ABCDEFG

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

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

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

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

EDITORIAL SECRETARY

Professor Mikko Siponen

Professor Harri Mantila

Professor Hannu Heusala

Professor Olli Vuolteenaho

Senior Researcher Eila Estola

Information officer Tiina Pistokoski

Senior Lecturer Seppo Eriksson

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-8833-3 (Paperback)ISBN 978-951-42-8834-0 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

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

MEDICA

ACTAD

OULU 2008

D 974

Virpi Muhonen

BONE–BIOMATERIAL INTERFACETHE EFFECTS OF SURFACE MODIFIEDNiTi SHAPE MEMORY ALLOY ONBONE CELLS AND TISSUE

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

D 974

ACTA

Virpi Muhonen

D974etukansi.kesken.fm Page 1 Thursday, May 29, 2008 11:44 AM

A C T A U N I V E R S I T A T I S O U L U E N S I SD M e d i c a 9 7 4

VIRPI MUHONEN

BONE–BIOMATERIAL INTERFACEThe effects of surface modified NiTi shape memory alloy on bone cells and tissue

Academic dissertation to be presented, with the assent ofthe Faculty of Medicine of the University of Oulu, forpublic defence in Auditorium A101 of the Department ofAnatomy and Cell Biology, on June 27th, 2008, at 12 noon

OULUN YLIOPISTO, OULU 2008

Copyright © 2008Acta Univ. Oul. D 974, 2008

Supervised byProfessor Juha Tuukkanen

Reviewed byProfessor Yrjö T. KonttinenResearch Professor Jari Koskinen

ISBN 978-951-42-8833-3 (Paperback)ISBN 978-951-42-8834-0 (PDF)http://herkules.oulu.fi/isbn9789514288340/ISSN 0355-3221 (Printed)ISSN 1796-2234 (Online)http://herkules.oulu.fi/issn03553221/

Cover designRaimo Ahonen

OULU UNIVERSITY PRESSOULU 2008

Muhonen, Virpi, Bone–Biomaterial Interface. The effects of surface modified NiTishape memory alloy on bone cells and tissueFaculty of Medicine, Institute of Biomedicine, Department of Anatomy and Cell Biology,Department of Medical Technology, University of Oulu, P.O.Box 5000, FI-90014 University ofOulu, Finland Acta Univ. Oul. D 974, 2008Oulu, Finland

Abstract

Whenever a foreign material is implanted into a human body an implant–tissue interface areaforms between them. In this microenvironment, interactions take place between the implant andthe surrounding tissue. The implantation of a biomaterial into tissue results in injury and initiationof the inflammatory response. This host response to biomaterials is an unavoidable series of eventsthat occur when tissue homeostasis is disturbed by the implantation process. In bone tissue,biocompatible implants must initially be capable of strong bone implant contact and subsequently,allow the normal bone remodeling cycle around the implant.

NiTi is a metal alloy composed of approximately a 50:50 ratio of nickel and titanium. Itpossesses shape memory and superelasticity properties, which make it an interesting biomaterial.NiTi has two phases: austenite and martensite. A decrease in temperature or applied stress inducethe austenite-to-martensite transformation. Heating or removing the stress restores the parentaustenite phase. The alloy in its martensite structure can be reshaped and strained several timesmore than a conventional metal alloy without irreversible deformation of the material. The alloyreturns to its original shape as it changes from martensite-to-austenite. This transformation is seenas the macroscopic shape memory effect.

This study further investigated the biocompatibility of NiTi, especially the bone cell responseto both austenite and martensite. Different surface treatments were investigated in order toimprove and possibly even control NiTi's bioactivity as a bone implant material.

Osteoclasts grew and attached well on the austenite NiTi phase, but the results indicated thatthe biocompatibility of martensite NiTi was compromised. Oxidation of the NiTi surfaceimproved osteoblast attachment and viability. This was due to the formation of a TiO2 surfacelayer of moderate thickness. Coating the NiTi surface with the extracellular matrix proteinfibronectin was shown to enhance osteoblast proliferation and increase the number of cells in theG1 cell cycle stage. Austenite was more prone to show these effects than martensite. A sol-gelderived titania-silica surface treatment was observed to increase the bone implant contact offunctional NiTi intramedullary nails. The surface treatment was most effective with the constantbending load provided by the NiTi nail.

Keywords: austenite, biocompatible materials, martensite, nickel-titanium alloy,osteoblasts, osteoclasts, prostheses and implants, surface treatment

To Plan B

6

7

Acknowledgements

This work was carried out in the Department of Anatomy and Cell Biology and the Department of Medical Technology, at the University of Oulu, during the years 2003–2008. There are many to thank, and I hereby wish to sincerely acknowledge everybody who has contributed to this work in one way or another. Especially, I would like to thank the following persons:

My supervisor, head of the Department of Anatomy and Cell Biology Professor Juha Tuukkanen, D.D.S., Ph.D. is most warmly acknowledged for his kind and humorous way of guiding me through this work.

The head of the Institute of Biomedicine (the Department of Medical Technology) Professor Timo Jämsä, M.Sc.Eng., Ph.D., is warmly acknowledged for all of his help and support.

Professor Hannu Rajaniemi, M.D., Ph.D. former Head of the Department of Anatomy and Cell Biology for supporting this work.

Professor Yrjö T. Konttinen and Research Professor Jari Koskinen for their critical and valuable comments on this thesis, and Dr. Deborah Kaska for the English revision of this thesis.

My co-authors Anatoli Danilov, Jean-Luc Duval, Caroline Fauveaux, Riikka Heikkinen, Joanna Ilvesaro, Sauli Kujala, Marie-Danielle Nagel and Gustavo Olivera for sharing their knowledge. Anna-Maija Ruonala is warmly acknowledged for her skillful help on the histological sectioning of the huge pile of rat legs.

The National Technology Agency of Finland (TEKES), the National Graduate School for Musculo-Skeletal Diseases and Biomaterials (TBGS) and Tauno Tönning Foundation are acknowledged for financial support.

Huge thanks go to my dear friend Annina Sipola for being there to share the sadness, happiness and the insanity of life. This would have been so much duller without you! My friends, Konna “tuh” Hakkonen and The Responsible Leader of the Project, should stop singing to me and concentrate on things that are more important. Konna - All little people should understand to be quiet when facing a threat a lot bigger than they are. Ave (Maria). Extra thanks to Mari for the three minutes on 3.8.2007 and for the red ball training of hand-eye coordination. And also for making me eat all the food on the plate and thus saving the polar bears. Keep up the good work (of studying the reproductive problems of gay seals and the regenerating pizza)! Thank you for the stimulating and intelligent conversations, especially on 12th of February 2008.

8

I also wish to thank my friends Zanna and Juxu for all the evenings and nights we have spend together eating good food and talking. You are übergreat. Molemolemolemole… Special thanks go also to Johanna Leppälä, Lumi Viljakainen, Marianne Kosonen and Tanja Pyhäjärvi for all the fun times since 1999, and for sharing the annoyances that come in the line of duty.

Wu Hsin Kuen Academy of Martial Arts is an amazing place filled with amazing individuals. I owe my deepest and warmest thanks to everybody who has been (literally) kicking my ass. My fellow Brownies Ilkka, Jukka and Reetta are warmly acknowledged for all the fun times of training and wondering how to stay alive. And special thanks to Jani, Janne and Ville for providing the reasons to wonder. And finally, I wish to thank sifu Risto Hietala for keeping it all together for us to discover the wonders of martial arts. Without the energy that you all have given to me during the past three years, this work would have taken so much more. It is truly a privilege to train with you guys!

The influence of my Super Great Family cannot be emphasized enough. Mom, dad – you are the best. My big brother Vesa… Did I actually win? Buah-hah-haa ;) Eevi, the gremlin of our family, you are the best little sister one can have. Thanks for supporting me in whatever I do in my life and forgiving me the fact that I will never grow up.

And finally, Tepi you ROCK! And I love you.

Oulu, May 2008 Virpi Muhonen

9

10

11

Abbreviations

αMEM Minimal essential medium; alpha modification Af Austenite finish temperature AFM Atomic force microscopy ALP Alkaline phosphatase Ang-1 Angiopoetin-1 ANOVA Analysis of variance AP Anteroposterior As Austenite start temperature BO Blue colored oxide BRC Bone remodeling compartment BMP Bone morphogenetic protein BSA Bovine serum albumin cAMP Cyclic adenosine monophosphate CaP Calcium phosphate CEL Carboxymethylcellulose Co-Cr Cobalt-chrome CSF-1 Colony stimulating factor-1 DMP-1 Dentin matrix protein-1 DMSO Dimethyl sulfoxide DSC Differential scanning calorimetry ECM Extracellular matrix EDTA Ethylenediaminetetraacetic acid EDX Energy dispersive X-ray analysis EthD-1 Ethidium homodimer-1 F-actin Filamentous actin FAK Focal adhesion kinase FBS Fetal bovine serum FGF Fibroblast growth factor FITC Fluorescein isothiocyanate FN Fibronectin HA Hydroxyapatite, Ca10(PO4)6(OH)2 HSC Hematopoietic stem cell IL Interleukin IM Intramedullary LDH Lactate dehydrogenase

12

LPS Lipopolysaccharide MAPK Mitogen activated protein kinase Mf Martensite finish temperature MMP Matrix metalloproteinase Ms Martensite start temperature MT Martensitic transformation MTT 3-[4, 5-dimethylthiazole-2-yl]-2, 5-diphenyltetrazolium

bromide Ni Nickel NiTi Nickel-Titanium shape memory alloy OPG Osteoprotegrin Pa Pascal (1 Pa = 1 N/m2) PBS Phosphate buffered saline PFA Paraformaldehyde PKC Protein kinase C PMMA Polymethylmethacrylate PMN Polymorphonuclear neutrophil granulocyte PTK Protein tyrosine kinase RANKL Receptor activator of nuclear factor κB ligand RGD Arginine-glycine-aspartic acid RMS Root mean square SCO Straw-colored oxide SDS Sodium dodecyl sulfate SiO2 Silica SMA Shape memory alloy SME Shape memory effect Ta Tantalum TCPS Tissue culture polystyrene Ti Titanium TiO2 Titania Ti6Al4V Titanium-aluminium-vanadium alloy TNF Tumor necrosis factor TRACP Tartrate-resistant acid phosphatase TRITC Tetramethyl-rhodamine isothiocyanate XDA X-ray diffraction analysis

13

Definitions

Austenite The high temperature (parent) phase of a material. Biocompatibility The ability of a material to perform with an appropriate

host response in a specific application. Biomaterial A nonviable material used in a medical device, intended to

interact with biological systems. Biomimetic A human-made process, substance, device or system that

imitates nature. Bone remodelation A process of reabsorption of pre-existing extracellular

matrix by osteoclasts, which is followed by osteoblastic secretion of new extracellular matrix that is later mineralized. Also referred to as bone turnover.

Hysteresis A retardation of the effect when the forces acting upon a body are changed.

Implant An artificial or natural object implanted into a tissue. Infection The state produced by the establishment of a pathogen in

or on a suitable host. Inflammation A local response to cellular injury that is marked by e.g.

capillary dilatation, leukocytic infiltration and swelling, and that serves as a mechanism initiating the elimination of noxious agents and of damaged tissue.

Load Describes the force applied to a component of a structure or to the structure as a unit.

Martensite The low temperature phase of a material. Osseoinduction The process by which osteogenesis is induced. Implies the

recruitment of immature cells and the stimulation of these cells to develop into preosteoblasts.

Osseointegration The direct anchorage of an implant due to the formation of bony tissue around the implant, without the growth of fibrous tissue at the bone–implant interface.

Osseoconduction In biomaterial science, the growth of bone on the surface of a foreign material.

Superelasticity The result of a stress-induced phase transformation. Shape memory effect The result of a thermal phase transformation. The ability

to recover the original predetermined shape under heating after prestraining in a low-temperature martensitic phase.

14

Shape memory alloy An alloy that possesses the shape memory effect. Sol-gel A wet-chemical technique for the fabrication of materials.

The sol-gel process involves the transition of a system from a liquid "sol" (mostly colloidal) into a solid "gel" phase.

Surface free energy The work required to increase the area of a substance by one unit area. In solids, it is sometimes also referred to as the "surface tension" of the solid substrate.

Strain A measurement of deformation, normalized by the original length of the specimen.

Stress Internal resistance to an applied force expressed in units of force (N) per unit area (m2) (1 N/m2 = 1 Pa).

Wear A loss of material with the generation of particles resulting from relative motion between two opposed surfaces.

Young's modulus Modulus of elasticity (E) measured as the slope of the linear portion of stress–strain curve. It is calculated by dividing the stress by the strain at any point along this portion of the curve.

15

List of original publications

This thesis is based on the following articles and additional unpublished results. The articles are referred to in the text by their Roman numerals:

I Muhonen V, Heikkinen R, Danilov A, Jämsä T, Ilvesaro J & Tuukkanen J (2005) The phase state of NiTi implant material affects osteoclastic attachment. J Biomed Mater Res A 75: 681–688.

II Muhonen V, Heikkinen R, Danilov A, Jämsä T & Tuukkanen J (2007) The effect of oxide thickness on osteoblast attachment and survival on NiTi alloy. J Mater Sci Mater Med 18(5): 959–67.

III Muhonen V, Fauveaux C, Olivera G, Vigneron P, Danilov A, Nagel M-D & Tuukkanen J (2008) Fibronectin modulates osteoblast behaviour on Nitinol. J Biomed Mater Res A (in press).

IV Muhonen V, Kujala S, Ääritalo V, Peltola T, Areva S, Närhi T & Tuukkanen J (2008) Biocompatibility of sol-gel coated orthopedic NiTi implants (in press).

16

17

Contents

Abstract Acknowledgements 7 Abbreviations 11 Definitions 13 List of original publications 15 Contents 17 1 Introduction 21 2 Review of the literature 23

2.1 Bone ........................................................................................................ 23 2.2 Bone cells................................................................................................ 23

2.2.1 Osteoblast ..................................................................................... 24 2.2.2 Osteoclast ..................................................................................... 25 2.2.3 Osteocytes and bone-lining cells .................................................. 26

2.3 Bone matrix............................................................................................. 28 2.4 Bone as a load sensing tissue .................................................................. 28 2.5 NiTi ......................................................................................................... 33

2.5.1 Shape memory property and superelasticity................................. 35 2.5.2 Biocompatibility of NiTi .............................................................. 40 2.5.3 Medical applications of NiTi ........................................................ 42 2.5.4 Surface modifications of NiTi ...................................................... 43

2.6 Cell adhesion........................................................................................... 44 2.6.1 Integrins........................................................................................ 45 2.6.2 Focal adhesions ............................................................................ 46 2.6.3 Fibronectin and fibrillar adhesions ............................................... 47

2.7 Cell–biomaterial interaction.................................................................... 48 2.7.1 Protein adsorption on an artificial surface (initial

interaction) ................................................................................... 49 2.7.2 Formation of adhesive bonds and triggering cell programs ......... 50

2.8 Bone implants ......................................................................................... 51 2.8.1 Biomaterials for bone applications ............................................... 52 2.8.2 Intramedullary nailing .................................................................. 54

2.9 Bone implant surface treatments............................................................. 57 2.10 Host response .......................................................................................... 59 2.11 Adaptive bone remodeling and loosening of orthopaedic bone

implants................................................................................................... 61

18

2.11.1 Peri-implant osteolysis ................................................................. 61 2.11.2 Mechanobiological regulation ...................................................... 64

3 Outline of the present study 67 4 Material and methods 69

4.1 Test Materials .......................................................................................... 69 4.1.1 NiTi samples – in vitro (I–III) ...................................................... 69 4.1.2 NiTi samples – in vivo (IV).......................................................... 70 4.1.3 Animals (IV)................................................................................. 71

4.2 Physical measurements ........................................................................... 72 4.2.1 Surface roughness......................................................................... 72 4.2.2 Contact angle measurements and surface free energy

calculations (I, III) ........................................................................ 74 4.2.3 Analysis of the surface oxide layer (II) ........................................ 76

4.3 Cell lines (I–III)....................................................................................... 78 4.4 Surgical procedure (IV)........................................................................... 79 4.5 Biological assays..................................................................................... 80

4.5.1 Cytotoxicity test (II) ..................................................................... 80 4.5.2 Preparation of fluorescein-fibronectin (FN-FITC) and FN-

FITC coating of NiTi (III) ............................................................ 81 4.5.3 Enzyme-linked immunosorbent assay (ELISA) (III).................... 81 4.5.4 Cell proliferation – LDH and MTT analyses (III) ........................ 82 4.5.5 Cell cycle analysis (III)................................................................. 83

4.6 Immunofluorescence staining ................................................................. 83 4.6.1 Staining of actin rings (I).............................................................. 83 4.6.2 Staining of focal adhesions (II)..................................................... 84

4.7 Histochemical staining and sample preparation...................................... 84 4.7.1 Staining of active osteoclasts (I)................................................... 84 4.7.2 Histological thin sections (IV)...................................................... 84

4.8 Imaging and image analysis .................................................................... 85 4.8.1 Light and fluorescence microscopy (I) ......................................... 85 4.8.2 Confocal microscopy (I–III) ......................................................... 85 4.8.3 Faxitron radiography (IV) ............................................................ 86 4.8.4 Digital image analysis (II–IV)...................................................... 86

4.9 Hybrid NiTi ............................................................................................. 87 4.10 Statistical analysis (I–IV) ........................................................................ 89

19

5 Results 91 5.1 Surface roughness of the NiTi samples (I–III) ........................................ 91 5.2 Contact angles (I, III) .............................................................................. 92 5.3 Surface free energy (I, III) ...................................................................... 92 5.4 Results of the physical measurements of the oxidized NiTi (II) ............. 93

5.4.1 X-ray diffraction........................................................................... 94 5.4.2 Resistivity..................................................................................... 94 5.4.3 Oxide thickness ............................................................................ 94

5.5 Success of the intramedullary NiTi nailing (IV) ..................................... 95 5.6 Biological assays..................................................................................... 97

5.6.1 Cytotoxicity (II)............................................................................ 97 5.6.2 Fibronectin rearrangement onto NiTi surface (III) ....................... 99 5.6.3 Analysis of fibronectin conformation on NiTi surfaces

(ELISA) (III) .............................................................................. 100 5.6.4 Cell proliferation on FN-coated NiTi (III).................................. 101 5.6.5 Cell cycle (III) ............................................................................ 103

5.7 Visualization of actin rings (I)............................................................... 104 5.8 Focal adhesions on osteoblasts attached to an oxidized NiTi

surface (II)............................................................................................. 106 5.9 Active osteoclasts on NiTi surface (I)................................................... 108 5.10 Osteoblast attachment to hybrid austenite–martensite NiTi

substrates............................................................................................... 109 5.11 Intramedullary NiTi nails – results of Faxitron radiography and

histological analysis (IV) .......................................................................110 6 Discussion 115

6.1 The phase state of NiTi implant material affects osteoclastic attachment (I) .........................................................................................115

6.2 The effect of oxide thickness on osteoblast attachment and survival (II) ............................................................................................117 6.2.1 Focal adhesions on oxidized NiTi .............................................. 120

6.3 Fibronectin modulates osteoblast behavior on Nitinol (III) .................. 120 6.4 Biocompatibility of sol-gel coated orthopaedic NiTi implants

(IV)........................................................................................................ 123 6.4.1 Intramedullary nailing and the effect of mechanical

loading........................................................................................ 123 6.4.2 Sol-gel surface treatment of NiTi ............................................... 127

20

7 Conclusions 129 References 131 Appendices 157 Original publications 181

21

1 Introduction

Biomaterials are materials intended to use in the human body to replace, augment or interact with the living tissue or function of the body. Vast numbers of different material types, such as metals, ceramics and polymers are used as biomaterials. Metals are commonly used in weight bearing applications, i.e. as bone implants. Unfortunately the perfect bone implant material has not yet been developed. All the current biomaterials used in bone tissue tend to loosen over time. And as life expectancy increases, the need for revision surgeries also increases. The interactions of implant materials and tissues are complex and they vary depending on both the implant material and target tissue. Although the principles of the processes that take place at the tissue–biomaterial interface are known, the details of these are still largely unknown. A comprehensive understanding of tissue–biomaterial interactions is important for the development of new and for the improvement of old biomaterial applications.

NiTi is an alloy of nickel and titanium in approximately a 50:50 ratio. NiTi is superelastic and can tolerate large strains without being irreversibly deformed. NiTi has two phases, called austenite and martensite. A decrease in temperature or applied stress induces the change of austenite into martensite. In the martensite form, the alloy can be largely strained and its macroscopic structure changed. When the alloy specimen is heated or the load removed, the specimen returns to its original shape. This phenomenon is seen macroscopically as the shape memory effect.

This study focused on biocompatibility investigations of both austenite and martensite phases by utilizing different bone cell cultures. The major part of the study concentrated on different surface treatments of NiTi in order to improve its biocompatibility and bioactivity. Biomaterial surface treatments are an interesting way to tailor the processes that occur after implantation, since the surface properties of an implant dictate biomolecule adsorption and subsequent cellular effects.

The use of NiTi as a load applying bone implant application was further examined. Bone is a load sensing tissue that remodels its structure constantly according to loading. Hypothetically, bone growth can be guided by applying a constant intramedullary load. A sol-gel derived titania-silica surface treatment was used to improve the osseoconduction and -integration of the NiTi implants.

22

23

2 Review of the literature

2.1 Bone

From a mechanical point of view, the skeleton is the supporting framework of all higher vertebrates. The skull and vertebral column gives shelter to the brain and spinal cord and the protective rib cage structure hides the vital soft tissue organs, including the lungs and heart. Bones also provide sites for muscle attachment thus enabling locomotion. On the other hand, bone and its marrow are responsible of blood cell production and the maintenance of calcium and phosphate homeostasis in serum. From a much larger perspective, bone tissue had an enormous relevance in the evolution of all higher animals. The subphylum Vertebrata, named according to the unique feature of the backbone, is associated with larger size and an active lifestyle, both features obviously giving these animals an edge in the battle for existence.

Bone tissue can be morphologically divided into two types: cortical and cancellous bone. The cortical bone structure is compact and serves mainly as mechanical support. In contrast, cancellous, or trabecular bone is structured as a sponge. Because of the structure, the synonym spongy bone is also used for this bone type. With a large surface area compared to volume, cancellous bone is mainly responsible of the metabolic functions of bone tissue. The spaces enclosed by calcified trabelulae of spongy bone and the medullary cavity surrounded by compact cortical bone are filled with hematopoetic bone marrow.

Bone is a type of highly specialized connective tissue, which is composed of cells and abundant extracellular matrix. The following chapters will describe these fundamental constituents of bone in more detail.

2.2 Bone cells

In principle, four types of bone cells can be found in bone tissue: osteoblasts, osteoclasts, osteocytes and bone-lining cells. In healthy adult bone tissue, the resorption of old bone and the formation of new bone are in balance. Once peak bone mass is achieved at early adulthood, bone remodeling (or bone turnover) occurs throughout life in an orderly fashion; where bone is eaten away by osteoclasts, and the same amount of bone is produced by osteoblasts. This process is called “coupling”. Healthy bone is a highly dynamic tissue with a constant and

24

controlled turnover. Peak bone mass is influenced by genetic heritage, adequate nutrition, normal sex steroid levels, and physical activity. The co-operation of different bone cell types in bone turnover is highly complex and can be disturbed by various diseases. These diseases manifest themselves in excess resorption or production of bone tissue.

2.2.1 Osteoblast

Osteoblasts synthesize new bone matrix, osteoid, and they are responsible for its later mineralization to new bone. These cells are of mesenchymal origin and differentiate under the influence of local growth factors such as fibroblast growth factors (FGFs), bone morphogenetic proteins (BMPs) and Wnt proteins (Komori 2008). Osteoblasts also require the transcription of Runx2 and Osterix transcription factors (Komori 2008). The plasma membrane of osteoblasts is rich in alkaline phosphatase (ALP), whose serum level is a common marker for bone formation. Osteoblast membranes have receptors for parathyroid hormone, prostaglandins, steroids (estrogen and vitamin D3) and for many adhesion molecules and cytokines. Osteoblasts also express cytokines at their plasma membrane. In particular, osteoblasts express colony stimulating factor 1 (CSF-1) and RANKL (receptor activator of nuclear factor κB ligand), which activate osteoclastogenesis. Osteoprotegrin (OPG) is a decoy RANK receptor secreted by osteoblasts in order to inhibit osteoclastic bone resorption. (Baron 2003.)

Endosteal osteoblasts are at least partly responsible for the maintenance of the self-renewal potential of the hematopoietic stem cells (HSC) transferring signals that anchor the stem cells in this special osteoblast-supportive microenvironment (Adams & Scadden 2006, Yin & Li 2006). Osteoblasts produce secreted and cell surface bound signaling ligands, such as Angiopoetin-1 (Ang-1) which is the ligand for Tie2-receptor tyrosine kinase expressed by HSCs (Huang et al. 2007). It has been shown that the Tie2 positive HSCs are localized to the bone surface and they are in contact with Ang-1-expressing osteoblasts (Arai et al. 2004). Ang-1/Tie2 signaling is important in sustaining quiescence and the long-term repopulating activity of hematopoetic stem cells in adult bone marrow (Puri & Bernstein 2003, Arai et al. 2004).

25

2.2.2 Osteoclast

Osteoclasts are of hematopoietic origin (Gothlin & Ericsson 1976). These large (20–100 µm) multinucleated cells are thought to form by fusion of mononuclear progenitors of the monocytes/macrophage family (Teitelbaum 2000, Boyle et al. 2003). This highly differentiated cell type has morphologically unique features, which function in the resorption process of bone tissue. The sealing zone is part of the osteoclast cell membrane (Väänänen & Horton 1995). According to its name, it seals the cell into the bone matrix. This structure can easily be detected in resorbing osteoclasts as a ring-like actin formation, whereas in non-resorbing osteoclasts, actin is merely arranged at the cell periphery. The sealing zone surrounds the ruffled border area of the osteoclast membrane (Stenbeck 2002). This part is filled with finger-like protrusions that penetrate to the bone matrix. Under the ruffled border, surrounded by the sealing zone, is located the resorption lacuna. This area of bone matrix is actively destroyed by the osteoclast by creating a low pH environment and by secreting matrix metalloproteinases (MMPs) (Delaisse & Vaes 1992, Hill et al. 1994). The degradation products are endocytosed and further processed by the osteoclasts and also by macrophages (Halleen et al. 1999). When bone resorption outstrips bone formation, as occurs in early menopause (Reid 2003, Rosen & Douglas 2003), osteopenia and osteoporosis can result. The structure of bone is weakened by loss of trabecular plates and bones become fragile and much more likely to fracture with low impact trauma. Similarly, if the rate of bone resorption is normal but bone formation is severely depressed, as is the case in patients on glucocorticoids (Lukert 2003), osteoporosis can also occur.

Osteoblasts fill the resorbed area (Howship's lacunae) produced by osteoclasts with new bone. Because of the joint function of osteoclasts and osteoblasts, these cells can usually be found in close proximity of each other in areas of active bone remodeling. More and more evidence has shown that osteoblast and osteoclast functions are coupled: osteoblasts are thought to control the formation and function of osteoclasts and vice versa (Rodan & Martin 1981, Lacey et al. 1998, Martin & Sims 2005). In addition to its requirement in osteoclast differentiation, the RANKL−OPG/RANK signaling route is evidently a crucial one in the osteoblast–osteoclast functional coupling (Theoleyre et al. 2004).

26

2.2.3 Osteocytes and bone-lining cells

When osteoblasts produce new bone matrix, they become trapped inside it. These trapped osteoblasts are called osteocytes. Osteocytes comprises 90−95% of all bone cells (Frost 1960, Palumbo et al. 1990). The osteocytes become phagocytosed and digested by osteoclasts during osteoclastic bone resorption. Osteocytes are in direct contact with each other and with the secreting side of osteoblasts. They also interact with the bone lining cells. Osteocytes communicate with the other bone cells by numerous long cellular membrane protrusions that lie in tunnels called canaliculi. This continuum network, which couples osteocytes metabolically and electrically, is formed by connections at the tip of the processes through gap junctions with integral membrane proteins called Connexins. (Lian et al. 2003.) Gap junctions are essential for osteocyte maturation, activity and survival (Furlan et al. 2001, Schiller et al. 2001).

The functions of these networked cells are still partly unknown. They are thought to be able to sense mechanical load on bone (Aarden et al. 1994, Tatsumi et al. 2007) and also participate in bone resorption (Shimizu et al. 1990). The mechanosensory function of osteocytes, i.e. the transduction of stress signals (stretching, bending, etc.) to biological activity, is thought to involve the strain of flow of extracellular fluid driven by extravascular pressure and mechanical forces throughout the canalicular routes (Nijweide et al. 2002, Bonewald 2006). Immobile, microtubule based primary cilia are suggested to function as the flow-sensing structures of osteocytes (Malone et al. 2007). Estrogen receptors of osteocytes might participate in the transduction of mechanical forces into cellular functions (Aguirre et al. 2007). The role of osteocyte apoptosis as a mechanosensory mechanism in intracortical remodeling has been recently suggested in the rabbit tibial midshaft (Hedgecock et al. 2007). Osteocyte apoptosis in micro-damage stimulated remodeling was also recently suggested (McNamara & Prendergast 2007).

In addition to mechanosensory function, recent papers on osteocyte function in mineral homeostasis (Feng et al. 2006) and mineralization of bone extracellular matrix (Lane et al. 2006) have been published. Dentin matrix protein 1 (DMP1; encoded by DMP1) is highly expressed in osteocytes (Toyosawa et al. 2001). Its deletion in mice results in a hypomineralized bone phenotype (Ling et al. 2005). The study of Feng and collegues (Feng et al. 2006) showed that DMP1 is critical for osteocyte maturation. This observation was supported by their findings that genetic removal of Dmp1 from the skeletal matrix (in mice) and loss-of-function

27

DMP1 mutations (in human autosomal recessive hypophosphatemic rickets kindreds) concurrently lead to independently altered skeletal mineralization and disturbed phosphate homeostasis, both due to defective osteocyte function. The results of Lane and co-workers (Lane et al. 2006) concerning the role of osteocytes in mineralization suggested that glucocorticoids might have direct effects on osteocytes, resulting in a modification of their microenvironment. These changes, including an enlargement of osteocytes lacunar space and the generation of a surrounding sphere of hypomineralized bone, seem to produce highly localized changes in bone material properties.

As with osteocytes, the functions of bone-lining cells are still partially unknown. In addition, the origin of these cells is still under debate. However, it is known that the majority of quiescent bone surfaces are covered with these cells (Jones & Boyde 1976, Zambonin Zallone et al. 1984, Hauge et al. 2001), but their role in bone remodeling sites are obscure. One hypothesis of their function in bone turnover is presented in figure 1.

Fig. 1. A schematic presentation of a cancellous bone remodeling compartment (BRC) during a complete remodeling cycle. According to one of the hypotheses concerning the bone remodeling site, osteoclastic bone resorption and osteoblastic bone formation occurs under the bone-lining cells. Osteocytes are in contact with the bone-lining cells, osteoblasts and with each other. The connecting network of osteocytes and bone-lining cells is re-established after the remodeling cycle.

28

According to one theory, bone-lining cells are an alternative osteoblast phenotype. After filling the bone remodeling compartment (BRC) with osteoid, the flattened osteoblasts differentiate into lining cells, as schematically presented in the right corner of figure 1, and the two lining cell layers become a single cell layer (Hauge et al. 2001). The function of BRC as a structure ensuring osteoclast–osteoblast interaction throughout bone remodeling is further discussed in the recent review of Matsua and Irie (Matsuo & Irie 2008). In a quite recent study (Eriksen et al. 2007), the authors investigated the BRC and the vasculature of trabecular bone, and came to interesting conclusions. They suggested that mechanosensory signals from the osteocyte network could be transmitted via the lining cells of quiescent bone to the bone remodeling compartment lining cells, where the signal could be translated into changes in osteoclast and osteoblast activity and eventually to the remodeling event of bone. Whether the cells of the outer wall of the BRC are differentiated osteoblasts, "real" lining cells, transformed endothelial cells or pericytes remains to be established. Also detailed knowledge of the interaction between BRC and adjacent sinusoidal structures with endothelial cells would be highly important to further clarify the functional relationships of the constituents of BRCs.

2.3 Bone matrix

Cells constitute only a minor part of the bone tissue. The majority of bone is extracellular matrix (ECM) that can be divided to organic (35%) and inorganic matrix (65%). The mineralized ECM is responsible for the rigidity and strength of the skeleton, while still having some degree of elasticity. The organic matrix is mainly composed of type I collagen (95%). Other compounds are proteoglycans and noncollagenous proteins, such as osteocalcin, osteonectin and osteopontin. The proteins of bone ECM are synthesized and secreted by osteoblasts. The inorganic matrix is mainly calcium and phosphate in the form of hydroxyapatite (HA). Bone is composed of approximately 70% mineral, 22% protein and 8% water (Lane 1979).

2.4 Bone as a load sensing tissue

Bone tissue architecture is influenced by mechanical stresses. This idea was first documented by Galilei in 1638 (Galilei 1638) when he suggested that the shape of bones is related to loading. The law of bone–stress relationship, the Wolff’s law

29

(Wolff 1892), and its further expansion by Frost (Frost 1987), establishes that the remodeling of bone occurs in response to physical stresses, and on the other hand, to the lack of them. Bone develops or adapts its structure to that most suited to resist the forces acting upon it (Wolff 1892). In principle, this means that bone is deposited in sites subjected to stress and is resorbed from sites where there is little stress (Robling et al. 2006). As with any structure, the mechanical properties of bone reflect the inherent material properties of its constituents and the way these are arranged and interact. Bone is highly anisotropic, i.e. it exhibits different properties when loaded from different directions (Turner et al. 1995), and it is also nonlinear and viscoelastic. These characteristics, together with the ability of bone to continually adapt to changes in its physiological or mechanical environment, make it difficult to establish universal constants to characterize the physical properties of bones (Einhorn 1996).

When forces in one direction exceeding the magnitude of bone tolerance are directed to bone, a fracture in its structure appears. A fracture is a specific site in bone where both the bone material and its structure failed. The applied force (i.e. load) causes a deformation of the bone tissue, resulting in an internal resistance to the applied force known as stress. Stress (δ) can be expressed in units of force (N) per unit area (m2) (1 N/m2 = 1 Pa). Stresses can be divided into three groups: tension, compression and shear (Fig. 2). Briefly, tension is produced in a material when two forces are directed away from each other along the same straight line. The ultimate tensile strength of a material is the sum of the cohesive intramolecular attractive forces that keep the material from being torn apart when subjected to tensile force. Compression results from two forces directed toward each other along the same straight line. Shear stress is an exception as it is produced when two forces are directed parallel to each other along different lines.

30

Fig. 2. Three groups of stresses. Arrows with “N” represents the direction of applied force. Black arrows show the direction of load.

The measurement of deformation, normalized to the original length of the specimen, is called strain (strain = ε = change in length / original length). Strain is the property of elastic materials, which are capable of returning to their original shape after deforming under load. When a moderate amount of stress is applied to normal, well-mineralized bone tissue, it will cause small strains, whereas the same stress applied to poorly mineralized bone will produce large strains (Einhorn 1996). In nature, loads are applied to bone from oblique angles, resulting in a variety of complex mechanical relationships. Thus, in addition to magnitude and direction, the orientation of a force must be known in order to determine the stresses and strains of a given tissue (Einhorn 1996). The relationship of load and deformation in bone tissue can be presented as a schematic stress−strain curve as in figure 3. This curve is independent of the size and shape of an object and is only a function of the material itself.

31

Fig. 3. Schematic presentation of a standard stress (δ) – strain (ε) curve of bone loaded in bending. The linear portion of the curve represents the elastic region and the slope of this part of the curve is used to derive the modulus of elasticity (E), also called the Young’s modulus of the material. Loading in this region will result in nonpermanent deformation. The nonlinear portion of the curve represents the plastic region in which the bone will be permanently deformed by the load, ultimately leading to fracture (fracture strength). The point of the curve which separates the two regions is called the yield point and the stress here is known as the elastic limit (δy), i.e. the amount of stress a material can undergo before it begins to permanently deform. The maximum amount of stress the material can handle, or in other words, the stress at the starting point of structural failure, is known as the ultimate strength of the material (δu). The strain at this point represents the bone’s maximum ductility. Loading after the point of the ultimate strength leads to bone fracture. The area under the curve is known as the strain energy, and the total energy stored at the point of fracture defines the toughness of the material.

Bones have optimal material properties in a particular orientation with minimum weight and metabolic load. As already mentioned, bones are anisotropic, having different mechanical properties in different directions. For example, the femoral bone in humans is oriented vertically and is subjected to a compressive load with every step taken. The structure of the femur can thereby resist high compressive loads, such as jumping from a modest height. On the contrary, the same load

32

applied from a transverse direction, causing high bending stresses, will not be as well resisted by the bone and may result in fracture. The bone as an entity is stronger in the direction of customary loading. This orientation of strongest resistance is especially seen in cortical bone where osteons are directed according to the direction of major stresses in the bone’s loading history (Hert et al. 1994). Evolutionary selection pressure has produced for load bearing bones a similar safety margin to failure (2–3 times) between the strains engendered during peak physical activity (2000–3000 microstrains), and those likely to cause yield or failure (Rubin 1984).

As seen in figure 3, the modulus of elasticity or Young’s modulus (E) is a measure of the slope of the linear portion of the curve. It is calculated by dividing the stress by the strain at any point along this portion of the curve. Thus, the Young’s modulus is an indicator of the behavior of a material under load. If the stress−strain curve were to be generated by testing a whole bone as opposed to uniform specimen, this measurement of linear slope would give the stiffness or rigidity of the bone. (Einhorn 1996.) The elastic modulus is the property of a constituent material and stiffness is the property of a solid body, thus elastic modulus is not principally the same as stiffness. However, the calculated value of the modulus of elasticity is often though of as a measure of the stiffness of a material body, such as the measurement of the elastic modulus of a bone sample as an indicator of the stiffness of the whole bone. Table 1 represents the estimated mechanical properties of femoral and tibial bone as a function of age. Table 2 gives estimates for the elastic moduli of cortical and trabecular bone as measured by a nanointendation protocol. However, it should be noted that the nanointendation method is likely to produce considerably greater values than more conventional tests (Currey 2008).

Table 1. Mechanical properties of femoral and tibial bone as a function of age.

Ultimate strength (MPa) Elastic modulus (GPa) Bone Age range

Tensile Compressive

Ultimate

tensile strain Tensile Compressive

Femur 20–29 140 209 0.034 17.0 18.1

Femur 70–79 129 190 0.025 16.3 18.0

Tibia 20–29 161 213a 0.040 18.9 35.3a

Tibia 70–79 145 183 0.027 19.9 26.7 aAge range 30 to 39. Modified from (Currey 2008).

33

Table 2. Mean values for elastic moduli of cortical and trabecular bone.

Bone type Direction of testing Elastic modulus (± SD), GPa

Longitudinal

Osteons 24.7 (2.5)

Interstitial lamellae 30.1 (2.4)

Cortical bone

Transverse 19.8 (1.6)

Longitudinal 20 (2.0) Trabecular bone

Transverse 14.7 (1.9)

Modified from (Wang et al. 2006).

Cortical bone is stronger in longitudinal compression than in tension, and after maturation the tensile strength and the modulus of elasticity of femoral cortical bone declines by approximately 2% per decade (Burstein et al. 1976). On the other hand, trabecular bone can withstand greater strains than cortical bone. The former fractures at changes of approximately 7% of its original length, and the latter already at 2%. In addition to the strength and direction of applied load, the rate of strain must also be identified when describing the behavior of bone, since bone tissue is viscoelastic (Einhorn 1992). Viscoelasticity simply means that a material’s mechanical properties differ according to the rate of loading.

Mechanotransduction in bone is mediated by bone cells. Which cells and how are they responsible of this, is still partly unknown. As already described in the earlier chapters, strong evidence has accumulated of the role and mechanisms of action of osteocytes as the mechanotransduction mediator in bone tissue (Bonewald 2006, Aguirre et al. 2007, Malone et al. 2007, Tatsumi et al. 2007).

Altogether, bone at both the tissue and macroscopic level is a very complex entity, in which the fine balance of structural geometry, composition and mechanical properties influence the metabolic (cell) and structural (skeletal) functions of bone. Thus, the form and function of bone is a three dimensional net of interactions which are inextricably bound together. (Einhorn 1996.)

2.5 NiTi

The nickel-titanium shape memory alloy NiTi (alias Nitinol or TiNi) has a long history in technical science and its applications, but a relatively short history in medicine. From its reporting in 1968 (Buehler & Wang 1968), NiTi has spread from submarines to the human body. The NiTi alloy is one of the most popular SMAs (shape memory alloys) in bioengineering applications owing to its

34

biocompatibility and its remarkable properties that allow recoverable mechanical energy to be stored in a compact delivery system (Cui et al. 2006). NiTi has a broad and continually expanding array of applications, including various prostheses used in vascular and general surgery. At the turn of the millennium, more than 227 tons of NiTi alloy were manufactured annually worldwide (Barras & Myers 2000). The NiTi shape memory alloys usually consist of binary alloys with Ni and Ti concentrations near an equiatomic composition. Most of the alloy is used in industrial applications.

The nickel-titanium alloys were first developed in 1962–1963 by the Naval Ordnance Laboratory in White Oak, Maryland and commercialized under the trade name Nitinol (an acronym for Nickel-Titanium Naval Ordnance Laboratory). There are at least two different stories about the discovery of the shape memory property of Nitinol: both stories state that the shape memory property was discovered by accident.

According to the first one, samples of the alloy were being subjected to strength tests by being pounded with hammers to see how much force was necessary to deform them. After several dents had been created, the researchers left the samples on a windowsill and went to lunch; upon their return, they discovered that the dents had “repaired” themselves. The second story says that the shape memory phenomenon of NiTi was discovered at a laboratory management meeting at the Naval Ordnance Laboratory. In the meeting, a strip of Nitinol was presented that was bent out of shape many times. One of the people present, Dr. David S. Muzzey, heated it with his pipe lighter, and surprisingly, the strip resumed its original form.

The unique properties of NiTi enable the alloy to undergo large mechanically induced deformations (a solid-state transformation) and when needed, to recover the original shape by thermal loading or by mechanical unloading, depending on the composition and processing history of the material. NiTi alloys have many important features including: resistant to corrosion, biocompatibility, fatigue resistance, magnetic resonance imaging compatibility and kink resistance with two unique thermomechanical behaviors: the shape memory effect (SME) and superelastic effect. (Petrini et al. 2005). These properties together make NiTi a very complex material in the sense of material specification and at least partly explain why ASTM (ASTM International, originally known as the American Society for Testing and Materials) specifications describing material composition and test methods have been issued as late as in the year 2002 (Stoeckel et al. 2004).

