Commun 316(2): 573–579.

Trueta J & Amato VP (1960) The vascular contribution to osteogenesis. III. Changes in the growth cartilage caused by experimentally induced ischaemia. J Bone Joint Surg Br 42(B): 571–587.

100

Trueta J & Buhr AJ (1963) The vascular contribution to osteogenesis V. The vasculature supplying the epiphysial cartilage inrachitic rats. J Bone Joint Surg Br 45: 572–581.

Trueta J & Trias A (1961) The vascular contribution to osteogenesis. IV. The effect of pressure upon the epiphysial cartilage of the rabbit. J Bone Joint Surg Br 43(B): 800–813.

Tuukkanen J, Koivukangas A, Jämsä T, Sundquist K, Mackay CA & Marks SC Jr (2000) Mineral density and bone strength are dissociated in long bones of rat osteopetrotic mutations. J Bone Miner Res 15(10): 1905–1911.

Uchida C & Haas TL (2009) Evolving strategies in manipulating VEGF/VEGFR signaling for the promotion of angiogenesis in ischemic muscle. Curr Pharm Des 15(4):411–421.

Uchida S, Sakai A, Kudo H, Otomo H, Watanuki M, Tanaka M, Nagashima M & Nakamura T (2003) Vascular endothelial growth factor is expressed along with its receptors during the healing process of bone and bone marrow after drill-hole injury in rats. Bone 32: 491–501.

Uusitalo H, Rantakokko J, Ahonen M, Jamsa T, Tuukkanen J, Kahari VM, Vuorio E & Aro HT (2001) A metaphyseal defect model of the femur for studies of murine bone healing. Bone 28: 423–429.

Väänänen HK (2005) Mechanism of osteoclast mediated bone resorption- rationale for the design of new therapeutics. Adv Drug Deliv Rev 57(7): 959–971.

Väänänen HK & Zhao H (2002) Osteoclast function: Biology and mechanisms. In Bilezikian JP, Raisz LG, & Rodan GA (eds) Principles of bone biology. 2 ed. San Diego CA, Academic Press: 127–140.

Väänänen HK & Horton M (1995) The osteoclast clear zone is a specialized cell-extracellular matrix adhesion structure. J Cell Sci 108(8): 2729–2732.

Väänänen HK, Karhukorpi EK, Sundquist K, Wallmark B, Roininen I, Hentunen T, Tuukkanen J & Lakkakorpi P (1990) Evidence for the presence of a proton pump of the vacuolar H(+)-ATPase type in the ruffled borders of osteoclasts. J Cell Biol 111(3): 1305–1311.

Wang DS, Miura M, Demura H & Sato K (1997) Anabolic effects of 1,25-dihydroxyvitamin D3 on osteoblasts are enhanced by vascular endothelial growth factor produced by osteoblasts and by growth factors produced by endothelial cells. Endocrinology 138(7): 2953–2962.

Wen W, Moses MA, Wiederschain D, Arbiser JL & Folkman J (1999) The generation of endostatin is mediated by elastase. Cancer Res 59(24): 6052–6056.

Wickström SA, Alitalo K, Keski-Oja J (2002) Endostatin associates with integrin alpha5beta1 and caveolin-1, and activates Src via a tyrosyl phosphatase-dependent pathway in human endothelial cells. Cancer Res 62: 5580–5589.

Wickström SA, Alitalo K & Keski-Oja J (2003) Endostatin associates with lipid rafts and induces reorganization of the actin cytoskeleton via down-regulation of RhoA activity. J Biol Chem 278(39): 37895–37901.

Wickström SA, Alitalo K & Keski-Oja J (2005) Endostatin Signaling and Regulation of Endothelial Cell–Matrix Interactions. Adv Cancer Res 94: 197–229.

101

Wickström SA, Veikkola T, Rehn M, Pihlajaniemi T, Alitalo K & Keski-Oja J (2001) Endostatin-induced modulation of plasminogen activation with concomitant loss of focal adhesions and actin stress fibers in cultured human endothelial cells. Cancer Res 61: 6511–6516.

Wodarz A & Nusse R (1998) Mechanisms of Wnt signaling in development. Annu Rev Cell Dev Biol 14: 59–88.

Wong BR, Rho J, Arron J, Robinson E, Orlinick J, Chao M, Kalachikov S, Cayani E, Bartlett FS 3rd, Frankel WN, Lee SY & Choi Y (1997) TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells. J Biol Chem 272(40): 25190–25194.

