Kian Eichholz, B.Eng 

 

DEVELOPMENT OF MECHANOBIOMIMETIC STRATEGIES 

TO DRIVE STEM CELL BEHAVIOUR FOR BONE REGENERATION 

Trinity College Dublin, January 2019 

 

 

 

A thesis submitted to the University of Dublin in partial fulfilment of  

the requirements for the degree of  

 

Doctor of Philosophy 

 

 

 

 

 

Supervisor: Prof. David Hoey 

Internal Examiner: Prof. Daniel Kelly 

External Examiner: Prof. Paul Dalton 

 

 



 

 

 



I 

 

Declaration 

I declare that this thesis has not been submitted as an exercise for a degree at this or 

any other university and it is entirely my own work.  

I agree to deposit this thesis in the University’s open access institutional repository 

or allow the library to do so on my behalf, subject to Irish Copyright Legislation and Trinity 

College Library conditions of use and acknowledgement. 

 

 

       

 ______________________ 

Kian Eichholz  



II 

 

Summary 

There are a host of cases where clinical intervention must be taken to treat diseased 

or damaged bone, including severe fractures, defects, tumours requiring tissue removal, and 

debilitating diseases such as osteoporosis. However, there are no current treatments which 

adequately achieve this goal, with current autografting approaches having severe limitations 

in terms of quantity of tissue available and additional surgical sites which damage healthy 

tissue and increase infection risk. There is thus a need to develop new strategies for bone 

regeneration. Understanding the mechanisms behind bone regeneration, and in particular, 

the key role of the mesenchymal stem cell (MSC) in this process, would provide invaluable 

information for the development of strategies to effectively regenerate bone in a 

physiologically appropriate manner.  

The overall aim of this thesis was to investigate the biophysical cues within the 

stem cell niche in bone which drive the recruitment and osteogenesis of MSCs, with the 

aim of developing strategies to recapitulate this behaviour and guide bone repair. The two 

primary means by which MSC behaviour is mediated were investigated in this thesis: 

indirect biophysical cues from osteocytes (stream 1 – chapter 3), and direct biophysical 

cues from the underlying fibrous tissue (stream 2 – chapter 4-6), which were subsequently 

combined to create a mechano-biomimetic scaffold for bone regeneration (chapter 7).  

In this thesis, the indirect biophysical cues from osteocyte signalling to MSCs were 

first investigated (chapter 3), where osteocytes were shown to release distinct mechanically 

activated osteocyte-derived extracellular vesicles (MAEVs) which contained unique cargo 

compared to extracellular vesicles (EVs) from statically cultured cells. These MAEVs 

significantly enhanced MSC recruitment and osteogenesis, with trends being almost 

identical to MSCs treated with conditioned medium from mechanically stimulated 

osteocytes. This confirmed that EVs are a key component in osteocyte-MSC 



III 

 

mechanosignaling, and reveals their potential alone for use as mechanotherapeutics to guide 

regeneration.   

Next, the role of direct biophysical cues in mediating MSC behaviour was 

investigated. To facilitate this, a melt electrowriting (MEW) printer was designed and built 

(chapter 4) to facilitate the fabrication of defined fibrous microenvironments upon which to 

study MSC behaviour. Various architectures with 10 μm fibre diameter were fabricated, 

where it was demonstrated that a 90° architecture enhanced MSC spreading and Yes-

associated protein (YAP) nuclear expression, a marker for osteogenesis (chapter 5). Long 

term culture of MSCs further revealed enhanced osteogenic differentiation in this scaffold 

architecture.  

The role of mineral modifications in driving MSC osteogenesis was further 

investigated (chapter 6). A biomimetic nano-needle hydroxyapatite (nnHA) coating was 

developed, characterised and compared to other mineral modification methods, including a 

commonly used coating method which yields a micro-plate hydroxyapatite (pHA) 

morphology, and a composite polycaprolactone-hydroxyapatite material to fabricate fibres 

with incorporated mineral. The nnHA coating significantly enhanced MSC mineralisation 

compared to all other groups and was also shown to facilitate a more controlled release of 

BMP2, demonstrating its potential for use in applications requiring controlled drug release. 

Finally, stream 1 and 2 were combined, with MAEVs being used for the 

functionalisation of nnHA scaffolds (chapter 7). The addition of MAEVS further 

significantly enhanced mineralisation almost 2-fold, in addition to a total 24-fold compared 

to the 90° scaffold developed in chapter 4 and a total of 52-fold compared to random 

fibrous scaffolds. This biomimetic scaffold which incorporates both direct and indirect 

biophysical cues inspired by the native stem cell niche in bone thus holds great potential as 

a more physiologically relevant and effective strategy to guide bone regeneration. 



IV 

 

Acknowledgements 

Undertaking this project over the past number of years has been a tough, but 

rewarding journey. I have been extremely lucky to have enormous support from mentors, 

family and friends, and have been lucky enough to meet a lot of new people along the way 

which have also become lifelong friends. Without all of this support, none of this work 

would have been possible. 

First and foremost, I would like to express my greatest thanks to Prof. David Hoey, 

and state that I genuinely could not have asked for a better supervisor. Dave has been a 

truly exceptional mentor, and I couldn’t have done this work to anywhere near this standard 

without his help. Most importantly, his endless friendly support has made this project a joy 

to do. I have learned a lot from Dave over these past few years and will always be grateful 

to him for giving me this opportunity. 

I was extremely lucky to work with a lot of great people throughout this PhD who 

have become great friends. To my original lab family; Jilly, Marie, Elena and Michele, who 

taught me from the level of not knowing what a pipette was, and Mathieu (Team Actimel), 

Ian, Angelica, Dan, Cairnan, Nian, Ivor and Conor, all of whom have helped me countless 

times throughout this project. I would also like to thank Kieran, Olwyn, Pedro, Paola, 

Pierluca, Fiona, Stan, Dinorath and Peter for their help and wisdom with various aspects of 

this project. Thanks also to Jess, Grá, Ian, Simon, Andy, Dave and Ross, my housemates 

Jenny, Julia and Stefan, and everyone else outside of my research group including those in 

my office in Parsons and everyone in TCBE, who have also helped along the way and made 

this journey much more fun than I thought possible. 

To some of my great and long-time friends, Killian, Kieron, Daniel, Darragh, Luke, 

Johnny, Billy, JP and Patrick. There’s been many times where things have gotten in the 



V 

 

way and we haven’t seen each other for long periods of time, but when we do it’s as if 

nothing has changed at all. So thank you all for always being there throughout the years.  

I would like to express my deepest thanks to my family; my parents Libby and Jörg, 

and my sisters Orlagh and Ciara. I’m sure I haven’t said this as much as I should, but I can’t 

thank you all enough for the continuous and endless support you have given me throughout 

my life, without which I would not be where I am today. I would also like to say a huge 

thank you to mam in particular, for all your help and time spent reading over this thesis. I 

am also very lucky to have a close and hugely supportive extended family. I would like to 

say a special thanks to my grandmother Betty and late grandfather James, who believed my 

mind for engineering came from him. To my extended family in Ireland and all over the 

world for your great support over the years; my Opa Jochen, late Oma Marlies, my aunts 

and uncles Mary, Gerry, Nicky, Seamie, Martin, Michael, Rita, Teresa, Tommy and Silke, 

and all of my close cousins of which there are too many to name here.  

Last but not least, I’d like to say a huge thank you to Sayuri, who has always given 

me tremendous support and has constantly been there for me throughout all the good times 

and the bad. I can’t imagine how difficult this would have been without you, and feel 

extremely lucky to always have you by my side. 

Thank you all. 

  



VI 

 

Table of Contents 

Declaration ............................................................................................................................. I 

Summary .............................................................................................................................. II 

Acknowledgements ............................................................................................................ IV 

Table of Contents ............................................................................................................... VI 

Publications ........................................................................................................................ XI 

Book chapters ..................................................................................................................... XI 

Conference abstracts ........................................................................................................ XII 

Nomenclature .................................................................................................................. XIV 

Chapter 1 Introduction ......................................................................................................... 1 

1.1 The need for new strategies to regenerate tissue ........................................................... 1 

1.2 Mechanobiological cues driving tissue regeneration .................................................... 2 

1.3 MEW as a means to control the tissue microenvironment ............................................ 5 

1.4 Objectives of thesis ....................................................................................................... 6 

Chapter 2 Literature review .............................................................................................. 11 

2.1 Introduction ................................................................................................................. 11 

2.2 Bone Anatomy ............................................................................................................ 12 

2.2.1 Cell types ........................................................................................................................ 12 

2.2.2 Development and healing ................................................................................................ 15 

2.2.3 Hierarchical structure and composition .......................................................................... 19 

2.3 Bone mechanobiology ................................................................................................. 24 

2.3.1 Introduction ..................................................................................................................... 24 

2.3.2 Role of the osteocyte in mechanoadaptation ................................................................... 24 

2.3.3 Osteocyte mechanosignalling ......................................................................................... 26 

2.4 Role of EVs in cell signalling ..................................................................................... 29 

2.4.1 Introduction ..................................................................................................................... 29 

2.4.2 EVs in bone ..................................................................................................................... 31 

2.5 Direct biophysical regulation of MSCs ....................................................................... 34 

2.5.1 Introduction ..................................................................................................................... 34 

2.5.2 Role of architecture in mediating MSC mechanobiology ............................................... 37 

2.6 Controlling architecture for bone regeneration ........................................................... 44 

2.6.1 Fibre architecture ............................................................................................................ 44 

2.6.2 Mineral architecture ........................................................................................................ 47 

2.7 MEW ........................................................................................................................... 50 

2.7.1 Introduction ..................................................................................................................... 50 

2.7.2 Development of MEW technology ................................................................................. 52 

2.7.3 Applications of MEW in regenerative medicine ............................................................. 58 



VII 

 

2.8 Summary ..................................................................................................................... 64 

Chapter 3 Osteocytes regulate human bone marrow mesenchymal stem cell behaviour 

via mechanically activated osteocyte-derived extracellular vesicles (MAEVs) ............. 67 

3.1 Introduction ................................................................................................................ 67 

3.2 Materials and methods ................................................................................................ 71 

