ATTACHMENT, POLARITY AND COMMUNICATION CHARACTERISTICS OF BONE CELLS

JO ANNA ILVESA RO
Department of Anatomy and Cell Biology, University of Oulu Biocenter Oulu, University of Oulu

OULU 2001

JOANNA ILVESARO

ATTACHMENT, POLARITY AND COMMUNICATION CHARACTERISTICS OF BONE CELLS

Academic Dissertation to be presented with the assent of the Faculty of Medicine, University of Oulu, for public discussion in the Auditorium L101 of the Department of Medical Biochemistry, on April 20th, 2001, at 12 noon.

O U L U N YL I O PI STO, O U L U 2 0 0 1

Copyright 2001 University of Oulu, 2001

Manuscript received 20 March 2001 Manuscript accepted 26 March 2001

Communicated by Docent Antero Salminen Professor Karl kerman

ISBN 951-42-5935-1

(URL: http://herkules.oulu.fi/isbn9514259351/)

ALSO AVAILABLE IN PRINTED FORMAT ISBN 951-42-5934-3 ISSN 0355-3221 (URL: http://herkules.oulu.fi/issn03553221/) OULU UNIVERSITY PRESS OULU 2001

Ilvesaro, Joanna, Attachment, polarity and communication characteristics of bone cells

Department of Anatomy and Cell Biology, Biocenter Oulu, University of Oulu, P.O.B. 5000, FIN-90014 University of Oulu, Finland 2001 Oulu, Finland (Manuscript received 20 March 2001)

Abstract
Bone resorbing osteoclasts require tight attachment of their plasma membrane to the bone surface in order to retain the specific microenvironment and thus to be able to dissolve the bone matrix underneath. Cadherins are transmembrane glycoproteins usually mediating homophilic calciumdependent cell-cell adhesion. In the present work we have studied the effects of the cadherin CAR sequence HAV-containing hexapeptide AHAVSE on osteoclasts. The primary attachment of osteoclasts to bone surface is not affected by the peptide, suggesting that it is not mediated by cadherins. Treatment of osteoclast cultures with AHAVSE decreased the number of resorption pits and the total resorbed area. Furthermore, we show rapid inactivation of osteoclasts with AHAVSE, which is seen as a decrease in the percentage of osteoclasts with actin rings. Pan-cadherin antibodies localized cadherin-like molecule in the sealing zone area of osteoclasts. These results suggest that cadherin-like molecules may mediate the tight attachment of osteoclasts in the sealing zone area and that the decrease of resorption in AHAVSE-treated osteoclast cultures is due to prevention of sealing zone formation. We studied the polarity of mesenchymal osteoblasts using osteosarcoma cell line UMR-108 and endosteal osteoblasts in situ in bone tissue cultures. Immunofluorescence confocal microscopy revealed that the vesicular stomatitis virus glycoprotein (VSV G) was targeted to the culture mediumfacing surface. In endosteal osteoblasts, VSV G protein was found in the surface facing the bone marrow and circulation. On the contrary, Influenza virus hemagglutinin (HA) was localized to the bone substrate-facing surface of the UMR-108 cells. Electron microscopy showed that VSV particles were budding from the culture medium-facing surface, whereas Influenza viruses budded from the bone substrate-facing plasma membrane. These findings suggest the bone attaching plasma membrane of osteoblasts is apical, and the circulation or bone marrow facing plasma membrane is basolateral in nature. Gap junctions often mediate communication between different cells and cell types. In the present work, we demonstrate that rat osteoclasts show connexin-43 staining localizing in the plasma membrane of the cells in cell-cell contacts and over the basolateral membrane of osteoclasts. The effects of heptanol and Gap 27, known gap- junctional inhibitors, were studied using the well-characterized pit formation assay. The inhibitors decreased the number and activity of osteoclasts, suggesting a defect in the fusion of mononuclear osteoclast precursors to multinucleated mature osteoclasts. Furthermore, the total resorbed area and the number of resorption pits also decreased in the cultures. These results suggest that gap-junctional connexin-43 plays a functional role in osteoclasts, and that the blocking of gap junctions decreases both the number and the activity of osteoclasts.

Keywords: cell adhesion, osteoclasts, osteoblasts, cell polarity, cell communication

To my husband Vesa

Acknowledgments

This work was carried out at the Department of Anatomy and Cell Biology and Biocenter Oulu, University of Oulu, during the years 1995-2001. It is my great pleasure to thank the following persons, who have taken part of this work and thus made it possible: I owe my deepest thanks to my mentor Professor Juha Tuukkanen, DDS, Ph.D. His encouragement, support, sense of humour and enthusiastic attitude towards research and life in general have been inspiring and guiding me during these years and have been more than I could have ever asked for. I thank my first teacher and supervisor Professor H. Kalervo Vnnen, M.D., Ph.D., for introducing me to the research world and especially to the world of bone. I wish to express my appreciation to Professor Hannu Rajaniemi, M.D., Ph.D., for kindly providing me the surroundings, thus making it possible for me to work in the Department of Anatomy and Cell Biology. I also thank Professor Juha Peltonen, M.D., Ph.D., for his interest towards my research. I am grateful to Professor Karl-Erik kerman, M.D., Ph.D., for his critical appraisal of this thesis. Also, I am in debt to Docent Antero Salminen, Ph.D. for carefully reviewing the thesis and for teaching me during my first years in the University of Jyvskyl and for once sending me to Oulu. The expertise of my co-authors, Doctors Pivi Lakkakorpi, Kalervo Metsikk and Pasi Tavi is most sincerely acknowledged. My humble thanks are due to the whole staff of Department of Anatomy and Cell Biology, and, in particularly the present and former Bone Heads for the stimulating working atmosphere. Especially Anita Kapanen, Jussi Halleen and Teuvo Hentunen are warmly acknowledged for fun times in and outside laboratory. I also wish to thank Sari Annunen, Allan Haimakainen, Minna Orrevetelinen, Eero Oja and Marja Vlimki for their skilful technical assistance. I owe my loving thanks to my best girlfriend and colleague Anne Tuomisto for the special bond we two share - there is nobody quite like her. I want to express my heartfelt gratitude to my parents for always encouraging me to aim as high as I possibly wanted and for always telling me that if I wanted to achieve something, I could do it, if I just put my heart into it. My husband Vesa, deserves my warmest thanks. I thank him for his love

and his never-ending optimism and encouragement during hard times. Vesas 24-hour ITsupport is also warmly acknowledged. I am priviliged to have him as my husband. This research was financially supported by Biocenter Oulu, Oulu University Scholarship Foundation and the Finnish Cultural Foundation.

Oulu, March 2001

Joanna Ilvesaro

Abbreviations
ALP BL BMU BRU BSA CAR CA-II CT Cx-43 DAB DMEM DMSO F-actin FBS FCS FITC FSD HAV Hepes KDa M-CSF M NCP OCIF ODF OPG OPGL PBS PFA PTH Alkaline Phosphatase Basolateral Bone multicellular unit Bone remodeling unit Bovine serum albumin Cell adhesion recognition Carbonic anhydrase II Calcitonin Connexin-43 3,3 diaminobenzidine tetrahydrochloride Dulbeccos modified Eagles medium Dimethyl sulfoxide Filamentous actin Fetal bovine serum Fetal calf serum Fluorescein isothiocyanate Functional secretory domain Histidine-alanine-valine N-(2-hydroxyethyl)piperazine-N-(2-ethanesulphonic acid) Kilodalton Macrophage colony-stimulating factor -6 Micromolar, 10 g/mol Non-collagen protein Osteoclastogenesis inhibitory factor Osteoclast differentiation factor Osteoprotegerin Osteoprotegerin ligand Phosphate-buffered saline Paraformaldehyde Parathyroid hormone

RANK RANKL RB RGD SZ TRANCE TRANCER TRAP TRITC SDS SDS-PAGE TEM VNR VSV WGA

Receptor activator for NF-B Receptor activator for NF-B ligand Ruffled border Arginine-glycine-aspartic acid Sealing zone TNF-related activator-induced cytokine TNF-related activator-induced cytokine receptor Tartrate-resistant acid phosphatase Tetramethylrhodamine isothiocyanate Sodium dodecyl sulfate SDS-polyacrylamide gel electrophoresis Transmission electron microscopy Vitronectin receptor Vesicular stomatitis virus Wheat germ agglutin

List of original publications

This thesis is based on the following articles, which are referred to in the text by their Roman numerals: I Ilvesaro JM, Lakkakorpi PT, Vnnen HK (1998) Inhibition of Bone Resorption by Cadherin Cell Adhesion Recognition Sequence Containing Peptide AHAVSE Is Due to Prevention of Sealing Zone Formation. Experimental Cell Research 242:75-83. Ilvesaro J, Metsikk K, Vnnen K, Tuukkanen J (1999) Polarity of osteoblasts and osteoblast-like UMR-108 cells. Journal of Bone and Mineral Research 14 (8): 13381344. Ilvesaro J, Vnnen K, Tuukkanen J (2000) Bone-Resorbing Osteoclasts Contain Gap-Junctional Connexin-43. Journal of Bone and Mineral Research 15 (5): 919-926. Ilvesaro J, Tavi P, Tuukkanen J Connexin-Mimetic Peptide Gap 27 Decreases Osteoclastic Activity. Submitted.

II

III IV

Contents
Abstract Dedication Acknowledgements Abbrevations List of original articles 1. Introduction 2.1 Structure and function of bone tissue ..................................................................... 15 2.1.1 Bone matrix components................................................................................. 16 2.1.2 Bone cells........................................................................................................ 17 2.2 Cellular mechanisms of bone resorption ................................................................ 20 2.2.1 Osteoclastic membrane domains ..................................................................... 23 2.2.2 Attachment of osteoclasts to bone matrix........................................................ 24 2.2.3 Cadherins ........................................................................................................ 25 2.2.4 Degradation of the mineral and organic matrix............................................... 27 2.2.5 Regulation of osteoclastic bone resorption ..................................................... 28 2.2.6 Disorders of bone modeling and remodeling .................................................. 29 2.3 Cell polarity ........................................................................................................... 30 2.4 Communication between bone cells....................................................................... 31 2.4.1 Gap junctions .................................................................................................. 31 4.1 Reagents (I-IV) ...................................................................................................... 35 4.1.1 Antibodies (I-IV)............................................................................................. 35 4.2 Cell culture (I-IV) .................................................................................................. 36 4.2.1 Isolation and culture of osteoclasts (I, III-IV)................................................. 36 4.2.2 Cell line (II)..................................................................................................... 36 4.2.3 Tissue culture (II) ............................................................................................ 36 4.3 Fluorescence and microscope techniques (I-IV) .................................................... 37 4.3.1 Light microscopy (I, III-IV) ............................................................................ 37 4.3.2 Conventional fluorescence microscopy (I, III-IV) .......................................... 37 4.3.3 Confocal laser scanning microscopy (I-IV) .................................................... 38 4.3.4 Immunoelectron microscopy (II) .................................................................... 38 4.3.5 Digital image analysis (I -IV).......................................................................... 38

4.4 Viral infections (II)................................................................................................. 39 4.4.1 Metabolic labeling (II) .................................................................................... 39 4.4.2 Cell surface biotinylation (II) .......................................................................... 39 4.5 Statistical analysis (I, III-IV).................................................................................. 39 5.1 Cadherin localization in osteoclasts (I) .................................................................. 40 5.2 Effects of HAV-containing peptide (I).................................................................... 40 5.2.1 Primary attachment of osteoclasts to bone surface.......................................... 40 5.2.2 Osteoclastic bone resorption ........................................................................... 41 5.2.3 Activity of osteoclasts ..................................................................................... 41 5.3 Polarity of osteoblasts (II)...................................................................................... 41 5.3.1 Immunofluorescence data ............................................................................... 41 5.3.2 Electron microscopy data ................................................................................ 42 5.3.3 Biochemical data............................................................................................. 42 5.4 Gap junctions in osteoclasts (III and IV)................................................................ 43 5.4.1 Connexin-43 localization in osteoclasts (III) .................................................. 43 5.4.2 Effects of gap-junctional inhibitors heptanol and Gap 27 ............................... 43 6.1 Methodological aspects.......................................................................................... 45 6.2 Cadherins and attachment of osteoclasts to bone ................................................... 46 6.3 Polarity of osteoblasts ............................................................................................ 48 6.4 Gap junctions in osteoclasts ................................................................................... 49

1 Introduction

Bone tissue has three major functions in the body: it offers mechanical support to the body, it protects major vital organs and it maintains calcium and phosphate homeostasis. Bone also provides the environment for hematopoiesis, which takes place in the marrow of long bones in adults. Bone is also a dynamic tissue, which remodels and repairs itself throughout life. In normal bone tissue the formation and resorption of bone are well balanced so that when old bone is resorbed by osteoclasts, similar amount of new bone is formed by osteoblasts. Excessive bone resorption results in net bone loss, whereas decreased bone resorption, or excess bone formation, results in increased bone mass. Osteoporosis is a metabolic bone disease leading to decreased bone mass and increased susceptibility to fractures. At the cellular level, osteoporosis is characterized by extensive bone resorption, which leads to disturbance in the bone microarchitecture, which in turn increases the probability of fracturing. Osteoporosis typically affects women after menopause, when serum estrogen levels decrease. The clinical complications of osteoporosis, fractures occur most often in the spine, femoral neck and at the wrist. The two cell types primarily involved in bone turnover are the bone-forming osteoblasts and bone-resorbing osteoclasts. Solving of the cellular and molecular mechanisms of bone formation and resorption is important because of the increasing rate of osteoporosis in the aging population. The experimental part of this thesis work was carried out to clarify some cell biological questions concerning different bone cells. Special interested was focused on the attachment of osteoclasts to bone surface, the polarization of the osteoblasts and the communication between osteoclasts and osteoblasts. All of these are interesting biological questions, but they might also appear to be useful targets for therapy of metabolic skeletal disorders.

2 Review of the literature

2.1 Structure and function of bone tissue
Bone consists of mineralized organic matrix and bone cells, including bone-forming osteoblasts, bone-lining cells, osteocytes and bone-resorbing osteoclasts. There are two morphologically distinct types of bone, cortical (compact) bone, which constitutes 85% of total bone in the body and trabecular (cancellous) bone that constitutes the rest. The core of each bone consists of cortex, a compact bone layer that forms the body of the bone, which is mainly responsible for the mechanical and protective function. Trabecular bone, a sponge-like structure of bone trabeculi, is found inside the cortex and it is responsible for the metabolic function of bone. The remaining volume, the spaces between the bone trabeculi in long bones, is composed of blood vessels and bone marrow, which is the site of erythrocyte synthesis. An extensive blood circulatory system provides bone nutrients, hormones and growth factors. Bone growth in embryonic development can occur in two mechanisms, either with intramembranous ossification or endochondral ossification. The flat bones of the body, such as bones of the skull, form by intramembranous ossification. This means that during embryonic development mesenchymal stem cells of connective tissue differentiate straight into bone forming osteoblasts and start to synthesize bone. On the other hand, long bones at joints, such as the long bones of legs and arms, which bear the weight of body, form via endochondral ossification. Mesenchymal stem cells differentiate into chondroblasts, which first form a cartilage model of the skeleton. Later on, the cartilage model is replaced by bone matrix with the help of osteoblasts and osteoclasts working synchronically. During the growth period (first 20-30 years of life), more bone is formed than is resorbed, resulting in an increase in bone mass. Peak bone mass, meaning the maximal amount of bone matrix an individual has attained, is reached by late adolescence (Matkovic et al. 1994, Mundy 1995). During the next three decades (and beyond) the adult skeleton is maintained by removing and replacing a fraction each year. This remodeling begins with a localized resorption that is succeeded by a precisely equal formation of bone at the same site (Parfitt 1994). Sometime after the fifth decade, the

16

formative phase of the remodeling sequence fails to keep the pace with resorptive activity and skeletal mass, including the connectivity of trabecular bone, decreases. This reduces skeletal strength and increases the risk of fracture over time, depending on the magnitude by which resorption and formation are uncoupled.

2.1.1 Bone matrix components

Bone matrix consists of an organic matrix that is strengthened by deposits of inorganic calcium salts. 70% of bone is mineral, 22% protein and 8% water (Lane 1979). Collagen, the primary secretory product of osteoblasts, comprises about 90% of the organic bone matrix. Type I collagen is the most abundant form of collagen found in the body and is widely distributed in connective tissue. Type I collagen represents the majority of bone collagen, but biochemical analysis have also indicated trace amounts of types III, V, XI and XIII (Gehron 1989). The bone collagen fibrils are highly insoluble because of their many covalent intra- and intermolecular cross-links, the type and pattern of which differs from that in soft connective tissues. In the matrix, individual collagen molecules are packed end to end with a short space (gap) between them. The molecules are packed laterally in a one-quarter stagger array so that each molecule is offset from its neighbor by approximately one-fourth of its length. This three-dimensional arrangement constitutes the fiber structure found in the bone extracellular space. Multiple crosslinking sites combine extracellularly to form pyridinium ring structures tying several collagen monomer molecules together within the formed collagen fiber, thereby rendering it completely insoluble (Eyre et al. 1988). These pyridinium cross-links are only released on degradation of the mineralized collagen fibrils during bone resorption, making it an excellent method to measure bone resorption rate from urine (Uebelhart et al. 1990). The remaining 10% of the organic matrix is composed of non-collagenous proteins (NCPs), which can be broken down into four general groups of proteins: 1) cell attachment proteins 2) proteoglycans 3) -carboxylated (gla) and 4) growth-related proteins. All connective tissue cells interact with their extracellular environment in response to chemical stimuli that direct and coordinate specific cell functions, such as proliferation, migration and differentiation. These actions require cell attachment and spreading by focal adhesions via integrins, that transduce signals to the cytoskeleton (Ruoslahti et al. 1987). Bone cells synthesize four proteins that affect cell attachment: fibronectin, thrombospondin, osteopontin and bone sialoprotein (Oldberg et al. 1986, Robey et al. 1989, Somerman et al. 1988). Of these, bone sialoprotein is exclusively found in bone, whereas the others can also be found in many tissues other than bone. Proteoglycans are macromolecules that contain acidic polysaccharide side chains (glycosaminoglycans) attached to central core protein. In bone, two types of glycosaminoglycan are found: chondroitin sulfate and heparin sulfate. -Carboxylation occurs in two bone NCPs, osteocalcin (bone gla-protein) and matrix gla-protein (Glowacki et al. 1987, Robey et al. 1989). Osteocalcin expression is restricted to bone, but matrix gla-protein can also be found in cartilage in addition to bone. Measurements of

17

osteocalcin in serum have proved to be valuable as markers of bone turnover in metabolic diseases (Price et al. 1980). A number of proteins in bone appear to be associated with the life cycle and function of osteoblasts (Ripamonti et al. 1992). These proteins may be bone morphogenetic factors or growth factors, such as TGF- and insulin-like growth factors, osteoblast secretion products that can stimulate osteoblast cell growth in an autocrine or paracrine fashion (Canalis et al. 1988, Hauschka et al. 1986, Hauschka et al. 1988, Robey et al. 1987). Other bone cell products may be associated with the growth or differentiation of osteoblasts in an indirect or as yet unidentified fashion. The most abundant NCP produced by osteoblasts is osteonectin, a phosphorylated glycoprotein accounting 2% of the total protein of developing bone in most animal species. 2+ Osteonectin has high affinity for binding Ca and hydroxyapatite and it also binds to collagen (Termine et al. 1981) and thrombospondin (Clezardin et al. 1988). The importance of the role of bone tissue as an ion reservoir can be appreciated from the fact that about 99% of the body calcium, about 85% phosphorous, and from 40-60% of the body sodium and magnesium are associated with the bone crystals. These consequently serve as the major source for the transport of these ions to and from the extracellular fluids. Rapid responses to calcium excesses and deficiencies in the body are crucial for a number of processes, including the regulation of muscle contraction, nerve impulse transmission, cell membrane permeability and blood coagulation. Indeed, the most abundant inorganic components of the bone matrix are phosphate and carbonate salts of calcium, with traces of calcium and magnesium fluorides (Levy 1894). The mineral is in the form of hydroxyapatite, a crystalline lattice in which the principal components are calcium and phosphate ions (DeJong 1926). With the mineral phase, the otherwise soft, pliable organic matrix now becomes relatively hard, rigid material which possesses the necessary mechanical properties that permit it to withstand the forces, stresses, and strains. The composite arrangement of bone tissue yields a material that is very strong for its weight and much stronger than either component would be alone.

