Adipose derived mesenchymal stem cells - DSOMK · 2016-11-08 · PhD Thesis Morten Højte Dahl, DDS...
104
PhD Thesis Morten Højte Dahl, DDS Adipose derived mesenchymal stem cells - Osteogenicity and osteoblast mineralization This thesis has been submitted to the Graduate School of Health and Medical Sciences, University of Copenhagen: 30/06/2016 FACULTY OF HEALTH AND MEDICAL SCIENCES UNIVERSITY OF COPENHAGEN
Transcript of Adipose derived mesenchymal stem cells - DSOMK · 2016-11-08 · PhD Thesis Morten Højte Dahl, DDS...
PhD Thesis
Morten Højte Dahl, DDS
Adipose derived mesenchymal stem cells
- Osteogenicity and osteoblast mineralization
This thesis has been submitted to the Graduate School of Health and Medical Sciences,
University of Copenhagen: 30/06/2016
F A C U L T Y O F H E A L T H A N D M E D I C A L S C I E N C E S UNIVERSITY OF COPENHAGEN
PhD Thesis
Morten Højte Dahl, DDS
Institute of Clinical Medicine
Faculty of Health and Medical Sciences
University of Copenhagen
Adipose derived mesenchymal stem cells
- Osteogenicity and osteoblast mineralization
Academic supervisors:
Professor Peter Schwarz,
MD, DMSci
Department of Endocrinology and Research Centre for
Ageing and Osteoporosis Rigshospitalet, University of
Copenhagen, Copenhagen, Denmark.
Professor Niklas Rye Jørgensen,
MD, PhD, DMSci
Department of Clinical Biochemistry Rigshospitalet,
Copenhagen, Denmark and Institute of Clinical
Research, University of Southern Denmark, Odense,
Denmark.
Professor Else Marie Pinholt,
DDS, MSci, Dr. odont
Institute of Regional Health Research, University of
Southern Denmark, Odense, Denmark.
Professor Robert Krarup Feidenhans’l,
Head of Institute, PhD
Niels Bohr Institute, University of Copenhagen,
Copenhagen, Denmark.
Table of contents
PREFACE ......................................................................................................................................................................... 5
LIST OF PAPERS ............................................................................................................................................................ 7
FUNDING.......................................................................................................................................................................... 8
SUMMARY ....................................................................................................................................................................... 9
DANISH SUMMARY (DANSK RESUMÉ) ................................................................................................................. 11
ABBREVIATIONS ......................................................................................................................................................... 13
1. INTRODUCTION ...................................................................................................................................................... 14
2. HYPOTHESIS ............................................................................................................................................................ 16
3. OBJECTIVES ............................................................................................................................................................. 17
4. BACKGROUND ......................................................................................................................................................... 18
4.1 BONE BIOLOGY ....................................................................................................................................................... 18
4.2 BONE CELLS ............................................................................................................................................................ 18
4.3 BONE REGULATORS ................................................................................................................................................. 21
4.4 BONE REMODELING ................................................................................................................................................. 24
4.5 BONE GRAFTING ...................................................................................................................................................... 27
5. MATERIALS AND METHODS ............................................................................................................................... 30
5.1 MEDLINE SEARCH (STUDY I) ................................................................................................................................... 30
5.2 TITANIUM CARRIER (STUDY II & III) ....................................................................................................................... 31
5.3 CULTURING OF ADMSCS (STUDY II & III) ............................................................................................................. 32
5.4 OSTEOGENIC POTENTIAL OF ADMSCS (STUDY II) .................................................................................................. 33
5.5 IN VITRO MINERALIZATION AND GENE EXPRESSION (STUDY II) ................................................................................ 35
5.6 EARLY IN VIVO EFFECT OF ADMSCS (STUDY III) .................................................................................................... 38
5.7 STATISTICS ............................................................................................................................................................. 42
6. RESULTS .................................................................................................................................................................... 43
6.1 MEDLINE SEARCH (STUDY I) ................................................................................................................................... 43
6.2 OSTEOGENIC POTENTIAL OF ADMSCS (STUDY II) .................................................................................................. 45
6.3 IN VITRO MINERALIZATION AND GENE EXPRESSION (STUDY II) ................................................................................ 46
6.4 EARLY IN VIVO EFFECT OF ADMSCS (STUDY III) .................................................................................................... 50
7. DISCUSSION .............................................................................................................................................................. 54
7.1 STRENGTHS AND LIMITATIONS (PAPER I) ................................................................................................................ 57
7.2 STRENGTHS AND LIMITATIONS (PAPER II) ............................................................................................................... 58
7.3 STRENGTHS AND LIMITATIONS (MANUSCRIPT III)................................................................................................... 59
7.4 IMPLICATIONS ......................................................................................................................................................... 60
8. CONCLUSION ........................................................................................................................................................... 61
9. PERSPECTIVES ........................................................................................................................................................ 62
10. REFERENCES ......................................................................................................................................................... 63
11. PAPER I .................................................................................................................................................................... 72
12. PAPER II ................................................................................................................................................................... 79
13. MANUSCRIPT III .................................................................................................................................................... 87
14. SUPPLEMENTARY DATA .................................................................................................................................. 101
14.1 OSTEOGENIC POTENTIAL OF ADMSCS ............................................................................................................... 101
14.2 TOTAL CALCIUM CONTENTS FOR MEDIUM VS. ADMSCS, TI- VS. TI+, AND TIO2- VS. TIO2+ ............................... 102
14.3 TOTAL CALCIUM CONTENT .................................................................................................................................. 103
14.4 RELATIVE ALPL, COL1Α1, AND RUNX2 GENE EXPRESSION ............................................................................. 104
5
Preface
The work presented in this thesis was mainly performed in the laboratory of Research Centre
for Ageing and Osteoporosis, Rigshospitalet Glostrup. The in vitro study was completed in 2010
under supervision of Professor Niklas Rye Jørgensen and Professor Else Marie Pinholt. Without the
help and support from Niklas and Else this project would not have been possible. A special thanks
to Niklas and Else, who gave me the opportunity to make this PhD.
I would like to give a special thanks to my principal supervisor Professor Peter Schwarz for
accepting me as his PhD student after changing principal supervisor. Outside our department I
would like to thank my co-supervisor Professor Robert Feidenhans’l from Niels Bohr Institute.
The in vivo study was prepared in the laboratory of Research Centre for Ageing and
Osteoporosis, Rigshospitalet Glostrup, and the surgery and live animal work were carried out at
Department of Experimental Medicine at Panum Institute, University of Copenhagen. I would like
to thank Cathrine Juel Bundgaard and Grete Østergaard for assisting my work in the animal
facilities.
I had the pleasure of spending four months during 2015 in the laboratory at Department of
Dentistry, Aarhus University alongside bioanalyst Sussi Madsen and associate professor David
Christian Evar Kraft. A great thanks to Sussi and David for assisting me with preparation of slides
and staining of the sections.
The process of evaluating the sections from the in vivo study was performed in collaboration
with laboratory technician Jette Barlach and Professor Ellen-Margrethe Hauge from Department of
Rheumatology, Aarhus University Hospital. I am grateful for your help and ideas for the evaluation
of my sections.
During the long process of this thesis I have faced many challenges. Fortunately, I have been
surrounded by supporting and cheerful people. I would therefore like to thank all my colleagues and
ex-colleagues at Research Centre of Ageing and Osteoporosis – Solveig Petersen, Torben Kvist,
6
Barbara Rubek Nielsen, Lars Schack Kruse, Trine Lund-Jacobsen, Anne Qvist Rasmussen, Ankita
Agrawal, Anne Lise Lysgaard, Bo Abrahamsen, Susanne Syberg Nielsen, Tina Kringelbach, Jeanne
Carlsen, Anja Gjødsbøl Frederiksen, Zanne Henriksen and Kasia Gurzawska. A special thanks to
Maria Ellegaard Larsen and Camilla Albeck Neldam for a joyful collaboration, support and
encouragement.
A special gratitude goes to Lasse Maretty Sørensen for assisting with some of my experiments
during my in vitro study. I would also like to thank my brother Anders Dahl for a thorough and
systematic review of my thesis with useful feedback.
Last but not least I would like to thank my family and friends for their never-ending support.
Especially, I would like to send lots of loving thoughts to my wife Kathrine Højte Dahl. Thank you
Kathrine, for always being there for me and being able to take my mind off work – and for giving
birth to our lovely daughter Olivia. If it was not for your support, this thesis would never have been
completed.
Morten Højte Dahl, June 2016
7
List of papers
This PhD thesis is based on the following papers and manuscript:
- Paper I:
Dahl M, Jørgensen NR, Hørberg M, Pinholt EM: Carriers in mesenchymal stem cell
osteoblast mineralization – State-of-the-art. J Craniomaxillofac Surg 42(1):41-47, 2014.
- Paper II:
Dahl M, Syberg S, Jørgensen NR, Pinholt EM: Adipose derived mesenchymal stem cells –
Their osteogenicity and osteoblast in vitro mineralization on titanium granule carriers. J
Craniomaxillofac Surg 41(8):e213-220, 2013.
- Manuscript III:
Dahl M, Hauge EM, Schwarz P, Jørgensen NR, Pinholt EM: Adipose derived mesenchymal
stem cells seeded on oxidized titanium granules increases osteoblast-covered surface in vivo
compared with titanium granules. Submitted to J Craniomaxillofac Surg, 2016.
The author of this thesis is the main contributor to all three papers.
8
Funding
The work of this thesis was kindly supported by The Danish Council for Independent
Research, Medical Sciences grant no. 09-067289 and by a grant from the UCPH2016 Excellence
Programme for interdisciplinary research to the Co-NeXT project. The PhD fee was partly funded
by Region H. Finally, porous titanium granule material was kindly provided by Tigran
Technologies AB, Malmø, Sweden.
9
Summary
Bone reconstruction is crucial in cranio-maxillofacial surgery. Although autologous bone
grafting is the preferred treatment for bone reconstruction, research within synthetic and natural
biomaterials combined with osteogenic cells is undergoing rapid and far-reaching developments.
Bone reconstruction with autologous bone grafting has the advantage of transferring
osteoprogenitor cells or osteoblasts, being both osteoconductive and osteoinductive. However,
harvesting of autologous bone grafts is associated with morbidity. Therefore, tissue engineering
with a combination of synthetic or natural biomaterials with osteogenic cells has been attempted to
promote osteogenesis. This involves the in vitro seeding of cells onto carrier material supporting
cell adhesion, migration, proliferation, and differentiation. The carrier defines the 3D shape of the
tissue to be engineered, and the optimal carrier should be biocompatible, biodegradable, and
osteoconductive to generate new bone formation. Many different carriers have been studied with
evaluation of gene expression and mineralized matrix formation. Nevertheless, the best carrier
material for bone reconstruction has not been found.
We hypothesized that adipose derived mesenchymal stem cells (ADMSCs) differentiated into
osteoblasts will result in significantly increased number of osteoblasts and mineralized matrix
formation when seeded on porous oxidized titanium (TiO2) granules compared with non-oxidized
titanium (Ti) granules. The aim was to demonstrate the osteogenic potential of ADMSCs and to
evaluate the effect of porous Ti and TiO2 granules as a carrier material for ADMSCs on formation
of mineralized matrix and gene expression in vitro and the early effect of in vivo implantation
subcutaneously in mice.
For this purpose, we isolated ADMSCs from mice, investigating their osteogenic potential
and the effect when seeded on porous Ti and TiO2 granules on mineralized matrix formation and
gene expression in vitro. To assess the early in vivo effect, we performed histomorphometry on
subcutaneously implanted ADMSCs seeded on porous Ti and TiO2 granules in mice.
In the in vitro study, we showed that ADMSCs express osteoblastic lineage genes and stain
strongly for alkaline phosphatase. Total calcium content as a measure for mineralized matrix
10
formation was significantly higher for ADMSCs seeded on porous TiO2 granules compared with Ti
granules. Expression of osteoblast-specific genes was significantly higher for ADMSCs seeded on
porous TiO2 granules, while expression of collagen type 1α1 was significantly higher for ADMSCs
alone.
In the in vivo study a higher osteoblast-covered surface ratio (p=0.002) and osteoclast-covered
surface ratio (p=0.036) was found in porous TiO2 granules with ADMSCs compared with Ti
granules with ADMSCs. Without ADMSCs, porous TiO2 granules also showed a higher osteoblast-
covered surface ratio (p=0.002) and osteoclast-covered surface ratio (p=0.003) compared with
porous Ti granules without ADMSCs. Likevise, a higher number of osteoblasts covering the
granule surface was found in porous TiO2 granules compared with Ti granules with ADMSCs
(p=0.001) and without ADMSCs (p=0.003). A higher number of osteoclasts covering the granules
surface was found in porous TiO2 granules compared with Ti granules with ADMSCs (p=0.023)
and without ADMSCs (p=0.009).
In conclusion, ADMSCs are osteogenic and culturing ADMSCs on porous TiO2 granules
resulted in a significantly higher mineralized matrix formation in vitro. In the in vivo study a higher
appearance of osteoblasts on porous TiO2 granules with and without ADMSCs was found,
suggesting porous TiO2 granules being more osteoinductive compared to porous Ti granules, and
that the pre-seeding of ADMSCs did not have any effect in our setup.
11
Danish summary (Dansk resumé)
Knoglerekonstruktion har afgørende betydning inden for kæbekirurgi. Selvom autolog
knogletransplantation er den foretrukne behandling, sker der store fremskridt inden for forskning i
syntetiske og naturlige biomaterialer kombineret med osteogene celler.
Knoglerekonstruktion med autolog knogletransplantation har den fordel, at der overføres
osteoprogenitor-celler eller osteoblaster, hvilket gør metoden både osteokonduktiv og
osteoinduktiv. Imidlertid er høst af autologe knogletransplantater forbundet med betydelig
morbiditet. Derfor har man forsøgt at kombinere syntetiske eller naturlige biomaterialer med
osteogene celler for at genskabe knoglevæv ved at fremme osteogenesen. Dette sker ved in vitro
udsåning af osteogene celler på et carrier-materiale, der understøtter celleadhæsion, migrering,
proliferering og differentiering. Carrierens funktion er at definere 3D formen af det væv, som skal
gendannes, og den optimale carrier bør være biokompatibel, bionedbrydelig og osteokonduktiv for
at genskabe ny knogle. Mange forskellige carriers er blevet undersøgt med evaluering af
genekspression og mineraliseret matrixdannelse. Ikke desto mindre er den bedste carrier til
knoglerekonstruktion ikke fundet.
Vores hypotese er, at osteoblaster differentieret fra adipøsderiverede mesenkymale stamceller
(ADMSC) vil resultere i et signifikant øget antal osteoblaster og mere mineraliseret matrixdannelse,
når cellerne dyrkes på porøst oxideret titanium (TiO2) granulat sammenlignet med ikke-oxideret
titanium (Ti) granulat. Formålet er at demonstrere det osteogene potentiale af ADMSC og evaluere
effekten af porøst Ti- og TiO2-granulat som carrier-materiale ved at vurdere mineraliseret
matrixdannelse og genekspression in vitro samt den tidlige in vivo effekt af subkutan implantation i
mus.
Vi isolerede ADMSC fra mus for at undersøge deres osteogene potentiale, deres evne til at
danne mineraliseret matrix og udtrykke osteoblast-specifikke gener ved dyrkning på henholdsvis
porøst Ti- og TiO2-granulat in vitro. Vi udførte histomorfometri for at vurdere den tidlige in vivo
effekt af ADMSC dyrket på porøst Ti- og TiO2-granulat implanteret subkutant i mus.