3.2.1 Cell culture ...................................................................................................................... 71 

3.2.2 Mechanical stimulation and conditioned medium collection .......................................... 71 

3.2.3 Effect of osteocyte conditioned media on hMSC recruitment ......................................... 72 

3.2.4 Effect of osteocyte conditioned media on hMSC osteogenesis ....................................... 72 

3.2.5 Sample preparation for mass spectrometric (MS) analysis .............................................. 73 

3.2.6 Liquid chromatography tandem mass spectrometry (LC MS/MS) .................................. 73 

3.2.7 MS data analysis .............................................................................................................. 74 

3.2.8 Bioinformatics and statistical analyses ............................................................................ 74 

3.2.9 Extracellular vesicle isolation from conditioned media ................................................... 75 

3.2.10 Characterisation of EVs ................................................................................................... 76 

3.2.10.1 TEM imaging ........................................................................................................ 76 

3.2.10.2 Immunoblotting ..................................................................................................... 76 

3.2.10.3 Quantification of EV content in conditioned medium ........................................... 76 

3.2.10.4 Particle size analysis .............................................................................................. 77 

3.2.11 Uptake of EVs by MSCs .................................................................................................. 77 

3.3 Results ........................................................................................................................ 78 

3.3.1 Osteocytes regulate human MSC recruitment and osteogenesis in response to fluid shear

 78 

3.3.2 Overview of identified proteins within the osteocyte sectretome .................................... 79 

3.3.3 Proteomic analysis of the osteocyte secretome reveals enrichment of proteins associated 

with EVs ..................................................................................................................................... 80 

3.3.4 Mechanical stimulation alters the protein release characteristics in osteocytes............... 83 

3.3.5 EVs are present within the osteocyte secretome and EV morphology and size 

distribution is not altered by mechanical stimulation ................................................................. 87 

3.3.6 Osteocytes regulate human MSC recruitment and osteogenesis in response to fluid flow 

shear via MAEVs ....................................................................................................................... 88 

3.4 Discussion ................................................................................................................... 90 

3.5 Conclusion .................................................................................................................. 96 

Chapter 4 Design and build of a melt electrowriting printer ......................................... 99 

4.1 Introduction ................................................................................................................ 99 

4.2 Materials and methods .............................................................................................. 101 

4.2.1 Transient heat transfer analysis...................................................................................... 101 

4.2.2 Heating-voltage assembly .............................................................................................. 106 

4.2.3 Polymer extrusion .......................................................................................................... 108 

4.2.4 Controlled deposition..................................................................................................... 109 

4.2.5 Enclosure and safety ...................................................................................................... 110 



VIII 

 

4.2.6 Process visualisation and monitoring ............................................................................ 111 

4.3 Results ....................................................................................................................... 111 

4.3.1 Completed MEW printer ............................................................................................... 111 

4.3.2 Characteristics of heating assembly .............................................................................. 112 

4.3.3 Extrusion characteristics of syringe pump and pressure driven systems ...................... 114 

4.3.4 Control of fibre deposition ............................................................................................ 116 

4.3.5 Scaffold fabrication ....................................................................................................... 118 

4.4 Discussion ................................................................................................................. 120 

4.5 Conclusion ................................................................................................................ 123 

Chapter 5 Mediating human stem cell behaviour via defined fibrous architectures by 

melt electrowriting ............................................................................................................ 125 

5.1 Introduction ............................................................................................................... 125 

5.2 Materials and methods .............................................................................................. 128 

5.2.1 Scaffold fabrication and characterisation ...................................................................... 128 

5.2.2 Mechanical characterisation of scaffolds ...................................................................... 129 

5.2.3 Cell culture and proliferation ........................................................................................ 130 

5.2.4 Immunofluorescence and YAP expression ................................................................... 131 

5.2.5 Characterisation of osteogenic differentiation .............................................................. 132 

5.2.5.1 Intracellular ALP ................................................................................................ 132 

5.2.5.2 Collagen production ........................................................................................... 132 

5.2.5.3 Mineral production ............................................................................................. 133 

5.2.6 Statistical analysis ......................................................................................................... 133 

5.3 Results ....................................................................................................................... 134 

5.3.1 Fibrous cellular micro-environment design and imaging ............................................. 134 

5.3.2 Mechanical characterisation of fibrous cellular micro-environments ........................... 135 

5.3.3 Attachment and proliferation of human MSCs ............................................................. 137 

5.3.4 Fibrous architecture mediates stem cell shape and mechano-signalling ....................... 138 

5.3.5 Fibrous architecture directs human stem cell osteogenesis ........................................... 140 

5.3.5.1 ALP activity ........................................................................................................ 140 

5.3.5.2 Collagen production ........................................................................................... 141 

5.3.5.3 Mineral production ............................................................................................. 142 

5.4 Discussion ................................................................................................................. 143 

5.5 Conclusion ................................................................................................................ 146 

Chapter 6 Melt electrospun written scaffolds with tailored fibrous and mineral 

architectures to enhance protein delivery and human MSC osteogenesis for bone 

regeneration ....................................................................................................................... 149 

6.1 Introduction ............................................................................................................... 149 

6.2 Materials and methods .............................................................................................. 154 

6.2.1 Scaffold fabrication ....................................................................................................... 154 

6.2.1.1 MEW of PCL ...................................................................................................... 154 

6.2.1.2 pHA coating ........................................................................................................ 154 



IX 

 

6.2.1.3 Nano-needle hydroxyapatite (nnHA) coating ..................................................... 155 

6.2.1.4 Hydroxyapatite incorporated (iHA) PCL MEW ................................................. 156 

6.2.2 Scaffold characterisation ............................................................................................... 156 

6.2.2.1 SEM imaging and Energy-dispersive X-ray spectroscopy .................................. 156 

6.2.2.2 X-ray diffraction .................................................................................................. 157 

6.2.2.3 Calcium staining .................................................................................................. 157 

6.2.2.4 Water contact angle ............................................................................................. 158 

6.2.2.5 Tensile testing ..................................................................................................... 158 

6.2.3 hMSC Cell culture ......................................................................................................... 158 

6.2.4 Proliferation ................................................................................................................... 159 

6.2.5 Characterisation of hMSC osteogenic differentiation .................................................... 159 

6.2.5.1 Intracellular alkaline phosphatase (ALP) ............................................................ 159 

6.2.5.2 Collagen production ............................................................................................ 159 

6.2.5.3 Calcium production ............................................................................................. 160 

6.2.6 BMP2 functionalisation of MEW scaffolds ................................................................... 160 

6.2.6.1 BMP2 adsorption onto scaffolds ......................................................................... 160 

6.2.6.2 BMP2 release kinetics ......................................................................................... 160 

6.2.6.3 Characterisation of osteogenic differentiation ..................................................... 161 

6.2.7 Statistical analysis .......................................................................................................... 161 

6.3 Results ...................................................................................................................... 161 

6.3.1 MEW fibre topography is significantly altered by bulk and surface modification ........ 161 

6.3.2 CaP treatments modify surface chemistry ..................................................................... 163 

6.3.2.1 Element analysis .................................................................................................. 163 

6.3.2.2 Crystal structure of surface and bulk modifications ............................................ 164 

6.3.2.3 Calcium content of PCL modifications ............................................................... 165 

6.3.2.4 Surface treatment enhances material hydrophilicity ........................................... 165 

6.3.3 Mechanical properties are enhanced via HA-incorporation into PCL MEW fibres ...... 166 

6.3.4 Attachment and proliferation of hMSCs ........................................................................ 167 

6.3.5 Surface treatment significantly enhances human stem cell osteogenesis ...................... 168 

6.3.5.1 ALP activity ........................................................................................................ 168 

6.3.5.2 hMSC Collagen production ................................................................................. 168 

6.3.5.3 hMSC Calcium production .................................................................................. 169 

6.3.6 Coating dissolution after long term culture ................................................................... 170 

6.3.7 Mineral architecture mediates BMP2 controlled release and further enhances hMSC 

osteogenesis ............................................................................................................................. 171 

6.4 Discussion ................................................................................................................. 173 

6.5 Conclusion ................................................................................................................ 178 

Chapter 7 Mechanically activated osteocyte-derived extracellular vesicle 

functionalised melt electrowritten materials for bone regeneration: A 

mechanobiomimetic scaffold ........................................................................................... 181 

7.1 Introduction .............................................................................................................. 181 



X 

 

7.2 Materials and methods .............................................................................................. 185 

7.2.1 MEW ............................................................................................................................. 185 

7.2.2 Nano-needle hydroxyapatite (nnHA) coating ............................................................... 186 

7.2.3 Osteocyte cell culture .................................................................................................... 186 

7.2.4 Extracellular vesicle isolation from conditioned media ................................................ 187 

7.2.5 Functionalising MEW constructs with EVs and CM .................................................... 187 

7.2.6 SEM imaging and EDX ................................................................................................ 188 

7.2.7 Immunofluorescent staining of MEW-EV constructs ................................................... 188 

7.2.8 hMSC culture ................................................................................................................ 188 

7.2.9 Proliferation .................................................................................................................. 189 

7.2.10 Characterisation of hMSC osteogenic differentiation ................................................... 189 

7.2.10.1 Intracellular ALP ................................................................................................ 189 

7.2.10.2 Collagen production ........................................................................................... 189 

7.2.10.3 Calcium production ............................................................................................ 190 

7.2.11 Statistical analysis ......................................................................................................... 190 

7.3 Results ....................................................................................................................... 191 

7.3.1 Fabrication of CM and EV functionalised scaffolds ..................................................... 191 

7.3.2 Characterisation of CM and EVs .................................................................................. 192 

7.3.3 Attachment and proliferation of hMSCs ....................................................................... 193 

7.3.4 MSC osteogenesis ......................................................................................................... 194 

7.3.4.1 ALP activity ........................................................................................................ 194 

7.3.4.2 Collagen production ........................................................................................... 195 

7.3.4.3 Mineral production ............................................................................................. 196 

7.3.4.4 Mineral characterisation ..................................................................................... 198 

7.4 Discussion ................................................................................................................. 199 

7.5 Conclusion ................................................................................................................ 204 

Chapter 8 Discussion ........................................................................................................ 207 

8.1 Stream 1: Indirect cues for bone regeneration .......................................................... 210 

8.2 Stream 2: Direct cues for bone regeneration ............................................................. 212 

8.3 Development of a bone mechano-biomimetic scaffold............................................. 215 

8.4 Limitations and future directions .............................................................................. 217 

8.5 Conclusions ............................................................................................................... 222 

References .......................................................................................................................... 225 

List of figures ..................................................................................................................... 247 

List of tables ....................................................................................................................... 255 

Supplementary figures and tables ................................................................................... 257 

 

  



XI 

 

Publications 

Eichholz, K.F. and Hoey, D.A. Mediating human stem cell behaviour via defined fibrous 

architectures by melt electrospinning writing. Acta Biomater, 2018. 75: p. 140-151. 