Wu HM, Huang Q, Yuan Y & Granger HJ (1996) VEGF induces NO-dependent hyperpermeability in coronary venules. Am J Physiol 271: 2735–2739.

Xia L, Kilb J, Wex H, Li Z, Lipyansky A, Breuil V, Stein L,. Palmer JT, Dempster DW & Bromme D (1999) Localization of rat cathepsin K in osteoclasts and resorption pits: inhibition of bone resorption and cathepsin K-activity by peptidyl vinyl sulfones. Biol Chem 380: 679–687.

Xiao Z, Zhang S, Mahlios J, Zhou G, Magenheimer BS, Guo D,Dallas SL, Maser R, Calvet JP, Bonewald L & Quarles LD (2006) Cilia-like structures and polycystin-1 in osteoblasts/osteocytes and associated abnormalities in skeletogenesis and Runx2 expression. J Biol Chem 281: 30884–30895.

Yamaguchi A & Kahn AJ (1991) Clonal osteogenic cell lines express myogenic and adipocytic developmental potential. Calcified Tissue Int 49: 221–225.

Yamaguchi N, Anand-Apte B, Lee M, Sasaki T, Fukai N, Shapiro R, Que I, Lowik C, Timpl R & Olsen BR (1999) Endostatin inhibits VEGF-induced endothelial cell migration and tumor growth independently of zinc binding. EMBO J 18: 4414–4423.

Yancopoulos GD, Davis S, Gale NW, Rudge JS, Wiegand SJ & Holash J (2000) Vascular-specific growth factors and blood vessel formation. Nature 407(6801): 242–248.

Yang Y (2008) Skeletal Morphogenesis and Embryonic Development. In Rosen V (ed) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. 7th ed. American Society for Bone and Mineral Research. Washington, USA: 2–10.

Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, Mochizuki S, Tomoyasu A, Yano K, Goto M, Murakami A, Tsuda E, Morinaga T, Higashio K, Udagawa N, Takahashi N & Suda T (1998) Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95(7): 3597–3602.

Yavropoulou MP & Yovos JG (2007) The role of the Wnt signaling pathway in osteoblast commitment and differentiation. Hormones 6(4): 279–294.

Yin T & Li L (2006) The stem cell niche in bone. J Clin Invest 116: 1195–1201. Yin G, Liu P, An p, Li P, Ding I & Planelles V (2002) Endostatin gene transfer inhibits

joint angiogenesis and pannus formation in inflammatory arthritis. Mol Ther 5: 547–554.

102

Yoshida H, Hayashi S, Kunisada T, Ogawa M, Nishikawa S, Okamura H, Sudo T, Shultz LD & Nishikawa S (1990) The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345(6274): 442–444.

Yu Y, Moulton KS, Khan MK, Vineberg S, Boye E, Davis VM, O'Donnell PE, Bischoff J & Milstone DS (2004) E-selectin is required for the antiangiogenic activity of endostatin. Proc Natl Acad Sci USA 101(21): 8005–8010.

Ylä-Herttuala S & Martin JF (2000) Cardiovascular gene therapy. Lancet 355: 213–222. Zelzer E, Glotzer DJ, Hartmann C, Thomas D, Fukai N, Soker S & Olsen BR (2001)

Tissue specific regulation of VEGF expression during bone development requires Cbfa1/Runx2. Mech Dev 106(1–2): 97–106.

Zelzer E, McLean W, Ng YS, Fukai N, Reginato AM, Lovejoy S, D’Amore PA & Olsen BR (2002) Skeletal defects in VEGF(120/120) mice reveal multiple roles for VEGF in skeletogenesis. Development 129(8): 1893–1904.

Zelzer E & Olsen BR (2005) Multiple roles of vascular endothelial growth factor (VEGF) in skeletal development, growth, and repair. Curr Top Dev Biol 65: 169–187.

103

Original publications

I Tarkka T, Sipola A, Jämsä T, Soini Y, Ylä-Herttuala S, Tuukkanen J & Hautala T (2003) Adenoviral VEGF-A gene transfer induces angiogenesis and promotes bone formation in healing osseous tissue. J Gene Medicine 5: 560–566.

