Biological interface of bone graft substitute materials

Umeå University Odontological Dissertation, New Series No 121

Biological interface of bone graft substitute materials:

Experimental studies on interactions between biomaterials and bone cells

Živko Mladenović

Department of Odontology

Umeå 2011

Cover photo: The picture on the front cover illustrates osteoclast derived from a

mouse bone marrow cell culture.

As a dental student I remember instantly becoming fascinated with the osteoclast

when we during a lecture in Oral Pathology were shown a video of an osteoclast in

action. I am very grateful that I have been given the opportunity to work with

these beautiful and fascinating cells. They still put a smile on my face every time I

see them.

Responsible publisher under Swedish law: the Dean of the Medical Faculty

This work is protected by the Swedish Copyright Legislation (Act 1960:729)

ISBN: 978-91-7459-286-3

ISSN: 0345-7532 New Series no 121

The electronic version is available at http://umu.diva-portal.org/

Printed by: Print & Media

Umeå, Sweden 2011

Copyright © Živko Mladenović

To my family and friends

虎穴に入らずんば虎児を得ず。 (Koketsu ni hairazunba koji o ezu)

Nothing ventured, nothing gained.

i

Table of Contents

Table of Contents i Abstract ii List of papers iii Abbreviations iv 1. Introduction 1

1.1 Bone – a short overview 2 1.1.2 Osteoblast 3 1.1.3 Osteoclast 4 1.1.4 Gap junction cell-cell communication 5 1.1.5 Bone healing and bone remodeling 6 1.1.6 Silicon 7

1.2 Bone graft substitute materials 8 1.3 Biological interface 9

2. Aims 10 3. Material and Methods 11

3.1 Animals 12 3.2 Osteoblast culture 12 3.3 Bone graft substitute materials 12 3.4 Dissolution extract medium 12 3.5 Silicon containing medium 13 3.6 ICP-OES 13 3.7 45Ca-isotop labeling 14 3.8 Surface analysis 14 3.9 Mineralization assay 15 3.10 Toluidine blue staining 15 3.11 Cell viability/toxicity 16

3.11.1 Flow cytometry PI 16 3.11.2 Neutral Red uptake assay 17

3.12 Bone resorption 17 3.13 Gene expression 18 3.14 Osteoclast 18

4. Results and Discussion 19 4.1 Experimental setup 19 4.1 Biomaterial-cell interface 19 4.2 Effects of Si on osteoclastogenesis and bone resorption 24

5. Conclusion 25 6. Clinical relevance and future perspectives 27 Acknowledgements 28 References 31 Appendix I

ii

Abstract

Bone graft substitute materials are becoming more common as alternative therapy strategies

when bone defects in patients need to be treated. The interaction between bone cells and

biomaterials occur at the surface of the materials. A great deal is known about the importance of

surface topography and physic-chemical properties of biomaterials. It is also known that cells

require proteins in order to interact with biomaterials. Less is known about how material

properties and proteins interact forming the biological interface that cells will be exposed to,

and that might determine if new bone is formed or not in the patient. The overall aim of the

present thesis was to systematically investigate bone graft substitute material surface reactions

and the interface in order to better understand how biomaterials may promote bone formation.

Bio-Oss (BO) is a commonly used bone graft substitute material in reconstruction of periodontal

and dentoalveolar bone defects. BO is mainly considered to be “osteoconductive”, but we could

show that it does interact with a biological fluid (α-MEM cell culture medium) through

dissolution/precipitation reactions. A significant reduction of calcium and phosphate levels in

the medium was obtained even with low concentrations of BO. A release of silicon from the

material was also demonstrated. An osteogenic response was seen in close contact to the BO

particles when cultured with different types of pre-osteoblastic cells (Paper I). X-Ray

Photoelectron Spectroscopy (XPS) with fast-frozen sample technique was used to further

characterize the surface of BO, Frios Algipore (AP) and 45S5 Bioglass (BG). These three bone

graft substitute materials are used as “model systems”, because they have all demonstrated

newly formed bone on the surface after implantation in patients. From the XPS analysis it can

be concluded that AP and BG acquired a positively charged surface while BO gained a negatively

charged surface. Only AP and BG adsorb organic components (amino acids) from the medium

(Paper II). Next we investigated initial surface reactions and the formation of a biological

interface in the presence of proteins (serum) for the three biomaterials. The major findings were

that in the presence of proteins BO underwent a surface charge reversal, all three biomaterials

adsorbed proteins on the surface and all three biomaterials altered the chemical composition of

the cell culture medium (Paper III-IV). Silicon (Si), which was released from BO as well as from

BG, is interesting in relation to bone health. Positive effects of BG Si dissolution products on

osteoblasts have been reported earlier. In the present study inhibitory interactions of Si on the

RANK/RANKL/OPG signaling pathway as well as with gap junction intercellular

communication in vitro are reported. These new findings implicate that Si could potentially be

beneficial for patients with imbalance in bone remodeling (osteoporosis) and treatments of bone

defects (Paper V). In conclusion, biomaterials of different origins interact with a solution

resembling the extracellular tissue fluid. The dissolution-precipitation reactions are influenced

by the material concentration used and should be taken into consideration when designing

experiments and when biomaterials are used clinically. The presence of proteins will influence

surface reactions, the formation of the biological interface and have implications on cellular

responses. Possible dissolution products from the biomaterials should be investigated. Si, a

dissolution product, is shown to have an inhibitory effect on osteoclastogenesis and bone

resorption in vitro. Potential clinical value of Si in treatment of patients with bone defects

should be further investigated.

iii

List of papers

This thesis is based on the following original papers, which will be referred to

by their Roman numerals:

I. Mladenović, Z., Sahlin-Platt, A., Andersson, B., Johansson, A., Björn, E.

and Ransjö, M. In vitro study of the biological interface of Bio-Oss®:

implications of the experimental setup. Clin Oral Impl Res, 2011; 1-7 doi:

10.1111/j.1600-0501.2011.02334.x

II. Mladenovic, Z., Sahlin-Platt, A., Bengtsson, Å., Ransjö, M. and

Shchukarev, A. Surface characterization of bone graft substitute

materials conditioned in cell culture medium. Surface and Interface

Analysis. 2010; 42: 452–456. doi: 10.1002/sia.3337

III. Mladenović, Ž., Sahlin-Platt, A., Andersson, M., Shchukarev, A. and

Ransjö, M. Investigation of surface reactions and solid-solution

interfaces of three bone graft substitute materials incubated in cell

culture medium. Submitted manuscript

IV. Shchukarev, A., Ransjö, M. and Mladenović Ž. To build or not to build:

The interface of bone graft substitute materials in biological media from

the view point of the cells. In Biomaterials Science and Engineering.

2011; 287-308. Rosario Pignatello (Ed.) IBSN: 978-953-307-609-6, In

Tech, (Open Access)

V. Mladenović, Ž., Johansson, A., Willman, B., Shahabi, K., Björn, E. and

Ransjö, M. Silicon inhibits signaling pathways and cell-cell

communication important for osteoclast formation and bone resorption

in vitro. Manuscript

The original papers are reprinted with the kind permission from the

publishers.

iv

Abbreviations

α-MEM Alpha-Minimum Essential Medium

ALP Alkaline phosphatase

AP Frios Algipore

BG 45S5 Bioglass

BMD Bone mineral density

BO Bio-Oss

Ca Calcium

cAMP Cyclic AMP

Ct Threshold cycle

Cath K Cathepsin K

CTR Calcitonin receptor

Cx Connexin

EFSA European food safety authority

EDX Energy dispersive X-ray analysis

FBS Fetal bovine serum

GJC Gap junction communication

GJIC Gap junction intercellular communication

ICP-OES Inductively coupled plasma optical emission spectrometry

IGF-II Insulin-like growth factor II

ITAM Immunoreceptor tyrosine-based activation motif

M-CSF Macrophage colony-stimulating factor

MMP Matrix metalloproteinase

NFATc1 Nuclear factor of activated T-cells, cytoplasmic 1

NR Neutral red

OPG Osteoprotegerin

OSA Orthosilicic acid

OVX Ovariectomy

RT-PCR Reverse transcriptase polymerase chain reaction

PI Propidium iodide

PLC Phospholipase C

PTH Parathyroid hormone

qRT-PCR Quantitative RT-PCR

RANK Receptor activator of NF-κβ

RANKL Receptor activator of NF-κβ ligand

RGD ArgGlyAsp

v

RUNX2 Runt related transcription factor 2

SBF Simulated body fluid

SEM Scanning electron microscopy

Si Silicon

TB Toluidine blue

TRAP Tartrate resistant acid phosphatase

XPS X-Ray Photoelectron Spectroscopy

1

1. Introduction

Bone graft substitute materials are becoming more common as alternative

therapy strategies when bone defects in patients need to be treated (Fig. 1).

The interaction between bone cells and biomaterials occur at the surface of

the materials. A great deal is known about the importance of surface

topography and physic-chemical properties of biomaterials. It is also known

that cells require proteins in order to interact with biomaterials. Less is

known about how material properties and proteins interact forming the

biological interface that cells will be exposed to, and that might determine if

new bone is formed or not in the patient. The overall aim of the present

thesis was to systematically investigate bone graft substitute material surface

reactions and the interface in order to better understand how biomaterials

may promote bone formation and possibly influence bone remodeling.