12
Resultaterne fra in vitro studiet viste, at ADMSC udtrykker osteoblast-specifikke gener og
farves positivt for indhold af basisk fosfatase. Det samlede calciumindhold, som udtryk for
mineraliseret matrixdannelse, var signifikant højere for ADMSC dyrket på porøst TiO2-granulat
sammenlignet med Ti-granulat. Samtidig var ekspressionen af osteoblast-specifikke gener
signifikant højere for ADMSC dyrket på porøst TiO2-granulat, mens ekspressionen af collagen type
1α1 var signifikant højere for ADMSC dyrket alene.
Resultaterne fra in vivo studiet viste en højere relativ osteoblast-dækket overflade (p=0.002)
og osteoklast-dækket overflade (p=0.036) i porøst TiO2-granulat med ADMSC sammenlignet med
Ti-granulat med ADMSC. Ligeledes havde porøst TiO2-granulat uden ADMSC en højere relativ
osteoblast-dækket overflade (p=0.002) og osteoklast-dækket overflade (p=0.003) sammenlignet
med Ti-granulat uden ADMSC. Endvidere fandt vi flere osteoblaster på granulatoverfladen af
porøst TiO2-granulat sammenlignet med Ti-granulat både med (p=0.001) og uden ADMSC
(p=0.003). Det samme gjaldt for osteoklaster, hvor vi fandt signifikant flere på granulatoverfladen
af porøst TiO2-granulat sammenlignet med Ti-granulat både med (p=0.023) og uden ADMSC
(p=0.009).
Afslutningsvis kan vi konkludere, at ADMSC er osteogene, og at in vitro dyrkning af
ADMSC på porøst TiO2-granulat resulterer i signifikant højere mineraliseret matrixdannelse.
Derudover kan vi konkludere, at porøst TiO2-granulat resulterer i flere osteoblaster in vivo, hvilket
tyder på, at porøst TiO2-granulat er et mere osteoinduktivt materiale sammenlignet med porøst Ti-
granulat. Samtidig fandt vi ingen effekt af at dyrke ADMSC på granulaterne i vores set-up.
13
Abbreviations
ADMSCs: Adipose derived mesenchymal stem cells
ALP (ALPL): Alkaline phosphatase
AR-S: Alizarin Red S
BMD: Bone mineral density
BMP: Bone morphogenic protein
BMSCs: Bone marrow mesenchymal stem cells
bOB: Mature osteoblasts extracted from mouse bone
β-TCP: β-tricalcium phosphate
Cbfa-1: Core-binding factor subunit α-1
CDHA: Calcium-deficient hydroxyapatite
COL1α1: Collagen type 1α1
CPC: Cetylpyridinium Chloride
CV: Coefficient of variation
DBM: Demineralized bone matrix
Fb.S: Fibrosis intersection
GAPDH: Glyceraldehyd 3-phosphate dehydrogenase
GS: Granule surface
GV: Granule volume
HA: Hydroxyapatite
M-CSF: Macrophage colony-stimulating factor
MeSH: Medical subject headings
mOB MEM: Mouse osteoblast medium
MSCs: Mesenchymal stem cells
NE.S: Not evaluable intersection
NFAT2: Nuclear factor for activated T cells 2
N.Ob: Number of osteoblasts
N.Oc: Number of osteoclasts
Ob.S: Osteoblast intersection
OC: Osteocalcin
Oc.S: Osteoclast intersection
OPG: Osteoprotegerin
PBS: Phosphate Buffered Saline w/o Ca2+ and Mg2+
PBS+: Phosphate Buffered Saline with Ca2+ and Mg2+
PCR: Polymerase chain reaction
PLGA: Polylactic acid-polyglycolic acid copolymer
PPAR-γ1: Peroxisome proliferator-activated receptor-γ1
PPAR-γ2: Peroxisome proliferator-activated receptor-γ2
PRP: Platelet-rich plasma
PTH: Parathyroid hormone
RANK: Receptor activator of nuclear factor κB
RANKL: Receptor activator of nuclear factor κB ligand
RT: Reverse transcription
RUNX2: Runt-related transcription factor 2
SCID-mice: Severe combined immunodeficiency mice
TGF-β: Transforming growth factor β
Ti (Ti+/-): Non-oxidized titanium (with/without cells)
TiO2 (TiO2+/-): Oxidized titanium (with/without cells)
Wnt: Wingless-related integration site
14
1. Introduction
Bone reconstruction is essential in cranio-maxillofacial surgery within traumatic head injuries,
guided bone regeneration in implant dentistry, and reconstructive surgery. Bone is one of the most
frequently transplanted tissues1,2. Autologous bone grafting is the preferred treatment for bone
reconstruction due to its osteoconductive and osteoinductive capacity, and transfer of bone cells3,4.
However, harvesting of autologous bone grafts is associated with morbidity in terms of pain, blood
loss, surgical scars, and necrosis. Additionally, it can be difficult to obtain sufficient volume of
grafting material5,6. The use of allografts carries the risk of transplant rejection and transfer of
pathogens. Therefore, researchers seek alternative methods through the use of synthetic or natural
biomaterials in spite of their inferior osteogenic potential. In an attempt to increase the osteogenic
potential of biomaterials, they have been combined with osteogenic cells7,8.
Tissue engineering involves the in vitro seeding of cells onto biomaterial carriers. These
carriers are supposed to support cell adhesion, migration, proliferation, and differentiation, and help
define the 3D structure of the engineered tissue. Therefore, the carrier should have surface
characteristics enabling mesenchymal stem cells (MSCs) to attach, proliferate, and differentiate.
The optimal carrier should be biocompatible, biodegradable, easy to use, cost-effective,
osteoinductive, and osteoconductive to generate new bone formation9-11.
It is preferable that carriers are 3D so the cells are able to proliferate while maintaining their
ability to differentiate. The success of tissue engineering is dependent on oxygen and nutrient
transport to the implanted cells12-14. A limitation of the 3D carriers is that the cells at the interior of
the carrier have decreased nutrient and oxygen transport and decreased removal of waste products15.
To secure a high density of colonizing cells and to promote neovascularization when implanted in
vivo, the carriers should have high porosity, large surface area, mechanical properties, pore size
appropriate for the application, and a highly interconnected porous structure6,13,16-19. Carrier
porosity may be inversely related to the mechanical properties of the material. It is important to find
a balance of securing mechanical needs of the tissue to be replaced and carrier porosity which
allows tissue growth6,20,21.
The ideal engineered cellular bone graft needs to exhibit the following features – the presence
of osteogenic cells to generate new bone directly, an appropriate extracellular matrix to provide an
15
osteoconductive carrier, osteoinductive growth factors to provide signals to the resident cells, and
an adequate blood supply to support cell growth and function22,23.
Studies have investigated different carriers such as collagen, titanium, calcium carbonates,
calcium phosphates, hydroxyapatite (HA), β-tricalcium phosphate (β-TCP), calcium-deficient
hydroxyapatite (CDHA), demineralized bone matrix (DBM), and polylactic acid-polyglycolic acid
copolymer (PLGA) in different combinations to assess mineralized matrix formation and gene
expression24. When measuring the total effect of a carrier combined with MSCs it includes reactions
from both carrier and cells. To do an accurate assessment of the true effect, measurements from
carrier and cells individually are necessary. Unfortunately, most published experiments do not use
control measurements on carrier without cells24.
Available titanium granules are manufactured in non-oxidized titanium (Ti) and oxidized
titanium (TiO2). They are irregular porous granules of commercially pure titanium. Titanium is
biocompatible and non-toxic, even in large doses. The porous properties of Tigran, a titanium
granule product, may lead to ingrowth of newly formed bone by interlocking them with each other
thus creating an uninterrupted structure25. Histology and scanning electron microscopy from both
clinical and experimental studies have revealed ingrowth of bone into implanted porous titanium
granules resulting in osseointegration26,27. Titanium granules provide the necessary initial
mechanical stability for bone ingrowth by keeping the blood clot, thereby initiating bone
development28,29. The osseointegration process of dental titanium implants is dependent of the
surface which mainly comprises titanium oxide30.
In the attempt to enhance the osteogenic potential of biomaterials, different MSCs have been
used. Previous studies have identified a cell population in adipose tissue called adipose derived
mesenchymal stem cells (ADMSCs), similar to the population of bone marrow mesenchymal stem
cells (BMSCs)31,32. ADMSCs have osteogenic potential in vitro31. These cells are able to undergo in
vitro differentiation into cell types found in bone, cartilage, fat, and muscle tissue depending on the
use of induction31. Besides that, human ADMSCs are shown to differentiate into endothelial cells in
vitro and to improve neoangiogenesis and collagen synthesis33,34.
Therefore, porous Ti and TiO2 granules are interesting to evaluate as carriers for ADMSCs for
bone regeneration.
16
2. Hypothesis
The main hypothesis of this thesis is that ADMSC derived osteoblasts can be seeded on
porous Ti and TiO2 granules to produce mineralized matrix.
The specific hypotheses are:
- The optimal carriers for MSC derived osteoblasts in mineralized matrix formation have not
been found.
- ADMSCs have a high osteogenic potential.
- Porous TiO2 granules are superior to porous Ti granules as carrier material for ADMSCs on
formation of mineralized matrix and gene expression in vitro.
- Porous TiO2 granules as carrier material for ADMSCs will result in an increased osteoblast-
covered surface with consequent more osteoid formation compared to porous Ti granules in
vivo.
17
3. Objectives
The overall aim of this thesis is to investigate the usefulness of biomaterial as carriers for
MSC derived osteoblasts for bone regeneration.
The specific aims are:
- To evaluate different carriers – calcium phosphate, titanium, collagen, calcium carbonate,
and PLGA – combined with MSCs in mineralized matrix formation and gene expression in
vitro and in vivo through a systematic review.
- To demonstrate the osteogenic potential of ADMSCs.
- To investigate the effect of porous Ti and TiO2 granules as a carrier material for ADMSCs
on formation of mineralized matrix and gene expression in vitro.
- To investigate the early in vivo effect of ADMSCs seeded on porous Ti and TiO2 granules
implanted subcutaneously in mice.
18
4. Background
4.1 Bone biology
Bone is a highly specialized connective tissue, consisting of cells and a mineralized
extracellular matrix. Bone is composed of an organic component, mainly collagen and
glycosaminoglycans, and an inorganic component in form of HA deposited within the matrix. The
normal architecture of bone is composed of an outer dense and compact zone, cortical bone, and an
inner spongier zone called trabecular bone. Cortical bone represents 80% of the skeleton, but only
20% of the bone turnover, meanwhile trabecular bone comprises 20% of the skeleton, but 80% of
the bone turnover as a result to its high surface area. The construction of bone matrix gives
mechanical stability, enables muscle attachment allowing locomotion, protects vital organs, and
serves as a reservoir for calcium and phosphate35,36.
Bone is a highly dynamic tissue. There are two biological processes in bone growth and
maintenance with coordinated actions between the bone cells – bone modeling and bone
remodeling. Bone modeling is the mechanism, where the skeletal elements form to get the right
morphology and mass during growth. Throughout life this mechanism continues at a low rate
assuring adaptation to changes in mechanical loading. Bone resorption and bone formation occur on
separate surfaces and are uncoupled37. Conversely, bone remodeling is the process that ensures
bone turnover with substitution of primary and infantile bone with stronger secondary bone,
renewal of old bone, repair of microfractures, adaptation to changes in mechanical loading in
smaller degree compared to bone modeling, and regulation of calcium and phosphate homeostasis.
The remodeling process is a coupled and balanced activity between bone resorbing osteoclasts and
bone forming osteoblasts on the same bone surface along specific sites called basic multicellular
units35-37.
4.2 Bone cells
MSCs are pluripotent cells located in bone marrow, muscles, and fat tissue. These cells can
differentiate into specific cells within different tissues including bone, cartilage, muscle, and fat.
The differentiation process is controlled by a number of factors included cytokines and hormones.
19
Differentiation toward an osteoblastic lineage involves among other transforming growth factor β
(TGF-β), bone morphogenic proteins (BMPs), wingless-related integration site (Wnts), and
parathyroid hormones (PTH). The differentiation process of MSCs into osteoblasts can be divided
into the following stages: Proliferation, commitment, differentiation, maturation (extracellular
matrix deposition), mineralization (matrix maturation and mineralization), and termination (Figure
1a). The osteoclasts on the other hand are differentiated from the hematopoietic stem cell (Figure
1b)38.
________________________________________________________________________________
________________________________________________________________________________
Figure 1: Bone cell differentiation
a) Mesenchymal stem cell differentiating toward an osteoblastic lineage. First mesenchymal stem cells proliferate and commit to a differentiation
lineage. Preosteoblasts differentiate into mature osteoblasts, which in turn terminate as osteocytes, lining cells, or undergo apoptosis. b)
Hematopoietic stem cells commit to the myeloid lineage as myeloid commited precursors, which then commit to the osteoclast lineage as osteoclast
precursors. Osteoclast precursors fuse to osteoclasts and become mature resorbing osteoclasts when activated.
20
4.2.1 Osteoblasts
Commitment of the MSCs into the osteoblastic lineage is controlled by extracellular stimuli
as mentioned above. These agents cause a specific response inside the cells with expression of
certain transcription factors. One of the key transcription factors leading to osteoblastic
differentiation is runt-related transcription factor 2 (RUNX2) also known as core-binding factor
subunit α-1 (Cbfa-1). This transcription factor among many others affects the alkaline phosphatase
(ALP) activity35,36,38.
Osteoblasts appear in clusters of 100-400 cells, organized in a multicellular network on the
bone surface. They are characterized by having a round nucleus at the base of the cell, with a strong
basophilic cytoplasma, a prominent Golgi complex, and a well-developed rough endoplasmic
reticulum. The latter contributes to the high level of biosynthesis and secretion activity. The
osteoblast has a high expression level of ALP, which is therefore a characteristic osteoblast marker
and a measure of bone formation activity. The main function of osteoblasts is to deposit non-
calcified bone matrix called osteoid. This is followed by matrix mineralization where the
osteoblasts deposit HA into the organic component. After finishing mineralization, the osteoblasts
undergo apoptosis, turn into lining cells, or differentiate into osteocytes (Figure 1a)35,36,38.
4.2.2 Osteocytes
Osteocytes originate from mature osteoblasts that remain trapped in the newly deposited
mineralized matrix. Osteocytes are the most abundant and longest-living cell type in the adult
skeleton. They appear as star-shaped cells encased in lacunae dispersed throughout the mineralized
matrix. They are connected to lining cells on the bone surface, other osteocytes, and bone marrow
through long dendritic processes running in canaliculi creating a network called syncytium. In this
way osteocytes are linked metabolically and electrically through gap junctions on these dendritic
processes enabling communication between cells. The osteocytes are mechanosensory cells,
responding to mechanical stimulation. As a respond to strain, a fluid-flow through the syncytium
stimulates osteocytes to synthesize factors that regulate bone metabolism by affecting osteoblast
and osteoclast activity35,36,39-41.
21
4.2.3 Osteoclasts
Osteoclasts are giant multinucleated cells with 4-50 nuclei, abundant mitochondria, and
numerous lysosomes. They originate from the hematopoietic stem cell (Figure 1b). The initial
differentiation committing to the myeloid lineage requires transcription factor PU.1. When
stimulated with macrophage colony-stimulating factor (M-CSF) and receptor activator of nuclear
factor κB ligand (RANKL), the cells commit to the osteoclast lineage as osteoclast precursors. Next
step is fusion of cells and activation into mature osteoclasts. Osteoclasts are responsible for
resorbing bone through polarizing, acidifying, and secretion of proteolytic enzymes. Osteoclasts are
characterized by its ruffled border with deep and irregular foldings of the membrane forming a
sealed compartment on the bone surface. After finishing resorbing, the osteoclasts move to another
site of resorption or undergo apoptosis35,36.