2.1.2 Bone cells 2.1.2.1 Osteoblastic cells

Bone formation is the responsibility of osteoblasts, cells of mesenchymal origin lining the new bone matrix. Osteoblasts arise from the same pluripotent stem cell with chondroblasts, adipocytes, myoblasts and fibroblasts (Bennett et al. 1991, Grigoriadis et al. 1988, Yamaguchi et al. 1991). Osteoblasts produce and secrete the major part of the organic bone matrix, which in a well-regulated process becomes calcified to form mineralized bone. Four maturational stages of osteoblast development can be distinguished in situ the preosteoblast, osteoblast, osteocyte and bone-lining cell (Aubin et al. 1996). Preosteoblast is considered the immediate precursor of the osteoblast and is identified in part by its localization in the adjacent of one or two cell layers distant from the osteoblasts lining the bone formation surfaces. Osteoblasts are postproliferative,

18

cuboidal, strongly alkaline phosphatase-positive cells lining the bone matrix at sites of active bone production. Osteoblasts can also be recognized by their ability to synthesize a number of phenotype-specific or -associated macromolecules, including the bone matrix proteins, such as osteocalcin, bone sialoprotein and osteopontin, proteoglycans, certain hormone receptors, and cytokines and growth factors. The precise physiological role of alkaline phosphatase is unknown, but it has been hypothesized to be involved in the mineralization process (Wennberg et al. 2000). After the completion of bone formation a small portion (10-20%) of osteoblasts incorporate themselves within the newly formed extracellular matrix; these cells trapped inside the matrix, which are considered the most mature form of osteoblastic lineage are osteocytes. Osteocytes are smaller than osteoblasts and they have lost many of their cytoplasmic organelles. Osteocyte cell bodies are found in lacunae inside the bone matrix and they have multiple long cytoplasmic processes extending through canaliculi by which they communicate with each other and possibly with other surface bone cells, such as osteoblasts and maybe even osteoclasts (Bonewald 1999, Doty 1981, Menton et al. 1984). The canaliculi are small tunnels in the bone, measuring about 1-2 m in diameter. Although the functions of this most abundant cell type of bone, osteocyte, are poorly understood, some evidence is gathering that osteocytes sense the mechanical stress to bone and therefore can participate in the bone remodeling process (Burger et al. 1999, Mikuni-Takagaki 1999). In the adult skeleton, the majority of bone surfaces that are not being remodeled (i.e. undergoing resorption or formation), are covered by flat, thin, elongated bone lining cells, which are thought to represent the inactive form of osteoblast in terms of matrix production. Bone lining cells detach from the surface of bone when osteoclasts start the resorbing of bone.

2.1.2.2 Osteoclasts
Osteoclasts, first named by Klliker (Klliker 1873), are multinucleated cells differentiating from haemopoietic cells. It is generally accepted that osteoclasts are formed by the fusion of mononuclear precursors derived from haemopoietic stem cells in the bone marrow, rather than incomplete cell divisions (Chambers 1989, Gthling et al. 1976, Kahn et al. 1975, Suda et al. 1992, Walker 1973a, Walker 1975b, Walker 1975a). They share a common stem cell with monocyte-macrophage lineage cells (Ash et al. 1980, Kerby et al. 1992). The differentiation of osteoclast precursors into mature multinucleated osteoclasts requires different factors including hormonal and local stimuli (Athanasou et al. 1988, Feldman et al. 1980, Walker 1975a, Zheng et al. 1991) and living bone and bone cells have been shown to play a critical role in osteoclast development (Hagenaars et al. 1989). Osteoblastic or bone marrow stromal cells are also required for osteoclast differentiation. One of the factors produced by these cells that support osteoclast formation is macrophage colony-stimulating factor, M-CSF (WiktorJedrzejczak et al. 1990, Yoshida et al. 1990). Receptor activator for NF-B ligand (RANKL, also known as TRANCE, ODF and OPGL) is the vital signal (Suda et al. 1992) through which osteoblastic/stromal cells stimulate osteoclast formation and resorption via

19

a receptor RANK (TRANCER) located in osteoclasts (Lacey et al. 1998, Tsuda et al. 1997, Wong et al. 1997a, Wong et al. 1997b, Yasuda et al. 1998a, Yasuda et al. 1998b). Osteoblasts also secrete protein that strongly inhibits osteoclast formation called osteoprotegerin (OPG, also known as OCIF), which acts as a decoy receptor for the RANKL thus inhibiting the true connection between osteoclasts and osteoblasts via RANK and RANKL (Fig.1).
OPG Decoy receptor, soluble TRANCE, RANKL, osteoblasts TRANCER, RANK, osteoclasts STIMULATION INHIBITION
OPG PTH 1,25(OH)2D3 M-CSF PGE2 PGE2, 1,25(OH)2D3

Fig. 1. Regulation of osteoclast formation and resorption by TRANCE/OPG. Osteoblasts express in their plasma membrane TRANCE. Osteoclast precursors in turn have a receptor for TRANCE, called TRANCER, by which osteoblastic or stromal cells can stimulate precursor differentiation into mature osteoclasts. Through this receptor osteoblastic cells can also stimulate mature osteoclastic bone resorption. Osteoblasts also secrete a soluble protein called OPG, which acts as a decoy receptor for TRANCE, thus blocking the stimulating effect of TRANCE to osteoclasts or their precursors.

Osteoclasts are responsible for dissolving both the mineral and organic bone matrix (Blair et al. 1986). Osteoclasts represent terminally differentiated cells expressing a unique polarized morphology with specialized membrane areas and several membrane and cytoplasmic markers, such as tartrate resistant acid phosphatase (TRAP) (Anderson et al. 1979), carbonic anhydrase II (Vnnen et al. 1983), calcitonin receptor (Warshafsky et al. 1985) and vitronectin receptor (Davies et al. 1989). Multinucleated osteoclasts usually contain less than 10 nuclei, but they may contain up to 100 nuclei being between 10 and 100 m in diameter (Gthling et al. 1976). This makes them relatively easy to identify by light microscopy. They are highly vacuolated when in active

20

state, and also contain many mitochondria, indicating a high metabolic rate (Mundy 1990).

2.2 Cellular mechanisms of bone resorption

The process of bone remodeling is the key phenomenon in bone cell biology. Understanding the mechanisms that control the remodeling process and its regulation will clarify not only the local control of the function of bone cells but also the pathophysiology of accelerated bone loss and osteoporosis. In the normal adult skeleton, bone formation occurs only at sites where bone has previously been resorbed. Throughout life old bone is continuously removed by osteoclasts and new bone is formed by osteoblasts in a process called bone remodeling cycle. The adult skeleton is in a dynamic state, being continuously broken down and reformed by the coordinated actions of osteoclasts and osteoblasts on trabecular surfaces and in Haversian systems. The remodeling cycle is a coordinated series of events, where the actions of osteoblasts and osteoclasts are tightly coupled to retain the balance between formation and resorption of bone. This turnover or remodeling of bone occurs in focal and discrete packets throughout the skeleton, the rate of which varies considerably both within and among different bones of the skeleton. The modeling of each packet takes a finite period of time resulting that approximately 10-15% of bone surface is continuously being remodeled. Bone remodeling occurs in focal and particular units, called basic multicellular units (BMU) or bone remodeling units (BRU) throughout the skeleton (Frost 1964, Parfitt 1983). The remodeling of each unit takes about 3-4 months. Bone resorption is recognized as being a multi-step process and the work done by each BRU is divided into four distinct phases: activation, resorption, reversal and formation (Baron 1977, Frost 1973, Frost 1986), as described below. The term activation is used to describe the initiating event that converts a resting bone surface into a remodeling surface. In the normal adult skeleton, a new BRU is activated every 10 s (Parfitt 1983). Activation involves recruitment of osteoclast precursors to the bone and their differentiation and fusion into functional osteoclasts. It also involves the penetration of the bone lining cells and recognition of the bone sites to be resorbed (Tran et al. 1982a, Tran et al. 1982b). The factors that determine why a particular site of bone is chosen for remodeling are not clear. One of the most attractive of the current hypotheses is that osteocytes may be involved in the activation (Chambers 1985, Frost 1973). The non-mineralized osteoid covers the mineralized bone matrix in vivo, preventing its resorption by osteoclasts. Thus, before osteoclasts can begin the actual resorption in vivo, the osteoid must be dissolved or mineralized, after which osteoclasts can attach to the mineralized matrix and initiate bone resorption (Jones et al. 1986). Proteases released by osteoblastic cells have been shown to be responsible for dissolving the osteoid and thereby inducing osteoclastic resorption and determining its location (Chambers et al. 1991). During the resorption phase osteoclasts work in a concerted fashion, removing both mineral and organic components of bone matrix (Blair et al. 1986). The hallmark of the

21

resorbing surface is the appearance of a scalloped erosion, also called an eroded surface or Howships or resorption lacuna. In cancellous bone, osteoclasts resorb saucer-shaped cavities some 50 m in depth towards its center. In cortical bone, the osteoclasts drill through the matrix in a direction that is parallel with the long axis of the bone, creating a tunnel about 2,5 mm long and 200 m in diameter (Eriksen 1986). The resorption phase takes about two-three weeks (Baron et al. 1984). Once the osteoclasts have resorbed most of the mineral and organic matrix, there is a reversal phase, which lasts from 7-14 days and heralds the switchover from destruction to repair. During the reversal phase macrophages may appear on the resorption surface (Raisz 1988). The roles of these macrophages are not known, but it has been suggested that they complete the task of bone resorption. Alternatively, they might produce factors that help initiate osteoblastic bone formation (Raisz 1988). If the fate of osteoclasts at the reversal site is apoptosis, the function of macrophages could be related to destruction of osteoclast corpses. This has been shown to occur with other cells (Arends et al. 1991). It is during the reversal phase that the coupling of bone resorption to formation takes place. The mechanism by which osteoblasts are summoned into the resorption lacuna is uncertain, but it is probable that a number of paracrine factors produced in or around the remodeling site are involved. These coupling factors could be elaborated by cells involved in the activation of resorption (lining cells) or by osteoclasts themselves or some other cell types present in the resorption lacunae. The factors could also be released from the bone matrix during the resorption phase (Mundy et al. 1987). Osteoclasts are highly motile and actively migrating cells, so that after completion of one resorption lacuna, they can move along the bone surface to another site and restart the resorption phase (Kanehisa et al. 1988). Formation of new bone begins after the short reversal phase. Bone formation is a twostage process beginning with the deposition of osteoid, an organic matrix consisting primarily of type I collagen and various other components. In normal adult bone, osteoid is laid down in discrete lamellae about 3 m thick. The second stage in bone formation is mineralization of the organic matrix, which occurs after a delay of about 20 days called the mineralization lag time. Two general theories exist for how calcification is initiated in skeletal tissue, not necessarily excluding one another: the nucleation theory and the matrix vesicle theory (Anderson 1984, Anderson 1989, Williams et al. 1991). The nucleation theory holds that the type I collagen fibril is the major site of crystal nucleation where calcium phosphate crystals are deposited in the hole regions of these fibrils. Originally it was believed that the initiation of bone mineralization is mainly an extracellular, biochemical process where the presence of collagens, certain glycosaminoglycans or lipids could trigger the initiation of calcification (Baylink et al. 1972, Irving 1958, Shuttleworth et al. 1972, Spector et al. 1972, Wuthier 1968, Wuthier 1969). The first line of evidence that cellular activity might be needed for the mineralization of bone arose out of the finding that mitochondria accumulate calcium (Engstrom et al. 1964). Later it was shown that mitochondria of bone-forming cells might produce a local rise in the levels of mineral ions, and also a matrix, which was capable of being mineralized (Shapiro et al. 1969). The mitochondria of columnar and hypertrophic zones of growth plate cells were found out to contain more enzymes associated with calcification (alkaline phosphatase, alkaline pyrophosphatase, lysozyme and lactate dehydrogenase) than did those of the resting zone (Arsenis 1972). The matrix vesicles

22

theory states that bone-forming cells produce small 100-200 nm in diameter organelles that have been observed in mineralizing tissues and cell cultures, often in contact with the initial mineral crystals (Anderson 1984, Anderson 1995, Schwartz et al. 1987). These vesicles have high alkaline phosphatase, alkaline pyrophosphatase and ATPase content, but hardly any acid phosphatase, and are thus not lysosomal origin (Ali et al. 1970). Plausible evidence has been demonstrated to support the matrix vesicle theory, showing that membrane bound vesicles bud from the long processes of osteoblastic cells and that 2+ the mineralization is induced by the release of Ca -containing vesicles (RingbomAnderson et al. 1994). However, later it has been speculated that the two mechanisms for bone mineralization are not alternative but parallel; one is predominant in both calcified cartilage and primitive woven bone, the other in lamellar bone (Gorski 1998, Termine 1993). Calcified cartilage and woven bone seem to mineralize via matrix vesicles (Bonucci 1971), membrane-bound bodies that exocytose from the plasma membrane and migrate to the loose extracellular matrix space. The lipid-rich inner membrane of these vesicles becomes the nidus for hydroxyapatite crystal formation and, eventually, crystallization proceeds to the point of obliteration of the vesicle membrane producing a spherulite of clustered, tiny, crystals. These spherulites conglomerate until a continuous mineralized mass is achieved throughout the matrix space. The driving force for the mineral cascade, once initiated, seems to be the mineral crystals themselves that are first associated with the matrix vesicle membrane. The rate of mineralization in both woven bone and in lamellar bone seems to depend on the presence of inhibitor molecules (e.g. pyrophosphate and acidic NCPs), which in solution seem to regulate the kinetics of the mineralization process (Termine et al. 1980). The cell buds off organelles capable of mineral accumulation and then synthesizes proteins that can control the rate at which crystallization proceeds. The extracellular matrix in non-fetal and more abundant lamellar bone is tightly packed with well-aligned collagen fibrils that are decorated with complexed NCPs (e.g. fibronectin and osteonectin). Either because there simply is insufficient space for them or for developmental reasons, matrix vesicles are rarely, if ever, seen in lamellar bone. Instead, mineralization proceeds in association with the heteropolymeric matrix fibrils themselves. Somewhat more mineral seems to be associated with aligned gap or hole regions of the fibers, which have more room for inorganic ions than rest of the fibril structure. Other loci for the bone mineral appear to be between collagen fibrils in a brick and mortar fashion (Weiner et al. 1986). It is unknown whether the driving force for mineralization is the bone collagen itself or its associated NCPs. There is no doubt that in mature calcified bone collagen and the inorganic phase are strongly bound (Irving 1973). After the mineralization process is triggered, the mineral content of the matrix increases rapidly over the first few days to 75% of the final mineral content (primary mineralization), but it takes from 3 months up to a year for the matrix to reach its maximum mineral content (secondary mineralization) (Amprino et al. 1952). The principal component of the mature mineral phase is hydroxyapatite.

23

2.2.1 Osteoclastic membrane domains

Inactive osteoclasts, although they can still adhere to the bone surface, are not polarized and fail to display specialized membrane domains (Holtrop et al. 1977). Actively resorbing osteoclasts, in turn, are highly polarized cells and they reveal distinct membrane domains and cytoskeletal structures that are not seen in inactive or immature osteoclasts (Baron et al. 1990, Holtrop et al. 1977, Lakkakorpi et al. 1989, Lakkakorpi et al. 1991a, Salo et al. 1996). The membrane areas of active osteoclasts can be separated into the following domains: sealing zone, ruffled border, basolateral membrane and functional secretory domain (Fig. 2). In the sealing zone, also called as clear zone, the cell membrane forms a tight attachment to the bone surface, thereby isolating the resorption lacuna from the extracellular fluid and permitting the maintenance of a specific microenvironment in the lacuna. The sealing zone is seen as an electro-dense area free of organelles (Schenk et al. 1967) and it has a striated appearance in electron microscopy, with alternating dark and light areas orientated perpendicular to the bone surface (Malkani et al. 1973) consisting of bundles of actin filaments (King et al. 1975). Ruffled border (RB) is a highly convoluted plasma membrane domain under which the actual bone resorption takes place. Sealing zone surrounds the RB, thus defining the outlines of the resorption lacuna. Acidic intracellular vesicles fuse to the cell membrane facing the bone matrix, thereby forming RB membrane. RB does not fit to any type of previously described plasma membrane domains, because it has both lysosomal and endosomal features (Akamine et al. 1993, Baron et al. 1988, Palokangas et al. 1997). Bone degradation occurs in the extracellular space between the bone matrix and the RB, which is called the resorption lacuna. In the middle of the basolateral membrane is the fourth membrane domain, the functional secretory domain (FSD), which appears when matrix degradation is started (Salo et al. 1997).

2
BL

3
FSD BL SZ RB SZ

SZ RB

SZ

Fig. 2. Osteoclastic membrane domains. 1) When an osteoclast is not resorbing bone, it shows no signs of polarized membrane domains. 2) Once the osteoclast starts the resorbing, it quickly polarizes its membrane into distinct domains. Ruffled border (RB) is a membrane domain facing the bone surface, where the actual resorption takes place. Sealing zone (SZ) forms a tight contact to the bone, sealing the proteolytic enzymes and acid into the forming resorption lacuna. Basolateral membrane (BL) faces towards the bone marrow. 3) When the osteoclast is actively resorbing bone, a fourth domain arises into the basolateral membrane, the functional secretory domain (FSD), which acts as a route of osteoclasts to exocytose the resorbed material.