Fig 1. Illustration of bone defects in patients later treated with bone graft materials. (a) Cranial reconstruction using adipose-derived stem cells and beta-tricalcium phosphate granules [1]. (b) Treatment of an intrabony periodontal defect using guided tissue regeneration and BO [2] . (c, d) Treatment of benign bone tumors and tumor-like lesions using biphasic bone substitute and fibrin sealant [3]. (Images reprinted with kind permission from copyright owner.)

2

1.1 Bone – a short overview

There are 206 named bones in the human body that make up the skeletal

system [4]. The skeletal system has several functions such as providing a

ridged framework to support the body, protection of inner organs, facilitate

body movement, harboring bone marrow involved in hematopoiesis and act

as a primary reservoir for mineral salts (e.g. calcium and phosphate) [4, 5].

The skeletal bones can be classified either by their shape (long, short, flat or

irregular bones) or be subdivided by the position they have in the skeletal

system. The two subdivisions are the axial and the appendicular skeleton.

The axial consists of the skull, the vertebral column, sternum and ribs,

whereas the appendicular are all other bones (upper and lower extremities)

[4, 5]. Every bone of the skeleton is an organ in itself since it comprises

several different tissues; bone tissue, hematopoietic tissue, nerve tissue,

adipose tissue, blood vessels and sometimes hyaline or articular cartilage [4-

6]. Bone tissue is a specialized form of connective tissue, consisting of bone

cells and a bone matrix (extracellular matrix) that can mineralize. Calcium

and phosphate are stored in the bone as hydroxyapatite-like mineral

[Ca10(PO4)6(OH)2] when newly formed bone matrix is mineralized. In the

bone tissue approximately 65-70% of the dry weight can be accounted to the

inorganic hydroxyapatite. The remaining, organic, 30-35% are comprised of

collagenous (90%; type I collagen) and noncollagenous proteins (10%) [6-9].

It is believed that the noncollagenous proteins are essential to bone and may

influence mineralization processes, cell proliferation, attachment and

activity. The three main groups of noncollagenous proteins are:

proteoglycans, glycoproteins and γ-glutamic acid-containing proteins

(vitamin K-dependent). In addition to the three major structural protein

groups there are also some growth factors and cytokines that can be assigned

to the noncollagenous proteins (e.g. bone morphogenetic proteins) [6, 10].

During embryonic development three distinct embryonic cell linages will

contribute to the formation of the different parts of the skeleton. Neural crest

cells will give rise to the craniofacial skeleton, axial skeleton will be derived

from the sclerotome cells from the somites and lateral plate mesoderm cells

will give rise to the remaining appendicular bones [11, 12]. Two distinct

forms of bone formation, osteogenesis, exists: endochondral ossification and

intramembranous ossification. The two processes are distinguished by

whether or not the ossification is preceded by a cartilage precursor (i.e.

endochondral ossification). Bones formed without previously being modeled

in cartilage (i.e. intramembranous ossification) are the flat bones of the skull

and face, the mandible and the clavicle [12]. Classically three types of bone

cells are defined: osteocytes, osteoblasts and osteoclast. However,

morphologically and histologically sometimes osteoprogenitor cells

(osteoblast precursor cells) and lining cells (inactive surface osteoblasts) are

3

also identified and discussed [6, 13, 14]. Osteocytes are the most common of

all bone cells and often regarded as a differentiated osteoblast embedded in

the mineralized bone matrix. These cells are dendritic in shape, enclosed

within a lacunocanalicular network and connect with neighboring osteocytes

through branching cytoplasmic processes connected via gap junctions

(intercellular transmembrane channels facilitating communication between

adjacent cells)[13, 15, 16].

1.1.2 Osteoblast

Osteoblasts arise from multipotential mesenchymal stem cells capable of

giving rise to a number of cell lineages such as adipocytes, myoblasts, or

chondrocytes [14]. Transcriptional factors control phenotype-specific gene

expression. Studies concerning the development of the osteoblast phenotype,

from the osteoprogenitor proliferative cell to the osteocyte embedded in the

extra cellular matrix, suggest a temporal sequence of differentiation

involving active cell proliferation, expression of osteoblastic markers,

Fig 2. Illustration of MSC differentiation into osteoblasts [17].

(Images reprinted with kind permission from copyright owner.)

4

synthesis, deposition and maturation of a collagenous extra cellular matrix

and, matrix mineralization. Runx2 is an essential transcription factor

necessary for osteoblasts differentiation and transcription of osteoblast-

related genes (e.g. alkaline phosphatase, osteocalcin, type I collagen,

osteopontin and bone sialoprotein). Osterix is another factor controlling

osteogenesis and acts downstream of Runx2 [14, 18]. Osteoblasts produce

both macrophage colony-stimulating factor (M-CSF) and receptor activator

of NF-kB ligand (RANKL), which are critical and necessary factors for

osteoclastogenesis. Osteoblasts/stromal cells also produce osteoprotegerin

(OPG), a soluble decoy receptor for RANKL that inhibits both differentiation

and function of osteoclasts. Thus, osteoclast differentiation, formation, and,

to a lesser degree, activation absolutely depend upon the proximity and

products of the osteoblast [19, 20].

1.1.3 Osteoclast

“The osteoclast remains one of the most complex and fascinating cells of the body. These cells

are destructive yet delicate, short-lived but highly active, modest in numbers but recognized as

the only cell of the body capable of degrading and removing large quantities of bone.” [21].

This introduction by Edwards & Mundy (2011) captures the essence of the

osteoclast. Originating from the hematopoietic cells of the monocyte-

macrophage linage the osteoclast progenitor cells are committed to

becoming active osteoclasts through a series of steps involving regulatory

factors as well as cell-to-cell interactions [22-24].

Fig 3. Illustration of osteoclastogenesis [21]. (Images reprinted with kind permission from copyright owner.)

5

RANKL, OPG and M-CSF are factors crucial in osteoclast differentiation and

activation produced by osteoblasts/stromal cells in response to different pro-

resorptive and calciotropic factors. RANKL binds to receptor activator of NF-

kB (RANK), expressed on osteoclast precursors and mature osteoclasts, and

activates several signaling pathways. One downstream pathway activates

PLC and ITAM-dependent co-stimulatory signals. This cause a rise in

intracellular Ca2+, and induction of the transcription factor NFATc1 and

other factors important in osteoclastogenesis (TRAF6, c-fos) and gene

expression for osteoclast specific markers (tartrate resistant acid

phosphatase (TRAP), calcitonin receptor (CTR), Cathepsin K (Cath K) and

RANK ) [23].

In order for osteoclast to resorb bone, osteoclasts need to bind tightly to the

mineralized bone surface. This binding is mediated by integrin aνβ3

interaction with matrix proteins containing the ArgGlyAsp (RGD) motifs.

The bone resorbing cell is polarized with a ruffled border membrane

exporting protons to the sealed zone, via a proton pump, thus creating a low

pH that can dissolve the mineral component of bone. In addition, osteoclast

contribute to the degradation of the organic part of the matrix by matrix

degrading proteases (Cath K, TRAP and matrix metalloproteinases (MMPs))

[21, 25].

1.1.4 Gap junction cell-cell communication

Bone cells are organized in cellular networks and this bone cell society is

crucially dependent on highly organized signaling systems in order to control

and coordinate the remodeling process. Morphological studies in rat bone

tissue have indicated gap junctions between osteoblasts, osteoblasts-

osteocytes, and between osteocytes [26]. Gap junctions provides a fast

communication system and permit the transit of ions, second messengers

(Ca2+, cAMP) and small molecules (up to 1 kDa) between adjacent cells. A

gap junction hemi channel, connexon, is formed by six gap junction protein

subunits, connexins, (Cx) which docks with a corresponding structure in a

neighboring cell. Gap junctions can be regulated at many levels by factors

such as membrane voltage, pH, phosphorylation state, and biochemical

signals. Gap junctional communication (GJC) plays a crucial role in many

aspects of cell differentiation, growth and proliferation. There is a large body

of evidence implicating that GJC is important in regulation of osteoblast

differentiation and function. A functional coupling between osteoblast-like

cells mainly composed of Cx 43 protein subunits has been demonstrated in

6

osteoblasts. In line with this, Cx 43-null mice exhibits delayed skeletal

mineralization, craniofacial abnormalities and osteoblast dysfunction. Thus,

gap junction coupling seems to be required for osteoblast differentiation and

the formation of bone matrix and mineralization [27, 28]. Far less is known

concerning the role of intercellular gap junction communication in the

regulation of osteoclast recruitment and the bone resorption process.

Ilvesaro et al. demonstrated Cx 43 immunoreactive sites in rat osteoclasts in

vitro and a gap-junction inhibitor decreased the number and size of bone

resorption pits [29]. We have reported expression of Cx 43 mRNA in mouse

bone marrow cultures and in a pure population of micro-isolated osteoclasts.

Moreover, Cx43 mRNA expression was demonstrated in bone resorption

cultures and was up-regulated when bone resorption stimulated with PTH or

vitamin-D3. Furthermore, pharmacological inhibitors of gap junction

intercellular communication clearly reduced the number of resorption pits

formed by osteoclasts incubated on bone slices without any reduction of the

number of TRAP-positive cells or signs of cytotoxic effects [30]. We have also

reported an up-regulation of Cx 43 gene expression in relation to increased

osteoclast formation in mouse bone marrow cultures. Gap junction

communication seems to be important in mediating signals crucial for

differentiation and fusion of osteoclasts. Our results suggest that gap

junction-mediated signals in osteoclastogenesis is acting downstream of

RANK-RANKL [30, 31]. An interesting question is if gap junction-mediated

signals interact with other signaling pathways and molecules important for

bone remodeling. Considering the important role for gap junction

transferred signals in normal osteogenic differentiation, it is suggested that

that GJIC is a platform to modulate craniofacial tissue engineering [32, 33].