35

2.5.1 Shape memory property and superelasticity

The phenomenon of shape memory property was observed in 1920 by Johnson from observations of a copper and zinc alloy (Johnson 1920), and in 1938 Greninger and Mooradian published a comprehensive article of the properties of SMAs (Greninger & Mooradian 1938). The first attempt to explain the shape memory effect of NiTi is attributed to Zijderveld and associates, who established in 1966 that the unique properties were due to a crystalline transition induced by temperature change or application of stress (Zijderveld et al. 1966). At least 15 alloys are now known to exhibit the shape memory effect. These are divided into three main types: the copper-zinc-aluminum, copper-aluminum-nickel and nickel-titanium alloys. Several SMAs consist of rare and expensive metals and few match the force of shape recovery and resistance to permanent deformation that have made NiTi of commercial interest (Hodgson 1988).

SMA presents two well-defined crystallographic phases, i.e. austenite and martensite. In SMAs, martensite is a phase that, in the absence of stress, is stable only at low temperatures (Hodgson et al. 1990). The austenite phase is the parent state of the metal alloy. The two crystal lattices stable under certain thermodynamic conditions for a given alloy, characterize the two ultimate products of the martensitic reaction so that after its completion, the whole volume of material is represented only by one or another phase. In the low-temperature (martensite) phase, the alloy can be easily bent and shaped. As the metal is reheated (austenite), its original shape and stiffness are restored. Table 3 summarizes some of the mechanical properties of austenite and martensite NiTi.

Table 3. Selected mechanical properties of austenite and martensite NiTi.

Mechanical property Austenite Martensite

Ultimate tensile strength (MPa) 800–1500 103–1100

Elastic limit (MPa) 100–800 50–300

Modulus of elasticity (GPa) 70–110 21–69

Strain at failure (%) 1–20 Up to 60

Modified from (Ryhänen 1999).

The martensitic (also called displacive or diffusionless) transformation is a classical cooperative phenomenon in solids. Although the displacement of each atom is not large, the transformation results in a macroscopic change in shape, since all atoms move in the same direction in a domain or variant. Interconversions of solids, liquids and gases are widely understood, but the shape-

36

memory effect that results from a temperature-dependent phase change that occurs exclusively in the solid state (known as thermoelastic martensitic transformation) still remains partly unknown.

At lower temperatures, SMAs have a readily deformable, and millions of times reliably repeatable, crystalline arrangement of martensite. As the martensite is deformed, no dislocation of the metallic lattice takes place and irreversible plastic deformation does not occur, provided the strain limits of the particular alloy are not exceeded. Even though bent out of shape while in the martensite form, the metal rapidly recovers its original shape in the austenite form, in as little as 0.2 s (Barras & Myers 2000).

The two phases, austenite and martensite, are named after scientists who described their structure: Sir William Chandler Roberts-Austen, 1843−1902 for austenite and the German metallurgist Adolf Martens, 1850−1914 for martensite. The use of the terms austenite and martensite is not limited to SMAs, but they are common nominators for particular metallurgical phases present in the microscopic structures of metals, e.g. stainless steel can also be either austenitic or martensitic (Newson 2002).

NiTi alloys must be processed to achieve the shape memory property. In the process called "shape memory training", a SMA is rolled and forged under heat, put through a cold working phase, and finally carefully controlled heat treated at 450–550°C for 1–5 minutes. During the heat treatment the alloy is constrained on a mandrel or fixture to form a desired shape. After this treatment, heating the alloy to above its characteristic transformation temperature range results in recovery of the "memorized" shape. The transformation temperature range can be manipulated to occur between -200 to +110°C by varying the alloy composition or heat treatments. (Barras & Myers 2000.)

The shape memory effect can be divided into one-way and two-way shape memory effects, of which the latter appears much rarer. In the one-way effect, the customized-shaped SMA sample has a fully austenite phase in a temperature above Af (“austenite finish”) (Fig. 4). When the temperature of the sample decreases, under the threshold value Ms (“martensite start”), the phase transformation starts to take place and austenite is replaced with twinned martensite. The phase change continues until the Mf (“martensite finish”) temperature is reached, the sample having now only the martensitic phase. Mechanical loading of the martensite alloy causes the appearance of detwinned martensite, which presents a residual strain after freeing the load. The original shape of the sample, the parent shape in the austenitic phase, can be recovered by

37

heating the detwinned martensitic sample from the austenite transformation start temperature As (“austenite start”) to the Af temperature, where the alloy again reaches the fully austenite state. (Machado & Savi 2003.) Af is the upper end of the transformation temperature range. Careful control of its value is crucial to commercial and medical applications of the alloy (Barras & Myers 2000).

Fig. 4. Schematic presentation of the shape memory effect. The high temperature austenite structure transforms into twinned martensite as the temperature is lowered. This phenomenon is known as the thermoelastic martensitic transformation. Twinned martensite can be easily deformed by outer force by de-twinning. Heating the deformed martensite restores the austenite structure. At this stage the macroscopic shape memory phenomenon is seen. Note that the transformation from austenite to martensite (cooling) and the reverse cycle from martensite to austenite (heating) does not occur at the same temperature. Af = austenite finish, As = austenite start, Ms = martensite start and Mf = martensite finish.

Two-way shape memory is also a martensitic transformation phenomenon. In this type, the sample has two different customized shapes, one in the austenite and one in the martensite state. In contrast to one-way shape memory, in two-way shape memory the change of temperature produces a change in sample shape without any mechanical loading. (Machado & Savi 2003.)

Martensitic transformations are not limited to inorganic materials; displacive phase transformations also take place in biological systems, e.g. in cylindrical

38

protein crystals of virus tail-sheaths and bacterial flagella (Olson & Hartman 1982). The transformations of these structures appear to be stress-assisted mechanically reversible martensitic transformations exhibiting a shape memory effect.

Perhaps the most fascinating property of SMAs is force hysteresis. For most engineering materials, load (or stress) increases in a linear relationship with deflection or stretching (strain) within the material. NiTi responds differently (Fig. 5). As part of hysteresis loop, both loading and unloading curves show plateaus, along which large strains can be accommodated on loading, or recovered on unloading, with only a small change in stress (Stoeckel et al. 2004). This behavior of NiTi is much like natural tissues such as bone, and results in a ‘‘superelastic’’ ability to withstand and recover from large deforming stresses. Natural materials, such as bone, can be elastically deformed, in some cases up to 10% strain in a non-linear way (Shabalovskaya 1996). In conventional metals, such as stainless steel, the elastic deformation is limited to approximately 1% strain, and elongation typically increases and decreases linearly with the applied force. NiTi behaves similarly to biological materials, and since deformation up to 10% strain can be elastically recovered, the alloy is called superelastic (Stoeckel et al. 2004).

39

Fig. 5. The stress (δ) − strain (ε) hysteresis loop of superelastic NiTi and some biological materials. When the deforming stress is released, the strain is recovered at lower stresses.

Superelasticity, which is also known as pseudoelasticity, is an inherent property of SMAs. Superelastic materials can be strained several times more than ordinary metal alloys without being plastically deformed due to fatigue. When the SMA sample is at a temperature above Ms, in which the austenite phase is the dominant phase in a stress-free specimen, the martensite phase can be induced by application of an outer force (Fig. 6). This martensite formation is called the stress-induced phase transformation, in contrast to the temperature-induced phase transformation seen in the shape memory events. In the stress-induced phase transformation, the original structure of the sample is recovered simply by removing the stress. The high elasticity of NiTi is not simply stretching the atomic bonds, but it is the result of changes in the crystal structure by stress. The largest ability to recover occurs close to the Af temperature and, moreover, the superelasticity is only observed in a distinct temperature range that is specific for each alloy composition. In addition to alloy composition, the material processing history and the ambient temperature also greatly influence the superelastic

40

properties of the material. The temperature at which martensite can no longer be stress-induced is called the Md-temperature. (Duerig et al. 1996.)

Fig. 6. Schematic presentation of superelasticity. Austenite structure is transformed into martensite by application of outer force. This type of austenite-to-martensite transformation is called the stress-induced phase transformation. The original shape of the parent austenite structure is restored when the outer force is removed.

Even though relatively many alloys have the shape memory property, only those that possess the ability to recover from substantial amounts of strain or that can generate significant force upon changing shape are of commercial interest. In medical applications, these materials must also have excellent biocompatibility in order to be used as biomaterials in a living body. Implant materials have to be inert enough not to disturb the fine balance of the surrounding tissues. Moreover, in the case of NiTi, both austenite and martensite phases need to be biocompatible.

2.5.2 Biocompatibility of NiTi

The biocompatibility, and especially allergenicity, toxicity and potential carcinogenic effects of NiTi have been the target of many studies. The adverse effects are thought to be caused by nickel, which is considered toxic (Sigel &

41

Sigel 1988, Karaczyn et al. 2006, Salnikow & Zhitkovich 2008). NiTi is an intermetallic compound and not an alloy in the metallurgic sense. The bonding force of nickel to titanium is much stronger in NiTi than, for example, to the alloy components of stainless steel, making nickel ion release from NiTi to the surrounding biological environment more rare. Furthermore, NiTi is a self-passivating material, i.e., it forms a stable nickel-free surface TiO2 layer that protects the base material from general corrosion (Chan et al. 1990, Wever et al. 1998). Polarization testing in physiological solution has repeatedly shown that NiTi is at least as chemically stable and corrosive resistant as stainless steel (Speck & Fraker 1980, Rondelli et al. 2006), although the integrity of the surface oxide layer is required for good corrosion resistance (Rondelli 1996). NiTi SMAs are also thought to have biocompatibility similar to that of pure titanium, provided that the microstructure and surface properties are well controlled in order to allow high corrosion resistance to be achieved (Shabalovskaya 1996, Wever et al. 1997, Es-Souni et al. 2005).

The tensile elastic modulus for cortical bone specimens is in the range of 11–20 GPa (Currey 2008). NiTi has a Young's modulus of around 70 GPa (Danilov et al. 2003), whereas for 316 stainless steel it is 193 GPa, and for titanium it is in the range of 105–120 GPa. Thus, the behavior of NiTi under load is closer to bone than that of more commonly used weight bearing bone implant materials. This property makes the alloy well suited for bone applications. NiTi has the potential to preserve the bone from an unwanted stress shielding effect, which is caused by uneven stress distribution due to “stiff” implant materials. In addition to the estimations of overall implant material elasticity or implant stiffness, microlevel stress analyses are important, since they can provide further details on loosening and bone resorption at the implant−bone interface (Rho et al. 1997, Rho et al. 1999, Hengsberger et al. 2002).

There are numerous studies concerning both the in vitro (Ryhänen et al. 1997, Assad et al. 1998, Kapanen et al. 2002a, Kapanen et al. 2002b) and in vivo (Ryhänen et al. 1998, Kapanen et al. 2001) biocompatibility of NiTi. The majority of these papers strongly indicate good biocompatibility of the alloy (Wever et al. 1997, Mantovani 2000, Es-Souni et al. 2005). Furthermore, in many of the studies (Shabalovskaya 1996, Trepanier et al. 1998, Armitage et al. 2003), the surface properties of NiTi were considered as an important aspect affecting the biocompatibility. Danilov and others pointed out the possible difference in the biocompatibility of the austenite and martensite phases of NiTi (Danilov et al.

42

2003). They concluded that biocompatibility decreased depending on martensite quantity and residual stresses in the parent austenite phase.

The average dietary intake of nickel is 200–300 mg/day and the critical concentration to induce allergy is around 600–2500 µg of nickel (Huang et al. 2003). Although the quantities of nickel released from NiTi are much lower than the above-mentioned critical values; there is a risk of induction of long-term inflammatory responses or altered cell behavior. For example, a significant decrease in ALP activity and DNA synthesis was found in osteoblasts with subtoxic concentrations of nickel. (Sun et al. 1997.) Other studies showed that Ni ions could be responsible for inducing the secretion of different cytokines involved in the inflammatory process (Wataha et al. 1996, Wataha et al. 1999, Cederbrant et al. 2003).

2.5.3 Medical applications of NiTi

Of all the commercial applications of NiTi, the most popular market today is the medical industry, in which the excellent malleability and ductility of NiTi enable it to be manufactured in the form of wires, ribbons, tubes, sheets or bars. It is particularly useful for very fine small devices. In the late 1980’s, minimally invasive endovascular medical applications of NiTi emerged (Dotter et al. 1983). At the beginning of this millennium, over 50% of all peripheral vascular stents available on the worldwide market were manufactured from NiTi alloys (Duerig et al. 2000).

Medical applications of NiTi exploit the superelasticity and the shape memory effect of the alloy. As an example, a NiTi stent is presented in figure 7. Due to the high kink resistance and superelasticity, NiTi stents are ideal for superficial vessels, e.g. carotid artery, where a risk of permanent stent deformation due to outside forces is a concern. NiTi stents can be completely crushed, but they still can return to their original diameter when the force is removed. (Duerig et al. 2000.)

43

Fig. 7. Deformation of a superelastic NiTi stent (Cordis SMART). The stent will recover to its original structure after the load is removed. Modified from (Stoeckel et al. 2004.)

Today, the list of NiTi-utilizing medical applications is long. These include: vascular, esophageal and biliary stents and filters, medical guidewires and -pins, surgical localization hooks, flexible steerable and hingeless laparoscopic surgical instruments, remote suturing and stapling devices, bone suture anchors, clamps for fixation of small bone fragments and applications in endo- and orthodontics where constant correcting loads are required (Musialek et al. 1998, Barras & Myers 2000, Machado & Savi 2003). Porous NiTi has also been investigated as a potential bone implant material (Kujala et al. 2003, Prymak et al. 2005, Bertheville 2006). The porosity is thought to allow bone ingrowth into the pores, thus enabling good osseointegration.

In contrast to stainless steel implants, NiTi devices produce no magnetic resonance imaging (MRI) “black-hole artifacts” or significant force of attraction which can cause a risk of implant displacement during MRI study (Teitelbaum et al. 1988a, Teitelbaum et al. 1988b, Taal et al. 1997).

2.5.4 Surface modifications of NiTi

Different surface modifications have been performed on NiTi in order to improve its biocompatibility. NiTi stents have been modified with oxide coatings (mainly TiO2) to e.g. achieve better corrosion resistance (Trepanier et al. 1998) and adhesion properties for endothelial cells (Wang et al. 2007). Also chitosan and

44

hyaluran have been used on NiTi stent surfaces to reduce the neointimal hyperplasia that is associated with stent implantation (Thierry et al. 2003).

In dental applications, NiTi archwires have been coated with diamond-like carbon (Kobayashi et al. 2007) and titanium (Ozeki et al. 2003) for improving biocompatibility in a corrosive environment. A diamond-like carbon coating has been shown to improve the corrosion resistance and hemocompatibility of NiTi without sacrificing its superelasticity (Sui & Cai 2006).

TiO2-coating followed by apatite layer induction in simulated body fluid (SBF) on porous nanocrystalline NiTi was observed to attract primary human osteoblast attachment and proliferation (Gu et al. 2006). Adhesion and proliferation of osteoblasts were also improved with carbon plasma immersion ion implantation and deposition treatment, which also improved the corrosion resistance and blocked the diffusion of Ni ions from the NiTi alloy (Poon et al. 2005, Chu 2006). Plasma surface treatment of NiTi has proven to be promising for enhancement of NiTi compatibility for orthopaedic applications (Yeung et al. 2007, Chu 2007a). For example, the surface properties of NiTi spine implants were improved using plasma-based methods (Chu 2007b).

Calcium phosphate (CaP) coating on NiTi inhibited apoptosis of polymorphonuclear neutrophil granulocytes (PMN), thus influencing the functional activities of PMN, such as local host defense (Bogdanski et al. 2004). A bonelike apatite-collagen composite coating on NiTi was observed to resemble the matrix of human thighbone, thus making the surface of the alloy more biocompatible (Cai et al. 2006). In addition, HA coated NiTi showed good biocompatibility and osseointegration in animal models (Chen et al. 2004, Li et al. 2007).

2.6 Cell adhesion

Cell adhesion can be considered as cell–cell, cell–ECM or cell–biomaterial adhesion. The latter two types of adhesions are the main interest in this study. Cell adhesion to a biomaterial surface is, in many cases, a special form of cell–ECM interaction, since the surface of a biomaterial is always covered by molecules with which the cells interact. The cells do not interact with the bare biomaterial (eg. metal) itself. Cell–cell and cell–ECM interactions are crucial for tissue morphogenesis, homeostasis and remodeling (Adams & Watt 1993, Gumbiner 1996, Damsky et al. 1997). Inadequate or inappropriate cell–ECM adhesions can even lead to anoikis, a special class of adhesion driven apoptosis (Frisch &

45

Screaton 2001). The extracellular matrix contains information that is interpreted by cells via adhesion structures and influences cell shape, cytoskeletal organization, cell motility and polarity, gene expression, proliferation and survival (Damsky & Werb 1992, Huhtala et al. 1995, Moursi et al. 1996, Moursi et al. 1997, Globus et al. 1998, Ilic et al. 1998).

2.6.1 Integrins

Cell adhesion requires transmembrane cell surface receptors to mediate the physical contact of cells to the outside matrix and for the propagation of signaling cascades from outside-to-inside, and vice versa. Integrins are a large family of homologous transmembrane cell–ECM adhesion receptors (Hynes 1987). They are heterodimeric structures that consist of a α- and β-subunits; the cytoplasmic domains of the β-subunits transduce the extracellular signal into the cytoplasm. The global conformation of integrin is considered important for its signaling functions (Takagi et al. 2002).

Different cell types express different mixtures of integrins; e.g. osteoblast and osteoclast plasma membranes have different integrin compositions (Clover et al. 1992). Integrins facilitate downstream signaling cascades by organizing the cytoskeletal ‘scaffold’ composed of different kinases for intracellular signaling and structural components like talin and paxillin (Hynes 1987, Turner 2000) in order to control cell morphology, proliferation, and differentiation (Menko & Boettiger 1987, Streuli et al. 1991). The interaction of ECM, integrins and the subsequent cellular responses are closely linked with the functions of different receptor and cellular protein tyrosine kinases (PTK), indicating that integrin and growth factor signaling pathways may converge (Boudreau & Jones 1999). The intracellular serine/threonine protein kinase C (PKC) family has been implicated as one of the important regulators of the adhesion mediated cellular functions (Miranti et al. 1999). In addition, integrins induce inside-out signals leading to e.g. ECM biosynthesis and to differentiation in osteoblasts (Knight et al. 2000, Luthen et al. 2005, Reyes et al. 2007). Integrins, with other adhesion components, function also as mechanotransducers, transmitting forces into cell reactions (Ingber 1997, Bershadsky et al. 2003, Luthen et al. 2005, Hirakawa et al. 2006, Balasubramanian et al. 2007).

The extracellular parts of the integrin dimers bind to ECM proteins. For example, in osteoblasts, α5β1 and α2β1 integrins bind to fibronectin controling multiple crucial osteoblast functions, such as proliferation (Gronthos et al. 1997,

46

Keselowsky et al. 2007b), osteogenesis in early stages of bone formation (Weiss & Reddi 1981, Cowles et al. 1998) and mineralization (Moursi et al. 1997, Schneider et al. 2001). The α2β1 integrin binds to type I collagen providing crucial signals for the induction of osteoblastic differentiation and matrix mineralization (Lynch et al. 1995, Takeuchi et al. 1997, Jikko et al. 1999, Mizuno et al. 2000, Mizuno & Kuboki 2001, Suzawa et al. 2002), such as the activation of the Runx2/Cbfa1 osteogenesis transcription factor (Xiao et al. 1998).

In addition to integrins, non-integrin receptors bind to ECM molecules. Non-integrin receptors include e.g. surface proteoglycans (Adams & Watt 1993), laminin-binding proteins, such as dystroglycan (Colognato et al. 2007) and also some receptor PTK's can function directly as receptors for ECM molecules (Shrivastava et al. 1997).

2.6.2 Focal adhesions

Focal adhesions are integrin-based molecular compositions of cells participating in adhesion-dependent signaling (Abercrombie et al. 1971, Izzard & Lochner 1976, Avnur & Geiger 1981). The structural and signaling proteins of focal adhesions constitute the diverse variety of complexes, which link the ECM to the actomyosin cytoskeleton of a cell (Sastry & Burridge 2000). The structure of focal adhesions is dynamic and heterogeneous, and these structures are motile and can assemble/disassemble, disperse and recycle according to the cell’s needs (Smilenov et al. 1999, Sastry & Burridge 2000). Figure 8 represents a simplified model of focal adhesions and the formation of fibrillar adhesion. In reality, focal contacts are very complex structures; at least 50 different proteins have been reported to be associated with these and related ECM structures (Zamir & Geiger 2001). Focal adhesions coincide with actin microfilaments on their cytoplasmic side. This association plays an important role in the regulation of actin organization, hence affecting cell spreading, morphogenesis and migration (Defilippi et al. 1999, Zamir & Geiger 2001).

47

Fig. 8. Hypothetical molecular model of focal and fibrillar adhesion. Fibronectin receptor α5β1 integrins translocate centripetally owing to the actomyosin-driven pulling. The translocation of the α5β1 integrins stretches the fibronectin matrix and promotes fibrillogenesis. The αvβ3 integrins bound to vitronectin remain stationary and stay in the focal adhesion sites although they are also subjected to actomyosin-driven contraction forces.

2.6.3 Fibronectin and fibrillar adhesions

Fibronectin (FN) is a large cell-adhesive glycoprotein composed of two 250 kDa subunits joined together by disulfide bonds. It is secreted by many normal and tumorigenic cells. (Hynes 1990.) It is a central component of the ECM, regulating adhesion-dependent survival signalling (Sechler & Schwarzbauer 1998), and taking a large role in cell proliferation and differentiation (Bourdoulous et al. 1998, Mercurius & Morla 1998, Garcia et al. 1999). FN binds primary to the α5β1 integrins, through the arginine-glycine-aspartic acid (RGD) cell-binding site (Takagi 2004). A recent study of plasma fibronectin knock-out mice by

48

Keselowsky et al. (Keselowsky et al. 2007a) showed that fibronectin has a role in the regulation of the foreign body reaction and chronic inflammatory responses induced by synthetic biomedical devices.

Fibronectin forms the primitive fibrillar matrix in both embryos and healing wounds. The process of fibrillogenesis of FN is dependent on the translocation of α5β1 integrins out of focal contacts in a centripetal direction into fibrillar adhesions (Pankov et al. 2000, Yamada et al. 2003, Luthen et al. 2005), as demonstrated in figure 8. The translocation of fibrillar adhesions is driven by actomyosin contractility and can be blocked with different inhibitors, such as protein kinase inhibitor H-7 (Zamir et al. 2000). Studies of fibronectin fibrillogenesis have provided information about matrix control of cellular functions (Mao & Schwarzbauer 2005). Fibronectin matrix formation has been shown to stimulate adhesion-dependent cell proliferation (Sottile et al. 1998). On the other hand, inhibition of FN matrix assembly decreased DNA synthesis in aortic smooth muscle cells (Mercurius & Morla 1998). In bone, the basic fibroblast growth factor (FGF) increases bone formation and fibronectin fibrillogenesis both in vitro and in vivo via a protein kinase C dependent pathway (Tang et al. 2004). Fibronectin networking is absent in tumor cells, a feature thought to be of importance for their invasive growth and formation of metastases (Ruoslahti & Giancotti 1989, Giancotti & Ruoslahti 1990, Hynes 1990). In addition, transformed cells are proposed to have a growth advantage in vivo, since they can survive and grow independent of the integrin-mediated mechanical input from the surrounding tissues (Wang et al. 2000).

2.7 Cell–biomaterial interaction

The quality of initial attachment and adhesion in a cell−biomaterial interaction will influence the cell’s capacity to proliferate and differentiate. The molecules responsible for these processes involve adhesion receptors such as integrins, the actin cytoskeleton, and the intracellular cytoskeleton associated proteins. Figure 9 demonstrates the three players in a tissue–biomaterial interaction (Kasemo & Lausmaa 1994).

49

Fig. 9. The players in tissue−biomaterial interactions.

The simplicity of the above figure (Fig. 9) is strongly misleading. In reality, a vast number of different interactions and reactions take place in the interface area, all of which ultimately define the relationship of the implant and the surrounding tissue.

2.7.1 Protein adsorption on an artificial surface (initial interaction)

When a solid implant material is surgically implanted into the physiological milieu, protein adhesion to the biomaterial surface takes place immediately (Horbett 1993). The protein coat forms in less than 30 seconds after exposure to the biological environment. The properties of the implant surface determine the kind and number of proteins that attach to the surface (Horbett 1981, Andrade & Hlady 1986, Kasemo & Lausmaa 1994, Ratner 1995). The protein-modified biomaterial surface is the substrate that the cells subsequently encounter, thus the understanding of protein adsorption phenomena and protein interactions on biomaterial surfaces are of great importance. For example, inflammatory cells react with the protein-modified biomaterial surface (Anderson & Miller 1984), and the nature of this interaction has a far-reaching influence on the success of the implantation. The fundamental interaction mechanisms of protein adsorption are well understood, but rather little is known about desorption and exchange of proteins on biomaterial surfaces.

Numerous substratum properties affect the initial protein adsorption and the subsequent reactions. For example, the substratum chemistry has been noted to have an influence on fibronectin matrix formation of human fibroblasts (Faucheux et al. 2006). Phosphorylation of the focal adhesion adaptor protein

50

paxillin during early osteoblast adhesion was observed to be influenced by surface-bound serum protein (Sommerfeldt et al. 2001). One important surface property that has an impact on protein adsorption is the surface free energy of the material. Surface free energy is an important parameter for cell attachment and spreading on artificial surfaces (Hallab et al. 2001), e.g. its role in osteoclast spreading has been emphasized (Redey et al. 1999, Kawai et al. 2004).

In cell culture experiments, the proteins that initialy adsorb come primary from the culture media. The media contains ECM and other proteins that adsorb to the material surface depending on the surface characteristics and subsequently affect the formation and quality of cell adhesions and cell responses (Faucheux et al. 2004). In in vivo procedures, the biomaterial surface properties, and hence the initial adsorption, are affected by the cleanliness, contamination and preparation procedures of the implant device (Kasemo & Lausmaa 1988).

2.7.2 Formation of adhesive bonds and triggering cell programs

The physical state of the ECM strongly contributes to the properties of cell–matrix adhesions (Katz et al. 2000). Biomaterial surface properties are reflected to the molecular architecture of the ECM, including protein conformation (Garcia et al. 1999, Thull 2002, Keselowsky et al. 2003b). This directly influences the cells ability to form adhesion complexes, and ultimately affects many aspects of cell behavior. For example, in fibroblasts, poor adsorption of proteins and hence, poor cell adhesion leads to the formation of cAMP second messengers and cAMP-induced cell aggregation and suppression of cell spreading and growth rate (Faucheux & Nagel 2002). The absence of adhesive proteins on a biomaterial surface can lead to the absence of focal contacts and stress fibers, which can lead to reduced cell growth (Faucheux et al. 2004).

As already discussed in the previous chapters, the formation of adhesive bonds between the ECM components and the cell cytoplasm via integrins is the starting point of vast numbers of signaling cascades. These cascades lead to cell proliferation, differentiation and other crucial cellular functions. The inflammatory response at the implant site is also dependent on the formation of adhesion. Macrocyte phagocytic activity (Kaplan & Gaudernack 1982, Newman & Tucci 1990), migration (Lanir et al. 1988), and macrophage fusion to form foreign body giant cells (Kaplan & Gaudernack 1982, Smetana et al. 1987) are influenced by adhesion to different proteins.

51

2.8 Bone implants

Biomaterial research is a multidisciplinary field that has grown exponentially during the past decades. In fact, biomaterial science as an “independent” research area has existed only just a half century. Today, biomaterials and medical devises are over a 100$ billion endeavor world wide (Ratner & Bryant 2004). A biomaterial can be defined as a “nonviable material used in a medical device, intended to interact with biological systems” (Williams 1987). Biomaterials have a vast scale of applications and uses, including about 25 different areas of medical and more than 15 of non-medical uses (table 4).

Table 4. Uses for biomaterials (in alphabetical order).

Medical uses Non-medical uses

Artery graft Arrays for DNA and diagnostics

Bone implants Bioremediation materials

Breast implant Biosensors

Cochlear implant Bioseparations, chromatography

Dental implant Biofouling-resistant materials

Ear drainage tube Biomimetics for new materials

Feeding tube Cell culture

Glaucoma drainage tube Controlled release for agriculture

Hydrocephalus shunt Electrophoresis materials

Intraocular lens Fuel cells (biomass)

Joint (hip. knee, shoulder) Microelectro mechanical systems (MEMS)

Keratoprosthesis Muscles (artificial) and actuators

Left ventricular assist device (LVAD) Nanofabrication

Mechanical heart valve Nano electro mechanical systems (NEMS)

Nerve guidance tube Neural computing/biocomputer

Ophthalmic drug delivery system Smart clothing

Pacemaker Yeast array chip

Renal dialyzer

Stent

Tissue adhesive

Urinary catheter

Valve, heart

Wound dressing

X-ray guide

Modified from (Ratner & Bryant 2004).

52

The population structure of developed countries is biased toward elderly people. The population/age structure of the developed world represents that of a stable (or nearly stable) population, with more a rectangular shaped “growth pyramid” than for a rapidly growing population where there is a much larger number of young people. The western population with its increasing number of elderly people is one of the main reasons behind the increasing need for bone implants. The process of “getting old” is accompanied by reduced bone density and bone loss with inadequate nutrition not compliant with calcium and vitamin D supplementation. Weaker bones together with reduced motoric skills are high risk factors for fall-related fractures. Hip fracture is a common fall-related fracture and according to the National Agency for Medicines ca. 5000 hip replacements are performed every year in Finland (Nevalainen et al. 1998). In addition to hip and other joint replacements, there are numerous types and usages for bone implants. The most common ones can be divided into different screws, nails, plates, wires and pins, which are usually manufactured from different metals and their alloys, such as stainless steel, titanium and its alloys.

2.8.1 Biomaterials for bone applications

Biomaterials can be classified by various criteria. The National Agency for Medicines classifies these as follows: 1) metals, 2) polymers, 3) ceramics, 4) composites and 5) biological materials. Biomaterials can also be classified according to their surface reactivity: I) almost inert, smooth surface; II) almost inert, rough surface; III) chemically reactive surfaces and IV) resorptive materials. Most of the materials in clinical use are in classes I and II. In these materials the tissue response is a foreign body reaction, where fibrous scar-like tissue forms around the implant, isolating the implant from the tissue. Biomaterials from class III, called bioactive materials, are e.g. capable of direct bonding to bone tissue, whereas type IV materials are completely replaced by regenerating tissue. Biomaterials for bone implants can also be classified according to the above mentioned classification by the material's chemical properties or by surface reactivity, since bone tissue related applications can be found in all of these groups. (Aho et al. 2003.)

Metals are common materials for weight bearing applications and these materials must tolerate wear, corrosion and fatigue even for the long term. Today, titanium and cobalt-chrome (Co-Cr) based alloys are used in various dental and orthopaedic applications, such as femoral stems in total hip replacement. In

53

addition, steel is still very commonly used, but mostly in applications where the implant is later removed from the body (Newson 2002). Cobalt-chrome has better rigidity, abrasion, fretting and corrosion properties when compared to titanium alloys (Carter & Spengler 1978, Dunn & Maxian 1994, Black & Hasting 1998) (Schmalzried et al. 1998), but titanium implants are a bit more osteoconductive than those made of Co-Cr. Titanium is also lighter than other biomaterial metals (4.51 kg/dm3), having approximately half the density of the others. In orthopaedic applications tantalum (Ta) is also used. It is mainly used in the porous form, providing a good matrix for bone ingrowth. The use of tantalum implants and augmentation devices has changed revision operations of the acetabulum part of the hip prosthesis, since with tantalum, the use of titanium screws and plates is reduced and also the need for bone allografts is diminished (Unger et al. 2005, Weeden & Schmidt 2007).

Polymers are chains of organic or inorganic monomers that react and form single- or multidimensional structures. Polymers can be synthetic, natural or modified from natural polymers. They can be either biostabile or biodegradable in tissue surroundings. Some of the most common polymers are silicones, polyurethanes, polyethylene, polyamides and polymethylmethacrylate. Polymethylmethacrylate (PMMA), or bone cement, is the common filling material between a hip prosthesis and bone. (Aho et al. 2003.)

Ceramic materials are solid compounds of metals and nonmetals, mainly oxygen. Ceramics are brittle by nature, hence they have limitations in weight bearing applications, but they are used, for example, as ceramic-on-ceramic, mainly alumina-on-alumina, bearing surfaces in total hip arthroplasty (Mehmood et al. 2008). Ceramic materials used in non-weight bearing applications include e.g. bone substitute materials (Välimäki & Aro 2006).

Composites contain two or more phases with different structural or characteristic properties at the macroscopic, microscopic or submicroscopic level. Composite materials are used in medical device fabrication, since rarely can only one material meet the physical requirements for successful implantation and function. Most of the natural biomaterials are composites, such as bone and cartilage. (Aho et al. 2003.)

Materials of biological origin can be divided into four categories: 1) tissue transplants, 2) blood products, 3) processed human or animal originated products, such as demineralized bone matrix and 4) hybrid products. Human or animal originated tissues or tissue products are designated auto-, allo- or xenografts according to their origin. An autograft is a transplant taken from and implanted

54

into the same individual. These are not usually considered as biomaterials. Allografts are taken from and given to individuals of the same species, but the donor and recipient are different individuals. Xenografts are interspecies tissue implants, thus the donor and recipient belong to different species. (Aho et al. 2003.) Regenerative medicine, i.e. the use of stem cell therapies, to restore lost, damaged, or aging cells and tissues in the human body, can be considered part of the category 3.

The focus of the next chapter is one of the bone implant applications. Intramedullary nailing is an important metal-based bone implant method and it has been one of the research topics of the present thesis work.

2.8.2 Intramedullary nailing

Intramedullary (IM) nailing is an operative approach for treatment of fractures. It allows fast recovery and reintegration, and is economically cost-effective with short hospitalization when compared to the traditional treatments such as traction and spica casting (body cast) that immobilizes the hips and thighs so that bones or tendons can heal properly (Buechsenschuetz et al. 2002). IM nails are used in fractured long bones (femur, tibia, humerus, ulna and radius) and also in unstable metacarpal and phalangeal shaft fractures. Intramedullary nailing is less invasive than extramedullary fixation of fractures with plating but the operation is rather technically demanding (Jiang et al. 2007). The nails stabilize the fracture fragments, allowing load transfer across the fracture site while maintaining anatomic alignment of the bone (Eveleigh 1995).

In principle, nailing can be performed either with locked or unlocked nails. The nails are either attached to the bone (locked) by interconnecting screws or they are designed to function without these fixation components (unlocked). IM nails have load-sharing characteristics (Cheung et al. 2004) and the amount of load supported by the nail depends on the stability of the fracture/implant construct (Bong et al. 2007). Unlocked nails can be considered as more biomechanically load-sharing devices and the locked nails as more load-bearing ones. In general, IM nails are assumed to bear most of the load after the operation, and then gradually transfer it to the healing bone (Bong et al. 2007). Large numbers of different commercial nails are available, but there are no universal guidelines stating the conditions at which each nail will perform at its optimum. The clinical choice of the nail type is strongly influenced by the bone type, and by the location and type of the fracture. Both locked and unlocked

55

methods are generally used and they have both advantages and disadvantages (Aro et al. 1982, Kempf et al. 1985, Lee et al. 2008). The two most frequently used materials for IM nails are titanium alloy and 316L stainless steel (Bong et al. 2007). Further information on the biomechanics and biology of IM nailing of the lower extremity can be found in the comprehensive article of Bong and colleagues (Bong et al. 2007).

IM femoral nailing is an old treatment for bone fractures. The concept of using a trochanteric entry point can be traced back at least to 1940, when the method was reintroduced by Küntscher (Küntscher 1940). The method – first described in 1918 by Hey Groves (Hey Groves 1918) – its modifications and possible limitations have been evaluated in many papers (de Belder 1968, Gonzalez-Herranz et al. 1995, Gadegone & Salphale 2006, Kanellopoulos et al. 2006). In contrast to antegrade nailing, the retrograde approach is used in various treatments, e.g. in distal femoral fractures (Janzing et al. 1998, Acharya & Rao 2006) and in obese patients (Herscovici & Whiteman 1996, Tucker et al. 2007). Femoral shaft fractures are one of the most common reasons for pediatric orthopaedic hospital admissions in many children’s hospitals (Buechsenschuetz et al. 2002). Flexible IM nailing seems to be safe for the treatment of these fractures (Carey & Galpin 1996), although there are some discrepancies in the nail material properties (Perez et al. 2007). In addition, retrograde elastic IM nail fixation can be a good choice in some pediatric distal-third femoral-shaft fractures (Mehlman et al. 2006), although the most common recommendation for these fractures is the antegrade insertion of the nail (Galpin et al. 1994, Carey & Galpin 1996, Huber et al. 1996, Bourdelat 1996).

The nailing site and the nail type for antegrade IM femoral implants has been a topic of discussion in the literature (Antonelli 1989, Miller et al. 1993, Gausepohl et al. 2002, Ostrum et al. 2005, Ricci et al. 2005), especially in the adolescent patient population (Townsend & Hoffinger 2000, Kanellopoulos et al. 2006, Caglar et al. 2006). The choice of entry point in antegrade nailing can have significant effect on the ease of nail insertion, the strength of the resulting fixation and on the inevitable soft tissue injury (Moein et al. 2005, Bong et al. 2007). According to one point of view, the fixation of the fractured femur by IM nailing in children less than 13 years of age should be altogether avoided, since it is practically impossible to avoid the proximal physeal system (Gonzalez-Herranz et al. 1995). There is also some discrepancies in the literature concerning the terminology of the entry sites, i.e. the piriform fossa and the trochanteric fossa (Papadakis et al. 2005). According to Papadakis and others (Papadakis et al.

56

2005, Bong et al. 2007), the groove of the greater trochanter, i.e. the trochanteric fossa, appears to be the standard entry point that most surgeons recommend (Fig. 10).

Fig. 10. The location of trochanteric fossa (t.f.) and piriformis fossa (p.f.) as viewed from the proximal part of the right femur: superior aspect.

Reaming of the intramedullary canal before nail insertion is part of a common protocol, which allows the insertion of a larger, stiffer nail with increased cortical contact. This obviously destructive procedure is considered to have positive effects on the outcome of the IM nailing. Reaming increases the local extraosseous and periosteal blood flow that is the most important source of nutrient flow during fracture healing. (Bong et al. 2007.)

57

2.9 Bone implant surface treatments

The main aim of bone implant coating methods is to improve the osseointegration of orthopaedic and dental implants. Osseointegration is crucial for long-term implant stability and inappropriate contact between implant and bone tissue results in loosening of the implant. Synthetic materials implanted into a biological environment induce host and foreign body reactions, and often also fibrous capsulation of the implant. The topic of implant loosening will be discussed in the following chapters. In order to postpone, and perhaps even inhibit, these unwanted reactions, new implant coating methods have been developed and the old ones further improved. Today, highly sophisticated material technology methods, together with other technological innovations, make it possible to tailor, even in nanoscale, the surface properties of an implant.

Two factors play an important role in osseointegration: primary stability (mechanical stability) and secondary stability (biological stability after bone remodeling) (Berglundh et al. 2003). Primary stability is the mechanical stability of the implant as soon as the implant is placed into the bone. It gradually decreases in the bone remodeling process. Secondary stability is the contact of new bone with the implant after bone remodeling. Primary stability is fully replaced by secondary stability when the healing process is completed. (Jimbo et al. 2007.) Implant coating can possibly advance and/or enhance the secondary stability.

It is important to mention that in addition to the coating, the material’s surface roughness itself has an effect on biocompatibility, such as on the ability to adsorb proteins like fibronectin (Luthen et al. 2005), and in that way improve the implant’s osseointegration. Micro/macroporous implant surfaces can be applied to directly enhance tissue ingrowth (Ryan et al. 2006). The mechanical properties of an implant, such as the flexibility of the substrate have an impact on the cell–biomaterial interaction (Pelham & Wang 1997).

The idea of coating biomaterials to improve their bioperformance is not new. As already mentioned, biomaterial research has existed as an independent research area for ca. 50 years and publications about surface treatments of biomaterials began to appear 30 years ago. Quick search from the PubMed database (on November 2007) gives over 3000 hits for the search term “biomaterial coating”. Searching solely with the term “biomaterial” gives approximately 50300 hits, including over 4200 reviews, and more than 3300 of

58

the papers are published solely in the year 2007. Therefore it is clear that biomaterials and their coating methods are an expanding area of research.

The table and its reference list in appendices 1 and 2 provide an overview of bone implant surface treatments. In the table of appendix 1, the coating type, coating substrates, possible modifications or additives of the coating and coating methods are given. It also summarizes the main aims of bone implant coatings, the most important being improved osseointegration. The classification used in the table is just one of the many possibilities. In this case, the grouping was done according to the implant surface processing (i.e. the coating). Coated substrates and coating methods are named in the table as they were presented in the original publications. Because of this, different names can be used to refer to principally the same method or substrate. For example, titanium alloys or TiAl6V4 alloy can be designated titanium in some of the publications. The references of Appendix 1 are summarized in Appendix 2.

One of the most recognized bone implant coating materials is hydroxyapatite. The osseoconductive and -inductive effects of HA have been demonstrated in experimental and clinical studies. However, there are ongoing developments to improve the quality of HA coatings. The poor bonding strength between ceramic HA and metal implants has been a concern. Since HA and other CaPs are the most studied coating materials for bone implants, appendices 1 and 2 can only provide a short overview of the literature of HA coatings, which covers hundreds of published papers. Although widely studied, it seems that the amount of research on mere HA and other CaP coatings is decreasing, while new methods, such as tailored protein coatings are gaining more interest in the research field.

Biomaterial surface modifications use ECM components in order to achieve more natural-like implant surfaces facing the biological milieu. These coatings have restrictions, since full-length proteins have a wide range of potential targets, i.e. their effect is more or less nonspecific, and they can induce immunogenicity. Short synthetic analogs that present the bioactive motif of a whole molecule have gained considerable attention in the past years. The most common of these analogs are the RGD-sequence peptides, which mediate cell attachment to several matrix proteins. The results of RGD-peptide coatings are controversial. Even though the actions of RGD-peptides are specific for integrins, they lack selectivity among integrins and thus they can trigger a wide range of interactions. Therefore, engineering peptides that selectively target specific integrins, thus inducing specific cellular responses, may be more advantageous in surface coating methods.