24

2.2.2 Attachment of osteoclasts to bone matrix

Bone resorption occurs in an extracellular compartment below the osteoclasts. This structure has been suggested to be a late endosome-type compartment (Palokangas et al. 1997), from which different hydrolytic enzymes, together with acid, are secreted (Baron 1989). In order to retain the specific microenvironment created by the cell organelles in the ruffled border membrane, osteoclasts need to have a tight attachment to bone matrix in the sealing zone. The mechanism by which osteoclasts attach to bone is not fully understood (Dresner-Pollak et al. 1994, Horton et al. 1989, Vnnen et al. 1995). A number of cell surface glycoproteins have been identified as intercellular adhesion molecules, and these have been classified into at least three major molecular families, the immunoglobulin (Ig) superfamily, the integrin superfamily, and the cadherin family. Integrins have been identified as a family of cell surface receptors that recognize extracellular matrices (Mundy et al. 1987). Integrins are adhesion molecules that mediate cell attachment to the substrate by binding to an arginine-glycine-aspartic acid (RGD) consensus sequence in their ligands (Ruoslahti et al. 1986). Osteoclasts express two members of integrin superfamily, the vitronectin receptor v3 (Davies et al. 1989, Horton 1990) and 21 (Horton et al. 1989). The first indications of the importance of vitronectin receptor came from the experiments in which monoclonal antibodies against osteoclasts (Chambers et al. 1986), later identified as anti-vitronectin receptor antibodies (Davies et al. 1989), inhibited osteoclastic bone resorption in vitro. It has been shown that peptides containing the RGD sequence potentially inhibit bone resorption (Horton et al. 1991, Lakkakorpi et al. 1991b, Sato et al. 1990). It has also been shown that osteoclasts adhere in vitro to a wide range of RGD peptide containing proteins, including bone sialoproteins, via a 3 integrin (Helfrich et al. 1992). Controversial data has been reported on the precise localization of vitronectin receptor in osteoclasts. It has been suggested that vitronectin receptor mediates the tight attachment of osteoclasts to bone matrix (Nakamura et al. 1996, Reinholt et al. 1990, Sato et al. 1990) and that osteopontin, a bone matrix component containing the RGD sequence (Oldberg et al. 1986) is the ligand of the osteoclast vitronectin receptor. Nevertheless, VNR has also been shown to be located only in the ruffled border, basolateral membranes and intracellular vesicles of osteoclasts, but missing from the area of the tight sealing zone (Lakkakorpi et al. 1991b, Suda et al. 1995). Although VNR is important in the function of osteoclasts, the final tight attachment of osteoclasts to the bone surface is mainly not mediated through it, according to their known structural features (Horton 1997, Lakkakorpi et al. 1991b). Interestingly, in echistatin-treated mice, where bone resorption is inhibited, the number of osteoclasts on the bone surface was increased rather than decreased and their electron microscopic morphology was found to be normal (Masarachia et al. 1998, Yamamoto et al. 1998). Hence, inhibition of integrin function did not cause detachment of osteoclasts from the bone surface or changes in the clear zone or ruffled borders.

25

Osteoclasts can attach to and migrate on bone surface in vitro without performing bone resorption. The attachment of these non-resorbing osteoclasts is referred as primary attachment, and it is thought to be mediated by integrins, which allow the cells to attach to the matrix by many simultaneous but weak bonds. This enables the cells to move along the bone surface (Vnnen et al. 1995). Expression of vitronectin receptor has also been detected in mononucleated tartrate resistant acid phosphatase (TRAP) positive cells during in vitro osteoclastogenesis in culture, and RGD-containing disintegrins, such as echistatin, inhibited the formation of multinucleated TRAP-positive cells. This observation suggests that v3 may play a role in osteoclast differentiation in vitro, possibly by interfering with the migration of osteoclast precursors for their fusion (Tanaka et al. 1991).

2.2.3 Cadherins
Cadherins are a family of structurally and functionally related molecules that usually mediate homophilic, calcium-dependent cell-cell interactions, and are predominantly localized at adherent junctions of epithelial cells (Boller et al. 1985, Geiger et al. 1992, Takeichi et al. 1988, Takeichi 1991). The demonstrated role of cadherins in mediating cell interactions, coupled with the complex spatio-temporal patterns of cadherin expression during embryonic development, suggest that cadherins play a critical role in embryogenesis and morphogenesis (Grunwald 1993, Takeichi 1991). The classical cadherins are represented by a group of highly conserved integral membrane proteins initially identified biochemically, immunochemically and functionally about 20 years ago. These classical N-, E- and P-cadherins, along with the more recently found R, B and EP-cadherins, posses similar structural and functional domains, which include sites for adhesive recognition, calcium binding, membrane integration, cytoskeletal interactions, and post-translational processing such as glycosylation, phosphorylation and proteolysis (Geiger et al. 1992, Grunwald 1993, Kemler 1992) (Fig. 3). Cadherins are protected by 2+ 2+ Ca against proteolysis but they are readily degraded if Ca is absent (Grunwald et al. 1981, Takeichi 1977).

26

Calciumbinding site

Glycosylation sites

Phosphorylation sites

Cytoskeletal binding site

NH2 - EC1

EC2

EC3

EC4

TM

CYTO - COOH

Adhesive recognition site

Catenins

Fig. 3. Structural organization of cadherins. Indicated are the five extracellular (EC1-EC5), transmembrane (TM), and cytoplasmic (CYTO) domains. The extracellular domains include multiple glycosylation sites, a proteolytic cleavage site near the cell membrane and functionally critical calcium-binding and adhesion recognition sites. The cytoplasmic domain includes sites for interaction with catenin cytoskeletal proteins. The cadherins and catenins both contain phosphorylation sites that regulate their function in intercellular junctions.

Different cadherins are similar in their overall primary structure: they have a single transmembrane domain that divides the molecules into the amino-terminal extracellular and the carboxy-terminal cytoplasmic domain. The most highly conserved region of cadherins is the cytoplasmic domain, and the second most conserved region is around the amino-terminal portion. Cadherins contain a common cell adhesion recognition (CAR) sequence HAV (His-Ala-Val) residing in the outermost region of the extracellular domain (Blaschuk et al. 1990a). HAV motifs have also been identified in Influenza virus hemagglutinins which are integral membrane proteins mediating attachment and fusion of the virus with the epithelia of the mammalian and avian upper respiratory tract (Blaschuk et al. 1990b). The HAV sequence is also found in fibroblast growth factor receptors and the deletion of this region completely abolishes fibroblast growth factor function (Byers et al. 1992). Cadherins have generally been associated with homotypic cell interactions among solid tissues. Interestingly, it has been suggested that cadherins may also be involved in heterotypic or even heterophilic interactions (Cepek et al. 1994, Tang et al. 1993). Moreover, a recent study has shown that cadherins can interact with collagen type I (Mbalaviele et al. 1996) and hence could act as cell-matrix adhesion molecules. Cadherins associate with the cytoskeleton of cells via proteins called catenins. This protein family has at least three major classes, , and , which interact with the cadherins (Knudsen et al. 1992, Ozawa et al. 1990). Even though cadherins have essential roles in embryogenesis, many mature tissues also continue to express cadherins. Cadherins may thus be partly responsible for maintaining tissue integrity. Indeed,

27

reduction of E-cadherin expression is positively correlated with progression of various cancers and increased cellular invasiveness (Umbas et al. 1992). It seems that E-cadherin could have tumor suppressor function. E-cadherin is expressed in osteoblasts (Babich et al. 1994) and a new cadherin called OB-cadherin has also been cloned from osteoblasts (Okazaki et al. 1994). E-cadherin has been suggested to be expressed in osteoclasts and to take part in the fusion of mononuclear osteoclast precursors to multinucleated mature osteoclasts. It has also been shown that the fusion of osteoclast precursors into multinucleated osteoclasts can be inhibited using a peptide containing the HAV sequence (Mbalaviele et al. 1995). It is, therefore, possible that since the HAV sequence inhibits osteoclast formation, the fusion of mononuclear osteoclast precursors mediated by E-cadherin may also involve cytoplasmic signaling events, as recently proposed (Cowin 1994).

2.2.4 Degradation of the mineral and organic matrix

Both the mineral and organic component of bone matrix are degraded by osteoclasts in resorption lacunae under ruffled border membranes. Osteoclasts resorb bone by generating a pH gradient between the cell and bone surface (Anderson et al. 1986, Baron et al. 1985, Blair et al. 1993). An acidic pH favors the dissolving of the bone mineral (hydroxyapatite) and provides optimal conditions for the action of the proteinases secreted by osteoclasts. Inorganic bone matrix is dissolved by the acidic environment in the resorption lacuna revealing the organic collagen network. Carbonic anhydrase II (CA II) is a cytoplasmic enzyme hydrolyzing carbon dioxide into bicarbonate and protons (Gay et al. 1974, Gay et al. 1983, Vnnen et al. 1983). CA II is suggested to be the main source of protons for the acidification of the resorption lacuna (Blair et al. 1989, Laitala et al. 1994, Minkin et al. 1972, Vnnen et al. 1990). A vacuolar-type proton pump, V-ATPase, is present in high amounts in the membranes of a population of intracellular vesicles. V-ATPase transports the protons generated by CA II into these vesicles, which are then transported and fused to the RB membrane (Bekker et al. 1990, Blair et al. 1989, Laitala et al. 1994, Sundquist et al. 1990, Vnnen et al. 1990) releasing their proton content to the lacuna. Acidification of the extracellular resorption lacuna is completed by passive, potential-driven chloride transport (Blair et al. 1991). RB thus contains large amounts of V-ATPase, which probably continues to function by transporting protons directly from the cytoplasm to the lacuna. Hydroxyapatite is first solubilized in the acidified lacuna, after which various proteolytic enzymes, such as lysosomal enzymes (Baron et al. 1988, Delaisse et al. 1984, Delaisse et al. 1987, Rifkin et al. 1991) and bone-derived collagenases (Chambers et al. 1985, Delaisse et al. 1988) secreted by osteoclasts through RB, digest the exposed organic matrix. When the osteoclast stops resorption and moves away from the resorption lacuna, phagocytes clean up the remains and make room for osteoblasts to begin bone formation in the newly formed resorption cavity. Only recently it has been shown that osteoclasts remove the degradation products of bone matrix from the resorption lacuna by transcytosis (Nesbitt et al. 1997, Salo et al.

28

1997). The degradation products, both organic and inorganic, are endocytosed to the resorbing osteoclast, transported through the cell in large vesicles, and finally released into the extracellular space through the FSD. This enables that osteoclasts can remove large amounts of degradation products via transcytosis without detaching from the bone surface and loosing the tight attachment of the sealing zone to the bone surface. This further supports the osteoclastic penetration deeper into the bone. The transcytosis route provides a possibility for osteoclasts to further process the endocytosed degradation products intracellularly during their passage through the cell. The bone-specific enzyme TRAP is located in cytoplasmic vesicles, which fuse to the transcytotic vesicles and participates in destroying the endocytosed material in the transcytotic route (Halleen et al. 1999).

2.2.5 Regulation of osteoclastic bone resorption

A balance between bone formation and bone resorption is a necessity for the normal function of bone. Although many acquired or environmental factors are known to affect the adult bone mineral density (BMD) (Vlimki et al. 1994), genetic factors play a major role as determinants of variation in BMD (Pocock et al. 1987). It has been estimated that up to 80% of BMD is genetically controlled, and it is the rate of bone formation rather than the rate of bone resorption that is influenced by genes (Kelly et al. 1991). Systemic stimulators of bone resorption include parathyroid hormone (PTH) (Barnicot 1948, Holtrop et al. 1974, Miller 1978, Raisz 1963), interleukin-1 (Chambers 1988, Gowen et al. 1983a, Gowen et al. 1983b), tumor necrosis factor (Abe et al. 1986, Thomson et al. 1987), transforming growth factor (Takahashi et al. 1986) and 1,25dihydroxyvitamin D3 (1,25(OH)2D3) (Raisz et al. 1972, Tanaka et al. 1971). On the other hand, calcitonin (CT) (Holtrop et al. 1974), gamma interferon (Gowen et al. 1986) and transforming growth factor (Bonewald et al. 1990, Chenu et al. 1988) are extremely potent in inhibiting osteoclast differentiation and activity. PTH and 1,25(OH)2D3 stimulate bone resorption by increasing the activity of existing osteoclasts and by promoting the differentiation of osteoclast precursors into mature multinucleated osteoclasts and 1,25(OH)2D3 or other vitamin D metabolites seem to play a role in the correction of calcium malabsorption, which is a common feature in osteoporosis (Lamberg-Allardt 1991, Taterossian 1973, Weisbrode et al. 1978). It has also been shown that peak bone mass is associated with vitamin D receptor polymorphism (Viitanen et al. 1996) and is also related to the rate of bone loss (Ferrari et al. 1995). Osteoblasts mediate the effects of PTH and 1,25(OH)2D3 on the osteoclasts (Narbaitz et al. 1983, Silve 1982). CT, a peptide hormone secreted by the thyroid, regulates blood calcium and phosphate levels (Kumar 1963) by causing short-lived falls in the plasma calcium. It does that by its effects to inhibit osteoclastic bone resorption (Friedman et al. 1986) and to promote renal calcium excretion (Ralston et al. 1985). In the past decades, it has become apparent that the skeleton is a major target tissue for reproductive steroid hormones, estrogen, androgen, testosterone, and to a lesser extent, progestin. The sex steroids, especially estrogen and testosterone, play a major role in the

29

sexual dimorphism of the skeleton in mineral homeostasis during reproduction and bone balance in adults. Estradiol and progesterone are the principal circulating sex steroids in females but also function in males (Landers et al. 1992). Sex steroids are essential for maintenance of normal bone volume and estrogen deficiency is an important risk factor for osteoporosis. Ovariectomy leads to a deficit in ash weight and bone mineral density in adult rats and monkeys (Jayo et al. 1990, Kalu et al. 1991, Lindsay et al. 1978) and it is generally believed that treatment with estrogen of ovariectomized rats prevents the increase of bone resorption (Anderson et al. 1970, Turner et al. 1988, Wronski et al. 1988).

2.2.6 Disorders of bone modeling and remodeling

Normal bone remodeling is initiated by local events that lead to an increase in osteoclastic activity. This is followed eventually by an increased attraction and proliferation of osteoblast precursors followed by the differentiation of these cells into mature osteoblasts, which lay down new bone and repair the resorption defects caused by the resorbing osteoclasts. Formation follows resorption at the resorption site, not at some other location, and the amount of bone formed is almost always nearly equal to the amount removed. Coupling is also manifest at a higher level of organization: global rate changes in resorption are invariably followed by similar global changes in formation. The remodeling process replaces aged bone tissue with new bone tissue. With the passage of time, repeated strain of skeletal tissue that occurs during ordinary mechanical usage results in the development of microdamage and the need to replace or renew bone tissue. Unless microdamage is repaired, it will accumulate and lead to catastrophic structural failure, namely, fracture. Several bone-related disorders have been mentioned in the literature, two of which affecting the bone mass and the mechanical competence, are shortly described here. Osteopetrosis is a metabolic bone disease characterized by decreased bone resorption due to reduced osteoclastic number and defects in the ruffled border membrane and leading to impaired function of osteoclasts (Marks, Jr. 1987, Walker 1973b). This leads to accumulation of excessive amounts bone seen as thickening of cortical region and decrease in the size of medullary space in the long bones, with sclerosis of the base of the skull (Elster et al. 1992a, Elster et al. 1992b). A reduced marrow space results in a decrease in hematopoiesis even to the point of complete bone marrow failure. Nerve compression and vascular compromise can also occur due to decrease in the caliber of cranial nerve and vascular canals. In the mildest form the disease can produce only few, or no symptoms, whereas the malignant form is fatal in early childhood. Although having higher bone mass, the osteopetrotic bones are easier to fracture (Tuukkanen et al. 2000). On the other hand, osteoporosis is a bone disease with reduction in bone mass and a change in the microarchitecture, which increases susceptibility to fractures. In osteoporosis, osteoclastic activity is highly increased leading to decreased bone mass. The causal role of estrogen deficiency in this condition is well established (Odell et al. 1993, Takano-Yamamoto et al. 1990, Tuukkanen et al. 1991). Immediately after

30

menopause, the rate of osteoclastic bone loss increases as much as 10-fold, the annual rate being about 2-3% in Caucasian women (Lindsay et al. 1980, Smith et al. 1975). Numerous investigations have been performed to study the therapeutic approaches treating osteoporosis (Rodan et al. 2000).

2.3 Cell polarity

Cellular polarization is best established in epithelium. The plasma membrane of epithelial cells is divided into two domains: an apical domain facing the external milieu and a basolateral domain in contact with the internal milieu and the blood supply. These plasma membrane domains have different lipid and protein compositions (Simons et al. 1988, van Meer et al. 1988). Epithelial cells have tight junctions, which separate the apical and basolateral membranes from each other. The general features defining cell polarity have been described by far in cultured (MDCK, Caco-2) and primary epithelial cells (Handler 1989, Rodriguez-Boulan et al. 1992). Many of the functions of epithelial, neuronal and certain other cells of the immunosystem are maintained by the characteristically polarized phenotype of these cells (Cereijido et al. 1989, Craig et al. 1992, Knust 1994, Nelson 1993, Rodriguez-Boulan et al. 1992) Polarization is also essential for embryonic development and tissue morphogenesis. Most of the plasma membrane proteins in epithelial cells are distributed primarily to either the apical or the basolateral plasma membrane domains, but very few are distributed to both of these surfaces. It is probable that only one of the two pathways to the polarized cell surface requires specific sorting signals (Matter et al. 1994). The membrane polarity of epithelial cells has been studied by using glycoproteins of enveloped viruses as markers (Rodriguez-Boulan et al. 1980, Simons et al. 1990). In epithelial cells, Influenza virus targets its glycoproteins to the apical membrane domain facing the external milieu, thus identifying it. Viral particles of VSV-infected epithelial cells bud from the basolateral membrane, which is in contact with the internal milieu and the blood supply. In epithelium, different plasma membrane domains are clearly separated by tight junctions enabling the maintenance of membrane polarity. Most of the osteoclast membrane has receptors for VSV, representing a basolateral membrane domain. The special cap structure on the top of an active osteoclast, the FSD, is the target of viral glycoproteins in osteoclasts infected by Influenza A and Sendai viruses, a feature characteristic of the apical membrane of epithelial cells, suggesting that FSD corresponds to the apical surface of epithelial cells (Salo et al. 1997). No exclusive reports have been presented on the polarity of osteoblastic cells. However, it has been shown that during the secretory phase, osteoblasts look highly polarized, and the matrix secretion is in the direction of the bone surface making it seem like a polarized event (Cooper et al. 1998, Marks, Jr. et al. 1996). In addition, RingbomAnderson et al. have presented that during matrix formation, several matrix vesicles can be found to bud from the long processes of U-2 OS osteoblastic cells, making the cells look highly polarized (Ringbom-Anderson et al. 1994).

31

2.4 Communication between bone cells

A series of complex and coordinated events is required for bone remodeling (Marcus 1987, Vnnen 1993). Osteoblasts, osteocytes and osteoclasts communicate with each other resulting in the elaborate balance of bone formation and resorption. It requires wellorganized actions by osteoblasts, osteoclasts, and osteocytes. This coordinated activity between bone cells suggests the existence of intercellular communication, but the mechanisms of this communication are not well characterized. Intercellular communication can occur via secretion of soluble factors, direct cell-cell contact through cell adhesion molecules or metabolic and electric coupling through gap junctions.