In line with this, GJIC could be a novel strategy to control the bone

developmental processes initiated by biomaterials [34, 35].

1.1.5 Bone healing and bone remodeling

Bone is a dynamic and living tissue with a remarkable capacity to repair and

regenerate in response to injury, fractures or during osseointegration. Three

to four consecutive healing phases have been suggested. Stage one involves

hematoma formation and inflammation (minutes-hours). This stage is

followed by the repair stage where a soft callus is formed (within a few days)

and transformed gradually into a hard callus, over a period of up to 8 weeks.

The final stage of healing involves the remodeling of the hard callus into

bone and over a time period of 3-6 months the new bone will gain the same

properties as original bone [4, 9, 36-38].

7

Bone remodeling is a process which involves both bone resorption and bone

formation in a locally coupled and balanced process between osteoblasts and

osteoclasts. Remodeling is important for mineral homeostasis as well as for

repair of bone tissue [12, 20, 39]. The bone remodeling can be described in

five stages: activation, resorption, reversal, formation and termination phase

[20].

1.1.6 Silicon

In 2004, the European food safety authority (EFSA) concluded that the

essentiality and functional role for silicon has not been established due to

insufficient data. However, accumulating evidence strongly suggest that

silicon intake is important for bone formation and skeletal health.

Silicon occurs in nature as silica and silicates which have very low

bioavailability. Dissolution products, mainly orthosilicic acid (OSA;

Si(OH)4), are found in plants and algae and drinking water. OSA is the most

readily bioavailable source of silicon and high levels of OSA are found in

fruits, grains, cereals, rice, seafood and beer [40]. The fasting serum levels of

silicon increase after dietary silicon intake in rats and humans. Although

silicon is eliminated from plasma into urine within 4-8 hr following

ingestion, some of the silicon is likely to be taken up by tissues. Studies in

rats demonstrate high levels of silicon present in bone and other connective

tissues and it is assumed to be the same in humans. Chemical analyses of

vertebrae in mice after 8 weeks feeding with excessive dietary silicon

demonstrated a significant increase in silicon content in bones[41]. The

relation between ingested silicon, plasma levels and accumulation in bone in

humans remains to be studied.

In animals a silicon deprivation result in skeletal defects, thinner legs and

decreased mineral content. Already in 1972, Carlisle and Schwartz reported

that silicon is essential for bone formation in the chick and rat. A short-term

supplementation of soluble silicon improved bone mineral density in rats

with OVX-induced osteoporosis [42]. In the Framingham offspring cohort a

higher intake of dietary silicon was significantly associated with higher bone

mineral density (BMD) in men and in pre-menopausal women. A significant

effect of silicon on BMD was demonstrated also in post-menopausal women

if combined with hormone replacement therapy [40], and, furthermore,

choline-stabilized OSA increased the stimulatory effect of calcium and

vitamin D3 on bone formation markers in osteopenic women [43]. It is not

clear what the mechanism is that mediates the positive effects of silicon on

bone. It is suggested that silicon crosslink extracellular matrix proteins in

8

bone but it is also reported to stimulate osteoblast synthesis of collagen I and

other markers of bone formation. Animal studies have shown that

supplemental Si inhibits bone loss and increases femoral BMD above that of

controls, suggesting that Si may have both anti-resorptive and anabolic

properties [44].

1.2 Bone graft substitute materials

The etiology of a bone defect varies, but regardless if it was acquired as a

congenital deformity, by injury or due to a pathological processes it is a

strain on the patient and on society [45]. It has been suggested that 5-10 % of

all fractures have a delayed or impaired healing [46]. Approximately 2.2

million bone grafting procedures are performed annually worldwide, making

bone the second most commonly transplanted tissue [47]. Harvesting bone

from the patient (autologous bone graft) is still considered the golden

standard [48]. Despite being the treatment of choice the amount of tissue

that can be harvested is limited. It has also been reported that the risk of

complications at the donor site ranges from 8-39% [47]. This has increased

the use of alternatives to autologous bone graft for bone repair such as

hydroxyapatite of natural origin (algae - and bovine bone-derived) and

synthetic bioceramic (calcium phosphate and bioactive glass) materials [49].

Bioactive glasses are silica-based ceramic materials that have different

bioactivity and bone bonding properties depending on chemical

composition. Bioactive glass 45S5 (BG) has been used in different shapes

and for reconstruction of craniofacial and periodontal bone defects. Biopsies

from maxillary sinus elevation with BG showed that the glass particles

became excavated and their centers gradually filled with bone tissue. All BG-

particles had disappeared 16 months after grafting and was completely

replaced by bone tissue. BG is known to release ionic dissolution products

(Na, Ca and Si) when incubated in a solution [50, 51]. Using the conditioned

medium containing the dissolution products Xynos et al (2000) could

demonstrate induction of insulin-like growth factor II (IGF-II) mRNA

expression in osteoblasts [52]. It was later also demonstrated that using BG

conditioned medium induced protein synthesis and formation of mineralized

bone nodules. The authors postulated that the effect seen of the conditioned

medium was due to the increase of Si ions in the medium [53]. Since silicon

is a major dissolution product from BG, it has been proposed to play a

critical role for the osteogenic effects of BG [54-57]. Several other inorganic

ions, which might be released as dissolution products from biomaterials,

9

have been suggested to be important for bone repair (e.g. magnesium,

strontium, zinc, copper, cobalt and boron) [58-60].

1.3 Biological interface

After a material is implanted in a bone defect a solid-solution interface is

rapidly established between the biomaterial surface and the surrounding

solution. The formation of a biological interface will start when protein are

adsorbed on the material surface. The presence of proteins is crucial for the

material-cell interaction(s) and the following attachment of cells [61, 62].

Interface properties (through physicochemical processes, surface topography

and nano-scale structures) will regulate cell functions and gene expression

[63-65].

Fig 4. Illustration of a solid-solution interface and a biological interface

Since bone strength is related to coordinated formation and resorption in a

physiological adaptation it is important to understand how the biomaterial

may regulate both bone forming osteoblasts as well as bone resorbing

osteoclasts. Before the material can be used for clinical applications, it

should have been studied in vitro to examine the biological responses. In

order to relate the results from in vitro experiments to the clinical situation,

it is crucial to consider the experimental setup, cell type and analyzing

techniques carefully when investigating cell/biomaterial interactions [34, 65-

68].

10

2. Aims

The overall aim was to systematically investigate bone graft substitute

material surface reactions in a biological medium in order to better

understand how biomaterials interact with bone cells activities and bone

remodeling.

Specific aims of the present thesis were:

To systematically investigate initial surface and interface reactions of

three bone graft substitute materials (BG, BO and AP) using an in vitro

model system based on a biological medium with or without the addition

of serum proteins (FBS).

To examine the biological interface of BO and the osteogenic response

when osteoblast-like cells were added to the in vitro model system

To study the solid-solution interfaces of bone graft substitute materials,

in the in vitro model system based on cell culture medium, using XPS

with fast-frozen technique.

To investigate effects on the biological interface when serum (proteins)

was added to the in vitro model system based on cell culture medium,

using XPS with fast-frozen technique.

To study the in vitro effects of BG ionic dissolution product and Si on

osteoclastogenesis and bone resorption and signaling pathways

important for osteoclastogenesis and bone resorption.

11

3. Material and Methods

The material and methods section has deliberately been kept short since the

methods and materials used have been more comprehensively described in

each paper.

Table 1. Summary of methods used

Method What is detected/ Used for Paper

ICP-OES Trace elements I,III,V

45Ca-isotop labeling Calcium isotope I,III

Osteoblast cultures

(a) Mouse marrow stromal cells

(b) Rat calvarial cells

(c) MC3T3T-E1

Mineralization assay

Cell morphology

I

Mineralization assay Nodule formation I

Toluidine blue staining Cell morphology I

von Kossa staining Phosphate binding I

Alizarin red S staining Calcium binding I

Light microscopy Cell morphology I,V

BET measurements Surface area and pore size II, IV

XPS (fast-frozen technique)

Surface chemical composition

Surface charge

Biological interface

II,III, IV

SEM/EDX Surface morphology

Surface chemical composition

III

RNA isolation, cDNA synthesis and

qPCR

Gene expression V

TRAP staining TRAP positive multinucleated cells V

Flow cytometry (Propidium Iodide) Toxicity (necrosis/apoptosis) V

Neutral red uptake assay Cell viability/toxicity V

Bone resorption assay (45Ca) 45Ca-release V

Osteoclast formation assays

(a) Mouse bone marrow cultures

(b) RAW264.7 cells

Osteoclast-like multinucleated cells

V

12

3.1 Animals

CsA mice from our own inbred colony were used in some of the experiments.

The animals were kept at the university animal facilities and any experiment

involving an animal was conducted after the animal had been sacrificed.

Animal care and experiments were approved by the Animal Ethics

Committee at Umeå University (approval no. A27-07 and A40-10) and are

also in accordance with European Communities Council Directive

(86/609/EEC).

3.2 Osteoblast culture

Mesenchymal stem cells differentiate into mature fully developed osteoblasts

that can produce and mineralize bone matrix. Gene expression will thus vary

with differentiation stages of the cells. Three differentiation stages of

osteoblast-like cells were used: mouse marrow stromal cells, primary rat

calvarial cells and MC3T3-E1 mouse pre-osteoblast cell line.