Eichholz, K.F. and Hoey, D.A. Design and build of a melt electrowriting apparatus. (In 

preparation) 

Eichholz, K.F., Woods, I., Johnson, G.J., Shen, N., Corrigan, M., Labour, M.N., Wynne, 

K., Lowry, M.C., O’Driscoll, L., Hoey, D.A. Osteocytes regulate human bone marrow 

mesenchymal stem cell behaviour via mechanically activated osteocyte-derived 

extracellular vesicles (MAEVs). (In preparation) 

Eichholz, K.F. and Hoey, D.A. Melt electrospun written scaffolds with tailored fibrous and 

mineral architectures to enhance protein delivery and human MSC osteogenesis for bone 

regeneration. (In preparation) 

Eichholz, K.F., Federici, A., Mahon, O.R., Hoey, D.A. Mechanically activated osteocyte-

derived extracellular vesicle functionalised melt electrowritten materials for bone 

regeneration: A mechanobiomimetic scaffold (In preparation) 

Book chapters 

Eichholz, K.F. and Hoey, D.A. The Role of the Primary Cilium in Cellular 

Mechanotransduction, in Mechanobiology, C.F. Rawlinson, Editor. 2017, John Wiley & 

Sons, Inc: Hoboken, New Jersey. 

  



XII 

 

Conference abstracts 

Eichholz, K.F. and Hoey, D.A. Mimicking Cellular and ECM Mechanobiological Cues to 

Drive MSC Osteogenesis and Bone Regeneration. 25th Annual Conference of the section of 

Bioengineering of the Royal Academy of Medicine in Ireland, Limerick, 2019.  

Eichholz, K.F. and Hoey, D.A. Mediating Human Stem Cell Behaviour Via Defined 

Fibrous Architectures By Melt Electrospinning Writing. In: Proc. of the European 

Orthopaedic Research Society, Galway, Ireland, 2018. 

Eichholz, K.F., Woods, I., Shen, N., Johnson, G.J., Corrigan, M., Labour, M.N., Wynne, 

K., Lowry, M.C., O’Driscoll, L., Hoey, D.A. Quantitative analysis of the osteocyte 

secretome following oscillatory fluid shear stimulation. In: Proc. of the European 

Orthopaedic Research Society, Galway, Ireland, 2018. 

Eichholz, K.F. and Hoey, D.A. Melt electrospun written constructs for skeletal tissue 

regeneration: The effect of fibre architecture on YAP signalling and hMSC differentiation. 

World Congress of Biomechanics, Dublin, Ireland, 2018.  

Eichholz, K.F., Johnson, G.J., Corrigan, M., Labour, M.N., Wynne, K., Hoey, D.A. 

Quantitative analysis of the osteocyte secretome following oscillatory fluid shear 

stimulation. In: Proc. of the Orthopaedic Research Society, New Orleans, USA, 2018. 

Eichholz, K.F. and Hoey, D.A. Melt electrospun written constructs for skeletal tissue 

regeneration: The effect of fibre architecture. In: Proc. of the Orthopaedic Research 

Society, New Orleans, USA, 2018. 

Eichholz, K.F. and Hoey, D.A. Melt Electrospun Written Constructs For Skeletal Tissue 

Regeneration: The Effect Of Fibre Architecture. 24th Annual Conference of the section of 

Bioengineering of the Royal Academy of Medicine in Ireland, Dublin, 2018.  



XIII 

 

Eichholz, K.F. and Hoey, D.A. Melt Electrospun Written Scaffolds As In vitro Platforms 

For Cell Culture: The Effect Of Scaffold Pore Architecture. TERMIS EU Meeting, Davos, 

Switzerland, 2017.  

Eichholz, K.F. and Hoey, D.A. Melt Electrospun Written Scaffolds As In vitro Platforms 

For Cell Culture: The Effect Of Scaffold Pore Architecture. 23rd Annual Conference of the 

section of Bioengineering of the Royal Academy of Medicine in Ireland, Belfast, 2017. 

Eichholz, K.F. Design and Manufacture of an Innovative Biofabrication Method for 3D 

Tissue Engineering Constructs. 19th Sir Bernard Crossland Symposium, Belfast, 2016.  

Eichholz, K.F., Mac Guinness, A.D., Hoey, D.A. Design and Fabrication of Tissue 

Engineering Scaffolds using Melt Electrospinning Writing. In: Proc. of the Orthopaedic 

Research Society, Florida, US, 2016. 

Eichholz, K.F. and Hoey, D.A. Design And Fabrication Of Tissue Engineering Scaffolds 

Using Melt Electrospinning Writing. 22nd Annual Conference of the section of 

Bioengineering of the Royal Academy of Medicine in Ireland, Galway, 2016. 

Eichholz, K.F. and Hoey, D.A. Design of a 3-axis Direct Writing Apparatus for Controlled 

Deposition of Melt Electrospun Fibres. 21st Annual Conference of the section of 

Bioengineering of the Royal Academy of Medicine in Ireland, Dublin, 2015. 

 

  



XIV 

 

Nomenclature 

This is a list of commonly used abbreviations used in this thesis. Any other abbreviations 

will be explained in the text when used.   

ALP Alkaline phosphatase 

ANOVA Analysis of variance 

BMP2 Bone morphogenic protein 2 

BSA Bovine serum albumin 

C Control 

CAD Computer aided design 

CaP Calcium phosphate 

COX2 Cyclooxygenase 2 

DMEM Dulbecco's modified eagle medium 

DNA Deoxyribonucleic Acid 

ECM Extracellular matrix 

EDX Energy-dispersive x-ray spectroscopy 

EV Extracellular vesicle 

FBS Fetal bovine serum 

FDM Fused deposition modelling 

HA Hydroxyapatite 

hMSC Human mesenchymal stem cell 

iHA Incorporated hydroxyapatite 

MAEV Mechanically activated osteocyte-derived extracellular vesicles 

MEW Melt electrowriting / melt electrospinning writing 

miRNA Micro RNA (ribonucleic acid) 

MLO-Y4 Osteocyte cell line from murine long bone osteocyte Y4 

MSC Mesenchymal stem cell 

NaOH Sodium hydroxide 

nHA Nano-hydroxyapatite 

NM Normal medium 

nnHA Nano-needle hydroxyapatite 

OCN Osteocalcin 

OM Osteogenic medium 

OPN Osteopontin 

OSX Osterix 



XV 

 

PBS Phosphate buffered saline 

PCL Polycaprolactone 

pHA Plate hydroxyapatite 

PID Proportional-integral-derivative 

RUNX2 Runt-related transcription factor 2 

S10 10° scaffold architecture 

S45 45° scaffold architecture 

S90 90° scaffold architecture 

SBF Simulated body fluid 

SEM Scanning electron microscope 

TEM Transmission electron microscope 

SD Standard deviation 

SR Random scaffold architecture 

TCP Tricalcium phosphate 

TE Tissue engineering 

TTCP Tetracalcium phosphate 

XRD X-ray diffraction 

YAP Yes-associated protein  

 

  



XVI 

 

  



1 

 

Chapter 1 

Introduction 

1.1 The need for new strategies to regenerate tissue 

There is an ever-increasing need for solutions to heal and replace damaged and 

diseased tissue. This can be appreciated by the millions of grafting procedures performed 

every year worldwide, including 2,000,000 bone graft procedures (Campana et al., 2014), 

800,000 coronary artery bypass grafts (Grand View Research, 2017) and 300,000 anterior 

cruciate ligament constructions (Giedraitis, Arnoczky, & Bedi, 2014), in addition to the 

11,000,000 people requiring medical attention for severe burns each year (World Health 

Organisation, 2017). The vast majority of these procedures utilise autografts, which have 

two substantial drawbacks due to the necessity of a second surgical donor site as well as the 

limited amount of tissue which may be harvested. Allografts are also commonly used, 

however, this approach is also limited in terms of the amount of tissue which may be taken, 

as well as issues associated with biocompatibility and the increased potential for triggering 

an immune response. Synthetic engineered scaffolds for tissue regeneration are thus an 

attractive solution and have the potential to overcome the above issues to provide a reliable 

and effective source of tissue for regenerative applications in vivo. Moreover, use of these 

scaffolds to engineer tissues in vitro has considerable potential for the development of 

tissue models or organoids. These may aid in the further understanding of native tissue 

behaviour, in addition to their use as powerful biological platforms for the study of 

development and disease, and the safe and ethical development of pharmaceutical therapies.  



2 

 

1.2 Mechanobiological cues driving tissue regeneration 

Throughout the body, cells are intimately intertwined with their surrounding 

environment. Both intrinsic and extrinsic forces play a fundamental role in defining this 

environment, which in turn guides cell behaviour in response to the constantly changing 

external cues. The mechanobiology of a cell’s surrounding micro-environment is thus 

crucial in mediating its behaviour and ensuring tissue specific physiological homeostasis. 

Two key contributors to mediating this behaviour are the biophysical 

architecture/mechanical properties of the surrounding matrix in which the cells reside, and 

the biochemical factors released by resident/adjacent cells in response to adaptations in the 

mechanical environment. In bone, the close relationship between these cues and stem cell 

behaviour is of particular interest, due to the crucial role of stem cells in ensuring the 

continued regeneration and repair of this tissue. Indirect biochemical cues released from 

osteocytes recruit stem cells from the surrounding periosteum and endosteum, and further 

drive osteogenic differentiation in conjunction with direct biophysical cues upon binding of 

the stem cell to the underlying matrix. 