2.4.1 Gap junctions

Most cells communicate with their immediate neighbors through exchange of cytosolic molecules, such as ions, second messengers, and small metabolites. This activity is made possible by clusters of intercellular channels called gap junctions, which connect adjacent cells (Kumar et al. 1996). Thus, apart from a few terminally differentiated cells, such as skeletal muscle, erythrocytes, and circulating lymphocytes, most cells in normal tissues generally communicate via gap junctions. Gap junctions are specialized regions of cell membrane, which connect cells and provide aqueous intercellular channels for bidirectional movement of small molecules and ions between the cytoplasms of adjacent cells (Warner 1988). Each gap junction pore is formed by a juxtaposition of two intercellular hemichannels in neighboring cells, which interact to span the plasma membranes of two adjacent cells and directly join the cytoplasm of one cell to that of another (Fig. 4 and 5). Each hemichannel is composed of a hexameric array of transmembrane proteins called connexins, a highly related multigene family consisting of at least 13 members (Beyer et al. 1990, Goodenough et al. 1996, Saez et al. 1993).

32

Extracellular

E1

E2

Gap 27

Membrane

M1

M2

M3

M4

Cytoplasm

NH2

COOH

Fig. 4. Connexin structure. The cylinders represent transmembrane domains (M1-M4). The loops between the first and second, as well as third and fourth, transmembrane domains are predicted to be extracellular (E1 and E2, respectively).

Connexin channels participate in the regulation of signaling between developing and differentiated cell types. The gap-junctional channels have an apparent selectivity based principally on molecular size, allowing the movement of molecules smaller than 1000 Da, such as cAMP, but preventing the movement of proteins and nucleic acids. Small informational molecules, such as certain morphogens could be directly transmitted between cells via gap junctions, which is important in, for instance, embryogenesis for regulating the events between cells (Warner et al. 1984). A spectrum of drugs has been shown to inhibit gap-junctional communication with varying degrees of efficacy and specificity. These include for instance volatile anesthetics such as halothane (Nedergaard 1994), the straight-chain alcohols heptanol and octanol (Donahue et al. 1995a, Nedergaard 1994), anandamide (Venance et al. 1995) and glycerrhetinic acid (Davidson et al. 1995). Although a variety of drugs have been used to block gap-junctional communication, the specificity of the blockers remains to be solved. Bone cells mediate signals to each other through the communicating gap junctions, which are found connecting adjacent osteoblasts together, between osteoblasts and overlying cells or osteocytes (Akisaka et al. 1990, Jeansonne et al. 1979, Jones 1974, Stanka 1975, Weinger et al. 1974, Yellowley et al. 2000). Osteocyte nutrition probably takes place across the gap junctions between the cytoplasmic extensions of neighboring osteocytes. Connexin-43 (Cx43) is widely expressed in different tissues, including brain, heart, kidney, smooth muscle, ovary, and some epithelia, and it is actually the most abundant connexin (Goodenough et al. 1996, Musil et al. 1990). Cx45 and Cx46 have been identified in bone, but Cx43 is the predominant gap junction protein expressed in bone (Donahue et al. 1995b, Vander et al. 1996).

33

On cell contact, connexons from neighboring cells may form gap junctions in a matter of minutes or even seconds, suggesting that connexons pre-exist in the plasma membrane (Rook et al. 1990). Gap-junctional communication would thus be an excellent method for bone cells to rapidly propagate locally generated signals throughout the skeletal tissue, as has been shown to happen in osteoblasts (Civitelli et al. 1993). A dependable means of communication is utterly important for correct coupling of osteoclasts and osteoblasts. The capability to exchange nutrients and regulatory molecules from one cell - or one cell type - to another throughout the skeletal environment may represent an efficient means of coordinating the action of a variety of cells in a highly complex tissue.

Homomeric Homotypic

Heteromeric Homotypic

Homomeric Heterotypic

Heteromeric Heterotypic

Fig. 5. Possible arrangements of connexons to form gap junction channels. Connexons consist of six subunits, connexins. A connexon may be homomeric, when it is composed of six identical connexin subunits or heteromeric, when it has more than one species of connexins. Connexons associate end to end to form a double membrane gap junction channel. The channel may be homotypic, if connexons are identical or heterotypic, when the two connexons are different.

3 Aims of the present study

Osteoclastic bone resorption as well as osteoblastic formation of new bone are both required for the correct composition of bone tissue. This is called bone remodeling. The bone multicellular unit works synchronically as a cell syncytium, which is controlled and regulated by hormonal signals and by mechanical forces adjusted to bone. The hypothesis of this work was that when the bone cells receive a signal to start the remodeling cycle, they have to communicate with each other in order to retain the balance between the activities of different bone cells. This requires attaching of bone cells to the matrix, polarizing of the cells plasma membranes into specific domains and contacting of bone cells to each other. The basic mechanism of this coupling has not been previously known. The aims of the present study work can be briefly outlined as follows: 1. To study how osteoclasts attach to bone surface. We wanted to learn what are the proteins involved in the tight attachment in osteoclasts and what proteins are needed from the bone surface. 2. To study if osteoblasts truly are polarized cells. The secretion of bone matrix happens in a polarized manner. We wanted to study the mechanisms underlying this feature of osteoblasts. 3. To study the communication mechanisms by which bone cells communicate with each other. The communication between osteoblasts and osteoclasts is essential for proper bone remodeling - we wanted to solve how the actual connection between the cells happens.

4 Materials and Methods

4.1 Reagents (I-IV)
All cell culture media and reagents were purchased from Gibco BRL (Paisley, England). Hexapeptides AHAVSE and VASAEH were synthesized in Department of Medical Biochemistry, University of Turku, Finland. Heptanol was purchased from Sigma Chemical Co. (St. Louis, MO). Synthetic peptide Gap 27 (SRPTEKTIFII) was synthesized in Keck Foundation Biotechnology Resource Laboratory, Yale University School of Medicine , USA.

4.1.1 Antibodies (I-IV)

Polyclonal rabbit antibody against pan-cadherin (Sigma Chemicals, St. Louis, MO), monoclonal antibodies against E-cadherin (Developmental Studies Hybridoma Bank, Iowa City, IA) and -catenin (Transduction Laboratories, Lexington, KY) were all used at 1:50 in PBS. A monoclonal mouse antibody against chicken gizzard smooth muscle vinculin was purchased from ICN Immuno Biologicals (Lisle, IL) and used at a dilution 1:50 (I). Antibodies against the gel-purified HA band of the WSN strain to Influenza virus hemagglutinin and an antibody against the Influenza virus nucleocapsid protein have been described earlier (Martin et al. 1991). The VSV G protein and the matrix protein were localized using specific polyclonal antibodies produced in our own laboratory (Metsikk et al. 1992) (II). Monoclonal antibodies raised against Connexin-43 (Transduction Laboratories, Lexington, KY) were used at a dilution 1:100 (III). Secondary antibodies, either tetramethylrhodamine isothiocyanate (TRITC) or fluoresceinisothiocyanate (FITC)-conjugated, rabbit anti-mouse or swine anti-rabbit or goat anti-rabbit were also obtained from DAKO, and used at a dilution of 1:100 (I-IV). TRITC or FITC-conjugated phalloidin was purchased from Molecular Probes Inc. (Eugene, OR) and was used at 5 units/ml. DNA-binding fluorochrome Hoechst 33258 (1

36

mg/ml stock, diluted 1:800, Sigma Chemical Co., St. Louis, MO) was used to visualize nuclei (I, III-IV).

4.2 Cell culture (I-IV)

4.2.1 Isolation and culture of osteoclasts (I, III-IV)

Osteoclasts from 1 or 2-day-old Sprague-Dawley rat pup long bones were scraped mechanically into DMEM (Dulbeccos modified Eagles medium) buffered with 20 mM Hepes and containing 0.84 g of sodium bicarbonate/liter, 2 mM L-glutamine, 100 IU of penicillin/ml, 100 g of streptomycin/ml and 7-10% heat-inactivated fetal calf serum (FCS). The cells were allowed to attach to sonicated bovine cortical bone slices for 30 minutes. After the attachment period, non-attached cells were rinsed away, and the bone slices with the remaining cells were transferred into 24-well plates containing fresh medium with appropriate substances to be tested (Boyde et al. 1984, Chambers et al. 1984, Lakkakorpi et al. 1989). Cells were cultured at +37C (5% CO2, 95% air) for up to 48 hours. In further studies some of the following changes were made to the culture medium: hexapeptides AHAVSE or VASAEH at concentration 1 mg/ml, heptanol 3 mM or Gap 27 at 500 M were added to the culture medium of the cells. The cultures were usually stopped by fixing the cells and bone samples with freshly made or refrigerated 3% paraformaldehyde (PFA), 2% sucrose in phosphate-buffered saline (PBS) for 10 minutes at room temperature.

4.2.2 Cell line (II)

UMR-108 osteoblastic rat osteosarcoma cells (ATCC CRL-1663 similar to UMR-106 CRL-1661) were cultured in MEM (Minimum Essential medium) containing 2 mM Lglutamine, 100 IU of penicillin/ml, 100 g of streptomycin/ml and 7-10% heatinactivated fetal calf serum (FCS), which will be later referred to as MEM-FCS. The cells were grown at +37C (5% CO2, 95% air) in small cell culture bottles in 5 ml of medium, and for the experiments the cells were trypsinized and plated on 150 m thick 1 cm2 bovine bone slices.

4.2.3 Tissue culture (II)

For studying endosteal osteoblasts, 3 to 4 days old rat pup femora, tibiae and humeri were dissected out, placed in MEM (Minimum Essential medium) containing 2 mM L-

37

glutamine, 100 IU of penicillin/ml, 100 g of streptomycin/ml and 7-10% heatinactivated fetal calf serum (FCS). After removing the epiphyseal cartilages the diaphyses, bones were bisected with a scalpel blade. The loose bone marrow was gently removed by injecting culture media over it with a pipette. The bones were cultured overnight at +37C (5% CO2, 95% air).

4.3 Fluorescence and microscope techniques (I-IV)

4.3.1 Light microscopy (I, III-IV)

Osteoclasts were identified by staining them for tartrate-resistant acid phosphatase (TRAP), a commonly accepted marker of osteoclasts (Minkin 1982), using a Sigma TRAP kit (No. 386-A, Sigma Chemicals, St. Louis, MO) according to manufacturer's instruction.

4.3.2 Conventional fluorescence microscopy (I, III-IV)

Osteoclasts were cultured in DMEM-FCS growth medium, with either hexapeptides AHAVSE or VASAEH (I), heptanol (III) or Gap 27 (IV). After growing the cells for 48 hours, the cells were fixed and permeabilized with various methods for the immunofluorescence staining.. With E-cadherin antibodies the cells were first fixed with methanol for 5 min at -20C and quickly rinsed in acetone and let air dry. When using catenin, cells were fixed with 10 min methanol at room temperature. With pan-cadherin, vinculin, the Influenza virus hemagglutinin, Influenza virus nucleocapsid protein, and the VSV G protein and the matrix protein and Connexin-43 antibodies, phalloidin or Hoechst 33258, the cells were fixed in 3 % paraformaldehyde in PBS containing 2 % sucrose for 10 min at room temperature. Then the samples were rinsed in PBS and permeabilized with 0.1% Triton X-100 in PBS for 10 minutes on ice. For immunostaining, the bone slices or bone samples were incubated with primary antibody one to two hours at +37C. The samples were rinsed thoroughly several times with PBS, and further incubated with TRITC- or FITC-labeled secondary antibodies for 30-60 min at +37C. Non-specific binding of antibodies was blocked with swine serum. The procedure for staining F-actin has previously been published in detail (Lakkakorpi et al. 1991a). F-actin was stained using either TRITC or FITC-conjugated phalloidin for 20 minutes at +37C. Finally all the samples were mounted in either Mowiol (Hoechst, Frankfurt) in PBS containing 2.5% 1,4-diazobicyclo(2,2,2)-octane or 50% PBS-glycerol. To visualize the nuclei, the cells were incubated with the DNA-binding fluorochrome Hoechst 33258 (1 mg/ml stock diluted 1:800 in PBS) for 10 minutes at room temperature. The samples were viewed with a Leitz Aristoplan microscope with appropriate filters.

38

4.3.3 Confocal laser scanning microscopy (I-IV)

Samples for confocal laser scanning microscope were prepared in the same way as for conventional epifluorescence microscopy. The confocal laser scanning microscope consisted of a 750 mW air-cooled argon-crypton laser (Omnichrome, Chino, CA), a Leitz Aristoplan microscope, and 1.05-software (Leica Lasertechnik Gmbh, Heidelberg, Germany). Scanning was performed in a XY-plane using either a 40 x 0.6 water immersion objective (Nikon, Tokyo Japan), a PLAPO 63x objective (NA 1.4) or Fluor 40x1.3 and 50x1.0 oil immersion objectives (Leitz, Germany) with suitable zoom factors (1x or 2x) of the scanner. Laser power and offset were adjusted according to the fluorescence label intensity. The localization of immunostaining was acquired by scanning serial optical xy and xz sections. In the bone tissue cultures the metaphyseal ends of the long bones were investigated with and the association of labeled cells on bone surface was confirmed by simultaneous observation with Nomarski differential interference contrast optics in transmitted light.

4.3.4 Immunoelectron microscopy (II)

The cell cultures were fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4, for 1 to 2 days at 4C. The bone slices were decalcified with 7% EDTA in phosphate buffer for 4 days at +4C with daily changes of the decalcifying medium. Ultrathin sections (100 nm) were made after removal of mineral from the bone slice. The samples were then postfixed with 4% OsO4, dehydrated and embedded in Epon. The samples were examined with a Philips 410 electron microscope (Philips, The Netherlands).

4.3.5 Digital image analysis (I -IV)

MCID image analyzers utilizing M1 or M2 software (Imaging Research Inc., Brock University, Ontario, Canada) consisting of an Intel 403 E microcomputer linked to an Image 1280 image processor (Matrox Dorral, Quebec, Canada), were used to the quantification of areas of the resorption lacunae. Resorption pits were visualized as brown-colored areas after staining with peroxidase-conjugated wheat germ agglutininlectin (WGA) (Sigma Chemical Co., St. Louis, MO) (Selander et al. 1994) and quantitated by combining the 10 X light microscopy image (Nikon TMD Diaphot) to a color-CCD camera (Sony XC 711P vision camera module).

39

4.4 Viral infections (II)

UMR-108 cells were cultured for two days to near confluency, and the cells were infected with wild type vesicular stomatitis virus (Indiana serotype) or Influenza virus of A/WSN strain. Adsorption was for 1 h at 37C with 10 plaque-forming units in MEM containing 0,1% BSA, after which the cells were further cultured for 3 hours (VSV) or 7 hours (WSN) in the presence of serum. The endosteal osteoblasts were infected after culturing the bisected bones overnight with same procedure as UMR-108 cells.

4.4.1 Metabolic labeling (II)

Cells cultured on bone slices were labeled with S-methionine (Amersham Corp., Bucks, England) at 4 hours postinfection (VSV) or 6 hours postinfection (WSN) for 10 minutes at 500 Ci/ml and chased for 90 minutes.
35

4.4.2 Cell surface biotinylation (II)

After the chase, biotinylation was performed. The plasma membrane proteins of the infected cells were labeled with sulpho-NHS-biotin (Pierce Chemical Co., Rockford, IL) at 0C as described by Lisanti et al. (Lisanti et al. 1988). The biotinylated proteins were isolated by the methods described by Hare & Lee (Hare et al. 1989). Briefly, membrane proteins labeled with sulfo-NHS-biotin at a concentration 0.5 mg/ml for 30 minutes at +0C were extracted from lysed cells and centrifuged at 30 000 x g for 1 h with a SW50.1 rotor. Streptavidin-agarose (Sigma) was used to adsorb the biotinylated proteins by using end-over-end rotation overnight at +5C. Proteins were eluted from the streptavidinagarose by heating for 3 minutes at +95C in SDS sample buffer. SDS-PAGE was carried out by the method of Laemmli (Laemmli 1970). After the SDS-PAGE, the gels were dried and the intensities of radioactivity of both the streptavidin bound fraction and cell lysates were quantified and calculated using a phosphoimager (Molecular Dynamics, Sunnyvale, CA) using MCID-M2 image analysis software (Imaging Research Inc., Brock University, St. Catherines, Canada).

4.5 Statistical analysis (I, III-IV)

All statistical analyses were performed by using Students t-test for independent samples and a computer program Microcal Origin version 4.00 (Microcal Software Inc., Northampton, MA).

5 Results
5.1 Cadherin localization in osteoclasts (I)
To localize cadherins in osteoclasts we stained the cells with different cadherin or cadherin-related antibodies. E-cadherin and -catenin antibodies were first tried in immunofluorescence staining of osteoclasts. Neither of the antibodies gave positive staining in osteoclasts even though MDCK-cells as a positive control showed similar staining patterns as described earlier. We could distinguish osteoclasts from the other cells in the cell culture by visualizing the nuclei with Hoechst stain. On the contrary, pan-cadherin antibody reacted with osteoclasts forming a ring-like structure localized in the sealing zone area. Some staining can also be seen in the cytoplasm. Double immunofluorescence staining of pan-cadherin with F-actin revealed that pan-cadherin antibodies co-localized with the actin ring in the sealing zone area. Pancadherin staining was also seen in between the vinculin double-rings, partly overlapping with them.

5.2 Effects of HAV-containing peptide (I)

5.2.1 Primary attachment of osteoclasts to bone surface

To reveal if the common cadherin CAR sequence motif, HAV, is involved in the primary attachment of osteoclasts to the bone surface, the CAR-containing peptide AHAVSE was added to the primary attachment medium of osteoclasts. When harvesting osteoclasts from rat pups the cells were allowed to attach to the bone slices with the attachment medium containing either AHAVSE or the control peptide VASAEH, both at a concentration of 1 mg/ml, or with no added peptide to act as zero-controls. Attached TRAP-positive multinucleated cells were counted after rinsing the non-attached cells

41

away. Similar amounts of TRAP-positive multinucleated cells were seen in AHAVSEtreated, VASAEH-treated and non-treated control cultures. Thus, the cadherin CAR sequence-containing peptide AHAVSE had no effect on the primary attachment of osteoclasts to bone slices.

5.2.2 Osteoclastic bone resorption

We studied the possible effect of AHAVSE on the ability of osteoclasts to form resorption pits. Both the number of resorption pits and the total resorbed area decreased significantly after addition of the cadherin CAR sequence-containing peptide AHAVSE. No significant differences in comparison with the zero-control were detected when the control peptide (VASAEH) was used. The mean size and the size distribution of resorption pits were similar in all the treatments.

5.2.3 Activity of osteoclasts

Since the AHAVSE peptide inhibited osteoclastic bone resorption, the numbers of active osteoclasts having actin rings were investigated. Osteoclasts were cultured for 24 hours, after which AHAVSE or VASAEH was added to the culture medium. The cells were allowed to grow for different periods of time from 20 minutes to 24 hours. After incubation, the cells were fixed and stained for actin rings. The percentage of active osteoclasts was determined as those with actin rings. The number of active osteoclasts decreased significantly after addition of AHAVSE. The decrease was significant at all time points from 20 minutes to 24 hours but it was not linear. No morphological changes in the appearance of remaining actin rings were detected.