3.3 Bone graft substitute materials

Three bone graft substitute materials were used as model system materials in

several of our studies. The three bone graft substitute materials used were: a

synthetic (45S5 Bioglass®), a natural bovine derived (Bio-Oss®) and an algae

derived material (Frios® Algipore®)

3.4 Dissolution extract medium

Samples for ICP-OES analysis were prepared by weighing 0.005, 0.025,

0.05, 0.1, 0.25, 0.5 g of Bio-Oss respectively in test tubes before adding 5 mL

of α-MEM. These sample concentrations correspond to 0.1, 0.5, 1.0, 2.0, 5.0

and 10.0% weight /volume. Samples were incubated in 37°C for 24 h, 72 h or

168h and thereafter the solutions were pipetted to new test tubes. Final

solutions were analyzed with Optima 2000 DV ICP-optical emission

spectrometer.

13

3.5 Silicon containing medium

Silicon containing medium was prepared by modifying a method previously

published by Sripanyakorn et al. [69]. The cell culture medium did not

contain any measurable amounts of silicon. At all times was the silicon

containing medium prepared and kept in plastic vessels (i.e. no glass vessels

were used for the silicon containing medium, due to the risk of

contamination). Concentrated sodium silicate solution was used to prepare

a stock solution containing 350 µg/mL of Si (0.1 mL of concentrated sodium

silicate solution to 49.9 mL of α-MEM). The pH was adjusted to 7.0-7.2

using 6M HCl before 10% FBS was added. Final concentrations of 0-200

µg/mL Si solutions were prepared by diluting the stock solution with α-MEM

+ 10% FBS. Silicon concentrations were confirmed using ICP-OES analysis.

3.6 ICP-OES

Spectra for emission wavelengths (nm) Calcium (Ca) 317.933, Phosphorous

(P) 213.617, Silicon (Si) 251.61, Sodium (Na) 330.273 and 589.592 were

recorded. Five different standard solutions containing 0, 1, 10, 50 and 100

μg/mL Ca, P, Si and 0, 20, 200, 1000 and 2000 μg/mL Na were prepared in

order to obtain calibration curves.

Fig 5. a) ICP-OES instrument. b) The plasma torch. c) Ionization of sample. d) Element specific light emission. e) Concentration of sample correlated to the standard curve.

14

In all standard solutions 10% α-MEM was added to resemble the sample

solution matrix. The stability of the calibration curve was assured by

periodical remeasurement of the standard solutions during analysis.

3.7 45Ca-isotop labeling

Calcium precipitation on material surface were assessed using an calcium-

45 isotope labeled medium (+10% FBS).

Fig 6. a) Biomaterial is added to the Calcium-45 containing medium. b) Calcium-45 precipitated on the material and the amount of isotope remaining in the medium is analyzed using a liquid scintillation system.

3.8 Surface analysis

There are number of different techniques available to study a surface.

Several properties of the surface (e.g. topography, chemical composition and

structure) are needed to understand surface behavior. No single technique

can provide all this information [70]. We have used Scanning Electron

Microscopy (SEM) give an image of the surface and elements can be detected

using Energy Dispersive X-ray (EDX) analysis. However, due the penetration

depth of the electron beam signals from the bulk will also be detected. EDX

is thus not always regarded as a surface analysis method [71]. X-ray

photoelectron Spectroscopy (XPS) was used as the principle technique to

study the material surfaces. Using the fast-frozen technique the solid-

solution interface could also be analyzed [72, 73].

15

Fig 7. Illustration of different methods used to study a surface [74]. (Images reprinted with kind permission from copyright owner.)

3.9 Mineralization assay

Mineralization was detected using Von Kossa staining and Alizarin red S

staining. Von Kossa detects phosphate and Alizarin red S was used for

calcium detection.

3.10 Toluidine blue staining

Toluidine blue (TB), also known as tolonium chloride, is a commonly used

dye in histology. TB is a basic (cationic) dye and reacts with anionic

components of cells and tissue. An acidic dye has a net negative charge on

the dye portion of the molecule and can be described by the general formula

[Na+dye-]. The basic dye has a net positive charge on the dye portion of the

molecule and can be described by the general formula [dye+Cl-] [75].

16

Fig. 8 MC3T3-E1 cells stained with Toluidine blue dye. Cells as well as cell nuclei are clearly outlined. Original magnification 40x.

TB can bind to anionic components of nucleic acid (phosphate groups),

proteoglycans and glycosaminoglycans (sulfate groups) as well as proteins

(carboxyl groups). The reaction of a basic dye is dependent on what anionic

groups are ionized and available for chemical reactions with the basic dye

and the availability of the anionic groups are pH dependent [75-77].

3.11 Cell viability/toxicity

A number of different methods exist to analyze cell viability/toxicity.

Common approaches can be to demonstrate metabolic activity and/or cell

membrane integrity as a prof of viable cells. Cell toxicity/ cell death could on

the other hand be proven by leakage of dyes into a cell or leakage of

cytoplasmic molecules out from a cell. We have used two methods: 1) flow

cytometry and Propidium Iodide (PI) labeling to assess cell toxicity and 2)

NR assay for viability/proliferation.

3.11.1 Flow cytometry PI

In flow cytometry each cell is passed through one or more laser beams

producing a light scatter that is detected. This means that for each cell data

will be collected regarding size (forward scatter), granulation (side scatter)

and if labeled with fluorochromes information about surface molecules, DNA

contents, proteins and intracellular activities can be detected [78]. PI

labeling is a well-established method for analyzing apoptosis. PI is a

fluorochrome that can only enter a disrupted cell membrane and then binds

to DNA [79]. The FL3 channel was used to detect PI binding to DNA.

17

3.11.2 Neutral Red uptake assay

Neutral Red (NR) uptake assay is also a commonly used method to assess

cell viability/cell toxicity. NR is a week cationic dye that can penetrate the

cell membrane by passive diffusion. The dye binds to lysosomes in the cell.

Once inside the lysosome the dye acquires a charge due to the internal

lysosomal pH and is only released if the pH gradient is reduces, as in the case

of cell death. It has been demonstrated that the amount of dye in a cell is

proportional to the number of cells [80]. The amount of dye in the cells is

quantified spectrophotometrically.

3.12 Bone resorption

Bone resorption was studied in a mouse calvarial model system. Release of

calcium-45 isotope from the pre-labeled bones to the medium was used as a

measure of osteoclast bone resorption activity.

Fig. 9 a) Calcium-45 prelabeld calvaria were dissected and cut into smaller pieces. b) After culturing the calvarials the amount of isotope released into the medium from the bones was assessed using a liquid scintillation system.

18

3.13 Gene expression

Using quantitative real-time RT-PCR mRNA expression important for

osteoclastogenesis as well as cell-cell communication was detected. Using the

relative standard curve method mRNA expression was computed from target

Ct values and β-actin values.

3.14 Osteoclast

Mouse bone marrow cells as well as RAW264.7 cell line was used to study

osteoclastogenesis. Multinucleated (>3 nuclei) and TRAP positive cells were

counted as osteoclast. Previously we have shown that multinucleated TRAP-

positive cells formed in the mouse marrow cultures can resorb bone slides

and show phenotypic markers (e.g. CTR and Cath K) for differentiated

osteoclasts [81].

Fig. 10 Mouse bone marrow isolation, multinucleated TRAP positive cells and phenotypic markers (e.g. CTR, Cath, K, TRAP and RANK). Pictures courtesy of Dr. Maria Ransjö

19

4. Results and Discussion

4.1 Experimental setup

To understand why and how bone graft substitute materials facilitate or

promote new bone formation in this complex environment it is important to

study materials effects on the biological environment. It is equally important

to study the environmental effects on the materials itself. Bone formation,

bone remodeling and osseointegration are all processes that depend on

cellular activities. The interactions are very complex and several parameters

influencing the bone formation have been suggested or described (e.g.

chemical and physical properties of the material, surface topography and cell

type). [61, 67, 82]. Physiological bone remodeling is dependent on an

intricate balance between the bone-forming cells and the bone-resorbing

cells (osteoclasts) as well as interactions with osteocytes. Thus, an important

question is how biomaterials may influence the remodeling process. All

cellular activities (cell adhesion, proliferation, differentiation and cell-cell

interactions) are regulated by systemic and local factors, signaling pathways

and cell-cell communication. We have aimed to systematically investigate

early surface reactions of three bone graft substitute materials using an in

vitro model system based on a biological medium. From our results we can

conclude that BG, BO and AP all altered the composition of the surrounding

solution (cell culture medium). This change in ion concentrations is

necessary to take into consideration when designing experimental system

with biomaterials and cells. We believe it is important to study material-cell

interactions require that all analyses are performed in a milieu where cells

can be cultured. In doing so the results obtained would be more relevant for

comparisons with the in vivo situation. The schematic illustrations presented

below are to be regarded as hypothetical and conceptual ideas of actions

occurring at different time points in order to give an overview of our results

in relation to what is already known.

4.1 Biomaterial-cell interface

Our model system is based on three clinically used and commercially

available bone graft substitute materials (BG, BO, and AP). The rationality

for the choice of materials is that size, physical and chemical properties are

known. An even more important reason is that all the materials have

demonstrated new bone formation in vivo [55, 83, 84]. We have used BG as

a reference material since it has a well-defined chemical composition and

has been extensively studied with regard to surface reactions [51]. The initial

20

step in our model system was to investigate material behavior in α-MEM cell

culture medium. In order to draw conclusions from the cell-biomaterial

studies, surface reactions of the materials need to be studied in the same

medium/solution as cell cultures. Material surfaces of BG, BO and AP (1%

w/v) “as received”, after 1 day incubation or after 7 days of incubation in α-

MEM (without FBS) were analyzed with SEM/EDX. SEM micrographs of BG

demonstrated a gradual loss of surface roughness after incubation in cell

culture medium. A Ca/P ratio, similar to what has previously been reported

for BG [85], could also be detected using EDX analysis.