4.3 Bone regulators
Regulation of bone remodeling is controlled by various local and systemic factors. PTH,
1,25(OH)2 vitamin D3, and estrogen are some of the essential hormonal regulators of bone
remodeling. Furthermore, several growth factors like TGF-β, Wnt, and BMP have a considerable
role in regulation of physiological bone remodeling (Figure 2)42.
PTH’s main function is to maintain calcium homeostasis and secretion is stimulated by low
serum calcium. Constant high levels of PTH promotes osteoclastogenesis indirectly by inducing the
osteoblasts’ RANKL and M-CSF expression, thereby releasing calcium from the bone to the blood
stream. Additionally, PTH stimulates proliferation and differentiation of preosteoblasts to mature
osteoblasts and inhibits expression of sclerostin from the osteocytes with subsequent increased bone
formation rate42. PTH acts through a G-protein-coupled receptor and activates protein kinase A,
which modulates gene transcription via phosphorylation (Figure 2)38.
Active 1,25(OH)2 vitamin D3 also has an important role in calcium and bone homeostasis. It
induces osteoblastogenesis and enhances mineralization42. Estrogen inhibits bone resorption by
attenuating RANKL- and M-CSF-induced osteoclast differentiation and stimulating osteoclast
apoptosis. Estrogen ensures maintenance of bone formation by inhibiting osteoblast apoptosis42.
Androgens such as testosterone can indirectly inhibit osteoclast activity and bone resorption by
22
affecting the receptor activator of nuclear factor κB (RANK), RANKL, and osteoprotegerin (OPG)
system. Furthermore, androgens increase differentiation of human osteoblastic cells and promote
mineralization42.
An essential event in osteoblast differentiation is the activation of the transcription factor
RUNX2. RUNX2 interacts with many transcriptional activators and repressors and other
coregulatory proteins, which can result in either positively or negatively regulated expression of a
variety of osteoblastic genes. Important differentiation markers are ALP, collagen type 1α1
(COL1α1), bone sialoprotein, osteopontin, osteocalcin (OC) and osteonectin. Besides that, RUNX2
regulates the expression of transcription factor osterix. Osterix interacts with nuclear factor for
activated T cells 2 (NFAT2). In cooperation, osterix and NFAT2 control the transcription of target
genes such as OC, osteopontin, osteonectin, and COL1α138.
TGF-β is a cytokine secreted by osteocytes in bone matrix, which is an essential part in the
control of proliferation, migration, differentiation and survival of bone cells. It has a key role in the
development and maintenance of the skeleton with both positive and negative effects on bone
formation. TGF-β binds to its specific receptors and induces activation of SMAD2/3. When
activated, SMAD2/3 interacts with SMAD4 and affects gene transcription including RUNX2
expression. This leads to an enhancement of bone matrix proteins, ALP activity, and bone
formation (Figure 2)38,42,43.
BMP is another signaling factor from the TGF-β superfamily with a central role in skeletal
development, maintenance of bone homeostasis, and fracture healing. BMPs are expressed in
skeletal tissue and bind as dimers to type-I and type-II serine/threonine receptor kinases, causing
phosphorylation of the intracellular signaling mediators SMAD1/5/8. The activated SMAD1/5/8
then associates with SMAD4 and translocate into the nucleus. This SMAD complex acts together
with RUNX2 in controlling osteoblast-specific gene expression and osteogenic differentiation
(Figure 2)38,42.
Wnts are glycoproteins that bind to Frizzled receptors and lipoprotein receptor-related
proteins 5/6 to transduce the signal. This leads to accumulation of cytoplasmic β-catenin, which
translocates to the nucleus to initiate transcription of target genes through complex formation with
T-cell factor and lymphoid enhancing factor 1. This enhances osteoblastic differentiation and plays
an important role in controlling normal bone homeostasis in osteocytes. When Wnt stimulation is
23
lacking, a destruction complex phosphorylates β-catenin resulting in proteosomal degradation
(Figure 2). With low levels of β-catenin the osteochondroprogenitor cells will tend to differentiate
toward the chondrocytic lineage. Moreover, Wnt-β-catenin signaling results in increased OPG,
which causes inhibition on osteoclast differentiation and bone resorption. Sclerostin is a
glycoprotein, which binds to lipoprotein receptor-related protein 5/6. The sclerostin binding inhibits
Wnt-stimulated bone formation (Figure 2)38,42.
________________________________________________________________________________
________________________________________________________________________________
Figure 2: Bone remodeling regulators
Transforming growth factor β (TGF-β) activates SMAD2/3. Activated SMAD2/3 interacts with SMAD4 and affects gene transcription. Bone
morphogenic protein (BMP) activates SMAD1/5/8. Activated SMAD1/5/8 interacts with SMAD4 and affects gene transcription. Wingless-related
integration site (Wnt) binds to Frizzled receptor, which leads to an accumulation of β-catenin and affection of gene transcription. Sclerostin inhibits
Wnt signaling through binding lipoprotein receptor-related protein 5/6. Parathyroid hormone (PTH) acts through a G-protein-coupled receptor and
activates protein kinase A (PKA), which modulates gene transcription through phosphorylation. Runt-related transcription factor 2 (RUNX2) is one of
the key transcription factors.
24
4.4 Bone remodeling
Bone remodeling is an ongoing process ensuring bone turnover, regulation of calcium and
phosphate homeostasis, reneweal of old bone, and repair of microfractures. The continuous
resorption and rebuilding occur in the approximately 1 million active basic multicellular units found
in healthy adults at any moment35,44. The remodeling process can be divided into five phases – the
activation phase, the resorption phase, the reversal phase, the formation phase, and finally the
termination phase (Figure 3).
________________________________________________________________________________
________________________________________________________________________________
Figure 3: Bone remodeling in trabecular bone
1) Activation of bone remodeling caused by a microfracture and stimulation of receptor activator of nuclear factor κB (RANK) with its ligand
(RANKL). Osteoprotegerin (OPG) regulates the stimulation of RANK. 2) Osteoclasts resorb bone matrix. 3) Reversal cells clean up the organic
matrix and prepare the surface for the osteoblasts. 4) Matrix formation by osteoblasts. 5) Osteoblasts are done creating mineralized matrix.
Osteoblasts turn into lining cells or if embedded in the mineralized matrix they differentiate into osteocytes.
25
4.4.1 The activation phase
The remodeling process can be activated by microfractures, alterations in mechanical loading
sensed by the osteocytes, or releasing of PTH and tumor necrosis factor. These stimuli can either
directly affect osteoclast precursors or indirectly via osteoblasts triggering differentiation and
activation of osteoclasts35.
The cytokine RANKL expressed by osteoblasts is a member of the tumor necrosis factor
superfamily and plays a decisive role in osteoclast differentiation and activation. RANKL interacts
with RANK on osteoclast precursors which stimulate the differentiation, fusion, and activation to
mature osteoclasts (Figure 3)35.
OPG is expressed by osteoblasts and is a decoy receptor for RANK. OPG regulates
stimulation of RANK signaling pathway by competing for RANKL. Binding of RANKL to OPG
will prevent the activation of RANK/RANKL interaction, which inhibits the activating effects on
the osteoclasts. The balance between these interactions will determine the osteoclastic response
(Figure 3)35,45. Several osteotropic factors like PTH and estrogen exert their effects on bone through
affecting the RANK/RANKL/OPG signaling pathway by regulating expression of RANKL and
OPG44.
Another important factor in the bone remodeling activation is the protein TGF-β which
inhibits osteoclastogenesis. When osteocytes undergo apoptosis due to microfractures, matrix
damage, or immobilization, the level of TGF-β is locally reduced. This leads to a disinhibition of
the osteoclastogenesis, resulting in activation of resorption43.
The last step in the activation phase is the osteoblastic degradation of the osteoid layer and the
withdrawal of the lining cells from the bone surface enabling the activated osteoclasts to gain access
to the mineralized surface46.
4.4.2 The resorption phase
The activated osteoclasts attach to the bone surface through vitronectin receptors and initiate
the resorbing phase by releasing acid, ions, and enzymes into the resorption lacunae. Acid to
dissolve the mineral is released from the osteoclast by fusion of acidic intracellular vesicles with the
membrane and by pumping protons and chloride ions out into the lacunae to create hydrochloric
26
acid. Furthermore, osteoclasts secrete different enzymes like cathepsin K and matrix
metalloproteases (collagenase and gelatinase), which degrade the remaining demineralized organic
matrix (Figure 3)47.
4.4.3 The reversal phase
When the resorption is completed, the osteoclasts leave the lacunae and less characterized
reversal cells appear to clean up the organic matrix and prepare the surface for the osteoblasts
(Figure 3)43,46.
4.4.4 The formation phase
While the osteoclasts are resorbing bone matrix, calcium and growth factors like TGF-β and
insulin-like growth factor I are released. This stimulates the recruitment of preosteoblasts to the
resorption site accompanied by increased proliferation and differentiation toward mature
osteoblasts. In addition, osteoclasts release coupling factors aiding the transition to the following
matrix formation by the recruited osteoblasts. The osteoblasts produce and secrete collagen type 1
and other proteins. After the bone matrix has been formed, the mineralization begins with the
deposition of HA (Figure 3)43,46.
An important part of the activation of matrix formation is the Wnt signaling pathway
described earlier (Figure 2). The pathway activates expression of specific osteoblastogenic
transcription factors like RUNX2 resulting in osteoblast proliferation, differentiation, and finally
mineralization. At the same time Wnt signaling increases the OPG/RANKL ratio and thereby
represses osteoclastogenesis48.
Sclerostin is a product of the SOST gene expressed by osteocytes. Sclerostin binds to
lipoprotein receptor-related protein-5/6 receptors on the osteoblasts and thereby inhibit Wnt
signaling pathway leading to decreased bone formation (Figure 2)49. Sclerostin production is
inhibited by PTH and mechanical loading50-52, and increased by calcitonin53.
27
4.4.5 The termination phase
When the osteoblasts are done creating the mineralized bone matrix the remodeling process is
terminated. It is not clarified what signals the osteoblast to stop mineralization. It may be caused by
an increased expression of sclerostin from the osteocytes. When the termination phase is over, the
osteoblasts turn into lining cells or differentiate into osteocytes embedded in the mineralized
matrix36,38,43. The remodeling process is coupled, and imbalances can lead to decreased or increased
bone mass.
4.5 Bone grafting
In the United States, more than 500,000 bone graftings are performed annually. There is a
rising demand of bone grafts and bone substitutes in the United States and therefore bone
regeneration is a very important area of research, especially within cranio-maxillofacial surgery9.
There are several different reasons for bone loss and alveolar process deformity including
traumas, severe periodontitis, traumatic teeth extractions, tumors, infections, malformations, and
edentulism54. A study showed that loss of anterior teeth had a high risk (91%) of alveolar
deformity55. According to Wolff’s Law, bone adapts to mechanical stimulation and if there is no
stimulation, the bone’s architecture will change56. Loss of posterior teeth is highly correlated with
reduced alveolar bone mass in elderly postmenopausal women38,57. Another study showed that a
lower bone mineral density (BMD) in the mandibular cortex corresponds to a lower number of teeth
in female subjects in their seventh decade58. Additionally, people with osteoporosis had a
significant lower number of teeth comparing to people with a normal BMD after adjusting for age
and smoking59. It is therefore reasonable to expect that loss of teeth and alveolar bone resorption is
well-correlated.
Genetic factors and certain lifestyles predispose for systemic bone loss as well as alveolar
bone loss. Smoking, diabetes, insufficient calcium and vitamin D intake are some of the risk factors.
These factors affect periodontal health and predispose to severe periodontitis and following bone
loss38. A study showed an age-related increase of cortical thinning and porosity after the age of 50
years old, why it is reasonable to assume an age-dependent decrease in the alveolar bone38,60.
28
Another study showed a significant decrease of mandibular teeth present for non-osteoporotic
women after the age of 50 years old compared to younger non-osteoporotic women61.
When there is damaged bone tissue e.g. following infections, severe periodontitis, and
extraction of teeth, a natural tissue repair will occur to replace lost or damaged tissue. This
replacement of the original tissue sometimes results in reduced function due to scar tissue or
insufficient bone regeneration and can affect the quality of life with worse chewing function,
esthetic problems, and insecurity. In order to optimize the natural tissue repair, treatment with tissue
regeneration is an option. Tissue regeneration is the attempt to replace lost or damaged tissue with
near-identical structures restoring the original function. Ingrowth and integration including vascular,
neural, and mechanical attachment is crucial to have successful tissue regeneration62.
Autologous bone grafting is a surgical procedure to rebuild damaged bone tissue with the
patient’s own bone graft. This ensures the most important properties for bone grafting material in
terms of osteoinduction, osteoconduction, and osteogenesis, and is still considered the gold
standard9,54,63,64. The most widely used donor site in cranio-maxillofacial surgery is the iliac crest
bone graft. The main advantages of iliac bone grafts are the availability of rather large quantities of
bone including osteoprogenitor cells and growth factors. However, iliac bone harvesting is, as any
other surgical procedure, associated with morbidity and complications9,54,63. A systematic review
reported an overall morbidity rate of 17% within autologous iliac bone grafting in cranio-
maxillofacial interventions. The average length of hospital stay was three days and the predominant
complications were postoperative haematomas, chronic pain, lateral femoral cutaneous nerve injury,
and sensory disturbances. A skilled surgical technique can decrease this morbidity rate54.
Instead of using autologous bone grafts, allografts are an easier accessible graft harvested
from cadavers in various shapes and sizes without donor site morbidity9,63. While allografts are
osteoconductive and can be osteoinductive like autografts, they lack viable osteogenic cells63,64.
As a consequence of the increasing demand for bone grafts, another approach with rising
clinical importance involves natural or synthetic biomaterials with osteoconductive
characteristics63,64. The ideal bone graft substitute is biocompatible, biodegradable,
osteoconductive, osteoinductive, structurally similar to bone, easy to use, and cost-effective9. An
optimal carrier has still not been determined63.
29
Currently osseous defects and insufficiency of alveolar bone are treated with bone substitutes
with promising results. Enhancement and acceleration of functional bone formation can be
increased with the use of osteogenic cells (e.g. ADMSCs and BMSCs) and growth factors (e.g.
TGF-β, BMP, and Wnt). In this way, the osteoinductive effect of the carrier material can be
increased63,65. Advanced tissue engineering with incorporation of BMP derived growth factors and
osteogenic cells demonstrate osteoinductive capacity9.
Porous Ti and TiO2 granules are used in bone tissue engineering with enhanced osteoblast
differentiation with promising results in sinus floor augmentation and implant installation66-68.
However, in a case series investigating porous titanium granules in the treatment of furcation
defects no significant improvements in clinical endpoints were found69. An interesting strategy to
improve the bone tissue engineering capacity of porous titanium granules is to pre-seed the granules
with osteogenic ADMSCs.
30
5. Materials and methods
5.1 Medline search (study I)
We identified relevant medical subject headings (MeSH) words from several articles dealing
with MSCs, tissue engineering, carriers, and bone regeneration. Based on the MeSH words, a
Medline search (Pub Med) was carried out May 2012, including studies published in English from
January 2000 to May 2012. Elleven different MeSH words were used in fourteen different search
combinations (Table 1).