5.3 Polarity of osteoblasts (II)

5.3.1 Immunofluorescence data

The VSV G protein localizes to the basolateral membrane of epithelial cells (RodriguezBoulan et al. 1980, Simons et al. 1990). We stained the osteoblast-like UMR-108 cells with the antibody against VSV G protein and found it to be localized to the culture medium facing plasma membrane of the cells. To elucidate the cytoplasm of the UMR108 cells, we used the VSV matrix protein antibody and found it to localize diffusely all over the cytoplasm (Nesbitt et al. 1993).

42

In endosteal osteoblasts, the VSV G protein localization was investigated using both unpermeabilized and Triton-permeabilized cells. In unpermeabilized cells, the VSV G protein was localized to the bone marrow facing plasma membrane. To prove that the VSV G antibodies truly localize to the marrow facing membrane, the cells were permeabilized with Triton. Antibodies against VSV G protein still localized to bone marrow facing plasma membrane, but as expected, they could also be seen in the Golgi complex. Double staining of VSV G protein and VSV matrix protein were also performed to distinguish the cells outlines. At the bone marrow or circulation side in the bone tissue cultures, the G protein immunolabeling had frequently an intense area towards circulation. In Triton-permeabilized osteocytes, the VSV G protein localized apart from the Golgi complex, at the plasma membrane in a nonpolarized manner, although in osteocyte processes extruded into the bone canaliculi, the ends of the processes probably communicating with neighboring cell processes showed somewhat condensed staining. In Influenza-infected UMR-108 cells the Influenza nuclear protein was used as a marker for the nucleus, as shown earlier (Stein et al. 1993). The Influenza hemagglutinin was sorted to the bone substrate-facing surface of UMR-108 cells. The hemagglutinin antibody localized between the nucleus and bone, which was different and opposite to the VSV G protein localization. Influenza HA was also found in the Golgi complex of permeabilized cells.

5.3.2 Electron microscopy data

We used electron microscopy to show that if the glycoprotein localization is polarized, the budding of viruses is also polarized, as shown earlier (Odell et al. 1993). Thus, the localization of viral glycoproteins determines the site for viral budding. Budding virus particles were detected at the plasma membrane of the infected cells. VSV particles were mainly budding on the medium side of UMR-108 cells, where the cells did not overlap. In the cells growing upon each other the cell surface against the bone was still deficient of virus particles, although the upper cells had nonpolarized budding. On the contrary, the Influenza virus was mainly found to bud from the plasma membrane facing the bone slice, where the cells were in close contact with the bone and in near confluency growing as a monolayer. In the cases where the growing cells overlapped, the superficial cells were not polarized, since Influenza was budding uniformly from all surfaces. However, even in these cases the cells in bone contact had virus particles budding mainly from the bone-facing surface.

5.3.3 Biochemical data

Sulfo-NHS-biotin binds to the cell surface proteins and does not internalize at 0C. Cell 35 surface biotinylation together with S-labeling revealed, Influenza HA was only 22%

43

biotinylated. This indicates, that Influenza HA was hidden in the bone facing plasma membrane, which was not available for biotin binding. On the contrary, VSV G protein was 55% biotinylated and thus found in the streptavidin eluate. This indicates that most of the VSV G protein was at the medium facing surface available for biotinylation. Even though the results are clear, they indicate that polarity is not achieved in all of the UMR108 cells.

5.4 Gap junctions in osteoclasts (III and IV)

5.4.1 Connexin-43 localization in osteoclasts (III)

Connexin-43 antibodies were used in immunofluorescence staining of rat bone cell cultures to obtain the precise localization of Cx-43 in osteoclasts. To visualize the cell outlines and the sealing zone area and thereby to distinguish active osteoclasts, double stainings with F-actin were performed. The indirect immunostaining of osteoclasts revealed punctuate distribution of Cx43 immunoreactivity. The staining could be seen in cell-cell contacts with neighboring bone marrow derived mononuclear cells at the interface of adjoining cells, but numerous immunoreactive sites were also apparent at free surfaces and within the cytoplasm of osteoclasts.

5.4.2 Effects of gap-junctional inhibitors heptanol and Gap 27 5.4.2.1 Number and Activity of Osteoclasts
Since osteoclasts express gap-junctional Cx-43, we wanted to test whether the addition of gap-junctional inhibitors, either heptanol or Gap 27, affects the number or activity of osteoclasts. A significant decrease in the number of TRAP-positive cells could be seen in the heptanol-treated (3 mM) and Gap 27-treated (500 M) cultures compared to controls. The numbers of both mononuclear and multinucleated TRAP-positive cells were significantly decreased both in the heptanol-treated and in the Gap 27-treated cultures. We could see no difference in the number or morphology of other cell types in the inhibitor-treated bone cell cultures, only the TRAP-positive cells seemed to be affected. The ratio of mononuclear to multinucleated TRAP-positive cells was altered in both inhibitor-treated groups. The control cultures contained approximately one to two times the number of mononuclear than multinucleated TRAP-positive cells, whereas in inhibitor-treated cultures the number of mononuclear TRAP-positive cells was three to four times greater than that of multinucleated TRAP-positive cells. The results of the Gap 27-treated bone cells were almost identical with heptanol-treated bone cells.

44

To find out why the number of TRAP-positive cells is decreased in the heptanoltreated bone cell cultures, we studied the nuclei of the cells. We could see numerous apoptotic osteoclasts in the inhibitor-treated bone cells cultures, whereas in control cultures we could not see nearly any apoptotic osteoclasts, but the cells looked absolutely healthy with intact nuclei. To reveal whether heptanol has an effect on the activity of the remaining osteoclasts, we counted how many of the TRAP-positive cells had actin rings. The activity of heptanol-treated osteoclasts was half of that of the controls. The activity of Gap 27treated osteoclasts was dramatically decreased when compared to control osteoclasts. Interestingly, the Gap 27 seemed to act as even more potent inhibitor of the activity.

5.4.2.2 Area Measurement of Excavated Pits

Since heptanol and Gap 27 decreased the number and activity of osteoclasts, we studied the effect of these inhibitors on the ability of osteoclasts to form resorption pits. A highly significant decrease could be seen in the total resorbed area in both heptanol-treated and Gap 27 treated cultures compared to the controls. In addition, the number of resorption pits was greatly decreased.

6 Discussion
6.1 Methodological aspects
We chose to use the pit formation assay as the research setting in our studies concerning osteoclasts. This assay, originally introduced by Boyde et al.(Boyde et al. 1984) and Chambers et al. (Chambers et al. 1984), enables the studying of resorption mechanism of isolated, mature osteoclasts. The setting mimics the situation in vivo: osteoclasts can resorb their natural target, mineralized bone. In addition, the events of the resorption cycle of isolated osteoclasts on bone slices are well characterized (Karhukorpi 1991, Lakkakorpi et al. 1991a, Lakkakorpi et al. 1993). Regarding to the cellular events taking place at the BMU in vivo, the pit formation assay is however, biased; bone resorption by osteoclasts is not followed by bone formation by osteoblasts. This may be due to the preparation and pretreatment of bone slices resulting in inactivation of proteins and cytokines, normally liberated via resorption and which are required for the cross-talk between osteoclasts and osteoblasts. This method delimits our studies to monitoring mature osteoclasts, and we thus miss the complicated maturation process of osteoclast precursors, at least during short period culture (Chambers et al. 1984). The effect of the agents used in this thesis work, cadherin CAR-sequence containing peptide AHAVSE or the two gap-junctional inhibitors, are results from the direct effects of these agents on osteoclasts and not of the complex cellular interactions occurring in vivo. However, we may safely conclude that some of the communication between osteoclasts and other bone cells occur in the pit formation assay since several cell types are present in this culture method. Different cellular mechanisms operate in osteoclasts cultured on glass, when compared to those cultured on bone slices. No special structural features of active osteoclasts, like ruffled border or sealing zone can be seen on artificial surfaces, thus the attachment of osteoclasts to glass may also be different from bone (Lakkakorpi et al. 1993). This is why we cultured the cells on bone slices instead of glass. When studying the polarity of osteoblasts we chose to use an osteoblast-like cell line UMR-108 cultured on bone slices, which are similar to UMR-106 cells that have been widely used as a model for osteoblastic cells (Koutsilieris et al. 1987, Matsumoto et al.

46

1985, Partridge et al. 1996). We then confirmed the results using endosteal osteoblast culture to study the osteoblastic cells in situ.

6.2 Cadherins and attachment of osteoclasts to bone

Since osteoclasts express the vitronectin receptor, it has been speculated, that the tight osteoclastic attachment to bone surface in the sealing zone area would be integrinmediated (Davies et al. 1989, Helfrich et al. 1992, Horton 1990). Nevertheless, VNR has been shown to be located only in the ruffled border and basolateral membranes, and it is missing from the area of the tight sealing zone (Lakkakorpi et al. 1991b, Suda et al. 1995). Although VNR is important in the function of osteoclasts, the final tight attachment of osteoclasts to the bone surface is mainly not mediated through it (Lakkakorpi et al. 1991b). The organization of the actin cytoskeleton in the sealing zone area of osteoclasts (Lakkakorpi et al. 1989, Lakkakorpi et al. 1991a) resembles that in epithelial adherens junctions. This led us study the role of cadherins, the main adherens junction proteins, in osteoclastic attachment to bone. We found that osteoclasts show a clear ring-like structure in their sealing zone area with the pan-cadherin antibody. The pan-cadherin antibodies should react with all (or, at least, most) cadherins, since the most homologous cytoplasmic region of the cadherin molecules, were selected for antibody production. It has been shown that these antibodies do react with a very broad spectrum of cadherins, such as N-, E- and P-cadherin (Geiger et al. 1990). We chose to use this antibody because it was not precisely known which cadherins, if any, can be found in osteoclasts and thus, this antibody proved to be excellent probe for the identification of cadherins from osteoclasts. The pan-cadherin ring became localized exactly around the resorption lacuna suggesting it marks the sealing zone. In addition, the staining co-localized with actin-ring and in between vinculin double rings, which further confirmed the cadherin localization to the sealing zone. The staining pattern of pan-cadherin antibodies also rules out the possibility that the sealing zone area cannot be reached by immunoglobulins, thus suggesting that the observation concerning the lack of vitronectin receptor in the same area is not due to steric hindrance (Lakkakorpi et al. 1991b). Mbalaviele et al. have suggested that mature murine osteoclasts contain E-cadherin along their membranes at the resorbing bone surfaces, and ruffled border area, with a lesser amount in the cytoplasm of these cells (Mbalaviele et al. 1995). In our studies, however, we detected no E-cadherin in mature rat osteoclasts. Thus the exact nature and identity of cadherin-like molecules in the osteoclastic sealing zone remains unknown. We used the cadherin CAR sequence containing peptide to study what the possible roles for cadherins in osteoclasts might be. Synthetic peptides containing the HAV sequence have been shown to inhibit compaction of eight-cell-stage mouse embryos, rat neurite outgrowth on astrocytes (Blaschuk et al. 1990a) and myoblast fusion (Mege et al. 1992), all of which are known to be mediated by cadherins. These data strongly emphasize the importance of the HAV sequence in the cellular functions that are dependent on cadherins. We noticed that the blocking of cadherin function by the peptide

47

does not affect the primary attachment of osteoclasts to bone matrix, as also demonstrated earlier (Mbalaviele et al. 1995). This could be expected, since the primary attachment is shown to be mediated by integrins (Duong et al. 1999, Lakkakorpi et al. 1993). When using the peptide AHAVSE, osteoclasts resorbed markedly less bone, which could be seen as a decrease in the number and total area of resorption pits when compared with the zero and peptide controls. Although the resorbed area is greatly decreased with the AHAVSE, we cannot quantify the depths of the resorption pits with the method used here. Thus, the true amount of bone resorbed remains open since we cannot know if the AHAVSE-treated osteoclasts have really resorbed less bone - as it seems when looking at the decrease in the area of resorption pits - or if the osteoclasts have just resorbed deeper resorption pits. Mbalaviele et al. have also shown a decrease in the total resorbed area when using HAV-containing peptide (Mbalaviele et al. 1995). They suggest that the decrease in the resorbed area is due to inhibition of osteoclast precursor fusion. The same report suggests that E-cadherin is involved in the generation of multinucleated osteoclasts from mononuclear precursors. As we wanted to learn if the activity of osteoclasts is truly decreased with AHAVSE or if it is only the mode of resorption that has changed, we measured the activity of osteoclasts by their morphology. The emerging of actin rings to the sealing zone of osteoclast is considered as a mark of osteoclastic activity (Lakkakorpi et al. 1991a). By counting the actin rings and the number of TRAP-positive cells, we can get a percentage of osteoclasts that are actively resorbing bone. We could see a great decrease in the activity of osteoclasts as soon as 20 minutes after AHAVSE peptide addition. The observed disorganization of actin rings leads to inactivation of osteoclasts. We believe that the disorganization of actin rings after the addition of AHAVSE is likely to be too rapid to be a result of inhibition of fusion of mononuclear osteoclast precursors. Our results thus suggest that AHAVSE has a specific effect on the sealing zone, since actin rings are an essential component of this tight attachment-mediating structure. The HAVcontaining peptide could either prevent the formation of new actin rings and tight attachment to the bone surface, or it could speed up the degradation of actin rings and subsequent detachment of the tight sealing. Since the sizes of the resorption pits in AHAVSE incubation were not changed when compared with those of the controls, the inhibition of formation of new attachments seems to be the main reason for the decreased percentage of active osteoclasts. We do not know what the counterpart of osteoclastic cadherin might be. Cadherins have generally been associated with homotypic cell-cell interactions, but it has been shown that cadherins might also be involved in interactions between different cell types or even different cell adhesion molecules (Cepek et al. 1994, Tang et al. 1993). These studies together with a recent report, where it was suggested that cadherins can interact with collagen type I (Mbalaviele et al. 1996) and could act as cell-matrix adhesion molecules, inspired us to investigate the possibility that cadherins might act as cell-matrix mediator in osteoclasts. Others have shown that osteoclasts express different cadherins including E-cadherin (Mbalaviele et al. 1995) and cadherin-6 (Mbalaviele et al. 1998).

48

6.3 Polarity of osteoblasts

Histologically, actively secreting mature osteoblasts are cuboidal or tall cells. When they start to secrete bone, the secretion is polarized towards the bone surface (Cooper et al. 1998, Marks, Jr. et al. 1996). This led us to study whether mesenchymal osteoblasts possess polarized plasma membrane domains, as epithelial cells do. Cellular polarization is best established in epithelium. Viral markers have been used in studying the different plasma membrane domains. The plasma membrane of epithelial cells is divided into two domains: an apical domain, which faces the external milieu and a basolateral domain in contact with the internal milieu and the blood supply. These plasma membrane domains have different lipid and protein compositions (Simons et al. 1988, van Meer et al. 1988). In epithelial cells the polarity is maintained with tight junctions, which separate the apical and basolateral membranes from each other. Our studies reveal that osteoblasts are truly polarized cells. Unlike in fibroblasts, different plasma membrane domains could be separated in osteoblasts by viruses. Fibroblastic cells do not polarize their plasma membrane as the epithelial cells do, and none of the fibroblastic cell lines studied, have distinct domains for VSV G protein or Influenza HA on their plasma membranes. After delivery to the cell surface these two protein species spread all over the surface in a non-polarized manner (Yoshimori et al. 1996). In osteoblastic UMR-108 cells, however, the antibodies against the basolateral marker VSV G protein localized clearly to the medium-facing plasma membrane of the cells. In addition, Vesicular Stomatitis viruses were found to bud from the upper membrane domain of these cells towards the medium corresponding the bone marrow in vivo. To further prove that the polarization of UMR-108 cells resembles that of endosteal osteoblasts and is a meaningful tool, we infected osteoblasts in situ with VSV. We could see that the VSV G protein antibodies localized to the bone marrow-facing plasma membrane of endosteal osteoblasts, hence strengthening our earlier results from UMR108 cells. As osteoblasts mature, the secretion becomes non-polarized and some of the osteoblasts become trapped into the matrix as osteocytes. Osteocytes seem not to secrete matrix components anymore or at least the secretion is probably not polarized. We wanted to see whether the localization of VSV G protein is changed in osteoblasts during this maturation. The VSV G protein seemed to localize in a non-polarized manner at the plasma membrane of osteocytes, apart from the Golgi complex, which could clearly be distinguished. Interestingly, the osteocyte processes branching into the bone canaliculi showed somewhat condensed staining in the ends of the processes which probably communicate with neighboring cell processes by gap junctions. On the contrary, Influenza glycoproteins localized in UMR-108 cells to the plasma membrane facing the bone slice. Also, the budding of Influenza viral particles took place in between the cell and the bone slice. These results were thus in agreement with each other indicating that the bone surface-facing membrane domain of osteoblasts is apical in nature. In the biotinylation studies we also confirmed this conclusion showing that whereas most of the VSV G protein was at the medium-facing surface, where it could be biotinylated, the Influenza HA was hidden in the bone facing plasma membrane, which was not available for biotin binding.

49

Taking these results together - the immunofluorescence data, observation of the viral particle budding and the biotinylation results we can safely conclude that osteoblastic cells clearly posses distinct polarized membrane domains. Still, when comparing the degree of polarity of osteoblastic cells to that of epithelial cells we can see a certain difference. In epithelial cells the polarization is absolute. No VSV-particles can be seen to bud anywhere but the basolateral membrane domain, whereas Influenza particles can only be seen in the apical membrane domain. Although a clear tendency of polarization is easily observed in osteoblastic cells, some viral particles bud randomly from opposite membrane surfaces, in a non-polarized manner. The same randomness is also detectable when looking at the biotinylation results: the biotinylation of VSV G protein is incomplete, indicating that not all of the osteoblasts bud viral particles in a polarized manner. It seems that when osteoblasts grow in a monolayer, the polarization is best achieved whereas some of it is lost when the cells start growing on top of each other. Very little is known about the polarity of other than epithelial cells. Mesenchymal osteoclasts have been shown to polarize its membrane into domains (Salo et al. 1996, Salo et al. 1997). Most of the osteoclast membrane has receptors for VSV indicating basolateral membrane domain, but the special cap structure on the top of an active osteoclast has receptors for Influenza virus, being thus apical in nature. The mechanism by which these non-epithelial cells organize their membrane domains and maintain the polarity is not known, since they do not have tight junctions that usually mediate the polarity in epithelial cells.