Fig 11. Concept illustration of initial solid-solution reactions after bone graft substitute

material implantation.

Formation of a Ca-P layer on the BG surface has also been reported using

SBF [51, 86, 87]. Apatite layer formation on a biomaterial surface in SBF has

broadly been accepted as a materials ability to bind to living bone. This

binding ability in vitro is used as a predictor of a materials “bioactivity” in

vivo [88]. Although tempting to use a method with the outcome measure

apatite layer formation or not, it would be to oversimplify the complexity of

the in vivo situation. SBF is lacking organic components and it has been

demonstrated that adding proteins to the solution will delay the formation of

the Ca-P layer [89-92]. Questions of how useful the SBF tests actually are

have been raised [93, 94]. The authors to the article “Apatite-formation

ability – Predictor of „„bioactivity”?” take the discussion one step further

when criticizing the use of SBF and write:

21

“Consequently, although apatitic mineral is the principal component of bone and tooth, a non biological prediction of bioactivity in vivo is essentially meaningless”. /…./ “However, even adding growth factors to SBF would be pointless in the absence of osteoblasts. Thus, the entire concept appears to be ill-conceived, being both chemically and biologically irrelevant to the

main question.”[95]

Material testing in SBF might have its merits. However, if the “true”

material-cell interactions are to be understood analyses have to be carried

out in a relevant system and cells need to be present. The concept of study

relevance for in vivo models has also been discussed for the relationship

between material architecture and biological responses and a new approach

in six steps was suggested by Bohner et al [96]. This concept is well in

agreement with what we have suggested.

Our SEM/EDX analysis of BO and AP did not demonstrate any major

changes of the surface morphology or chemical composition when studied

for up to 7 days in α-MEM. However, detecting small changes in surface Ca

and P concentrations of materials with a calcium phosphate bulk can prove

to be somewhat problematic due to the probing depth the EDX data

originate from. The formation of a Ca-P layer on the BG surface has been

studied comprehensively and at least five dissolution-precipitation reaction

stages have been described [50, 97]. To our knowledge, dissolution-

precipitation reactions for BO and AP have not been studied earlier. Our

results demonstrate time and does-dependent changes of ions in the cell

culture medium after incubation with the materials. This finding is

interesting since a number of in vitro studies on BO have reported negative

effects of the material regarding cellular activities (e.g. proliferation,

attachment and gene expressions) [98-101]. It is also problematic that the

amount of material used in different studies is not always reported, which

makes it hard to draw any conclusions about the results reported. Si release

from BO was a surprising finding and might reflect that BO has a biological

origin. Had we not actively been monitoring for Si ions, then we might not

have detected it. This raises the question if any other ions of importance for

bone function are released and what clinical relevance this might have. Next

step in our model system was to add serum and cells.

22

Fig 12. Cell-material interaction. Proteins are needed for cells to attach. Integrin-RGD mediated signaling, cell-cell communication, as well as local and systemic factors will initiate cellular gene expressions.

Fig 13. Formation of bone matrix

When material effects (dissolution-precipitation reactions) on the

experimental setup were taken into account an osteogenic response could be

detected in close contact with the BO particles. It can be concluded that the

23

amount of material used in the experimental model will be a key parameter

for the in vitro results obtained.

Fig 14. Osteoid formation and initiation of the mineralization process.

XPS analyses with fast frozen technique demonstrate that BG and AP acquire

a positive surface charge and adsorb organic molecules from the cell culture

medium. BO on the other hand gained a negative surface charge and no

organic components could be detected on the surface. Adding 10% serum

resulted in a charge reversal for BO and adsorption of organic molecules

despite differences in chemical composition and surface structure. Protein

conformation and orientation at the interface could determine what part of

the protein that will be exposed to cells. Serum proteins, such as vitronectin

and fibronectin, contain the RDG sequence needed for cell attachment.

RGD-Integrin mediated attachments are known to induce intra cellular

signaling [102-105]. It should also be further investigated if the protein layer

in itself could function as barrier regulating the interaction at the interface.

Surface properties such as topography might influence the protein

adsorption and conformation. It has also been demonstrated that surface

topography can influence gap junction communication which could also

depend on cytoskeletal tensions [34, 106]. The osteogenic response seen in

close contact with BO (demonstrated with Von Kossa and Alizarin red S)

could be due to surface properties and protein adsorption.

24

4.2 Effects of Si on osteoclastogenesis and bone resorption

Earlier studies have mainly focused on effects of dietary Si on bone health

and in relation to bone formation and osteoblast functions [44, 107-109].

There are results suggesting that Si might suppress bone resorption and

number of osteoclast in ovariectomized rats [110-112]. However, direct

effects of Si on osteoclasts formation and functions have not been reported

earlier. In our study we demonstrated that BG dissolution extract inhibits

osteoclastic bone resorption in an organ culture system. The BG extract also

inhibited osteoclast formation in the mouse bone marrow system as well as

in the RAW264.7 cell line. Although Si was released from the BG to the

medium it cannot be excluded that changes in pH, Ca and P ion

concentrations could have contributed to the inhibitory effects on bone

resorption and osteoclast formation. To investigate if the inhibitory effects

seen with the BG extract were caused by Si, media with known

concentrations of the ion were prepared. The Si containing medium

inhibited bone resorption and osteoclast formation. Results suggest that Si

acts on osteoclast precursors through interaction with the

RANK/RANKL/OPG signaling pathway and that gap junction intracellular

communication could also be involved. Si concentrations used were

nontoxic. Si releasing materials that promote osteoblast bone formation and

inhibits osteoclast formation and function might be of clinical relevance at

an initial stage. Si is not considered to be an essential trace element for

human health today. However, should molecular mechanism for Si on bone

cells be established it could prove to be a cost effective way of treating bone

loss in patients.

25

5. Conclusion The overall conclusion is that the three bone graft substitute materials (BG,

BO and AP), despite different origin and chemical composition, all interact

with surrounding solution through dissolution/precipitation reactions. The

formation of a biological interface is dependent on the chemical composition

of the solution used as well as on the presence of serum proteins. Ionic

dissolution products released from the bone graft substitute materials can

also interact with the surroundings. To understand cellular/biological

responses it is important to choose a biologically relevant experimental

model system. It is necessary to study bone graft substitute materials both in

vivo and in vitro in order to understand how the materials interact with

bone cells and bone remodeling.

From the results in the thesis it can be concluded:

The experimental setup has a profound impact on the formation of the

biological interface. BG, BO and AP interact with surrounding solutions

through dissolution- precipitation reactions. Dose and time-dependent

behavior of bone graft substitute materials should be examined in a

biologically relevant model system prior to performing studies with bone

cells.

BO, generally considered as being osteoconductive, interacted through

dissolution-precipitation reactions with the surrounding solution. Si was

released from BO, but the concentrations varied and were batch

dependent. An osteogenic response could be obtained in the presence of

osteoblast-like cells. A key factor for results obtained in an experimental

study is the concentration of the bone graft substitute material used, and

might also be of importance when treating bone defects in patients

XPS with fast-frozen technique is a useful method to acquire additional

information of chemical interactions occurring at the solid-solution

interface of bone graft substitute materials.

Serum proteins adsorbed on the three bone graft materials and in the

case of BO cause a surface charge reversal at the interface. Proteins

adsorbed on the biomaterial might not only be essential for cell

attachment, proliferation and differentiation, but could also function as

a barrier regulating material behavior in a solution. Proteins are an

integral part of the biological interface and should not be omitted in

26

experimental studies where bone cell responses to biomaterials are to be

studied.

BG dissolution extract and particles had an inhibitory effect on

osteoclast formation and bone resorption. Results suggest that the

inhibitory effects of BG/extract were caused by the release of Si from the

material. However, it cannot be excluded that the inhibitory effects were

caused by change of pH and/or concentrations of Ca and P in the cell

culture medium.

Si, added to the cell culture medium, inhibited osteoclast formation and

bone resorption. Results demonstrate that the inhibitory effects of Si are

related to the RANK/RANKL/OPG signaling pathway as well as gap

junction intercellular communication. Direct effects of Si on osteoclast

precursor are suggested, although regulation via osteoblasts cannot be

excluded.

27

6. Clinical relevance and future perspectives

From a clinical perspective:

Handling of the material before implantation might be of importance to the

biological response.

It is important that the practitioner has a solid and sound knowledge of what

can be expected of a bone graft substitute material after implantation in a

patient.

A good understanding of the biological effects which bone graft substitute

could exert on the surrounding tissues and cells could improve treatment

outcome.

Future perspectives:

Future studies should address the identification of the specific proteins

species adsorbed as well as protein confirmation in relation to material

surface properties.

Further studies need to be conducted to establish the molecular mechanisms

for the Si actions on bone formation, bone resorption as well as bone

remodeling process.

28

Acknowledgements

Science is serious business. A lot of time and effort, devotion, patience,

meticulousness and money is put into the search for new knowledge.