Table 1: Results of the Medline search
Search MeSH word MeSH word MeSH word MeSH word Number
of articles
Presented in
previous search
#1 Mesenchymal
stem cells
Alkaline
phosphatase
Bone
regeneration
Drug carriers 1 0
#2 Mesenchymal
stem cells
Alkaline
phosphatase
Bone
regeneration
Tissue scaffolds 15 0
#3 Mesenchymal
stem cells
Alkaline
phosphatase
Bone
regeneration
Titanium 0 0
#4 Mesenchymal
stem cells
Alkaline
phosphatase
Bone
regeneration
Calcium carbonate 0 0
#5 Mesenchymal
stem cells
Alkaline
phosphatase
Bone
regeneration
Calcium phosphates 13 9
#6 Mesenchymal
stem cells
Alkaline
phosphatase
Bone
regeneration
Collagen 7 2
#7 Mesenchymal
stem cells
Alkaline
phosphatase
Bone
regeneration
Polylactic acid-
polyglycolic acid
copolymer
2 2
#8 Mesenchymal
stem cells
Alkaline
phosphatase
Tissue
engineering
Drug carriers 2 1
#9 Mesenchymal
stem cells
Alkaline
phosphatase
Tissue
engineering
Tissue scaffolds 30 7
#10 Mesenchymal
stem cells
Alkaline
phosphatase
Tissue
engineering
Titanium 0 0
#11 Mesenchymal
stem cells
Alkaline
phosphatase
Tissue
engineering
Calcium carbonate 0 0
#12 Mesenchymal
stem cells
Alkaline
phosphatase
Tissue
engineering
Calcium phosphates 28 20
#13 Mesenchymal
stem cells
Alkaline
phosphatase
Tissue
engineering
Collagen 34 14
#14 Mesenchymal
stem cells
Alkaline
phosphatase
Tissue
engineering
Polylactic acid-
polyglycolic acid
copolymer
7 4
MeSH, medical subject headings.
31
Each search included four MeSH words. The key MeSH words “mesenchymal stem cells” and
“alkaline phosphatase” were used in all searches. They were combined with either “bone
regeneration or tissue engineering”, and one of the following carrier MeSH words “drug carriers,
tissue scaffolds, titanium, calcium carbonate, calcium phosphates, collagen, or polylactic acid-
polyglycolic acid copolymer” (Table 1).
Titles and abstracts were screened, and full-text analysis was performed in relevant
publications dealing with formation of mineralized matrix or gene expression. The following
inclusion and exclusion criteria were used (Table 2).
Table 2: Inclusion and exclusion criteria
Inclusion criteria Exclusion criteria
Published after January 2000 Diffuse statistics
Published in English Other carriers than included
Control (carrier without cells) Size or volume of carrier not presented
One of the five carriers:
- Calcium phosphate
- Titanium
- Collagen
- Calcium carbonate
- PLGA
Number of cells not presented
PLGA, polylactic acid-polyglycolic acid copolymer.
5.2 Titanium carrier (study II & III)
Porous titanium granule material (Natix®) (Tigran Technologies AB, Malmö, Sweden)
(Figure 4) is a bone substitute intended for use in oral and maxillofacial osseous defects for
augmentational purposes. Natix® consists of porous, unalloyed, irregularly shaped titanium
particles, 0.7-1.0 mm in size, with a porosity of about 80%. Natix® granules are provided sterile for
single use by gamma radiation at minimum 25 kGy, ISO 11137-1, and a sterility assurance level of
10-6. Natix® granules are fabricated in two versions, porous Ti and TiO2 granules. However, all
titanium oxidizes when exposed to air. The oxide layer that surrounds a porous Ti granule is only 1
µm. By heating the porous Ti version of Natix® granules, the oxide layer increases and becomes
white. The white granules are composed almost entirely of titanium dioxide, TiO2 granules70.
32
________________________________________________________________________________
________________________________________________________________________________
Figure 4: Porous titanium carrier
Scanning electron microscopy, magnification ×295, of a porous titanium granule (Natix®, Tigran Technologies AB, Malmö, Sweden).
5.3 Culturing of ADMSCs (study II & III)
The current studies were approved by the Danish Animal Experiments Inspectorate. Eight-
week-old Balb/cJ mice were euthanized and abdominal adipose tissue was collected. The tissue was
divided into fine particles and washed in Dulbecco’s Phosphate Buffered Saline w/o Ca2+ and Mg2+
(PBS) (Lonza, Verviers, Belgium), and centrifuged in 5 minutes at speed 2170 rpm at 20°C. After
centrifugation, PBS was removed and the tissue was collagenase treated (7.5 mg collagenase in 10
mL PBS) (SIGMA-ALDRICH, St. Louis, USA). After rotation for 30 minutes at speed 170 rpm at
37°C, collagenase was inactivated by adding 10 mL mouse osteoblast medium (mOB MEM: 435
mL MEM Earle’s w/o Phenol red (Invitrogen, Auckland, New Zealand), 50 mL Fetal Calf Serum
heat inactivated (SIGMA-ALDRICH), 5 mL Penicillin/Streptomycin (Invitrogen), 5 mL glutamax
(Invitrogen), 500 µL L-Ascorbic Acid Phosphate Magnesium Salt n-Hydrate (Wako Chemicals,
Virginia, USA) (50 mg/mL) and 5 mL Glycerol 2-phosphate disodium salt hydrate (SIGMA-
ALDRICH) (1 M)). Hereafter centrifugation in 10 minutes at speed 4340 rpm at 20°C, after which
medium was removed and the precipitate was dissolved in 10 mL mOB MEM, and 106 ADMSCs
33
were plated in 100 mm culture dishes (NUNC, Roskilde, Denmark). Cells were maintained in a
humidified atmosphere at 37°C and 5% CO2. After two weeks, osteoblast-like morphology of the
cells was confirmed by visual inspection in a light microscope.
5.4 Osteogenic potential of ADMSCs (study II)
Expression of osteoblast-specific genes was investigated in order to assess the osteogenic
potential of ADMSCs. Furthermore, ALP activity for ADMSC derived osteoblasts were compared
with mature osteoblasts extracted from mouse bone (bOB) and osteoblasts differentiated from
BMSCs. ADMSCs were cultured for eight weeks in mOB MEM (Invitrogen) and transferred to 6-
well-plates (NUNC) for ALP assay and RNA isolation, and onto 2-Chamber cell culture slides
(SIGMA-ALDRICH) for ALP staining. The cells were cultured for an additional four weeks. The
osteogenic potential of ADMSCs was shown by Susanne Syberg and Lasse Maretty Sørensen as
preliminary study for the in vitro mineralization and gene expression study.
5.4.1 RNA isolation
RNA was isolated using the RNeasy Mini kit (QIAGEN, Hilden, Germany) according to the
manufacturer’s recommendations. The RNA content in the samples was measured using NanoDrop
2000c (Thermo Scientific, Søborg, Denmark) and RNA was stored at -80°C. RNA was isolated at
week 12 to show the osteogenic potential of ADMSCs.
5.4.2 cDNA and polymerase chain reaction
Mastermix was produced according to the manufacturer’s recommendation (Omniscript
Reverse Transcription, QIAGEN) corresponding to 15 μg RNA. 100 μL Mastermix was added to
each tube containing 5 μg RNA. Hereafter the reverse transcription (RT) process was done using an
Eppendorf Mastercycler. To determine the expression of genes related to osteoblast differentiation,
activity and osteogenesis, primers for the following genes were used (Applied Biosystems, Foster
City, USA): Cbfa-1 (RUNX2), OC, peroxisome proliferator-activated receptor-γ1 (PPAR-γ1) and -
γ2 (PPAR-γ2), and housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The
polymerase chain reaction (PCR) was as follows: 10 minutes at 95°C, then 40 cycles with two
34
minutes at 94°C, 45 seconds at hybridizing temperature, 60 seconds at 72°C, and finally 10 minutes
at 72°C.
5.4.3 ALP staining assay
Medium was removed from the chamber slides (SIGMA-ALDRICH). The ADMSCs cell
layer was dried and fixated with 70% ethanol for five minutes, after which they were incubated in
ALP solution (substrate 1: 150 mg variamin blausaltz B and substrate 2: 75 mg sodium-naphthyl-P
dissolved in a buffer, 75 mL 2.1% amino-methyl-propandiol 0.2 M, 15 mL HCl 0.1 M and 210 mL
distilled H2O) (SIGMA-ALDRICH) for 20 minutes at 4°C. Next, the cells were rinsed in tap water
for five minutes and counter-stained in Meyers haematoxylin (SIGMA-ALDRICH) for 30 seconds,
and again rinsed in tap water for 10 minutes after which the coverslip was attached to the slide and
the extent of staining examined in a light microscope.
5.4.4 Protein corrected ALP activity assay
First, cells were lyzed to obtain the intracellularly located ALP. Next, assay buffer, 7 mL
Sigma 211 Buffer (1.5 M), 1 mL MgCl2 (1 M), and 92 mL distilled H2O, was prepared. Substrate
solution was prepared by dissolving p-Nitrophenyl-Phosphat tablet in 10 mL assay buffer. Sigma p-
Nitrophenyl-Phosphat standard solutions were made (0, 25, 50, 75, 100, 125, and 150 nmol/mL) by
diluting with assay buffer and by adding 200 μL NaOH (1.0 M). Assay buffer and substrate solution
was heated in a waterbath at 37°C for 10 minutes. The cell lysate was diluted 25 times (20 μL lysate
+ 480 μL TBS buffer). Prediluted lysate (20 μL), control or standard was transferred into a 96-well-
plate (NUNC) in duplicate. Hereafter 80 μL assay buffer (37°C) and 100 μL substrate was added to
each well, and the plate was incubated at 37°C for 30 minutes. The reaction was terminated by
adding 100 μL NaOH (1.0 M). Absorbance was read at 405 nm.
To determine protein content, the Pierce BCA Protein Assay (Thermo Scientific) was used.
Cell lysate was thawed. The reaction solution, BCA “Working Reagent” composed of 25 mL BCA
reagent A and 500 μL reagent B. An albumin standard was prepared with the following
concentrations: 0, 50, 100, 200, 300, 400, and 500 μg/mL by diluting albumin standard with TBS
buffer. Lysate 50 μL, control or standard was transferred to a 96-well-plate (NUNC) in duplicate.
35
BCA “Working Reagent” (200 μL) was added to each well. The reaction was incubated at 37°C for
60 minutes. After adjusting the plate to room temperature for 10 minutes, the absorbance was read
at 562 nm.
5.5 In vitro mineralization and gene expression (study II)
For the evaluation of the effect of porous Ti and TiO2 granules as carrier material for
ADMSCs, cells were cultured for an initial two weeks in mOB MEM to stimulate differentiation to
osteoblasts. Thereafter, they were trypsinized and transferred to the well-plates with and without a
carrier.
For mineralization assays 24-well-plates (NUNC) were used with inserts (FALCON, Franklin
Lakes, NJ, USA) with a pore size of 3.0 μm. The amount of the carrier, granules in one insert, was
0.20 mL. Four different combinations of cells and carriers were used: Ti granules combined with
4×104 ADMSCs (Ti+) and TiO2 granules combined with 4×104 ADMSCs (TiO2+), and as controls
Ti granules without ADMSCs (Ti-) and TiO2 granules without ADMSCs (TiO2-). As additional
controls mOB MEM with and without ADMSCs were measured.
For RNA isolation 6-well-plates (NUNC) were used with inserts (FALCON) with a pore size
of 3.0 μm. The amount of the carrier granules in one insert was 1.5 mL. Cells were kept in mOB
MEM (Invitrogen) and maintained at 37°C and 5% CO2. The in vitro mineralization and gene
expression study comprised a five-week period (Figure 5).
36
________________________________________________________________________________
________________________________________________________________________________
Figure 5: Time line for the in vitro mineralization and gene expression study
Adipose tissue was collected at day -14 for isolating and culturering of adipose derived mesenchymal stem cells (ADMSCs). Well-plates were
prepared with and without ADMSCs at day 0. Measures of gene expression were done on day 1, 4, and 8. Measures of mineralization with Alizarin
Red S (AR-S) assay were done on day 1, 3, 7, 10. 14, 17, and 21.
5.5.1 Alizarin Red S Assay
The Alizarin Red S (AR-S) assay71-73 was used to determine the amount of mineralized matrix
formed by the osteoblasts. It was performed on day 1, 3, 7, 10, 14, 17, and 21. All test groups (mOB
MEM without cells, ADMSCs, Ti-, Ti+, TiO2-, and TiO2+) in the inserts were fixated in 70%
ethanol for 1 hour. Next, they were stained in 1 mL 40 mM AR-S, pH 4.2 (684.6 mg AR-S
(SIGMA-ALDRICH) dissolved in 50 mL distilled water, pH adjusted to 4.2 with NaOH/HCl) for
10 minutes at room temperature. Hereafter, all test groups were washed 12 times with distilled
water, and then washed with Dulbecco’s Phosphate Buffered Saline with Ca2+ and Mg2+ (PBS+)
(Lonza) for 15 minutes at room temperature with rotation (50 rpm) to reduce non-specific AR-S
staining. After that all test groups were destained in 1 mL 10% Cetylpyridinium Chloride (CPC)
(SIGMA-ALDRICH) (55 g CPC dissolved in 550 mL 10 mM sodium phosphate) for 15 minutes at
room temperature with rotation (150 rpm). Standard dilutions and AR-S-extracts (1 mM AR-S
standard: 34.2 mg AR-S dissolved in 10 mL 10% CPC, from where 1 mL was added another 9 mL
10% CPC, fresh solution for each assay-day) were pipetted into a 96-well-plate (NUNC) (200
37
µL/well in duplicate), and the AR-S concentration was determined on a plate reader at 562 nm
(infinite M200, TECAN).
5.5.2 RNA isolation
RNA was isolated using the RNeasy Mini kit (QIAGEN) according the manufacturer’s
recommendations. The RNA content in the samples was measured using NanoDrop 2000c (Thermo
Scientific) and RNA was stored at -80°C. RNA was isolated at day 1, 4, and 8.
5.5.3 RT-PCR
PCR OneStep High Capacity cDNA RT Kit (Applied Biosystems) was used to prepare 2×RT
master mix (per 20 µL reaction) on ice calculated for 5 µg RNA (1 mL 10×RT Buffer, 0.4 mL 25×
dNTP Mix (100 mM), 1 mL 10×RT Random Primers, 0.5 mL MultiscribeTM Reverse Transcriptase,
2.1 mL Nuclease-free H2O). PCR reactions were performed using a PTC-100TM Programmable
Thermal Controller (MJ Research, Inc.). The program, ABI-RT, was as follows: 10 minutes at
25°C, 120 minutes at 37°C, 5 seconds at 85°C, final extension – hold at 4°C and cDNA RT tubes
(Eppendorf) were stored at -18°C.
5.5.4 Quantitative PCR
A comparative CT (relative standard curve) method was selected for quantitative PCR
performed by OneStep Real Time PCR System (Applied Biosystems). TaqMan Gene Expression
Assays (Applied Biosystems) of ALPL, COL1α1, and RUNX2 were used as target genes and
GAPDH was used as endogenous control. TaqMan Universal PCR master mix 2x (10 µL) (Applied
Biosystems), sterile water Mini-Plasco (6 µL) (B. Braun Melsungen AG, Melsungen, Germany) and
gene assay on-demand Expression mix 20x (1 µL) (Applied Biosystems) were used in order to
prepare 17 µL master mix. Manually prepared cDNA (3 µL) and master mix (17 µL) were pipetted
to each well of 48-well MicroAmp Fast Optical Reaction Plate (Applied Biosystems) to reach a
volume of 20 µL in each well. The plates were covered with MicroAmp 48-Well Optical Adhesive
Film PCR Compatible, DNA/RNA/RNase Free (Applied Biosystems), after which the plates were
38
spinned for 5 minutes at 3575 rpm at 20°C (Heraeus, Labofuge 400 R Centrifuge, Thermo
Scientific) right before measuring.