59

2.10 Host response

The implantation of a biomaterial into tissue results in an injury and initiation of the inflammatory response. A host response to biomaterials is an unavoidable series of events that occur whenever tissue homeostasis is disturbed by the implantation process. Even the most biocompatible materials trigger unwanted reactions such as inflammation, fibrosis, infection and thrombosis, at least to some extent. The host response following implantation of a medical device proceeds through nine consecutive events: injury, blood–biomaterial interactions, provisional matrix formation, acute inflammation, chronic inflammation, formation of granulation tissue, foreign body reaction and fibrosis/fibrous capsule formation. It is important to note, that in the normal physiological process of wound healing, i.e. without the presence of an intrusive implant, no non-specific protein adsorption occurs. The normal events of acute inflammation, the formation of granulation and eventual scar formation are all highly complex and well-coordinated processes that occur in wound healing in the same way regardless of the cause of wounding. In contrast, wounding with the presence of an implant is an interplay of normal physiological responses and “nonphysiological” protein adsorption dictated by the implant properties. This may lead to the upregulation of proinflammatory processes and result in chronic inflammation. (Anderson 2001.)

The adhesion of monocytes to the naturally protein-modified biomaterial surface and concomitant phenotypic change to macrophages and their fusion to multinucleated giant cells is a key event in biocompatibility. This process is complicated and event-rich and parts of its molecular and cellular basis are still unknown (Anderson 2000). Giant cells formed in the presence of foreign bodies are called foreign body giant cells. These cells are observed to interface with e.g. orthopaedic prostheses for periods extending up to 15 years (Anderson 1988). This indicates ongoing chronic inflammation at biomaterial sites for prolonged time periods. The diagram in figure 11 shows some important aspects of host tissue reactions to implanted synthetic materials (Keselowsky et al. 2007a).

60

Fig. 11. Illustration of host response to biomaterials. The figure represents a simplified diagram of the processes occurring after implantation of synthetic material.

It is not in the scope of the present study to go into further details of the host response to biomaterials, but more detailed information on this topic can be found in the following articles on the pathology of implant failure (Bauer & Schils 1999), coagulation and complement (Gorbet & Sefton 2004), molecular basis of foreign body reaction (Tang & Eaton 1993, Hu et al. 2001), connective tissue formation (Anderson 2001) and on monocytes/macrophages/foreign body giant cell activity after implantation (McNally & Anderson 1994, McNally & Anderson 1995, McNally & Anderson 2002).

61

2.11 Adaptive bone remodeling and loosening of orthopaedic bone implants

Despite numerous biomaterials and their applications, non-biodegradable bone implants have an overall tendency to detach over time. This “over time” can be as long as many decades, but even today, no truly non-loosening application exists. Even if an implant is strongly osteointegrative, and thus forms a strong attachment to peri-implant bone, the implant is always at risk of loosening.

Before going into detail on aseptic bone implant loosening, it should be noted that infection is one cause of implant failure (Gristina & Costerton 1985). After total joint arthroplasty operations, the frequency of infection is about 1–5%, the prevalence being smaller in hip arthroplasties than in knee arthroplasties, and they are more likely to occur after revision surgery (Bauer & Schils 1999). These infections are affected by many factors, such as the complexity of the surgical procedure and the characteristics of the operating room. Overall, bacteria (such as Staphylococcus epidermidis) attach to the implant surface, including titanium and polymethylmethacrylate cement. The bacterial fauna is the cause of infection that usually takes place quite soon after the implantation. The division of implant failure to aseptic and micro-organism based reasons is not completely clear, since there is still some debate over the long term effect of bacteria on implant loosening (Nelson et al. 2005). Implant-related infections can be further divided into early (contamination-related), and intermediate and late infections of hematogenetic origin.

2.11.1 Peri-implant osteolysis

There are a number of causes for implant failure, but the most important is the aseptic osteolysis occurring because of a chronic inflammatory response to implant derived wear particles. A vast number of implant design modifications have tried to overcome this problem, but aseptic osteolysis is still the limiting factor of implant longevity. For example, aseptic loosening of cemented implants causes more than 80% of clinical revisions (Erli et al. 2003).

Implant wear can be defined as a loss of material with the generation of particles resulting from relative motion between two opposed surfaces (Wright & Goodman 1996). Small particles that detach from the implant as a result of abrasion, adhesion or fatigue induce a very complex, and partially unknown, mechanism of action that leads to the activation of the inflammatory response and

62

ultimately to implant loosening. There are many modes of wear involved in implant damage (McKellop 1995) and the damage to implants, such as components of total joint arthroplasty, are rarely caused by a single wear mechanism.

The major cause of bone resorption around bone implants is the inflammatory reaction to particles of wear debris (Schmalzried et al. 1992, Wright & Goodman 1996). Adherent endotoxin, such as lipopolysaccharide (LPS), the primary outer cell wall component of Gram-negative bacteria, might also be involved in many of the effects of orthopaedic wear particles (Bi et al. 2001). Osteoclasts are the major cause of the bone lysis leading to implant failure (Haynes et al. 2004). In vitro studies have shown that cells incubated with wear particles produce cytokines that participate in osteoclast differentiation and activation. Osteolysis of peri-implant bone results indirectly from an inflammatory reaction amplified by macrophages in response to the phagocytosis of debris particles (Santavirta et al. 1990a, Santavirta et al. 1990b, Jiranek et al. 1993, Jiranek et al. 1995). The mechanisms of inflammation-induced bone loss are partly unknown, and recent studies have attempted to determine the exact molecular biology behind these reactions (Alnaeeli et al. 2007).

The initial response to particulate wear debris includes a subtle inflammatory response, which becomes more pronounced as osteolysis progresses. The inflammatory environment provokes a cellular response characterized by elevated levels of factors such as TNFs (tumor necrosis factors), RANK/RANKL, OPG, and interleukins IL-6, IL-1 and IL-11. Most of these affect osteoclast differentiation and activity directly, and result in enhanced osteolysis. (Konttinen et al. 2005.) Figure 12 shows some of the important cells and pro-inflammatory factors that are involved in bone resorption due to implant wear debris.

63

Fig. 12. Diagram of the cells and pro-inflammatory cytokines thought to be involved in biomaterial derived debris mediated bone resorption. MMPs = matrix metalloproteinases, IL-6/1-β/10 = interleukin-6/1-β/10, PGE2 = prostaglandin-E2, MCP-1 = monocyte chemoattractant protein-1, TNF-α = tumor necrosis factor α, CFU-GM = colony-forming units granulocyte-macrophage, TGF-β = transforming growth factor-β, OPG = osteoprotegrin, RANK/RANKL = receptor activator of nuclear factor κB / ligand, FGF = fibroblast growth factor, PDGF, platelet-derived growth factor. Modified from (Bauer & Schils 1999).

Although the majority of the wear debris particles remain in the tissues around the implant, some of these may enter the lymphatic vessels and be transported to lymph nodes and other tissues (Jacobs et al. 1991). In addition to debris particles, failing implants can give rise to ions that are systemically distributed and can be detected from blood and urine (Bauer et al. 1993, Stulberg et al. 1994).

64

2.11.2 Mechanobiological regulation

Since 1638 it has been know that bone responds structurally to mechanical load (Galilei 1638), a phenomenon later known as Wolff’s law (Wolff 1892). After bone tissue is subjected to an implant, the distribution of stress changes in the bone. Bone tissue adapts to this new situation by sensing the changes in the mechanical stimuli and responding with the appropriate balance between bone formation and resorption. As described in previous chapters, mechano-transduction in bone is a complex phenomenon (Bonewald 2006, Aguirre et al. 2007, Malone et al. 2007, Tatsumi et al. 2007).

The success of implantation depends on the initial fixation and the maintenance of adequate fixation over time. Maintenance of the initial mechanical fixation depends in part on maintaining local bone mass and minimizing bone resorption around the implant. The factors that influence bone density around an implant are, on the bone formation side, appropriate load magnitude, and from the net bone resorption side, inappropriate load, inflammatory reactions, implant motion and hydrodynamic pressure (Bauer & Schils 1999).

As mentioned, the distribution of stress changes when an implant is inserted into the bone. For example, if a conventional femoral stem implant is inserted into the medulla, the proximal bone is subjected to less stress as load is transmitted from the middle or distal aspects of the implant stem to the diaphyseal cortex. This is called stress shielding, since the implant shields the proximal bone from stress (Harris & Tarr 1979, Engh et al. 1992). The result of stress shielding is resorption of bone in the “shielded” areas (Mueller et al. 2007). The magnitude of this adaptive bone remodeling is determined by the magnitude of change in the mechanical load, which in turn is influenced by the mechanical properties of the implant (Huiskes et al. 1989, Sumner & Galante 1992, Glassman et al. 2006). Thus, stress shielding caused by the implant can be affected by biomaterial design. For example, manufacturing porous materials with a tailored elastic modulus can be used to create implants which match better to the elastic modulus of bone, thus reducing the problems associated with stress shielding (Ryan et al. 2006). Although stress shielding influences local bone remodeling, it is not an isolated cause of implant loosening. The process of adaptive bone remodeling is of multi-factorial origin (Sychterz & Engh 1996, Braun et al. 2003), where direct connection between the cause and effect resulting in implant loosening cannot readily be made.

65

Micromotion of implants is an important mechanism of loosening. Repetitive loading can cause damage to the implant material or to the interface between the implant and bone (Huiskes 1993). Even though bone can remodel itself in response to altered mechanical load, excess accumulation of mechanical damage can result in destruction of the implant−bone bond, and result in micromotion of the implant. Implant motion is associated with bone resorption, even in the absence of wear debris (Aspenberg & Herbertsson 1996), although micromotion is not thought to be solely responsible for large and erosive patterns of peri-implant osteolysis (Jasty et al. 1986). Total joint implants that do not achieve adequate initial fixation are most likely to fail because of micromotion in response to load (Ryd et al. 1995). Implant micromotion can also generate hydrodynamic pressure that might further assist the loosening progress through osteocyte mediated increase of bone resorption (Bonewald 2006).

66

67

3 Outline of the present study

This study has concentrated on investigation of the effects of austenite and martensite phases on bone cells and on implant coating as a way to improve and guide cell and tissue reactions after implantation.

The specific aims of the present research were:

1. to evaluate the effect of austenite and martensite phases of NiTi on osteoclast responses, and to study whether osteoclasts can be used as sensitive cell probes to investigate the effects of the fine physical differences between the two NiTi phases (I)

2. to investigate the effect of different levels of NiTi surface oxidation on osteoblastic cell attachment and survival in order to study the eventual effect of the thickness of the oxide layer on these cellular functions (II)

3. to determine if the two phases of NiTi have different effects on osteoblast attachment, proliferation and stage of the cell cycle in response to a fibronectin coating, and to the fibronectin molecule itself (III)

4. to further investigate the usefulness of functional NiTi intramedullary nailing and the possible advantages of coating the implant with a sol-gel derived TiO2-SiO2 ceramic thin film (IV).

68

69

4 Material and methods

4.1 Test Materials

NiTi test materials and possible control materials used in the original studies I–IV are briefly explained in the following chapters.

4.1.1 NiTi samples – in vitro (I–III)

Study I: Two disks 40 mm-in diameter and 5 mm in height were made of hot-rolled binary NiTi. In the austenitic sample, the percentages of the elements were 44 wt-% titanium and 56 wt-% nickel. In the martensitic sample, the alloy composition was 45 wt-% titanium and 55 wt-% nickel. Sonicated cortical bovine bone slices (A = 0.5 cm2; 150 μm thick) and autoclaved standard glass cover slips were used as control materials.

Study II: Fifteen disks of six mm and one of 40 mm-in diameter (5 mm height) samples were made of hot-rolled binary NiTi. The samples were composed of 44.3% Ti and 55.7% Ni by weight, thus having the austenite phase at the experimental temperatures. Six of the small sample disks were left untreated, having only the natural thin oxide layer at the surface. The rest of the small disks were oxidized in air either at 350°C (straw-colored oxide, SCO; 5 samples) or at 450°C (blue oxide, BO; 4 samples) for two hours. The 40 mm-in diameter sample was oxidized at 600°C during two hours. In addition, special flat NiTi control samples (80 × 10 × 2 mm3 and 15 × 15 × 2 mm3) and samples of thin (0.5 mm in thickness) NiTi wire were used for the study of electrical properties.

Study III: NiTi sample disks of 20 mm-in diameter and 5 mm height were made with the following melted compositions: alloy I (austenite) – 44% titanium and 56% nickel, alloy II (martensite) – 46% titanium and 54% nickel (by weight). Six samples disks of both phase type were made. Differential scanning calorimetric (DSC) measurements were carried out (DSC 822e, Mettler-Toledo Corporation) to determine the characteristic martensitic transformation (MT) temperatures, which were: Ms = 15°C; Mf = 6°C; As = 23°C; Af = 32°C for the alloy I and Ms = 65°C; Mf = 45°C; As = 70°C; Af = 94°C for the alloy II. Tissue culture polystyrene (TCPS) and carboxymethylcellulose (CEL, as an anti-adhesive coatings (Hindie et al. 2005, Hindie et al. 2006), were used as control materials in part of the experiments.

70

All NiTi samples used in the original studies I–III, were produced in the Institute of Light Metals (Moscow, Russia). To provide equal surface roughness, the samples were ground by the Exakt® micro grinding system (Exakt, Kulzer, Germany) with carbide silicon of decreasing grade (from 150 to 1200), followed by manual polishing with a rubber wheel and chromium oxide paste. Before use in the experiments, the samples were washed with detergent, rinsed and degreased 10 minutes with 70% ethanol. Finally, the samples were washed in ultrasonic bath and autoclaved for 30 minutes at 120°C.

4.1.2 NiTi samples – in vivo (IV)

The material used was NiTi (44.3% Ti and 55.7% Ni by weight) fabricated to a wire with a diameter of 1.4 mm. The initial round pieces of curvature radii of 15 and 28 mm were made by constraining the wires in special mandrels while annealing at 450°C. Intramedullary nails of 26 mm in length were cut from the round pieces. Straight nails were made from the unconstrained wires.

To determine characteristic MT temperatures and thus, to be sure of the samples phase state in the range of experimental temperatures used (martensite at 0°C and fully austenite at body temperature), differential scanning calorimetric measurements were carried out (DSC 822e). The MT temperatures determined were: Ms = 15°C; Mf = 6°C; As = 23°C and Af = 32°C.

Part of the NiTi samples were coated with TiO2-SiO2 (molar ratio of 70:30) as described in (Jokinen et al. 1998, Ääritalo et al. 2007). The sol-gel method of coating is a wet-chemical technique that starts from a chemical solution that reacts to produce colloidal particles (sol). The colloid sol is composed of solid particles ranging from 1 nm to 1 µm dispersed in a solvent. The sol further evolves to an inorganic network containing liquid phase (gel). Drying of the gel removes the liquid phase, thus forming a porous material, which can be further processed in order to improve its mechanical and other properties. In this study the TiO2-SiO2 sol was prepared by mixing TiO2 and SiO2 to obtain the predetermined molar ratio. SiO2 was prepared by mixing tetra-ethyl ortho-silicate, ethanol and water at room temperature having the EtOH:Si(OR)4, and H2O:Si(OR)4 molar ratios of 3.7 and 0.27, respectively. The prepared sol was aged at 40°C for 60 minutes. TiO2 was prepared by mixing tetra-isopropyl titanate, ethanol, nitric acid, and water at room temperature and aged at 40°C for 30 minutes having EtOH:Ti(OR)4, H2O:Ti(OR)4 and HNO3:Ti(OR)4 molar ratios of 3.4, 0.11 and 0.70, respectively. The prepared TiO2-SiO2 sols were aged at

71

40°C for 24 hours and cooled to 0°C before the dipping process. The NiTi samples were dipped into the sols and withdrawn at a speed of 0.3 mm/s at ambient temperature. After that the substrates were calcined at 500°C for 10 minutes, cooled and washed ultrasonically 5 min in acetone and 5 min in ethanol, then dried in air. This cycle was repeated five times, which resulted in films having a thickness of roughly 280 nm. (Jokinen et al. 1998.) The proper attachment and durability of the sol-gel derived coating on the NiTi wires during bending was confirmed by SEM-imaging (JEOL JSM 5500, Tokyo, Japan).

After the surface finishing, the implants were degreased with 70% ethanol and autoclaved for 30 minutes at 120°C, after which they were stored in sterile packages at room temperature. The chemical compound and the technological itinerary of the alloy resulted in implants that could be deformed at about 0°C (fully martensitic state) and that restored their initial shape at about 25–30°C (fully austenitic state). Table 5 summarizes the implant types used in the original study IV.

Table 5. NiTi implants used in the original study IV. The nail length was 26 mm and thickness 1.4 mm in every case.

Group name Sol-gel derived coating Curvature radius (mm) n

I – 28 7

I-sol-gel TiO2-SiO2 28 7

II – 15 7

II-sol-gel TiO2-SiO2 15 7

STR – 0 6

STR-sol-gel TiO2-SiO2 0 6

4.1.3 Animals (IV)

Forty male Sprague-Dawley rats (Laboratory Animal Center, University of Oulu, Finland) were randomized into six groups. The age of the animals was 14 weeks and the mean weight (± SD) 385 ± 26 g. The animals were housed in groups of 3–4 in Macrolon IV polycarbonate cages in a thermostatically controlled room at 20 ± 1°C with a relative humidity of 50 ± 10%. The room was artificially illuminated with a light and dark cycle of 12 hours. Aspen chips (Fintapway, Finland) were used as bedding. A wooden tunnel made of the same material as the bedding chips was provided to the animals as a stimulus-toy for the improvement of animal welfare. In addition, cellulose paper was offered for fulfilling the rats nesting

72

behavior. Pelleted rat feed (SDS R3 (E), Special Diet Service Ltd., Great Britain) and tap water were available ad libitum. The animal tests were performed after approval by the National Laboratory Animal Center. All aspects of animal care complied with the Animal Welfare Act and the recommendations of the NIH-PHS Guide for the Care and Use of Laboratory animals.

4.2 Physical measurements

All the physical measurements performed in the original articles I–III are briefly summarized in the following chapters.

4.2.1 Surface roughness

Atomic force microscopy (I, II)

The atomic force microscope (AFM) is a high-resolution type of scanning probe microscope with demonstrated resolution of only a fraction of a nanometer. The term 'microscope' in the name is actually a misnomer because it implies looking, while in fact the information is gathered by "feeling" the surface with a mechanical probe. The probe is raster scanned across the surface under study while a constant force is maintained between the probe and sample. By monitoring the motion of the scanning probe, a three-dimensional image of the surface can be constructed.

Surface roughness of the NiTi samples was measured with AFM (Explorer, Thermomicroscopes, Sunnyvale, CA, USA). Images were collected in the contact mode using triangular silicon nitride cantilevers (Veeco Instruments, USA) with a z-scanner of 13.6 µm range. Image resolution was 300 × 300. SPMLabNT software version 5.01 (Thermomicroscopes, Sunnyvale, CA, USA) was used to calculate the average surface roughness from first-order plane leveled images. In study I, the image size analyzed was 100 × 50 µm and the number of images analyzed was 10 for both sample types. In study II, the image size was 100 × 90 µm. The number of images analyzed was 40 for austenite, 60 for BO and 80 for SCO.

73

Phase-contrast interferential microscopy (III)

Interferential microscopy is a well-known technique, with a simple founding principle: a reference mirror is associated to the surface under study in order to produce a luminous interference, from which a huge number of fringes describing the surface can be obtained. A surface-relief map is produced by computer-based treatment of the images obtained from the surface.

Even though the interference-based technique provides considerable information about a surface, the method leaves the slope sign of the analyzed profile unknown. This problem can be overcome with a phase-contrast interferential microscopy technique, which describes the surface roughness in terms of statistical parameters, such as the root mean square (RMS) roughness.

In the original study III, surface roughness was measured using a surface profiler (Alpha-Step 500, Tencor Instruments, San Jose, California, U.S.A), which computes the roughness parameters of the Z range (the difference between Zmax, the maximum amplitude of roughness, and Zmin, the minimum amplitude of roughness) and the RMS roughness (R). The RMS, corresponding to the standard deviation of the Z values within a given area, is determined from the equation below:

2( )i aveZ Z

RN−

= ∑ , (1)

where Zave is the average of Z values within a given area, Zi the current Z value and N the number of points within a given area.

The RMS values depend only on the amplitude of the peaks/troughs, and therefore do not include any quantitative information about how frequently such irregularities appear on the material’s surface. A computational post-treatment of the roughness profiles can be implemented in order to obtain more relevant roughness estimates. After selecting the most convenient position for setting two perpendicular axes, the computer identifies every single peak/trough present along the profile, storing the corresponding coordinates in a linear array of data. An approximated profile is then generated by simply connecting such points with straight lines, whose individual lengths are progressively added until a quantitative representation is obtained, not only of the height of the peaks/troughs, but also of the frequency with which they appear. In this study, 81 images from austenite and 39 from martensite were analyzed.

74

4.2.2 Contact angle measurements and surface free energy calculations (I, III)

Contact angle analysis characterizes the wettability of a surface by measuring the surface tension of a solvent droplet at its interface with a homogenous surface. In other words, contact angle measures the attraction of molecules within the droplet to each other versus the attraction or repulsion those molecules experience towards the molecules of the surface under study. There are many methods and modifications to measure contact angles, but in all the methods, the contact angle (Ө) is the angle of the liquid at the interface relative to the plane of the test surface (Fig. 13).

Fig. 13. Schematic demonstration of the difference in contact angle of a liquid droplet between (A) hydrophobic and (B) hydrophilic surfaces.

The sessile drop method (measurement of static contact angles) is an optical contact angle method used to estimate wetting properties of a localized region on a solid surface. The static contact angle is the contact angle when all participating

75

phases, as in the studies presented here i.e. gas, liquid and solid, have reached their natural equilibrium positions, that is to say, when the phase lines of the three phases no longer move. (Neumann & Godd 1979, Ulman 1991, Ratner et al. 1996.)

Measurement of contact angles yield data that reflect the thermodynamics of a liquid/solid interaction and characterize the wetting behavior of a particular liquid/solid pair. The thermodynamics of a solid surface can be characterized with various methods, which all are based on the same principle: a solid is tested against a series of liquids and the contact angles are measured. Calculations based on these measurements produce a parameter (critical surface tension or surface free energy), which quantifies the characteristics of the solid and mediates the properties of the solid substrate. The surface free energy obtained in this way can be regarded as the "surface tension" of the solid substrate, which is a characteristic property of the solid in the same way as the surface tension is for a liquid; defining the work required to increase the area of a substance by one unit area. (KSV Instruments 2008.)

Study I: A digital optical contact angle meter CAM 200 equipped with Optical contact angle and pendant drop surface tension software, version 3.50 (KSV Instruments Ltd., Finland) was used for measuring static contact angles.

Briefly, a series of images of the sessile drop placed at the sample surface at +21 ± 1°C was collected for 20 seconds at 1-second intervals. The left and right contact angles of each drop were automatically calculated and the contact angle of each image was determined as an average of the two angles. The final contact angles used for the surface free energy calculation were the averages of the contact angles measured between 10 and 20 seconds. Five liquids, whose surface tensions are known, were used: deionized, autoclaved water, glycerol (BDH, England), dimethyl sulfoxide (Merck, Germany), ethylene glycol (Merck) and diiodine methane (Fluka, Germany). Each liquid was measured eight times. NiTi samples were cleaned between measurements with 70% ethanol, wiped with cellulose wadding, blown by compressed air and let dry in air for 3 minutes.

The surface free energy parameters were calculated using the average static contact angles of the test liquids with CAM 200 Surface free energy analysis software, version 2.41 (KSV Instruments Ltd, Finland), using a harmonic mean approach (Wu 1971).

Study III: Contact angle measurements were performed using the sessile drop method with a goniometer (Carl Zeiss, Jena, Germany). Drop images of three different liquids: water (reference liquid), ethylene glycol and diiodine methane

76

were collected in room temperature with a digital camera and analyzed by using a curve fitting method and tangent approximation.

The material’s surface energy was calculated using the three different liquids mentioned above, whose surface tensions are known. For every single sample, a total of nine values were obtained. For surface energy values calculations, the theoretical approach called the Lifshitz-van der Waal approach or acid-base approach was employed, which is presented below:

( )1 cos 2 2 2LW LWL i S L S L S Lγ θ γ γ γ γ γ γ+ − − ++ = ⋅ + ⋅ + ⋅ , (2)

where θi is contact angle corresponding to liquid I, γLLW, γL

+, γL- energy

components of liquid I and γSLW, γS

+, γS- energy components of the surface.

γ describes the total surface free energy. γLW is the Lifshitz-van der Waals component of the surface energy. The Lewis acid-base component is further divided into acidic electron-acceptor (γ+) and basic electron-donating (γ-) parameters. (Gindl et al. 2001, Sharma & Rao 2002, Chibowski & Perea-Carpio 2002, Mykhaylyk et al. 2003.)

4.2.3 Analysis of the surface oxide layer (II)

X-ray diffraction analysis

X-rays are electromagnetic radiation of wavelength about 1 Å, which is about the same size as an atom. This enables the study of the atomic and molecular structure of crystalline substances. X-ray diffraction is one of the most important characterization tools used in materials science. In principle, a sample is exposed to X-rays at various angles and the diffraction patterns produced are compared with reference standards for identification.

The phase composition of the oxidized NiTi samples was controlled by X-ray diffraction analysis (XDA), using filtered CuKα radiation. To analyze the surface structure, a grazing beam of 1o incidence was used. The crystal lattice parameters were determined by extrapolation to θ = 90o using the extrapolating function f(θ) = 1/2(cos2θ / θ+cos2θ / sinθ). The shifts of diffraction lines under the grazing beam were controlled by experiments with a thin layer of vacuum annealed fine tungsten powder at the surface of the oxidized samples. The diffraction peaks that did not correspond to austenite, martensite or Ni3Ti were used as primary

77

information for interpretation of the surface oxides phase compositions and structure.

Electrical resistance measurements

The electrical properties of a material, especially its surface electrical properties have a strong influence on a material's adsorption ability. This is an important aspect of a material’s biocompatibility, since the initial interactions of a biomaterial surface facing a biological environment (i.e. electrically charged macromolecules) are underlined by their electrical properties. Thus electrical properties of the surface have a tremendous effect on the chemical, biological and, ultimately, on tissue reactions.

The differences in electrical properties of the oxidized NiTi samples were assessed by changes in electrical resistance before and after oxidation. The measurements of electrical resistance were performed with an industrial digital milliohm meter MEGGER BT51 (Megger, USA). The specific resistance of the material surface layers after oxidation was determined by gradual etching with simultaneous control of changes in the sample's electrical resistance and thickness, considering that the resistances of the layer removed and the remaining material as parallel connected.

Calculation of the oxide thickness

Oxide thickness (δ) after two hours of oxidation was computed from the sample mass gain per area unit in accordance with the formula δ=Δm/ρ, where Δm is mass augmentation per sample area unit and ρ is oxide density. This approach was based on the assumption that the mass gain was a result of oxygen adsorption only, and that all of this oxygen was used to produce the TiO2 oxide. The validity of this assumption was confirmed by the results of X-ray diffraction (XRD) analysis, which demonstrated the predominant content of TiO2 (rutile) in the sample surface after oxidation. The value of ρ = 4.2 g/cm3 for stoichiometric rutile TiO2 was used (ASTM Diffraction Data File 1969).

To provide higher accuracy, the values of mass augmentations were interpolated from the equations fitted to the experimental data mass gain vs. oxidation time. The experimental oxidation curves were described by the following equations:

78

Δm = 0.2205 × 10-4 × lg t for oxidation at 350°C, (3)

Δm1.67 = 0.83311 × 10-8 × t for oxidation at 450°C, (4)

Δm2 = 4.744 × 10-8 × t for oxidation at 600°C. (5)

The related augmentations of oxide thickness (δ) after two hours of oxidation were calculated using the mass augmentations from the equations 3–5.

4.3 Cell lines (I–III)

Study I: The original method of osteoclast isolation and culture was slightly modified (Boyde et al. 1984, Chambers et al. 1984, Lakkakorpi et al. 1989). 1- or 2-day-old Sprague-Dawley rat pups were sacrificed by decapitation. A total of seven rat pups were used in one experiment. Femora, tibiae and humeri were dissected out and quickly rinsed in 70% ethanol. Bones were submerged in sterile phosphate buffered saline (PBS) and the soft tissues were removed around the bones as thoroughly as possible. The bones with their cartilage ends were placed into 10 ml of alpha minimal essential medium (αMEM; BioWhittaker, Belgium) and cut in half in a longitudinal direction. The bone cavity was scraped and the tissue harvested in the culture medium. The tissue mass was suspended and the suspension centrifuged at 1000 rpm for 7 minutes. The sediment was dissolved in 10 ml of αMEM and 50 µl, 100 µl and 500 µl of the suspension was placed on a cortical bovine bone slice, glass coverslip or NiTi sample, respectively. The samples were submerged in the culture medium before adding the cell suspension. After 30 minute incubation at +37°C (5% CO2, 95% air), non-attached cells were rinsed away. The remaining cells on the substrates were cultured for 48 hours in αMEM buffered with 20 mM Hepes containing 0.84 g sodium bicarbonate/liter, 2 mM L-glutamine (Gibco), 100 IU of penicillin/ml (Sigma-Aldrich), and 7-10% heat-inactivated fetal bovine serum (FBS; Bioclear), at +37°C (5% CO2, 95% air).

Study II: The rat osteosarcoma cell line ROS-17/2.8 (Merck Research Laboratories, West Point, PA, USA) was cultured at 37oC in a humidified atmosphere (5% CO2, 95% air) in αMEM (Gibco), supplemented with 10% FBS (Bioclear), antibiotics (100 U of penicillin/ml, 100 µg of streptomycin/ml; Sigma-Aldrich) and L-glutamine (2 mM; Gibco). This cell line closely resembles differentiated osteoblasts (Majeska et al. 1980, Thiede et al. 1988).

79

Study III: An immature osteoblast-like mouse cell line, MC3T3-E1 Subclone 4 (ATCC CRL-2593), was used for these experiments. The cells were cultured under conditions similar to those described above in study II.

Between cell culture experiments (I, II), the NiTi samples were gently wiped with cellulose paper, washed with detergent, rinsed in sterile water and degreased 10 minutes with 70% ethanol. In the original study III, after wiping the sample surface with cellulose paper and washing with sterile water, the samples were incubated in a 2.5% sodium dodecyl sulfate (SDS, in PBS) solution at 37°C for one hour, followed by rinsing several times with sterile water. Finally, the samples were washed in an ultrasonic bath and autoclaved for 30 minutes at 120°C (I–III). In addition, in study III, energy dispersive X-ray analysis (EDX) was used to detect possible contamination left from the cell culture on the material surface after the cleaning procedure.

4.4 Surgical procedure (IV)

Anesthesia was induced with 4.5% Isoflurane (Forene®, inhalation vapor, Abbot Scandinavia AB, Sweden) and maintained during the operation with a 2.5% concentration in a special anesthesia unit (Univentor 400 Anaesthesia Unit, Zejtum, Malta). Immediately after the induction of anesthesia, rats received Buprenorfin 0.3 mg/kg subcutaneously (Temgesic® 0.3 mg/ml, Reckitt Benckiser Healthcare Ltd., Hull, Great Britain) as an analgesic.

Before the operation, all the IM NiTi nails and other instruments used during the implantation were submerged in sterile iced saline to reach and maintain the martensite form of NiTi throughout the implantation process.

A skin incision was made proximal to the greater trochanter and the entry site for the NiTi nails was made through the skin and muscle layers into the medullary canal via a hole drilled through the revealed trochanteric fossa. The femoral canal was reamed using various needles (up to 18-gauge) with a rotating motion. In their martensitic form (on ice bath), the bent nails were straightened and inserted quickly into a special implantation device modified from a bone anchor Mini Mitek® device (DePuy Mitek Inc., Raynham, Massachusetts, USA). The implants were pushed rapidly into the reamed medullary canal and allowed to settle in an incidental position. Straight nails were also implanted with the device. The wound was closed in layers using resorbable sutures (3-0 Vicryl®, Ethicon, Inc., Somerville, NJ, USA).

80

After the operation, the rats were kept isolated, but were allowed to move freely with no external support on the operated limb. During the three days of isolation, Buprenorfin (Temgesic®, 0.3 mg/ml, Reckitt Benckiser Healthcare Ltd., Hull, Great Britain) 0.06 mg/kg body weight subcutaneously was used as a postoperative analgesic twice a day. After the isolation period, the rats were grouped into groups of 3–4 animals and observed daily for activity and weight bearing on the implanted leg.

The experiment ended after 24 weeks and the rats were killed using carbon dioxide followed by cervical dislocation. The animals were sedated by anesthetic (Dormicum®, Roche, France: 5 mg/ml Midazolam and Hypnorm®, Janssen, Brussels, Belgium: 0.315 mg/ml Fentanyl citrate + 10 mg/ml Fluanisone) before euthanasia. All femurs with implants, and also tibias and fibulas were dissected, as were the contralateral limbs. The bones were fixed in 70% ethanol and stored at +4°C.

The rats were weighed before the operation (14-weeks of age) and after euthanasia (38-weeks old). The length of both operated and control femurs were measured using a digital vernier caliper.

4.5 Biological assays

The descriptions of the biological assays used in the original articles II and III are summarized in the following chapters.

4.5.1 Cytotoxicity test (II)

The ROS-17/2.8 cells cultured for 48 hours on the oxidized NiTi samples (5×103 cell/sample) were stained with a LIVE/DEAD®Viability/Cytotoxicity kit (Molecular Probes, Oregon, USA) for 10 minutes at 37°C. This assay is based on the simultaneous determination of live and dead cells with the detection of intracellular esterase activity by calceinAM and plasma membrane integrity for ethidium homodimer-1 (EthD-1). The optimal concentrations of the EthD-1 dye were 0.1 µM and for the calceinAM dye 1.0 µM (Kapanen et al. 2002b).

81

4.5.2 Preparation of fluorescein-fibronectin (FN-FITC) and FN-FITC coating of NiTi (III)

Fluorescein isothiocyanate (FITC; Sigma-Aldrich) was dissolved in dimethyl sulfoxide (DMSO, 10 mg/ml; Sigma-Aldrich) and 10 µl were incubated with 1 ml of human plasma fibronectin (FN, 1 mg/ml; Roche) in 0.1 M sodium bicarbonate buffer (pH 9.0) at room temperature for 1 hour. The resulting FN-FITC conjugate was separated from unconjugated dye using a SephadexTM G-25 PD-10 desalting column (Amersham Biosciences Limited, GE Healthcare) equilibrated with PBS. The FN-FITC concentration was estimated by measuring the absorbance at 280 nm. The final product was stored at +4°C.

500 µl of FN-FITC (20 µg/ml) solution was added to the NiTi surface (20 mm-in diameter) and the protein-dye conjugate coating was allowed to form for 30 minutes at +37°C in a humidified atmosphere. After PBS-rinsing, 500 µl of sterile filtered 1% bovine serum albumin (BSA) was added and incubated 30 minutes at room temperature, followed by washing with PBS.

MC3T3-E1 cells (6.8×104) were seeded on freshly made FN-FITC NiTi surfaces and cultured for 2 hours under conditions described in previous chapters. The cultures were stopped by adding 3% (w/v) paraform aldehyde (PFA) for 10 minutes at +4°C. The formation and rearrangement of fibronectin fibrils was imaged with confocal microscopy and digital image analysis, both methods are described in later chapters.

The 2-hour time point was chosen since it was considered to be long enough for fibronectin fibrillogenesis (Keselowsky et al. 2003b, Faucheux et al. 2006), but short enough to minimize the possible impact of matrix proteins secreted by osteoblasts and incorporated into the cellular microenvironment.

4.5.3 Enzyme-linked immunosorbent assay (ELISA) (III)

The NiTi samples were coated with 320 ng/cm2 of human plasma FN (in PBS; Roche) followed by blocking in a BSA-Tween solution (0.25% BSA and 0.1% Tween 20 in PBS), both for 30 minutes at +37°C. Tissue culture polystyrene and carboxymethylcellulose control samples were coated with 120 ng/cm2 of FN. These FN concentrations were chosen because they correspond to the respective saturation limit concentrations (beginning of plateau) of the anti-FN polyclonal antibody reaction.

82

The samples were incubated in a solution of anti-FN antibodies [polyclonal F3648 (Sigma-Aldrich), diluted 1/5000 in PBS; monoclonal antibodies 1926, 1935, 1936, 1892 (Chemicon) 1/5000] and with horseradish peroxidase-conjugated secondary antibodies [A6154 or Up 446330 (Sigma-Aldrich) 1/10000] both for 1 hour at 37°C. After BSA-Tween rinsing, 500 μl of chromogen solution [5 mg o-Phenylenediamine dihydrochloride (Sigma-Aldrich) dissolved in 10 ml of citrate buffer and 2 μl of hydrogen peroxide (30% v/v)] was added to each sample and incubated in the dark for 2 minutes at room temperature. The reaction products (100 μl/well) were transferred to a 96-well plate together with 100 μl of 1 M HCl. Absorbance (490 nm) was read in triplicate in a microwell plate reader. The test for each antibody was repeated 3–4 times. A control test was performed for all the samples to estimate the absorbance in the absence of the antibody.

4.5.4 Cell proliferation – LDH and MTT analyses (III)

Cell proliferation was studied in two ways: by LDH (lactate dehydrogenase) assay and by a MTT colorimetric assay (3-[4, 5-dimethylthiazole-2-yl]-2, 5-diphenyltetrazolium bromide). The assays were done from the same samples: LDH from the supernatant and MTT from the cells.

Lactate dehydrogenase is a cytoplasmic enzyme that catalyzes the interconversion of lactate to pyruvate. LDH's enzymatic activity is relatively stable in culture medium and can be measured easily after leakage out of nonviable cells with compromised membranes. The LDH assay, therefore, is a measure of cell membrane integrity. LDH released into the culture medium is measured with a coupled enzymatic assay that results in the conversion of resazurin into a fluorescent resorufin product. The amount of fluorescence produced is proportional to the leaked enzyme and thus, to cell damage. (Korzeniewski & Callewaert 1983, Decker & Lohmann-Matthes 1988.).

MTT is a yellow water-soluble tetrazolium salt that is reduced by live cells to a water-insoluble purple formazan that can be spectrophotometrically detected. The MTT assay measures the respiratory activity and thus, the cell viability (Alley et al. 1988).

The MC3T3-E1 cells (104 cells/cm2) seeded on FN-coated (320 ng/cm2) or uncoated NiTi and TCPS were incubated at 37°C for 2 and 48 h. The LDH assay (CytoTox-ONETM 96 non-radioactive cytotoxicity assay, Promega, Wisconsin, USA) was carried out in accordance with the manufacturer's instructions, from

83

100 µl of the culture supernatant. Absorbance was read at 490 nm in triplicate and three independent experiments were performed.

After removing the supernatant, 500 μl of MTT (0.5 mg/ml in PBS; Sigma-Aldrich) was added to the PBS-washed cell layer. The samples were incubated at 37°C for 3 h, the MTT solution was removed and the formazan crystals were solubilized in 500 μl of DMSO (Sigma-Aldrich). The reaction products were transferred to a 96-well plate for absorbance reading (550 nm).

4.5.5 Cell cycle analysis (III)

A cell cycle analysis was performed on the MC3T3-E1 cells (104 cells/cm2) grown on FN-coated (320 ng/cm2) or uncoated TCPS, austenite and martensite NiTi for 48 h. The cells were washed, removed by trypsinization and suspended in 250 µl of PBS containing 5 mM of ethylenediaminetetraacetic acid (EDTA, Prolab), and fixed by applying 750 µl of ice-cold absolute ethanol for 45 min at 4°C. The cells were washed and suspended in PBS-EDTA containing 0.1% Triton X-100 (Promega), mixed with 40 µg of RNase A (Sigma-Aldrich) and 25 µg of propidium iodide (Sigma-Aldrich), and incubated in the dark for 15 minutes. The stained samples were analyzed in an Epics XL-MCL flow cytometer (Beckman Coulter).

4.6 Immunofluorescence staining

Two staining protocols were used in original studies I and III. They are briefly described in the following chapters:

4.6.1 Staining of actin rings (I)

In nature, phalloidin is a toxin from the mushroom death cap (Amanita phalloides). It belongs to a large group of toxins known as phallotoxins. Phalloidin binds at the interface between filamentous actin (F-actin) subunits of actin, thus preventing its depolymerization, which results in poisoning the cell. This toxic property of phalloidin is a useful tool for investigating the distribution of F-actin in cells by labeling phalloidin with fluorescent tags, and using them to visualize the actin filaments.

In the original study I, staining of primary rat osteoclasts for F-actin was carried out by using tetramethyl-rhodamine isothiocyanate (TRITC)-conjugated

84

phalloidin (Sigma Chemical Co., St. Louis, MO, USA) for 20 minutes at room temperature. To visualize the nuclei, the cells were incubated with the DNA-binding fluorochrome Hoechst 33258 (1 mg/ml stock diluted 1/1000 in PBS; Sigma Chemical Co., St. Louis, MO, USA) for 10 minutes at room temperature.

4.6.2 Staining of focal adhesions (II)

The PFA-fixed, permeabilized (0.1% (v/v) Triton-X-100 in PBS) and BSA treated (0.2% BSA) ROS-17/2.8 cells were stained by using mouse monoclonal paxillin antibody (diluted 1/100 in PBS; ZYMED Laboratories, San-Francisco, CA, USA) for 45 minutes on +4°C. Staining was carried out with secondary antibody (1/100; ALEXA Fluor 488 goat anti-mouse IgG, Molecular Probes, Oregon, USA) for 30 minutes at +4°C and with DNA-binding fluorochrome Hoechst 33258 (Sigma Chemical Co., St. Louis, MO, USA) for 10 minutes at room temperature.

4.7 Histochemical staining and sample preparation

Histological staining and sample preparation methods used in the original studies I and IV, are as follows:

4.7.1 Staining of active osteoclasts (I)

Tartrate-resistant acid phosphatase (TRACP), a commonly accepted marker for osteoclasts (Minkin 1982) was used for visualization of the primary rat osteoclasts using a commercial TRACP kit (Sigma Chemical Co., St. Louis, MO, USA) according to the manufacturer’s instructions.

4.7.2 Histological thin sections (IV)

Preparation of histological thin sections of the NiTi functional IM nail implanted rat femurs were performed by the cutting-grinding technique for hard tissue as instructed by the manufacturer (EXAKT Appartebau, Germany).

In short, the operated and control femurs, stored in 70% ethanol at +4°C, were cut by the Exakt® cutting-grinding system. The first cut was made at the minor trochanter and the femur was cut into about 4 mm pieces until the distal end of the implant was reached. The cut bone pieces were dehydrated in a graded ethanol series and infiltrated in plastic infiltration liquid (Technovit® 7200 VLC,

85

Kulzer GmbH, Germany). After infiltration, the sample blocks were embedded and polymerized (Technovit® 7200 VLC), and finally, the blocks were attached between two parallel plastic slides and cut to about 100 µm thick sections by the Exakt® cutting-grinding system. After the final micro grinding by the Exakt® micro-grinding system to about 89 µm, the samples were stained with a standard Masson’s trichrome Goldner staining protocol.