Nevertheless, science is also about passion, curiosity, opened mindness,

willingness to share knowledge and “playing” in the lab. I owe my deepest

gratitude to all the people that have been a part of this thesis in one way or

another. In particular I particularly would like to acknowledge the following

persons:

Associate Prof. Maria Ransjö, Supervisor. Thank you for not only being

my supervisor, but also a mentor in its true meaning. It has been an honor

and a privilege to be your PhD-student. You are truly and inspiring person

with your knowledge, passion for science, curiosity, dedication and strive for

perfection. Every smile I get from seeing an osteoclat I owe you. I am grateful

for all the chances and opportunities you have given me and will always be in

debt.

Associate Prof. Anders Johansson, Co-Supervisor. Thank you for being

my co-supervisor and co-author. It has been inspiring working with you and

I am thankful for all methodological knowledge you have shared with me.

Your door has always been open to me for which I am grateful.

Dr .Britta Willman, Co-Supervisor. Thank you for being my co-supervisor

and co-author. Thank you for the many laughs we have had in the lab and for

always caring.

Dr. Annika Sahlin-Platt, Co-Author. Thank you for being my co-author,

former fellow PhD-student, and dear friend. It has truly been a journey with

many unexpected turns. Thank you for all the adventures and the good

laughs.

Associate Prof. Andrey Shchukarev, Co-Author. Thank you for being

my co-author. You contribution to the “interface” has been substantial. I am

grateful for all the help and discussions about XPS and surfaces.

Associate Prof. Erik Björn, Co-Author. Thank you for being my co-

author. I am grateful for all the help and discussions about the ICP-OES.

Associate Prof. Martin Andersson, Co-Author. Thank you for being my

co-author. I am grateful for all the help and discussions about the

SEM/EDX.

29

Kaveh Shahabi, Co-Author. Thank you for being my co-author and good

friend. Keep up the good work and passion for science.

Dr. Åsa Bengtsson, Co-Author. Thank you for being my co-author.

Prof. Jan van Dijken, Examiner. Thank you for being my examiner and

for the interest you have shown for this project.

Lillemor Hägglund. You are devoted, helpful and every PhD-students best

friend in the jungle of administrative papers. I am thankful for all the

administrative help I have received.

Anita Lie. Thank you for always taking your time to help me, to answer any

questions about laboratory work I have had and for your friendship.

Forskarvåningen: Prof. Ulf Lerner, Associate Prof. Pernilla

Lundberg, Birgit Andertun, Ingrid Boström, Dr. Rolf Claesson,

Dr. Carola Höglund-Åberg, Dr. Ali Kassem, Dr. Cecilia Koskinen,

Dr. Malin Larsson Brundin, Inger Lundgren, Dr. Emelie Persson,

Chrissie Roth, Fredrik Strålberg. Thank you for discussions, good

laughs and friendship.

Biomaterialgruppen: Dr. Ingrid Andersson Wenckert, Associate

Prof. Berit Ardlin, Dr. Anders Berglund, Dr. Ulrika Funegård,

Prof. Margareta Molin Thoren, Dr. Lena Mårell, Dr. Karin

Sunnegårdh-Grönberg, Dr. Per Tidehag, Associate Prof. Göran

Sjögren and Associate Prof. Ylva-Britt Wahlin. Thank you for many

good scientific discussions.

Hjördis Olsson and Ann-Sofie Strandberg. Thank you for all the

administrative help.

I am thankful for all the friendship, discussions, and social activities over the

years with past and current fellow PhD-students not earlier mentioned. Dr.

Georgios Belibasakis, Monica Brage, Dr. Anna Brechter, Dr. Karin

Danielsson, Dr. Patrik Danielson, Vincy Eklöf, Dr. Maria Garoff,

Dr. Susanne Granholm, Nina Gennebäck, Lisa Harryson, Dr.

Pamela Hasslöf, Dr. Anna Karin Hulterström, Karin Håberg, Dr.

Anders Johansson, Dr. Eleni Kanasi, Dr. Wen Kou, Kristina

Lindvall, Dr. Stephen Matemba, Angelica Olsson, Dr. Py

Palmqvist, Dr. Emma Persson, Dr. Mattias Pettersson, Matilda

Rentoft, Dr. Pedro Souza, Dr. Ann Sörlin, Bernard Thay, Dr. Nelly

30

Romani Vestman, Katarina Wikén Albertsson, Dr. Daniel Öhlund

and Dr. Catharina Österlund.

I would also like to thank my family and friends. You are all very drear to me

and although maybe not mentioned you are not forgotten. However, there

are some people I do want to mention. Marcus E, Ana K, Jacob B,

Tomas, A, Martin C, Bronius R, Erik J, Emelie SN, Jonas N, Anto

R, Dario S, Mats A, David von B, Gazelle RK and Andria K. Thank

you for reminding me that there is a life outside of the lab as well.

Thank you to all my friends at IKSU Kampsport for many good laughs and

fights. You have been my “extended” family here in Umeå.

To my mother Jadranka, father Milovan, sister Slavica and brother

Saša. Thank you very much for all the love and support. Пуно вас волим.

Ви сте ми све!

Finaly I would just like to say ”Kanelbullen, jag älskar dig”.

31

References

[1] Thesleff T, Lehtimaki K, Niskakangas T, Mannerstrom B, Miettinen S,

Suuronen R, et al. Cranioplasty with adipose-derived stem cells and biomaterial: a novel method for cranial reconstruction. Neurosurgery. 2011;68:1535-40.

[2] Stavropoulos A, Karring T. Guided tissue regeneration combined with a deproteinized bovine bone mineral (Bio-Oss) in the treatment of intrabony periodontal defects: 6-year results from a randomized-controlled clinical trial. Journal of clinical periodontology. 2010;37:200-10.

[3] Reppenhagen S, Reichert JC, Rackwitz L, Rudert M, Raab P, Daculsi G, et al. Biphasic bone substitute and fibrin sealant for treatment of benign bone tumours and tumour-like lesions. International orthopaedics. 2011.

[4] Marieb EN, Mallatt J, Wilhelm PB. Human anatomy. 5th ed. San

Francisco ; London: Pearson Benjamin Cummings; 2008.

[5] Andrew C K. Chapter 3 - Embryonic Development of Bone and Regulation of Intramembranous and Endochondral Bone Formation. In: John PB, Lawrence GR, T. John MartinA2 - John P. Bilezikian LGR, Martin TJ, editors. Principles of Bone Biology (Third Edition). San Diego: Academic Press; 2008. p. 53-84.

[6] Ross MH, Pawlina W. Histology : a text and atlas : with correlated cell and molecular biology. 5th ed. Baltimore, MD: Lippincott Wiliams & Wilkins; 2006.

[7] Weatherholt AM, Fuchs RK, Warden SJ. Specialized Connective Tissue:

Bone, the Structural Framework of the Upper Extremity. Journal of Hand Therapy.

[8] Datta HK, Ng WF, Walker JA, Tuck SP, Varanasi SS. The cell biology of bone metabolism. Journal of clinical pathology. 2008;61:577-87.

[9] Kalfas IH. Principles of bone healing. Neurosurgical focus. 2001;10:E1.

[10] Pamela GR. Chapter 17 - Noncollagenous Bone Matrix Proteins. In:

John PB, Lawrence GR, T. John MartinA2 - John P. Bilezikian LGR, Martin TJ, editors. Principles of Bone Biology (Third Edition). San Diego: Academic Press; 2008. p. 335-49.

[11] Bilezikian JP, Raisz LG, Martin TJ. Principles of bone biology. 3rd ed. Amsterdam ; Boston: Elsevier/Academic Process; 2008.

32

[12] Teti A. Bone development: overview of bone cells and signaling. Current osteoporosis reports. 2011;9:264-73.

[13] Rochefort GY, Pallu S, Benhamou CL. Osteocyte: the unrecognized side

of bone tissue. Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 2010;21:1457-69.

[14] Jane E A. Chapter 4 - Mesenchymal Stem Cells and Osteoblast Differentiation. In: John PB, Lawrence GR, T. John MartinA2 - John P. Bilezikian LGR, Martin TJ, editors. Principles of Bone Biology (Third Edition). San Diego: Academic Press; 2008. p. 85-107.

[15] Bonewald LF. The Amazing Osteocyte. Journal of Bone and Mineral Research. 2011;26:229-38.

[16] Klein-Nulend J, Bonewald L. Chapter 8 - The Osteocyte. In: John PB,

Lawrence GR, T. John MartinA2 - John P. Bilezikian LGR, Martin TJ, editors. Principles of Bone Biology (Third Edition). San Diego: Academic Press; 2008. p. 153-74.

[17] Harada S, Rodan GA. Control of osteoblast function and regulation of bone mass. Nature. 2003;423:349-55.

[18] Marie PJ. Transcription factors controlling osteoblastogenesis. Archives of biochemistry and biophysics. 2008;473:98-105.

[19] Matsuo K, Irie N. Osteoclast-osteoblast communication. Archives of

biochemistry and biophysics. 2008;473:201-9.

[20] Raggatt LJ, Partridge NC. Cellular and molecular mechanisms of bone remodeling. The Journal of biological chemistry. 2010;285:25103-8.

[21] Edwards JR, Mundy GR. Advances in osteoclast biology: old findings and new insights from mouse models. Nature reviews Rheumatology. 2011;7:235-43.

[22] Crockett JC, Rogers MJ, Coxon FP, Hocking LJ, Helfrich MH. Bone

remodelling at a glance. J Cell Sci. 2011;124:991-8.

[23] Takahashi N, Udagawa N, Kobayashi Y, Takami M, Martin TJ, Suda T. Chapter 9 - Osteoclast Generation. In: John PB, Lawrence GR, T. John MartinA2 - John P. Bilezikian LGR, Martin TJ, editors. Principles of Bone Biology (Third Edition). San Diego: Academic Press; 2008. p. 175-92.