5.6 Early in vivo effect of ADMSCs (study III)
The ADMSCs were used for evaluating the early in vivo effect seeded on porous Ti and TiO2
granules subcutaneously in mice. ADMSCs were cultured for an initial two weeks in mOB MEM to
stimulate differentiation to osteoblasts. Thereafter, they were trypsinized and transferred to the well-
plates with and without a carrier. Osteoblast phenotype was confirmed in AR-S Assay on day -4
and -1 before implantation and results were similar to study II (Figure 6)74.
________________________________________________________________________________
________________________________________________________________________________
Figure 6: Time line for the early in vivo effect of adipose derived mesenchymal stem cells
implanted subcutaneously in mice
Adipose tissue was collected at day -21 for isolating and culturering of adipose derived mesenchymal stem cells (ADMSCs). Well-plates were
prepared with and without ADMSCs at day -7. Osteoblast phenotype was confirmed on day -4 and -1 before implantation with Alizarin Red S (AR-S)
assay. All surgical interventions were performed at day 0. Two weeks after implantation, all mice were euthanized and samples collected for later
evaluation.
5.6.1 Study design
Twenty-four 12-week-old Balb/cJ mice (Figure 7a) were anesthetized using inhaled Isofluran
(1-2%) (Baxter A/S, Allerød, Denmark). Each mouse was prepared for surgical intervention with
39
application of ophthalmic ointment (Neutral Ophtha) (Ophtha A/S, Gentofte, Denmark),
subcutaneous prophylactic antibiotic 0.05 mL Norostrept vet (200000 ie/mL
benzylpenicillinprocain, 200 mg/mL dihydrostreptomycin) (Scanvet, Fredensborg, Denmark),
shaving and disinfection with 2.5% iodine alcohol. Four subcutaneous pockets on the back of each
mouse were created through skin incisions. The subcutaneous pockets were filled in random order
with 1) mOB MEM without ADMSCs, 2) mOB MEM with ADMSCs, 3) Ti- (12 mice) or TiO2- (12
mice), and 4) Ti+ (12 mice) or TiO2+ (12 mice), and sutured with Prolene 4.0 (Figure 7b). All mice
received postoperatively analgesic 0.04 mL Rimadyl vet (50 mg/mL carprofen) (Pfizer Aps,
Ballerup, Denmark) and 1 mL sterile saline (sodium chloride isotonic SAD 9 mg/mL) (Amgros I/S,
Copenhagen, Denmark) subcutaneously. All mice were kept single caged for four hours at
temperature 26°C, hereafter at temperature 22°C. All mice had a natural ingredient diet ad libitum
and free access to water.
________________________________________________________________________________
________________________________________________________________________________
Figure 7: Study design
a) Twenty-four mice allocated in two groups. Four subcutaneous pockets in each mouse were filled in random order with 1) mouse osteoblast medium
(mOB MEM) without adipose derived mesenchymal stem cells (ADMSCs), 2) mOB MEM with ADMSCs, 3) titanium granules without pre-seeded
ADMSCs (either Ti- (12 mice) or TiO2- (12 mice)), and 4) titanium granules with pre-seeded ADMSCs (either Ti+ (12 mice) or TiO2+ (12 mice)). b)
Surgical intervention: TiO2+ granules inserted in one of the subcutaneous pockets.
40
Two weeks after the in vivo implantation, all mice were euthanized by cervical dislocation
(Figure 6). All four samples from each animal were sampled including surrounding muscle tissue
and fixated in absolute ethanol 20:1. Preparation for non-decalcified specimens were performed at
Sahlgrensska Institute, Department of Biomaterials, University of Gothenburg by sequential
dehydration and embedded in methylmetacrylate75. Approximately 5-9 decalcified sections, 250 µm
thick, perpendicular to the normal anatomy were produced from each embedded specimen by means
of a Leiden saw (KDG-95, Meprotech, Heerhugowaard, The Netherlands). Afterwards, grinding
and polishing of sections to a final thickness of 30-60 µm were completed76,77. All sections were
stained with basic fuchsin 0.3% and toluidine blue 1%.
5.6.2 Quantitative histology
All sections were evaluated using a light microscope (Nikon Eclipse 80i, Nikon, Tokyo,
Japan) equipped with a digital camera (Olympus DP72, Olympus, Tokyo, Japan) connected to a PC
running Visiopharm Integrator System – NewCAST version 5.3.0 interactive stereology software
system (Visiopharm, Hørsholm, Denmark), magnification ×464. The following in-plane parameters
were determined: Granule surface (GS) to granule volume (GV) ratio (GS/GV), the fraction of
granule surface covered with fibrosis (Fb.S/GS), osteoclasts (Oc.S/GS), osteoblasts (Ob.S/GS) and
not evaluable surface (NE.S/GS), and the profile number per granule surface of osteoblasts
(N.Ob/GS) and osteoclasts (N.Oc/GS)78.
Region of interest was defined as the tissue surrounding the carrier material (Figure 8a). A
meander sampling was performed (fraction 100% with random orientation). The aim was to achieve
around 200 points and 200 intersections per specimen for the parameter being studied. Area per
point was 5936 µm2 and area per length was 396.5 µm. These were superimposed to the histological
section in the microscope. Point-grid was used for GV estimation (all black/dark areas were defined
as carrier material), while line-grid was used for GS estimation (total intersections with the granule
surface). An estimation of the in-plane granule surface density (GS/GV) was calculated as
2×GS/(GV×dgrid), where dgrid is the distance between the lines (74.85 µm)79. All granule surface
intersections were defined in one of four categories: Fibrosis (Fb.S), osteoclasts (Oc.S), osteoblasts
(Ob.S), or not evaluable (NE.S) (Figure 8b). Fibrosis was defined as collagen with small spindle
shaped fibroblasts (Figure 8d), large multinucleated cells were defined as osteoclasts (Figure 8d),
41
and groups of connected mono-nucleated cuboidal cells were defined as osteoblasts (Figure 8c). NE
was defined as blurred areas either because of excessive staining or dusty areas because of grinding
and polishing of the granules.
________________________________________________________________________________
________________________________________________________________________________
Figure 8: Cross-section of specimen containing porous oxidized titanium granules with pre-
seeded adipose derived mesenchymal stem cells (TiO2+), stained with basic fuchsin 0.3% and
toluidine blue 1%
a) Region of interest is defined as the tissue surrounding the titanium granules. b) Intersections defined as fibrosis
(Fb.S), osteoclasts (Oc.S), osteoblasts (Ob.S), and not evaluable (NE.S). Magnification ×464. c) Osteoblast-covered
(Ob) TiO2+ surface. d) An osteoclast- (Oc) and fibrosis-covered (Fb) TiO2+ surface.
42
Group allocation was blinded to the evaluator, Morten Dahl, who carried out all
measurements. The coefficient of variation (CV), CVGV=4.0% and CVFb.S=7.5% were fully
acceptable. CVOc.S=55.6%, CVOb.S=73.1%, and CVNE.S=27.7% were relatively high, why profile
number of osteoblasts and osteoclasts on the granule surface were analyzed80. Point-grid and line-
grid were found relatively acceptable due to the few and scattered bone cells in the section.
5.7 Statistics
In study II data from the first part regarding the osteogenic potential are presented as mean
and 95% confidence interval. Differences between BMSCs (n=29), ADMSCs (n=8), and bOB
(n=29) were analyzed with one-way analysis of variance and students T-test for post hoc
comparison. In the second part regarding in vitro mineralization and gene expression data were
provided as mean±SEM (n=5-8 for each condition and time point). All results were analyzed with
one-way analysis of variance and by students T-test for post hoc comparison.
In study III data are provided as median and interquartile range. Comparisons between groups
were performed using Kruskal-Wallis one-way analysis of variance for non-parametric data to test
whether the medians of the four groups were equal. If we found significant differences between the
groups a post hoc Mann-Whitney U Test was performed for pairwise comparison of groups.
Significance levels of 0.05 were used throughout the thesis. All statistical tests were
performed using IBM SPSS Statistics 22.
43
6. Results
6.1 Medline search (study I)
The MeSH search resulted in 80 different articles. Screening of titles and abstracts were done
with the inclusion and exclusion criteria in hand (Table 2), resulting in 51 potentially relevant
publications where full-text analysis was performed. Out of the 51 articles, nine articles met the
criteria and were included in the present study 81-89. Numerical values of defined outcomes were not
presented in any of the publications. Outcomes in all included articles were exclusively presented as
figures. The corresponding authors of the included nine articles were contacted to obtain numerical
values from their studies. Only two out of the nine studies were able to provide us with numerical
values (Table 3)81,82.
In the first study, Kasten et al.81 investigated ectopic in vivo bone formation with different
carriers in combination with and without BMSCs and induced BMSCs. Kasten et al.81 examined the
resorbable CDHA carrier and compared with other biomaterials β-TCP, HA, and DBM in severe
combined immunodeficiency mice (SCID-mice). The outcome measures were ALP activity and
expression of OC as indicators of mineralized matrix formation assessed after four and eight weeks.
They found a significant effect when pre-seeding with BMSCs compared with empty controls, and
no significant effect between BMSCs and induced BMSCs within ALP activity. Additionally, they
found a significantly higher ALP activity when comparing CDHA, β-TCP, and HA with DBM.
Expression of OC was significantly higher for DBM compared with the other biomaterials. No
significant difference was found between the other biomaterials and no effect of pre-seeding with
BMSCs or induced BMSCs within expression of OC. Kasten et al.81 did not find any bone
formation after four weeks. After eight weeks, they confirmed bone formation with immunostaining
in CDHA with BMSCs (1 out of 8), CDHA with induced BMSCs (3 out of 8), β-TCP with BMSCs
(2 out of 8), DBM with BMSCs (2 out of 8), and finally DBM with induced BMSCs (1 out of 4)81.
44
Table 3: Outcome measures from the two included studies
Author Type of cell Type of carrier ALP assay
ng p-nitrophenol/µg protein
4 weeks 8 weeks
ELISA kit, OC
#
4 weeks 8 weeks
Kasten et
al.81
Human
BMSC
2×105
cells/carrier
CDHA
-empty
-with BMSC
-with induced BMSC
β-TCP
-empty
-with BMSC
-with induced BMSC
DBM
-empty
-with BMSC
-with induced BMSC
HA
-empty
-with BMSC
-with induced BMSC
12.54
74.65
94.65
19.67
188.91
132.19
20.59
22.16
8.75
36.65
52.94
34.30
16.70
68.09
126.98
18.51
168.32
92.88
15.74
10.27
12.14
20.65
88.03
280.40
0.739
1.072
1.258
1.089
1.015
0.792
6.984
6.819
7.601
1.075
0.919
0.735
0
0.227
0.703
0.409
0.264
0.081
2.496
2.098
6.728
0
0
0
Kasten et
al.82
Human
BMSC
2×105
cells/carrier
CDHA
-empty
-with BMSC
-with induced BMSC
CDHA with PRP
-empty
-with BMSC
-with induced BMSC
β-TCP
-empty
-with BMSC
-with induced BMSC
β-TCP with PRP
-empty
-with BMSC
-with induced BMSC
12.54
74.65
94.65
23.05
148.90
124.30
19.67
188.91
132.19
26.14
297.09
149.59
16.70
68.09
126.98
13.24
180.41
116.85
18.51
168.32
92.88
19.16
140.68
124.55
0.739
1.072
1.258
0.084
1.577
0.885
1.089
1.015
0.792
0.713
0.729
0.629
0
0.227
0.703
0
1.183
2.672
0.409
0.264
0.081
0
0
0.102 BMSC, bone marrow mesenchymal stem cells; ALP, alkaline phosphatase; OC, osteocalcin; CDHA, calcium-deficient hydroxyapatite; β-TCP, β-
tricalcium phosphate; DBM, demineralized bone matrix; HA, hydroxyapatite; PRP, platelet-rich plasma. # unit not presented.
In the other study, Kasten et al.82 investigated osteogenesis and ectopic in vivo bone formation
in SCID-mice by combining CDHA and β-TCP carriers with and without BMSCs and platelet-rich
plasma (PRP), respectively. They used the same outcome measures in terms of ALP activity and
expression of OC after four and eight weeks. They found a significant increased ALP activity when
pre-seeding with BMSCs compared with empty controls. Additionally, they found an increased
ALP activity for β-TCP seeded with BMSCs compared with β-TCP seeded with induced BMSCs.
No significant difference was found between CDHA seeded with BMSCs and CDHA seeded with
45
induced BMSCs. Furthermore, they found a significant effect on ALP activity when combining
PRP with CDHA carrier both with and without BMSCs. On the other hand, they did not find any
effect of combining PRP with β-TCP. Expression of OC was significantly higher in CDHA carrier
with BMSCs compared with the empty control. No significant difference in OC expression was
found between BMSCs and induced BMSCs in CDHA carrier. Evaluating the OC expression with
the β-TCP carrier no significant differences were found between any of the groups. Kasten et al.82
did not discover any bone formation after four weeks. After eight weeks, they confirmed bone
formation with immunostaining in CDHA with BMSCs (1 out of 8), CDHA with induced BMSCs
(3 out of 8), CDHA with PRP and BMSCs (3 out of 8), and finally β-TCP with BMSCs (2 out of
8)82.
6.2 Osteogenic potential of ADMSCs (study II)
To investigate the osteogenic potential, we analyzed the gene expression of osteoblast-
specific genes (Figure 9a). The gene of the key transcription factor from the osteoblastic lineage
Cbfa-1 (RUNX2) was highly expressed in the ADMSCs indicating osteoblast differentiation. The
adipocyte-related genes PPAR-γ1 and PPAR-γ2 were only expressed weakly and not at all,
respectively (Figure 9a). We found no expression of OC, which is the last step before
mineralization (Figure 9a). The ADMSCs stained strongly for ALP, which is specific for
osteoblastic cells (Figure 9b).
When comparing the ALP activity of bOB, ADMSC derived osteoblasts, and BMSC derived
osteoblasts, we found no significant difference between the adipose tissue derived osteoblasts and
the other two osteoblastic cell lineages (p=1.00 and p=0.947, respectively) (Figure 9c). However,
the bone marrow derived osteoblasts had significantly lower ALP activity than the bone derived
osteoblasts (p=0.015) (Figure 9c).
46
________________________________________________________________________________
________________________________________________________________________________
Figure 9: Osteogenic potential of adipose derived mesenchymal stem cells
a) Reverse transcription polymerase chain reaction (RT-PCR) assay for core-binding factor subunit α-1 (Cbfa-1), osteocalcin (OC), peroxisome
proliferator-activated receptor-γ1 and -γ2 (PPAR-γ1 and -γ2), and housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Adipose
derived mesenchymal stem cells (ADMSCs) highly express Cbfa-1 and small amounts of PPAR-γ1 expression. ADMSCs show strong osteoblast
phenotype. b) Light microscopy, magnification ×50. ADMSCs stain positively for alkaline phosphatase (ALP), shown by arrows. c) ALP acitivty
presented as mean with 95% confidence interval for bone marrow mesenchymal stem cells (BMSC) (n=29), ADMSCs (n=8), and mature osteoblasts
extracted from mouse bone (bOB) (n=29). Numerical values can be found in Supplementary data 14.1.
6.3 In vitro mineralization and gene expression (study II)
6.3.1 Mineralized matrix
The quantity of calcium per insert was determined at day 1, 3, 7, 10, 14, 17, and 21 using AR-
S staining. Total calcium content was significantly higher when ADMSCs were seeded compared to
mOB MEM without cells (Figure 10a). Background level of absorbance for the porous Ti- granules
was higher compared to total calcium content for ADMSCs. When pre-seeding of ADMSCs on
47
porous Ti granules (Ti+) a significantly higher total calcium content was found compared with Ti-
granules (Figure 10b). The porous TiO2- granules has an even higher background level of
absorbance compared to porous Ti+ granules, and when pre-seeded with ADMSCs (TiO2+) a
significantly higher total calcium content was found compared with TiO2- granules (Figure 10c).