4.8 Imaging and image analysis

All the imaging equipment and image analysis methods used in the original articles I–IV are briefly described in the following chapters.

4.8.1 Light and fluorescence microscopy (I)

The numbers of mononuclear and multinucleated TRACP-positive cells were counted, using a Leica DMLM microscope (NPLAN 20x/0.40BD objective). The cells were counted from 30 microscopy images of austenite and martensite NiTi from two experiments.

The primary osteoclasts stained for the visualization of actin rings were studied using a Nikon Eclipse 600 microscope with a Nikon Plan Fluor 20x/0.50 objective with an appropriate filterset. The numbers of actin rings were determined by manual counting from an area of 15 consecutive microscopy fields. The counting was repeated 15 times from different areas of the sample. Two parallel tests were done.

4.8.2 Confocal microscopy (I–III)

Actin rings (I), focal adhesions (II) and the amount of live and dead cells (II) were viewed in a confocal microscope LSM 510 equipped with an inverted microscope Axiovert 100M (Zeiss, Germany), with appropriated objectives. The same equipment was used to visualize the FN-FITC coating on the NiTi surface (III).

Actin rings (I), focal adhesions (II) and fibronectin rearrangement (III) were viewed with 20x (only in the study I) and 63x objectives (NA 1.2/water; Zeiss, Germany). The optical sections were scanned with a 1024 × 1024 frame size (pixel size 0.14 μm × 0.14 μm) close to the metal surface. The optical sections in study II were scanned next to the metal alloy surface at the level containing

86

approximately most of the focal adhesions. Images of actin rings (I) and fibronectin fibrils (III) were taken from the most descriptive focal plane, illustrating the structures of interest as clearly as possible.

The rat osteosarcoma cells stained for visualization of live and dead cells in the original study II were immediately viewed under the microscope with a 10x objective (NA 1.2/water). For each disk, the pictures were taken from randomly chosen areas (1024 × 1024 frame size, pixel size 0.9 × 0.9 µm).

4.8.3 Faxitron radiography (IV)

Radiographs of the dissected rat femurs of study IV were taken in anteroposterior (AP) projection on a Faxitron Specimen Radiography System (Model MX-20, Wheeling, IL, USA) with a high-resolution algorithm. The imaging parameters were identical for all the specimens and the following settings were used: 12 s, 30 kV; pixel matrix 1024 × 1054; resolution 200 pixels mm-1.

4.8.4 Digital image analysis (II–IV)

A digital image analyzer (MCID M4 v.3.0.re.1.1, Imaging Research Inc., Brock University, St. Catharines, Canada) was used for the analysis of the confocal microscopy (II, III) and Faxitron X-ray (IV) images.

Study II: The confocal microscope images of live and dead osteosarcoma cells were analyzed. From each image (921.3 × 921.3 µm), the numbers of live (green) and dead (red) cells were calculated and differences between sample groups were estimated. The proportional area (per one confocal picture) occupied by live cells was also obtained. The total number of analyzed pictures (after three experimental runs) was 91 for austenite, 75 for SCO and 60 for the BO sample type.

The numbers of paxillin-containing focal adhesions were measured. The measured area of interest was 146.2 × 146.2 µm, corresponding to one confocal microscope image. The images were segmented on green color intensity, hue and saturation, representing the emission color of the secondary antibody. These interactively defined focal adhesions were automatically counted from the region of interest. The total number of analyzed pictures (after three experimental runs) was 134 for austenite, 111 for SCO and 93 for the BO sample type. Nineteen pictures were analyzed from the additional NiTi disk oxidized at 600°C.

87

Study III: The confocal images (146.2 × 146.2 µm) of fibronectin fibers were analyzed. The images were segmented by green color intensity, hue and saturation, representing the emission color of fluorescein isothiocyanate, and the fibrils were automatically counted from the region of interest. The number, length and area of the fibronectin fibrils were defined from 34 confocal images from the austenite and 22 from the martensite sample.

Study IV: The standard Masson’s trichrome Goldner stained thin sections of the intramedullary nailed rat femurs were digitalized with a Sony 930 DXC camera (Sony, Tokyo, Japan) coupled to the MCID M4 image analysis system by using a Nikon Optiphot II microscope, with a 2.5x plan objective (Nikon, Tokyo, Japan). From each thin section, the perimeter of bone and fibrous tissue adjacent to the implant perimeter was measured and the following calculations were carried out:

Total bone–implant contact (%) = (total length of bone contact (µm) / total length of implant perimeter (µm)) × 100,

Total fibrous tissue–implant contact (%) = (total length of fibrous tissue contact (µm) / total length of implant perimeter (µm)) × 100.

The total amounts of analyzed thin sections were: 33 for group I, 34 for I-sol-gel, 44 for II, 54 for II-sol-gel, 48 for STR and 39 for STR-sol-gel.

4.9 Hybrid NiTi

An additional series of tests for the analysis of the effect of austenite and martensite phases on osteoblast behavior was performed. In these experiments hybrid NiTi disks were used. The 6 mm discs had a 3 mm martensitic core surrounded by a tightly adjusted austenitic shell. Figure 14 represents one of these samples.

88

Fig. 14. An image of the hybrid NiTi sample (diameter 6 mm). The inner disk is composed of martensitic NiTi, which is compressed by the outer shell of austenitic NiTi.

These samples were unique, since they allowed us to examine the impact of NiTi phases on osteoblast behavior directly with samples having both the austenite and martensite phase in the same sample. The contact area of the two phases was also under study, since in this area, compression and tension forces create an interesting area of stresses. In principle, the inner plate of martensite was under compression caused by “shrinking” of the outer austenite ring. On the other hand, the austenite shell was subjected to tension from the expansion of the inner martensite plate.

On these samples, approximately 5×103 osteoblast-like ROS-17/2.8 cells were cultured for 48 hours under conditions similar to those already described in previous chapters. After cultivation, the cells were PFA-fixed and the focal adhesions were stained with paxillin antibody followed by fluorescent secondary antibody. In addition, actin filaments were stained with TRITC-conjugated phalloidin in order to observe the effect of material stresses on the actin cytoskeleton. The samples were viewed with a confocal microscope and the number of cell adhesions estimated from the confocal images with an M4 image analysis system. These methods were also described in the previous chapters.

89

4.10 Statistical analysis (I–IV)

An analysis of variance (ANOVA) and the Student′s t-test were utilized to assess the level of significance of the differences between the experimental groups (I–III). The statistical analyses were performed with commercial software (Origin 6.0, Microcal Software, Inc., USA). Most of the data (fibronectin conformation, cell proliferation and cell cycle studies) in study III were analyzed with One-way ANOVA and the Kruskal-Wallis test to distinguish the differences between the groups. The analyses were followed by a Tukey-Kramer test for multiple comparisons. These statistical computations were performed with Graphpad Instat® 2.00 software and some were analyzed using Wincycle software.

The difference in femur length between the operated and control (contralateral) femurs in the original study IV were tested versus zero by a one-sample t-test. Comparisons between experimental groups were evaluated with independent samples t-test. In addition, the distribution of the data of bone formation and fibrous tissue adjacent to the implant surface within each test group were tested with a 1-sample Kolmogorov-Smirnov test. Differences between experimental groups were tested with the Mann-Whitney nonparametric test. Statistical analyses were performed with commercial software (SPSS version 14.0, Chicago, IL, USA), and probabilities of p < 0.05 were considered significant in all the analyses.

90

91

5 Results

5.1 Surface roughness of the NiTi samples (I–III)

The surface roughness of the NiTi samples used in the original studies (I–III) was measured either with AFM (I, II) or by a surface profiler (III). The results of surface roughness measurements are summarized in table 6.

Table 6. Average surface roughness of the NiTi samples [nm, mean (± SD)].

Original article Austenite Martensite SCO BO

I 25.4 (4.04) 26.3 (4.50)

II 849 262 310

III 124 (23) 188 (33)

SCO = straw-colored oxide (austenite NiTi oxidized at 350°C), BO = blue colored oxide (austenite NiTi

oxidized at 450°C). I) The number of analyzed images was 10 for both austenite and martensite. II) The

number of analyzed images was 40 for austenite, 80 for SCO and 60 for BO. III) The number of analyzed

images was 42 for austenite and 39 for martensite.

The values in study I were not statistically different, hence the surface of austenite and martensite samples were approximately equally rough. The results of study II showed the following statistical differences: the austenite samples were rougher than the oxidized surfaces (p < 0.001) and, in addition, the BO samples were rougher than the SCO (p < 0.01) samples. Despite these differences, the actual effect on cell behavior was probably insignificant, since it is not generally thought that nm roughness levels will affect cell functionality to the same extent as µm variations (Clark et al. 1991). Other results also support this hypothesis: only subtle differences in focal adhesion length and no differences in focal adhesion movement were observed with titanium samples having roughness differences ranging from 190 nm to over 6000 nm (Diener et al. 2005). In study II, roughness varied at most by only approximately 600 nm.

The difference between the RMS roughness values of austenitic and martensitic samples (III) was statistically significant (p < 0.01). Since the RMS values are limited to only one dimension (the direction perpendicular to the surface under study), a bidimensional analysis that takes into account the frequency of occurrence of every single peak/valley, thus calculating the additional active surface generated by the roughness was used. The corrected

92

results demonstrated that the martensite surface was 1.6 times rougher than the austenite surface.

5.2 Contact angles (I, III)

The results of contact angle measurements performed in studies I and III are summarized in table 7.

Table 7. Average static contact angles [°, mean (± SD)] for austenite and martensite NiTi and for glass coverslips as a reference material. The differences between the contact angles of austenite and martensite samples are also presented.

I III Liquid

Aus Mar Diff. P Glass Aus Mar Diff. P

Water 65.4 (2.8) 67.3 (2.4) -1.9 ns 57.8 (4.3) 79.1 (1.9) 86.6 (1.8) -7.5 ns

Dimethyl sulfoxide 46.2 (0.9) 43.9 (1.5) 2.3 < 0.01 64.9 (6.1)

Ethylene glycol 50.5 (2.0) 43.5 (1.4) 7.0 < 0.001 68.4 (5.0) 54.9 (2.2) 66.6 (1.1) -11.7 ns

Glycerol 68.6 (3.8) 58.5 (3.0) 10.1 < 0.001 92.5 (2.0)

Diiodine methane 47.2 (2.5) 44.4 (1.9) 2.8 < 0.05 68.6 (1.0) 43.4 (1.5) 55.8 (2.4) -12.4 ns

Original articles are referred to by the Roman numerals I and III. Aus = austenite NiTi, Mar = martensite

NiTi. The values presented in the “Diff” columns are calculated as the difference between the mean value

of austenite and martensite samples. I) The number of contact angles analyzed was 176 for every sample

type. III) A total of 54 angles were analyzed for both austenite and martensite sample types.

In addition to the statistical differences presented in the above table, the contact angles measured on glass cover slips were significantly different from those of austenite or martensite with all five liquids, including water (p-values not shown).

5.3 Surface free energy (I, III)

The results of surface free energy calculation in the original studies I (harmonic mean approach) and III (acid-base approach) are summarized in table 8. It is important to note that these results are strongly dependent on the particular method used to calculate the surface free energy.

93

Table 8. Surface free energy and its components [mN/m, mean (± SD)] for austenite and martensite NiTi and for glass coverslips as a reference material.

I III Energy

components Aus Mar Glass Aus Mar

γD 15.7 19.42 6.75

γP 22.52 20.00 30.67

γLW 37.9 (0.8) 30.9 (1.4)

γ+ 0.01 (0.3) 4×10-4 (5×10-3)

γ- 7.6 (0.7) 6.1 (1.5)

γ 38.21 39.41 37.42 38.5 (1.2) 31.1 (1.5)

Original articles are referred to by the Roman numerals I and III. Aus = austenite NiTi, Mar = martensite

NiTi. γD is the dispersive component of the surface energy. γP refers to the polar component. γLW is the

Lifshitz-van der Waal component of the surface energy. The Lewis acid-base component is divided into

acidic electron-acceptor (γ+) and basic electron-donating (γ-) parameters. γ describes the total surface free

energy.

No significant differences in total surface free energies or its components were found between austenite and martensite (I). The total surface free energy of glass coverslips was similar to that of the NiTi samples. In contrast, the polar component was higher and the dispersive component lower for the glass control samples compared to austenite or martensite NiTi.

Since the static contact angles measured in study III were smaller on the austenitic NiTi than on the martensitic phase, the calculated values of the materials’ surface energy were larger on austenite than on martensite. Even though the measured differences were not statistically significant, the total surface energy, mostly due the difference in the LW component, was somewhat different between austenite and martensite. The results show that the acid components were null, since the calculated errors are greater than the component itself. Thus, the surfaces of the austenite and martensite samples were only basic, as is the case for the thin metal oxide layer present at the NiTi alloy surface.

5.4 Results of the physical measurements of the oxidized NiTi (II)

The following chapters present the results from original article II, concerning the analyses of the oxidized NiTi sample surface.

94

5.4.1 X-ray diffraction

The X-ray patterns obtained in study II showed that the non-oxidized NiTi was characterized in the austenite state with small quantities of Ti2Ni. The basic phase in the bulk of NiTi after its oxidation at 350oC and at 450oC was also austenite, but quantities of martensite phase had emerged in the alloy structure, with higher amounts in the samples oxidized at 450°C. In addition, after oxidation in both temperatures, the metal alloy next to the oxide layer contained higher amounts of martensite than the bulk of the alloy.

5.4.2 Resistivity

In the original study II, the electrical properties of the oxidized NiTi were measured. The results are summarized in table 9.

Table 9. Changes in electrical resistance (∆R, ohm) of NiTi samples as a result of their oxidation at 350°C or 450°C for 1 hour. Resistance of the basic NiTi alloy (ρNiTi, ohm m) and the oxide layers (ρOX, ohm m) are also presented.

Oxidation

temperature (°C)

∆R ρNiTi initial ρNiTi after oxidation change in ρNiTi ρOX

350 0.107 0.769 x10-6 0.804 x10-6 -0.035 x10-6 2.61 x10-6

450 0.019 0.791 x10-6 0.770 x10-6 0.021 x10-6 8.48 x10-6

5.4.3 Oxide thickness

The plots of mass gain per surface area unit (Δm) as a function of time for NiTi samples oxidized at 350, 450 and 600°C demonstrated that NiTi oxidation followed logarithmic time dependence at 350°C and parabolic dependence at 450 and 600°C. The calculated oxide thicknesses were: 15.6 nm for SCO, 51.0 nm for BO and 725.0 nm for the additional NiTi sample oxidized at 600°C.

The values of the oxide thicknesses estimated were consistent with the oxide thickness obtained by direct measurements in a study of similar oxidation regimes (Firstov et al. 2002). Since the oxidized surfaces include the initial natural oxide layer of non-oxidized NiTi, the final values should be added to with the initial oxide thickness (δ0). In the absence of available methods for its determination in the study I, a value of 5 nm was ascribed as δ0, corresponding to the spontaneous oxide crystal size. A similar value (about 3 nm) was estimated previously

95

(Kofstad 1966) and has also been reported for mechanically polished NiTi (Armitage & Grant 2003).

5.5 Success of the intramedullary NiTi nailing (IV)

First of all, the adhesion between the NiTi wire and sol-gel derived coating was considered sufficient (Fig. 15). Since the coating did not crack due to the bending but rather adopted the elastic property of the NiTi wire, the coated NiTi intramedullary nails were considered to be suitable for implantation.

Fig. 15. SEM image of the bend NiTi reference wire (A) and sol-gel coated NiTi wire (B). Scale bar 200 µm.

96

Forty male Sprague-Dawley rat femurs were implanted with the functional NiTi intramedullary nails. The implantation was performed in a proximal to distal direction into the medullary canal of the right femur. Twenty-four weeks postoperatively, the animals were weighed (this was done also preoperatively) and the lengths of the dissected operated and contralateral femurs were measured. The results are summarized on table 10.

Table 10. Summary of the femur lengths (mm) and animal weights (g) in the six experimental groups. Data are represented as mean (± SD).

I I-sol-gel II II-sol-gel STR STR-sol-gel Average

n 5 (-2) 5 (-2) 6 (-1) 7 6 6

Operated

femur length

4.22 (0.15) 4.26 (0.09) 4.15 (0.14) 4.19 (0.09) 4.22 (0.12) 4.28 (0.15) 4.22 (0.12)

Control femur

length

4.22 (0.11) 4.22 (0.04) 4.18 (0.12) 4.16 (0.08) 4.18 (0.08) 4.27 (0.08) 4.20 (0.09)

Initial weight 376 (22) 376 (20) 392 (28) 375 (22) 392 (33) 405 (24) 385 (26)

Weight after 24

weeks

485 (25) 522 (36) 501 (29) 489 (26) 528 (35) 529 (29) 509 (33)

(-)l indicates animals sacrificed at 2 weeks postoperatively.

The length of the operated femurs did not differ significantly compared to the control contralateral femurs in any of the six test groups. In addition, there were no significant differences in femur lengths between the six experimental groups.

Even though there were differences in average animal weights between the test groups, the average weight gain of the animals was statistically equal after the 24 weeks, approximately 123 ± 40.6 g in all the six experimental groups.

Two weeks after the implantation, five of the 40 animals had to be sacrificed, since they did not use their operated legs normally due to an infection at the implantation site. These rats were stored at -20°C until the time point of 24 weeks was reached. Dissection of the five rats showed an infected area with a large amount of pus at the implantation site. The contralateral legs were normal as were all of the limbs in the animals maintained for the full 24 weeks. The infection of the operated femurs was most probably due to an error in the nailing procedure. Bostman et al. (Bostman et al. 1989) reported septic arthritis of the hip joint after intramedullary nailing, in which a deep wound infection had occurred after the operation in which the hip capsule has been accidentally opened. In the present case, the hip joint area of the five rats showed a highly resorbed joint surface, indicating septic arthritis (visual observation, data not shown).

97

In addition, on one operated rat (STR-sol-gel), the implant had penetrated approximately 2 mm out of the distal femur. This was clearly due to an operational error, in which the reaming was accidentally done through the cortex of the bone.

5.6 Biological assays

The following chapters represent the main results obtained from the biological assays used in the original articles II and III.

5.6.1 Cytotoxicity (II)

Cytotoxicity of the oxidized NiTi surfaces was tested with a commercial test kit that divides the cells grown on the sample surface into live (green) and dead (red) cells. The results are presented in figure 16.

98

Fig. 16. Results of the cytotoxicity test. (A) The percentages of live cells in the ROS-17/2.8 cell populations on the three different NiTi alloy samples. Statistical analyses showed a significant difference in the percentages between SCO and austenite NiTi disks (p < 0.001). (B) The proportional area occupied by live cells. The mean values differed significantly between austenitic and BO (oxidized at 450°C) sample groups (p < 0.05). A similar difference is also observed between the oxidized samples. These results are presented per one confocal microscopy field. (C) The percentages of dead cells in the ROS-17/2.8 cell populations on the three sample groups. A significant difference in the number of dead cells between austenitic and SCO NiTi was observed (p < 0.001). The results are combined from three experimental runs. The numbers of confocal images analyzed were: austenite = 91, SCO = 75 and BO = 60. *** p < 0.001; * p < 0.05

99

The results of the cytotoxicity assay demonstrated that most of the cells grown on the NiTi surfaces were alive and only a very small portion of the whole cell population was dead; approximately 4.7% on austenite, 1.9% on SCO and 4.1% on BO. However, the interpretation of these results requires caution, since there is also a possibility that the dead cells were flushed away from the sample surface during the staining procedure. This might have had an effect on the results.

5.6.2 Fibronectin rearrangement onto NiTi surface (III)

Fluorescein-labeled fibronectin molecules were allowed to attach onto austenite and martensite NiTi samples and osteoblast-like cells were grown on the coated surfaces for 2 hours. The FN-FITC surface coating was clearly reorganized into a fibronectin-fibril network on the NiTi samples after the 2-hour culture (Fig. 17).

Fig. 17. Confocal images of FN-FITC coated A) austenite and B) martensite NiTi after 2-hour cell culture. Green fibronectin fibrils are clearly shown around cell perimeters and between adjacent osteoblasts (the cells are not labeled for visualization). Magnification 63x, frame size 146.2 × 146.2 µm.

Table 11 summarizes the results of fibronectin remodeling on the NiTi surfaces. As the table shows, the average number, length and total area of fibronectin fibers were larger on the austenite sample type compared to martensite. This was also visually observed from the confocal images of FN-coated NiTi, as represented in

100

figure 17. The amount of fibronectin adsorbed onto biomaterial surfaces vary in the literature (Keselowsky et al. 2005, Jimbo et al. 2007), but the ~3 µg/cm2 of FITC-labeled FN used in this study was clearly enough to create an uniform fibronectin coat over the NiTi sample disks.

Table 11. Results of fibronectin remodeling on austenite and martensite NiTi. The results are represented as mean (± SD) per one confocal image field.

Fibronectin Austenite Martensite P

Number 191 (91) 115 (78) < 0.01

Length (µm) 1.48 (0.24) 1.28 (0.18) < 0.01

Covered area (µm2) 283 (144) 147 (104) < 0.001

Total numbers of analyzed confocal images were 36 for austenite and 25 for martensite.

5.6.3 Analysis of fibronectin conformation on NiTi surfaces (ELISA) (III)

These analyses were performed in order to study the adsorption of fibronectin and possible conformational differences in the fibronectin structure between austenite and martensite NiTi. The results are summarized in table 12.

Table 12. Fibronectin adsorption and conformation onto austenite and martensite NiTi assessed by ELISA. The results are presented as mean (± SD) absorbances/cm2.

Antibodies Austenite Martensite TCPS CEL Polyclonal Anti-FN 0.0658 (0.0019) 0.0687 (0.0010) 0.0576 (0.0026) 0.0195 (0.0004)

MAb anti-RGD 0.0368 (0.0031) 0.0359 (0.0021) 0.0332 (0.0014) 0.0010 (0.0001)

MAb anti-N-terminal 0.0243 (0.0019) 0.0236 (0.0015) 0.0250 (0.0016) 0.0011 (0.0001)

MAb anti-C-terminal 0.0802 (0.0074) 0.0837 (0.0097) 0.0424 (0.0006) 0.0028 (0.0002)

MAb anti-gelatin 0.0375 (0.0022) 0.0328 (0.0029) 0.0229 (0.0008) 0.0001 (0.0002)

MAb = monoclonal antibody. Controls samples are tissue culture polystyrene (TCPS) and anti-adhesive

carboxymethylcellulose (CEL). Absorbance (490 nm) was read in triplicate and the test for each antibody

was repeated 3–4 times. Each group had six parallel samples.

The results using an anti-FN polyclonal antibody showed that the total FN adsorption was similar onto austenite and martensite NiTi. Control materials gave as expected results: FN adsorbed well on TCPS when referred to the FN concentration used for the coating, in addition, the adsorption was very low onto the anti-adhesive CEL (carboxymethylcellulose).

101

The results of FN conformation, assessed by monoclonal antibodies specific for distinct epitopes, indicated similar conformations of fibronectin on both NiTi phases. There were no statistically significant differences in the binding of RGD, N-terminal, C-terminal, or gelatin sites by the respective MAbs onto austenite and martensite NiTi.

5.6.4 Cell proliferation on FN-coated NiTi (III)

Cell proliferation on austenite and martensite was studied with LDH and MTT assays. The assays were done from the same samples: LDH from the supernatant and MTT from the MC3T3-E1 cells. The results of both LDH and MTT assays for 2 and 48 hours cell culture on austenite and martensite NiTi, and TCPS control, are summarized on figure 18.

Fig. 18. MC3T3-E1 cell proliferation on uncoated and FN-coated tissue culture polystyrene (TCPS), austenite and martensite NiTi. A) LDH at 2 h, B) LDH at 48 h, C) MTT at 2 h and D) MTT at 48 h. The data are shown as mean, the error bars indicate the standard deviation. White bars represent uncoated and grey bars FN-coated samples. * p < 0.05, ** p < 0.01, *** p < 0.001. In addition to the statistical differences shown in the figure, in both assays and time points used, the TCPS control substrate differed significantly (p < 0.001) when compared to the corresponding NiTi samples.

102

The results of the LDH cell damage study showed after 2-hour culture that the LDH activity was higher, i.e. more cell membrane damage present, on NiTi samples than on the TCPS controls (p < 0.001). The fibronectin coating seemed to have no effect on osteoblast survival on the control TCPS surface. On the contrary, on NiTi samples, both austenite and martensite, the presence of fibronectin significantly improved the surface, resulting in significantly less LDH activity on the FN-coated surfaces (p < 0.001). In addition, on the martensite surface, cell damage was higher than on austenite. This difference was significant both without (p < 0.01) and with (p < 0.05) fibronectin (Fig. 18A).

After two days, the fibronectin coating still did not have an effect on the LDH activity on the TCPS surface. However, as after two hours, LDH activity after two days was very significantly lower on TCPS than on the NiTi surfaces. On austenite, fibronectin significantly reduced cell damage (p < 0.001), thus improving the biocompatibility of the parent phase of NiTi. In contrast, on martensite, fibronectin did not have a statistically significant effect on the survival of the cells. In addition, the protein coating significantly (p < 0.001) reduced the LDH activity on austenite when compared to the coated martensite surface (Fig. 18B).

In conclusion, surface modification with FN reduced cell damage, as measured by an LDH assay on austenite and martensite NiTi; by the two-hour time point, the biocompatibility of both austenite and martensite was significantly improved by the presence of FN. After two days, fibronectin kept the cell damage lower on austenite, but on martensite, the difference between the coated and uncoated surfaces disappeared. In addition, after both 2 and 48 hours, with fibronectin on the alloy surface, the cells preferred austenite to martensite.

The results of the MTT analysis showed that two hours after seeding, mitochondrial activity was significantly lower (p < 0.001) on the NiTi samples (both uncoated and FN-coated) compared with the TCPS control (Fig 18C). Forty-eight hours after seeding, mitochondrial activity increased in each experimental situation, but the difference between TCPS and NiTi remained (Fig 18D).

The fibronectin coating did not have an effect on cell proliferation measured by the MTT assay on both austenite and martensite NiTi. After two hours (Fig 18C), fibronectin increased cell proliferation only on TCPS (p < 0.05), but after 48 hours this difference was reversed (Fig 18D), resulting in a decrease of osteoblast survival on the fibronectin coated TCPS surface (p < 0.001). The results observed on TCPS might be due to the fact that TCPS is optimized for cell

103

adhesion, thus it is reasonable that cells do prefer this surface type over the NiTi ones. The reason why fibronectin lowers the biocompatibility of TCPS might also be due to the surface optimization: the pure fibronectin surface is not as ideal for cell attachment as is the untreated TCPS surface.

Together these analyses indicate that although cell growth is similar on austenite and martensite NiTi, cell damage is more prevalent on the martensite phase, indicating better biocompatibility of the austenitic NiTi. A fibronectin coating further improves the biocompatibility of austenite.

5.6.5 Cell cycle (III)

The analysis of the cell cycle stage of MC3T3-E1 cells grown on FN-coated or uncoated TCPS, austenite or martensite NiTi for 48 hours was performed in the original study III. Comparisons of the cell cycle phases of MC3T3-E1 cells after a 48 h culture with or without the fibronectin-coating are shown in table 13.

Table 13. Comparisons of MC3T3-E1 cells in different phases of the cell cycle after 48-hour culture with or without a fibronectin coating on austenite and martensite NiTi and on tissue culture polystyrene (TCPS). The data are shown as mean % (± SD).

Aus Mar TCPS Cell cycle w/o FN FN P

w/o FN FN P w/o FN FN P

G1 69.0 (4.5) 79.0 (1.9) < 0.01 71.9 (3.3) 79.4 (1.3) < 0.01 69.7 (2.8) 72.4 (3.7) ns

S 15.8 (2.9) 11.9 (1.5) < 0.01 14.0 (2.6) 11.5 (1.0) < 0.01 19.5 (1.8) 19.8 (3.0) ns

G2/M 15.2 (1.8) 9.2 (0.8) < 0.01 14.1 (1.3) 9.1 (1.1) < 0.01 10.7 (1.4) 7.8 (1.5) < 0.01

w/o FN = without fibronectin coating, FN = with fibronectin coating. Two independent tests were analyzed.

Each group had six parallel samples.

The cell cycle analysis showed similar results on both austenite and martensite NiTi. With the fibronectin coating on NiTi, the number of cells in the G1 phase increased significantly and the number of cells in the S and G2/M phases decreased when compared with the uncoated sample surface. The elevated number of G1 cells on NiTi might indicate an increase in the number of differentiating cells on both austenite and martensite samples. On TCPS, FN had no effect on the percentage of cells in the G1 or S phases; instead, a decrease in the population of cells in the G2/M phase was observed.

104

5.7 Visualization of actin rings (I)

The organization of actin in adherent osteoclasts depends on the mineralization of the matrix. In resorbing osteoclasts filamentous actin is organized as a constant large band, a part of a formation called the sealing zone, which is easily detected by microscopic methods. Podosome structures of actin are present in non-resorbing osteoclasts, corresponding to numerous F-actin columns arranged at the cell periphery. Osteoclasts may sense the surrounding physical properties and react accordingly by changing the actin organization: actin is organized as podosome belts in spread osteoclasts adhering onto glass, as the sealing zone is seen in osteoclasts adherent on mineralized matrix. (Saltel et al. 2004.) Hence, actin organization can be used as an indicator of osteoclast tolerance to the underlying substratum. The results of using actin ring formation as a marker of osteoclast tolerance towards the two NiTi phases are represented in figure 19.

105

Fig. 19. Confocal microscopy images of F-actin rings using TRITC-conjugated phalloidin in rat bone cells grown 48 h on the austenitic NiTi disk (A, B), martensitic NiTi disk (C, D), bovine bone slice (E), and glass coverslip (F) as a control substratum. Magnification 20x, with the exception of B and D 63x. Scale bar 50 μm.

As figure 19 shows, the overall cell morphology and level of cell confluence were different in the NiTi samples compared to the control materials. On the bone slices, the cultivated cell layer appeared notably more confluent and more actin rings were present compared to the other sample types (data not shown). Some

106

amount of deterioration in cell morphology and confluence could be seen on the austenitic NiTi, but clearly, the martensitic NiTi provoked the most abnormal cellular responses, seen as distinctively reduced confluence and the appearance of round-shaped cells.

The numbers of actin rings was calculated from 15 consecutive microscopy views, repeated 15 times on each sample disks. The average amount (± SD) of actin rings in these 0.38 cm2 areas, after two separate tests, were 279 ± 31 for austenite and 91 ± 24 for the martensite sample type. The difference was statistically very significant (p < 0.001).

5.8 Focal adhesions on osteoblasts attached to an oxidized NiTi surface (II)

The osteoblast-like cells cultured for 48 hours on unoxidized and oxidized NiTi samples were immunofluorescently stained and the number of focal adhesions was calculated. Figure 20 and table 14 summarize the results obtained.

Figure 20 represents images of ROS-17/2.8 osteoblastic cells at the surface of the different NiTi alloy samples. It is evident, that focal adhesion formation and average size is reduced when cells are grown on NiTi surfaces oxidized at 450°C and 600°C compared to the other sample types. Despite the fact that the BO samples and the one oxidized at 600°C contained the thickest surface oxide layers (51.0 nm for BO and 725.0 nm for 600°C oxidation), the cells seem to attach best to the SCO surface with a thin oxide layer (15.6 nm) or to the unoxidized austenite surface.

107

Fig. 20. Immunofluorescence staining of focal adhesion with paxillin antibody on ROS-17/2.8 cells. (A) Austenitic NiTi, (B) NiTi oxidized at 350oC (SCO), (C) at 450oC (BO) and (D) at 600°C. Numerous white patches represent focal adhesion plaques. The average thickness of the surface oxide layer: (A) 5 nm, (B) 20.6 nm, (C) 56.0 nm and (D) 730.0 nm. Magnification 63x, frame size 146.2 × 146.2 µm.

Table 14. The average number and total area (µm2) of focal adhesions per ROS-17/2.8 cell. The data are shown as mean (± SD).

Cell adhesion Austenite SCO BO

Number of FA / cell 40 (19) 47 (20) 18 (13)

Total area of FA / cell 23 (14) 21 (11) 6 (5)

FA = focal adhesion. The numbers of confocal images were analyzed: austenite = 134, SCO = 111 and

BO = 93. Each sample type had approximately 1100 attached cells.

108

The average number of focal adhesions (table 14) was very significantly (p < 0.001) higher in the austenitic and SCO samples compared to the BO ones. In addition, more focal adhesions were present on the cells at the SCO surface compared to the austenite (p < 0.01). The average area (µm2) occupied by focal adhesions, i.e. the total area of cell membrane attach to NiTi surface, was similarly higher in austenitic and SCO samples compared to the NiTi treated in 450°C (p < 0.001).

The additional 40 mm-in diameter sample was also analyzed (Fig. 19D). The results showed the same decreasing trend in focal adhesion number with increasing oxidation temperature as seen in the other NiTi samples. The average amount of focal adhesion was approximately 16 plaques per cell and the area covered by them was approximately 11 µm2.

5.9 Active osteoclasts on NiTi surface (I)

The results of TRACP-staining of primary rat osteoclasts cultured on austenite and martensite NiTi are presented in table 15.

Table 15. The effect of NiTi alloy phases on the numbers of TRACP-positive cells and on the percentages of active osteoclasts. The data are shown as mean (± SD).

TRACP-staining Austenite Martensite P

Mononuclear TRACP+ cells 339 (54) 204 (27) < 0.001

Multinuclear TRACP+ cells 94 (25) 59 (19) < 0.001

Mononuclear /Multinuclear TRACP+ cells 3.9 (1.5) 3.7 (1.1)

% Osteoclasts active 66 (13) 35 (11) < 0.001

The percentages of active osteoclasts were determined by counting the total number of osteoclasts with

actin rings divided by the number of TRACP-positive cells after 48-hour cultivation on the austenitic and

martensitic NiTi disks. n = 30 representing the number of analyzed areas of 0.38 cm2 on NiTi disks

combined from two experiments.

In this case, “an active osteoclast” is used as a term for those osteoclasts that developed an actin ring and were TRACP-positive. The cells’ ability to etch or modify the underlying metal or construct the full resorbing machinery remains unknown.

109

5.10 Osteoblast attachment to hybrid austenite–martensite NiTi substrates

The results of ROS-17/2.8 cell culture on hybrid NiTi samples clearly showed that these osteoblast-like cells were more spread in the austenite areas of the samples. In addition, the cells on the austenite part of the sample had more focal adhesion, which were also more elongated in structure. From visual observations of the confocal images, a conclusion can be draw that although the actin cytoskeleton was normally organized on both phases, on the austenite part of the samples the cells had a thicker, stronger-looking cytoskeleton with large focal adhesions at the ends of the filaments. These can be seen in figure 21. The contact area of the two phases was observed to have strong cell attachment inhibitory effects.

Fig. 21. ROS-17/2.8 cells on a hybrid NiTi sample. Green represents focal adhesions and red actin filaments, yellow is the area where these two colocalized. (A) austenitic shell, (B) the contact area of the two phases, and (C) martensitic core. Magnification 63x, frame size 146.2 × 146.2 µm.

No statistical analysis was performed on the data, since the number of confocal images analyzed was low due to the preliminary nature of the experiment. The amounts of focal contacts were counted from a few confocal images: 26 on the outer austenite ring, 4 on the contact area and 8 on the inner martensite area of the sample. The preliminary results were (mean ± SD): 550 ± 227 focal adhesions on austenite, 327 ± 226 on the contact area and 490 ± 208 on martensite (per one confocal image of 146.2 × 146.2 µm). The cells on the contact area and on the martensite inner disk were round and diffusively stained, and thus, paxillin-stained focal adhesions were "countable" only in parts of the confocal images.

110

5.11 Intramedullary NiTi nails – results of Faxitron radiography and histological analysis (IV)

Figures 22–24 and table 16 presents an overview of the results of the antegrade intramedullary nailing from the original study IV.

The IM nails were randomly placed into the medullary canal; as a result, the curved nails were bowed either in a medial (5) or a distal (18) direction. The radiographs often showed that 1–2 mm of the nail tip extented from the bone at the proximal end: 8 of the 23 curved nails and 6 of the 12 straight nails were partly outside the femur. This had no effect on the health of the rats due to adverse tissue reactions that might have been possible because of the rubbing of the implant against surrounding soft tissue. The curved IM nails were typically in contact with the bone tissue at the distal tip of the implant, where a cup-like formation of bone was very common: 80% of groups I and I-sol-gel, 50% of group II and 86% of group II-sol-gel. This effect was not observed with the straight nails.

The results showed that the femurs with curved or straight nails grew equally when compared to the unoperated contralateral femurs. In other words, the presence of a NiTi nail did not cause any bowing of the bone. This might have been due to the fact that the nailing procedure was not completely successful, since the intramedullary NiTi nails were not completely inserted into the medullary canal. Because of this, the bending nails did not direct the bending load towards the diaphysis of the femur, but more on the proximal part of the bone. Even though optimization of the nailing procedure is needed, the NiTi nails did not cause fractures in the bone. In some cases, the bending nails clearly followed the structure of the femur, as seen in the figure 23.

111

Fig. 22. Overview of the radiographic results of antegrade intramedullary NiTi nailing. The upper three images (A–C) are from the uncoated sample groups I, II and STR, respectively. The lower images (D–F) represent the TiO2-SiO2 coated samples of groups’ I-sol-gel, II-sol-gel and STR-sol-gel. At the distal tip of the implant, a cup-like bone structure can be seen at the bone−implant contact area (white arrow).

112

Fig. 23. Left hand side: An AP radiograph of sol-gel surface treated intramedullary NiTi nail (group II-sol-gel). The austenite nail clearly follows the intramedullary bone structure without any bending of the bone. Right hand side: AP radiograph of the contralateral control femur.

As seen in the previous figure (Fig. 23), the IM NiTi nail adapted to the inner surface of the medullary canal. The figure demonstrates that the curved NiTi nails did actually push the medullary surface, since there was not enough space to allow the full shape change when transforming from martensite to austenite.

113

Fig. 24. Histological sections of the implanted femoral bones from the six test groups. (A) I, (B) II, (C) STR, (D) I-sol-gel, (E) II-sol-gel and (F) STR-sol-gel. An AP radiograph of a femur with a type II-sol-gel nail is presented (G): the upper mark shows the first cut made at the minor trochanter, the latter is the bone site where the histological sections shown in this figure are taken (17.5 mm from the proximal end of the bone). Green represents bone and orange near the implant surface (black sphere; 1.4 mm-in diameter) represents fibrous tissue.

Femur thicknesses were measured from the radiographs using the MCID M4 image analysis system in triplicate from the area of the minor trochanter (Fig. 24G, the upper line drawn in the figure). The results of every test group showed that the diameter of the operated femur (5.96 ± 0.37 mm) was significantly larger (0.01 < p < 0.05) than the unoperated contralateral counterpart (5.14 ± 0.29 mm). The same effect was observed with both curved and straight nails. Since this effect was evident in every group and in addition, there were no differences between the six test groups, the thickening of the bone was probably the consequence of natural bone remodeling after the disturbance caused by the implantation procedure.

Table 16. The percentages of bone and fibrous tissue attached to the intramedullary NiTi nails. The data are shown as mean % (± SD).

Tissue–implant contact I I-sol-gel II II-sol-gel STR STR-sol-gel

Bone–implant contact 79 (19) 75 (23) 77 (18) 89 (10) 47 (34) 57 (30)

Fibrous tissue–implant contact 20 (20) 28 (28) 24 (18) 11 (10) 54 (35) 41 (30)

The numbers of analyzed samples were: I = 33, I-sol-gel = 34, II = 44, II-sol-gel = 54, STR = 48 and STR-

sol-gel = 39.

114

As seen in the above table 16, group II-sol-gel had the most bone attached around the implant and also had the smallest percentage of fibrous tissue present. These properties were significantly different when compared to any other group tested (0.001 < p < 0.05). On the contrary, the straight non-coated nails (group STR) had the smallest amount of neocortex and highest amount of fibrous tissue compared to any other group (0.001 < p < 0.01) besides group STR-sol-gel, which had statistically the same amount of these tissues present in proximity to the NiTi implant.

According to the histological results, the higher bending force together with sol-gel surface treatment (group II-sol-gel) was the most effective in the sense of bone attachment onto the implant surface. Without the coating (group II), there was much less bone–implant attachment present (p < 0.001). The strength of the bending force was another significant factor in the bone–implant attachment. The implants with smaller bending force (group I-sol-gel) had less bone attached (p < 0.01) and more fibrous tissue present (p < 0.001) at the peri-implant area than the group with the higher force (II-sol-gel). In addition, there were no differences between the uncoated and coated nails with the smaller curvature radius (groups I and I-sol-gel). Also, no differences were found when comparing the two types of curved nails without the sol-gel surface treatment (i.e. groups I and II).

Straight nails were inferior to the curved nails, demonstrating the positive effect of bending on the bone–implant attachment. Although not statistically significant, sol-gel surface treatment also seemed to improve the biocompatibility of the straight nails. As an overall conclusion, sol-gel treatment seems to enhance the biocompatibility of the NiTi IM nails. The surface treatment might even play a more important role in bone bonding to the NiTi implant surface than the constant bending force applied to the surrounding bone, although the bending itself is also a contributing factor of bone–implant attachment.

115

6 Discussion

The following discussion is divided into four sections according to the titles of the original articles (I–IV).

6.1 The phase state of NiTi implant material affects osteoclastic attachment (I)

Medical or dental implants made from NiTi alloy can have both austenite and martensite phase present, depending on the design and use of the implant. Even though extensive research concerning the biocompatibility of NiTi has been performed, the two phases have not been studied separately to distinguish their independent effects on cells and tissues. Since the two phases have different crystal lattice organizations, they are bound to have different stress levels, which in turn cause different surface states for austenite and martensite. The properties of the implant surface influence the attachment and subsequent reactions of proteins and other extracellular molecules interacting with the implant surface in a biological environment. In a bony environment, adequate osseointegration is required for bone implants to perform correctly. In addition to good initial attachment, proper bone turnover must function at the bone–implant interface. Thus, osteoblasts and osteoclasts have to be tolerant of the implanted material, since their collaborative function at the vicinity/contact of the implant is required for long-term success of the implantation.