Changes in specific tissue mechanical environments can have both local and 

systemic effects through the release of mechanically-mediated biochemical cues from 

resident cells that act via an autocrine or paracrine manner to influence neighbouring cells 

within the tissue, as well as to much greater distances to different tissues and organs 

throughout the body. This indirect biophysical regulation of cell behaviour is of particular 

importance in bone, which comprises a range of cell types working in unison to 

continuously maintain the tissue in response to the complex loading regimes to which it is 

subjected throughout everyday activity. It is known that cells within the matrix of bone, in 

particular the osteocyte which is by far the most abundant cell in bone, can detect forces via 

a combination of means including direct mechanical strain and interstitial fluid flow 



3 

 

induced shear forces (van Oers, Wang, & Bacabac, 2015; Weinbaum, Cowin, & Zeng, 

1994). In response. the osteocyte subsequently sends signals to other bone residing cells to 

infer a response and help the tissue adapt to the changing mechanical environment (Dallas, 

Prideaux, & Bonewald, 2013). Mechanically stimulated osteocytes have been shown to 

promote osteoblast proliferation and recruitment (Brady, O'Brien, & Hoey, 2015) and 

inhibit osteoclast formation (Tan et al., 2007; You et al., 2008) via a paracrine mechanism. 

Importantly, osteocytes are also known to enhance the recruitment and osteogenic 

differentiation of stem cells (Brady et al., 2015; Hoey, Kelly, & Jacobs, 2011) via the 

release of various factors (P. M. Govey et al., 2014; Moester, Papapoulos, Löwik, & van 

Bezooijen, 2010), which is of great importance for the continued regeneration of bone. The 

specific means by which these factors mediate cell behaviour, along with the complete 

identification of the factors themselves, is yet to be fully understood however. The 

importance of these signalling cues in mediating tissue appropriate behaviour can truly be 

appreciated in cases of disease. For example, sclerostin expression is mediated by the 

osteocyte, with mechanical loading resulting in the inhibition of osteocyte expressed 

sclerostin and increased bone formation (A. G. Robling et al., 2008). This has been 

exploited in a treatment for osteoporosis, where antibody treatment to inhibit sclerostin 

expression enhances bone formation in vivo (Xiaodong Li et al., 2009), and has been shown 

in clinical trials to increase bone mineral density and reduce fracture risk (McClung, 2017). 

The extracellular matrix (ECM) imposes a direct, biophysical constraint on the 

degree to which cells can interact with and spread within their surrounding matrix. In bone, 

this matrix is comprised of a composite structure composed primarily of collagen fibres and 

mineral crystals, with collagen fibres at the endosteum and periosteum aligning in a load 

dependent manner in response to extrinsic forces (Foolen et al., 2008; McMahon, Boyde, & 

Bromage, 1995). Cell density, shape and degree of spreading are fundamentally defined by 

the underlying matrix in bone, which provides the blueprints to guide appropriate cellular 



4 

 

behaviour in addition to driving stem cell commitment towards the osteogenic lineage to 

ensure constant regeneration (Dupont et al., 2011; McBeath, Pirone, Nelson, Bhadriraju, & 

Chen, 2004). A secondary variable of great importance which is defined by the matrix 

architecture is matrix stiffness, which is 25–40 kPa at the collagen precursor to bone laid 

down by osteoblasts, with this being the optimal range upon which MSCs undergo 

differentiation (Engler, Sen, Sweeney, & Discher, 2006). It is known that cells can sense 

this stiffness and exert cytoskeletal tension in response, which in turn can trigger a cascade 

of responses from initial alterations in cell shape through to controlling the cell’s final fate 

and lineage commitment (Engler et al., 2006; Wells, 2008). This has been demonstrated in 

a range of cell types within bone, including stem cells, osteoblasts and endothelial cells, in 

addition to other cells throughout the body including neurons, myoblasts, fibroblasts, which 

may behave in vastly different ways and differentiate towards different lineages in response 

to altered matrix architecture and stiffness (Y. Yang, Wang, Gu, & Leong, 2017). As with 

biochemical cues, the importance of tissue specific architecture in maintaining 

physiologically appropriate behaviour can truly be appreciated by considering cases of 

disease, and how this can influence the architectural and mechanical properties of native 

tissue. For example, inappropriately aligned fibres surrounding stroma can facilitate tumour 

cell migration and metastasis (Hogrebe, Reinhardt, & Gooch, 2017b), while osteoarthritis is 

associated with altered fibrous architecture in the superficial zone of cartilage resulting in 

the amplification of tissue strain causing further cell damage and death (Saarakkala et al., 

2010). 

Appropriate signalling cues are thus fundamental for maintaining physiologically 

relevant behaviour and guiding appropriate tissue growth and regeneration, with these 

being present in bone in the form of indirect, mechanically mediated biochemical cues and 

direct, matrix mediated biophysical cues. It is thus of utmost importance to consider and 

understand the physical and biochemical cues present in the native environment of bone 



5 

 

and how these influence stem cell behaviour. This knowledge can then be used to design 

safe, effective and powerful bioengineering strategies for biologically inspired bone tissue 

regeneration.  

1.3 MEW as a means to control the tissue microenvironment 

Although cellular derived biochemical cues can be isolated from conditioned media, 

the ability to fabricate platforms that mimic the precise architecture and scale of the ECM 

of tissues is a considerable. MEW is a recently developed 3D printing technique which has 

the capability of fabricating fibres on the micron to sub-micron scale and precisely 

controlling their deposition in three dimensions (Brown, Dalton, & Hutmacher, 2011; 

Hochleitner et al., 2015). The unique combination of ECM like fibre production and 3D 

printing control yields an extremely powerful biofabrication technology with the potential 

to create complex microenvironments which guide cell behaviour in a specific and desired 

way. Fibres may be precisely stacked on top of one another to fabricate large volume 

constructs with precise microarchitectures (F.M. Wunner, Wille, et al., 2018), collected on 

rotating mandrels to form micro-fibrous tubular structures (McColl, Groll, Jungst, & 

Dalton, 2018) and used to fabricate scaffolds with intricate non-linear fibre architectures to 

define complex mechanical environments (Miguel Castilho, Mil, et al., 2018; Hochleitner, 

Chen, et al., 2018). The great potential of this technology can be seen in the rapid 

acceleration of MEW based constructs which have been developed for biomedical 

engineering applications since the technology was first described in 2011 (Brown et al., 

2011). Constructs have been developed for applications ranging from tissue engineering 

strategies for a host of tissue types including bone (Baldwin et al., 2017), cartilage (Visser 

et al., 2015), skin (Farrugia BL, 2013) and cardiac tissue (Miguel Castilho, Mil, et al., 

2018), to models for the study of prostate (Holzapfel et al., 2014), breast (Thibaudeau et al., 

2014) and bone (Wagner et al., 2016) tumours, to constructs to be used as platforms for 



6 

 

delivering cancer immunotherapy treatments (Delalat et al., 2017). Many of these 

constructs have used MEW scaffolds as the foundation upon which to define the 

microarchitecture perceived by the cells, before the further addition of components such as 

growth factors, drugs, hydrogels and minerals to further define the cell microenvironment.  

This technology thus has the potential to yield powerful and effective strategies in 

combination with specific growth factors and signalling molecules to precisely mediate cell 

behaviour for specific biomedical engineering applications. Despite this potential, there are 

only a few studies which have investigated the role of the defined fibrous architectures, 

achievable via MEW, on cell behaviour (F. Tourlomousis, Boettcher, Ding, & Chang, 

2017; Filippos Tourlomousis & Chang, 2017). Further understanding of the mechanisms by 

which complex fibrous microarchitectures mediate cell behaviour will undoubtedly 

progress the already immense potential of this technology in the field of bioengineering. 

This, in combination with further functionalisation with specific growth factors and 

signalling molecules which are present in the native tissue environment, will allow for the 

fabrication of powerful, highly specific microenvironments to intricately guide the desired 

cell behaviour for a given biomedical engineering or tissue engineering application.   

1.4 Objectives of thesis 

Bone is characterised by a complex microenvironment consisting of a range of cell 

types signalling to one another in a constantly changing highly dynamic environment. Stem 

cells are a fundamental component of this system, responsible for bone development and 

playing key roles in bone growth, regeneration and healing via constant replenishment of 

the native cell population. Stem cell behaviour is driven by both direct and indirect 

biophysical cues in bone, with these cues providing the blueprints for guided stem cell 

recruitment and differentiation. The objective of this thesis is to study the complex 

microenvironment of bone and elucidate the appropriate direct and indirect biophysical 



7 

 

cues to drive osteogenic stem cell behaviour. This knowledge will be used to engineer safe, 

effective biomimetic strategies to regenerate bone. The ultimate goal is to design a next 

generation bone tissue regeneration strategy via the fabrication of an optimised 

microarchitecture as a foundation upon which to define specific mineral and biological 

components to elicit an enhanced osteogenesis response from a native stem cell population. 

This will be achieved via five specific objectives in two main streams, A and B, as shown 

below. A schematic of this thesis project outline is also shown in Figure 0-1.  

A. Investigate the role of the osteocyte in mediating stem cell behaviour, facilitating 

the identification of biological factors for scaffold functionalisation: 

1. Study the role of the osteocyte in mechanically mediated signalling to 

human mesenchymal stem cells (hMSCs). A proteomic analysis will be 

conducted on the secretome of the osteocyte under static and dynamic 

conditions to form a map of potential mechanically regulated therapeutic 

targets to mediate stem cell behaviour. The key role of EVs in these 

processes will also be investigated.   

B. Fabricate an optimised MEW microarchitecture to mediate stem cell behaviour: 

2. Design and build a MEW apparatus. The effect of system parameters on 

deposition characteristics will be investigated, and the system tuned for 

precise fibre deposition. This apparatus will be used to fabricate scaffolds 

with precise microenvironments for tissue engineering applications.  