________________________________________________________________________________
________________________________________________________________________________
Figure 10: Alizarin Red S – mineralized matrix
Total calcium contents as absorbance for a) osteoblast medium with adipose derived mesenchymal stem cells (ADMSCs) and without ADMSCs
(medium), b) porous titanium granules with pre-seeded ADMSCs (Ti+) and without pre-seeded ADMSCs (Ti-), and c) porous oxidized titanium
granules with pre-seeded ADMSCs (TiO2+) and without pre-seeded ADMSCs (TiO2-). Data are presented as mean±SEM (n=8 for each condition and
time point). p-level below 0.05 is considered as significant (*: p<0.05; **: p<0.01; ***: p<0.001, comparing the two treatment groups). Numerical
values can be found in Supplementary data 14.2.
To evaluate the effect of ADMSCs alone and on the two porous Ti and TiO2 granules we
plotted the difference of the paired curves above (Figure 11). Total calcium content was presented
as mean±SEM after correction for the values of controls without ADMSCs. Total calcium content
from ADMSCs increased very slowly toward day 7, after which it stabilized. After an initial
increase in the total calcium content on ADMSCs seeded on Ti granules, the curve stabilized.
ADMSCs seeded on TiO2 granules showed an initial increase followed by an even larger increase
48
after day 7. The total calcium content for cells seeded on TiO2 granules appeared to peak between
day 14 and 21 (Figure 11).
________________________________________________________________________________
________________________________________________________________________________
Figure 11: Alizarin Red S – mineralized matrix
Total calcium contents as absorbance for adipose derived mesenchymal stem cells (ADMSCs) and porous titanium granules with pre-seeded
ADMSCs (Ti and TiO2) after correction for the values of the controls without ADMSCs. Data are presented as mean±SEM (n=8 for each condition
and time point). p-level below 0.05 is considered as significant (*: p<0.05; **: p<0.01; ***: p<0.001, comparing treatment groups TiO2 with Ti and
ADMSCs). Numerical values can be found in Supplementary data 14.3.
6.3.2 Gene expression of osteoblast-specific genes
Quantitative PCR was performed to evaluate the gene expression of ALPL, COL1α1, and
RUNX2 at day 1, 4, and 8 for ADMSCs in mOB MEM, and porous Ti+ and TiO2+ granules (Figure
12). Already at day 1, a significantly increased gene expression of ALPL was noticed for TiO2+
49
granules compared with ADMSCs. At day 4, a significantly higher ALPL expression was found in
Ti+ granules compared with ADMSCs (Figure 12a).
The expression of COL1α1 for ADMSCs in mOB MEM was significantly higher than for
porous Ti+ and TiO2+ granules at all time points. The COL1α1 expression decreased over time for
all three groups. At day 4 and 8, a significantly higher COL1α1 expression was found in porous Ti+
granules compared with TiO2+ granules (Figure 12b).
Expression of RUNX2 was significantly increased for TiO2+ granules at day 1 compared with
ADMSCs and porous Ti+ granules. At day 4 and 8, RUNX2 was comparable in all three groups
(Figure 12c).
________________________________________________________________________________
________________________________________________________________________________
Figure 12: Gene expression of osteoblast-specific genes
a) Relative ALPL gene expression for osteoblast medium with adipose derived mesenchymal stem cells (ADMSCs), porous titanium granules with
pre-seeded ADMSCs (Ti+), and porous oxidized titanium granules with pre-seeded ADMSCs (TiO2+). Data are presented as mean±SEM (n=8 for
each condition and time point, except for ADMSCs at day 8, n=7, and TiO2+ at day 4, n=7, and day 8, n=5). b) Relative collagen type 1α1 (COL1α1)
gene expression for ADMSCs, Ti+, and TiO2+. Data are presented as mean±SEM (n=8 for each condition and time point). c) Relative runt-related
transcription factor 2 (RUNX2) gene expression for ADMSCs, Ti+, and TiO2+. Data are presented as mean±SEM (n=8 for each condition and time
point, except for ADMSCs at day 8, n=7). p-level below 0.05 is considered as significant (*: p<0.05; **: p<0.01; ***: p<0.001, comparing treatment
groups). Numerical values can be found in Supplementary data 14.4.
50
6.4 Early in vivo effect of ADMSCs (study III)
In sections from the samples where only mOB MEM with and without ADMSCs were
implanted we did not find any bone cells or site effects from the surgical intervention. Statistical
tests therefore only include the four groups Ti-, TiO2-, Ti+, and TiO2+.
6.4.1 Surface and volume estimation of the carrier
In order to determine the volume of porous Ti and TiO2 granules implanted subcutaneously,
GS/GV was estimated from histomorphometric analysis meander sampling (Table 4). There was no
statistically significant difference in GS/GV across the four groups (p=0.330).
Table 4: Surface and volume estimation, fractional surfaces, and numbers of osteoblast and
osteoclast profiles
Without ADMSCs With ADMSCs
Ti- n=12
TiO2- n=12
p-value Ti- vs. TiO2-
Ti+ n=12
TiO2+ n=11
p-value Ti+ vs. TiO2+
p-value Across all groups#
GS/GV 0.021
[0.020-0.237]
0.019
[0.017-0.022]
--- 0.018
[0.015-0.021]
0.018
[0.015-0.019]
--- 0.330
Ob.S/GS 0.000
[0.000-0.000]
0.0021
[0.0002-0.0056]
0.002 0.000
[0.000-0.000]
0.0028
[0.000-0.0055]
0.002 <0.001
Oc.S/GS 0.071
[0.061-0.089]
0.124
[0.084-0.155]
0.003 0.079
[0.053-0.094]
0.090
[0.084-0.144]
0.036 0.006
Fb.S/GS 0.91
[0.88-0.92]
0.85
[0.83-0.87]
0.001 0.88
[0.86-0.91]
0.83
[0.80-0.89]
0.042 0.002
NE.S/GS 0.015
[0.003-0.037]
0.012
[0.002-0.051]
--- 0.045
[0.006-0.060]
0.0239
[0.009-0.050]
--- 0.495
N.Ob/GS 0.000
[0.000-0.000]
0.013
[0.002-0.038]
0.003 0.000
[0.000-0.000]
0.022
[0.004-0.072]
0.001 <0.001
N.Oc/GS 0.131
[0.107-0.159]
0.258
[0.144-0.325]
0.009 0.152
[0.086-0.165]
0.182
[0.165-0.277]
0.023 0.004
GS/GV, granule surface to volume ratio; Ob.S/GS, fractional osteoblast surface ratio; Oc.S/GS, fractional osteoclast surface ratio; Fb.S/GS, fractional
fibrosis surface ratio; NE.S/GS, fractional not evaluable surface ratio; N.Ob/GS, number of osteoblast profiles per granule surface; N.Oc/GS, number
of osteoclast profiles per granule surface; ADMSCs, adipose derived mesenchymal stem cells; Ti-, non-oxidized porous titanium granules without
ADMSCs; TiO2-, oxidized porous titanium granules without ADMSCs; Ti+, non-oxidized porous titanium granules with ADMSCs; TiO2+, oxidized
porous titanium granules with ADMSCs. No significant difference could be found between Ti- vs. Ti+ or TiO2- vs. TiO2+. # Kruskal-Wallis one-way
analysis of variance.
51
6.4.2 Fractional surfaces
When evaluating whether the titanium surface as well as the pre-seeding of ADMSCs on the
granules affected the fractional surfaces covered by osteoblasts, we found significant differences
between the groups. Comparing the carriers without ADMSCs, osteoblast-covered surface
(Ob.S/GS) was significantly higher in TiO2- than Ti- (p=0.002). The same was the case for the
carriers with ADMSCs, where Ob.S/GS was significantly higher in TiO2+ than Ti+ (p=0.002)
(Figure 13a, Table 4). The osteoclast-covered surface (Oc.S/GS) was also significantly higher in
TiO2- compared with Ti- (p=0.003) and in TiO2+ compared with Ti+ (p=0.036) (Figure 13b, Table
4). On the contrary, fibroblast-covered surface (Fb.S/GS) was significantly more common in Ti-
compared with TiO2- (p=0.001) and in Ti+ compared with TiO2+ (p=0.042) (Figure 13c, Table 4).
Comparing the individual carriers with and without cells, we found no significant differences
between Ti- vs. Ti+ or TiO2- vs. TiO2+ evaluating Ob.S/GS (p=0.952 and p=0.877, respectively),
Oc.S/GS (p=0.603 and p=0.667, respectively), and Fb.S/GS (p=0.094 and 0.758, respectively).
Finally, we wanted to control whether there was difference in not evaluable granule surfaces
between the titanium surface and pre-seeding of ADMSCs. There was no significant difference in
NE.S/GS across the four groups (p=0.495) (Table 4).
52
________________________________________________________________________________
________________________________________________________________________________
Figure 13: Fractional surfaces
Fractional surface ratio of a) osteoblasts (Ob.S/GS), b) osteoclasts (Oc.S/GS), and c) fibrosis (Fb.S/GS) for porous titanium granules with (Ti+ and
TiO2+) and without (Ti- and TiO2-) adipose derived mesenchymal stem cells.
6.4.3 Number of osteoblast and osteoclast profiles
Investigating if the titanium surface as well as the pre-seeding of ADMSCs on the granules
affected the number of osteoblasts covering the granule surface, we found significant differences
between the groups. Comparing the carriers without ADMSCs, number of osteoblasts (N.Ob/GS)
was significantly higher in TiO2- than Ti- (p=0.003). The same was the case for the carriers with
ADMSCs, where N.Ob/GS was significantly higher in TiO2+ than Ti+ (p=0.001) (Figure 14a, Table
4). The number of osteoclasts covering the granule surface (N.Oc/GS) was also significantly higher
in TiO2- compared with Ti- (p=0.009) and in TiO2+ compared with Ti+ (p=0.023) (Figure 14b,
53
Table 4). In contrast no significant differences in N.Ob/GS and N.Oc/GS were found between Ti-
vs. Ti+ (p=0.616 and p=0.686, respectively) and TiO2- vs. TiO2+ (p=0.665 and p=0.356,
respectively).
________________________________________________________________________________
________________________________________________________________________________
Figure 14: Number of osteoblast and osteoclast profiles
Number of a) osteoblasts per granule surface (N.Ob/GS) and b) osteoclasts per granule surface (N.Oc/GS) for porous titanium granules with (Ti+ and
TiO2+) and without (Ti- and TiO2-) adipose derived mesenchymal stem cells.
54
7. Discussion
The overall aim of this thesis was to investigate the usefulness of biomaterial as carriers for
MSC derived osteoblasts for bone regeneration through a systematic review, an in vitro study, and
an in vivo study.
In the systematic review, numerical values of the outcomes of mineralization and gene
expression of MSCs combined with a carrier were only available from two studies where carriers
and cells were tested separately81,82. Due to the sparse data, we were unable to conduct a meta-
analysis and instead we did a descriptive analysis of the two included studies.
In the first study, Kasten et al.81 concluded an overall similarity in ALP activity, expression of
OC, and bone formation between CDHA and β-TCP as carrier matieral in ectopic in vivo bone
formation. They found no effect on bone formation or osteogenic differentiation when BMSCs were
cultured in osteoblast medium for two weeks prior to the in vivo implantation. Expression of OC
was very low in all biomaterials except for DBM, where the higher expression is probably due to
DBM containing OC itself81. In the other study, Kasten et al.82 concluded that CDHA carrier
combined with PRP and BMSCs had a positive effect on osteogenic differentiation, while there was
no improvement when combining β-TCP with PRP82.
Expression of OC in the two studies was low, which may be explained by insufficient RNA
isolation from the samples, or by an insensitive assay with difficulties in detechting the expression
of OC. Furthermore, they used human derived BMSCs from donors aged 58-79 years. Human
derived BMSCs obtained from this age group have an inferior osteogenic potential compared with
younger donors. The human derived BMSCs were collected in three heterogenic cell pools with
high variability within the different assays, which can affect the obtained results.
Unfortunately, different carriers were superior depending on which outcome measures Kasten
et al.81,82 used and hence no specific carrier was clearly superior to the others. Combined with the
very limited amount of thorough studies available this makes it difficult to conclude anything
regarding the optimal carrier.
55
In the in vitro study, we isolated ADMSCs and showed that these cells could be differentiated
into functional osteoblasts. This was confirmed in a preliminary study that differentiated ADMSCs
highly expressed Cbfa-1 in RT-PCR, stained positive for ALP, and showed a high ALP activity.
In the in vitro mineralization and gene expression study, we showed that ADMSCs
differentiated toward an osteoblastic lineage were able to produce mineralized matrix in smaller
amounts when seeded without a carrier and to a higher degree when seeded on porous titanium
granules (Ti and TiO2 granules). The primary finding from this study was a significantly higher
amount of matrix formation when ADMSCs were seeded on porous TiO2 granules compared with
porous Ti granules. Another interesting finding, though not statistically significant for all values,
was that the numerical values for total calcium content were increased for all days for porous TiO2
granules compared with Ti granules. It is also worth noticing that the porous Ti and TiO2 granules
without cells have a relatively high background level of absorbance (non-specific AR-S staining),
most pronounced in porous TiO2 granules, which emphasizes the importance of measuring the
carrier without cells.
With regard to the relative gene expression our results indicate an initially higher ALPL
expression for ADMSCs seeded on porous Ti and TiO2 granules compared with ADMSCs without
a carrier. At day 1, the RUNX2 expression was significantly increased for ADMSCs seeded on
porous TiO2 granules compared with ADMSCs seeded on porous Ti granules and ADMSCs without
a carrier. These results suggest a better osteogenic differentiation of ADMSCs when seeded on
titanium carrier especially porous TiO2 granules. On the other hand, COL1α1 expression was
significantly higher for ADMSCs seeded without a carrier compared with ADMSCs seeded on a
titanium carrier.
In the in vivo study, we investigated the early effect of ectopic implantion of porous Ti and
TiO2 granules with and without osteogenic ADMSCs in Balb/cJ mice. It allows us to investigate the
use of ADMSCs and porous Ti and TiO2 granules under non-loaded conditions and without the
presence of growth factors. Additionally, the mouse model allows a cost effective in vivo model for
comparison of different biomaterials.
We had expected osteoid formation after two weeks of ectopic in vivo implantation because of
the high metabolic index of mice90. However, there were no signs of osteoid formation around the
56
granules after two weeks. In contrast, Supronowicz et al.91 found bone matrix within 2 weeks using
DBM with ADMSCs implanted intramuscular in rats91. Their setup may offer more osteoinductive
factors both from the increased vascularization in muscle tissue and from the DBM carrier,
explaning why they were able to show bone formation.
We found that the use of porous TiO2 granules was superior to porous Ti granules in terms of
more osteoblasts and osteoclasts on the granule surface. Initially we evaluated the fractional surface
of the carrier covered by osteoblasts and osteoclasts and found it significantly higher in TiO2
granules compared with Ti granules. When analyzing these results, it is important to take the
coefficient of variation into account. Due to rather high coefficient of variation for both osteoblasts
and osteoclasts we decided to count the profile number of osteoblasts and osteoclasts as well. The
porous TiO2 carrier was still superior with significantly higher numbers of osteoblasts and
osteoclasts than the Ti granules. In the comparison of the individual carriers, with and without
ADMSCs, there was no effect on the appearance of osteoblasts and osteoclasts when pre-seeding
the carriers with ADMSCs.