6.4 Gap junctions in osteoclasts

To establish and maintain the balance between bone formation and bone resorption, bone cells must have a way to communicate with each other. Most cells communicate with their immediate neighbors through exchange of cytosolic molecules, such as ions, second messengers, and small metabolites. This activity is made possible by clusters of intercellular channels called gap junctions, which connect adjacent cells (Kumar et al. 1996). Others have shown that of the bone cells osteoblasts and osteocytes express gap junctional connexins (Donahue et al. 1995b, Jones et al. 1993, Yellowley et al. 2000). Some reports of connexins in osteoclasts have also been published earlier (Jones et al. 1993, Jones et al. 1994, Su et al. 1997), but the functional significance has not been demonstrated earlier. It has been speculated that osteoblasts, osteocytes and osteoclasts may adjust the number of communicating gap junctions throughout the bone cell network. This way the cells could regulate the transmission of molecular and electrical information to support the balance of bone remodeling (Su et al. 1997). We clearly demonstrate that osteoclasts contain Cx43. The immunofluorescence data leaves no doubt of the Cx43 expression in osteoclasts. The Cx43 staining localizes mostly in the plasma membrane of the cells with some cytoplasmic staining and shows a staining pattern typical of a gap-junctional protein. Cx43 staining could be seen when osteoclasts were in cell-cell contacts with bone marrow derived mononuclear cells. Numerous punctuate connexin stainings could be seen in the basolateral plasma membrane of

50

osteoclasts, some of them being rather large. This finding indicates that osteoclasts are connected to the other bone cell types and thus take part in the remodeling of bone. Since the Cx43 staining was so intense in osteoclasts, we tested the effect of two gapjunctional inhibitors, heptanol and Gap 27, on osteoclasts with dramatic results. Great variety of drugs have been shown to inhibit gap-junctional communication with varying degrees of efficacy and specificity, including volatile anesthetics such as halothane (Nedergaard 1994), the straight-chain alcohols heptanol and octanol (Donahue et al. 1995a, Nedergaard 1994), anandamide (Venance et al. 1995), glycerrhetinic acid (Davidson et al. 1995) and Gap 27 (Chaytor et al. 1997, Chaytor et al. 1998). Of these we chose heptanol and Gap 27. The number of osteoclasts decreased markedly in the treated cultures. The results between the two inhibitors were very similar. When we further studied the reason for this decrease, we found to our surprise that the share of apoptotic osteoclasts was multiple to that of controls. It seems that the communication between osteoclasts and other bone marrow derived cells was essential for the survival of osteoclasts. The fact, that several osteoclasts undergo apoptosis as a result of blocking of the gap junctions, indicates the importance gap-junctional communication to the boneresorbing cells. None of the other cell types present in this bone cell culture model, excluding osteoclasts, are affected but their number is unchanged and they seem normal with healthy looking intact nuclei. When we further examined the inhibitor-treated osteoclast cultures we noticed that even though the numbers of both mononuclear and multinucleated TRAP-positive cells had reduced with the treatments, the number of TRAP-positive multinucleated cells was even more radically reduced. This could indicate that to be able to survive, multinucleated osteoclasts need contacts with the neighboring cells even more than their mononuclear precursors. Furthermore, the fusion of osteoclast precursors into mature multinucleated osteoclasts could somehow be gap junction dependent. This is by no means a far-fetched idea, since involvement of gap junctions has been demonstrated during the progress of macrophage multinucleation in vitro (Ejiri et al. 1987). In addition, Jones and Boyde (Jones et al. 1994) have also suggested the involvement of gap junctions in the fusion of osteoclast precursors to mature osteoclasts. Since the number of osteoclasts decreased significantly with the inhibitors, it was only expected that the resorption activity be decreased as well. The decrease in the resorbed area was in consistent with our expectations. We measured the activity of osteoclasts also by determining the activity of individual osteoclasts by the actin ring staining. Osteoclastic activity was found to be decreased by this method as well. The reduced activity of the remaining osteoclasts might suggest that even though some TRAP-positive mononuclear and multinucleated cells are found after Gap 27-treatment in these cultures, they may already be undergoing some changes leading eventually to apoptosis and thus inactivating the cells. Musil et al. (Musil et al. 1990) have shown that cells lacking morphologically and physiologically recognizable gap junctions can still express Cx43. This illustrates that connexin synthesis cannot automatically be taken as proof that a cell possesses functioning intercellular channels if it expresses connexin. Nevertheless, our results clearly show that inhibition of gap-junctional communication with either a commonly used gap-junctional inhibitor heptanol or with a more specific gap-junctional inhibitor Gap 27, causes osteoclastic apoptosis and hence a decrease in the bone resorption. It has

51

been suggested that the presence of Cx43 in gap junctions between osteoblasts may be a prerequisite for effective metabolic coupling between osteoblastic cells in bone cell networks. The precise role of intercellular communication in bone function is not clear, but gap junctions may modulate bone development and remodeling by coordinating the response of bone cell network to extracellular signals such as physical or hormonal signals (Palumbo et al. 1990). Indeed, it has been demonstrated that gap-junctional communication is critical for the functional responsiveness of osteoblastic network (Vander et al. 1996). Another possible role of gap-junctional intercellular communication in bone may be in the propagation of biophysical signals. A signal from cells that sense a change in mechanical load may be transmitted to other cells that do not sense the signal but are effectors of remodeling (Su et al. 1997).

7 Conclusions
Tight osteoclastic attachment to bone surface in the sealing zone is sensitive to the cadherin CAR sequence-containing peptide AHAVSE. The peptide-treatment decreased activity of osteoclasts seen as a decrease in the number of resorption pits and the total resorbed area. Staining of osteoclasts with a cadherin-recognizing antibody localized cadherin-like molecule in the sealing zone area of osteoclasts. The primary attachment of osteoclasts to bone surface is not affected by the peptide, suggesting that it is not mediated by cadherins. These results suggest that cadherin-like molecules may mediate the tight attachment of osteoclasts to the bone surface in the sealing zone and that the decrease of resorption in AHAVSE-treated osteoclast cultures is due to prevention of sealing zone formation. We demonstrate for the first time that mesenchymal osteoblastic cells have polarized plasma membrane domains. Our findings suggest that osteoblasts are polarized at some point of their life cycle. The bone attaching or bone matrix-facing plasma membrane of osteoblasts has apical characteristics, while the circulation or bone marrow-facing plasma membrane is basolateral in nature. The data from the bone cell communication studies suggest a functional role for gap junctions in osteoclastic bone resorption. We believe, that during bone remodeling, cell contacts and junctions play an important role in modulating osteoblast and osteoclast activity, recruiting osteoblasts that transform into osteocytes and forming an open canalicular network among osteocytes. Osteoclasts show connexin staining in cell-cell contacts, suggesting a possible communication mechanism to mononuclear cells. Osteoclastic activity is clearly reduced by gap junctional inhibitors and the number of osteoclasts also decreases with the treatment, which indicates the vital role of gapjunctional communication for the survival of osteoclasts. The proportional number of mononuclear TRAP-positive cells increases, which may mean that gap junctions are needed for the fusion of precursors to mature multinucleated osteoclasts. On the basis of these results we conclude that gap-junctional communication is essential for proper remodeling for bone and to the action of bone-resorbing osteoclasts.

8 References
Abe E, Tanaka H, Ishimi Y, Miyaura C, Hayashi T, Nagasawa H, Tomida M, Yamaguchi Y, Hozumi M & Suda T (1986) Differentiation-inducing factor purified from conditioned medium of mitogen-treated spleen cell cultures stimulates bone resorption. Proc Natl Acad Sci U S A 83: 5958- 5962. Akamine A, Tsukuba T, Kimura R, Maeda K, Tanaka Y, Kato K & Yamamoto K (1993) Increased synthesis and specific localization of a major lysosomal membrane sialoglycoprotein (LGP107) at the ruffled border membrane of active osteoclasts. Histochemistry 100: 101- 108. Akisaka T, Yoshida H, Kogaya Y, Kim S, Yamamoto M & Kataoka K (1990) Membrane modifications in chick osteoclasts revealed by freeze-fracture replicas. Am J Anat 188: 381392. Ali SY, Sajdera SW & Anderson HC (1970) Isolation and characterization of calcifying matrix vesicles from epiphyseal cartilage. Proc Natl Acad Sci U S A 67: 1513- 1520. Amprino R & Engstrom A (1952) Studies on x-ray absorption and diffraction of bone tissue. Acta Anat 15: 1- 22. Anderson HC (1984) Mineralization by matrix vesicles. Scan Electron Microsc 953- 964. Anderson HC (1989) Mechanism of mineral formation in bone. Lab Invest 60: 320- 330. Anderson HC (1995) Molecular biology of matrix vesicles. Clin Orthop 266- 280. Anderson JJ, Greenfield JW, Posada JR & Crackel WC (1970) Effect of estrogen on bone mineral turnover in mature female rats as measured by strontium-85. Proc Soc Exp Biol Med 135: 883886. Anderson RE, Woodbury DM & Jee WS (1986) Humoral and ionic regulation of osteoclast acidity. Calcified Tissue Int 39: 252- 258. Anderson TR & Toverud SU (1979) Purification and partial characterization of two acid phosphatases from rat bone. Calcified Tissue Int 27: 219- 226. Arends MJ & Wyllie AH (1991) Apoptosis: mechanisms and roles in pathology. Int Rev Exp Pathol 32: 223- 254. Arsenis C (1972) Role of mitochondria in calcification, Mitochondrial activity distribution in the epiphyseal plate and accumulation of calcium and phosphate ions by chondrocyte mitochondria. Biochem Biophys Res Commun 46: 1928- 1935. Ash P, Loutit JF & Townsend KM (1980) Osteoclasts derived from haematopoietic stem cells. Nature 283: 669- 670. Athanasou NA, Quinn J & McGee JO (1988) Immunocytochemical analysis of the human osteoclast: phenotypic relationship to other marrow-derived cells. Bone Miner 3: 317- 333. Aubin JE & Liu F (1996) The osteoblast lineage. In: Bilezikian JP, Raisz LG & Rodan GA (eds) Principles of Bone Biology. Academic Press, San Diego, p 51- 67. Babich M & Foti LR (1994) E-cadherins identified in osteoblastic cells: effects of parathyroid hormone and extracellular calcium on localization. Life Sci 54: PL201- 8.

54

Barnicot NA (1948) The local action of the parathyroid and other tissues on the bone in intacelebral grafts. J Anat 233- 248. Baron R (1977) Importance of the intermediate phases between resorption and formation in the measurement and understanding of the bone remodeling sequence. In: Meunier PJ (ed) Bone histomorphometry: proceedings of the 2nd international workshop. Socit de la Nouvelle Imprimerie Fournie, Toulouse, p 179- 183. Baron R (1989) Polarity and membrane transport in osteoclasts. Connect Tissue Res 20: 109- 120. Baron R, Neff L, Brown W, Courtoy PJ, Louvard D & Farquhar MG (1988) Polarized secretion of lysosomal enzymes: co-distribution of cation- independent mannose-6-phosphate receptors and lysosomal enzymes along the osteoclast exocytic pathway. J Cell Biol 106: 1863- 1872. Baron R, Neff L, Brown W, Louvard D & Courtoy PJ (1990) Selective internalization of the apical plasma membrane and rapid redistribution of lysosomal enzymes and mannose 6-phosphate receptors during osteoclast inactivation by calcitonin. J Cell Sci 97 (Pt 3): 439- 447. Baron R, Neff L, Louvard D & Courtoy PJ (1985) Cell-mediated extracellular acidification and bone resorption: evidence for a low pH in resorbing lacunae and localization of a 100-kD lysosomal membrane protein at the osteoclast ruffled border. J Cell Biol 101: 2210- 2222. Baron R, Vignery A & Horowitz M (1984) Lymphocytes, macrophages and the regulation of bone remodeling. In: Peck WA (ed) Bone and Mineral Research, Annual 2. Edition. Elsevier, Amsterdam, p 175- 243. Baylink D, Wergedal J & Thompson E (1972) Loss of proteinpolysaccharides at sites where bone mineralization is initiated. J Histochem Cytochem 20: 279- 292. Bekker PJ & Gay CV (1990) Biochemical characterization of an electrogenic vacuolar proton pump in purified chicken osteoclast plasma membrane vesicles. J Bone Miner Res 5: 569- 579. Bennett JH, Joyner CJ, Triffitt JT & Owen ME (1991) Adipocytic cells cultured from marrow have osteogenic potential. J Cell Sci 99 (Pt 1): 131- 139. Beyer EC, Paul DL & Goodenough DA (1990) Connexin family of gap junction proteins. J Membrane Biol 116: 187- 194. 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: 855- 857. Blair HC, Teitelbaum SL, Grosso LE, Lacey DL, Tan HL, McCourt DW & Jeffrey JJ (1993) Extracellular-matrix degradation at acid pH. Avian osteoclast acid collagenase isolation and characterization. Biochem J 290 (Pt 3): 873- 884. Blair HC, Teitelbaum SL, Tan HL, Koziol CM & Schlesinger PH (1991) Passive chloride permeability charge coupled to H(+)-ATPase of avian osteoclast ruffled membrane. Am J Physiol 260: C1315- C1324. Blaschuk OW, Pouliot Y & Holland PC (1990b) Identification of a conserved region common to cadherins and influenza strain A hemagglutinins. J Mol Biol 211: 679- 82. Blaschuk OW, Sullivan R, David S & Pouliot Y (1990a) Identification of a cadherin cell adhesion recognition sequence. Dev Biol 139: 227- 9. Boller K, Vestweber D & Kemler R (1985) Cell-adhesion molecule uvomorulin is localized in the intermediate junctions of adult intestinal epithelial cells. J Cell Biol 100: 327- 332. Bonewald LF (1999) Establishment and characterization of an osteocyte-like cell line, MLO- Y4. J Bone Miner Metab 17: 61- 65. Bonewald LF & Mundy GR (1990) Role of transforming growth factor-beta in bone remodeling. Clin Orthop Relat R 261- 276. Bonucci E (1971) The locus of initial calcification in cartilage and bone. Clin Orthop Relat R 78: 108- 139. Boyde A, Ali NN & Jones SJ (1984) Resorption of dentine by isolated osteclasts in vitro. Brit Dent J 156: 216- 220. Burger EH & Klein-Nulend J (1999) Mechanotransduction in bone--role of the lacuno-canalicular network. FASEB J 13 Suppl: S101- S112.

55

Byers S, Amaya E, Munro S & Blaschuk O (1992) Fibroblast growth factor receptors contain a conserved HAV region common to cadherins and influenza strain A hemagglutinins: a role in protein-protein interactions? Dev Biol 152: 411- 414. Canalis E, McCarthy T & Centrella M (1988) Isolation and characterization of insulin-like growth factor I (somatomedin-C) from cultures of fetal rat calvariae. Endocrinology 122: 22- 27. Cepek KL, Shaw SK, Parker CM, Russell GJ, Morrow JS, Rimm DL & Brenner MB (1994) Adhesion between epithelial cells and T lymphocytes mediated by E- cadherin and the alpha E beta 7 integrin. Nature 372: 190- 193. Cereijido M, Contreras RG & Gonzalez-Mariscal L (1989) Development and alteration of polarity. Annu Rev Physiol 51: 785- 795. Chambers TJ (1985) The pathobiology of the osteoclast. J Clin Pathol 38: 241- 252. Chambers TJ (1988) The regulation of osteoclastic development and function. Ciba Found Symp 136: 92- 107. Chambers TJ (1989) The origin of osteoclasts. In: Peck WA (ed) Bone and Mineral Research, Annual 6. Edition. Elsevier, Amsterdam, p 1- 25. Chambers TJ, Darby JA & Fuller K (1985) Mammalian collagenase predisposes bone surfaces to osteoclastic resorption. Cell Tissue Res 241: 671- 675. Chambers TJ, Fuller K, Darby JA, Pringle JA & Horton MA (1986) Monoclonal antibodies against osteoclasts inhibit bone resorption in vitro. Bone Miner 1: 127- 135. Chambers TJ & Hall TJ (1991) Cellular and molecular mechanisms in the regulation and function of osteoclasts. Vitam Horm 46: 41- 86. Chambers TJ, Revell PA, Fuller K & Athanasou NA (1984) Resorption of bone by isolated rabbit osteoclasts. J Cell Sci 66: 383- 399. Chaytor AT, Evans WH & Griffith TM (1997) Peptides homologous to extracellular loop motifs of connexin 43 reversibly abolish rhythmic contractile activity in rabbit arteries. J Physiol (Lond) 503 (Pt 1): 99- 110. Chaytor AT, Evans WH & Griffith TM (1998) Central role of heterocellular gap junctional communication in endothelium-dependent relaxations of rabbit arteries. J Physiol (Lond) 508 (Pt 2): 561- 573. Chenu C, Pfeilschifter J, Mundy GR & Roodman GD (1988) Transforming growth factor beta inhibits formation of osteoclast-like cells in long-term human marrow cultures. Proc Natl Acad Sci USA 85: 5683- 5687. Civitelli R, Beyer EC, Warlow PM, Robertson AJ, Geist ST & Steinberg TH (1993) Connexin43 mediates direct intercellular communication in human osteoblastic cell networks. J Clin Invest 91: 1888- 1896. Clezardin P, Malaval L, Ehrensperger AS, Delmas PD, Dechavanne M & McGregor JL (1988) Complex formation of human thrombospondin with osteonectin. Eur J Biochem 175: 275- 284. Cooper L, Yliheikkil P, Felton D & Whitson S (1998) Spatiotemporal assessment of Fetal Bovine Osteoblast Culture Differentiation Indicates a Role for BSP in Promoting Differentiation. J Bone Miner Res 13: 620- 632. Cowin P (1994) Unraveling the cytoplasmic interactions of the cadherin superfamily [see comments]. Proc Natl Acad Sci U S A 91: 10759- 10761. Craig AM, Jareb M & Banker G (1992) Neuronal polarity. Curr Opin Neurobiol 2: 602- 606. Davidson JS & Baumgarten IM (1995) Glycerrhetinic acid derivates: a novel class of inhibitors of gap-junctional intercellular communication. Structure-activity relationships. J Pharmacol Exp Ther 246: 1104- 1107. Davies J, Warwick J, Totty N, Philp R, Helfrich M & Horton M (1989) The osteoclast functional antigen, implicated in the regulation of bone resorption, is biochemically related to the vitronectin receptor. J Cell Biol 109: 1817- 1826. DeJong WF (1926) La substance minrale dans les os. Recl Trav Chim Pays-Bas Belg 45: 445448. Delaisse JM, Boyde A, Maconnachie E, Ali NN, Sear CH, Eeckhout Y, Vaes G & Jones SJ (1987) The effects of inhibitors of cysteine-proteinases and collagenase on the resorptive activity of isolated osteoclasts. Bone 8: 305- 313.