In this study, we used primary rat osteoclast cultured on separate austenite and martensite samples in order to reveal the impact of the two NiTi phases on the function of these bone resorbing cells. We found that martensite was clearly less biocompatible than austenite, when the formation of actin rings and positive staining of tartrate-resistant acid phosphatase were measured. Danilov et al. (Danilov et al. 2003) concluded that the presence of martensite in the alloy decreases parameters of biocompatibility depending on martensite quantities and residual structural stresses in the parent phase; higher stresses promote better cell viability. This might have been the reason, or at least part of the reason, for the low biocompatibility of the martensite phase. In addition to osteoclasts, osteoblastic cells cultured on the hybrid NiTi samples, although only briefly examined, exhibited different cell behavior depending on the cell’s location on the sample surface. When the cells approached the contact area of the two phases, the morphology of the cells started to deteriorate and when they were completely at

116

the martensite area of the samples, cell survival was clearly compromised. Taken together, it is evident that despite the same chemical composition, the two phases of NiTi have different impacts on bone cell behavior.

Although the damaged morphology of both osteoclasts and osteoblasts at the martensite surface is evident, we do not know the cellular mechanism(s) behind it. The reduced number of cells at the martensite samples might give us a hint of what is going on, indicating reduced cell adhesion onto this phase of NiTi. Cells attach to artificial biomaterials by forming adhesive structures with a macromolecular layer adsorbed onto the biomaterial surface. The structure and conformation of this ECM contains information that influences cell shape, cytoskeletal organization, cell motility and polarity, gene expression, proliferation and survival (Damsky & Werb 1992, Huhtala et al. 1995, Moursi et al. 1996, Moursi et al. 1997, Globus et al. 1998, Ilic et al. 1998). The cell culture media used in culturing the osteoclasts contained vitronectin and vitronectin receptor αvβ3 integrin is highly expressed in osteoclasts. Rat osteoclasts are known to adhere to ECM proteins containing RGD sequences mainly through αvβ3 integrins (Helfrich et al. 1992). If the integrins present in osteoclasts are unable to recognize their ECM ligands, the primary attachment of osteoclasts is prevented. This prevents cytoskeletal microfilament organization of the actin ring and hence obstructs osteoclast activation towards the resorbing state (Duong & Rodan 1998). This might be one of the reasons why martensitic NiTi contained fewer actin rings and TRACP-positive osteoclasts than austenitic NiTi.

The good biocompatibility of austenite NiTi, almost comparable to that of bone slices, demonstrated that there were no unwanted effects caused by the material chemistry, that is, especially the high amount of nickel in the alloy. Osteoclasts were not negatively stimulated by NiTi. If the alloy had released significant amounts of nickel into the culture media, this would probably have had an effect on osteoclasts. Osteoclasts express a high level of protein tyrosine kinase (PTK) c-src (Horne et al. 1992, Tanaka et al. 1992), and inhibition of this and other PTKs disturbs the actin organization and resorbing activity of osteoclasts (Kajiya et al. 2000). It has been show that Ni2+ binds to specific domains of PTKs and may activate its function (Ahmadibeni et al. 2007). Nickel cations can mimic the effects of calcium (Shankar et al. 1993), which function as a negative feedback mechanism through which an osteoclast monitors its own activity using resorbed Ca2+ as an extracellular signal (Zaidi et al. 1993). In one study (Moonga et al. 2002), the authors stated that Ni2+ might be also required for the rise of cytosolic Ca2+ by transmembrane Ca2+ influx in depolarized

117

osteoclasts. Since osteoclasts had well organized resorptive machinery on austenite NiTi, the negative effect of martensite must be the result of factors other than chemical properties of the alloy. The use of separate austenite and martensite samples revealed the existence of the different effects of the material's chemical and physical properties on the biocompatibility of the NiTi alloy.

Allen et al. (Allen et al. 2003) stated that the assessment of biomaterial compatibility relies heavily on the analysis of macroscopic cellular responses to material interaction. In their study, they used both conventional phenotypic and contemporary transcriptomic (DNA microarray-based) analysis techniques to examine the interaction of cells with soft condensed biomaterial surfaces. As a result, they elucidated putative links between phenotypic responses to cell–biomaterial interactions and global gene expression profile alterations. They also showed that even discrete modifications with respect to the physiochemistry of the amorphous materials studied can lead to significant impacts on the phenotype of interacting cells (Allen et al. 2003). It would be very interesting to study the impact of phase transformation and different surface modifications of NiTi on the gene expression profile since, as noted in the studies presented in this thesis, even small differences in material properties have proved to have significant impact on cellular behavior. It would be important to combine the conventional cellular phenotypic analysis with the application of DNA microarray-based gene expression profiling to provide a comprehensive and integrated overview of cell–biomaterial interaction.

6.2 The effect of oxide thickness on osteoblast attachment and survival (II)

Hundreds of different materials, methods and improvements of a biomaterial’s surface characteristics have been developed to improve the biocompatibility and functionality of implants. In metal and metal alloy implants, the surface of the implant must be resistant to corrosion and other material-consuming chemical, biological or physical processes that take place when the implant is in physiological environment. Oxidation of an implant surface in order to achieve bioinert, -active, -stable or tolerant surfaces is a common method of modifying metal implants. In many studies concerning the oxidization of a metal biomaterial, ion release experiments are being carried out (Speck & Fraker 1980, Michiardi et al. 2006). In these studies, improvement of the biocompatibility has been tested by measurements of ion release into simulated physiological

118

solutions. In our study of the oxidation of NiTi, the oxide layer has been in direct contact with cells, and thus, the results reflect the effect of oxidation on cell behavior. It has previously been proposed that the formation of a thick and homogenous stoichiometric TiO2 on the surface of NiTi alloy would improve the corrosion resistance of the material and improve its biocompatibility (Michiardi et al. 2006). In the study of Plant and co-workers (Plant et al. 2005), the authors showed that when brought into contact with the human endothelium, nickel ions from the surface of NiTi are capable of inducing pathological levels of oxidative stress sufficient to cause breakdown of the barrier function of the endothelium. The exposure of human umbilical vein endothelial cells to NiTi also caused loss of vascular-endothelial cadherin and F-actin, thus significantly increasing the paracellular permeability. These pathological phenomena were not found in cells grown on NiTi which had been oxidized at 600°C. In another study, oxidation of NiTi at 600°C and above resulted in a titania surface layer structure, and subsequent immersion into simulated body fluid was observed to induce the deposition of biomimetic bone-like apatite onto the TiO2 coated NiTi surface (Gu et al. 2005). Interestingly, no TiO2 was detected on the surface of NiTi after heat treatment at 300 and 400°C and the formation of apatite was also observed to be dependent of the oxidation temperature. The formation of titania and apatite was most likely dependent on the presence of anatase and/or rutile in the 600 and 800°C heat-treated NiTi which could provide atomic arrangements in their crystal structures suitable for the epitaxy of apatite crystals, and anatase had better apatite-forming ability than rutile (Gu et al. 2005). The results of the present study indicated that the crystal lattice strains were different between the oxidized samples (300, 450 and 600°C), and thus, the lattice distortions were different between the samples studied. This probably had an effect on the structure of the oxide layer; the one produced at 350°C was, for some reason, preferred by the osteoblastic cells.

Although NiTi is always covered with a natural thin oxide layer, mainly composed of TiO2 (Brunette et al. 2001), thickening of the oxidative layer is thought to have protective effects on the cells and tissues in contact with the metal (Firstov et al. 2002, Armitage et al. 2003). We observed that thickening of the surface oxide layer does not improve the NiTi surface biocompatibility when measured as the attachment and cytotoxicity of osteoblastic ROS-17/2.8 cells. The reason for this might lie in the changes of the amount of martensite next to the oxide layer at the oxide–metal contact, since the Gibbs energies of austenite and martensite are different (Wayman 1975). The samples oxidized at the higher

119

temperature (450°C vs. 350°C) had thicker surface oxide layers, but they also had more martensite NiTi at the oxide–metal interface area. It can be hypothesized that these alterations in the phase composition caused changes in the oxide–metal contact potential leading to variations in external surface potential. This probably led to changes in the surface adsorption ability. Our results indicated differences in the lattice structure and properties of the oxidized NiTi, and that this somehow caused different cell behavior. In addition, as already mentioned, lattice distortions in the oxidized samples were different. Hence, the stresses exerted on the surface of the NiTi samples probably were different between the three sample types investigated. Supporting these arguments are results of a recent study of Michiardi and others (Michiardi et al. 2007a), which indicated that the oxide formed by oxidation treatment on the austenitic NiTi did not have the same chemical properties as the one formed on the martensitic phase. The authors suggested that the difference between austenitic and martensitic NiTi was probably due to the structural difference between the phases, which had an influence on the oxidation process and on the growth of the corresponding oxide. Even though the mechanisms behind the observed differences in cell behavior are unclear, the results and hypotheses presented here indicate the need for a deeper understanding of the interface area of a material and its surface. In addition to the chemical composition of the NiTi surface, surface oxide hydration degrees, types of conductivity, surface charge, catalytic activity, structural defects caused by oxygen deficiency and their effects on the electronic structure of Ti surface oxides modified by the presence of Ni, must be explored in order to gain a full understanding of NiTi surface modifications (Shabalovskaya et al. 2008).

Even though oxidation of NiTi alloys seems to be promising for further improvement of NiTi biocompatibility, the oxide layer produced on the alloy surface must withstand the tribological environment after implantation in order to be usable in clinical applications. Oxide layers are brittle and crack under tensile stress (McGuigan et al. 2003). In addition, the thicker the layers, the higher the tendency for crack formation. Thus, thin oxide layers might in this sense be better suited for applications, and this is supported by the present study in which a thin oxide layer was formed on NiTi (< 16 nm) with cell protective properties. However, these results presented here require caution and are only preliminary, since detailed knowledge of the electronic structure of NiTi oxides and their properties are lacking in the literature.

120

6.2.1 Focal adhesions on oxidized NiTi

The use of paxillin staining for studying focal adhesions of osteoblasts on oxidized NiTi surfaces was considered a valid method, since paxillin staining of bone cells has been used in various biomaterial studies for investigating the suitability of a biomaterial (Lim et al. 2007, Finke et al. 2007, Woo et al. 2007). Cell adhesion to a biomaterial surface is a complex and dynamic process, which occurs through a sequence of events that propagate outwards from a surface along both spatial and temporal coordinates. The separate “components of biocompatibility” overlap in time and space, which can cause subsequent events to partially or wholly obscure the preceding. Because of the complexity, no single assay sampling of a particular period of time or space can hope to resolve spatiotemporal components of biocompatibility. (Williams 1998, Lim et al. 2004, Woodruff et al. 2007.) Although it is clear, and in many ways scientifically tested that surface properties of a biomaterial affect cell reactions in multiple ways, the exact molecular mechanisms behind these are mostly still unknown. Despite these difficulties, the results of the different quantity of paxillin-containing focal adhesions on austenite and martensite NiTi demonstrate that these phases have different effects on the formation of adhesive bonds of osteoblasts.

6.3 Fibronectin modulates osteoblast behavior on Nitinol (III)

It is well know that fibronectin binds nonspecifically to substrata of extremely varied chemical composition (Klebe et al. 1981) and fibronectin coating of a biomaterial may improve bone–implant adhesion (Seitz et al. 1982). Once fibronectin is adsorbed onto a TiO2 surface, as in the case of NiTi, which is covered by a naturally formed TiO2 layer, the adsorbed molecules do not desorb easily and retain their tertiary conformation (MacDonald et al. 2002).

Fibronectin has been successfully used as a coating material for NiTi. In the study of Kong and others (Kong et al. 2002), where fibronectin was used as a candidate molecule for coating NiTi devices designated for the closure of intra-atrial communications, a fibronectin coating fully supported endothelial cell adhesion, proliferation and fibronectin matrix formation. The authors concluded that NiTi biocompatibility can be improved by coating with fibronectin with regard to anti-thrombogenicity and endothelialization. A recombinant fragment of FN coated on titanium disks was observed to attract more osteoblasts on the sample surface when compared to vitronectin coated or untreated surfaces (Ku et

121

al. 2005). The recent in vivo and in vitro results of Jimbo et al. (Jimbo et al. 2007) suggested that the adsorption of plasma FN onto the implant surface was effective for earlier osseointegration due to the release of plasma FN and the subsequent chemotaxis of cellular FN-positive bone marrow stromal cells with an osteogenic potential.

Fibronectin interacts with cells through integrin receptors. The integrin binding specificity is dictated by surface chemistry-induced changes in the structure of the adsorbed fibronectin (Keselowsky et al. 2005). The differences in integrin binding profiles most likely regulate osteoblastic gene expression and mineralization via modulation of intracellular signaling pathways that regulate transcriptional activity. Thus, the integrin binding specificity regulates the effects of surface chemistry on osteoblast differentiation (Keselowsky et al. 2003a, Keselowsky et al. 2004, Keselowsky et al. 2005). In our study, fibronectin adsorption and conformation did not differ between austenite and martensite NiTi. Nevertheless, these phases somehow caused differences in cell behavior: fibronectin coated austenite NiTi demonstrated abundant FN rearrangement into fibrils and improved osteoblast proliferation/survival. Fibronectin also elevated the percentage of cells in the G1 phase on both austenite and martensite NiTi.

It would be interesting to study the integrin composition and integrin-mediated signaling proteins of osteoblasts on austenite and martensite in order to investigate the possible differences in integrin profiles. Integrin mediated adhesion-dependent cellular signals are key events in regulating multiple crucial cellular functions such as cell migration, survival, cell cycle progression, proliferation and differentiation. Because of this, it might be postulated that the differences observed in MC3T3-E1 osteoblast behavior on NiTi was most probably due to the differences in integrin profiles, but the reason for this still remain unknown. In the study of Krause et al. (Krause et al. 2000) the authors found that adhesion to FN and Ti6Al4V (titanium-aluminum-vanadium alloy) activated uniform integrin-mediated signaling pathways and similar phosphorylation of MAPK (mitogen-activated protein kinase), resulting in similar cell morphology and proliferation on both FN and Ti6Al4V. Thus, the differences in cell proliferation on FN coated austenite and martensite NiTi might reflect differences in the phosphorylation pattern and/or interaction of a protein tyrosine kinase FAK [focal adhesion kinase; activation requires cell attachment; (Clark & Brugge 1995)] with other signaling molecules (such as MAPK) due to the differences between the substrates, which causes alternate or defective signaling pathways.

122

Oxidation treatment of NiTi was observed to enhance the hydrophilic character of NiTi surfaces, mainly by increasing the polar component of their surface free energy (Michiardi et al. 2007b). The polar component is the interaction energy of a material with water, and it can be used to determine cellular adhesion strength (Hallab et al. 2001). The oxidation treatment increased the absorption levels and affinity of fibronectin for NiTi surfaces (Michiardi et al. 2007b). In the study of Altankov and Groth (Altankov & Groth 1996), the authors demonstrated that the decreasing wettability of the materials significantly reduced endogenous fibronectin deposition on the substratum. In addition, the fibrillar organization of fibronectin appeared only on relatively hydrophilic materials, while on more hydrophobic materials cells secreted some fibronectin but were not able to organize it into a fibronectin matrix. Although not statistically significant, a difference existed between the contact angles measured from austenite and martensite. Martensite was more hydrophobic, i.e. the wettability of the surface was lower than that of austenite. This might have had an effect on the differences in fibronectin reorganization on the two NiTi phases. The two hour time period for cell–FN-coating–NiTi interaction was probably so short that the results were not affected by endogenous fibronectin production, but were merely the results of the reorganization of the artificial FN coat.

As discussed in a number of papers (Vitte et al. 2004, Garcia & Reyes 2005) the determination of cell behavior on an artificial surface is influenced by a number of nonbiochemical signals. Parallel with ligand-receptor mediated cues for cell behavior, the mechanical, physico-chemical and topographical features of the substratum also have an input on the initial adhesion and thus for all the following cellular reactions. As the present study showed, in austenite and martensite NiTi, the physical differences of these two phases had a strong impact on both osteoclasts and osteoblasts. The experiments with FN coating demonstrated that these phase state-dependent differences of NiTi did not have an effect on the adsorption and conformation of human plasma fibronectin, and thus the FN coated samples had the same adsorbed surface protein structure regardless of the sample’s phase state. For some unknown reason FN was more fibrillated by MC3T3-E1 osteoblasts on austenite than on martensite. Osteoblast survival was improved with a fibronectin coating; especially at the early stage of attachment (measured 2 h after cell contact with the FN-NiTi surface). Unlike on martensite, the protecting quality of fibronectin remained on austenite NiTi after 48 hours. Fibronectin is known to regulate osteoblast differentiation and mineralization (Moursi et al. 1996, Globus et al. 1998, Keselowsky et al. 2005). It is thus

123

understandable that the fibronectin coating elevated the number of MC3T3-E1 cells at the G1 phase of the cell cycle, possibly indicating cell differentiation.

As Keselowsky and others (Keselowsky et al. 2005) pointed out, the effects of biomaterial surface properties on cellular responses are generally attributed to material-dependent differences in adsorbed protein species, concentration, and/or biological activity. However, the molecular mechanisms modulating these cellular activities remain poorly understood. The lack of fundamental understanding of cell–material interactions hinders progress toward the development of synthetic materials that elicit the desired cellular responses. In the present study, the fibronectin coating on NiTi probably had different biological activities on austenite and martensite, even though its adsorption and conformation appeared to be similar.

6.4 Biocompatibility of sol-gel coated orthopaedic NiTi implants (IV)

6.4.1 Intramedullary nailing and the effect of mechanical loading

Intramedullary nailing has become a treatment of choice for various bone fractures. Because of their location in the medullary canal, nails do not have nearly as adverse an effect on callus strength as bone fixation plates, despite the fact that the nails are substantially stiffer than bone (Chapman 1998). IM nailing using the groove of the greater trochanter as an entry point to the femoral canal is considered to be a safe method to treat various bone fractures (Papadakis et al. 2005, Ostrum et al. 2005) even in adolescent patients (Townsend & Hoffinger 2000, Kanellopoulos et al. 2006).

Functional IM NiTi nails have been previously used in the studies of Kujala and co-authors (Kujala et al. 2002a, Kujala et al. 2002b). In these investigations, significant retardation of longitudinal growth of the implanted femoral bone was observed, but this effect was evident in both curved and straight NiTi nails, thus it seemed to be due to the nailing itself rather than the bending force. In the present study, two different bending strengths were used, and one provided a higher bending force than those used in the studies of Kujala and others. The results of the present study indicated that the bending force applied inwardly to the bone increased the biocompatibility of the functional NiTi nails without any effect on the macroscopic structure of the bone.

124

The increased biocompatibility, detected as increased bone−implant attachment, might be, at least partly, due to better immobilization of the nail to its position. As Bond et al. wrote, the unlocked IM nails acts to guide the motion of the bone along the longitudinal axis of the implant. The friction of the implant within the medullary cavity determines this resistance to motion (Bong et al. 2007). In this study, the bending NiTi nails were tightly attached to the endosseus surface, and thus provided good anchorage of the nail to the bone surface.

The observed enhancement of bone attachment to the bending NiTi implants is very interesting, since the influence of static forces around osseointegrated implants is a topic of debate (McDonald et al. 1994, Carr et al. 1996, Michaels et al. 1997, Duyck et al. 2001a). It is known that certain prosthesis mismatches induce a static load continuously on the implant and causes strains and stresses to the bone (Duyck et al. 2001a). Bone has a high biological tolerance against static forces (Akin-Nergiz et al. 1998, Duyck et al. 2001a), and this might have contributed to the “unbending” observed in this study despite the use of load inducing NiTi nails. In an in vivo study of the quantity and quality of pre-load on implants supporting a fixed full prosthesis, the authors found bending moments and forces caused by mismatch, but no compromised osseointegration was observed (Duyck et al. 2001b). Others have noted that care should be taken to avoid prosthesis mismatch since significant bone deformation was found around such implants, which could have an effect on the initial marginal bone loss (Jemt & Lekholm 1998). In addition to mismatches creating static load, poor initial fit may decrease the longevity of the implant. The results of Dalton and colleagues (Dalton et al. 1995) indicated that attachment strength and bone ingrowth are significantly affected by gaps in the bone–implant interface, particularly those of more than 1.0 mm.

In the study of Duyck and others, the authors concluded that excessive load can trigger resorption through the induction of micro-damage in the bone (Duyck et al. 2001a). Although excessive loading of bone due to implant-related reasons can cause adverse effects on the normal bone remodeling cycle, an appropriate amount of load is important for successful osseointegration. As Brånemark stated, it is critical for the maintenance of osseointegration that the carefully controlled and prosthetic-induced loading of the implant−tissue interface is maintained (Brånemark et al. 1985). The controlled mechanical environment is necessary to assure an adequate remodeling stimulus for maintenance of integration. Since bone modeling and remodeling are the result of the interactions of the mechanical environment (e.g. load magnitudes, load directions, loading rates, and loading

125

frequencies) and the biological milieu (e.g. bone quality, bone quantity, and the ability of cells to respond) (Hoshaw et al. 1994), it is difficult to define the properties of the implant required to achieve the desired bone response, in this case, the "overload" needed to induce the bending of the femoral bone by the intramedullary NiTi nail. The precise “overload” of bone required to induce negative effects on osseointegration is yet to be determined. It is also important to note that the majority of the studies concerning loading effects on bones have been done with dental implants, where the cyclic loading of occlusion has great importance to the prosthesis/implant−tissue interaction. Bone implants in the weight bearing parts of the body are also subjected to cyclic loading, e.g. walking, but comprehensive studies of the load effects on these implant types and their surrounding tissues seem to be rare.

The overall success of osseointegrated implants depends on factors concerning biological, biomechanical and clinical practice. Of these aspects the biomechanical properties of the implant and the peri-implant bone is of great significance. Bone implants must be capable of transferring the load to peri-implant tissue, thus enabling the bearing capacity of the system even after the introduction of an implant. The stress/strain level and its history in the bone act as mechanical stimuli for bone remodeling (Bonewald 2006, Aguirre et al. 2007, Malone et al. 2007, Tatsumi et al. 2007). The magnitude and frequency of stress is closely related to the possible improvement or decrease of bone biomechanical properties. In addition, the variable elastic properties of bone tissue surrounding the implant affect the deformability of the bone−implant system and stress/strain distribution on the bone tissue (Natali et al. 2006). In the present study, the stress distribution of the implants differed due to the structure of the austenite NiTi nail. Curved nails created compressive forces on the concave side of the nail and tension forces on the convex side. Although not measured, the highest stresses were probably present in the middle part of the twisting nail and also in the tips of the nail pushing towards the bone endosteum. The magnitude of force directed to the bone by the bending NiTi nail was unfortunately not studied. This information would have been an important addition to the results and would have helped in gaining more accurate and detailed results of the effect of the intramedullary functional NiTi nail. Overall, the results presented here did not reveal loss of bone contact in different areas of the bending implants.

When the distal tips of the curved NiTi nails were in contact with the bone cortex, a cup-like bone structure could be seen in most of the antero-posterior radiographs. This "cup" was possibly due to a mismatch in the radius of curvature

126

between the implant and the femur, resulting in a partial cortical penetration (Bong et al. 2007). In addition to nail penetration into the cortex, there seemed to be bone formation below the tip of the nail, thus creating a cup-like structure around the tip. The interpretations of these observations require caution, since the radiographic findings could not be histologically visualized. It is possible that the NiTi nail pushing toward the endosteum caused both bone formation and resorption. The bones with a curved IM NiTi nail were subjected to multiple forces, such as local loading, uneven stress transfer and bending strains – all of these affecting bone turnover and site specific bone formation (Engh et al. 1987, Raab-Cullen et al. 1994).

Distortion of the physiological milieu by a surgical procedure induces the normal regional acceleratory phenomenon (RAP) (Frost 1989). RAP is an increase in all metabolic activities in a part of the body that has been attacked by any provocative stimulus. In the study of Ryhänen and colleagues (Ryhänen et al. 1999), the authors found no negative effect on the normal RAP of bone after NiTi periosteal implantation. In the present study, RAP might have played a role in the observed thickening of the proximal third of all of the operated femurs by inducing bone turnover with a tendency toward excess bone formation. In addition, we might hypothesize that in this study the increase of bone formation after implantation of the curved intramedullary NiTi nails was partly due to the stronger stimulus of RAP when compared to the straight nails. The curved nails provided increased mechanical loading, which might have increased the RAP, resulting in higher bone formation. In addition to early osseointegration, bone implants have to allow normal bone turnover in order to be fully biocompatible. Thus, the maintenance of bone−implant contact and the quality of bone around the implant are important factors that determine the long-term biocompatibility of an implant.

Even though this study failed to produce macroscopic changes in the architecture of healthy femoral bone, the use of NiTi-based shape memory IM implants can provide interesting future advancements in IM applications (Bong et al. 2007). Due to the martensitic transformation, even bent implants are easy to insert. The human femur has an average radius of curvature of 120 ± 36 cm, but the nails currently used as femoral IM implants have radii ranging from 186 to 300 cm (Egol et al. 2004). With NiTi, femoral nails that follow the natural curvature radius of the human femur can be manufactured and implanted as straightened in their martensitic form. The histological result of this study indicates good osseointegration of bent intramedullary NiTi nails.

127

6.4.2 Sol-gel surface treatment of NiTi

The sol-gel surface treatment with TiO2-SiO2 improved the biocompatibility of the intramedullary NiTi nails. This surface treatment increased bone–implant attachment and decreased fibrous tissue–implant contact. The improved biocompatibility of might be due to the ability of the modified surface to attract calcium and phosphate from body fluids, depositing as CaP on the implant surface and rendering it more suitable for direct bone bonding. In the study by Areva and colleagues, the authors found that sol-gel derived nano-porous titania coated titanium implants formed close contacts with adjacent connective tissue, whereas a gap and fibrous capsule formed at the control implant–tissue interface (Areva et al. 2004). In a earlier article, the authors (Areva et al. 2002) suggested that the formation of bonelike apatite on a sol-gel derived titania coating occurs via continuous dissolution/reprecipitation processes. However, proteins might quench this formation; i.e. if proteins from surrounding tissue adsorb to the initial CaP layer, this could inhibit further growth of the CaP to bonelike apatite. In addition, sol-gel processing of Ti-alloy has been shown to promote the capacity to initiate CaP nucleation, and to increase the attachment of osteoblasts as well as the number of mineralized nodules under cell culture conditions (Advincula et al. 2006).

Nanophase ceramics, such as the sol-gel derived titania coating, have been shown to adsorb vitronectin, which partly explains the subsequent enhanced adhesion of osteoblasts on these ceramics (Webster et al. 2000). Furthermore, sol-gel deposition of TiO2 and subsequent steam treatment on NiTi has been shown to improve the corrosion resistance of the alloy (Chiu et al. 2007). In addition to improved corrosion properties, sol-gel derived titania was also observed to increase the blood compatibility of the NiTi alloy (Liu et al. 2003). The results of a recent study of cell responses on titanium surfaces with different ratios of TiO2/SiO2 in a sol-gel derived coating showed good fibroblast and osteoblast response to the surfaces (Areva et al. 2007). These results demonstrated osteoblast activity on both 70:30 and 30:70 TiO2-SiO2 surfaces; the latter prolonged osteoblast activity and calcified nodule formation in vitro. The in vivo results of the present study support the use of a 70:30 ratio of TiO2-SiO2 as an osteoinductive coating, but other ratios of titania and silica should be tested in order to optimize the coating method. In addition, the cellular mechanisms behind the improved biocompatibility due to the sol-gel derived implant coating should be further examined. Furthermore, the sol-gel derived surface finishing can be

128

further developed with methods such as CO2 laser for direct densification and local activation of the coating, and for creating implants with varying bioactivity (Moritz et al. 2003).

The findings of the present study, supported by other research in the field, offer promising indications for the safe and versatile use of sol-gel surface treated NiTi bone implants. The results encourage further studies concerning the sol-gel derived nonresorbable TiO2-SiO2 nanocoating as a reactive and biofunctional implant surface treatment.

129

7 Conclusions

In study I, the results showed that austenite and martensite phases of NiTi SMA had different impacts on rat primary osteoclasts. On martensite, both osteoclast morphology and activity deteriorated. The results indicate that the biocompatibility of NiTi is different in the austenite and martensite phase in respect to osteoclast adhesion and growth. Additional studies are needed to further investigate the mechanisms behind the differences observed in osteoclast behavior.

Study II showed that NiTi biocompatibility can be improved with a TiO2 surface treatment, although the thickness of the coating does not necessarily correlate with the improvement. Oxidation of austenite NiTi caused a martensite phase presence in the alloy structure, the higher the oxidation temperature, the higher the amount of martensite. This caused differences in the nano-structure of the oxidized samples, which probably resulted in different macromolecule adsorption onto the alloy surface. As a result, osteoblasts adhered better on nonoxidized austenite and to the modest oxide coating compared to the thicker oxide surface.

The results of study III demonstrated that a fibronectin coating improves NiTi biocompatibilily in osteoblast culture. The protective effect was stronger on the austenite phase. Fibronectin remodeling was evident on both phases, but it appeared to be more pronounced on the austenite surface. Nevertheless, the comformation of fibronectin showed no differences between the phases when studied by antibody-binding capacity. The fibronectin coating increased the number of cells in the G1 stage of the cell cycle on both phases, possibly indicating cell differentiation.

In study IV, functional IM NiTi nails were implanted in a proximal to distal direction into rat femurs. Some of the nails were TiO2-SiO2 coated by sol-gel deposition. The implanted nails were either curved or straight in their customized austenite shape. Two types of curved nails were used with two different curvature radii. The results showed that sol-gel surface treatment of titania-silica coating improved the bone–implant attachment. The constant load applied by the curved nails enhanced this attachment, but did not result in changes in the macroscopic structure of the bones.

As an overall conclusion, NiTi SMA is well tolerated but the mechanically induced martensite phase should be avoided when implants are designed. Also the possibility of enhancing IM nail fixation by careful control of the austenite

130

implant design should be further studied. Various degrees of titanium oxide and titania-silica coating can improve the biocompatibility of the NiTi implants probably by influencing the adsorption of ECM proteins on the surface.

131

References Aarden EM, Burger EH & Nijweide PJ (1994) Function of osteocytes in bone. J Cell

Biochem 55: 287-299. Ääritalo V, Areva S, Jokinen M, Linden M & Peltola T (2007) Sol-gel-derived TiO2-SiO2

implant coatings for direct tissue attachment. Part I: Design, preparation and characterization. J Mater Sci: Mater Med 18: 1863-1873.

Abercrombie M, Heaysman JE & Pegrum SM (1971) The locomotion of fibroblasts in culture. IV. Electron microscopy of the leading lamella. Exp Cell Res 67: 359-367.

Acharya KN & Rao MR (2006) Retrograde nailing for distal third femoral shaft fractures: a prospective study. J Orthop Surg (Hong Kong ) 14: 253-258.

Adams GB & Scadden DT (2006) The hematopoietic stem cell in its place. Nat Immunol 7: 333-337.

Adams JC & Watt FM (1993) Regulation of development and differentiation by the extracellular matrix. Development 117: 1183-1198.

Advincula MC, Rahemtulla FG, Advincula RC, Ada ET, Lemons JE & Bellis SL (2006) Osteoblast adhesion and matrix mineralization on sol-gel-derived titanium oxide. Biomaterials 27: 2201-2212.

Aguirre JI, Plotkin LI, Gortazar AR, Martin MM, O'Brien CA, Manolagas SC & Bellido T (2007) A novel ligand-independent function of the estrogen receptor is essential for osteocyte and osteoblast mechanotransduction. J Biol Chem 31: 25501-25508.

Ahmadibeni Y, Hanley M, White M, Ayrapetov M, Lin X, Sun G & Parang K (2007) Metal-binding properties of a dicysteine-containing motif in protein tyrosine kinases. Chembiochem 8: 1592-1605.

Aho A, Anderson Ö, Heikkilä J, Keränen J, Konttinen Y, Lappalainen R, Lepojärvi M, Nevalainen J, Santavirta S, Salenius J, Tarvainen T, Törmälä P, Vallittu P, Viljanen V, Waris E & Waris V (2003) Yleiskatsaus terveydenhuollon laitteissa ja tarvikkeissa käytettyihin biomateriaaleihin. Lääkelaitoksen julkaisusarja.

Akin-Nergiz N, Nergiz I, Schulz A, Arpak N & Niedermeier W (1998) Reactions of peri-implant tissues to continuous loading of osseointegrated implants. Am J Orthod Dentofacial Orthop 114: 292-298.

Allen LT, Fox EJ, Blute I, Kelly ZD, Rochev Y, Keenan AK, Dawson KA & Gallagher WM (2003) Interaction of soft condensed materials with living cells: phenotype/transcriptome correlations for the hydrophobic effect. Proc Natl Acad Sci U S A 100: 6331-6336.

Alley MC, Scudiero DA, Monks A, Hursey ML, Czerwinski MJ, Fine DL, Abbott BJ, Mayo JG, Shoemaker RH & Boyd MR (1988) Feasibility of drug screening with panels of human tumor cell lines using a microculture tetrazolium assay. Cancer Res 48: 589-601.

Alnaeeli M, Park J, Mahamed D, Penninger JM & Teng YT (2007) Dendritic cells at the osteo-immune interface: implications for inflammation-induced bone loss. J Bone Miner Res 22: 775-780.

132

Altankov G & Groth T (1996) Fibronectin matrix formation by human fibroblasts on surfaces varying in wettability. J Biomater Sci Polym Ed 8: 299-310.

Anderson JM (1988) Inflammatory response to implants. ASAIO Trans 34: 101-107. Anderson JM (2000) Multinucleated giant cells. Curr Opin Hematol 7: 40-47. Anderson JM (2001) Biological responses to materials. Annu Rev Mater Res 31: 81-110. Anderson JM & Miller KM (1984) Biomaterial biocompatibility and the macrophage.

Biomaterials 5: 5-10. Andrade JD & Hlady V (1986) Protein adsorption and materials biocompatibility - a

tutorial review and suggested hypotheses. Adv Polym Sci 79: 1-63. Antonelli L (1989) Closed intramedullary nailing of diaphyseal fractures of the femur.

Problems related to anatomical variations of the greater trochanter. Ital J Orthop Traumatol 15: 67-74.

Arai F, Hirao A, Ohmura M, Sato H, Matsuoka S, Takubo K, Ito K, Koh GY & Suda T (2004) Tie2/angiopoietin-1 signaling regulates hematopoietic stem cell quiescence in the bone marrow niche. Cell 118: 149-161.

Areva S, Ääritalo V, Tuusa M, Jokinen M, Linden M & Peltola T (2007) Sol-gel-derived TiO2-SiO2 implant coatings for direct tissue attachment. Part II: Evaluation of cell responce. J Mater Sci: Mater Med 18: 1633-1643.

Areva S, Paldan H, Peltola T, Närhi T, Jokinen M & Linden M (2004) Use of sol-gel-derived titania coating for direct soft tissue attachment. J Biomed Mater Res A 70: 169-178.

Areva S, Peltola T, Sailynoja E, Laajalehto K, Linden M & Rosenholm JB (2002) Effect of albumin and fibrinogen on calcium phosphate formation on sol-gel-derived titania coatings in vitro. Chem Mater 14: 1614-1621.

Armitage DA & Grant DM (2003) Characterisation of surface-modified nickel titanium alloys. Mat Sci Eng A-Struct 349: 89-97.

Armitage DA, Parker TL & Grant DM (2003) Biocompatibility and hemocompatibility of surface-modified NiTi alloys. J Biomed Mater Res 66A: 129-137.

Aro H, Eerola E & Aho AJ (1982) Osteolysis after rigid fixation. The possible role of periosteal neural mechanoreceptors in bone remodeling. Clin Orthop Relat Res: 292-300.

Aspenberg P & Herbertsson P (1996) Periprosthetic bone resorption. Particles versus movement. J Bone Joint Surg Br 78: 641-646.

Assad M, Yahia LH, Rivard CH & Lemieux N (1998) In vitro biocompatibility assessment of a nickel-titanium alloy using electron microscopy in situ end-labeling (EM-ISEL). J Biomed Mater Res 41: 154-161.

ASTM Diffraction Data File (1969) ASTH, Philadelphia, 73-2224, 21-1276. Avnur Z & Geiger B (1981) The removal of extracellular fibronectin from areas of cell-

substrate contact. Cell 25: 121-132. Balasubramanian L, Ahmed A, Lo CM, Sham JS & Yip KP (2007) Integrin-mediated

mechanotransduction in renal vascular smooth muscle cells : activation of calcium sparks. Am J Physiol Regul Integr Comp Physiol 293: R1586-R1594.

133

Baron R (2003) General principles of bone biology. In Primer on the Metabolic Bone Disease and Disorders of Mineral Metabolism, 5 ed. American Society for Bone and Mineral Research,p 1-8.

Barras CD & Myers KA (2000) Nitinol - its use in vascular surgery and other applications. Eur J Vasc Endovasc Surg 19: 564-569.

Bauer TW, Saltarelli M, McMahon JT & Wilde AH (1993) Regional dissemination of wear debris from a total knee prosthesis. A case report. J Bone Joint Surg Am 75: 106-111.

Bauer TW & Schils J (1999) The pathology of total joint arthroplasty.II. Mechanisms of implant failure. Skeletal Radiol 28: 483-497.

Berglundh T, Abrahamsson I, Lang NP & Lindhe J (2003) De novo alveolar bone formation adjacent to endosseous implants. Clin Oral Implants Res 14: 251-262.

Bershadsky AD, Balaban NQ & Geiger B (2003) Adhesion-dependent cell mechanosensitivity. Annu Rev Cell Dev Biol 19: 677-695.

Bertheville B (2006) Porous single-phase NiTi processed under Ca reducing vapor for use as a bone graft substitute. Biomaterials 27: 1246-1250.

Bi Y, Seabold JM, Kaar SG, Ragab AA, Goldberg VM, Anderson JM & Greenfield EM (2001) Adherent endotoxin on orthopedic wear particles stimulates cytokine production and osteoclast differentiation. J Bone Miner Res 16: 2082-2091.

Black J, Hasting G. (1998). Handbook of Biomaterial Properties.Chapman & Hall, Springer.

Bogdanski D, Esenwein SA, Prymak O, Epple M, Muhr G & Koller M (2004) Inhibition of PMN apoptosis after adherence to dip-coated calcium phosphate surfaces on a NiTi shape memory alloy. Biomaterials 25: 4627-4632.

Bonewald LF (2006) Mechanosensation and transduction in osteocytes. BoneKEy Osteovision 3: 7-15.

Bong MR, Kummer FJ, Koval KJ & Egol KA (2007) Intramedullary nailing of the lower extremity: biomechanics and biology. J Am Acad Orthop Surg 15: 97-106.

Bostman O, Varjonen L, Vainionpää S, Majola A & Rokkanen P (1989) Incidence of local complications after intramedullary nailing and after plate fixation of femoral shaft fractures. J Trauma 29: 639-645.

Boudreau NJ & Jones PL (1999) Extracellular matrix and integrin signalling: the shape of things to come. Biochem J 339: 481-488.

Bourdelat D (1996) Fracture of the femoral shaft in children: advantages of the descending medullary nailing. J Pediatr Orthop B 5: 110-114.

Bourdoulous S, Orend G, MacKenna DA, Pasqualini R & Ruoslahti E (1998) Fibronectin matrix regulates activation of RHO and CDC42 GTPases and cell cycle progression. J Cell Biol 143: 267-276.

Boyde A, Ali NN & Jones SJ (1984) Resorption of dentine by isolated osteclasts in vitro. Brit Dent J 156: 216-220.

Boyle WJ, Simonet WS & Lacey DL (2003) Osteoclast differentiation and activation. Nature 423: 337-342.

134

Brånemark PI, Zarb GA, Albrektsson T. (1985). Tissue-integrated prostheses: osseointegration in clinical dentistry. Chicago: Quintessence.

Braun A, Papp J & Reiter A (2003) The periprosthetic bone remodelling process - signs of vital bone reaction. Int Orthop 27 Suppl 1: S7-10.

Brunette DM et al. (2001). Titanium in Medicine. Berlin-Heidelberg-New York: Springer-Verlag.

Buechsenschuetz KE, Mehlman CT, Shaw KJ, Crawford AH & Immerman EB (2002) Femoral shaft fractures in children: traction and casting versus elastic stable intramedullary nailing. J Trauma 53: 914-921.

Buehler WJ & Wang FE (1968) A summary of recent research on the Nitinol* alloys and their potential application in ocean engineering. Ocean Engng 1: 105-120.

Burstein AH, Reilly DT & Martens M (1976) Aging of bone tissue: mechanical properties. J Bone Joint Surg Am 58: 82-86.

Caglar O, Aksoy MC, Yazici M & Surat A (2006) Comparison of compression plate and flexible intramedullary nail fixation in pediatric femoral shaft fractures. J Pediatr Orthop B 15: 210-214.

Cai YL, Liang CY, Zhu SL, Cui ZD & Yang XJ (2006) Formation of bonelike apatite-collagen composite coating on the surface of NiTi shape memory alloy. Scripta Mater 54: 89-92.

Carey TP & Galpin RD (1996) Flexible intramedullary nail fixation of pediatric femoral fractures. Clin Orthop Relat Res: 110-118.

Carr AB, Gerard DA & Larsen PE (1996) The response of bone in primates around unloaded dental implants supporting prostheses with different levels of fit. J Prosthet Dent 76: 500-509.

Carter DR & Spengler DM (1978) Mechanical properties and composition of cortical bone. Clin Orthop Relat Res 135: 192-217.

Cederbrant K, Anderson C, Andersson T, Marcusson-Stahl M & Hultman P (2003) Cytokine production, lymphocyte proliferation and T-cell receptor Vbeta expression in primary peripheral blood mononuclear cell cultures from nickel-allergic individuals. Int Arch Allergy Immunol 132: 373-379.

Chambers TJ, Revell PA, Fuller K & Athanasou NA (1984) Resorption of bone by isolated rabbit osteoclasts. J Cell Sci 66: 383-399.

Chan CM, Trigwell S & Duerig T (1990) Oxidation of an NiTi alloy. Surf Interface Anal 15: 349-354.

Chapman MW (1998) The effect of reamed and nonreamed intramedullary nailing on fracture healing. Clin Orthop Relat Res 355: S230-S238.

Chen MF, Yang XJ, Hu RX, Cui ZD & Man HC (2004) Bioactive NiTi shape memory alloy used as bone bonding implants. Mat Sci Eng C-Bio S 24: 497-502.

Cheung G, Zalzal P, Bhandari M, Spelt JK & Papini M (2004) Finite element analysis of a femoral retrograde intramedullary nail subject to gait loading. Med Eng Phys 26: 93-108.

Chibowski E & Perea-Carpio R (2002) Problems of contact angle and solid surface free energy determination. Adv Colloid Interface Sci 98: 245-264.

135

Chiu KY, Wong MH, Cheng FT & Man HC (2007) Characterization and corrosion studies of titania-coated NiTi prepared by sol-gel technique and steam crystallization. Appl Surf Sci 253: 6762-6768.

Chu PK (2006) Bioactivity of plasma implanted biomaterials. Nucl Instrum Meth B 242: 1-7.