3. Investigate the role of fibrous architecture in mediating human stem cell 

behaviour. Various scaffold architectures will be fabricated via MEW after 

which the mechanisms by which stem cells are influenced by the scaffolds 

will be investigated in terms of cell morphology, cell behaviour, 

proliferation and osteogenic differentiation. An optimal design for further 

development in this thesis will finally be identified. 



8 

 

4. Develop methods by which to modify MEW scaffolds with a mineral 

component to enhance osteogenic differentiation for bone regeneration. The 

previously identified optimal architecture will be utilised as a foundation for 

this work. Scaffolds with incorporated calcium phosphates will also be 

fabricated by forming a composite material for MEW. The protein loading 

characteristics of these modified constructs will also be investigated, to 

assess drug loading capacity and release profiles for further therapeutic 

applications.  

Finally, these two streams will be combined to achieve the ultimate goal of this 

thesis: 

5. Develop a biomimetic fibrous scaffold with controlled nanotopography 

functionalised with MAEVs to harness the regenerative potential of 

mechanobiological cues and significantly enhance bone repair 



9 

 

 

Figure 0-1 Schematic illustrating outline of thesis. Two project streams will be followed with the aim of recapitulating the bone stem cell niche, with the final goal of developing a 

mechano-biomimetic scaffold for bone regeneration.   



10 

 

  



11 

 

Chapter 2 

Literature review 

2.1 Introduction 

Direct and indirect biophysical cues jointly orchestrate cell behaviour throughout 

the body. This is particularly seen to be the case within bone, where these cues define the 

dynamic stem cell niche, which is a fundamental for the continuous, controlled regeneration 

of bone tissue (P. M. Govey, Loiselle, & Donahue, 2013). An understanding of the 

mechanisms by which the stem cell niche within bone behaves will provide powerful, 

biomimetic tools for the developments of therapeutics to regenerate diseased and damaged 

bone. To achieve this, a thorough understanding of the mechanobiology of bone tissue is 

key. This literature review will thus begin with an introduction to bone biology, before 

going into greater detail on the crosstalk between the different cells involved in bone 

regeneration, with a focus on signalling between stem cells and the osteocyte. Furthermore, 

the role of the surrounding matrix and associated mechanical forces in mediating osteocyte 

mechanosignalling will also be discussed. Next, the role of tissue architecture and mineral 

topography will be discussed, with a focus on how stem cells interact with their 

surrounding matrix in bone, and the implications for this in driving appropriate cell 

behaviour. How tissue architecture may be altered with disease and the implications of this 

for further local and systemic disease development will also be discussed, with this 

providing key insights into the importance of physiologically appropriate cellular 

behaviour. Taken together, this information will provide key insight for the development of 

tissue regeneration strategies which can be closely integrated with the biology and 

mechanics of native healthy tissue.  



12 

 

Previous approaches by which tissue regeneration strategies have been developed 

will then be discussed, with a focus on two technologies in particular: electrospinning and 

3D printing. These two technologies will be discussed due to the sub-micron scale fibrous 

architectures which can be achieved and the great level of control which can be gained over 

architecture respectively. These encompass key capabilities which together give substantial 

control over the architectures which may be fabricated for regenerative medicine 

applications. A relatively new technology, MEW, does just this, and combines these 

technologies to form a powerful fabrication strategy with enormous potential. This 

biofabrication strategy will be introduced and discussed in terms of the underlying 

technology, fabrication capabilities, and current applications in regenerative medicine. 

Finally, advanced approaches which have developed this technology for applications in 

bone regeneration in particular will be discussed, including the development of composite 

materials, multi-phasic scaffolds, and incorporation of biochemical cues for enhanced 

therapeutic benefit. 

2.2 Bone Anatomy 

2.2.1 Cell types 

There are several types of cells within bone which collectively regulate the 

formation, growth and repair of the tissue. One such family of cells is derived from 

mesenchymal cells (MSCs), which are precursor cells found within the marrow and 

periosteum. These MSCs are critical components of bone development and are the main 

source of bone marrow stromal cells, which in turn differentiate into osteoblasts and 

subsequently into osteocytes (Manolagas, 2000). The osteoclast is another cell resident in 

bone and is derived from a separate lineage of monocyte/macrophage cells originating from 

hematopoietic stem cells. These cells and their basic functions will be introduced below, 



13 

 

with a more in-depth discussion of signalling and crosstalk between these cell types being 

conducted in section 2.3.  

Mesenchymal stem cells (MSCs) are precursor cells which play key roles in bone 

development, differentiating into osteoblasts which lay down bone matrix, and which in 

turn differentiate into osteocytes. This class of cells have been isolated from bone marrow 

and expanded in culture and have demonstrated a capacity to differentiate in vitro, leading 

to them being given the name MSCs. These cells are now known to arise from pericytes 

and adventitial progenitors, collectively known as perivascular stem cells (PSCs), with 

evidence also suggesting a possible non-blood vessel source in cranial sutures (Murray & 

Péault, 2015). There is still much debate over what exactly constitutes an MSC, and if these 

cells may truly be considered stem cells for regenerative purposes. The name was originally 

coined by Caplan (A. I. Caplan, 1991) due to his and other researchers’ findings that this 

cell could give rise to osteoblasts, chondrocytes, adipocytes and fibroblasts among others 

(Charbord, 2010). Recently, a more specific cell which can differentiate into bone, cartilage 

and stroma alone has been identified and characterised, with this being termed a skeletal 

stem cell (SSC) (Chan et al., 2018).  While isolated MSCs or SSCs can self-renew and 

differentiate in vitro, there is a lack of evidence for  long term skeletal regeneration or 

engraftment of these cells to bone or bone marrow in vivo and whether they behave as bona 

fide stem cells is still in question (Murray & Péault, 2015). Nonetheless, their importance in 

bone biology and key roles in signalling with other cells is well appreciated, leading to an 

alternative name of medicinal signalling cells being proposed (Arnold I. Caplan, 2017). For 

the remainder of this thesis, the term MSC will be used to denote “mesenchymal stem cell”, 

with this remaining the most accepted and commonly used term in the literature to date.    

Osteoblasts are cuboidal cells which are located on the surface of bone in a single 

layer. Their primary function is the synthesis of bone matrix which occurs by first 

depositing an organic matrix of collagen, other non-collagenous proteins such as 



14 

 

osteocalcin (OCN) and osteopontin (OPN), and proteoglycans (Florencio-Silva, Sasso, 

Sasso-Cerri, Simões, & Cerri, 2015). Subsequently, matrix vesicles are released, which due 

to their negative charge, immobilize calcium ions. ALP is also released by osteoblasts, 

resulting in the degradation of phosphate-containing compounds and release of phosphate 

ions inside the matrix vesicles. The calcium and phosphate ions within the vesicles then 

nucleate to form hydroxyapatite (HA) crystals, which after growth, eventually leads to their 

rupture and spreading of mineral crystals to the surrounding matrix. A quiescent variation 

of this cell is present on bone surfaces where bone remodelling does not occur, with these 

being given the name bone lining cells.  

Osteocytes are the most abundant cell type in bone, comprising 95% of all bone 

cells with an average density of 20,000 - 80,000 cells/mm3 bone tissue and a life span of up 

to several decades (Franz-Odendaal, Hall, & Witten, 2005). They originate from 

osteoblasts, which become terminally differentiated as osteocytes after becoming embedded 

in secreted matrix. They play key roles in coordinating the behaviour of other cell types in 

bone and throughout the body via communication by secreted factors (Bonewald, 2011). 

Osteocytes extend numerous dendritic processes which connect adjacent osteocytes 

forming a sensory network throughout bone tissue termed the lacuna-canalicular (LC) 

network. This network provides the platform through which osteocytes can sense changes 

in the external mechanical environment such as fluid flow induced shear stress, and signal 

to other cells within bone to adapt the tissue to these changes. The osteocyte can be 

considered an orchestrator of bone biology and remodelling and is thus of great relevance 

for tissue function. The osteocyte and its specific roles in signalling and tissue regulation 

will be discussed in greater detail in section 2.3.  

In contrast to the above cell types, osteoclasts are derived from the hematopoietic 

stem cell lineage. They are large, multinucleated cells, formed via osteoclastogenesis which 

is mediated via RANKL (Florencio-Silva et al., 2015). When this factor binds to RANK 



15 

 

receptors in osteoclast precursors, osteoclasts are formed. Conversely, osteoblasts release 

osteoprotegerin (OPG) which acts as a decoy receptor also binding to RANKL, inhibiting 

osteoclastogenesis. It is through this process that osteoclast numbers are controlled and the 

fine balance between bone resorption and deposition is maintained (Boyle, Simonet, & 

Lacey, 2003).  

2.2.2 Development and healing 

Understanding the mechanisms by which bone naturally develops and heals is 

useful for the development of regenerative strategies, which may exploit the natural 

behaviour of cells in order to form a strategy to heal bone. Bone development may occur 

via two methods; intramembranous ossification and endochondral ossification. 

Intramembranous ossification, which creates cortical and cancellous bone, is responsible 

for creating most of the cranial bones, flat bones of the face and the clavicles (K. A. Young 

et al., 2013). It is initiated when MSCs proliferate and form compact nodules, before 

developing into capillaries as well as osteogenic cells (Gilbert, 2000) (Figure 2-1). 

Osteoblasts become separated from the region of calcification via a secreted osteoid matrix 

which calcifies due to mineral deposition, with some cells also becoming trapped in this 

matrix and differentiating into osteocytes. Trabeculae begin to form around this ossification 

region, with mesenchymal cells on the outer regions forming the periosteum. Blood vessels 

form within the region forming red bone marrow, with remodelling also taking place via 

osteoclasts. 