II Sipola A, Nelo K, Hautala T, Ilvesaro J & Tuukkanen J (2006) Endostatin inhibits VEGF-induced osteoclastic bone resorption in vitro. BMC Musculoskelet Disord 13(7): 56.

III Sipola A, Ilvesaro J, Birr E, Jalovaara P, Petterson RF, Stenbäck F, Ylä-Herttuala S, Hautala T & Tuukkanen J (2007) Endostatin inhibits endochondral ossification. J Gene Medicine: 1057–1064

IV Sipola A, Seppinen L, Pihlajaniemi T & Tuukkanen J (2009) Endostatin affects osteoblast behaviour in vitro, but collagenXVIII/endostatin is not essential for skeletal development in vivo. Calcified Tissue International 85(5): 412–420.

Reprinted with permissions from BioMed Central (II), John Wiley & Sons (I, III)

and Springer Science + Business Media (IV).

Original publications are not included in the electronic version of the dissertation.

104

A C T A U N I V E R S I T A T I S O U L U E N S I S

Distributed byOULU UNIVERSITY LIBRARY

P.O. Box 7500, FI-90014University of Oulu, Finland

Book orders:OULU UNIVERSITY PRESSP.O. Box 8200, FI-90014University of Oulu, Finland

S E R I E S D M E D I C A

1016. Hietaniemi, Mirella (2009) Studies on novel and traditional risk factors ofatherosclerosis

1017. Kyllönen, Heli (2009) Gelatinases, their tissue inhibitors and p53 in lymphomas

1018. Kämäräinen, Olli-Pekka (2009) The search for susceptibility genes inosteoarthritis

1019. Karppinen, Sanna-Maria (2009) The role of BACH1, BARD1 and TOPBP1 genes infamilial breast cancer

1020. Niinimäki, Jaakko (2009) Diffusion-weighted MRI and delayed contrastenhancement of degenerated intervertebral disc

1021. Habjanic, Ana (2009) Quality of institutional elderly care in Slovenia

1022. Hugg, Timo (2009) Exposure to environmental tobacco smoke, animals andpollen grains as determinants of atopic diseases and respiratory infections

1023. Sormunen, Erja (2009) Repetitive work in the cold. Work ability, musculoskeletalsymptoms and thermal and neuromuscular responses in food industry workers

1024. Saarnio, Reetta (2009) Fyysisten rajoitteiden käyttö vanhusten laitoshoidossa

1025. Lampela, Pekka (2009) Keuhkoahtaumataudin sairaalahoitoperusterveydenhuollossa ja erikoissairaanhoidossa

1026. Kokkonen, Salla-Maarit (2009) MRI of intracranial tumours in adults. Oedema-attenuated inversion recovery MR sequence in low-field MRI, diffusion-weightedMRI and BOLD fMRI

1027. Matinolli, Maarit (2009) Balance, mobility and falls in Parkinson’s disease

1028. Raunio, Taina (2009) Gene polymorphism and systemic inflammatory response inchronic periodontitis

1029. Pohjola, Vesa (2009) Dental fear among adults in Finland

1030. Kellokoski, Eija (2009) Ghrelin and atherosclerosis. Human, experimental animaland cell culture studies

1031. Kortesluoma, Riitta-Liisa (2009) Hospitalized children as social actors in theassessment and management of their pain

1032. Niinimäki, Maarit (2009) Medical compared with surgical management in inducedabortions and miscarriages

1033. Yan, Ying (2009) The antichlamydial effects of drugs used in cardiovasculardiseases

ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

S E R I E S E D I T O R S

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

PUBLICATIONS EDITOR

Professor Mikko Siponen

University Lecturer Elise Kärkkäinen

Professor Hannu Heusala

Professor Helvi Kyngäs

Senior Researcher Eila Estola

Information officer Tiina Pistokoski

University Lecturer Seppo Eriksson

University Lecturer Seppo Eriksson

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-9317-7 (Paperback)ISBN 978-951-42-9318-4 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1034

ACTA

Annina Sipola

OULU 2009

D 1034

Annina Sipola

EFFECTS OF VASCULAR ENDOTHELIAL GROWTH FACTOR (VEGF-A) AND ENDOSTATIN ON BONE

FACULTY OF MEDICINE,INSTITUTE OF BIOMEDICINE,DEPARTMENT OF ANATOMY AND CELL BIOLOGY,UNIVERSITY OF OULU