Chu PK (2007a) Plasma surface treatment of artificial orthopedic and cardiovascular biomaterials. Surf Coat Tech 201: 5601-5606.

Chu PK (2007b) Enhancement of surface properties of biomaterials using plasma-based technologies. Surf Coat Tech 201: 8076-8082.

Clark EA & Brugge JS (1995) Integrins and signal transduction pathways: the road taken. Science 268: 233-239.

Clark P, Connolly P, Curtis AS, Dow JA & Wilkinson CD (1991) Cell guidance by ultrafine topography in vitro. J Cell Sci 99: 73-77.

Clover J, Dodds RA & Gowen M (1992) Integrin subunit expression by human osteoblasts and osteoclasts in situ and in culture. J Cell Sci 103: 267-271.

Colognato H, Galvin J, Wang Z, Relucio J, Nguyen T, Harrison D, Yurchenco PD & Ffrench-Constant C (2007) Identification of dystroglycan as a second laminin receptor in oligodendrocytes, with a role in myelination. Development 134: 1723-1736.

Cowles EA, Derome ME, Pastizzo G, Brailey LL & Gronowicz GA (1998) Mineralization and the expression of matrix proteins during in vivo bone development. Calcif Tissue Int 62: 74-82.

Cui J, Chu YS, Famodu OO, Furuya Y, Hattrick-Simpers J, James RD, Ludwig A, Thienhaus S, Wuttig M, Zhang Z & Takeuchi I (2006) Combinatorial search of thermoelastic shape-memory alloys with extremely small hysteresis width. Nat Mater 5: 286-290.

Currey JD (2008) Ontogenetic changes in compact bone material properties. In Cowin SC (ed) Bone Mechanics Handbook, 2 ed. Informa Healtcare USA, Inc., New York, p 19 1-16.

Dalton JE, Cook SD, Thomas KA & Kay JF (1995) The effect of operative fit and hydroxyapatite coating on the mechanical and biological response to porous implants. J Bone Joint Surg Am 77: 97-110.

Damsky CH, Moursi A, Zhou Y, Fisher SJ & Globus RK (1997) The solid state environment orchestrates embryonic development and tissue remodeling. Kidney Int 51: 1427-1433.

Damsky CH & Werb Z (1992) Signal transduction by integrin receptors for extracellular matrix: cooperative processing of extracellular information. Curr Opin Cell Biol 4: 772-781.

Danilov A, Kapanen A, Kujala S, Saaranen J, Ryhänen J, Pramila A, Jämsä T & Tuukkanen J (2003) Biocompatibility of austenite and martensite phases in NiTi-based alloys. J Phys IV France 112: 1117-1120.

de Belder KR (1968) Distal migration of the femoral intramedullary nail. Report of seven cases. J Bone Joint Surg Br 50: 324-333.

136

Decker T & Lohmann-Matthes ML (1988) A quick and simple method for the quantitation of lactate dehydrogenase release in measurements of cellular cytotoxicity and tumor necrosis factor (TNF) activity. J Immunol Methods 115: 61-69.

Defilippi P, Olivo C, Venturino M, Dolce L, Silengo L & Tarone G (1999) Actin cytoskeleton organization in response to integrin-mediated adhesion. Microsc Res Tech 47: 67-78.

Delaisse J-M & Vaes G (1992) Mechanism of mineral solubility and matrix degradation in osteoclastic bone resorption. In Riflein BR& Gay C (eds) Biology and Physiology of the Osteoclast. p 289-314.

Diener A, Nebe B, Luthen F, Becker P, Beck U, Neumann HG & Rychly J (2005) Control of focal adhesion dynamics by material surface characteristics. Biomaterials 26: 383-392.

Dotter CT, Buschmann RW, McKinney MK & Rosch J (1983) Transluminal expandable nitinol coil stent grafting: preliminary report. Radiology 147: 259-260.

Duerig TW, Pelton AR & Stockel D (1996) The utility of superelasticity in medicine. Biomed Mater Eng 6: 255-266.

Duerig TW, Tolomeo DE & Wholey M (2000) An overview of superelastic stent design. Min Invas Ther Allied Technol 9: 235.

Dunn MG & Maxian SH (1994) Biomaterials used in orthopaedic surgery. In Creco RS (ed) Implantation Biology: the Host Responce and Biomedical Devices. CRC Press, Bona Raton, p 229-252.

Duong LT & Rodan GA (1998) Integrin-mediated signaling in the regulation of osteoclast adhesion and activation. Front Biosci 3: d757-d768.

Duyck J, Ronold HJ, Van Oosterwyck H, Naert I, Vander SJ & Ellingsen JE (2001a) The influence of static and dynamic loading on marginal bone reactions around osseointegrated implants: an animal experimental study. Clin Oral Implants Res 12: 207-218.

Duyck J, Van Oosterwyck H, Vander SJ, De Cooman M, Puers R & Naert I (2001b) Pre-load on oral implants after screw tightening fixed full prostheses: an in vivo study. J Oral Rehabil 28: 226-233.

Egol KA, Chang EY, Cvitkovic J, Kummer FJ & Koval KJ (2004) Mismatch of current intramedullary nails with the anterior bow of the femur. J Orthop Trauma 18: 410-415.

Einhorn TA (1996) Biomechanics of bone. In Bilizikian JP, Raisz LG & Rodan GA (eds) Principles of Bone Biology. Academic Press, San Diego, CA, USA, p 25-50.

Einhorn TA (1992) Bone strength: the bottom line. Calcif Tissue Int 51: 333-339. Engh CA, Bobyn JD & Glassman AH (1987) Porous-coated hip replacement. The factors

governing bone ingrowth, stress shielding, and clinical results. J Bone Joint Surg Br 69: 45-55.

Engh CA, McGovern TF, Bobyn JD & Harris WH (1992) A quantitative evaluation of periprosthetic bone-remodeling after cementless total hip arthroplasty. J Bone Joint Surg Am 74: 1009-1020.

137

Eriksen EF, Eghbali-Fatourechi GZ & Khosla S (2007) Remodeling and vascular spaces in bone. J Bone Miner Res 22: 1-6.

Erli HJ, Marx R, Paar O, Niethard FU, Weber M & Wirtz DC (2003) Surface pretreatments for medical application of adhesion. Biomed Eng Online 2: 15.

Es-Souni M, Es-Souni M & Fischer-Brandies H (2005) Assessing the biocompatibility of NiTi shape memory alloys used for medical applications. Anal Bioanal Chem 381: 557-567.

Eveleigh RJ (1995) A review of biomechanical studies of intramedullary nails. Med Eng Phys 17: 323-331.

Faucheux N & Nagel MD (2002) Cyclic AMP - dependent aggregation of Swiss 3T3 cells on a cellulose substratum (Cuprophan) and decreased cell membrane Rho A. Biomaterials 23: 2295-2301.

Faucheux N, Schweiss R, Lutzow K, Werner C & Groth T (2004) Self-assembled monolayers with different terminating groups as model substrates for cell adhesion studies. Biomaterials 25: 2721-2730.

Faucheux N, Tzoneva R, Nagel MD & Groth T (2006) The dependence of fibrillar adhesions in human fibroblasts on substratum chemistry. Biomaterials 27: 234-245.

Feng JQ, Ward LM, Liu S, Lu Y, Xie Y, Yuan B, Yu X, Rauch F, Davis SI, Zhang S, Rios H, Drezner MK, Quarles LD, Bonewald LF & White KE (2006) Loss of DMP1 causes rickets and osteomalacia and identifies a role for osteocytes in mineral metabolism. Nat Genet 38: 1310-1315.

Finke B, Luethen F, Schroeder K, Mueller PD, Bergemann C, Frant M, Ohl A & Nebe BJ (2007) The effect of positively charged plasma polymerization on initial osteoblastic focal adhesion on titanium surfaces. Biomaterials 28: 4521-4534.

Firstov GS, Vitchev RG, Kumar H, Blanpain B & Van Humbeeck J (2002) Surface oxidation of NiTi shape memory alloy. Biomaterials 23: 4863-4871.

Frisch SM & Screaton RA (2001) Anoikis mechanisms. Curr Opin Cell Biol 13: 555-562. Frost HM (1989) The biology of fracture healing. An overview for clinicians, Part II. Clin

Orthop 248: 294-309. Frost HM (1960) In vivo osteocyte death. J Bone Joint Surg Am 42-A: 138-143. Frost HM (1987) The mechanostat: a proposed pathogenic mechanism of osteoporoses and

the bone mass effects of mechanical and nonmechanical agents. Bone Miner 2: 73-85. Furlan F, Lecanda F, Screen J & Civitelli R (2001) Proliferation, differentiation and

apoptosis in connexin43-null osteoblasts. Cell Commun Adhes 8: 367-371. Gadegone WM & Salphale YS (2006) Proximal femoral nail - an analysis of 100 cases of

proximal femoral fractures with an average follow up of 1 year. Int Orthop. Galilei G. (1638). Discoursi e demonstrazione matematiche intorno á due nuove scienze.

Leiden: Elzevir Galpin RD, Willis RB & Sabano N (1994) Intramedullary nailing of pediatric femoral

fractures. J Pediatr Orthop 14: 184-189. Garcia AJ & Reyes CD (2005) Bio-adhesive surfaces to promote osteoblast differentiation

and bone formation. J Dent Res 84: 407-413.

138

Garcia AJ, Vega MD & Boettiger D (1999) Modulation of cell proliferation and differentiation through substrate-dependent changes in fibronectin conformation. Mol Biol Cell 10: 785-798.

Gausepohl T, Pennig D, Koebke J & Harnoss S (2002) Antegrade femoral nailing: an anatomical determination of the correct entry point. Injury 33: 701-705.

Giancotti FG & Ruoslahti E (1990) Elevated levels of the alpha 5 beta 1 fibronectin receptor suppress the transformed phenotype of Chinese hamster ovary cells. Cell 60: 849-859.

Gindl M, Sinn G, Gindl W, Reiterer A & Tschegg S (2001) A comparison of different methods to calculate the surface free energy of wood using contact angle measurements. Colloid Surface A 181: 279-287.

Glassman AH, Bobyn JD & Tanzer M (2006) New femoral designs: do they influence stress shielding? Clin Orthop Relat Res 453: 64-74.

Globus RK, Doty SB, Lull JC, Holmuhamedov E, Humphries MJ & Damsky CH (1998) Fibronectin is a survival factor for differentiated osteoblasts. J Cell Sci 111: 1385-1393.

Gonzalez-Herranz P, Burgos-Flores J, Rapariz JM, Lopez-Mondejar JA, Ocete JG & Amaya S (1995) Intramedullary nailing of the femur in children. Effects on its proximal end. J Bone Joint Surg Br 77: 262-266.

Gorbet MB & Sefton MV (2004) Biomaterial-associated thrombosis: roles of coagulation factors, complement, platelets and leukocytes. Biomaterials 25: 5681-5703.

Gothlin G & Ericsson JL (1976) The osteoclast: review of ultrastructure, origin, and structure-function relationship. Clin Orthop Relat Res: 201-231.

Greninger AB & Mooradian VG (1938) Strain transformation in metastable beta copper-zinc and beta copper-tin alloy. Trans Met Soc AIME 128: 337-368.

Gristina AG & Costerton JW (1985) Bacterial adherence to biomaterials and tissue. The significance of its role in clinical sepsis. J Bone Joint Surg Am 67: 264-273.

Gronthos S, Stewart K, Graves SE, Hay S & Simmons PJ (1997) Integrin expression and function on human osteoblast-like cells. J Bone Miner Res 12: 1189-1197.

Gu YW, Li H, Tay BY, Lim CS, Yong MS & Khor KA (2006) In vitro bioactivity and osteoblast response of porous NiTi synthesized by SHS using nanocrystalline Ni-Ti reaction agent. J Biomed Mater Res A 78: 316-323.

Gu YW, Tay BY, Lim CS & Yong MS (2005) Biomimetic deposition of apatite coating on surface-modified NiTi alloy. Biomaterials 26: 6916-6923.

Gumbiner BM (1996) Cell adhesion: the molecular basis of tissue architecture and morphogenesis. Cell 84: 345-357.

Hallab NJ, Bundy KJ, O'Connor K, Moses RL & Jacobs JJ (2001) Evaluation of metallic and polymeric biomaterial surface energy and surface roughness characteristics for directed cell adhesion. Tissue Eng 7: 55-71.

Halleen JM, Räisänen S, Salo JJ, Reddy SV, Roodman GD, Hentunen TA, Lehenkari PP, Kaija H, Vihko P & Väänänen HK (1999) Intracellular fragmentation of bone resorption products by reactive oxygen species generated by osteoclastic tartrate-resistant acid phosphatase. J Biol Chem 274: 22907-22910.

139

Harris LJ & Tarr RR (1979) Implant failures in orthopaedic surgery. Biomater Med Devices Artif Organs 7: 243-255.

Hauge EM, Qvesel D, Eriksen EF, Mosekilde L & Melsen F (2001) Cancellous bone remodeling occurs in specialized compartments lined by cells expressing osteoblastic markers. J Bone Miner Res 16: 1575-1582.

Haynes DR, Crotti TN & Zreiqat H (2004) Regulation of osteoclast activity in peri-implant tissues. Biomaterials 25: 4877-4885.

Hedgecock NL, Hadi T, Chen AA, Curtiss SB, Martin RB & Hazelwood SJ (2007) Quantitative regional associations between remodeling, modeling, and osteocyte apoptosis and density in rabbit tibial midshafts. Bone 40: 627-637.

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: 335-343.

Hengsberger S, Kulik A & Zysset P (2002) Nanoindentation discriminates the elastic properties of individual human bone lamellae under dry and physiological conditions. Bone 30: 178-184.

Herscovici DJr & Whiteman KW (1996) Retrograde nailing of the femur using an intercondylar approach. Clin Orthop Relat Res: 98-104.

Hert J, Fiala P & Petrtyl M (1994) Osteon orientation of the diaphysis of the long bones in man. Bone 15: 269-277.

Hey Groves EE (1918) Ununited fractures with special reference to gunshot injuries and the use of bone grafting. Br J Surg 6: 203-247.

Hill PA, Murphy G, Docherty AJ, Hembry RM, Millican TA, Reynolds JJ & Meikle MC (1994) The effects of selective inhibitors of matrix metalloproteinases (MMPs) on bone resorption and the identification of MMPs and TIMP-1 in isolated osteoclasts. J Cell Sci 107: 3055-3064.

Hindie M, Legeay G, Vayssade M, Warocquier-Clerout R & Nagel MD (2005) Culture of melanoma cells as aggregates on cellulose substratum. Biomol Eng 22: 205-208.

Hindie M, Vayssade M, Dufresne M, Queant S, Warocquier-Clerout R, Legeay G, Vigneron P, Olivier V, Duval JL & Nagel MD (2006) Interactions of B16F10 melanoma cells aggregated on a cellulose substrate. J Cell Biochem.

Hirakawa M, Oike M, Watanabe M, Karashima Y & Ito Y (2006) Pivotal role of integrin alpha5beta1 in hypotonic stress-induced responses of human endothelium. FASEB J 20: 1992-1999.

Hodgson DE. (1988). Using Shape Memory Alloys. Shape Memory Applications, Inc., CA. Engineering Kit Instruction Manual.

Hodgson DE, Wu MH, Biermann RJ. (1990). Shape Memory Alloys, Metals Handbook. Vol. 2. Ohio, USA: ASM International.

Horbett TA (1993) Principles Underlying the Role of Adsorbed Plasma-Proteins in Blood Interactions with Foreign Materials. Cardiovasc Pathol 2: S137-S148.

Horbett TA (1981) Adsorption of proteins from plasma to a series of hydrophilic-hydrophobic copolymers. II. Compositional analysis with the prelabeled protein technique. J Biomed Mater Res 15: 673-695.

140

Horne WC, Neff L, Chatterjee D, Lomri A, Levy JB & Baron R (1992) Osteoclasts express high levels of pp60c-src in association with intracellular membranes. J Cell Biol 119: 1003-1013.

Hoshaw SJ, Brunski JB & Cochran GVB (1994) Mechanical loading of Brånemark implants affects interfacial bone modelling and remodelling. Int J Oral Maxillofac Implants 9: 345-360.

Hu WJ, Eaton JW, Ugarova TP & Tang L (2001) Molecular basis of biomaterial-mediated foreign body reactions. Blood 98: 1231-1238.

Huang HH, Chiu YH, Lee TH, Wu SC, Yang HW, Su KH & Hsu CC (2003) Ion release from NiTi orthodontic wires in artificial saliva with various acidities. Biomaterials 24: 3585-3592.

Huang X, Cho S & Spangrude GJ (2007) Hematopoietic stem cells: generation and self-renewal. Cell Death Differ 14: 1851-1859.

Huber RI, Keller HW, Huber PM & Rehm KE (1996) Flexible intramedullary nailing as fracture treatment in children. J Pediatr Orthop 16: 602-605.

Huhtala P, Humphries MJ, McCarthy JB, Tremble PM, Werb Z & Damsky CH (1995) Cooperative signaling by alpha 5 beta 1 and alpha 4 beta 1 integrins regulates metalloproteinase gene expression in fibroblasts adhering to fibronectin. J Cell Biol 129: 867-879.

Huiskes R (1993) Failed innovation in total hip replacement. Diagnosis and proposals for a cure. Acta Orthop Scand 64: 699-716.

Huiskes R, Weinans H & Dalstra M (1989) Adaptive bone remodeling and biomechanical design considerations for noncemented total hip arthroplasty. Orthopedics 12: 1255-1267.

Hynes RO (1987) Integrins: a family of cell surface receptors. Cell 48: 549-554. Hynes RO. (1990). Fibronectins. New York: Springer-Verlag. Ilic D, Almeida EA, Schlaepfer DD, Dazin P, Aizawa S & Damsky CH (1998)

Extracellular matrix survival signals transduced by focal adhesion kinase suppress p53-mediated apoptosis. J Cell Biol 143: 547-560.

Ingber DE (1997) Integrins, tensegrity, and mechanotransduction. Gravit Space Biol Bull 10: 49-55.

Izzard CS & Lochner LR (1976) Cell-to-substrate contacts in living fibroblasts: an interference reflexion study with an evaluation of the technique. J Cell Sci 21: 129-159.

Jacobs JJ, Skipor AK, Black J, Urban R & Galante JO (1991) Release and excretion of metal in patients who have a total hip-replacement component made of titanium-base alloy. J Bone Joint Surg Am 73: 1475-1486.

Janzing HM, Stockman B, Van Damme G, Rommens P & Broos PL (1998) The retrograde intramedullary supracondylar nail: an alternative in the treatment of distal femoral fractures in the elderly? Arch Orthop Trauma Surg 118: 92-95.

Jasty MJ, Floyd WE, III, Schiller AL, Goldring SR & Harris WH (1986) Localized osteolysis in stable, non-septic total hip replacement. J Bone Joint Surg Am 68: 912-919.

141

Jemt T & Lekholm U (1998) Measurements of bone and frame-work deformations induced by misfit of implant superstructures. A pilot study in rabbits. Clin Oral Implants Res 9: 272-280.

Jiang LS, Shen L & Dai LY (2007) Intramedullary fixation of subtrochanteric fractures with long proximal femoral nail or long gamma nail: technical notes and preliminary results. Ann Acad Med Singapore 36: 821-826.

Jikko A, Harris SE, Chen D, Mendrick DL & Damsky CH (1999) Collagen integrin receptors regulate early osteoblast differentiation induced by BMP-2. J Bone Miner Res 14: 1075-1083.

Jimbo R, Sawase T, Shibata Y, Hirata K, Hishikawa Y, Tanaka Y, Bessho K, Ikeda T & Atsuta M (2007) Enhanced osseointegration by the chemotactic activity of plasma fibronectin for cellular fibronectin positive cells. Biomaterials 28: 3469-3477.

Jiranek W, Jasty M, Wang JT, Bragdon C, Wolfe H, Goldberg M & Harris W (1995) Tissue-Response to Particulate Polymethylmethacrylate in Mice with Various Immune Deficiencies. J Bone Joint Surg Am 77A: 1650-1661.

Jiranek WA, Machado M, Jasty M, Jevsevar D, Wolfe HJ, Goldring SR, Goldberg MJ & Harris WH (1993) Production of cytokines around loosened cemented acetabular components. Analysis with immunohistochemical techniques and in situ hybridization. J Bone Joint Surg Am 75: 863-879.

Johnson F (1920) Some features in the behavior of beta-brass when cold-rolled. J Inst Metals 24: 301-310.

Jokinen M, Pätsi M, Rahiala H, Peltola T, Ritala M & Rosenholm JB (1998) Influence of sol and surface properties on in vitro bioactivity of sol-gel-derived TiO2 and TiO2-SiO2 films deposited by dip-coating method. J Biomed Mater Res 42: 295-302.

Jones SJ & Boyde A (1976) Experimental study of changes in osteoblastic shape induced by calcitonin and parathyroid extract in an organ culture system. Cell Tissue Res 169: 499-65.

Kajiya H, Okabe K, Okamoto F, Tsuzuki T & Soeda H (2000) Protein tyrosine kinase inhibitors increase cytosolic calcium and inhibit actin organization as resorbing activity in rat osteoclasts. J Cell Physiol 183: 83-90.

Kanellopoulos AD, Yiannakopoulos CK & Soucacos PN (2006) Closed, locked intramedullary nailing of pediatric femoral shaft fractures through the tip of the greater trochanter. J Trauma 60: 217-222.

Kapanen A, Danilov A, Lehenkari P, Ryhänen J, Jämsä T & Tuukkanen J (2002a) Effect of metal alloy surface stresses on the viability of ROS-17/2.8 osteoblastic cells. Biomaterials 23: 3733-3740.

Kapanen A, Ilvesaro J, Danilov A, Ryhänen J, Lehenkari P & Tuukkanen J (2002b) Behaviour of nitinol in osteoblast-like ROS-17 cell cultures. Biomaterials 23: 645-650.

Kapanen A, Ryhänen J, Danilov A & Tuukkanen J (2001) Effect of nickel-titanium shape memory metal alloy on bone formation. Biomaterials 22: 2475-2480.

142

Kaplan G & Gaudernack G (1982) In vitro differentiation of human monocytes. Differences in monocyte phenotypes induced by cultivation on glass or on collagen. J Exp Med 156: 1101-1114.

Karaczyn AA, Golebiowski F & Kasprzak KS (2006) Ni(II) affects ubiquitination of core histones H2B and H2A. Exp Cell Res 312: 3252-3259.

Kasemo B & Lausmaa J (1988) Biomaterial and implant surfaces: on the role of cleanliness, contamination, and preparation procedures. J Biomed Mater Res 22: 145-158.

Kasemo B & Lausmaa J (1994) Material-tissue interfaces: the role of surface properties and processes. Environ Health Perspect 102 Suppl 5: 41-45.

Katz BZ, Zamir E, Bershadsky A, Kam Z, Yamada KM & Geiger B (2000) Physical state of the extracellular matrix regulates the structure and molecular composition of cell-matrix adhesions. Mol Biol Cell 11: 1047-1060.

Kawai H, Shibata Y & Miyazaki T (2004) Glow discharge plasma pretreatment enhances osteoclast differentiation and survival on titanium plates. Biomaterials 25: 1805-1811.

Kempf I, Grosse A & Beck G (1985) Closed locked intramedullary nailing. Its application to comminuted fractures of the femur. J Bone Joint Surg Am 67: 709-720.

Keselowsky BG, Bridges AW, Burns KL, Tate CC, Babensee JE, LaPlaca MC & Garcia AJ (2007a) Role of plasma fibronectin in the foreign body response to biomaterials. Biomaterials 28: 3626-3631.

Keselowsky BG, Collard DM & Garcia AJ (2003a) Surface chemistry modulates fibronectin conformation and directs integrin binding and specificity to control cell adhesion. J Biomed Mater Res A 66: 247-259.

Keselowsky BG, Collard DM & Garcia AJ (2004) Surface chemistry modulates focal adhesion composition and signaling through changes in integrin binding. Biomaterials 25: 5947-5954.

Keselowsky BG, Collard DM & Garcia AJ (2003b) Surface chemistry modulates fibronectin conformation and directs integrin binding and specificity to control cell adhesion. J Biomed Mater Res A 66: 247-259.

Keselowsky BG, Collard DM & Garcia AJ (2005) Integrin binding specificity regulates biomaterial surface chemistry effects on cell differentiation. Proc Natl Acad Sci U S A 102: 5953-5957.

Keselowsky BG, Wang L, Schwartz Z, Garcia AJ & Boyan BD (2007b) Integrin alpha(5) controls osteoblastic proliferation and differentiation responses to titanium substrates presenting different roughness characteristics in a roughness independent manner. J Biomed Mater Res A 80: 700-710.

Klebe RJ, Bentley KL & Schoen RC (1981) Adhesive substrates for fibronectin. J Cell Physiol 109: 481-488.

Knight CG, Morton LF, Peachey AR, Tuckwell DS, Farndale RW & Barnes MJ (2000) The collagen-binding A-domains of integrins alpha(1)beta(1) and alpha(2)beta(1) recognize the same specific amino acid sequence, GFOGER, in native (triple-helical) collagens. J Biol Chem 275: 35-40.

143

Kobayashi S, Ohgoe Y, Ozeki K, Hirakuri K & Aoki H (2007) Dissolution effect and cytotoxicity of diamond-like carbon coatings on orthodontic archwires. J Mater Sci Mater Med 18: 2263-2268.

Kofstad P. (1966). High-temperature oxidation in metals. New York: J. Wiley & Sons. Komori T (2008) Regulation of bone development and maintenance by Runx2. Front

Biosci 13: 898-903. Kong X, Grabitz RG, van Oeveren W, Klee D, van Kooten TG, Freudenthal F, Qing M,

von Bernuth G & Seghaye MC (2002) Effect of biologically active coating on biocompatibility of Nitinol devices designed for the closure of intra-atrial communications. Biomaterials 23: 1775-1783.

Konttinen YT, Zhao D, Beklen A, Ma G, Takagi M, Kivela-Rajamaki M, Ashammakhi N & Santavirta S (2005) The microenvironment around total hip replacement prostheses. Clin Orthop Relat Res: 28-38.

Korzeniewski C & Callewaert DM (1983) An enzyme-release assay for natural cytotoxicity. J Immunol Methods 64: 313-320.

Krause A, Cowles EA & Gronowicz G (2000) Integrin-mediated signaling in osteoblasts on titanium implant materials. J Biomed Mater Res 52: 738-747.

KSV Instruments (2008). Surface Free Energy (solid), www.ksvltd.com/content/index/ keysfe. webpage 22-2-2008.

Ku Y, Chung CP & Jang JH (2005) The effect of the surface modification of titanium using a recombinant fragment of fibronectin and vitronectin on cell behavior. Biomaterials 26: 5153-5157.

Kujala S, Ryhänen J, Danilov A & Tuukkanen J (2003) Effect of porosity on the osteointegration and bone ingrowth of a weight-bearing nickel-titanium bone graft substitute. Biomaterials 24: 4691-4697.

Kujala S, Ryhänen J, Jämsä T, Danilov A, Saaranen J, Pramila A & Tuukkanen J (2002a) Bone modeling controlled by a nickel-titanium shape memory alloy intramedullary nail. Biomaterials 23: 2535-2543.

Kujala S, Tuukkanen J, Jämsä T, Danilov A, Pramila A & Ryhänen J (2002b) Comparison of the bone modeling effects caused by curved and straight nickel-titanium intramedullary nails. J Mater Sci Mater Med 13: 1157-1161.

Küntscher G (1940) Die Marknagelung von Knockenbrucken. Arch Klin Chir 200: 443-455.

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.

Lakkakorpi P, Tuukkanen J, Hentunen T, Järvelin K & Väänänen K (1989) Organization of osteoclast microfilaments during the attachment to bone surface in vitro. J Bone Miner Res 4: 817-25.

Lane JM (1979) Biochemistry of fracture repair. AAOS Monterey Seminar Am Acad, Orthop Surg, Chicago: 141-165.

144

Lane NE, Yao W, Balooch M, Nalla RK, Balooch G, Habelitz S, Kinney JH & Bonewald LF (2006) Glucocorticoid-treated mice have localized changes in trabecular bone material properties and osteocyte lacunar size that are not observed in placebo-treated or estrogen-deficient mice. J Bone Miner Res 21: 466-476.

Lanir N, Ciano PS, van de Water L, McDonagh J, Dvorak AM & Dvorak HF (1988) Macrophage migration in fibrin gel matrices. II. Effects of clotting factor XIII, fibronectin, and glycosaminoglycan content on cell migration. J Immunol 140: 2340-2349.

Lee YS, Lo TY & Huang HL (2008) Intramedullary fixation of tibial shaft fractures: a comparison of the unlocked and interlocked nail. Int Orthop 32: 69-74.

Li CY, Yang XJ, Zhang LY, Chen MF & Cui ZD (2007) In vivo histological evaluation of bioactive NiTi alloy after two years implantation. Materials Science & Engineering C-Biomimetic and Supramolecular Systems 27: 122-126.

Lian JB, Stein GS & Aubin JE (2003) Bone formation: maturation and functional activities of osteoblast lineage cells. In Primer on the Metabolic Bone Disease and Disorders of Mineral Metabolism, 5 ed. American Society for Bone and Mineral Research,p 13-28.

Lim JY, Dreiss AD, Zhou Z, Hansen JC, Siedlecki CA, Hengstebeck RW, Cheng J, Winograd N & Donahue HJ (2007) The regulation of integrin-mediated osteoblast focal adhesion and focal adhesion kinase expression by nanoscale topography. Biomaterials 28: 1787-1797.

Lim JY, Liu X, Vogler EA & Donahue HJ (2004) Systematic variation in osteoblast adhesion and phenotype with substratum surface characteristics. J Biomed Mater Res A 68: 504-512.

Ling Y, Rios HF, Myers ER, Lu Y, Feng JQ & Boskey AL (2005) DMP1 depletion decreases bone mineralization in vivo: an FTIR imaging analysis. J Bone Miner Res 20: 2169-2177.

Liu J-X, Yang D-Z, Shi F & Cai Y-J (2003) Sol-gel deposited TiO2 film on NiTi surgical alloy for biocompatibility improvement. Thin Solid Films 429: 225-230.

Lukert BP (2003) Glucocorticoid-induced osteoporosis. In Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 5 ed. American Society for Bone and Mineral Research,p 364-370.

Luthen F, Lange R, Becker P, Rychly J, Beck U & Nebe JGB (2005) The influence of surface roughness of titanium on beta 1-and beta 3-integrin adhesion and the organization of fibronectin in human osteoblastic cells. Biomaterials 26: 2423-2440.

Lynch MP, Stein JL, Stein GS & Lian JB (1995) The influence of type I collagen on the development and maintenance of the osteoblast phenotype in primary and passaged rat calvarial osteoblasts: modification of expression of genes supporting cell growth, adhesion, and extracellular matrix mineralization. Exp Cell Res 216: 35-45.

MacDonald DE, Deo N, Markovic B, Stranick M & Somasundaran P (2002) Adsorption and dissolution behavior of human plasma fibronectin on thermally and chemically modified titanium dioxide particles. Biomaterials 23: 1269-1279.

Machado LG & Savi MA (2003) Medical applications of shape memory alloys. Braz J Med Biol Res 36: 683-691.

145

Majeska RJ, Rodan SB & Rodan GA (1980) Parathyroid hormone-responsive clonal cell lines from rat osteosarcoma. Endocrinology 107: 1494-1503.

Malone AM, Anderson CT, Tummala P, Kwon RY, Johnston TR, Stearns T & Jacobs CR (2007) Primary cilia mediate mechanosensing in bone cells by a calcium-independent mechanism. Proc Natl Acad Sci U S A 104: 13325-13330.

Mantovani D (2000) Shape memory alloys: properties and biomedical applications. JOM - J Miner Metals Mater Soc 52: 36-44.

Mao Y & Schwarzbauer JE (2005) Fibronectin fibrillogenesis, a cell-mediated matrix assembly process. Matrix Biol 24: 389-399.

Martin TJ & Sims NA (2005) Osteoclast-derived activity in the coupling of bone formation to resorption. Trends Mol Med 11: 76-81.

Matsuo K & Irie N (2008) Osteoclast-osteoblast communication. Arch Biochem Biophys 473: 201-209.

McDonald F, Yettram AL & MacLeod K (1994) The response of bone to external loading regimens. Med Eng Phys 16: 384-397.

McGuigan AP, Briggs GAD, Burlakov V, Yanaka M & Tsukahara Y (2003) An elastic-plastic shear lag model for fracture of layered coatings. Thin Solid Films 424: 219-223.

McKellop HA (1995) Wear modes, mechanisms, damage, and debris: separating cause from effect in the wear of total hip replacements. In Galante JO, Rosenberg AG & Callaghan JJ (eds) Total hip revision surgery. Raven Press, New York, p 21-39.

McNally AK & Anderson JM (2002) Beta1 and beta2 integrins mediate adhesion during macrophage fusion and multinucleated foreign body giant cell formation. Am J Pathol 160: 621-630.

McNally AK & Anderson JM (1995) Interleukin-4 induces foreign body giant cells from human monocytes/macrophages. Differential lymphokine regulation of macrophage fusion leads to morphological variants of multinucleated giant cells. Am J Pathol 147: 1487-1499.

McNally AK & Anderson JM (1994) Complement C3 participation in monocyte adhesion to different surfaces. Proc Natl Acad Sci U S A 91: 10119-10123.

McNamara LM & Prendergast PJ (2007) Bone remodelling algorithms incorporating both strain and microdamage stimuli. J Biomech 40: 1381-1391.

Mehlman CT, Nemeth NM & Glos DL (2006) Antegrade versus retrograde titanium elastic nail fixation of pediatric distal-third femoral-shaft fractures: a mechanical study. J Orthop Trauma 20: 608-612.

Mehmood S, Jinnah RH & Pandit H (2008) Review on ceramic-on-ceramic total hip arthroplasty. J Surg Orthop Adv 17: 45-50.

Menko AS & Boettiger D (1987) Occupation of the extracellular matrix receptor, integrin, is a control point for myogenic differentiation. Cell 51: 51-57.

Mercurius KO & Morla AO (1998) Inhibition of vascular smooth muscle cell growth by inhibition of fibronectin matrix assembly. Circ Res 82: 548-556.

146

Michaels GC, Carr AB & Larsen PE (1997) Effect of prosthetic superstructure accuracy on the osteointegrated implant bone interface. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 83: 198-205.

Michiardi A, Aparicio C, Planell JA & Gil FJ (2006) New oxidation treatment of NiTi shape memory alloys to obtain Ni-free surfaces and to improve biocompatibility. J Biomed Mater Res B 77: 249-256.

Michiardi A, Aparicio C, Planell JA & Gil FJ (2007a) Electrochemical behaviour of oxidized NiTi shape memory alloys for biomedical applications. Surf Coat Tech 201: 6484-6488.

Michiardi A, Aparicio C, Ratner BD, Planell JA & Gil J (2007b) The influence of surface energy on competitive protein adsorption on oxidized NiTi surfaces. Biomaterials 28: 586-594.

Miller SD, Burkart B, Damson E, Shrive N & Bray RC (1993) The effect of the entry hole for an intramedullary nail on the strength of the proximal femur. J Bone Joint Surg Br 75: 202-206.

Minkin C (1982) Bone acid phosphatase: tartrate-resistant acid phosphatase as a marker of osteoclast function. Calcif Tissue Int 34: 285-290.

Miranti CK, Ohno S & Brugge JS (1999) Protein kinase C regulates integrin-induced activation of the extracellular regulated kinase pathway upstream of Shc. J Biol Chem 274: 10571-10581.

Mizuno M, Fujisawa R & Kuboki Y (2000) Type I collagen-induced osteoblastic differentiation of bone-marrow cells mediated by collagen-alpha2beta1 integrin interaction. J Cell Physiol 184: 207-213.

Mizuno M & Kuboki Y (2001) Osteoblast-related gene expression of bone marrow cells during the osteoblastic differentiation induced by type I collagen. J Biochem (Tokyo) 129: 133-138.

Moein CM, Verhofstad MH, Bleys RL & van der Werken C (2005) Soft tissue injury related to choice of entry point in antegrade femoral nailing: piriform fossa or greater trochanter tip. Injury 36: 1337-1342.

Moonga BS, Li S, Iqbal J, Davidson R, Shankar VS, Bevis PJ, Inzerillo A, Abe E, Huang CL & Zaidi M (2002) Ca(2+) influx through the osteoclastic plasma membrane ryanodine receptor. Am J Physiol Renal Physiol 282: F921-F932.

Moritz N, Jokinen M, Peltola T, Areva S & Yli-Urpo A (2003) Local induction of calcium phosphate formation on TiO2 coatings on titanium via surface treatment with a CO2 laser. J Biomed Mater Res A 65: 9-16.

Moursi AM, Damsky CH, Lull J, Zimmerman D, Doty SB, Aota S, Globus & RK. (1996) Fibronectin regulates calvarial osteoblast differentiation. J Cell Sci 109: 1369-1380.

Moursi AM, Globus RK & Damsky CH (1997) Interactions between integrin receptors and fibronectin are required for calvarial osteoblast differentiation in vitro. J Cell Sci 110: 2187-2196.

Mueller LA, Voelk M, Kress A, Pitto RP & Schmidt R (2007) An ABJS Best Paper: Progressive cancellous and cortical bone remodeling after press-fit cup fixation: a 3-year followup. Clin Orthop Relat Res 463: 213-220.

147

Musialek J, Filip P & Nieslanik J (1998) Titanium-nickel shape memory clamps in small bone surgery. Arch Orthop Trauma Surg 117: 341-344.

Mykhaylyk TA, Evans SD, Fernyhough CM, Hamley IW & Henderson JR (2003) Surface energy of ethylene-co-1-butene copolymers determined by contact angle methods. J Colloid Interf Sci 260: 234-239.

Natali AN, Pavan PG & Ruggero AL (2006) Analysis of bone-implant interaction phenomena by using a numerical approach. Clin Oral Implants Res 17: 67-74.

Nelson CL, McLaren AC, McLaren SG, Johnson JW & Smeltzer MS (2005) Is aseptic loosening truly aseptic? Clin Orthop Relat Res: 25-30.

Neumann AW & Godd RJ (1979) Techniques of measuring contact angles. In Good RJ& Stromberg RR (eds) Surface and colloid sciene - Experimental methods, vol 11. Plenum Publishing, New York, p 31-61.

Nevalainen J, Hirvonen J, Pulkkinen P. (1998). The 1997 implant yearbook on orthopeadic endoprosthesis. Finnish Arthoplasty Register. Vol. 6. National Agency for Medicines.

Newman SL & Tucci MA (1990) Regulation of human monocyte/macrophage function by extracellular matrix. Adherence of monocytes to collagen matrices enhances phagocytosis of opsonized bacteria by activation of complement receptors and enhancement of Fc receptor function. J Clin Invest 86: 703-714.

Newson T (2002) Stainless Steel - A Family of Medical Device Materials. In Business Briefing: Medical Device Manufacturing & Technology. World Martkets Research Centre, Business Briefings Ltd.p 1-3.

Nijweide PJ, Burger EH & Klein-Nulend J (2002) The Osteocytes. In Bilezikian JP, Raisz LG & Rodan GA (eds) Principles of Bone Biology, 2 ed. Academic Press, San Diego, CA, USA, p 93-107.

Olson GB & Hartman H (1982) Martensite and life: displacive transformations as biological processes. J Physique 43: C4-855-865.

Ostrum RF, Marcantonio A & Marburger R (2005) A critical analysis of the eccentric starting point for trochanteric intramedullary femoral nailing. J Orthop Trauma 19: 681-686.

Ozeki K, Yuhta T, Aoki H, Asaoka T, Daisaku T & Fukui Y (2003) Deterioration in the superelasticity of Ti sputter coated on NiTi orthodontic wire. Biomed Mater Eng 13: 355-362.

Palumbo C, Palazzini S, Zaffe D & Marotti G (1990) Osteocyte differentiation in the tibia of newborn rabbit: an ultrastructural study of the formation of cytoplasmic processes. Acta Anat (Basel) 137: 350-358.

Pankov R, Cukierman E, Katz BZ, Matsumoto K, Lin DC, Lin S, Hahn C & Yamada KM (2000) Integrin dynamics and matrix assembly: tensin-dependent translocation of alpha(5)beta(1) integrins promotes early fibronectin fibrillogenesis. J Cell Biol 148: 1075-1090.

Papadakis SA, Shepherd L, Babourda EC & Papadakis S (2005) Piriform and trochanteric fossae. A drawing mismatch or a terminology error? A review. Surg Radiol Anat 27: 223-226.

148

Pelham RJJr & Wang Y (1997) Cell locomotion and focal adhesions are regulated by substrate flexibility. Proc Natl Acad Sci U S A 94: 13661-13665.

Perez A, Mahar A, Negus C, Newton P & Impelluso T (2007) A computational evaluation of the effect of intramedullary nail material properties on the stabilization of simulated femoral shaft fractures. Med Eng Phys [Epub ahead of print].

Petrini L, Migliavacca F, Massarotti P, Schievano S, Dubini G & Auricchio F (2005) Computational studies of shape memory alloy behavior in biomedical applications. J Biomech Eng 127: 716-725.

Plant SD, Grant DM & Leach L (2005) Behaviour of human endothelial cells on surface modified NiTi alloy. Biomaterials 26: 5359-5367.

Poon RW, Yeung KW, Liu XY, Chu PK, Chung CY, Lu WW, Cheung KM & Chan D (2005) Carbon plasma immersion ion implantation of nickel-titanium shape memory alloys. Biomaterials 26: 2265-2272.

Prymak O, Bogdanski D, Koller M, Esenwein SA, Muhr G, Beckmann F, Donath T, Assad M & Epple M (2005) Morphological characterization and in vitro biocompatibility of a porous nickel-titanium alloy. Biomaterials 26: 5801-5807.

Puri MC & Bernstein A (2003) Requirement for the TIE family of receptor tyrosine kinases in adult but not fetal hematopoiesis. Proc Natl Acad Sci U S A 100: 12753-12758.

Raab-Cullen DM, Akhter MP, Kimmel DB & Recker RR (1994) Periosteal bone formation stimulated by externally induced bending strains. J Bone Miner Res 9: 1143-1152.

Ratner BD (1995) Surface modification of polymers - chemical, biological and surface analytical challenges. Biosens Bioelectron 10: 797-804.

Ratner BD & Bryant SJ (2004) Biomaterials: where we have been and where we are going. Annu Rev Biomed Eng 6: 41-75.

Ratner BD et al. (1996). Biomaterial science: an introduction to materials in medicine. San Diego, CA: Academic Press.

Redey SA, Razzouk S, Rey C, Bernache-Assollant D, Leroy G, Nardin M & Cournot G (1999) Osteoclast adhesion and activity on synthetic hydroxyapatite, carbonated hydroxyapatite, and natural calcium carbonate: relationship to surface energies. J Biomed Mater Res 45: 140-147.