16 

 

 
Figure 2-1 The stages of intramembranous ossification. Clusters of mesenchymal cells form in the embryonic 

skeleton (A), which begin to differentiate into osteogenic cells and secrete matrix which becomes mineralised 

(B). Trabeculae radiate from this ossification region, with mesenchymal cells lining the periphery and blood 

vessels branching through them (C). The periosteum is formed and red bone marrow forms from blood 

vessels, with remodelling continuing via osteoclasts (D). (K. A. Young et al., 2013) 

Endochondral ossification is the process but which long bones are formed within 

the body, in addition to the bones at the base of the skull. Cartilage tissue first forms from 

an aggregation of mesenchymal cells which differentiate into chondrocytes, with this 

forming the template for subsequent ossification and bone formation. Chondrocytes in the 

centre begin to increase in size, and more matrix is produced and becomes calcified, 

eventually limiting nutrient transfer resulting in cell death (K. A. Young et al., 2013). The 

cartilage matrix then begins to break down, with blood vessels invading the resulting voids 

and expanding them to form a cavity. Blood vessels also supply osteogenic cells, which 

form a periosteal collar consisting of cortical bone around the diaphysis at the centre of the 

developing bone. A primary ossification centre forms within the diaphysis, resulting in 

cortical bone formation radiating towards the epiphyses at the ends of the bone. 



17 

 

Simultaneously, chondrocytes continue proliferating and forming cartilage, increasing the 

length of the developing bone before cartilage is replaced by bone. Secondary ossification 

centres form postnatally in the epiphyses, resulting in cartilage becoming replaced with 

cancellous bone in these regions with the exception of the surface, at which articular 

cartilage forms. An epiphyseal plate, also called the growth plate, allows further 

proliferation of cartilage and subsequent replacement with bone to continue in this region 

until bones have fully formed in length. Bone modelling also occurs during bone growth, 

with this resulting in an increase in bone diameter.  Resorption of endosteal bone at the 

outer surface of the medullary cavity occurs via osteoclasts, along with deposition of bone 

beneath the periosteum via osteoblasts. 

 
Figure 2-2 The stages of endochondral ossification. A cartilage template first develops via an aggregation of 

mesenchymal cells which differentiate into chondrocytes (1). Cells in the centre become hypertrophic, with 

nutrient transfer becoming limited after further matrix deposition, resulting in cell death and infiltration of 

voids by blood vessels (2). Blood vessels supply osteogenic cells, which form a primary ossification centre (3), 

which expands towards the epiphyses replacing the cartilage template with bone. Secondary ossification 

centres form after birth (4), and bone increases in length via chondrocyte proliferation at the epiphyseal 

(growth) plate, while osteoblasts simultaneously develop cortical bone. This continues until the bone has 



18 

 

reached its final length and structure with complete fusion of the epiphyseal plates (6).  (Mescher & 

Junqueira, 2013) 

Bones are continuously remodelled in adulthood, with old tissue and damaged bone 

being resorbed by osteoclasts, before deposition of new bone matrix by osteoblasts 

allowing defects and microfractures to be replaced by new tissue. The collaboration of 

osteoclasts and osteoblasts in this process is called a basic multicellular unit (BMU). In 

cortical bone, a cylindrical canal approximately 2 mm long and 175 μm wide penetrates the 

bone at a speed of approximately 20-40 μm/day (Hadjidakis & Androulakis, 2006). 

Approximately 10 osteoclasts form a tunnel in the dominant direction of loading, with 

thousands of osteoblasts subsequently filling the tunnel and forming new tissue. The 

process is similar in cancellous bone, with osteoclasts travelling along the surface of 

trabeculae and forming trenches which are then refilled with osteoblasts. In the event of 

serious trauma such as fracture, bone has the capability to regenerate itself. In less serious 

cases, this may occur via primary (contact) healing. If there is stable contact at the fracture 

point and a gap of under 0.01 mm, healing may occur similar to bone remodelling, with 

osteoclasts crossing the fracture point and osteoblasts remodelling the region as outlined 

above (Marsell & Einhorn, 2011). However secondary (indirect) healing occurs much more 

commonly. A hematoma first forms, providing a large range of growth factors and cells, 

while inflammatory factors promote angiogenesis. MSCs are recruited from bone marrow, 

surrounding tissues and systemically from the circulatory system. Endochondral 

ossification occurs within the hematoma, with cartilaginous tissue providing stability until 

bone formation and remodelling occurs. Intramembranous ossification also occurs directly 

adjacent to the fracture ends, which when bridged, increases rigidity and allows a degree of 

weight bearing (Marsell & Einhorn, 2011). Tissue is disorganised at first, with remodelling 

occurring via the osteoblast/osteoclast BMUs to restore the loading optimised structure of 

the tissue. 



19 

 

2.2.3 Hierarchical structure and composition 

Bone is a hierarchical structure with two types on the macro scale: cortical 

(compact) and cancellous (spongy). Both of these types of bone are found throughout the 

body. Due to its greater strength, cortical bone found in regions which experience high 

levels of loading or which require greater protection such as in the diaphysis of long bones 

and in the skull. Cortical bone consists of osteons, which consist of sheets of 

circumferentially arranged tissue called lamellae. These contain networks of blood vessels 

and nerves through their centre as well as vessels which run perpendicularly to adjacent 

osteons, with this network providing the tissue and cells with a means for nutrient and 

waste transfer (Figure 2-3A).  Osteocytes are regularly distributed through the tissue inside 

lacunae. The highly compact nature of this osteon derived architecture allows the tissue to 

withstand the great forces present during movement and loading. Cancellous bone is found 

at the ends of long bones, in addition to other bones including the ribs, vertebrae, pelvic 

bones and skull. The basic sub-unit of cancellous bone is the trabecula, which contain 

lamellae and osteocytes as in cortical bone, however with a more irregular architecture and 

distribution (Figure 2-3B). As stated by Wolff’s law, healthy bone will adapt in response to 

loading (Frost, 1994). While this may occur in both cortical and cancellous bone, the 

greater capacity for trabeculae to arrange themselves in response to the load experienced in 

cancellous bone owes this tissue well to regions which experience complex loading 

patterns, for example in the femoral neck of the hip.  



20 

 

 
Figure 2-3 Structure of cortical (compact) bone (A), and cancellous (spongy) bone (B). Adapted from (K. A. 

Young et al., 2013) 

Further down the hierarchical structure of bone on the micro-scale, the organic 

component of the tissue is composed of collagen fibres of diameter 1 – 10 μm, which are 

arranged parallel to trabeculae in cancellous bone and in sheets of 90° offset between 

subsequent layers to form osteons in cortical bone (Figure 2-4) (Kane & Ma, 2013). 

Collagen provides the largest contribution towards the organic component of bone at 90%, 

with this comprising of primarily collagen type I. Collagen is a structural protein which 

provides the toughness for bone, as well as influencing bone cell proliferation, 

differentiation and apoptosis (M. F. Young, 2003). Collagen fibres are composed of fibrillar 

aggregations approximately 500 nm in diameter which in turn are composed of collagen 

molecules approximately 2 nm in diameter and 200 nm long. There are many other proteins 

which comprise the organic component of bone, including fibronectin, osteocalcin, 

osteopontin, alkaline phosphatase (ALP) and bone morphogenic proteins (BMPs), along 

with growth factors including insulin like growth factors (IGFs), transforming growth 

factors (TGFs), fibroblast growth factors (FGFs) and platelet derived growth factors 

(PDGFs) (Florencio-Silva et al., 2015). All of these components play various roles within 

bone. BMPs play a key role in bone formation in addition to regeneration in fracture 

healing (Solheim, 1998). IGFs are known to be involved in the recruitment and 



21 

 

proliferation of MSCs, while TGFs, FGFs, and PDGFs are all involved in maintaining 

normal cell physiology while also demonstrating roles in promoting tissue repair (Solheim, 

1998).  

 
Figure 2-4 Hierarchical structure of bone (Kane & Ma, 2013) 

A large proportion of bone is composed of inorganic mineral, which in turn has a 

hierarchical structure closely integrated with the collagen component, with proteins and 

growth factors dispersed throughout this composite structure.  Bone mineral comprises 

approximately 60-70% of the tissue by weight (Boskey, 2013), with this providing the 

majority of the stiffness and hardness of the tissue. The basic mineral unit of bone consists 

of nano-scale needles with approximate dimensions of base 5 nm and length 50 – 100 nm 

(Reznikov, Bilton, Lari, Stevens, & Kröger, 2018). These needles form platelets composed 

of partly merging crystals of the same base and diameter, with width 20 – 30 nm, which in 

turn form stacks and aggregates with irregular 3D structures of size 200 – 300 nm. Water is 

known to play a role in this crystal organisation, and contributes to the inorganic structure 

via mediating the orientation of mineral crystals (Y. Wang et al., 2013). Crystals are neither 



22 

 

exclusively intrafibrillar nor extrafibrillar but form a continuous crossfibrillar organisation 

which spans throughout and between collagen fibrils. This mineral organisation not only 

facilitates the high strength and stiffness of bone, but also provides a nano-scale structure to 

stabilise proteins and maintain their conformational structure (Yu et al., 2017). 

Remarkably, proteins within the mineral matrix of bone can be preserved for centuries (Yu 

et al., 2017), with the inverse relationship between protein stability and mineral feature size 

attesting to the great stability provided by the nano-structural features of bone.  

In terms of mineral composition, bone is often approximated to stoichiometric 

hydroxyapatite (HA). This is a member of the apatite family, referring to a group of 

phosphate minerals with general formula Ca5(PO4)3(F,OH,Cl), with F-, OH- and Cl- present 

in the most common forms being fluorapatite, hydroxyapatite and chlorapatite respectively, 

with geological apatite, the most abundant phosphate mineral in the crust of the Earth 

containing various proportions of these ions (Wopenka & Pasteris, 2005). However, 

biological bone apatite has several characteristics which differ significantly from HA, such 

as the presence of varying amounts of chemical substitutions, the significantly smaller 

percentage of hydroxyl groups (Loong et al., 2000), and the more disordered and 

protonated environment at the surface of crystals in biological minerals (Rey, Combes, 

Drouet, & Glimcher, 2009). Determining the precise chemical composition of bone brings 

many challenges, in part due to its complexity, presence of ions which are difficult to 

accurately measure to define their composition, and presence of a hydrated surface layer 

composed of bone and synthetic crystals containing a great variety of ions. Thus, several 

researchers have used synthetic calcium apatites, derived from natural geological apatites 

for example, to help gain a greater understanding of biological apatites. Remarkably, apatite 

is able to incorporate 50% of the elements of the periodic table, with substitutions that are 

known to occur in bone and tooth mineral including F, Cl, Na+, K+, Fe2+, Zn2+, Sr2+, Mg2+, 

carbonate (CO3
2-) and citrate (Wopenka & Pasteris, 2005). Of particular relevance is the 



23 

 

substitution of PO4
3 anions in the general hydroxyapatite formula with HPO4

2- and CO3
2-, 

which have also been detected in large amounts in bone mineral samples (Rey et al., 2009). 