Many in vitro and in vivo studies have shown promising results within mineralized matrix
formation, bone formation, and gene expression of osteoblast-specific genes of carriers combined
with osteogenic cells6,15,19,92-99. Unfortunately, many of these publications did not reveal control
measurements of the carrier itself and cells separately. Hence we cannot estimate the true effect of
the different osteogenic cells or the osteoinductive and osteoconductive effect of the carrier
material. Our in vitro74 and in vivo study as well as Kasten et al.81,82 demonstrate a surprisingly high
effect of the carriers without cells. Studies with included control measurements of the carrier
without osteogenic MSCs enable us to investigate the actual effect of the carrier and the osteogenic
MSCs, respectively. For this reason, it should be mandatory to measure the carrier both with and
without seeding of MSCs to achieve an accurate assessment.
When comparing porous Ti and TiO2 granules, Sabetrasekh et al.100 found higher porosity,
larger pore size, higher surface area-to-volume ratio, and a greater compressive strength of porous
TiO2 granules. Sabetrasekh et al.100 also compared proliferation activity of human MSCs on porous
Ti and TiO2 granules. They found that the proliferation rate was similar on day 1 whereas it was
57
significantly higher in porous TiO2 granules on day 3100. These results are well in accordance with
our findings of porous TiO2 granules being superior in promoting mineralized matrix formation in
vitro, and being a more attractive surface for recruiting osteoblasts from the blood stream and
therefore more osteoinductive in vivo. A likely explanation to the above is the increased porosity,
larger pores, and higher surface area-to-volume ratio of the heat treated TiO2 granules.
Different studies have showed osteogenic potential of ADMSCs differentiated into osteoblasts
in combination with different carriers. Supronowicz et al.91 discovered increased bone matrix
formation in vitro and in vivo when combining ADMSCs with DBM91. Another study101 found
improved osteogenic regeneration in rabbits when using osteoblast-differentiated ADMSCs
compared with controls and undifferentiated ADMSCs101. Marini et al.102 compared ADMSCs
seeded on either nanostructured titanium or polystyrene and concluded that although not superior to
polystyrene, nanostructured titanium still functioned as a basis for ADMSCs to differentiate into
osteoblasts and produce bone matrix102. However, this study did not include control measurements
of the nanostructure titanium carrier without the ADMSCs to evaluate the actual effect of adding
the stem cells.
In our in vitro and in vivo study, we did control measurements of the carriers without pre-
seeding of ADMSCs. In our in vitro study, we found a significantly higher mineralized matrix
formation with the pre-seeding of ADMSCs on porous TiO2 granules. However, in our in vivo
study, we did not find any effect of pre-seeding with ADMSCs, neither on the porous Ti or TiO2
granules. A possible explanation is that the overall amount of bone cells on the carrier material was
rather low. Another reason may be that the carrier itself plays the most important role in recruiting
and maintaining the osteoblasts on the surface and therefore the addition of ADMSCs does not
contribute significant changes.
7.1 Strengths and limitations (Paper I)
A strength in our review was the objective systematic search based on predefined criteria and
different combinations of relevant MeSH words. This reduces the likelihood of selection bias when
searching for articles.
58
A limitation is that we only found nine published studies that included control measurements
of their carrier material without MSCs, and only two of these were able to provide us with
numerical data. Due to the limited amount of data we were unable to perform the statistics needed
to do a regular meta-analysis. We could have compared the studies without control measurement to
see if a tendency could be obtained toward a better carrier material, however, we considered it
would result in assumptions based on insufficient information. Instead we had to suffice with a
descriptive review of the data at hand.
If we were going to do a review again, we would have to consider whether we are using the
right MeSH words. Maybe the addition of the fourth MeSH word regarding the carrier limited the
number of published studies too much, and this may explain why we did not found any publications
with porous Ti and TiO2 granules.
7.2 Strengths and limitations (Paper II)
A strength of our in vitro study is the systematic use of control measurements for each
condition and time point regarding to our AR-S mineralization assay. This enables us to assess the
pure effect of the ADMSCs when seeded on a titanium carrier. Another strength is that we culture
the ADMSCs in inserts with and without porous Ti and TiO2 granules instead of culturing the
ADMSCs directly in the wells. In this way, we can change the mOB MEM easier with reduced risk
of disturbing the cells. The 3.0 μm pore size of the inserts is sufficiently small to avoid the risk of
cells escaping through the net.
A limitation in the in vitro study is that we did not estimate cell number to evaluate whether
the mineralized matrix formation was caused by an increase in the number of cells or mostly by the
osteogenic effectiveness of the differentiated cells. Another limitation is that the AR-S staining
could be bound to the granules and give a false positive signal. To compensate for this, we
optimized the washing stage with 12 repeats to minimize the excess dye in the granules.
Additionally, it is uncertain if we have isolated all RNA in ADMSCs seeded on porous titanium
granules, since it is possible that cells can hide in the connected pores of the granules porosity.
59
We did not use control measurement regarding to relative gene expression, because we did
not find it relevant in isolating RNA, when there were no cells. We do not think the carrier itself can
affect the result regarding to gene expression in any way in vitro.
7.3 Strengths and limitations (Manuscript III)
A strength of our in vivo study is the systematic use of control measurements both with mOB
MEM alone and porous Ti and TiO2 granules without pre-seeded ADMSCs. Another strength is the
blinded evaluation of the cells on the surface of the carriers.
A limitation of the study is that the volume of carrier material was only approximately the
same in each implanted specimen. This increases the possible variation between the samples.
Due to our relatively high coefficient of variation regarding to intersections of osteoblasts and
osteoclasts, we chose to count profile numbers of osteoblasts and osteoclasts on the granule surface.
This induces the possibility of sampling bias when we count the osteoblast and osteoclast profiles
on the carrier surface. Solely because osteoclasts are considerably larger cells than osteoblasts the
number of osteoclast profiles will tend to be overestimated. Since we are not comparing the number
of osteoblasts and osteoclasts this does not have any impact on our conclusions. Further increasing
of test-lines will not change the relative high coefficient of variation substantially.
An additional challenge when using porous Ti and TiO2 granules is that our decalcified
sections were 250 µm thick to ensure the sections stability and to avoid fractures. Evaluation of the
tissue and cells surrounding the granule surface on 250 µm thick sections is almost impossible and
therefore we had the sections grinded and polished to improve visualization. Unfortunately, this
includes the risk of losing the part of the section with the granule that we were trying to evaluate.
Moreover, we could not define region of interest more precisely than the tissue surrounding the
carrier material.
For future experiments, we could use SCID-mice with human ADMSCs making it possible to
locate the seeded cells with immunostaining. Furthermore, prolonging the in vivo implantation in
our animal model to 8 weeks might have revealed mineralized matrix formation.
60
7.4 Implications
An implication of our studies is that to ensure an accurate evaluation of the effect of
osteogenic MSCs and different carriers, it is crucial that future studies include control
measurements of carriers and cells separately as well as together.
Another implication of our studies is that ADMSCs have an osteogenic potential and because
they are easy accessible with a lower morbidity compared with harvesting of BMSCs, it is worth
considering them for future use.
Finally, our studies have forwarded understanding of the complex area of using porous Ti and
TiO2 granules in combination with ADMSCs in vitro and in vivo for bone regeneration purposes.
Since our studies suggest that porous TiO2 granules are superior to Ti granules, this promising
carrier material should be further investigated in more detailed future studies.
61
8. Conclusion
Based on the findings in this thesis we can conclude that:
- An outcome comprises reactions from the carrier as well as from the cells, why
measurements of each of these – carrier and cells – are deemed necessary for evaluation.
More thorough studies with proper controls are warranted for evaluation of an optimal
carrier.
- ADMSCs are osteogenic with similar ALP activity as BMSCs and bOB.
- ADMSC derived osteoblasts are capable of producing mineralized matrix when cultured on
porous Ti and TiO2 granules with significantly higher mineralized matrix formation for the
porous TiO2 granules. Furthermore, ADMSC derived osteoblasts seeded on porous Ti and
TiO2 granules are able to express osteoblast-specific genes.
- Porous TiO2 granules result in an increased osteoblast-covered surface, indicating better
osteoinductive and osteoconductive conditions compared to porous Ti granules. There is no
additive effect of pre-seeding with ADMSCs on porous titanium granules.
Thus, based on our findings, TiO2 granules may be a promising bone grafting material. The
use of ADMSCs seeded on porous TiO2 granules have to be studied further to document the
potential use in vivo.
62
9. Perspectives
Bone regeneration within tissue engineering is a complex treatment. Although, many different
carriers have shown promising results, the best carrier material has not been found. In this thesis we
have shown that porous TiO2 granules are a promising carrier. Nevertheless, more studies are
warranted to document the potential use in vivo for a longer period. It is relevant to do an in vivo
study with histomorphometry after 2, 4, 8, and 12 weeks for evaluation of new bone formation.
Furthermore, by using human derived ADMSCs combined with porous TiO2 granules implanted in
SCID-mice, it would enable us to locate the cultured cells with immunostaining to evaluate the
actual effect of seeding osteogenic cells.
Tissue engineering is an interesting alternative to autologous bone grafts within bone
regeneration in maxillofacial surgery. In the clinic, it would be an advantage to create a 3D template
similar to the patient’s bone defect and use it to design the carrier for implantation.
In the future, more focus on the 3D physiological environment for the cells is relevant. This
could be done by creating 3D collagen gels consisting of several layers of cells, which creates a
gradient between the cells, and facilitate cell to cell interaction mimicking physiological conditions.
Moreover, 3D collagen gels have the advantage of designing each layer with different cells and
including hydroxyapatite in the bottom and top layer to increase the mechanical strength.
63
10. References
1 Marx RE: Bone harvest from the posterior ilium. Atlas Oral Maxillofac Surg Clin North Am 13(2):109-118,
2005.
2 Montespan F, Deschaseaux F, Sensébé L, Carosella ED, Rouas-Freiss N: Osteodifferentiated mesenchymal stem
cells from bone marrow and adipose tissue express HLA-G and display immunomodulatory properties in HLA-
mismatched settings: implications in bone repair therapy. J Immunol Res 2014. Epub 2014 Apr 30.
3 Goldberg VM, Stevenson S. Natural history of autografts and allografts. Clin Orthop Relat Res (225):7-16, 1987.
4 Behnia H, Khojasteh A, Soleimani M, Tehranchi A, Atashi A. Repair of alveolar cleft defect with mesenchymal
stem cells and platelet derived growth factors: A preliminary report. J Craniomaxillo fac Surg 40(1):2-7, 2012.
5 Younger EM, Chapman MW: Morbidity at bone graft donor sites. J Orthop Trauma 3(3):192-195, 1989.
6 Martins AM, Pham QP, Malafaya PB, Raphael RM, Kasper FK, Reis RL, Mikos AG: Natural stimulus
responsive scaffolds/cells for bone tissue engineering influence of lysozyme upon scaffold degradation and
osteogenic differentiation of cultured marrow stromal cells induced by CaP coatings. Tissue Eng Part A
15(8):1953-1963, 2009.
7 Chen J, Sotome S, Wang J, Orii H, Uemura T, Shinomiya K: Correlation of in vivo bone formation capability
and in vitro differentiation of human bone marrow stromal cells. J Med Dent Sci 52(1):27-34, 2005.
8 Seitz S, Ern K, Lamper G, Docheva D, Drosse I, Milz S, Mutschler W, Schieker M: Influence of in vitro
cultivation on the integration of cell-matrix constructs after subcutaneous implantation. Tissue Eng 13(5):1059-
1067, 2007.
9 Greenwald AS, Boden SD, Goldberg VM, Khan Y, Laurencin CT, Rosier RN: Bone-graft substitutes: facts,
fictions, and applications. J Bone Joint Surg Am 83-A Suppl 2 Pt 2:98-103, 2001.
10 Leong NL, Jiang J, Lu HH: Polymer-ceramic composite scaffold induces osteogenic differentiation of human
mesenchymal stem cells. Conf Proc IEEE Eng Med Biol Soc 1:2651-2654, 2006.
11 Ben-David D, Kizhner T, Livne E, Srouji S: A tissue-like construct of human bone marrow MSCs composite
scaffold support in vivo ectopic bone formation. J Tissue Eng Regen Med 4(1):30-37, 2010.
64
12 Laschke MW, Harder Y, Amon M, Martin I, Farhadi J, Ring A, Torio-Padron N, Schramm R, Rücker M, Junker
D, Häufel JM, Carvalho C, Heberer M, Germann G, Vollmar B, Menger MD: Angiogenesis in tissue
engineering: breathing life into constructed tissue substitutes. Tissue Eng 12(8):2093-2104, 2006.
13 Schumann P, Tavassol F, Lindhorst D, Stuehmer C, Bormann KH, Kampmann A, Mülhaupt R, Laschke MW,
Menger MD, Gellrich NC, Rücker M: Consequences of seeded cell type on vascularization of tissue engineering
constructs in vivo. Microvasc Res 78(2):180-190, 2009.
14 Brady MA, Sivananthan S, Mudera V, Liu Q, Wiltfang J, Warnke PH. The primordium of a biological joint
replacement: Coupling of two stem cell pathways in biphasic ultrarapid compressed gel niches. J
Craniomaxillofac Surg 39(5):380-386, 2011.
15 Stiehler M, Bünger C, Baatrup A, Lind M, Kassem M, Mygind T: Effect of dynamic 3-D culture on
proliferation, distribution, and osteogenic differentiation of human mesenchymal stem cells. J Biomed Mater Res
A 89(1):96-107, 2009.
16 Langer R, Vacanti JP: Tissue engineering. Science 260(5110):920-926, 1993.
17 Leong KF, Cheah CM, Chua CK: Solid freeform fabrication of three-dimensional scaffolds for engineering
replacement tissues and organs. Biomaterials 24(13):2363-2378, 2003.
18 Rezwan K, Chen QZ, Blaker JJ, Boccaccini AR: Biodegradable and bioactive porous polymer/inorganic
composite scaffolds for bone tissue engineering. Biomaterials 27(18):3413-3431, 2006.
19 Heo SJ, Kim SE, Wei J, Kim DH, Hyun YT, Yun HS, Kim HK, Yoon TR, Kim SH, Park SA, Shin JW, Shin
JW: In vitro and animal study of novel nano-hydroxyapatite/poly(epsilon-caprolactone) composite scaffolds
fabricated by layer manufacturing process. Tissue Eng Part A 15(5):977-989, 2009.
20 Salgado AJ, Coutinho OP, Reis RL: Bone tissue engineering: state of the art and future trends. Macromol Biosci
4(8):743-765, 2004.
21 Hutmacher DW, Schantz JT, Lam CX, Tan KC, Lim TC: State of the art and future directions of scaffold-based
bone engineering from a biomaterials perspective. J Tissue Eng Regen Med 1(4):245-260, 2007.
22 Goldberg VM: Selection of bone grafts for revision total hip arthroplasty. Clin Orthop Relat Res (381):68-76,
2000.
23 Lund AW, Bush JA, Plopper GE, Stegemann JP: Osteogenic differentiation of mesenchymal stem cells in
defined protein beads. J Biomed Mater Res B Appl Biomater 87(1):213-221, 2008.
65
24 Dahl M, Jørgensen NR, Hørberg M, Pinholt EM. Carriers in mesenchymal stem cell osteoblast mineralization –
State-of-the-art. Journal of CranioMaxillofacial Surgery 42(1):41-47, 2014.
25 Bystedt H, Rasmusson L: Porous titanium granules used as osteoconductive material for sinus floor
augmentation: a clinical pilot study. Clin Implant Dent Relat Res 11(2):101-105, 2009.
26 Alffram PA, Bruce L, Bjursten LM, Urban RM, Andersson GB: Implantation of the femoral stem into a bed of
titanium granules using vibration: a pilot study of a new method for prosthetic fixation in 5 patients followed for
up to 15 years. Ups J Med Sci 112(2):183-189, 2007.