Reid IR (2003) Menopause. In Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 5 ed. American Society for Bone and Mineral Research,p 86-89.

Reyes CD, Petrie TA, Burns KL, Schwartz Z & Garcia AJ (2007) Biomolecular surface coating to enhance orthopaedic tissue healing and integration. Biomaterials 28: 3228-3235.

Rho JY, Roy ME, Tsui TY & Pharr GM (1999) Elastic properties of microstructural components of human bone tissue as measured by nanoindentation. J Biomed Mater Res 45: 48-54.

Rho JY, Tsui TY & Pharr GM (1997) Elastic properties of human cortical and trabecular lamellar bone measured by nanoindentation. Biomaterials 18: 1325-1330.

149

Ricci WM, Devinney S, Haidukewych G, Herscovici D & Sanders R (2005) Trochanteric nail insertion for the treatment of femoral shaft fractures. J Orthop Trauma 19: 511-517.

Robling AG, Castillo AB & Turner CH (2006) Biomechanical and molecular regulation of bone remodeling. Annu Rev Biomed Eng 8: 455-498.

Rodan GA & Martin TJ (1981) Role of osteoblasts in hormonal control of bone resorption - a hypothesis. Calcif Tissue Int 33: 349-351.

Rondelli G (1996) Corrosion resistance tests on NiTi shape memory alloy. Biomaterials 17: 2003-2008.

Rondelli G, Torricelli P, Fini M, Rimondini L & Giardino R (2006) In vitro corrosion study by EIS of an equiatomic NiTi alloy and an implant quality AISI 316 stainless steel. J Biomed Mater Res B Appl Biomater 79: 320-324.

Rosen CJ & Douglas PK (2003) Age-related osteoporosis. In Primer on the Metabolic Bone Disease and Disorders of Mineral Metabolism, 5 ed. American Society for Bone and Mineral Research,p 89-92.

Rubin CT (1984) Skeletal strain and the functional significance of bone architecture. Calcif Tissue Int 36 Suppl 1: S11-S18.

Ruoslahti E & Giancotti FG (1989) Integrins and tumor cell dissemination. Cancer Cells 1: 119-126.

Ryan G, Pandit A & Apatsidis DP (2006) Fabrication methods of porous metals for use in orthopaedic applications. Biomaterials 27: 2651-2670.

Ryd L, Albrektsson BE, Carlsson L, Dansgard F, Herberts P, Lindstrand A, Regner L & Toksvig-Larsen S (1995) Roentgen stereophotogrammetric analysis as a predictor of mechanical loosening of knee prostheses. J Bone Joint Surg Br 77: 377-383.

Ryhänen J (1999) Biocompatibility evaluation of nickel-titanium shape memory metal alloy, Acta Univ Oul D 518.

Ryhänen J, Kallioinen M, Tuukkanen J, Junila J, Niemelä E, Sandvik P & Serlo W (1998) In vivo biocompatibility evaluation of nickel-titanium shape memory metal alloy: muscle and perineural tissue responses and encapsule membrane thickness. J Biomed Mater Res 41: 481-488.

Ryhänen J, Kallioinen M, Tuukkanen J, Lehenkari P, Junila J, Niemelä E, Sandvik P & Serlo W (1999) Bone modeling and cell-material interface responses induced by nickel- titanium shape memory alloy after periosteal implantation. Biomaterials 20: 1309-1317.

Ryhänen J, Niemi E, Serlo W, Niemelä E, Sandvik P, Pernu H & Salo T (1997) Biocompatibility of nickel-titanium shape memory metal and its corrosion behavior in human cell cultures. J Biomed Mater Res 35: 451-457.

Salnikow K & Zhitkovich A (2008) Genetic and epigenetic mechanisms in metal carcinogenesis and cocarcinogenesis: nickel, arsenic, and chromium. Chem Res Toxicol 21: 28-44.

Saltel F, Destaing O, Bard F, Eichert D & Jurdic P (2004) Apatite-mediated actin dynamics in resorbing osteoclasts. Mol Biol Cell 15: 5231-5241.

150

Santavirta S, Hoikka V, Eskola A, Konttinen YT, Paavilainen T & Tallroth K (1990a) Aggressive granulomatous lesions in cementless total hip arthroplasty. J Bone Joint Surg Br 72: 980-984.

Santavirta S, Konttinen YT, Bergroth V, Eskola A, Tallroth K & Lindholm TS (1990b) Aggressive granulomatous lesions associated with hip arthroplasty. Immunopathological studies. J Bone Joint Surg Am 72: 252-258.

Sastry SK & Burridge K (2000) Focal adhesions: a nexus for intracellular signaling and cytoskeletal dynamics. Exp Cell Res 261: 25-36.

Schiller PC, D'Ippolito G, Brambilla R, Roos BA & Howard GA (2001) Inhibition of gap-junctional communication induces the trans-differentiation of osteoblasts to an adipocytic phenotype in vitro. J Biol Chem 276: 14133-14138.

Schmalzried TP, Clarke IC & McKellop HA (1998) Bearing surfaces. In Callaghan JJ, Rosenberg AG & Rubash HE (eds) The Adult Hip. p 247-264.

Schmalzried TP, Jasty M & Harris WH (1992) Periprosthetic bone loss in total hip arthroplasty. Polyethylene wear debris and the concept of the effective joint space. J Bone Joint Surg Am 74: 849-863.

Schneider GB, Zaharias R & Stanford C (2001) Osteoblast integrin adhesion and signaling regulate mineralization. J Dent Res 80: 1540-1544.

Sechler JL & Schwarzbauer JE (1998) Control of cell cycle progression by fibronectin matrix architecture. J Biol Chem 273: 25533-25536.

Seitz TL, Noonan KD, Hench LL & Noonan NE (1982) Effect of fibronectin on the adhesion of an established cell line to a surface reactive biomaterial. J Biomed Mater Res 16: 195-207.

Shabalovskaya S, Anderegg J & Van HJ (2008) Critical overview of Nitinol surfaces and their modifications for medical applications. Acta Biomater 4: 447-467.

Shabalovskaya SA (1996) On the nature of the biocompatibility and on medical applications of NiTi shape memory and superelastic alloys. Biomed Mater Eng 6: 267-289.

Shankar VS, Bax CM, Bax BE, Alam AS, Moonga BS, Simon B, Pazianas M, Huang CL & Zaidi M (1993) Activation of the Ca2+ "receptor" on the osteoclast by Ni2+ elicits cytosolic Ca2+ signals: evidence for receptor activation and inactivation, intracellular Ca2+ redistribution, and divalent cation modulation. J Cell Physiol 155: 120-129.

Sharma PK & Rao KH (2002) Analysis of different approaches for evaluation of surface energy of microbial cells by contact angle goniometry. Adv Colloid Interface Sci 98: 341-463.

Shimizu H, Sakamoto M & Sakamoto S (1990) Bone resorption by isolated osteoclasts in living versus devitalized bone: differences in mode and extent and the effects of human recombinant tissue inhibitor of metalloproteinases. J Bone Miner Res 5: 411-418.

Shrivastava A, Radziejewski C, Campbell E, Kovac L, McGlynn M, Ryan TE, Davis S, Goldfarb MP, Glass DJ, Lemke G & Yancopoulos GD (1997) An orphan receptor tyrosine kinase family whose members serve as nonintegrin collagen receptors. Mol Cell 1: 25-34.

151

Sigel, Sigel. (1988). Nickel and Its Role in Biology. Vol. 23. New York: Mercel Deccer, Inc.

Smetana KJr, Sulc J & Krcova Z (1987) Physicochemical aspects of the giant multinucleate cell formation. Exp Mol Pathol 47: 271-278.

Smilenov LB, Mikhailov A, Pelham RJ, Marcantonio EE & Gundersen GG (1999) Focal adhesion motility revealed in stationary fibroblasts. Science 286: 1172-1174.

Sommerfeldt DW, McLeod KJ, Rubin CT & Hadjiargyrou M (2001) Differential phosphorylation of paxillin in response to surface-bound serum proteins during early osteoblast adhesion. Biochem Biophys Res Commun 285: 355-363.

Sottile J, Hocking DC & Swiatek PJ (1998) Fibronectin matrix assembly enhances adhesion-dependent cell growth. J Cell Sci 111: 2933-2943.

Speck KM & Fraker AC (1980) Anodic polarization behavior of Ti-Ni and Ti-6A1-4V in simulated physiological solutions. J Dent Res 59: 1590-1595.

Stenbeck G (2002) Formation and function of the ruffled border in osteoclasts. Semin Cell Dev Biol 13: 285-292.

Stoeckel D, Pelton A & Duerig T (2004) Self-expanding nitinol stents: material and design considerations. Eur Radiol 14: 292-301.

Streuli CH, Bailey N & Bissell MJ (1991) Control of mammary epithelial differentiation: basement membrane induces tissue-specific gene expression in the absence of cell-cell interaction and morphological polarity. J Cell Biol 115: 1383-1395.

Stulberg BN, Merritt K & Bauer TW (1994) Metallic wear debris in metal-backed patellar failure. J Appl Biomater 5: 9-16.

Sui JH & Cai W (2006) Effect of diamond-like carbon (DLC) on the properties of the NiTi alloys. Diam Relat Mater 15: 1720-1726.

Sumner DR & Galante JO (1992) Determinants of stress shielding: design versus materials versus interface. Clin Orthop Relat Res: 202-212.

Sun ZL, Wataha JC & Hanks CT (1997) Effects of metal ions on osteoblast-like cell metabolism and differentiation. J Biomed Mater Res 34: 29-37.

Suzawa M, Tamura Y, Fukumoto S, Miyazono K, Fujita T, Kato S & Takeuchi Y (2002) Stimulation of Smad1 transcriptional activity by Ras-extracellular signal-regulated kinase pathway: a possible mechanism for collagen-dependent osteoblastic differentiation. J Bone Miner Res 17: 240-248.

Sychterz CJ & Engh CA (1996) The influence of clinical factors on periprosthetic bone remodeling. Clinical Orthopaedics and Related Research: 285-292.

Taal BG, Muller SH, Boot H & Koops W (1997) Potential risks and artifacts of magnetic resonance imaging of self-expandable esophageal stents. Gastrointest Endosc 46: 424-429.

Takagi J (2004) Structural basis for ligand recognition by RGD (Arg-Gly-Asp)-dependent integrins. Biochem Soc Trans 32: 403-406.

Takagi J, Petre BM, Walz T & Springer TA (2002) Global conformational rearrangements in integrin extracellular domains in outside-in and inside-out signaling. Cell 110: 599-11.

152

Takeuchi Y, Suzawa M, Kikuchi T, Nishida E, Fujita T & Matsumoto T (1997) Differentiation and transforming growth factor-beta receptor down-regulation by collagen-alpha2beta1 integrin interaction is mediated by focal adhesion kinase and its downstream signals in murine osteoblastic cells. J Biol Chem 272: 29309-29316.

Tanaka S, Takahashi N, Udagawa N, Sasaki T, Fukui Y, Kurokawa T & Suda T (1992) Osteoclasts express high levels of p60c-src, preferentially on ruffled border membranes. FEBS Lett 313: 85-89.

Tang CH, Yang RS, Huang TH, Liu SH & Fu WM (2004) Enhancement of fibronectin fibrillogenesis and bone formation by basic fibroblast growth factor via protein kinase C-dependent pathway in rat osteoblasts. Mol Pharmacol 66: 440-449.

Tang L & Eaton JW (1993) Fibrin(ogen) mediates acute inflammatory responses to biomaterials. J Exp Med 178: 2147-2156.

Tatsumi S, Ishii K, Amizuka N, Li M, Kobayashi T, Kohno K, Ito M, Takeshita S & Ikeda K (2007) Targeted ablation of osteocytes induces osteoporosis with defective mechanotransduction. Cell Metab 5: 464-475.

Teitelbaum GP, Bradley WG, Jr. & Klein BD (1988a) MR imaging artifacts, ferromagnetism, and magnetic torque of intravascular filters, stents, and coils. Radiology 166: 657-664.

Teitelbaum GP, Ortega HV, Vinitski S, Stern H, Tsuruda JS, Mitchell DG, Rifkin MD & Bradley WG, Jr. (1988b) Low-artifact intravascular devices: MR imaging evaluation. Radiology 168: 713-719.

Teitelbaum SL (2000) Bone resorption by osteoclasts. Science 289: 1504-1508. Theoleyre S, Wittrant Y, Tat SK, Fortun Y, Redini F & Heymann D (2004) The molecular

triad OPG/RANK/RANKL: involvement in the orchestration of pathophysiological bone remodeling. Cytokine Growth F R 15: 457-475.

Thiede MA, Yoon K, Golub EE, Noda M & Rodan GA (1988) Structure and expression of rat osteosarcoma (ROS 17/2.8) alkaline phosphatase: product of a single copy gene. Proc Natl Acad Sci U S A 85: 319-323.

Thierry B, Winnik FM, Merhi Y, Silver J & Tabrizian M (2003) Bioactive coatings of endovascular stents based on polyelectrolyte multilayers. Biomacromolecules 4: 1564-1571.

Thull R (2002) Physicochemical principles of tissue material interactions. Biomol Eng 19: 43-50.

Townsend DR & Hoffinger S (2000) Intramedullary nailing of femoral shaft fractures in children via the trochanter tip. Clin Orthop Relat Res: 113-118.

Toyosawa S, Shintani S, Fujiwara T, Ooshima T, Sato A, Ijuhin N & Komori T (2001) Dentin matrix protein 1 is predominantly expressed in chicken and rat osteocytes but not in osteoblasts. J Bone Miner Res 16: 2017-2026.

Trepanier C, Tabrizian M, Yahia LH, Bilodeau L & Piron DL (1998) Effect of modification of oxide layer on NiTi stent corrosion resistance. J Biomed Mater Res 43: 433-440.

153

Tucker MC, Schwappach JR, Leighton RK, Coupe K & Ricci WM (2007) Results of femoral intramedullary nailing in patients who are obese versus those who are not obese: a prospective multicenter comparison study. J Orthop Trauma 21: 523-529.

Turner CE (2000) Paxillin and focal adhesion signalling. Nat Cell Biol 2: E231-E236. Turner CH, Chandran A & Pidaparti RM (1995) The anisotropy of osteonal bone and its

ultrastructural implications. Bone 17: 85-89. Ulman A. (1991). An introduction to ultrathin organic films: From Langmuir-Blodgett to

self-assembly. Boston, MA: Academic Press. Unger AS, Lewis RJ & Gruen T (2005) Evaluation of a porous tantalum uncemented

acetabular cup in revision total hip arthroplasty: clinical and radiological results of 60 hips. J Arthroplasty 20: 1002-1009.

Väänänen HK & Horton M (1995) The osteoclast clear zone is a specialized cell-extracellular matrix adhesion structure. J Cell Sci 108: 2729-2732.

Välimäki VV & Aro HT (2006) Molecular basis for action of bioactive glasses as bone graft substitute. Scand J Surg 95: 95-102.

Vitte J, Benoliel AM, Pierres A & Bongrand P (2004) Is there a predictable relationship between surface physical-chemical properties and cell behaviour at the interface? Eur Cell Mater 7: 52-63.

Wang GX, Shen Y, Zhang H, Quan XJ & Yu QS (2007) Influence of surface microroughness by plasma deposition and chemical erosion followed by TiO(2) coating upon anticoagulation, hydrophilicity, and corrosion resistance of NiTi alloy stent. J Biomed Mater Res A [Epub ahead of print].

Wang HB, Dembo M & Wang YL (2000) Substrate flexibility regulates growth and apoptosis of normal but not transformed cells. Am J Physiol Cell Physiol 279: C1345-C1350.

Wang XJ, Chen XB, Hodgson PD & Wen CE (2006) Elastic modulus and hardness of cortical and trabecular bovine bone measured by nanoindendation. Trans Nonferrous Met Soc China 16: 744-748.

Wataha JC, Lockwood PE, Marek M & Ghazi M (1999) Ability of Ni-containing biomedical alloys to activate monocytes and endothelial cells in vitro. J Biomed Mater Res 45: 251-257.

Wataha JC, Ratanasathien S, Hanks CT & Sun Z (1996) In vitro IL-1 beta and TNF-alpha release from THP-1 monocytes in response to metal ions. Dent Mater 12: 322-327.

Wayman CM (1975) Proceedings of the international symposium of shape memory effects and applications, Toronto 1975. In Perkins J (ed). Plenum Press, New York.

Webster TJ, Ergun C, Doremus RH, Siegel RW & Bizios R (2000) Specific proteins mediate enhanced osteoblast adhesion on nanophase ceramics. J Biomed Mater Res 51: 475-483.

Weeden SH & Schmidt RH (2007) The use of tantalum porous metal implants for Paprosky 3A and 3B defects. J Arthroplasty 22: 151-155.

Weiss RE & Reddi AH (1981) Role of fibronectin in collagenous matrix-induced mesenchymal cell proliferation and differentiation in vivo. Exp Cell Res 133: 247-254.

154

Wever DJ, Veldhuizen AG, de Vries J, Busscher HJ, Uges DR & van Horn JR (1998) Electrochemical and surface characterization of a nickel-titanium alloy. Biomaterials 19: 761-769.

Wever DJ, Veldhuizen AG, Sanders MM, Schakenraad JM & van Horn JR (1997) Cytotoxic, allergic and genotoxic activity of a nickel-titanium alloy. Biomaterials 18: 1115-1120.

Williams DF. (1987). Definitions in Biomaterials, Progress in Biomedical Engineering. Vol. 4. Amsterdam: Elsevier.

Williams DF (1998) General concepts of biocompatibility. In Black J& Hastings G (eds) Handbook of Biomaterial Properties. Chapman & Hall, London, p 481-488.

Wolff J. (1892). Das gesetz der transformation der knochen. Berlin: Springer-Verlag. Woo KM, Jun JH, Chen VJ, Seo J, Baek JH, Ryoo HM, Kim GS, Somerman MJ & Ma PX

(2007) Nano-fibrous scaffolding promotes osteoblast differentiation and biomineralization. Biomaterials 28: 335-343.

Woodruff MA, Jones P, Farrar D, Grant DM & Scotchford CA (2007) Human osteoblast cell spreading and vinculin expression upon biomaterial surfaces. J Mol Histol 38: 491-499.

Wright TM, Goodman SB. (1996). Implant wear: the future of total joint replacement. Rosemont, Illinois.: American Academy of Orthopaedic Surgeons Publications.

Wu S (1971) Calculation of interfacial tension in a polymer system. J Polym Sci (Pt C) 34: 19-30.

Xiao G, Wang D, Benson MD, Karsenty G & Franceschi RT (1998) Role of the alpha2-integrin in osteoblast-specific gene expression and activation of the Osf2 transcription factor. J Biol Chem 273: 32988-32994.

Yamada KM, Pankov R & Cukierman E (2003) Dimensions and dynamics in integrin function. Braz J Med Biol Res 36: 959-966.

Yeung KWK, Poon RWY, Liu XM, Chu PK, Chung CY, Liu XY, Chan S, Lu WW, Chan D, Luk KDK & Cheung KMC (2007) Nitrogen plasma-implanted nickel titanium alloys for orthopedic use. Surf Coat Tech 201: 5607-5612.

Yin T & Li L (2006) The stem cell niches in bone. J Clin Invest 116: 1195-1201. Zaidi M, Alam AS, Huang CL, Pazianas M, Bax CM, Bax BE, Moonga BS, Bevis PJ &

Shankar VS (1993) Extracellular Ca2+ sensing by the osteoclast. Cell Calcium 14: 271-277.

Zambonin Zallone A, Teti A & Primavera MV (1984) Resorption of vital or devitalized bone by isolated osteoclasts in vitro. The role of lining cells. Cell Tissue Res 235: 561-564.

Zamir E & Geiger B (2001) Molecular complexity and dynamics of cell-matrix adhesions. J Cell Sci 114: 3583-3590.

Zamir E, Katz M, Posen Y, Erez N, Yamada KM, Katz BZ, Lin S, Lin DC, Bershadsky A, Kam Z & Geiger B (2000) Dynamics and segregation of cell-matrix adhesions in cultured fibroblasts. Nature Cell Biology 2: 191-196.

155

Zijderveld JA, de Lange RG & Verbraak CA (1966) La transformation martensitique des alliages titane–nickel au voisinage de la composition e´quiatomique. Mem Sci Rev Met 63: 888.

156

157

Appendix 1

Tabl

e 17

.

Coa

ting

type

S

ubst

rate

sA

dditi

onal

mod

ifica

tions

C

oatin

g m

etho

ds

Aim

s of

Coa

ting

Ref

s.

Sur

face

pro

cess

ing

Por

ous

Co-

Cr

Co-

Cr-

Mo

Tita

nium

Hyd

roxy

apat

ite (H

A)

Pol

yeth

ylen

e pa

rticl

es

Circ

umfe

rent

ial

Non

-circ

umfe

rent

ial

Pla

sma

spra

ying

Pre

dict

ion

by m

echa

no-r

egul

ator

y

algo

rithm

Sin

terin

g

Tita

nium

fibe

r ope

n-po

re c

oatin

g

Yttr

ium

-alu

min

ium

-gar

net l

aser

sca

nnin

g

Impr

oved

fatig

ue

char

acte

ristic

s

Oss

eoco

nduc

tion

Oss

eoin

tegr

atio

n

Pre

vent

ion

of w

ear d

ebris

Pro

tect

ion

from

ost

eoly

sis

1–15

Mic

ro-p

lasm

a te

xtur

ing

Tita

nium

O

xide

laye

r with

HA

A

nodi

c pl

asm

a-ch

emic

al tr

eatm

ent

Enh

ance

d bo

ne a

ppos

ition

16

Aci

d-et

chin

g Ti

tani

um

San

dbla

stin

g w

ith H

A

Sin

tere

d tit

aniu

m b

eads

Aci

d-et

chin

g

Oss

eoin

tegr

atio

n 17

,

18

Grit

-bla

stin

g Ti

tani

um

Ti6A

l4V

Am

orph

ous

/cry

stal

line

CaP

HA

Grit

-bla

stin

g w

ith T

iO2 p

artic

les

Oss

eoin

tegr

atio

n 19

–

23

Mul

tipha

se a

nodi

c sp

ark

depo

sitio

n

Tita

nium

–

Thre

e-st

ep a

nodi

c sp

ark

depo

sitio

n

Oss

eoin

tegr

atio

n 24

–

26

158

Coa

ting

type

S

ubst

rate

s A

dditi

onal

mod

ifica

tions

C

oatin

g m

etho

ds

Aim

s of

Coa

ting

Ref

s.

Hyd

roxy

apat

ite (H

A) a

nd o

ther

calc

ium

pho

spha

tes

(CaP

)

Cel

lulo

se s

pong

e

Cr-

Co-

Mo

HA

-gla

ss-ti

tani

um

Pol

yeth

ylen

e

Por

ous

poly

mer

Pyr

olyt

ic c

arbo

n

Tant

alum

Tita

nium

Ti6A

l4V

Zirc

onia

-alu

min

a

nano

com

posi

te

Am

elog

enin

Bis

phos

phan

ate(

s)

Bon

e al

logr

aft

Cal

cium

tita

nate

Car

bona

te

Chl

orhe

xidi

ne a

djus

tmen

t

Fluo

rinat

ed a

patit

e

Grit

-bla

stin

g

Gro

wth

and

diff

eren

tiatio

n

fact

or-5

Sili

con

Stro

ntiu

m

Ther

mal

ly s

pray

ing

TiO

2 w

ith d

iffer

ent

crys

talli

nitie

s

Zirc

onia

-rei

nfor

cem

ent

Zn(2

+)

Ano

dic

spar

k ox

idat

ion

Bila

yere

d C

aP-H

A

Bla

stin

g

Dip

coa

ting

Ele

ctro

chem

ical

dep

ositi

on

Ele

ctro

n be

am e

vapo

ratio

n

Ele

ctro

phor

etic

dep

ositi

on

Ele

ctro

plat

ing

Ele

ctro

poly

mer

izat

ion

Ele

ctro

stat

ic s

pray

depo

sitio

n

In v

ivo

coat

ing

Ion

beam

dyn

amic

mix

ing

Mag

netic

co-

sput

terin

g

Mag

netro

n sp

utte

ring

Mol

ecul

ar p

recu

rsor

met

hod

Nan

o st

ruct

urin

g /

nano

grai

ning

Pla

sma

spra

ying

Spu

tterin

g

Pul

sed

lase

r dep

ositi

on

Sol

-gel

der

ived

Sol

utio

n-de

rived

bio

mim

etic

met

hod

Spa

rk d

isch

argi

ng

Enh

ance

d bo

ne c

ell a

dhes

ion

Ext

race

llula

r mat

rix p

rodu

ctio

n

Impr

oved

bio

com

patib

ility

/

bioa

ctiv

ity

Impr

oved

mec

hani

cal s

treng

th

Oss

eoco

nduc

tion

Oss

eoin

duct

ion

Ost

eobl

ast d

iffer

entia

tion

Ost

eocl

ast s

uppr

essi

on

Pre

vent

ion

of in

fect

ion

Pre

vent

ion

of o

steo

lysi

s

Pre

vent

ion

of w

ear d

ebris

Impl

ant u

se in

ost

eopo

rotic

and

aged

indi

vidu

als

27–

78

159

Coa

ting

type

S

ubst

rate

s A

dditi

onal

mod

ifica

tions

Coa

ting

met

hods

A

ims

of C

oatin

g R

efs.

Cer

amic

mat

eria

ls, o

xide

s

and

met

als

Bio

activ

e gl

ass

/ Gla

ss

cera

mic

Alu

min

a

Co-

Cr

Gla

ss-fi

ber-

com

posi

te

Sta

inle

ss s

teel

Tita

nium

Ti6A

l4V

Zirc

onia

Apa

tite

Wol

last

onite

(CaS

iO2)

Zirc

onia

CO

2 las

er

Ele

ctro

phor

etic

dep

ositi

on

Ena

mel

ing

tech

niqu

e

Flam

e sp

rayi

ng

Pla

sma

spra

ying

Pul

sed

lase

r dep

ositi

on

Bon

e re

cons

truct

ural

mat

eria

l

Oss

eoco

nduc

tion

Oss

eoin

tegr

atio

n

79–9

0

Tita

nium

dio

xide

/ tit

ania

Ti

tani

um

Ti6A

l4V

Ana

tase

tita

nium

oxid

e

Loca

l den

sific

atio

n of

titan

ia

SiO

2

Che

mic

al v

apor

dep

ositi

on

Col

loid

al s

uspe

nsio

n

Ele

ctro

chem

ical

act

ivat

ion

in N

aOH

Pla

sma

spra

ying

Sol

-gel

der

ived

CaP

-indu

ctio

n

Enh

ance

d bo

ne m

iner

aliz

atio

n

Enh

ance

d os

teob

last

pro

lifer

atio

n an

d

adhe

sion

Oss

eoin

tegr

atio

n

91–9

5

Alu

min

ium

oxi

de /

alum

ina

Alu

min

a

Tita

nium

Alu

min

a ce

ram

ic

bead

s

Alu

min

a ce

ram

ic b

inde

r

Ano

disa

tion

of e

lect

ron

beam

evap

orat

ed a

lum

inum

Enh

ance

d bo

ne c

ell a

dhes

ion

Oss

eoin

tegr

atio

n

96, 9

7

Zirc

oniu

m o

xide

/

zirc

onia

Tita

nium

Ti6A

l4V

Zirc

oniu

m

– C

ollo

idal

sus

pens

ion

Sol

-gel

der

ived

Oss

eoin

tegr

atio

n

Und

erst

andi

ng o

steo

blas

t act

ivity

in

bone

form

atio

n

98–

100

160

Coa

ting

type

S

ubst

rate

s A

dditi

onal

mod

ifica

tions

Coa

ting

met

hods

A

ims

of C

oatin

g R

efs.

Tita

nium

(-ca

rbid

e, -n

itrid

e)

Car

bon

fiber

-rei

nfor

ced

poly

mer

s

Pol

ymer

ic m

ater

ials

(UH

MW

PE

, PTF

E)

PM

MA

Sta

inle

ss s

teel

Tita

nium

Ti6A

l4V

Alk

ali t

reat

men

t

Nat

rium

tita

nate

Ioni

c pl

asm

a de

posi

tion

Pla

sma

spra

ying

Pow

der i

mm

ersi

on re

actio

n

assi

sted

coa

ting

Pul

sed

lase

r dep

ositi

on

Enh

ance

d ca

lciu

m

depo

sitio

n

Enh

ance

d os

teob

last

func

tion

Oss

eoco

nduc

tion

Oss

eoin

duct

ion

101–

107

Tant

alum

, tan

talu

m o

xide

C

arbo

n fib

er c

ompo

site

Sta

inle

ss s

teel

Fibr

illar

col

lage

n C

hem

ical

vap

or d

epos

ition

Mag

netro

n sp

utte

ring

Impr

oved

cor

rosi

on

resi

stan

ce

Oss

eoin

tegr

atio

n

108,

109

Gol

d Ti

tani

um

Ti6A

l4V

Por

ous

coat

ing

Oxi

datio

n

L-cy

stei

ne a

nd ty

pe I

colla

gen

Ele

ctro

lytic

stri

ke p

latin

g fo

llow

ed

by a

cid

gold

bat

h

Pho

tolit

hogr

aphy

met

hod

Gui

ded

tissu

e fo

rmat

ion

Ear

ly o

sseo

inte

grat

ion

Red

uced

infla

mm

ator

y

reac

tion

110,

111

Pol

ymer

s

Pol

y(D

, L-la

ctid

e), P

oly(

L-la

ctid

e),

slow

ly a

dsor

babl

e po

lym

ers

Sel

f-rei

nfor

ced

poly

glyc

olid

e

Sta

inle

ss s

teel

Tita

nium

Ti6A

l4V

Cal

cium

car

bona

te

Sur

face

fatty

aci

d

side

cha

ins

Dip

coa

ting

Dire

ct s

olve

nt p

olym

er d

epos

ition

Pla

sma-

poly

mer

ized

ally

lam

ine

Enh

ance

d os

teob

last

adhe

sion

Oss

eoin

tegr

atio

n

Ost

eobl

ast

diffe

rent

iatio

n

Ost

eoto

my

fixat

ion

112–

118

161

Coa

ting

type

S

ubst

rate

s A

dditi

onal

mod

ifica

tions

Coa

ting

met

hods

A

ims

of C

oatin

g R

efs.

Pol

ypho

spha

zene

(PTF

E)

Sili

con

Tita

nium

Dia

mon

d-lik

e

carb

on

Por

ous

coat

ed

Ads

orpt

ion

Pla

sma

acce

lera

ting

filte

red

puls

ed

arc

disc

harg

e m

etho

d

Oss

eoin

tegr

atio

n

Red

uced

infla

mm

ator

y re

actio

n

119,

120

Pol

ypyr

role

S

tain

less

ste

el

Tita

nium

Ti6A

l4V

Bov

ine

seru

m

albu

min

Ele

ctro

chem

ical

ly g

row

n

Ele

ctro

coat

ing

Enh

ance

d os

teob

last

prol

ifera

tion

and

adhe

sion

Impr

oved

bio

com

patib

ility

121,

122

Am

orph

ous

diam

ond

/ dia

mon

d-lik

e

carb

on (D

CL)

Cr-

Co-

Mo

Sta

inle

ss s

teel

Ti6A

l4V

SiC

C

hem

ical

vap

or d

epos

ition

Pla

sma

vapo

r dep

ositi

on m

agne

tron

sput

terin

g

Pul

sed

plas

ma

disc

harg

e m

etho

d

Impr

oved

cor

rosi

on re

sist

ance

Impr

oved

fric

tiona

l beh

avio

ur

Oss

eoin

tegr

atio

n

Pre

vent

ion

of w

ear d

ebris

123–

128

Tiss

ue e

ngin

eerin

g an

d bi

olog

ical

appr

oach

es

Ost

eoge

nic

bone

mar

row

cel

ls a

nd

auto

logo

us o

steo

blas

ts

CaP

sca

ffold

s

Tita

nium

Ti6A

l4V

CaP

bio

mim

etic

coat

ing

Cel

l cul

turin

g O

sseo

inte

grat

ion

Oss

eoco

nduc

tion

129,

130

Aut

o- a

nd a

llogr

aft b

one

trans

plan

tatio

n

Tita

nium

fibe

r

Tota

l kne

e

arth

ropl

asty

Por

ous

coat

ing

Aut

ogen

ous

canc

ello

us b

one

Aut

olog

ous

bone

chi

ps

Free

ze-d

ried

allo

geni

c ca

ncel

lous

bone

Bon

e m

arro

w g

rafti

ng

Oss

eoin

tegr

atio

n 13

1–

133

162

Coa

ting

type

S

ubst

rate

s A

dditi

onal

mod

ifica

tions

C

oatin

g m

etho

ds

Aim

s of

Coa

ting

Ref

s.

Col

lage

n C

obal

t allo

y

Sta

inle

ss s

teel

Tita

nium

Ti6A

l4V

Car

bodi

imid

e in

duce

d cr

oss-

linki

ng

Cho

ndro

itin

sulfa

te

Dec

orin

and

bio

glyc

an

HA

Tant

alum

oxi

de

Ads

orpt

ion

Bio

mim

etic

fibril

loge

nesi

s

Cov

alen

t

imm

obili

zatio

n

Col

lage

n/H

A

nano

com

posi

te

Enh

ance

d (e

arly

) bon

e

rem

odel

ing

Impr

oved

bio

com

patib

ility

Impr

oved

bon

e m

atur

atio

n an

d

min

eral

izat

ion

Oss

eoin

tegr

atio

n

Ost

eobl

ast d

iffer

entia

tion

134–

141

Fibr

onec

tin

Acr

ylic

impl

ant

Bio

glas

s

TiO

2/per

lite

Tita

nium

Tita

nium

-ion

plat

ing

Ads

orpt

ion

Cov

alen

t coa

ting

Enh

ance

d os

teob

last

func

tion

(Ear

ly) o

sseo

inte

grat

ion

Ost

eobl

ast c

hem

otax

is

142–

145

Bon

e si

alop

rote

in

Tita

nium

Gel

atin

D

ip c

oatin

g O

sseo

indu

ctio

n 14

6

Gro

wth

fact

ors:

Fibr

obla

st g

row

th fa

ctor

Tran

sfor

min

g gr

owth

fact

or

beta

1 an

d 2

Insu

lin-li

ke g

row

th fa

ctor

1

Hep

atoc

yte

grow

th fa

ctor

HA

Tita

nium

Tita

nium

-fibe

r-m

etal

coat

ed ro

d

Bio

degr

adab

le p

oly(

D, L

-

lact

ide)

HA

HA

-tric

alci

umph

osph

ate

Hum

an fi

bron

ectin

frag

men

t

rhB

MP

-2

Ads

orpt

ion

Loca

l dru

g de

liver

y sy

stem

Oss

eoin

duct

ion

Oss

eoin

tegr

atio

n

Ost

eobl

ast d

iffer

entia

tion

147–

153

163

Coa

ting

type

S

ubst

rate

sA

dditi

onal

mod

ifica

tions

Coa

ting

met

hods

A

ims

of C

oatin

g R

efs.

Bon

e m

orph

ogen

etic

pro

tein

/ B

MP

Ti

tani

um

Ti6A

l7N

b

CaP

HA

Nan

ocry

stal

line

diam

ond

Ads

orpt

ion

Bio

mim

etic

co-

prec

ipita

tion

with

CaP

CaP

car

rier

Col

lage

n sp

onge

car

rier

Pol

y(D

, L-la

ctid

e) c

arrie

r

Loca

l dru

g de

liver

y sy

stem

Oss

eoin

tegr

atio

n

Spi

ne fu

sion

154–

159

Cel

l adh

esio

n m

olec

ules

GFO

GE

R c

olla

gen

mim

etic

pept

ide

Tetra

-cel

l adh

esio

n m

olec

ule

(T-

CA

M)

Arg

-Gly

-Asp

pep

tide

(RG

D)

Cyc

lic R

GD

Arg

-Gly

-Asp

-Cys

pep

tide

(RG

DC

)

PM

MA

Tita

nium

Ti6A

l4V

HA

-bla

stin

g A

dsor

ptio

n

Acr

ylic

aci

d co

uplin

g

Thio

l anc

horin

g

SA

MP

pla

tform

Enh

ance

d os

teob

last

func

tion

Oss

eoin

tegr

atio

n

160–

167

Pho

spho

lipid

s Ti

tani

um

Ti6A

l7N

b

Ti6A

l4V

– A

dsor

ptio

n

Dip

coa

ting

Bio

mim

etic

cel

l mem

bran

e

Bon

e fo

rmat

ion

Oss

eoin

tegr

atio

n

168–

-

170

Chi

tosa

n Ti

tani

um

Bio

glas

s

HA

S

olut

ion

cast

ing

and

bond

ing

via

sila

ne

reac

tions

Chi

tosa

n-H

A c

ompo

site

Oss

eoin

tegr

atio

n

Bon

e fo

rmat

ion

171–

174

164

Coa

ting

type

S

ubst

rate

s A

dditi

onal

mod

ifica

tions

Coa

ting

met

hods

A

ims

of C

oatin

g R

efs.

Bis

phos

phan

ate

HA

Sta

inle

ss

stee

l

Tita

nium

– In

corp

orat

ion

into

CaP

coa

ting

Inco

rpor

atio

n in

to p

oly(

D, L

-lact

ide)

coat

ing

Incr

ease

d m

echa

nica

l fix

atio

n

Oss

eoin

tegr

atio

n

Red

uced

ost

eocl

ast r

esor

ptio

n ac

tivity

Use

in o

steo

poro

tic in

divi

dual

s

175–

179

Ant

imic

robi

al /

antib

iotic

s

Tita

nium

–

Pol

y(D

, L-la

ctid

e) lo

aded

with

the

antib

iotic

Sol

vent

cas

ting

tech

niqu

e

Loca

l ant

ibio

tic th

erap

y

Pre

vent

ion

of im

plan

t ass

ocia

ted

oste

omye

litis

Pro

tect

ion

agai

nst i

nfec

tion

180–

183

165

166

167

Appendix 2 1. Liu X & Niebur GL (2007) Bone ingrowth into a porous coated implant predicted by a

mechano-regulatory tissue differentiation algorithm. Biomech Model Mechanobiol [Epub ahead of print].

2. Rahbek O, Kold S, Zippor B, Overgaard S & Soballe K (2005) The influence of surface porosity on gap-healing around intra-articular implants in the presence of migrating particles. Biomaterials 26: 4728-4736.

3. Frenkel SR, Jaffe WL, Dimaano F, Iesaka K & Hua T (2004) Bone response to a novel highly porous surface in a canine implantable chamber. J Biomed Mater Res B Appl Biomater 15: 387-391.

4. von Knoch M, Engh CA Sr., Sychterz CJ, Engh CA Jr. & Willert HG (2000) Migration of polyethylene wear debris in one type of uncemented femoral component with circumferential porous coating: an autopsy study of 5 femurs. J Arthroplasty 15: 72-78.

5. Emerson RH Jr., Sanders SB, Head WC & Higgins L (1999) Effect of circumferential plasma-spray porous coating on the rate of femoral osteolysis after total hip arthroplasty. J Bone Joint Surg Am 81: 1291-1298.

6. Simmons CA, Valiquette N & Pilliar RM (1999) Osseointegration of sintered porous-surfaced and plasma spray-coated implants: An animal model study of early postimplantation healing response and mechanical stability. J Biomed Mater Res 47: 127-138.

7. Guldberg RE, Richards M, Caldwell NJ, Kuelske CL & Goldstein SA (1997) Trabecular bone adaptation to variations in porous-coated implant topology. J Biomech 30: 147-153.

8. Urban RM, Jacobs JJ, Sumner DR, Peters CL, Voss FR & Galante JO (1996) The bone-implant interface of femoral stems with non-circumferential porous coating. J Bone Joint Surg Am 78: 1068-1081.

9. Engh CA, O'Connor D, Jasty M, McGovern TF, Bobyn JD & Harris WH (1992) Quantification of implant micromotion, strain shielding, and bone resorption with porous-coated anatomic medullary locking femoral prostheses. Clin Orthop Relat Res 285: 13-29.

10. Jacobs JJ, Latanision RM, Rose RM & Veeck SJ (1990) The effect of porous coating processing on the corrosion behavior of cast Co-Cr-Mo surgical implant alloys. J Orthop Res 8: 874-882.

11. Cook SD, Thongpreda N, Anderson RC & Haddad RJ Jr. (1998) The effect of post-sintering heat treatments on the fatigue properties of porous coated Ti-6Al-4V alloy. J Biomed Mater Res 22: 287-302.

12. Manley MT, Stern LS & Gurtowski J (1985) The load carrying and fatigue properties of the stem-cement interface with smooth and porous coated femoral components. J Biomed Mater Res 19: 563-575.

13. Engh CA (1983) Hip arthroplasty with a Moore prosthesis with porous coating. A five-year study. Clin Orthop Relat Res 176: 52-66.

168

14. Cameron HU (1981) Essential design considerations for microporous implants: preliminary communication. J R Soc Med 74: 887-891.

15. Pilliar RM, Cameron HU & Macnab I (1975) Porous surface layered prosthetic devices. Biomed Eng 10: 126-131.

16. Beck U, Lange R & Neumann HG (2007) Micro-plasma textured Ti-implant surfaces. Biomol Eng 24: 47-51.

17. Hacking SA, Harvey EJ, Tanzer M, Krygier JJ & Bobyn JD (2003) Acid-etched microtexture for enhancement of bone growth into porous-coated implants. J Bone Joint Surg Br 85: 1182-1189.

18. Giavaresi G, Fini M, Chiesa R, Rimondini L, Rondelli G, Borsari V, Martini L, Nicolialdini N, Guzzardella GA & Giardino R (2002) Osseointegration of sandblasted or anodised hydrothermally-treated titanium implants: mechanical, histomorphometric and bone hardness measurements. Int J Artif Organs 25: 806-813.

19. Eckardt A, Aberman HM, Cantwell HD & Heine J (2003) Biological fixation of hydroxyapatite-coated versus grit-blasted titanium hip stems: a canine study. Arch Orthop Trauma Surg 123: 28-35.

20. Coathup MJ, Blunn GW, Flynn N, Williams C & Thomas NP (2001) A comparison of bone remodelling around hydroxyapatite-coated, porous-coated and grit-blasted hip replacements retrieved at post-mortem. J Bone Joint Surg Br 83: 118-123.

21. Vercaigne S, Wolke JG, Naert I & Jansen JA (2000) A histological evaluation of TiO2-gritblasted and Ca-P magnetron sputter coated implants placed into the trabecular bone of the goat: Part 2. Clin Oral Implants Res 11: 314-324.