Studies on synthetic apatites have also shown that there is a presence of non-apatite based 

mineral ions, as well as loosely bound ions on a hydrated surface layer as well as other ions 

and proteins in solution surrounding the mineral nanocrystals (Figure 2-5). In summary, 

while bone mineral may largely be approximated as stoichiometric hydroxyapatite, it is 

more accurately loosely defined as comprising a nano-plate/needle apatite base structure 

surrounded by a hydrated layer, disordered calcium phosphate (CaP) and loosely bound 

non-apatitic components. 

 
Figure 2-5 Schematic of the structure of bone mineral. The nanocrystal core is composed of non-

stoichiometric apatite. A hydrated surface layer contains loosely bound ions, with other ions and proteins (Pr) 

present in the surrounding fluid. (Rey et al., 2009). Transmission electron microscope image illustrating the 

needle morphology of the basic mineral unit of bone (Reznikov et al., 2018). 

In conclusion, bone is a hierarchical tissue comprised primarily of micron scale 

collagen fibres and nano scale apatite crystals, with its structure being maintained by the 

native cell population including osteocytes, osteoblasts, osteoclasts and MSCs. 

Understanding this structure, and the role of the cells within, is critical for the development 

of strategies for tissue regeneration, and may additionally facilitate the development of bio-

inspired scaffold designs to achieve this.  



24 

 

2.3 Bone mechanobiology 

2.3.1 Introduction 

It is well known that the structure of bone adapts in response to changes in its 

mechanical environment (Kontulainen, Sievänen, Kannus, Pasanen, & Vuori, 2003). These 

adaptive changes are regulated by the resident cells within bone, which sense mechanical 

forces and transduce this information to generate signals and infer a response. An important 

component in this system is signalling from the osteocyte to osteoblasts and osteoclasts and 

their precursors, which mediate bone remodelling. Osteocytes, which are the most abundant 

cells in bone, sense mechanical forces and send signals to osteoclasts to target microcracks 

and osteoblasts to lay down new bone matrix (Alexander G. Robling & Turner, 2009). 

Osteocytes also play a key role in mechanosignalling to MSCs to enhance recruitment and 

osteogenesis (Birmingham et al., 2012). This process is fundamental for the replenishment 

of the cell population in bone, ensuring continuous bone regeneration. In addition, MSCs 

have been demonstrated to exhibit excellent therapeutic potential following in vivo 

transplantation (Parekkadan & Milwid, 2010). Understanding the specific mechanisms 

behind osteocyte mechanosignalling to MSCs thus has great potential to uncover novel 

mechano-therapeutics to guide MSC behaviour, with this being discussed in greater detail 

in the following sections. 

2.3.2 Role of the osteocyte in mechanoadaptation 

Osteocytes are the primary sensing and metabolism controlling cells within bone 

tissue, and are key to directing the processes of bone formation and resorption via the 

release of various signalling molecules which act upon osteoblasts and osteoclasts or their 

progenitors (Dallas et al., 2013). The two overarching processes of interest in this regard 

are the means by which osteocytes infer information from their surrounding environment, 

and the mechanisms by which they communicate the changing stimulus in response. There 



25 

 

are several mechanisms by which it has been theorized that osteocytes detect the forces 

they are subjected to during bone loading, including sensing of direct mechanical strain, 

piezoelectric effects, electric ions generated by loading induced fluid flow, and fluid flow 

itself between osteocyte cell processes and canaliculi (van Oers et al., 2015). Fluid flow in 

vivo is one of the primary mechanisms thought to be responsible for the stimulation of 

osteocytes in the presence of macro scale loading, with flow induced fluid shear across the 

membrane of osteocyte processes believed trigger downstream mechanosignalling 

(Weinbaum et al., 1994). More recently it has been proposed that fluid flow through the 

lacunar-canalicular network induces strains in actin filament bundles more than an order of 

magnitude greater than strains on the tissue level (Han, Cowin, Schaffler, & Weinbaum, 

2004). Another mechanism which may contribute to osteocyte mechanosensing is the 

primary cilium, an immobile solitary antennae like organelle which is present in osteocytes 

(Malone et al., 2007) and may translate information to the cell via bending and 

transmission of strain to the cell in the presence of fluid flow. Regardless of the precise 

mechanism or combinations thereof, fluid flow has been shown to be a key stimulus of 

osteocytes with this being demonstrated via both fluid structure interaction (FSI) modelling 

(Verbruggen, Vaughan, & McNamara, 2014) and in vitro experimentation.  

A range of studies have illustrated the role of the osteocyte in mediating behaviour 

in other cells, primarily via the collection of conditioned medium from osteocytes following 

loading via parallel plate flow bioreactors. The majority of these in vitro osteocyte studies 

use an osteocyte-like cell called MLO-Y4, which is commonly used due to its versatility as 

a cell line, similarity to primary osteocytes, and ability to respond to mechanical stimulation 

(Kato, Windle, Koop, Mundy, & Bonewald, 1997). While other cell lines have been 

developed which more closely resemble primary osteocytes, including IDG-SW3 (Woo, 

Rosser, Dusevich, Kalajzic, & Bonewald, 2011) and Ocy454 (Spatz et al., 2015), MLO-Y4 

cells are still the most commonly used due to their well-established culture methods, ease of 



26 

 

use and well characterised behaviour. Co-culture studies has shown that fluid flow 

stimulation of osteocytes significantly reduces osteoblastic proliferation, and enhance 

differentiation as demonstrated by elevated ALP activity (Vezeridis, Semeins, Chen, & 

Klein-Nulend, 2006). Similar studies have validated these early findings while also 

demonstrating a role for gap junctions in this process (Taylor et al., 2007). In contrast, 

another study has demonstrated that mechanically stimulated osteocytes promote osteoblast 

proliferation as well as recruitment (Brady et al., 2015). Flow conditioned medium from 

osteocytes is also known to inhibit osteoclast formation, as demonstrated using 

multinucleated osteoclast progenitors isolated from bone marrow (Tan et al., 2007) and co-

cultures of RAW264.7 macrophage cells with osteocytes and ST2 stromal cells (You et al., 

2008). A role for the osteocyte in mediating stem cell behaviour following mechanical 

stimulation has also been shown, with enhanced migration, proliferation and osteogenic 

differentiation of MSCs following treatment with fluid shear stimulated osteocyte 

conditioned medium (Brady et al., 2015; Hoey et al., 2011). Remarkably, the osteocyte has 

also been shown to be involved in a large host of systemic functions, including regulating 

lymphoid organs and fat metabolism, increasing muscle myogenesis, influencing heart and 

liver function and suppressing growth of breast cancer and bone metastasis (Bonewald, 

2017).  

2.3.3 Osteocyte mechanosignalling 

The means by which the osteocyte communicates these biophysical stimuli to bone 

forming and resorbing effector cells is of great interest, with a host of studies investigating 

the possible factors released by osteocytes in response to mechanical stimulation. Nitric 

oxide (NO) (Klein-Nulend, Semeins, Ajubi, Nijweide, & Burger, 1995), prostaglandin E2 

(Cheng et al., 2001; Cherian et al., 2005), ATP (Genetos, Kephart, Zhang, Yellowley, & 

Donahue, 2007), RANKL (Nakashima et al., 2011), OPG and macrophage colony-

stimulating factor (M-CSF) (Zhao et al., 2002) have all be shown to be mechanosignalling 



27 

 

factors released by osteocytes in response to fluid shear, with these mediating a range of 

functions both within bone and systemically. One factor which has gained significant 

interest is sclerostin (SOST), which is a protein that is continually released resulting in the 

inhibition of Wnt-mediated bone formation (Moester et al., 2010). Expression of SOST is 

inhibited following mechanical loading (A. G. Robling et al., 2008; Shu et al., 2017), with 

this releasing the inhibition of Wnt signalling and stimulating bone formation. Anti-

sclerostin therapies which are currently being investigated are in stage 3 clinical trials 

(Lewiecki, 2014).  

To gain a greater understanding of the role of osteocyte signalling in response to 

loading, several studies have performed transcriptome analyses via microarrays and 

proteomic analyses via mass spectrometry.  These methods provide the advantage of 

facilitating the identification of a large datasets of gene/protein targets and their relative 

expression in response to loading. They also allow further detailed functional and pathway 

analyses to greater understand enrichments and interactions of defined groups of targets, 

thus significantly expanding our knowledge of osteocyte mechanobiology. One such study 

by Chen et al. investigated the transcriptome of the osteocyte following cyclic compressive 

force stimulation for several periods from 10 min to 6 h, with the greatest differential 

expression of targets (both upregulation and downregulation) compared to statically 

cultured cells occurring following 6 h stimulation (W. Chen et al., 2010). One cluster of 

genes with significantly upregulated expression following stimulation was highlighted in 

particular, containing a large number of chemokines and cytokines and which were 

associated with the significantly enriched “cytokine-cytokine receptor interaction pathway”. 

Interestingly, many of these genes were associated with bone resorptive signalling, 

indicating that the osteocyte may trigger resorption via osteoclasts and thus initiate bone 

remodelling in response to cyclic compressive forces.  



28 

 

A later study aimed to combine transcriptomic and proteomic analyses of the 

osteocyte to further reveal mechanosensitive signalling pathways (P. M. Govey et al., 

2014). Cells were cultured in parallel plate flow chambers with a flow regime of 1 Pa and 1 

Hz and were either lysed immediately or further cultured for 2, 8 or 24 h post stimulation. 

This revealed a range of transcripts and proteins which were differentially expressed with 

flow stimulation, as well as identifying a time course effect, where the greatest differential 

expression of transcripts and protein levels occurred at 2 hr and 8 hr post flow respectively. 