33

[24] Shinohara M, Takayanagi H. Novel osteoclast signaling mechanisms. Current osteoporosis reports. 2007;5:67-72.

[25] Väänänen HK, Zhao H. Chapter 10 - Osteoclast Function: Biology and

Mechanisms. In: John PB, Lawrence GR, T. John MartinA2 - John P. Bilezikian LGR, Martin TJ, editors. Principles of Bone Biology (Third Edition). San Diego: Academic Press; 2008. p. 193-209.

[26] Doty SB. Morphological Evidence of Gap-Junctions between Bone-Cells. Calcif Tissue Int. 1981;33:509-12.

[27] Watkins M, Grimston SK, Norris JY, Guillotin B, Shaw A, Beniash E, et al. Osteoblast connexin43 modulates skeletal architecture by regulating both arms of bone remodeling. Molecular Biology of the Cell. 2011;22:1240-51.

[28] Lecanda F, Warlow PM, Sheikh S, Furlan F, Steinberg TH, Civitelli R.

Connexin43 deficiency causes delayed ossification, craniofacial abnormalities, and osteoblast dysfunction. The Journal of cell biology. 2000;151:931-44.

[29] Ilvesaro J, Vaananen K, Tuukkanen J. Bone-resorbing osteoclasts contain gap-junctional connexin-43. J Bone Miner Res. 2000;15:919-26.

[30] Ransjo M, Sahli J, Lie A. Expression of connexin 43 mRNA in microisolated murine osteoclasts and regulation of bone resorption in vitro by gap junction inhibitors. Biochem Biophys Res Commun. 2003;303:1179-85.

[31] Matemba SF, Lie A, Ransjo M. Regulation of osteoclastogenesis by gap

junction communication. Journal of cellular biochemistry. 2006;99:528-37.

[32] Rossello RA, Kohn DH. Gap Junction Intercellular Communication: A Review of a Potential Platform to Modulate Craniofacial Tissue Engineering. J Biomed Mater Res B. 2009;88B:509-18.

[33] Rossello RA, Wang Z, Kizana E, Krebsbach PH, Kohn DH. Connexin 43 as a signaling platform for increasing the volume and spatial distribution of regenerated tissue. Proc Natl Acad Sci U S A. 2009;106:13219-24.

[34] Kirmizidis G, Birch MA. Microfabricated Grooved Substrates Influence

Cell-Cell Communication and Osteoblast Differentiation In Vitro. Tissue Eng Pt A. 2009;15:1427-36.

34

[35] Heng BC, Cao T, Stanton LW, Robson P, Olsen B. Strategies for directing the differentiation of stem cells into the osteogenic lineage in vitro. Journal of Bone and Mineral Research. 2004;19:1379-94.

[36] Kolar P, Schmidt-Bleek K, Schell H, Gaber T, Toben D, Schmidmaier G,

et al. The early fracture hematoma and its potential role in fracture healing. Tissue engineering Part B, Reviews. 2010;16:427-34.

[37] Tsiridis E, Upadhyay N, Giannoudis P. Molecular aspects of fracture healing:Which are the important molecules? Injury. 2007;38:S11-S25.

[38] Terheyden H, Lang NP, Bierbaum S, Stadlinger B. Osseointegration – communication of cells. Clin Oral Implants Res. 2011:n/a-n/a.

[39] Sims NA, Gooi JH. Bone remodeling: Multiple cellular interactions

required for coupling of bone formation and resorption. Seminars in cell & developmental biology. 2008;19:444-51.

[40] Sripanyakorn S, Jugdaohsingh R, Dissayabutr W, Anderson SH, Thompson RP, Powell JJ. The comparative absorption of silicon from different foods and food supplements. The British journal of nutrition. 2009;102:825-34.

[41] Izu A, Kumai T, Tohno Y, Tohno S, Minami T, Yamada G, et al. Silicon intake to vertebral columns of mice after dietary supply. Biological trace element research. 2006;113:297-316.

[42] Bae YJ, Kim JY, Choi MK, Chung YS, Kim MH. Short-term

administration of water-soluble silicon improves mineral density of the femur and tibia in ovariectomized rats. Biological trace element research. 2008;124:157-63.

[43] Spector TD, Calomme MR, Anderson SH, Clement G, Bevan L, Demeester N, et al. Choline-stabilized orthosilicic acid supplementation as an adjunct to calcium/vitamin D3 stimulates markers of bone formation in osteopenic females: a randomized, placebo-controlled trial. BMC musculoskeletal disorders. 2008;9:85.

[44] Jugdaohsingh R. Silicon and bone health. The journal of nutrition, health & aging. 2007;11:99-110.

[45] Szpalski C, Barr J, Wetterau M, Saadeh PB, Warren SM. Cranial bone

defects: current and future strategies. Neurosurgical focus. 2010;29:E8.

35

[46] Einhorn TA. Enhancement of fracture-healing. The Journal of bone and joint surgery American volume. 1995;77:940-56.

[47] Mauffrey C, Seligson D, Lichte P, Pape HC, Al-Rayyan M. Bone graft

substitutes for articular support and metaphyseal comminution: what are the options? Injury. 2011;42 Suppl 2:S35-9.

[48] Giannoudis PV, Dinopoulos H, Tsiridis E. Bone substitutes: an update. Injury. 2005;36 Suppl 3:S20-7.

[49] Helms JA, Amasha RR, Leucht P. Bone voyage: An expedition into the molecular and cellular parameters affecting bone graft fate. Bone. 2007;41:479-85.

[50] Jones JR, Sepulveda P, Hench LL. Dose-dependent behavior of

bioactive glass dissolution. J Biomed Mater Res. 2001;58:720-6.

[51] Sepulveda P, Jones JR, Hench LL. In vitro dissolution of melt-derived 45S5 and sol-gel derived 58S bioactive glasses. J Biomed Mater Res. 2002;61:301-11.

[52] Xynos ID, Edgar AJ, Buttery LD, Hench LL, Polak JM. Ionic products of bioactive glass dissolution increase proliferation of human osteoblasts and induce insulin-like growth factor II mRNA expression and protein synthesis. Biochem Biophys Res Commun. 2000;276:461-5.

[53] Tsigkou O, Jones JR, Polak JM, Stevens MM. Differentiation of fetal

osteoblasts and formation of mineralized bone nodules by 45S5 Bioglass conditioned medium in the absence of osteogenic supplements. Biomaterials. 2009;30:3542-50.

[54] Radin S, Ducheyne P, Falaize S, Hammond A. In vitro transformation of bioactive glass granules into Ca-P shells. J Biomed Mater Res. 2000;49:264-72.

[55] Tadjoedin ES, de Lange GL, Holzmann PJ, Kuiper L, Burger EH. Histological observations on biopsies harvested following sinus floor elevation using a bioactive glass material of narrow size range. Clin Oral Implants Res. 2000;11:334-44.

[56] Radin S, Reilly G, Bhargave G, Leboy PS, Ducheyne P. Osteogenic

effects of bioactive glass on bone marrow stromal cells. J Biomed Mater Res A. 2005;73A:21-9.

36

[57] Xynos ID, Edgar AJ, Buttery LDK, Hench LL, Polak JM. Ionic products of bioactive glass dissolution increase proliferation of human osteoblasts and induce insulin-like growth factor II mRNA expression and protein synthesis. Biochem Biophys Res Commun. 2000;276:461-5.

[58] Habibovic P, Barralet JE. Bioinorganics and biomaterials: bone repair.

Acta Biomater. 2011;7:3013-26.

[59] Hoppe A, Guldal NS, Boccaccini AR. A review of the biological response to ionic dissolution products from bioactive glasses and glass-ceramics. Biomaterials. 2011;32:2757-74.

[60] Gentleman E, Fredholm YC, Jell G, Lotfibakhshaiesh N, O'Donnell MD, Hill RG, et al. The effects of strontium-substituted bioactive glasses on osteoblasts and osteoclasts in vitro. Biomaterials. 2010;31:3949-56.

[61] Meyer U, Buchter A, Wiesmann HP, Joos U, Jones DB. Basic reactions

of osteoblasts on structured material surfaces. European cells & materials. 2005;9:39-49.

[62] Hayes J, Czekanska E, Richards R. The Cell–Surface Interaction

[Without Title]. Springer Berlin / Heidelberg; 2011. p. 1-31.

[63] Puleo DA, Nanci A. Understanding and controlling the bone-implant

interface. Biomaterials. 1999;20:2311-21.

[64] Mager MD, LaPointe V, Stevens MM. Exploring and exploiting chemistry at the cell surface. Nat Chem. 2011;3:582-9.

[65] Wilson CJ, Clegg RE, Leavesley DI, Pearcy MJ. Mediation of biomaterial-cell interactions by adsorbed proteins: a review. Tissue Eng. 2005;11:1-18.

[66] Ohtsuki C, Kamitakahara M, Miyazaki T. Bioactive ceramic-based

materials with designed reactivity for bone tissue regeneration. J R Soc Interface. 2009;6:S349-S60.

[67] Barrere F, van Blitterswijk CA, de Groot K. Bone regeneration: molecular and cellular interactions with calcium phosphate ceramics. International journal of nanomedicine. 2006;1:317-32.

[68] Engel E, Del Valle S, Aparicio C, Altankov G, Asin L, Planell JA, et al. Discerning the role of topography and ion exchange in cell response of bioactive tissue engineering scaffolds. Tissue Eng Pt A. 2008;14:1341-51.