56

Delaisse JM, Eeckhout Y & Vaes G (1984) In vivo and in vitro evidence for the involvement of cysteine proteinases in bone resorption. Biochem Biophys Res Commun 125: 441- 447. Delaisse JM, Eeckhout Y & Vaes G (1988) Bone-resorbing agents affect the production and distribution of procollagenase as well as the activity of collagenase in bone tissue. Endocrinology 123: 264- 276. Donahue HJ, Guilak F, Vander MM, McLeod KJ, Rubin CT, Grande, DA & Brink PR (1995a) Chondrocytes isolated from mature articular cartilage retain the capacity to form functional gap junctions. J Bone Miner Res 10: 1359- 1364. Donahue HJ, McLeod KJ, Rubin CT, Andersen J, Grine EA, Hertzberg EL & Brink PR (1995b) Cell-to-cell communication in osteoblastic networks: cell line- dependent hormonal regulation of gap junction function. J Bone Miner Res 10: 881- 889. Doty SB (1981) Morphological evidence of gap junctions between bone cells. Calcified Tissue Int 33: 509- 512. Dresner-Pollak R & Rosenblatt M (1994) Blockade of osteoclast-mediated bone resorption through occupancy of the integrin receptor: a potential approach to the therapy of osteoporosis. J Cell Biochem 56: 323- 30. Duong LT & Rodan GA (1999) The role of integrins in osteoclast function. J Bone Miner Metab 17: 1- 6. Ejiri S, Segawa A, Miyaura C, Abe E, Suda T & Ozawa H (1987) An ultrastructural study on the multinucleation process of mouse alveolar macrophages induced by 1 alpha,25dihydroxyvitamin D3. J Bone Miner Res 2: 547- 557. Elster AD, Theros EG, Key LL & Chen MY (1992b) Cranial imaging in autosomal recessive osteopetrosis. Part I. Facial bones and calvarium. Radiology 183: 129- 135. Elster AD, Theros EG, Key LL & Chen MY (1992a) Cranial imaging in autosomal recessive osteopetrosis. Part II. Skull base and brain. Radiology 183: 137- 144. Engstrom GW & Deluca HF (1964) The nature of Ca++ binding by kidney mitochondria. Biochemistry 3: 379 Eriksen EF (1986) Normal and pathological remodeling of human trabecular bone: three dimensional reconstruction of the remodeling sequence in normals and in metabolic bone disease. Endocr Rev 7: 379- 408. Eyre DR, Dickson IR & Van Ness K (1988) Collagen cross-linking in human bone and articular cartilage. Age- related changes in the content of mature hydroxypyridinium residues. Biochem J 252: 495- 500. Feldman RS, Krieger NS & Tashjian AH, Jr. (1980) Effects of parathyroid hormone and calcitonin on osteoclast formation in vitro. Endocrinology 107: 1137- 1143. Ferrari S, Rizzoli R, Chevalley T, Slosman D, Eisman JA & Bonjour JP (1995) Vitamin-Dreceptor-gene polymorphisms and change in lumbar-spine bone mineral density. Lancet 345: 423- 424. Friedman J, Au W & Raisz LG (1986) Responses of fetal rat bone to thyrocalcitonin in tissue culture. Endocrinology 82: 149- 156. Frost HM (1964) Dynamics of bone remodeling. In: Bone biodynamics. Little & Brown, Boston Frost HM (1973) In: Bone remodeling and its relationship to metabolic bone disease. Charles C Thomas, Springfield, IL Frost HM (1986) In: Intermediary organization of the skeleton. CRC Press, Boca Raton, FL Gay CV, Ito MB & Schraer H (1983) Carbonic anhydrase activity in isolated osteoclasts. Metab Bone Dis Relat Res 5: 33- 39. Gay CV & Mueller WJ (1974) Carbonic anhydrase and osteoclasts: localization by labeled inhibitor autoradiography. Science 183: 432- 434. Gehron RP (1989) The biochemistry of bone. Endocrinol Metab Clin North Am 18: 858- 902. Geiger B & Ayalon O (1992) Cadherins. Annu Rev Cell Biol 8: 307- 32. Geiger B, Volberg T, Ginsberg D, Bitzur S, Sabanay I & Hynes RO (1990) Broad spectrum pancadherin antibodies, reactive with the C- terminal 24 amino acid residues of N-cadherin. J Cell Sci 97: 607- 14. Glowacki J & Lian JB (1987) Impaired recruitment and differentiation of osteoclast progenitors by osteocalcin-deplete bone implants. Cell Differ 21: 247- 254.

57

Goodenough DA, Goliger JA & Paul DL (1996) Connexins, connexons, and intercellular communication. Annu Rev Biochem 65: 475- 502. Gorski JP (1998) Is all bone the same? Distinctive distributions and properties of non- collagenous matrix proteins in lamellar vs. woven bone imply the existence of different underlying osteogenic mechanisms. Crit Rev Oral Biol Med 9: 201- 223. Gowen M, Meikle MC & Reynolds JJ (1983a) Stimulation of bone resorption in vitro by a nonprostanoid factor released by human monocytes in culture. Biochim Biophys Acta 762: 471474. Gowen M & Mundy GR (1986) Actions of recombinant interleukin 1, interleukin 2, and interferon- gamma on bone resorption in vitro. J Immunol 136: 2478- 2482. Gowen M, Wood DD, Ihrie EJ, McGuire MK & Russell RG (1983b) An interleukin 1 like factor stimulates bone resorption in vitro. Nature 306: 378- 380. 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. Grunwald GB (1993) The structural and functional analysis of cadherin calcium- dependent cell adhesion molecules. Curr Opin Cell Biol 5: 797- 805. Grunwald GB, Bromberg RE, Crowley NJ & Lilien J (1981) Enzymatic dissection of embryonic cell adhesive mechanisms. II. Developmental regulation of an endogenous adhesive system in the chick neural retina. Dev Biol 86: 327- 338. Gthling G & Ericsson JLE (1976) The osteoclast. Review of the ultrastructure, origin, and structure-function relationship. Clin Orthop Relat R 120: 201- 228. Hagenaars CE, van der Kraan AA, Kawilarang-de Haas EW, Visser JW & Nijweide PJ (1989) Osteoclast formation from cloned pluripotent hemopoietic stem cells. Bone Miner 6: 179- 189. Halleen JM, Risnen S, Salo JJ, Reddy SV, Roodman GD, Hentunen TA, Lehenkari PP, Kaija H, Vihko P & Vnnen HK (1999) Intracellular fragmentation of bone resorption products by reactive oxygen species generated by osteoclastic tartrate-resistant acid phosphatase. J Biol Chem 274: 22907- 22910. Handler JS (1989) Overview of epithelial polarity. Annu Rev Physiol 51: 729- 740. Hare JF & Lee E (1989) Metabolic behavior of cell surface biotinylated proteins. Biochemistry 28: 574- 580. Hauschka PV, Chen TL & Mavrakos AE (1988) Polypeptide growth factors in bone matrix. Ciba F Symp 136: 207- 225. Hauschka PV, Mavrakos AE, Iafrati MD, Doleman SE & Klagsbrun M (1986) Growth factors in bone matrix. Isolation of multiple types by affinity chromatography on heparin-Sepharose. J Biol Chem 261: 12665- 12674. Helfrich MH, Nesbitt SA, Dorey EL & Horton MA (1992) Rat osteoclasts adhere to a wide range of RGD (Arg-Gly-Asp) peptide-containing proteins, including the bone sialoproteins and fibronectin, via a beta 3 integrin. J Bone Miner Res 7: 335- 43. Holtrop ME & King GJ (1977) The ultrastructure of the osteoclast and its functional implications. Clin Orthop 177- 196. Holtrop ME, Raisz LG & Simmons HA (1974) The effects of parathyroid hormone, colchicine, and calcitonin on the ultrastructure and the activity of osteoclasts in organ culture. J Cell Biol 60: 346- 355. Horton M (1990) Vitronectin receptor: tissue specific expression or adaptation to culture? Int J Exp Pathol 71: 741- 759. Horton MA (1997) The alpha v beta 3 integrin "vitronectin receptor". Int J Biochem Cell Biol 29: 721- 725. Horton MA & Davies J (1989) Perspectives: adhesion receptors in bone. J Bone Miner Res 4: 803808. Horton MA, Taylor ML, Arnett TR & Helfrich MH (1991) Arg-Gly-Asp (RGD) peptides and the anti-vitronectin receptor antibody 23C6 inhibit dentine resorption and cell spreading by osteoclasts. Exp Cell Res 195: 368- 75. Irving JT (1958) A histological stain for newly calcified tissues. Nature (London) 181: 704 Irving JT (1973) Theories of mineralization of bone. Clin Orthop 97: 225- 236.

58

Jayo MJ, Weaver DS, Adams MR & Rankin SE (1990) Effects on bone of surgical menopause and estrogen therapy with or without progesterone replacement in cynomolgus monkeys. Am J Obstet Gynecol 163: 614- 618. Jeansonne BG, Feagin FF, McMinn RW, Shoemaker RL & Rehm WS (1979) Cell-to-cell communication of osteoblasts. J Dent Res 58: 1415- 1423. Jones SJ (1974) Secretory territories and rate of matrix production of osteoblasts. Calcified Tissue Res 14: 309- 315. Jones SJ & Boyde A (1994) Questions of quality and quantity--a morphological view of bone biology. Kaibogaku Zasshi - J Anat 69: 229- 243. Jones SJ, Boyde A & Ali NN (1986) The interface of cells and their matrices in mineralized tissues: a review. Scanning Electron Microscopy 1555-1569. Jones SJ, Gray C, Sakamaki H, Arora M, Boyde A, Gourdie R & Green C (1993) The incidence and size of gap junctions between the bone cells in rat calvaria. Anat Embryol 187: 343- 352. Kahn AJ & Simmons DJ (1975) Investigation of cell lineage in bone using a chimaera of chick and quial embryonic tissue. Nature 258: 325- 327. Kalu DN, Salerno E, Liu CC, Echon R, Ray M, Garza-Zapata M & Hollis BW (1991) A comparative study of the actions of tamoxifen, estrogen and progesterone in the ovariectomized rat. Bone Miner 15: 109- 123. Kanehisa J & Heersche JN (1988) Osteoclastic bone resorption: in vitro analysis of the rate of resorption and migration of individual osteoclasts. Bone 9: 73- 79. Karhukorpi EK (1991) Carbonic anhydrase II in rat acid secreting cells: comparison of osteoclasts with gastric parietal cells and kidney intercalated cells. Acta Histochem 90: 11- 20. Kelly PJ, Hopper JL, Macaskill GT, Pocock NA, Sambrook PN & Eisman JA (1991) Genetic factors in bone turnover. J Clin Endocrinol Metab 72: 808- 813. Kemler R (1992) Classical cadherins. Semin Cell Biol 3: 149- 155. Kerby JA, Hattersley G, Collins DA & Chambers TJ (1992) Derivation of osteoclasts from hematopoietic colony-forming cells in culture. J Bone Miner Res 7: 353- 62. King GJ & Holtrop ME (1975) Actin-like filaments in bone cells of cultured mouse calvaria as demonstrated by binding to heavy meromyosin. J Cell Biol 66: 445- 451. Knudsen KA & Wheelock MJ (1992) Plakoglobin, or an 83-kD homologue distinct from betacatenin, interacts with E-cadherin and N-cadherin. J Cell Biol 118: 671- 9. Knust E (1994) Control of epithelial cell polarity in Drosophila. Trends Genet 10: 275- 280. Koutsilieris M, Rabbani SA & Goltzman D (1987) Effects of human prostatic mitogens on rat bone cells and fibroblasts. J Endocrinol 115: 447- 454. Kumar MA (1963) Further evidence for calcitonin - a rapid activating hormone which lowers plasma calcium. Lancet 2: 480- 492. Kumar NM & Gilula NB (1996) The gap junction communication channel. Cell 84: 381- 388. Klliker A (1873) In: Die normale Resorption des Knochengewebes und ihre Bedeutung fur die Entstehung der typischen Knochenformen. F C W Vogel, Leipzig. 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: 165- 176. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680- 685. Laitala T & Vnnen HK (1994) Inhibition of bone resorption in vitro by antisense RNA and DNA molecules targeted against carbonic anhydrase II or two subunits of vacuolar H(+)ATPase. J Clin Invest 93: 2311- 238. Lakkakorpi P, Tuukkanen J, Hentunen T, Jrvelin K & Vnnen K (1989) Organization of osteoclast microfilaments during the attachment to bone surface in vitro. J Bone Miner Res 4: 817- 25. Lakkakorpi PT, Helfrich MH, Horton MA & Vnnen HK (1993) Spatial organization of microfilaments and vitronectin receptor, alpha v beta 3, in osteoclasts. A study using confocal laser scanning microscopy. J Cell Sci 104: 663- 70.

59

Lakkakorpi PT, Horton MA, Helfrich MH, Karhukorpi EK & Vnnen HK (1991b) Vitronectin receptor has a role in bone resorption but does not mediate tight sealing zone attachment of osteoclasts to the bone surface. J Cell Biol 115: 1179- 186. Lakkakorpi PT & Vnnen HK (1991a) Kinetics of the osteoclast cytoskeleton during the resorption cycle in vitro. J Bone Miner Res 6: 817- 26. Lamberg-Allardt C (1991) Is there a role for vitamin D in osteoporosis? Calcified Tissue Int 49 Suppl: S46- S49. Landers JP & Spelsberg TC (1992) New concepts in steroid hormone action: transcription factors, proto- oncogenes, and the cascade model for steroid regulation of gene expression. Crit Rev Eukaryot Gene Expr 2: 19- 63. Lane JM (1979) Biochemistry of fracture repair. AAOS Monterey Seminar Am Acad, Orthop Surg , Chicago 141- 165. Levy M (1894) Chemische untersuchungen uber osteomalacische knochen. Hoppe-Seyler's Z Physiol Chem 19: 239- 270. Lindsay R, Aitken JM, Hart DM & Purdie D (1978) The effect of ovarian sex steroids on bone mineral status in the oophorectomized rat and in the human. Postgrad Med J 54 Suppl 2: 50- 58. Lindsay R, Hart DM, Forrest C & Baird C (1980) Prevention of spinal osteoporosis in oophorectomised women. Lancet 2: 1151- 1154. Lisanti MP, Sargiacomo M, Graeve L, Saltiel AR & Rodriguez-Boulan E (1988) Polarized apical distribution of glycosyl-phosphatidylinositol- anchored proteins in a renal epithelial cell line. Proc Nat Acad Sci USA 85: 9557- 9561. Malkani K, Luxembourger MM & Rebel A (1973) Cytoplasmic modifications at the contact zone of osteoclasts and calcified tissue in the diaphyseal growing plate of foetal guinea-pig tibia. Calcified Tissue Res 11: 258- 264. Marcus R (1987) Normal and abnormal bone remodeling in man. Annu Rev Med 38: 129- 141. Marks SC, Jr. (1987) Osteopetrosis--multiple pathways for the interception of osteoclast function. Appl Pathol 5: 172- 183. Marks SC, Jr. & Hermey DC (1996) The Structure and Development of Bone. In: Bilezikian JP, Raisz LG & Rodan GA (eds) Principles in Bone Biology. Academic Press, San Diego, p 3- 14. Martin K & Helenius A (1991) Nuclear transport of influenza virus ribonucleoproteins: the viral matrix protein (M1) promotes export and inhibits import. Cell 67: 117- 130. Masarachia P, Yamamoto M, Leu CT, Rodan G & Duong L (1998) Histomorphometric evidence for echistatin inhibition of bone resorption in mice with secondary hyperparathyroidism. Endocrinology 139: 1401- 1410. Matkovic V, Jelic T, Wardlaw GM, Ilich JZ, Goel PK, Wright JK, Andon MB, Smith KT & Heaney RP (1994) Timing of peak bone mass in Caucasian females and its implication for the prevention of osteoporosis. Inference from a cross-sectional model. J Clin Invest 93: 799- 808. Matsumoto T, Kawanobe Y, Morita K & Ogata E (1985) Effect of 1,25-dihydroxyvitamin D3 on phospholipid metabolism in a clonal osteoblast-like rat osteogenic sarcoma cell line. J Biol Chem 260: 13704- 13709. Matter K & Mellman I (1994) Mechanisms of cell polarity: sorting and transport in epithelial cells. Curr Opin Cell Biol 6: 545- 554. Mbalaviele G, Chen H, Boyce BF, Mundy GR & Yoneda T (1995) The role of cadherin in the generation of multinucleated osteoclasts from mononuclear precursors in murine marrow. J Clin Invest 95: 2757- 65. Mbalaviele G, Nishimura R, Myoi A, Niewolna M, Reddy SV, Chen D, Feng J, Roodman D, Mundy GR & Yoneda T (1998) Cadherin-6 mediates the heterotypic interactions between the hemopoietic osteoclast cell lineage and stromal cells in a murine model of osteoclast differentiation. J Cell Biol 141: 1467- 1476. Mbalaviele G, Nishimura R, Williams P, Dallas M, Mundy GR & Yoneda T (1996) The role of Ecadherin-catenin interaction in cell-matrix adhesion and breast cancer metastasis to bone. J Bone Miner Res 11 (Suppl. 1): S163. Mege RM, Goudou D, Diaz C, Nicolet M, Garcia L, Geraud G & Rieger F (1992) N-cadherin and N-CAM in myoblast fusion: compared localisation and effect of blockade by peptides and antibodies. J Cell Sci 103 (Pt 4): 897- 906.

60

Menton DN, Simmons DJ, Chang SL & Orr BY (1984) From bone lining cell to osteocyte--an SEM study. Anat Rec 209: 29- 39. Metsikk K, Hentunen T & Vnnen K (1992) Local expression and exocytosis of viral glycoproteins in multinucleated muscle cells. J Cell Biol 117: 987- 995. Mikuni-Takagaki Y (1999) Mechanical responses and signal transduction pathways in stretched osteocytes. J Bone Miner Metab 17: 57- 60. Miller SC (1978) Rapid activation of the medullary bone osteoclast cell surface by parathyroid hormone. J Cell Biol 76: 615- 618. Minkin C (1982) Bone acid phosphatase: tartrate-resistant acid phosphatase as a marker of osteoclast function. Calcified Tissue Int 34: 285- 290. Minkin C & Jennings JM (1972) Carbonic anhydrase and bone remodeling: sulfonamide inhibition of bone resorption in organ culture. Science 176: 1031- 1033. Mundy GR (1990) Bone resorbing cells. In: Favus MJ (ed) Primer on the metabolic bone diseases and disorders of mineral metabolism. American Society of Bone and Mineral Research, Kelseyville, p 18- 22. Mundy GR (1995) In: Bone remodeling and its disorders. Martin Dunitz, London. Mundy GR & Roodman GD (1987) Osteoclast ontogeny and function. In: Peck WA (ed) Bone and Mineral Research. Elsevier, Amsterdam, p 209- 279. Musil LS, Cunningham BA, Edelman GM & Goodenough DA (1990) Differential phosphorylation of the gap junction protein connexin43 in junctional communication-competent and -deficient cell lines. J Cell Biol 111: 2077- 2088. Nakamura I, Gailit J & Sasaki T (1996) Osteoclast integrin alphaVbeta3 is present in the clear zone and contributes to cellular polarization. Cell Tissue Res 286: 507- 515. Narbaitz R, Stumpf WE, Sar M, Huang S & Deluca HF (1983) Autoradiographic localization of target cells for 1 alpha, 25- dihydroxyvitamin D3 in bones from fetal rats. Calcified Tissue Int 35: 177- 182. Nedergaard M (1994) Direct signaling from astrocytes to neurons in cultures of mammalian brain cells. Science 263: 1768- 1771. Nelson WJ (1993) Regulation of cell surface polarity in renal epithelia. Pediatr Nephrol 7: 599604. Nesbitt S, Nesbit A, Helfrich M & Horton M (1993) Biochemical characterization of human osteoclast integrins. Osteoclasts express alpha v beta 3, alpha 2 beta 1, and alpha v beta 1 integrins. J Biol Chem 268: 16737- 1645. Nesbitt SA & Horton MA (1997) Trafficking of matrix collagens through bone-resorbing osteoclasts. Science 276: 266- 269. Odell WD & Heath H3d (1993) Osteoporosis: pathophysiology, prevention, diagnosis, and treatment. Disease-A-Month 39: 789- 867. Okazaki M, Takeshita S, Kawai S, Kikuno R, Tsujimura A, Kudo A & Amann E (1994) Molecular cloning and characterization of OB-cadherin, a new member of cadherin family expressed in osteoblasts. J Biol Chem 269: 12092- 12098. Oldberg A, Franzen A & Heinegard D (1986) Cloning and sequence analysis of rat bone sialoprotein (osteopontin) cDNA reveals an Arg-Gly-Asp cell-binding sequence. Proc Natl Acad Sci USA 83: 8819- 8823. Ozawa M, Ringwald M & Kemler R (1990) Uvomorulin-catenin complex formation is regulated by a specific domain in the cytoplasmic region of the cell adhesion molecule. Proc Natl Acad Sci USA 87: 4246- 50. Palokangas H, Mulari M & Vnnen HK (1997) Endocytic pathway from the basal plasma membrane to the ruffled border membrane in bone-resorbing osteoclasts. J Cell Sci 110 (Pt 15): 1767- 1780. Palumbo C & Rodan GA (1990) Morphological study of intercellular junctions during osteocyte differentiation. Bone 11: 401- 406. Parfitt AM (1983) The physiological and clinical significance of bone histomorphometric data. In: Recker RR (ed) Bone histomorphometry: techniques and interpretation. CRC Press, Boca Raton, FL, p 143- 223.