Many of the identified transcripts agreed with the previous work by Chen et al., while 24 

proteins out of 558 were differentially expressed with fluid flow and several of these 

including NDK, calcyclin and GRK-6 being identified as possibly playing key roles in 

mechanotransduction. This study also identified for the first time a combined 

transcriptome/proteome signalling network which identified key signalling nodes which 

would not have been revealed by either method alone. The complex microenvironment of 

bone was later investigated in two further studies from this research group, which re-

analysed the samples via RNA sequencing to more accurately quantify differential gene 

expression and further analyse network interactions to demonstrate fluid flow induced 

changes in genes which corresponded to downregulated osteoclast differentiation (Peter M. 

Govey, Kawasawa, & Donahue, 2015; F. Meng, Murray, Kurgan, & Donahue, 2018). 

The mechanobiology of the osteocyte has also been investigated via a more 

biologically accurate in vivo loading model (Wasserman et al., 2013). In this study, the C5 

vertebra of mice were either loaded in a single treatment, loaded three times weekly over a 

four-week period, or unloaded in control mice. Medullary tissue was then isolated and 

digested to remove all cells except osteocytes with trabecular bone. RNA was isolated from 

pulverised trabeculae and analysed via microarray analyses. Continuous loading was found 

to result in the most differentially expressed genes, with a total of 1339 compared to 339 in 

mouse loaded in a single treatment only. Two of the most enriched biological processes in 



29 

 

differentially expressed genes were “cellular movement” and “cell-to-cell signalling and 

interaction”, again highlighting the role of the loaded osteocyte in mediating cell behaviour 

and recruitment. The authors also highlighted the differentially expressed genes which code 

for extracellular proteins, identifying a total of 55 with continuous loading. While several 

have of these had previously implicated with osteoblast and osteoclast behaviour, many had 

not, again highlighting the value of large scale proteomic and genomic studies to further our 

understanding of cell behaviour. Of note are the identification of several thrombospondins, 

which promote osteoblast mineralisation and Wnt5a, which is a Wnt signalling pathway 

agonist, a pathway which plays key roles in bone formation and regulation. Upregulation of 

Dmp1 upregulation with loading is also of interest, which plays a role in maintaining 

phosphate homeostasis in bone (Y. Lu, Yuan, et al., 2011).  

In summary, the osteocyte plays a primary role in sensing external forces and 

sending signals to other bone residing cells to respond and adapt to the dynamic mechanical 

environment. Osteocyte mechanosignalling to MSCs is of particular importance due to the 

latter’s role in constantly replenishing the cell population in bone. Understanding and 

exploiting how osteocytes achieve this has great potential to reveal novel therapeutics for 

bone regeneration.  

2.4 Role of EVs in cell signalling 

2.4.1 Introduction 

Much of the work investigating means of cell signalling has previously focused on 

factors such as cytokines, chemokines and hormones. Recently however, another means by 

which cells communicate via EVs has gained significant interest, with EVs being 

demonstrated to be fundamental to an extensive range of signalling functions throughout 

the body (van Niel, D'Angelo, & Raposo, 2018). EVs are structures with a lipid bilayer 

membrane which can further be categorised as exosomes or microvesicles (MVs) 



30 

 

depending on their origin. Exosomes are formed when multivesicular endosomes (MVEs) 

fuse with the plasma membrane and release intraluminal vesicles (ILVs) from the cell, 

while MVs are formed in a separate process via budding of the plasma membrane (Figure 

2-6A-B). EVs have been shown to carry a wide range of cargo depending on conditions and 

cell type, including proteins, nucleic acids and lipids, with their composition directly 

influencing their fate and function thus highlighting their capacity for specific cell targeting 

and communication. This is achieved through a multitude of methods including surface 

binding of EVs to initiate signalling, fusion of EVs to the membrane for release of vesicle 

contents, and complete internalisation of EVs by the cell (Figure 2-6C).  

 
Figure 2-6 Summary of extracellular vesicle (EV) formation and signalling. EVs may be categorised as 

exosomes or microvesicles (MVs) (A). Exosomes are formed by the release of release of intraluminal vesicles 

contained within multivesicular endosomes, while microvesicles are formed via budding of the plasma 

membrane (B). These EVs can carry signals to a recipient cell, where they can initiate a response via surface 

binding, release of vesicle contents, or complete internalisation and further processing of EV contents (C). 

Adapted from (van Niel et al., 2018).  



31 

 

2.4.2 EVs in bone 

There are a range of studies highlighting the importance of EVs in bone in 

particular; with these primarily investigating osteoblast released EVs and their role in 

signalling with other cell types. Osteoblast EVs have been well characterised via proteomic 

analyses (J. Morhayim et al., 2015; Thouverey et al., 2011), with these studies providing 

novel insights into in contents of osteoblast EVs and their likely release mechanism from 

apical microvilli. One study investigated the use of RANKL by osteoblast EVs as a 

signalling mechanism to communicate with osteoclast precursors (L. Deng et al., 2015), 

where they were found to transfer RANKL to osteoclast precursor cells thus stimulating 

osteoclast formation. It was also found that these EVs interact with cells in a highly specific 

manner indicating targeted signalling, where they would bind to RANK-expressing-

NIH3T3 fibroblasts, however in NIH3T3 fibroblasts they were transferred to the cytosol for 

possible degradation within lysosomes or dismissed via transcytosis. One interesting study 

investigated the role of osteoblast EVs in signalling and differentiation of MSCs (Davies et 

al., 2017a). In this study, the osteogenic differentiation of MSCs cultured with osteoblast 

EVs was investigated and compared to treatment of MSCs with BMP2. ALP levels were 

found to be comparable in cells treated with EVs and BMP2, however even more 

remarkably, calcium deposition was significantly enhanced with EV treatment compared to 

BMP2. The authors then carried out a proteomic analysis on the EVs to further investigate 

the mechanisms behind these results, where a great number of proteins involved in calcium 

binding, osteogenesis and collagen modification were identified. One group of proteins of 

particular interest which was identified were annexins, due to their interaction with 

phospholipids to form Ca2+ complexes, which act as centres for crystallisation and 

nucleation, thus initiating localised mineralisation at EVs.   

Recently, work on EVs released by osteocytes has also emerged, with one such 

study investigating miRNA expression in EVs isolated from the plasma of osteocyte 



32 

 

ablated mice and wild-type mice, where altered miRNA levels suggest a role for osteocytes 

in systemic EV signalling (Mari Sato, Suzuki, Kawano, & Tamura, 2017). Another study 

investigated the influence of unidirectional fluid flow stimulation on osteocyte EV release 

(Morrell et al., 2018). Mechanical stimulation was not found to influence EV size, 

however, an upregulation in particle concentration was demonstrated. This increased EV 

production with flow was blunted when cells were treated with neomycin, an inhibitor of 

Ca2+ signalling, indicating the role of this mechanosensitive signalling pathway in EV 

release. This was further demonstrated in vivo, where Lysosomal-associated membrane 

protein 1 (LAMP1), a secretory vesicle marker, was upregulated in loaded mice tibia and 

attenuated in neomycin treated mice. More broadly, another study investigated the role of 

exercise in regulating systemic EV release (Whitham et al., 2018), with over 300 proteins 

upregulated, suggesting an EV based means by which osteocytes may communicate 

systemically to other cells following mechanical stimulation. 

Due to the involvement and specificity of EVs in cell signalling, several researchers 

are beginning to investigate the use of EVs for therapeutic purposes. EVs have been 

isolated from human dental pulp stem cells (DPSCs), and can be up taken by hMSCs and 

undifferentiated primary DPSCs, resulting in significantly enhanced differentiation as seen 

by enhanced BMP2 gene expression (C.-C. Huang, Narayanan, Alapati, & Ravindran, 

2016). In this study, EVs were also added to collagen membranes in human root tooth slices 

and implanted subcutaneously in mice, where they were seen to promote vascularisation 

and dental pulp-like tissue regeneration.  In another study, osteoclast EV interaction with 

osteoblasts was investigated, with EVs being enriched with a specific miRNA (miR-214), 

which are taken up by osteoblasts, inhibiting their activity (Weijia Sun et al., 2016). This 

miRNA was found to be elevated in an osteoporotic mouse model as well as osteoporotic 

patients further revealing its likely role in the regulation of osteoblast behaviour and as a 

potential EV based target for osteoporosis therapies. The researchers further demonstrated 



33 

 

this by inhibiting EV release via systemic treatment with Rab27a siRNA, which was shown 

to promote osteoblast activity. Work is also ongoing into the loading of EVs with drugs for 

specific therapeutic delivery. In one such paper, two different anti-osteoclast drugs, 

dasatinib and the clinically approved zoledronate for osteoporosis treatment, were 

internalised within osteoblast EVs (Cappariello et al., 2017). Acute osteoclast over 

activation was initiated in mice, after which they were treated with free drugs, as well as 

drugs loaded within EVs. Drugs loaded within EVs were shown to be just as effective as 

free drugs in inhibiting osteoclast activity and inducing apoptosis, both in vitro and in vivo. 

EVs, either unmodified or loaded with drugs, are thus highly attractive biological tools for 

the development of effective cell-free therapeutics with targeted delivery. 

Due to the regenerative and therapeutic potential of EVs, their use for the 

functionalisation of tissue engineering scaffolds has recently been investigated. In a paper 

by Xie et al., EVs were isolated from MSCs and used to functionalise decalcified bone 

matrix scaffolds, which were subsequently implanted subcutaneously in mice to assess their 

angiogenic and regenerative potential (H. Xie et al., 2017b). After 1 and 2 months, bone 

volume did not change with EV treatment, however, scaffolds functionalised with both EVs 

and cells displayed significantly enhanced bone formation, demonstrating their synergistic 

effect in bone tissue regeneration. Remarkably, blood vessel formation with EV 

functionalised scaffolds was not only significantly greater than control scaffolds, but also 

just as effective as cell seeded scaffolds, while scaffolds with a combination of EVs and 

cells resulted in even greater vessel formation. Another study modified 3D printed scaffolds 

with EVs and implanted them in rat calvarial defects to investigate their capacity for bone 

repair (Diomede et al., 2018). Polyethy