37

[69] Sripanyakorn S, Jugdaohsingh R, Mander A, Davidson SL, Thompson RP, Powell JJ. Moderate ingestion of alcohol is associated with acute ethanol-induced suppression of circulating CTX in a PTH-independent fashion. J Bone Miner Res. 2009;24:1380-8.

[70] Vickerman JC. Intoduction. In: vickerman JC, Gilmore I, editors.

Surface Analysis: Principal Techniques: Wiley; 2009. p. 1-8.

[71] Carew EO, Cooke FW, Lemons JE, Ratner BD, Vesely I, Vogler E. Properties of Materials. In: Ratner BD, Hoffman AS, Schoen FJ, Lemons JE, editors. Biomaterials Science: An introduction to Materials in Medicine. Second Edition ed: Elsevier; 2004. p. 23-65.

[72] Shchukarev A. XPS at solid-solution interface: experimental approaches. Surf Interface Anal. 2006;38:682-5.

[73] Shchukarev A. XPS at solid-aqueous solution interface. Advances in

colloid and interface science. 2006;122:149-57.

[74] Hook AL, Anderson DG, Langer R, Williams P, Davies MC, Alexander MR. High throughput methods applied in biomaterial development and discovery. Biomaterials. 2010;31:187-98.

[75] Ross MH, Pawlina W. Histology : a text and atlas with correlated cell and molecular biology. 5th ed. Baltimore, MD: Lippincott Williams & Wilkins; 2006.

[76] Balabanova M, Popova L, Tchipeva R. Dyes in dermatology. Clin

Dermatol. 2003;21:2-6.

[77] Schmitz N, Laverty S, Kraus VB, Aigner T. Basic methods in histopathology of joint tissues. Osteoarthritis Cartilage. 2010;18 Suppl 3:S113-6.

[78] Cram LS. Flow cytometry, an overview. Methods in cell science : an official journal of the Society for In Vitro Biology. 2002;24:1-9.

[79] Riccardi C, Nicoletti I. Analysis of apoptosis by propidium iodide

staining and flow cytometry. Nature protocols. 2006;1:1458-61.

[80] Repetto G, del Peso A, Zurita JL. Neutral red uptake assay for the estimation of cell viability/cytotoxicity. Nature protocols. 2008;3:1125-31.

38

[81] Ransjö M, Lie A, Mackie EJ. Cholera toxin and forskolin stimulate formation of osteoclast-like cells in mouse marrow cultures and cultured mouse calvarial bones. European journal of oral sciences. 1999;107:45-54.

[82] Engel E, Del Valle S, Aparicio C, Altankov G, Asin L, Planell JA, et al.

Discerning the role of topography and ion exchange in cell response of bioactive tissue engineering scaffolds. Tissue engineering Part A. 2008;14:1341-51.

[83] Ewers R. Maxilla sinus grafting with marine algae derived bone forming material: A clinical report of long-term results. J Oral Maxil Surg. 2005;63:1712-23.

[84] Mordenfeld A, Hallman M, Johansson CB, Albrektsson T. Histological and histomorphometrical analyses of biopsies harvested 11 years after maxillary sinus floor augmentation with deproteinized bovine and autogenous bone. Clin Oral Implants Res. 2010;21:961-70.

[85] Xynos ID, Hukkanen MV, Batten JJ, Buttery LD, Hench LL, Polak JM.

Bioglass 45S5 stimulates osteoblast turnover and enhances bone formation In vitro: implications and applications for bone tissue engineering. Calcif Tissue Int. 2000;67:321-9.

[86] Bhakta S, Pattanayak DK, Takadama H, Kokubo T, Miller CA, Mirsaneh M, et al. Prediction of osteoconductive activity of modified potassium fluorrichterite glass-ceramics by immersion in simulated body fluid. Journal of materials science Materials in medicine. 2010;21:2979-88.

[87] Zhong JP, Greenspan DC, Feng JW. A microstructural examination of apatite induced by Bioglass in vitro. Journal of materials science Materials in medicine. 2002;13:321-6.

[88] Kokubo T, Takadama H. How useful is SBF in predicting in vivo bone

bioactivity? Biomaterials. 2006;27:2907-15.

[89] Mei J, Shelton RM, Marques PM. Changes in the elemental composition of Bioglass during its surface development in the presence or absence of proteins. J Mater Sci-Mater M. 1995;6:703-7.

[90] Juhasz JA, Best SM, Auffret AD, Bonfield W. Biological control of apatite growth in simulated body fluid and human blood serum. Journal of materials science Materials in medicine. 2008;19:1823-9.

[91] Stenport V, Kjellin P, Andersson M, Currie F, Sul YT, Wennerberg A, et

al. Precipitation of calcium phosphate in the presence of albumin on

39

titanium implants with four different possibly bioactive surface preparations. An in vitro study. Journal of materials science Materials in medicine. 2008;19:3497-505.

[92] Combes C, Rey C. Adsorption of proteins and calcium phosphate

materials bioactivity. Biomaterials. 2002;23:2817-23.

[93] Bohner M, Lemaitre J. Can bioactivity be tested in vitro with SBF solution? Biomaterials. 2009;30:2175-9.

[94] Lee JTY, Leng Y, Chow KL, Ren FZ, Ge X, Wang KF, et al. Cell culture medium as an alternative to conventional simulated body fluid. Acta Biomater. 2011;7:2615-22.

[95] Pan H, Zhao X, Darvell BW, Lu WW. Apatite-formation ability--

predictor of "bioactivity"? Acta Biomater. 2010;6:4181-8.

[96] Bohner M, Loosli Y, Baroud G, Lacroix D. Commentary: Deciphering the link between architecture and biological response of a bone graft substitute. Acta Biomater. 2011;7:478-84.

[97] Hench LL. Bioceramics - from Concept to Clinic. J Am Ceram Soc. 1991;74:1487-510.

[98] Trubiani O, Fulle S, Traini T, Paludi M, la Rovere R, Orciani M, et al.

Functional assay, expression of growth factors and proteins modulating bone-arrangement in human osteoblasts seeded on an anorganic bovine bone biomaterial. European cells & materials. 2010;20:72-83.

[99] Beloti MM, Martins W, Jr., Xavier SP, Rosa AL. In vitro osteogenesis induced by cells derived from sites submitted to sinus grafting with anorganic bovine bone. Clin Oral Implants Res. 2008;19:48-54.

[100] Herten M, Rothamel D, Schwarz F, Friesen K, Koegler G, Becker J. Surface- and nonsurface-dependent in vitro effects of bone substitutes on cell viability. Clinical oral investigations. 2009;13:149-55.

[101] Bernhardt A, Lode A, Peters F, Gelinsky M. Novel ceramic bone

replacement material Osbone(R) in a comparative in vitro study with osteoblasts. Clin Oral Implants Res. 2011;22:651-7.

[102] Kundu AK, Putnam AJ. Vitronectin and collagen I differentially regulate osteogenesis in mesenchymal stem cells. Biochem Biophys Res Commun. 2006;347:347-57.

40

[103] Yee KL, Weaver VM, Hammer DA. Integrin-mediated signalling through the MAP-kinase pathway. IET systems biology. 2008;2:8-15.

[104] Zambuzzi WF, Ferreira CV, Granjeiro JM, Aoyama H. Biological

behavior of pre-osteoblasts on natural hydroxyapatite: a study of signaling molecules from attachment to differentiation. J Biomed Mater Res A. 2011;97:193-200.

[105] Stevens MM, George JH. Exploring and engineering the cell surface interface. Science. 2005;310:1135-8.

[106] Tang J, Peng R, Ding J. The regulation of stem cell differentiation by cell-cell contact on micropatterned material surfaces. Biomaterials. 2010;31:2470-6.

[107] Zou S, Ireland D, Brooks RA, Rushton N, Best S. The effects of silicate

ions on human osteoblast adhesion, proliferation, and differentiation. Journal of Biomedical Materials Research Part B: Applied Biomaterials. 2009;90B:123-30.

[108] Jugdaohsingh R, Tucker KL, Qiao N, Cupples LA, Kiel DP, Powell JJ. Dietary silicon intake is positively associated with bone mineral density in men and premenopausal women of the Framingham Offspring cohort. Journal of Bone and Mineral Research. 2004;19:297-307.

[109] Shie M-Y, Ding S-J, Chang H-C. The role of silicon in osteoblast-like cell proliferation and apoptosis. Acta Biomater. 2011;7:2604-14.

[110] Kim MH, Bae YJ, Choi MK, Chung YS. Silicon supplementation

improves the bone mineral density of calcium-deficient ovariectomized rats by reducing bone resorption. Biological trace element research. 2009;128:239-47.

[111] Rico H, Gallego-Lago JL, Hernandez ER, Villa LF, Sanchez-Atrio A, Seco C, et al. Effect of silicon supplement on osteopenia induced by ovariectomy in rats. Calcif Tissue Int. 2000;66:53-5.

[112] Hott M, de Pollak C, Modrowski D, Marie PJ. Short-term effects of organic silicon on trabecular bone in mature ovariectomized rats. Calcif Tissue Int. 1993;53:174-9.

I

Appendix

Permission granted from copyright owner through RightsLink®

Illustration Ref License number

[1] 2783390020893

Fig 1 [2] 2783391213104

[3] 2783381311894

Fig 2 [17] 2786550747855

Fig 3 [21] 2787131042058

Fig 7 [74] 2780271177216

Servier Medical Art

Fig 4, 6, 10, 11, 12, 13 and 14 were produced using Servier Medical Art.

Photograph

Fig 10: Pictures courtesy of Dr. Maria Ransjö