61

Parfitt AM (1994) Osteonal and hemi-osteonal remodeling: the spatial and temporal framework for signal traffic in adult human bone. J Cell Biochem 55: 273- 86. Partridge NC, Walling HW, Bloch SR, Omura TH, Chan PT, Pearman AT & Chou WY (1996) The regulation and regulatory role of collagenase in bone. Crit Rev Eukaryotic Gene Expr 6: 1527. Pocock NA, Eisman JA, Hopper JL, Yeates MG, Sambrook PN & Eberl S (1987) Genetic determinants of bone mass in adults. A twin study. J Clin Invest 80: 706- 710. Price PA, Parthemore JG & Deftos LJ (1980) New biochemical marker for bone metabolism. Measurement by radioimmunoassay of bone GLA protein in the plasma of normal subjects and patients with bone disease. J Clin Invest 66: 878- 883. Raisz LG (1963) Stimulation of bone resorption by parathyroid hormone in tissue culture. Nature 197: 1015- 1016. Raisz LG (1988) Local and systemic factors in the pathogenesis of osteoporosis. N Engl J Med 318: 818- 828. Raisz LG, Trummel CL, Holick MF & Deluca HF (1972) 1,25-dihydroxycholecalciferol: a potent stimulator of bone resorption in tissue culture. Science 175: 768- 769. Ralston SH, Gardner MD, Dryburgh FJ, Jenkins AS, Cowan RA & Boyle IT (1985) Comparison of aminohydroxypropylidene diphosphonate, mithramycin, and corticosteroids/calcitonin in treatment of cancer-associated hypercalcaemia. Lancet 2: 907- 910. Reinholt FP, Hultenby K, Oldberg A & Heinegard D (1990) Osteopontin--a possible anchor of osteoclasts to bone. Proc Natl Acad Sci USA 87: 4473- 445. Rifkin BR, Vernillo AT, Kleckner AP, Auszmann JM, Rosenberg LR & Zimmerman M (1991) Cathepsin B and L activities in isolated osteoclasts. Biochem Biophys Res Commun 179: 6369. Ringbom-Anderson T, Jantti J & Akerman KE (1994) Production and release of matrix vesicles in the cell processes of TPA- treated human osteoblast-like cells. J Bone Mineral Res 9: 661- 670. Ripamonti U & Reddi AH (1992) Growth and morphogenetic factors in bone induction: role of osteogenin and related bone morphogenetic proteins in craniofacial and periodontal bone repair. Crit Rev Oral Biol Med 3: 1- 14. Robey PG, Young MF, Fisher LW & McClain TD (1989) Thrombospondin is an osteoblastderived component of mineralized extracellular matrix. J Cell Biol 108: 719- 727. Robey PG, Young MF, Flanders KC, Roche NS, Kondaiah P, Reddi AH, Termine JD, Sporn MB & Roberts AB (1987) Osteoblasts synthesize and respond to transforming growth factor-type beta (TGF-beta) in vitro. J Cell Biol 105: 457- 463. Rodan GA & Martin TJ (2000) Therapeutic approaches to bone diseases. Science 289: 1508- 1514. Rodriguez-Boulan E & Pendergast M (1980) Polarized Distribution of Viral Envelope Proteins in the Plasma Membrane of Infected Epithelial Cells. Cell 20: 45- 54. Rodriguez-Boulan E & Powell SK (1992) Polarity of Epithelial and Neuronal Cells. Annu Rev Cell Biol 8: 395- 427. Rook MB, de JB, Jongsma HJ & Masson-Pevet MA (1990) Gap junction formation and functional interaction between neonatal rat cardiocytes in culture: a correlative physiological and ultrastructural study. J Membrane Biol 118: 179- 192. Ruoslahti E & Pierschbacher MD (1986) Arg-Gly-Asp: a versatile cell recognition signal. Cell 44: 517- 518. Ruoslahti E & Pierschbacher MD (1987) New perspectives in cell adhesion: RGD and integrins. Science 238: 491- 497. Saez JC, Berthoud VM, Moreno AP & Spray DC (1993) Gap junctions. Multiplicity of controls in differentiated and undifferentiated cells and possible functional implications. Adv Sec Mess Phosph Res 27: 163- 198. Salo J, Lehenkari P, Mulari M, Metsikk K & Vnnen HK (1997) Removal of osteoclast bone resorption products by transcytosis. Science 276: 270- 273. Salo J, Metsikk K, Palokangas H, Lehenkari P & Vnnen HK (1996) Bone-resorbing osteoclasts reveal a dynamic division of basal plasma membrane into two different domains. J Cell Sci 109: 301- 307.

62

Sato M, Sardana MK, Grasser WA, Garsky VM, Murray JM & Gould RJ (1990) Echistatin is a potent inhibitor of bone resorption in culture. J Cell Biol 111: 1713- 123. Schenk RK, Spiro D & Wiener J (1967) Cartilage resorption in the tibial epiphyseal plate of growing rats. J Cell Biol 34: 275- 291. Schwartz Z, Amir D, Weinberg H & Sela J (1987) Extracellular matrix vesicle distribution in primary mineralization two weeks after injury to rat tibial bone (ultrastructural tissue morphometry). Eur J Cell Biol 45: 97- 101. Selander K, Lehenkari P & Vnnen HK (1994) The effects of bisphosphonates on the resorption cycle of isolated osteoclasts. Calcified Tissue Int 55: 368- 75. Shapiro IM & Greenspan JS (1969) Are mitochondria directly involved in biological mineralisation? Calcified Tissue Res 3: 100- 102. Shuttleworth A & Veis A (1972) The isolation of anionic phosphoproteins from bovine cortical bone via the periodate solubilization of bone collagen. Biochim Biophys Acta 257: 414- 420. Silve CH (1982) Parathyroid hormone receptor in intact embryonic chicken bone: Characterization and cellular localization. J Cell Biol 94: 379- 386. Simons K & van Meer G (1988) Lipid sorting in epithelial cells. Biochemistry 27: 6197- 6202. Simons K & Wandinger-Ness A (1990) Polarized sorting in epithelia. Cell 62: 207- 210. Smith DM, Khairi MR & Johnston CC, Jr. (1975) The loss of bone mineral with aging and its relationship to risk of fracture. J Clin Invest 56: 311- 318. Somerman MJ, Fisher LW, Foster RA & Sauk JJ (1988) Human bone sialoprotein I and II enhance fibroblast attachment in vitro. Calcified Tissue Int 43: 50- 53. Spector AR & Glimcher MJ (1972) The extraction and characterization of soluble anionic phosphoproteins from bone. Biochim Biophys Acta 263: 593Stanka P (1975) Occurrence of cell junctions and microfilaments in osteoblasts. Cell Tissue Res 159: 413- 422. Stein GS & Lian JB (1993) Molecular mechanisms mediating proliferation/differentiation interrelationships during progressive development of the osteoblast phenotype. Endocr Rev 14: 424- 442. Su M, Borke JL, Donahue HJ, Li Z, Warshawsky NM, Russell CM, Lewis & JE (1997) Expression of connexin 43 in rat mandibular bone and periodontal ligament (PDL) cells during experimental tooth movement. J Dent Res 76: 1357- 1366. Suda T, Takahashi N & Martin TJ (1992) Modulation of osteoclast differentiation [published erratum appears in Endocr Rev 1992 May;13(2):191]. Endocr Rev 13: 66- 80. Suda T, Udagawa N, Nakamura I, Miyaura C & Takahashi N (1995) Modulation of osteoclast differentiation by local factors. Bone 17: 87S- 91S. Sundquist K, Lakkakorpi P, Wallmark B & Vnnen K (1990) Inhibition of osteoclast proton transport by bafilomycin A1 abolishes bone resorption. Biochem Biophys Res Commun 168: 309- 13. Takahashi N, MacDonald BR, Hon J, Winkler ME, Derynck R, Mundy GR & Roodman GD (1986) Recombinant human transforming growth factor-alpha stimulates the formation of osteoclast-like cells in long-term human marrow cultures. J Clin Invest 78: 894- 898. Takano-Yamamoto T & Rodan GA (1990) Direct effects of 17 beta-estradiol on trabecular bone in ovariectomized rats. Proc Natl Acad Sci USA 87: 2172- 2176. Takeichi M (1977) Functional correlation between cell adhesive properties and some cell surface proteins. J Cell Biol 75: 464- 474. Takeichi M (1991) Cadherin cell adhesion receptors as a morphogenetic regulator. Science 251: 1451- 5. Takeichi M, Hatta K, Nose A & Nagafuchi A (1988) Identification of a gene family of cadherin cell adhesion molecules. Cell Differ Dev 25 Suppl: 91- 4. Tanaka H, Sato M, Shinar D & Rodan GA (1991) Echistatin inhibits osteoclast-like cell generation in co-culture system of mouse osteoblastic cells and bone marrow cells. J Bone Miner Res 6: S148 Tanaka Y & Deluca HF (1971) Bone mineral mobilization activity of 1,25dihydroxycholecalciferol, a metabolite of vitamin D. Arch Biochem Biophys 146: 574- 578.

63

Tang A, Amagai M, Granger LG, Stanley JR & Udey MC (1993) Adhesion of epidermal Langerhans cells to keratinocytes mediated by E- cadherin. Nature 361: 82- 85. Taterossian A (1973) Effect of parathyroid extract on blood calcium and osteoclast count in mice. Calcified Tissue Int 11: 251- 257. Termine JD (1993) Bone Matrix Proteins and the Mineralization Process. 2: 21- 25. Termine JD, Eanes ED & Conn KM (1980) Phosphoprotein modulation of apatite crystallization. Calcified Tissue Int 31: 247- 251. Termine JD, Kleinman HK, Whitson SW, Conn KM, McGarvey ML & Martin GR (1981) Osteonectin, a bone-specific protein linking mineral to collagen. Cell 26: 99- 105. Thomson BM, Mundy GR & Chambers TJ (1987) Tumor necrosis factors alpha and beta induce osteoblastic cells to stimulate osteoclastic bone resorption. J Immunol 138: 775- 779. Tran VP, Vignery A & Baron R (1982a) An electron-microscopic study of the bone-remodeling sequence in the rat. Cell Tissue Res 225: 283- 292. Tran VP, Vignery A & Baron R (1982b) Cellular kinetics of the bone remodeling sequence in the rat. Anat Rec 202: 445- 451. Tsuda E, Goto M, Mochizuki S, Yano K, Kobayashi F, Morinaga T & Higashio K (1997) Isolation of a novel cytokine from human fibroblasts that specifically inhibits osteoclastogenesis. Biochem Biophys Res Co 234: 137- 142. Turner RT, Wakley GK, Hannon KS & Bell NH (1988) Tamoxifen inhibits osteoclast-mediated resorption of trabecular bone in ovarian hormone-deficient rats. Endocrinology 122: 11461150. Tuukkanen J, Koivukangas A, Jamsa 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: 1905- 1911. Tuukkanen J, Peng Z & Vnnen HK (1991) Effect of exercise on osteoporosis induced by ovariectomy in rats. Calcified Tissue Int 49 Suppl: S80. Uebelhart D, Gineyts E, Chapuy MC & Delmas PD (1990) Urinary excretion of pyridinium crosslinks: a new marker of bone resorption in metabolic bone disease. Bone Miner 8: 87- 96. Umbas R, Schalken JA, Aalders TW, Carter BS, Karthaus HF, Schaafsma HE, Debruyne FM & Isaacs WB (1992) Expression of the cellular adhesion molecule E-cadherin is reduced or absent in high-grade prostate cancer. Cancer Res 52: 5104- 5109. van Meer G & Simons K (1988) Lipid polarity and sorting in epithelial cells. J Cell Biochem 36: 51- 58. Vander MM, Rubin CT, McLeod KJ, McCauley LK & Donahue HJ (1996) Gap junctional intercellular communication contributes to hormonal responsiveness in osteoblastic networks. J Biol Chem 271: 12165- 12171. Venance L, Piomelli D, Glowinski J & Giaume C (1995) Inhibition by anandamide of gap junctions and intercellular calcium signalling in striatal astrocytes. Nature 376: 590- 594. Viitanen A, Krkkinen M, Laitinen K, Lamberg-Allardt C, Kainulainen K, Rsnen L, Viikari J, Vlimki MJ & Kontula K (1996) Common polymorphism of the vitamin D receptor gene is associated with variation of peak bone mass in young finns. Calcified Tissue Int 59: 231- 234. Vlimki MJ, Krkkinen M, Lamberg-Allardt C, Laitinen K, Alhava E, Heikkinen J, Impivaara O, Mkel P, Palmgren J & Seppnen R (1994) Exercise, smoking, and calcium intake during adolescence and early adulthood as determinants of peak bone mass. Cardiovascular Risk in Young Finns Study Group. BMJ 309: 230- 235. Vnnen HK (1993) Mechanism of bone turnover. Ann Med 25: 353- 9. Vnnen HK & Horton M (1995) The osteoclast clear zone is a specialized cell-extracellular matrix adhesion structure. J Cell Sci 108 ( Pt 8): 2729- 2732. Vnnen 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: 1305- 111. Vnnen HK & Parvinen EK (1983) High active isoenzyme of carbonic anhydrase in rat calvaria osteoclasts. Immunohistochemical study. Histochemistry 78: 481- 485. Walker DG (1973b) Nicolas Andry award for 1973. Experimental osteopetrosis. Clin Orthop 97: 158- 174.

64

Walker DG (1973a) Osteopetrosis cured by temporary parabiosis. Science 180: 875Walker DG (1975a) Bone resorption restored in osteopetrotic mice by transplants of normal bone marrow and spleen cells. Science 190: 784- 785. Walker DG (1975b) Spleen cells transmit osteopetrosis in mice. Science 190: 785- 787. Warner A (1988) The gap junction. J Cell Sci 89: 1- 7. Warner AE, Guthrie SC & Gilula NB (1984) Antibodies to gap-junctional protein selectively disrupt junctional communication in the early amphibian embryo. Nature 311: 127- 131. Warshafsky B, Aubin JE & Heersche JN (1985) Cytoskeleton rearrangements during calcitonininduced changes in osteoclast motility in vitro. Bone 6: 179- 185. Weiner S & Traub W (1986) Organization of hydroxyapatite crystals within collagen fibrils. FEBS Lett 206: 262- 266. Weinger JM & Holtrop ME (1974) An ultrastructural study of bone cells: the occurrence of microtubules, microfilaments and tight junctions. Calcified Tissue Res 14: 15- 29. Weisbrode SE, Capen CC & Norman AW (1978) Ultrastructural evaluation of the effects of 1,25dihydroxyvitamin D3 on bone of thyroparathyroidectomized rats fed a low-calcium diet. Am J Pathol 92: 459- 472. Wennberg C, Hessle L, Lundberg P, Mauro S, Narisawa S, Lerner UH & Millan JL (2000) Functional characterization of osteoblasts and osteoclasts from alkaline phosphatase knockout mice. J Bone Miner Res 15: 1879- 1888. Wiktor-Jedrzejczak W, Bartocci A, Ferrante AW, Jr., Ahmed-Ansari A, Sell KW, Pollard JW & Stanley ER (1990) Total absence of colony-stimulating factor 1 in the macrophage- deficient osteopetrotic (op/op) mouse [published erratum appears in Proc Natl Acad Sci USA 1991 Jul 1;88(13):5937]. Proc Natl Acad Sci USA 87: 4828- 4832. Williams DC & Frolik CA (1991) Physiological and pharmacological regulation of biological calcification. Int Rev Cytol 126: 195- 292. Wong BR, Josien R, Lee SY, Sauter B, Li HL, Steinman RM & Choi Y (1997a) TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a dendritic cell-specific survival factor. J Exp Med 186: 2075- 2080. Wong BR, Rho J, Arron J, Robinson E, Orlinick J, Chao M, Kalachikov S, Cayani E, Bartlett FS, III, Frankel WN, Lee SY & Choi Y (1997b) TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells. J Biol Chem 272: 2519025194. Wronski TJ, Cintron M, Doherty AL & Dann LM (1988) Estrogen treatment prevents osteopenia and depresses bone turnover in ovariectomized rats. Endocrinology 123: 681- 686. Wuthier RE (1968) Lipids of mineralizing epiphyseal tissues in the bovine fetus. J Lipid Res 9: 6878. Wuthier RE (1969) A zonal analysis of inorganic and organic constituents of the epiphysis during endochondral calcification. Calcified Tissue Res 4: 20- 38. Yamaguchi A & Kahn AJ (1991) Clonal osteogenic cell lines express myogenic and adipocytic developmental potential. Calcified Tissue Int 49: 221- 225. Yamamoto M, Fisher JE, Gentile M, Seedor JG, Leu CT, Rodan SB & Rodan GA (1998) The integrin ligand echistatin prevents bone loss in ovariectomized mice and rats. Endocrinology 139: 1411- 1419. Yasuda H, Shima N, Nakagawa N, Mochizuki SI, Yano K, Fujise N, Sato Y, Goto M, Yamaguchi K, Kuriyama M, Kanno T, Murakami A, Tsuda E, Morinaga T & Higashio K (1998a) Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): a mechanism by which OPG/OCIF inhibits osteoclastogenesis in vitro. Endocrinology 139: 1329- 1337. 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 (1998b) Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesisinhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95: 35973602. Yellowley CE, Li Z, Zhou Z, Jacobs CR & Donahue HJ (2000) Functional gap junctions between osteocytic and osteoblastic cells. J Bone Miner Res 15: 209- 217.

65

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: 442- 444. Yoshimori T, Keller P, Roth MG & Simons K (1996) Different biosynthetic transport routes to the plasma membrane in BHK and CHO cells. J Cell Biol 133: 247- 256. Zheng MH, Papadimitriou JM & Nicholson GC (1991) A quantitative cytochemical investigation of osteoclasts and multinucleate giant cells. Histochem J 23: 180- 188.