22. Vercaigne S, Wolke JG, Naert I & Jansen JA (2000) A mechanical evaluation of TiO2-gritblasted and Ca-P magnetron sputter coated implants placed into the trabecular bone of the goat: Part 1. Clin Oral Implants Res 11: 305-313.

23. Overgaard S, Lind M, Glerup H, Bunger C & Soballe K (1998) Porous-coated versus grit-blasted surface texture of hydroxyapatite-coated implants during controlled micromotion: mechanical and histomorphometric results. J Arthroplasty 13: 449-458.

24. Giavaresi G, Fini M, Chiesa R, Giordano C, Sandrini E, Bianchi AE, Ceribelli P & Giardino R (2008) A novel multiphase anodic spark deposition coating for the improvement of orthopedic implant osseointegration: An experimental study in cortical bone of sheep. J Biomed Mater Res A 85: 1022-1031.

25. Giordano C, Chiesa R, Sandrini E, Cigada A, Giavaresi G, Fini M & Giardino R (2005) Physical and biological characterizations of a novel multiphase anodic spark deposition coating to enhance implant osseointegration. J Mater Sci Mater Med 16: 1221-1229.

26. Sandrini E, Morris C, Chiesa R, Cigada A & Santin M (2005) In vitro assessment of the osteointegrative potential of a novel multiphase anodic spark deposition coating for orthopaedic and dental implants. J Biomed Mater Res B Appl Biomater 73: 392-399.

169

27. Bernstein A, Nobel D, Mayr HO, Berger G, Gildenhaar R & Brandt J (2008) Histological and histomorphometric investigations on bone integration of rapidly resorbable calcium phosphate ceramics. J Biomed Mater Res B Appl Biomater 84: 452-462.

28. Yamada Y, Ito A, Kojima H, Sakane M, Miyakawa S, Uemura T & LeGeros RZ (2008) Inhibitory effect of Zn2+ in zinc-containing beta-tricalcium phosphate on resorbing activity of mature osteoclasts. J Biomed Mater Res A 84: 344-352.

29. Laure B, Besnier JM, Bergemer-Fouquet AM, Marquet-Van Der Mee N, Damie F, Quentin R, Favard L & Rosset P (2008) Effect of hydroxyapatite coating and polymethylmethacrylate on stainless steel implant-site infection with Staphylococcus epidermidis in a sheep model. J Biomed Mater Res A 84: 92-98.

30. Malhotra R, Dua A, Kiran EK & Bhan S (2008) Femoral revision using long hydroxyapatite-coated interlocking stem. Arch Orthop Trauma Surg 128: 355-362.

31. Oliveira AL, Reis RL & Li P (2007) Strontium-substituted apatite coating grown on Ti6Al4V substrate through biomimetic synthesis. J Biomed Mater Res B Appl Biomater 83: 258-265.

32. Yang CY, Chen CR, Chang E & Lee TM (2007) Characteristics of hydroxyapatite coated titanium porous coatings on Ti-6Al-4V substrates by plasma sprayed method. J Biomed Mater Res B Appl Biomater 82: 450-459.

33. Chen F, Lam WM, Lin CJ, Qiu GX, Wu ZH, Luk KD & Lu WW (2007) Biocompatibility of electrophoretical deposition of nanostructured hydroxyapatite coating on roughen titanium surface: in vitro evaluation using mesenchymal stem cells. J Biomed Mater Res B Appl Biomater 82: 183-191.

34. Suh JY, Jeung OC, Choi BJ & Park JW (2007) Effects of a novel calcium titanate coating on the osseointegration of blasted endosseous implants in rabbit tibiae. Clin Oral Implants Res 18: 362-369.

35. Balamurugan A, Rebelo A, Kannan S, Ferreira JM, Michel J, Balossier G & Rajeswari S (2007) Characterization and in vivo evaluation of sol-gel derived hydroxyapatite coatings on Ti6Al4V substrates. J Biomed Mater Res B Appl Biomater 81: 441-447.

36. McLeod K, Anderson GI, Dutta NK, Smart RS, Voelcker NH, Sekel R & Kumar S (2006) Adsorption of bisphosphonate onto hydroxyapatite using a novel co-precipitation technique for bone growth enhancement. J Biomed Mater Res A 79: 271-281.

37. Hayakawa T, Takahashi K, Yoshinari M, Okada H, Yamamoto H, Sato M & Nemoto K (2006) Trabecular bone response to titanium implants with a thin carbonate-containing apatite coating applied using the molecular precursor method. Int J Oral Maxillofac Implants 21: 851-858.

38. Yildirim OS, Aksakal B, Hanyaloglu SC, Erdogan F & Okur A (2006) Hydroxyapatite dip coated and uncoated titanium poly-axial pedicle screws: an in vivo bovine model. Spine 31: E215-E220.

170

39. Goyenvalle E, Aguado E, Nguyen JM, Passuti N, Le GL, Layrolle P & Daculsi G (2006) Osteointegration of femoral stem prostheses with a bilayered calcium phosphate coating. Biomaterials 27: 1119-1128.

40. Du C, Schneider GB, Zaharias R, Abbott C, Seabold D, Stanford C & Moradian-Oldak J (2005) Apatite/amelogenin coating on titanium promotes osteogenic gene expression. J Dent Res 84: 1070-1074.

41. Ekholm E, Tommila M, Forsback AP, Martson M, Holmbom J, Ääritalo V, Finnberg C, Kuusilehto A, Salonen J, Yli-Urpo A & Penttinen R (2005) Hydroxyapatite coating of cellulose sponge does not improve its osteogenic potency in rat bone. Acta Biomater 1: 535-544.

42. 42. Oh S, Tobin E, Yang Y, Carnes DL Jr. & Ong JL (2005) In vivo evaluation of hydroxyapatite coatings of different crystallinities. Int J Oral Maxillofac Implants 20: 726-731.

43. Coathup MJ, Blackburn J, Goodship AE, Cunningham JL, Smith T & Blunn GW (2005) Role of hydroxyapatite coating in resisting wear particle migration and osteolysis around acetabular components. Biomaterials 26: 4161-4169.

44. Thian ES, Huang J, Best SM, Barber ZH & Bonfield W (2005) Magnetron co-sputtered silicon-containing hydroxyapatite thin films - an in vitro study. Biomaterials 26: 2947-2956.

45. Yang Y, Kim KH & Ong JL (2005) A review on calcium phosphate coatings produced using a sputtering process - an alternative to plasma spraying. Biomaterials 26: 327-337.

46. Lee TM, Yang CY, Chang E & Tsai RS (2004) Comparison of plasma-sprayed hydroxyapatite coatings and zirconia-reinforced hydroxyapatite composite coatings: in vivo study. J Biomed Mater Res A 71: 652-660.

47. Xie J, Baumann MJ & McCabe LR (2004) Osteoblasts respond to hydroxyapatite surfaces with immediate changes in gene expression. J Biomed Mater Res A 71: 108-117.

48. Bhadang KA & Gross KA (2004) Influence of fluorapatite on the properties of thermally sprayed hydroxyapatite coatings. Biomaterials 25: 4935-4945.

49. Heimann RB, Schurmann N & Muller RT (2004) In vitro and in vivo performance of Ti6Al4V implants with plasma-sprayed osteoconductive hydroxylapatite-bioinert titania bond coat "duplex" systems: an experimental study in sheep. J Mater Sci Mater Med 15: 1045-1052.

50. Nguyen HQ, Deporter DA, Pilliar RM, Valiquette N & Yakubovich R (2004) The effect of sol-gel-formed calcium phosphate coatings on bone ingrowth and osteoconductivity of porous-surfaced Ti alloy implants. Biomaterials 25: 865-876.

51. Barrere F, van de Valk CM, Dalmeijer RA, Meijer G, van Blitterswijk CA, de Groot K & Layrolle P (2003) Osteogenecity of octacalcium phosphate coatings applied on porous metal implants. J Biomed Mater Res A 66: 779-788.

52. Goyenvalle E, Guyen NJ, Aguado E, Passuti N & Daculsi G (2003) Bilayered calcium phosphate coating to promote osseointegration of a femoral stem prosthesis. J Mater Sci Mater Med 14: 219-227.

171

53. Daculsi G, Laboux O, Malard O & Weiss P (2003) Current state of the art of biphasic calcium phosphate bioceramics. J Mater Sci Mater Med 14: 195-200.

54. Du C, Meijer GJ, van de Valk CM, Haan RE, Bezemer JM, Hesseling SC, Cui FZ, de Groot K & Layrolle P (2002) Bone growth in biomimetic apatite coated porous Polyactive 1000PEGT70PBT30 implants. Biomaterials 23: 4649-4656.

55. Ter Brugge PJ, Wolke JG & Jansen JA (2002) Effect of calcium phosphate coating crystallinity and implant surface roughness on differentiation of rat bone marrow cells. J Biomed Mater Res 60: 70-78.

56. Le IS, Kim DH, Kim HE, Jung YC & Han CH (2002) Biological performance of calcium phosphate films formed on commercially pure Ti by electron-beam evaporation. Biomaterials 23: 609-615.

57. Moursi AM, Winnard AV, Winnard PL, Lannutti JJ & Seghi RR (2002) Enhanced osteoblast response to a polymethylmethacrylate-hydroxyapatite composite. Biomaterials 23: 133-144.

58. Lind M, Overgaard S, Bunger C & Soballe K (1999) Improved bone anchorage of hydroxypatite coated implants compared with tricalcium-phosphate coated implants in trabecular bone in dogs. Biomaterials 20: 803-808.

59. Soballe K, Overgaard S, Hansen ES, Brokstedt-Rasmussen H, Lind M & Bunger C (1999) A review of ceramic coatings for implant fixation. J Long Term Eff Med Implants 9: 131-151.

60. Morscher EW, Hefti A & Aebi U (1998) Severe osteolysis after third-body wear due to hydroxyapatite particles from acetabular cup coating. J Bone Joint Surg Br 80: 267-272.

61. Okamoto K, Matsuura T, Hosokawa R & Akagawa Y (1998) RGD peptides regulate the specific adhesion scheme of osteoblasts to hydroxyapatite but not to titanium. J Dent Res 77: 481-487.

62. Pazzaglia UE, Brossa F, Zatti G, Chiesa R & Andrini L (1998) The relevance of hydroxyapatite and spongious titanium coatings in fixation of cementless stems. An experimental comparative study in rat femur employing histological and microangiographic techniques. Arch Orthop Trauma Surg 117: 279-285.

63. Ban S, Maruno S, Arimoto N, Harada A & Hasegawa J (1997) Effect of electrochemically deposited apatite coating on bonding of bone to the HA-G-Ti composite and titanium. J Biomed Mater Res 36: 9-15.

64. D'Antonio JA, Capello WN & Manley MT (1996) Remodeling of bone around hydroxyapatite-coated femoral stems. J Bone Joint Surg Am 78: 1226-1234.

65. Harada Y, Wang JT, Doppalapudi VA, Willis AA, Jasty M, Harris WH, Nagase M & Goldring SR (1996) Differential effects of different forms of hydroxyapatite and hydroxyapatite/tricalcium phosphate particulates on human monocyte/macrophages in vitro. J Biomed Mater Res 31: 19-26.

66. Hasegawa S (1996) Characterization and canine bone tissue reaction of hydroxyapatite-coated titanium using thermal decomposition method. Bull Tokyo Med Dent Univ 43: 25-44.

172

67. Caulier H, van der Waerden JP, Paquay YC, Wolke JG, Kalk W, Naert I & Jansen JA (1995) Effect of calcium phosphate (Ca-P) coatings on trabecular bone response: a histological study. J Biomed Mater Res 29: 1061-1069.

68. Hetherington VJ, Lord CE & Brown SA (1995) Mechanical and histological fixation of hydroxylapatite-coated pyrolytic carbon and titanium alloy implants: a report of short-term results. J Appl Biomater 6: 243-248.

69. Dalton JE, Cook SD, Thomas KA & Kay JF (1995) The effect of operative fit and hydroxyapatite coating on the mechanical and biological response to porous implants. J Bone Joint Surg Am 77: 97-110.

70. De Bruijn JD, Bovell YP & van Blitterswijk CA (1994) Structural arrangements at the interface between plasma sprayed calcium phosphates and bone. Biomaterials 15: 543-550.

71. Dhert WJ (1994) Retrieval studies on calcium phosphate-coated implants. Med Prog Technol 20: 143-154.

72. Soballe K, Hansen ES, Brockstedt-Rasmussen H & Bunger C (1993) Hydroxyapatite coating converts fibrous tissue to bone around loaded implants. J Bone Joint Surg Br 75: 270-278.

73. Soballe K, Hansen ES, Brockstedt-Rasmussen H, Pedersen CM & Bunger C (1992) Bone graft incorporation around titanium-alloy- and hydroxyapatite-coated implants in dogs. Clin Orthop Relat Res 274: 282-293.

74. Cook SD, Thomas KA & Kay JF (1991) Experimental coating defects in hydroxylapatite-coated implants. Clin Orthop Relat Res 265: 280-290.

75. Thomas KA, Cook SD, Haddad RJ Jr., Kay JF & Jarcho M (1989) Biologic response to hydroxylapatite-coated titanium hips. A preliminary study in dogs. J Arthroplasty 4: 43-53.

76. Geesink RG, de Groot K & Klein CP (1987) Chemical implant fixation using hydroxyl-apatite coatings. The development of a human total hip prosthesis for chemical fixation to bone using hydroxyl-apatite coatings on titanium substrates. Clin Orthop Relat Res 225: 147-170.

77. Thomas KA, Kay JF, Cook SD & Jarcho M (1987) The effect of surface macrotexture and hydroxylapatite coating on the mechanical strengths and histologic profiles of titanium implant materials. J Biomed Mater Res 21: 1395-1414.

78. Ducheyne P, Hench LL, Kagan A, Martens M, Bursens A & Mulier JC (1980) Effect of hydroxyapatite impregnation on skeletal bonding of porous coated implants. J Biomed Mater Res 14: 225-237.

79. Tuusa SM, Peltola MJ, Tirri T, Puska MA, Röyttä M, Aho H, Sandholm J, Lassila LV & Vallittu PK (2008) Reconstruction of critical size calvarial bone defects in rabbits with glass-fiber-reinforced composite with bioactive glass granule coating. J Biomed Mater Res B Appl Biomater 84:510-519.

80. Borrajo JP, Serra J, Gonzalez P, Leon B, Munoz FM & Lopez M (2007) In vivo evaluation of titanium implants coated with bioactive glass by pulsed laser deposition. J Mater Sci Mater Med 18: 2371-2376.

173

81. Radice S, Kern P, Burki G, Michler J & Textor M (2007) Electrophoretic deposition of zirconia-Bioglass composite coatings for biomedical implants. J Biomed Mater Res A 82: 436-444.

82. Xue W, Liu X, Zheng X & Ding C (2005) In vivo evaluation of plasma-sprayed wollastonite coating. Biomaterials 26: 3455-3460.

83. Ignatius A, Peraus M, Schorlemmer S, Augat P, Burger W, Leyen S & Claes L (2005) Osseointegration of alumina with a bioactive coating under load-bearing and unloaded conditions. Biomaterials 26: 2325-2332.

84. Moritz N, Rossi S, Vedel E, Tirri T, Ylänen H, Aro H & Närhi T (2004) Implants coated with bioactive glass by CO2-laser, an in vivo study. J Mater Sci Mater Med 15: 795-802.

85. Merolli A, Cacchioli A, Giannotta L & Tranquilli LP (2001) Energy dispersive analysis (EDX) of a degradable bioactive-glass coating on Ti6Al4V in-vivo. J Mater Sci Mater Med 12: 727-730.

86. Wheeler DL, Montfort MJ & McLoughlin SW (2001) Differential healing response of bone adjacent to porous implants coated with hydroxyapatite and 45S5 bioactive glass. J Biomed Mater Res 55: 603-612.

87. Price N, Bendall SP, Frondoza C, Jinnah RH & Hungerford DS (1997) Human osteoblast-like cells (MG63) proliferate on a bioactive glass surface. J Biomed Mater Res 37: 394-400.

88. Li ZL, Kitsugi T, Yamamuro T, Chang YS, Senaha Y, Takagi H, Nakamura T & Oka M (1995) Bone-bonding behavior under load-bearing conditions of an alumina ceramic implant incorporating beads coated with glass-ceramic containing apatite and wollastonite. J Biomed Mater Res 29: 1081-1088.

89. Ido K, Matsuda Y, Yamamuro T, Okumura H, Oka M & Takagi H (1993) Cementless total hip replacement. Bio-active glass ceramic coating studied in dogs. Acta Orthop Scand 64: 607-612.

90. Lacefield WR & Hench LL (1986) The bonding of Bioglass to a cobalt-chromium surgical implant alloy. Biomaterials 7: 104-108.

91. Xie J & Luan BL (2008) Nanometer-scale surface modification of Ti6Al4V alloy for orthopedic applications. J Biomed Mater Res A 84: 63-72.

92. Areva S, Ääritalo V, Tuusa M, Jokinen M, Linden M & Peltola T (2007) Sol-gel-derived TiO2-SiO2 implant coatings for direct tissue attachment. Part II: Evaluation of cell responce. J Mater Sci Mater Med 18: 1633-1643.

93. Giavaresi G, Ambrosio L, Battiston GA, Casellato U, Gerbasi R, Finia M, Aldini NN, Martini L, Rimondini L & Giardino R (2004) Histomorphometric, ultrastructural and microhardness evaluation of the osseointegration of a nanostructured titanium oxide coating by metal-organic chemical vapour deposition: an in vivo study. Biomaterials 25: 5583-5591.

94. Kim HK, Jang JW & Lee CH (2004) Surface modification of implant materials and its effect on attachment and proliferation of bone cells. J Mater Sci Mater Med 15: 825-830.

174

95. Moritz N, Jokinen M, Peltola T, Areva S & Yli-Urpo A (2003) Local induction of calcium phosphate formation on TiO2 coatings on titanium via surface treatment with a CO2 laser. J Biomed Mater Res A 65: 9-16.

96. Briggs EP, Walpole AR, Wilshaw PR, Karlsson M & Palsgard E (2004) Formation of highly adherent nano-porous alumina on Ti-based substrates: a novel bone implant coating. J Mater Sci Mater Med 15: 1021-1029.

97. Takagi H, Yamamuro T, Hyakuna K, Nakamura T, Kotoura Y & Oka M (1989) Bone bonding behavior of bead-coated alumina ceramic under load-bearing conditions. J Biomed Mater Res 23: 161-181.

98. Sollazzo V, Pezzetti F, Scarano A, Piattelli A, Bignozzi CA, Massari L, Brunelli G & Carinci F (2008) Zirconium oxide coating improves implant osseointegration in vivo. Dent Mater 24: 357-361.

99. Sollazzo V, Palmieri A, Pezzetti F, Bignozzi CA, Argazzi R, Massari L, Brunelli G & Carinci F (2008) Genetic effect of zirconium oxide coating on osteoblast-like cells. J Biomed Mater Res B Appl Biomater 84: 550-558.

100. Filiaggi MJ, Pilliar RM, Yakubovich R & Shapiro G (1996) Evaluating sol-gel ceramic thin films for metal implant applications. I. Processing and structure of zirconia films on Ti-6AI-4V. J Biomed Mater Res 33: 225-238.

101. Reising A, Yao C, Storey D & Webster TJ (2007) Greater osteoblast long-term functions on ionic plasma deposited nanostructured orthopedic implant coatings. J Biomed Mater Res A [Epub ahead of print].

102. Shalabi MM, Wolke JG, Cuijpers VM & Jansen JA (2007) Evaluation of bone response to titanium-coated polymethyl methacrylate resin (PMMA) implants by X-ray tomography. J Mater Sci Mater Med 18: 2033-2039.

103. Brama M, Rhodes N, Hunt J, Ricci A, Teghil R, Migliaccio S, Rocca CD, Leccisotti S, Lioi A, Scandurra M, De Maria G, Ferro D, Pu F, Panzini G, Politi L & Scandurra R (2007) Effect of titanium carbide coating on the osseointegration response in vitro and in vivo. Biomaterials 28: 595-608.

104. Freeman CO & Brook IM (2006) Bone response to a titanium aluminium nitride coating on metallic implants. J Mater Sci Mater Med 17: 465-470.

105. Xue W, Liu X, Zheng X & Ding C (2005) In vivo evaluation of plasma-sprayed titanium coating after alkali modification. Biomaterials 26: 3029-3037.

106. Sovak G, Weiss A & Gotman I (2000) Osseointegration of Ti6Al4V alloy implants coated with titanium nitride by a new method. J Bone Joint Surg Br 82: 290-296.

107. Harman MK, Banks SA & Hodge WA (1997) Wear analysis of a retrieved hip implant with titanium nitride coating. J Arthroplasty 12: 938-945.

108. Li H, Zou X, Woo C, Ding M, Lind M & Bunger C (2007) Experimental lumbar spine fusion with novel tantalum-coated carbon fiber implant. J Biomed Mater Res B Appl Biomater 81: 194-200.

109. Muller R, Abke J, Schnell E, Macionczyk F, Gbureck U, Mehrl R, Ruszczak Z, Kujat R, Englert C, Nerlich M & Angele P (2005) Surface engineering of stainless steel materials by covalent collagen immobilization to improve implant biocompatibility. Biomaterials 26: 6962-6972.

175

110. Zainali K, Danscher G, Jakobsen T, Jakobsen SS, Baas J, Moller P, Bechtold JE & Soballe K (2008) Effects of gold coating on experimental implant fixation. J Biomed Mater Res A [Epub ahead of print].

111. Zhao JM, Tsuru K, Hayakawa S & Osaka A (2008) Modification of Ti implant surface for cell proliferation and cell alignment. J Biomed Mater Res A 84: 988-993.

112. Lagoa AL, Wedemeyer C, von Knoch M, Löer F & Epple M (2008) A strut graft substitute consisting of a metal core and a polymer surface. J Mater Sci Mater Med 19: 417-424.

113. Finke B, Luethen F, Schroeder K, Mueller PD, Bergemann C, Frant M, Ohl A & Nebe BJ (2007) The effect of positively charged plasma polymerization on initial osteoblastic focal adhesion on titanium surfaces. Biomaterials 28: 4521-4534.

114. Porter K, Hossain M, Wang M, Radano CP, Baker GL, Smith MR III & McCabe LR (2006) Regulation of osteoblast gene expression and phenotype by polylactide-fatty acid surfaces. Mol Biol Rep 33: 1-12.

115. Partale K, Klein P, Schell H, Schmidmaier G, Wildemann B, Bail H, Schiller R, Braqulla H & Duda GN (2005) Poly(D,L-lactide) coating is capable of enhancing osseous integration of Schanz screws in the absence of infection. J Biomed Mater Res B Appl Biomater 74: 608-616.

116. Bagno A, Genovese M, Luchini A, Dettin M, Conconi MT, Menti AM, Parniqotto PP & Di Bello C (2004) Contact profilometry and correspondence analysis to correlate surface properties and cell adhesion in vitro of uncoated and coated Ti and Ti6Al4V disks. Biomaterials 25: 2437-2445.

117. Merolli A, Gabbi C, Cacchioli A, Ragionieri L, Caruso L, Giannotta L & Tranguilli Leali P (2001) Bone response to polymers based on poly-lactic acid and having different degradation times. J Mater Sci Mater Med 12: 775-778.

118. Vasenius J, Vainionpää S, Vihtonen K, Mero M, Mäkelä A, Tormälä P & Rokkanen P (1990) A histomorphological study on self-reinforced polyglycolide (SR-PGA) osteosynthesis implants coated with slowly absorbable polymers. J Biomed Mater Res 24: 1615-1635.

119. Zeifang F, Grunze M, Delling G, Lorenz H, Heisel C, Tosounidis G, Sabo D, Simank HG & Holstein JH (2008) Improved osseointegration of PTFEP-coated titanium implants. Med Sci Monit 14: BR35-BR40.

120. Kinnari TJ, Soininen A, Esteban J, Zamora N, Alakoski E, Kouri VP, Lappalainen R, Konttinen YT, Gomez-Barrena E & Tiainen VM (2007) Adhesion of staphylococcal and Caco-2 cells on diamond-like carbon polymer hybrid coating. J Biomed Mater Res A [Epub ahead of print].

121. De Giglio E, Cometa S, Calvano CD, Sabbatini L, Zambonin PG, Colucci S, Benedetto AD & Colaianni G (2007) A new titanium biofunctionalized interface based on poly(pyrrole-3-acetic acid) coating: proliferation of osteoblast-like cells and future perspectives. J Mater Sci Mater Med 18: 1781-1789.

122. Khan W, Kapoor M & Kumar N (2007) Covalent attachment of proteins to functionalized polypyrrole-coated metallic surfaces for improved biocompatibility. Acta Biomater 3: 541-549.

176

123. Santavirta S (2003) Compatibility of the totally replaced hip. Reduction of wear by amorphous diamond coating. Acta Orthop Scand Suppl 74: 1-19.

124. Brizuela M, Garcia-Luis A, Viviente JL, Braceras I & Onate JI (2002) Tribological study of lubricious DLC biocompatible coatings. J Mater Sci Mater Med 13: 1129-1133.

125. Rupprecht S, Bloch A, Rosiwal S, Neukam FW & Wiltfang J (2002) Examination of the bone-metal interface of titanium implants coated by the microwave plasma chemical vapor deposition method. Int J Oral Maxillofac Implants 17: 778-785.

126. Affatato S, Frigo M & Toni A (2000) An in vitro investigation of diamond-like carbon as a femoral head coating. J Biomed Mater Res 53: 221-226.

127. Lappalainen R, Anttila A & Heinonen H (1998) Diamond coated total hip replacements. Clin Orthop Relat Res 352: 118-127.

128. Aspenberg P, Anttila A, Konttinen YT, Lappalainen R, Goodman SB, Nordsletten L & Santavirta S (1996) Benign response to particles of diamond and SiC: bone chamber studies of new joint replacement coating materials in rabbits. Biomaterials 17: 807-812.

129. Frosch KH, Sondergeld I, Dresing K, Rudy T, Lohmann CH, Rabba J, Schild D, Breme J & Stuermer KM (2003) Autologous osteoblasts enhance osseointegration of porous titanium implants. J Orthop Res 21: 213-223.

130. De Bruijn JD, van den Brink I, Mendes S, Dekker R, Bovell YP & van Blitterswijk CA (1999) Bone induction by implants coated with cultured osteogenic bone marrow cells. Adv Dent Res 13: 74-81.

131. Kim KJ, Iwase M, Kotake S & Itoh T (2007) Effect of bone marrow grafting on the titanium porous-coated implant in bilateral total knee arthroplasty. Acta Orthop 78: 116-122.

132. Kienapfel H, Sumner DR, Turner TM, Urban RM & Galante JO (1992) Efficacy of autograft and freeze-dried allograft to enhance fixation of porous coated implants in the presence of interface gaps. J Orthop Res 10: 423-433.

133. Hofmann AA, Bloebaum RD, Rubman MH, Bachus KN & Plaster RL (1992) Microscopic analysis of autograft bone applied at the interface of porous-coated devices in human cancellous bone. Int Orthop 16: 349-358.

134. Teng SH, Lee EJ, Park CS, Choi WY, Shin DS & Kim HE (2008) Bioactive nanocomposite coatings of collagen/hydroxyapatite on titanium substrates. J Mater Sci Mater Med [Epub ahead of print].

135. Douglas T, Hempel U, Mietrach C, Viola M, Vigetti D, Heinemann S, Bierbaum S, Scharnweber D & Worch H (2008) Influence of collagen-fibril-based coatings containing decorin and biglycan on osteoblast behavior. J Biomed Mater Res A 84: 805-816.

136. Rammelt S, Heck C, Bernhardt R, Bierbaum S, Scharnweber D, Goebbels J, Ziegler J, Biewener A & Zwipp H (2007) In vivo effects of coating loaded and unloaded Ti implants with collagen, chondroitin sulfate, and hydroxyapatite in the sheep tibia. J Orthop Res 25: 1052-1061.

177

137. Muller R, Abke J, Schnell E, Scharnweber D, Kujat R, Englert C, Taheri D, Nerlich M & Angele P (2006) Influence of surface pretreatment of titanium- and cobalt-based biomaterials on covalent immobilization of fibrillar collagen. Biomaterials 27: 4059-4068.

138. Morra M, Cassinelli C, Meda L, Fini M, Giavaresi G & Giardino R (2005) Surface analysis and effects on interfacial bone microhardness of collagen-coated titanium implants: a rabbit model. Int J Oral Maxillofac Implants 20: 23-30.

139. Rammelt S, Schulze E, Bernhardt R, Hanisch U, Scharnweber D, Worch H, Zwipp H & Biewener A (2004) Coating of titanium implants with type-I collagen. J Orthop Res 22: 1025-1034.

140. Morra M, Cassinelli C, Cascardo G, Cahalan P, Cahalan L, Fini M & Giardino R (2003) Surface engineering of titanium by collagen immobilization. Surface characterization and in vitro and in vivo studies. Biomaterials 24: 4639-4654.

141. Becker D, Geissler U, Hempel U, Bierbaum S, Scharnweber D, Worch H Wenzel KW (2002) Proliferation and differentiation of rat calvarial osteoblasts on type I collagen-coated titanium alloy. J Biomed Mater Res 59: 516-527.

142. Jimbo R, Sawase T, Shibata Y, Hirata K, Hishikawa Y, Tanaka Y, Bessho K, Ikeda T & Atsuta M (2007) Enhanced osseointegration by the chemotactic activity of plasma fibronectin for cellular fibronectin positive cells. Biomaterials 28: 3469-3477.

143. Erli HJ, Ruger M, Ragoss C, Jahnen-Dechent W, Hollander DA, Paar O & von Walter M (2006) The effect of surface modification of a porous TiO2/perlite composite on the ingrowth of bone tissue in vivo. Biomaterials 27: 1270-1276.

144. Ku Y, Chung CP & Jang JH (2005) The effect of the surface modification of titanium using a recombinant fragment of fibronectin and vitronectin on cell behavior. Biomaterials 26: 5153-5157.

145. Seitz TL, Noonan KD, Hench LL & Noonan NE (1982) Effect of fibronectin on the adhesion of an established cell line to a surface reactive biomaterial. J Biomed Mater Res 16: 195-207.

146. O'Toole GC, Salih E, Gallagher C, FitzPatrick D, O'Higgins N & O'Rourke SK (2004) Bone sialoprotein-coated femoral implants are osteoinductive but mechanically compromised. J Orthop Res 22: 641-646.

147. Li Y, Lee IS, Cui FZ & Choi SH (2008) The biocompatibility of nanostructured calcium phosphate coated on micro-arc oxidized titanium. Biomaterials 29: 2025-2032.

148. Park JM, Koak JY, Jang JH, Han CH, Kim SK & Heo SJ (2006) Osseointegration of anodized titanium implants coated with fibroblast growth factor-fibronectin (FGF-FN) fusion protein. Int J Oral Maxillofac Implants 21: 859-866.

149. Sumner DR, Turner TM, Urban RM, Virdi AS & Inoue N (2006) Additive enhancement of implant fixation following combined treatment with rhTGF-beta2 and rhBMP-2 in a canine model. J Bone Joint Surg Am 88: 806-817.

178

150. Wildemann B, Sander A, Schwabe P, Lucke M, Stockle U, Raschke M, Haas NP & Schmidmaier G (2005) Short term in vivo biocompatibility testing of biodegradable poly(D,L-lactide) - growth factor coating for orthopaedic implants. Biomaterials 26: 4035-4040.

151. Hossain M, Irwin R, Baumann MJ & McCabe LR (2005) Hepatocyte growth factor (HGF) adsorption kinetics and enhancement of osteoblast differentiation on hydroxyapatite surfaces. Biomaterials 26: 2595-2602.

152. Blom EJ, Klein-Nulend J, Yin L, van Waas MA & Burger EH (2001) Transforming growth factor-beta1 incorporated in calcium phosphate cement stimulates osteotransductivity in rat calvarial bone defects. Clin Oral Implants Res 12: 609-616.

153. Sumner DR, Turner TM, Purchio AF, Gombotz WR, Urban RM & Galante JO (1995) Enhancement of bone ingrowth by transforming growth factor-beta. J Bone Joint Surg Am 77: 1135-1147.

154. Kloss FR, Gassner R, Preiner J, Ebner A, Larsson K, Hächl O, Tuli T, Rasse M, Moser D, Laimer K, Nickel EA, Laschober G, Brunaver R, Klima G, Hinterdorfer P, Steinmuller-Nethl D & Lepperdinger D (2008) The role of oxygen termination of nanocrystalline diamond on immobilisation of BMP-2 and subsequent bone formation. Biomaterials 29: 2433-2442.

155. Liu Y, Li JP, Hunziker EB & de Groot K (2006) Incorporation of growth factors into medical devices via biomimetic coatings. Philos Transact A Math Phys Eng Sci 364: 233-248.

156. Aebli N, Stich H, Schawalder P, Theis JC & Krebs J (2005) Effects of bone morphogenetic protein-2 and hyaluronic acid on the osseointegration of hydroxyapatite-coated implants: an experimental study in sheep. J Biomed Mater Res A 73: 295-302.

157. Liu Y, de Groot K & Hunziker EB (2005) BMP-2 liberated from biomimetic implant coatings induces and sustains direct ossification in an ectopic rat model. Bone 36: 745-757.

158. Bragdon CR, Doherty AM, Rubash HE, Jasty M, Li XJ, Seeherman H & Harris WH (2003) The efficacy of BMP-2 to induce bone ingrowth in a total hip replacement model. Clin Orthop Relat Res 417: 50-61.

159. Kandziora F, Bail H, Schmidmaier G, Schollmeier G, Scholz M, Knispel C, Hiller T, Pflugmacher R, Mittlmeier T, Raschke M & Haas NP (2002) Bone morphogenetic protein-2 application by a poly(D,L-lactide)-coated interbody cage: in vivo results of a new carrier for growth factors. J Neurosurg 97: 40-48.

160. Shannon FJ, Cottrell JM, Deng XH, Crowder KN, Doty SB, Avaltroni MJ, Warren RF, Wright TM & Schwartz J (2007) A novel surface treatment for porous metallic implants that improves the rate of bony ongrowth. J Biomed Mater Res A [Epub ahead of print].

161. Reyes CD, Petrie TA, Burns KL, Schwartz Z & Garcia AJ (2007) Biomolecular surface coating to enhance orthopaedic tissue healing and integration. Biomaterials 28: 3228-3235.

179

162. Park JW, Lee SG, Choi BJ & Suh JY (2007) Effects of a cell adhesion molecule coating on the blasted surface of titanium implants on bone healing in the rabbit femur. Int J Oral Maxillofac Implants 22: 533-541.

163. Rammelt S, Illert T, Bierbaum S, Scharnweber D, Zwipp H & Schneiders W (2006) Coating of titanium implants with collagen, RGD peptide and chondroitin sulfate. Biomaterials 27: 5561-5571.

164. Elmengaard B, Bechtold JE & Soballe K (2005) In vivo effects of RGD-coated titanium implants inserted in two bone-gap models. J Biomed Mater Res A 75: 249-255.

165. Elmengaard B, Bechtold JE & Soballe K (2005) In vivo study of the effect of RGD treatment on bone ongrowth on press-fit titanium alloy implants. Biomaterials 26: 3521-3526.

166. Ferris DM, Moodie GD, Dimond PM, Gioranni CW, Ehrlich MG & Valentini RF (1999) RGD-coated titanium implants stimulate increased bone formation in vivo. Biomaterials 20: 2323-2331.

167. Schaffner P, Meyer J, Dard M, Wenz R, Nies B, Verrier S, Kessler H & Kantlehner M (1999) Induced tissue integration of bone implants by coating with bone selective RGD-peptides in vitro and in vivo studies. J Mater Sci Mater Med 10: 837-839.

168. Willumeit R, Schossig M, Clemens H & Feyerabend F (2007) In-vitro interactions of human chondrocytes and mesenchymal stem cells, and of mouse macrophages with phospholipid-covered metallic implant materials. Eur Cell Mater 13: 11-25.

169. Willumeit R, Schuster A, Iliev P, Linser S & Feyerabend F (2007) Phospholipids as implant coatings. J Mater Sci Mater Med 18: 367-380.

170. Santin M, Rhys-Williams W, O'Reilly J, Davies MC, Shakesheff K, Love WG, Lloyd AW & Denyer SP (2006) Calcium-binding phospholipids as a coating material for implant osteointegration. J R Soc Interface 3: 277-281.

171. Vogt JC, Brandes G, Kruger I, Behrens P, Nolte I, Lenarz T & Stieve M (2008) A comparison of different nanostructured biomaterials in subcutaneous tissue. J Mater Sci Mater Med 19: 2629-2636.

172. Martin HJ, Schulz KH, Bumgardner JD & Walters KB (2007) XPS study on the use of 3-aminopropyltriethoxysilane to bond chitosan to a titanium surface. Langmuir 23: 6645-6651.

173. Bumgardner JD, Chesnutt BM, Yuan Y, Yang Y, Appleford M, Oh S, McLaughlin R, Elder SH & Ong JL (2007) The integration of chitosan-coated titanium in bone: an in vivo study in rabbits. Implant Dent 16: 66-79.

174. Bumgardner JD, Wiser R, Gerard PD, Bergin P, Chestnutt B, Marin M, Ramsey V, Elder SH & Gilbert JA (2003) Chitosan: potential use as a bioactive coating for orthopaedic and craniofacial/dental implants. J Biomater Sci Polym Ed 14: 423-438.

175. Wermelin K, Suska F, Tengvall P, Thomsen P & Aspenberg P (2008) Stainless steel screws coated with bisphosphonates gave stronger fixation and more surrounding bone. Histomorphometry in rats. Bone 42: 365-371.

180

176. Greiner S, Kadow-Romacker A, Wildemann B, Schwabe P & Schmidmaier G (2007) Bisphosphonates incorporated in a poly(D,L-lactide) implant coating inhibit osteoclast like cells in vitro. J Biomed Mater Res A 83: 1184-1191.

177. Greiner S, Kadow-Romacker A, Lubberstedt M, Schmidmaier G & Wildemann B (2007) The effect of zoledronic acid incorporated in a poly(D,L-lactide) implant coating on osteoblasts in vitro. J Biomed Mater Res A 80: 769-775.

178. Peter B, Gauthier O, Laib S, Bujoli B, Guicheux J, Janvier P, van Lenthe GH, Muller R, Zambelli PY, Bouler JM & Pioletti DP (2006) Local delivery of bisphosphonate from coated orthopedic implants increases implants mechanical stability in osteoporotic rats. J Biomed Mater Res A 76: 133-143.

179. Ganguli A, Henderson C, Grant MH, Meikle ST, Lloyd AW & Goldie I (2002) The interactions of bisphosphonates in solution and as coatings on hydroxyapatite with osteoblasts. J Mater Sci Mater Med 13: 923-931.

180. Lawson MC, Bowman CN & Anseth KS (2007) Vancomycin derivative photopolymerized to titanium kills S. epidermidis. Clin Orthop Relat Res 461: 96-105.

181. Darouiche RO, Mansouri MD, Zakarevicz D, Alsharif A & Landon GC (2007) In vivo efficacy of antimicrobial-coated devices. J Bone Joint Surg Am 89: 792-797.

182. Lucke M, Schmidmaier G, Sadoni S, Wildemann B, Schiller R, Haas NP & Raschke M (2003) Gentamicin coating of metallic implants reduces implant-related osteomyelitis in rats. Bone 32: 521-531.

183. Gollwitzer H, Ibrahim K, Meyer H, Mittelmeier W, Busch R & Stemberger A (2003) Antibacterial poly(D,L-lactic acid) coating of medical implants using a biodegradable drug delivery technology. J Antimicrob Chemother 51: 585-591.

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

Distributed byOULU UNIVERSITY LIBRARY

P.O. Box 7500, FI-90014University of Oulu, Finland

Book orders:OULU UNIVERSITY PRESSP.O. Box 8200, FI-90014University of Oulu, Finland

S E R I E S D M E D I C A

958. Heikkinen, Jouni (2008) Outcome after mitral valve surgery for mitral valveregurgitation

959. Papponen, Hinni (2008) The muscle specific chloride channel ClC-1 and myotoniacongenita in Northern Finland

960. Heikura, Ulla (2008) Intellectual disability in the Northern Finland Birth Cohort1986

961. Ylöstalo, Pekka (2008) Dental health, lifestyle and cardiovascular risk factors—astudy among a cohort of young adult population in northern Finland

962. Dunder, Teija (2008) Environment and atopy and asthma in childhood. The effectof dietary fats, common infections and asthma treatment practises on morbidityrates

963. Juvonen, Raija (2008) Respiratory infections and cold exposure in asthmatic andhealthy military conscripts

964. Rysä, Jaana (2008) Gene expression profiling in experimental models of cardiacload

965. Ikäheimo, Pekka (2008) Suomalaisen aikuisen astma – kysely- ja rekisteritutkimusvuonna 2000

966. Junttila, Juhani (2008) Characteristics of subjects with Brugada syndrome typeelectrocardiogram

967. Huilaja, Laura (2008) Collagen XVII and pathomechanisms of junctionalepidermolysis bullosa and gestational pemphigoid

968. Heikka, Helena (2008) Sosiaali- ja terveysjohtajan työn sisältö ja kompetenssit

969. Juntti, Hanna (2008) Association of respiratory syncytial virus infection withasthma and atopic allergy

970. Winqvist, Satu (2008) Alcohol misuse in relation to traumatic brain injury. TheNorthern Finland 1966 birth cohort study

971. Makkonen, Saara (2008) Teknillisestä apulaisesta laboratoriohoitajaksi. Turunlaboratoriohoitajakoulutuksen kehitys vuosina 1955–1990

972. Liljeroos, Mari (2008) Toll-like receptor 2 (TLR2) and TLR4 signaling in the innateresponse against bacterial components

973. Vierimaa, Outi (2008) Multiple Endocrine Neoplasia Type 1 (MEN1) and PituitaryAdenoma Predisposition (PAP) in Northern Finland

D974etukansi.kesken.fm Page 2 Thursday, May 29, 2008 11:44 AM

ABCDEFG

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

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

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

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

EDITORIAL SECRETARY

Professor Mikko Siponen

Professor Harri Mantila

Professor Hannu Heusala

Professor Olli Vuolteenaho

Senior Researcher Eila Estola

Information officer Tiina Pistokoski

Senior Lecturer Seppo Eriksson

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-8833-3 (Paperback)ISBN 978-951-42-8834-0 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

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

MEDICA

ACTAD

OULU 2008

D 974

Virpi Muhonen

BONE–BIOMATERIAL INTERFACETHE EFFECTS OF SURFACE MODIFIEDNiTi SHAPE MEMORY ALLOY ONBONE CELLS AND TISSUE

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

D 974

ACTA

Virpi Muhonen

D974etukansi.kesken.fm Page 1 Thursday, May 29, 2008 11:44 AM