27 Turner TM, Urban RM, Hall DJ, Andersson GB: Bone ingrowth through porous titanium granulate around a
femoral stem. Ups J Med Sci 112(2):191-197, 2007.
28 Masuda T, Salvi GE, Offenbacher S, Felton DA, Cooper LF: Cell and matrix reactions at titanium implants in
surgically prepared rat tibiae. Int J Oral Maxillofac Implants 12(4):472-485, 1997.
29 Hong J, Andersson J, Ekdahl KN, Elgue E, Axén N, Larsson R, Nilsson B: Titanium is a highly thrombogenic
biomaterial: possible implications for osteogenesis. Thromb Haemost 82(1):58-64, 1999.
30 Sul YT, Johansson C, Wennerberg A, Cho LR, Chang BS, Albrektsson T: Optimum surface properties of
oxidized implants for reinforcement of osseointegration: surface chemnistry, oxide thickness, porosity,
roughness, and crystal structure. Int J Oral Maxillofac Implants 20(3):349-359, 2005.
31 Zuk PA, Zhu M, Mizuno H, Huang J, Futrell JW, Katz AJ, Benhaim P, Lorenz HP, Hedrick MH: Multilineage
cells from human adipose tissue: implications for cell-based therapies. Tissue Eng 7(2):211-228, 2001.
32 Zuk PA, Zhu M, Ashjian P, De Urgate DA, Huang JI, Mizuno H, Alfonso ZC, Fraser JK, Benhaim P, Hedrick
MH: Human adipose tissue is a source of multipotent stem cells. Mol Biol Cell 13(12):4279-4295, 2002.
33 Cao Y, Sun Z, Liao L, Meng Y, Han Q, Zhao RC: Human adipose tissue-derived stem cells differentiate into
endothelial cells in vitro and improve postnatal neovascularization in vivo. Biochem Biophys Res Commun
332(2):370-379, 2005.
34 Meruane M, Rojas M, Marcelain K, Chile S: The use of adipose tissue-derived stem cells within a dermal
substitute improves skin regeneration by increasing neoangiogenesis and collagen synthesis. Plast Reconstr Surg
130(1):53-63, 2012.
35 Hadjidakis DJ, Androulakis II: Bone remodeling. Ann N Y Acad Sci 1092:385-396, 2006.
66
36 Del Fattore A, Teti A, Rucci N. Bone cells and the mechanisms of bone remodeling. Front Biosci (Elite Ed)
4:2302-2321, 2012.
37 Baron R, Kneissel M. WNT signaling in bone homeostasis and disease: from human mutations to treatments. Nat
Med 19(2):179-192, 2013.
38 Rosen CJ. Primer on the metabolic bone diseases and disorders of mineral metabolism. The American Society
for Bone and Mineral Research; Eight edition, 2013. Chapter 2-4, 6, 113-114.
39 Klein-Nulend J, Nijweide PJ, Burger EH. Osteocyte and bone structure. Curr Osteoporos Rep1:5-10, 2003.
40 Henriksen K, Neutzsky-Wulff AV, Bonewald LF, Karsdal MA. Local communication on and within bone
controls bone remodeling. Bone 44:1026-1033, 2009.
41 Neve A, Corrado A, Cantatore FP. Osteocytes: central conductors of bone biology in normal and pathological
conditions. Acta Physiol (Oxf) 204:317-330, 2012.
42 Siddiqui JA, Partridge NC: Physiological bone remodeling: systemic regulation and growth factor involvement.
Physiology 31(3):233-245, 2016.
43 Raggatt LJ, Partridge NC. Cellular and molecular mechanisms of bone remodeling. J Biol Chem 285:25103-
25108, 2010.
44 Manolagas SC: Birth and death of bone cells: basic regulatory mechanisms and implications for the pathogenesis
and treatment of osteoporosis. Endocr Rev 21:115-137, 2000.
45 Kearns AE, Khosla S, Kostenuik PJ. Receptor activator of nuclear factor kappaB ligand and osteoprotegerin
regulation of bone remodeling in health and disease. Endocr Rev 29:155-192, 2008.
46 Lerner UH: Bone remodeling in post-menopausal osteoporosis. J Dent Res 85:584-595, 2006.
47 Väänänen K. Mechanism of osteoclast mediated bone resorption – rationale for the design of new therapeutics.
Adv Drug Deliv Rev 57(7):959-971, 2005.
48 Fei Y, Hurley MM. Role of fibroblast growth factor 2 and Wnt signaling in anabolic effects of parathyroid
hormone on bone formation. J Cell Physiol 227(11):3539-3545, 2012.
49 Kramer I, Keller H, Leupin O, Kneissel M: Does osteocytic SOST suppression mediate PTH bone anabolism?
Trends Endocrinol Metab 21:237-244, 2010.
67
50 Bellido T, Ali AA, Gubrij I, Plotkin LI, Fu Q, O’Brien CA, Manolagas SC, Jilka RL. Chronic elevation of
parathyroid hormone in mice reduces expression of sclerostin by osteocytes: a novel mechanism for hormonal
control of osteoblastogenesis. Endocrinology 146(11):4577-4583, 2005.
51 Bellido T, Saini V, Pajevic PD. Effects of PTH on osteocyte function. Bone 54(2):250-257, 2013.
52 Robling AG, Niziolek PJ, Baldridge LA, Condon KW, Allen MR, Alam I, Mantila SM, Gluhak-Heinrich J,
Bellido TM, Harris SE, Turner CH. Mechanical stimulation of bone in vivo reduces osteocyte expression of
Sost/sclerostin. J Biol Chem 283(9):5866-5875, 2008.
53 Gooi JH, Pompolo S, Karsdal MA, Kulkarni NH, Kalajzic I, McAhren SH, Han B, Onvia JE, Ho PW, Gillespie
MT, Walsh NC, Chia LY, Quinn JM, Martin TJ, Sims NA. Calcitonin impairs the anabolic effect of PTH in
young rats and stimulates expression of sclerostin by osteocytes. Bone 46(6):1486-1497, 2010.
54 Dimitriou R, Mataliotakis GI, Angoules AG, Kanakaris NK, Giannoudis PV: Complications following
autologous bone graft harvesting from the iliac crest and using the RIA: a systemativ review. Injury 42 Suppl.
2:S3-S15, 2011.
55 Abrams H, Kopczyk RA, Kaplan AL: Incidence of anterior ridge deformities in partially edentulous patients. J
Prosthet Dent 57(2): 191-194, 1987.
56 Frost HM: Wolff’s Law and bone’s structural adaptations to mechanical usage: an overview for clinicians. Angle
Orthod 64(3),175-188, 1994.
57 Taguchi A, Suei Y, Ohtsuka M, Otani K, Tanimoto K, Hollender LG: Relationship between bone mineral
density and tooth loss in elderly Japanese women. Dentomaxillofac Radiol 28(4):219-223, 1999.
58 Taguchi A, Tanimoto K, Suei Y, Wada T: Tooth loss and mandibular osteopenia. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 79(1):127-132, 1995.
59 Nicopoulou-Karayianni K, Tzoutzoukos P, Mitsea A, Karayiannis A, Tsiklakis K, Jacobs R, Lindh C, van der
Stelt P, Allen P, Graham J, Horner K, Devlin H, Pavitt S, Yuan J: Tooth loss and osteoporosis: the osteodent
study. J Clin Periodontol 36(3):190-197, 2009.
60 von Wowern N, Stoltze K: Pattern of age related bone loss in mandibles. Scand J Dent Res 88(2):134-146, 1980.
61 Kribbs PJ, Chesnut CH 3rd, Ott SM, Kilcoyne RF: Relationships between mandibular and skeletal bone in a
population of normal women. J Prosthet Dent 63(1):86-89, 1990.
68
62 Brown RA. Extreme tissue engineering. Concepts and strategies for tissue fabrication. Wiley-Blackwell; First
edition, 2013.
63 Lohmann H, Grass G, Rangger C, Mathiak G: Economic impact of cancellous bone grafting in trauma surgery.
Arch Orthop Trauma Surg 127(5):345-348, 2007.
64 Giannoudis PV, Chris Art JJ, Schmidmaier G, Larsson S: What should be the characteristics of the ideal bone
graft substitute? Injury 42 Suppl 2:S1-2, 2011.
65 Vo TN, Kasper FK, Mikos AG: Strategies for controlled delivery of growth factors and cells for bone
regeneration. Adv Drug Deliv Rev 64(12):1292-1309, 2012.
66 Pullisaar H, Tiainen H, Landin MA, Lyngstadaas SP, Haugen HJ, Reseland JE, Ostrup E: Enhanced in vitro
osteoblast differentiation on TiO2 scaffold coated with alginate hydrogel containing simvastatin. J Tissue Eng
4:2041731413515670, 2013.
67 Pullisaar H, Reseland JE, Haugen HJ, Brinchmann JE, Ostrup E: Simvastatin coating of TiO2 scaffold induces
osteogenic differentiation of human adipose tissue-derived mesenchymal stem cells. Biochem Biophys Res
Commun 447(1):139-144, 2014.
68 Lyngstadaas SP, Verket A, Pinholt EM, Mertens C, Haanæs HR, Wall G, Wallström M, Rasmusson L: Titanium
granules for augmentation of the maxillary sinus – a multicenter study. Clin Implant Dent Relat Res Suppl
2:e594-600, 2015.
69 Wohlfahrt JC, Lyngstadaas SP, Heijl L, Aass AM: Porous titanium granules in the treatment of mandibular class
II furcation defects: a consecutive case series. J Periodontol 83(1):61-69, 2012.
70 Tigran. Informative letter. Malmö, Sweden: Tigran Technologies AB, 2010.
71 Puchtler H, Meloan SN, Terry MS: On the history and mechanism of alizarin and alizarin red S stains for
calcium. J Histochem Cytochem 17(2):110-124, 1969.
72 Bonewald LF, Harris SE, Rosser J, Dallas MR, Dallas SL, Camacho NP, Boyan B, Boskey A: Von Kossa
staining alone is not sufficient to confirm that mineralization in vitro represents bone formation. Calcif Tissue Int
72:537-547, 2003.
73 Wang YH, Liu Y, Maye P, Rowe DW: Examination of mineralized nodule formation in living osteoblastic
cultures using fluorescent dyes. Biotechnol Prog 22:1697-1701, 2006.
69
74 Dahl M, Syberg S, Jørgensen NR, Pinholt EM. Adipose derived mesenchymal stem cells – their osteogenicity
and osteoblast in vitro mineralization on titanium granule carriers. J Craniomaxillofac Surg 41(8):e213-220,
2013.
75 Erben RG: Embedding of bone samples in methylmethacylate: an improved method suitable for bone
histomorphometry, histochemistry, and immunohistochemistry. J Histochem Cytochem 45(2):307-313, 1997.
76 Donath K: Preparation of histologic sections by the cutting grinding technique for hard tissue and other material
not suitable to be sectioned by routine methods. EXACT-Kulzer-Publication, Norderstedt, 1-15, 1993.
77 Luzi C, Verna C, Melsen B: Immediate loading of orthodontic mini-implants: a histomorphometric evaluation of
tissue reaction. Eur J Orthod 31(1):21-29, 2009.
78 Dempster DW, Compston JE, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM, Recker
RR, Parfitt M: Standardized nomenclature, symbols, and units for bone histomorphometry: a 2012 update of the
report of the ASBMR histomorphometry nomenclature committee. J Bone Miner Res 28(1):2-17, 2013.
79 Vesterby A, Kragstrup J, Gundersen HJG, Melsen F: Unbiased stereologic estimation of surface density in bone
using vertical sections. Bone 8:13-17, 1987.
80 Sterio DC: The unbiased estimation of number and sizes of arbitrary particles using the dissector. J Microsc
134(Pt 2):127-136, 1984.
81 Kasten P, Vogel J, Luginbühl R, Niemeyer P, Tonak M, Lorenz H, Helbig L, Weiss S, Fellenberg J, Leo A,
Simank HG, Richter W: Ectopic bone formation associated with mesenchymal stem cells in a resorbable calcium
deficient hydroxyapatite carrier. Biomaterials 26(29):5879-5889, 2005.
82 Kasten P, Vogel J, Luginbühl R, Niemeyer P, Weiss S, Schneider S, Kramer M, Leo A, Richter W: Influence of
platelet-rich plasma on osteogenic differentiation of mesenchymal stem cells and ectopic bone formation in
calcium phosphate ceramics. Cells Tissues Organs 183(2):68-79, 2006.
83 Champa Jayasuriya A, Bhat A: Mesenchymal stem cell function on hybrid organic/inorganic microparticles in
vitro. J Tissue Eng Regen Med 4(5):340-348, 2010.
84 Korda M, Hua J, Little NJ, Heidari N, Blunn GW: The effect of mesenchymal stromal cells on the
osseoinduction of impaction grafts. Tissue Eng Part A 16(2):675-683, 2010.
85 Weir MD, Xu HH: Culture human mesenchymal stem cells with calcium phosphate cement scaffolds for bone
repair. J Biomed Mater Res B Appl Biomater 93(1):93-105, 2010.
70
86 Xu HH, Zhao L, Weir MD: Stem cell-calcium phosphate constructs for bone engineering. J Dent Res
89(12):1482-1488, 2010.
87 Chen M, Le DQ, Baatrup A, Nygaard JV, Hein S, Bjerre L, Kassem M, Zou X, Bünger C: Self-assembled
composite matrix in a hierarchical 3-D scaffold for bone tissue engineering. Acta Biomater 7(5):2244-2255,
2011.
88 Kruger EA, Im DD, Bischoff DS, Pereira CT, Huang W, Rudkin GH, Yamaguchi DT, Miller TA: In vitro
mineralization of human mesenchymal stem cells on three-dimensional type I collagen versus PLGA scaffolds: a
comparative analysis. Plast Reconstr Surg 127(6):2301-2311, 2011.
89 Barhanpurkar AP, Gupta N, Srivastava RK, Tomar GB, Naik SP, Joshi SR, Pote ST, Mishra GC, Wani MR: IL-3
promotes osteoblast differentiation and bone formation in human mesenchymal stem cells. Biochem Biophys
Res Commun 418(4):669-675, 2012.
90 Coulson RA: Relationship between fluid flow and O2 demand in tissues in vivo and in vitro. Perspect Biol Med
27(1):121-126, 1983.
91 Supronowicz P, Gill E, Trujillo A, Thula T, Zhukauskas R, Ramos T, Cobb RR: Human adipose-derived side
population stem cells cultured on demineralized bone matrix for bone tissue engineering. Tissue Eng Part A
17(5-6):789-798, 2011.
92 Hosseinkhani H, Inatsugu Y, Hiraoka Y, Inoue S, Tabata Y: Perfusion culture enhances osteogenic
differentiation of rat mesenchymal stem cells in collagen sponge reinforced with poly(glycolic Acid) fiber.
Tissue Eng 11(9-10):1476-1488, 2005.
93 Turhani D, Watzinger E, Weissenbock M, Yerit K, Cvikl B, Thurnher D, Ewers R: Three-dimensional
composites manufactured with human mesenchymal cambial layer precursor cells as an alternative for sinus
floor augmentation: an in vitro study. Clin Oral Implants Res 16(4):417-424, 2005.
94 Shih YR, Chen CN, Tsai SW, Wang YJ, Lee OK: Growth of mesenchymal stem cells on electrospun type I
collagen nanofibers. Stem cells 24(11):2391-2397, 2006.
95 Weissenboeck M, Stein E, Undt G, Ewers R, Lauer G, Turhani D: Particle size of hydroxyapatite granules
calcified from red algae affects the osteogenic potential of human mesenchymal stem cells in vitro. Cells Tissues
Organs 182(2):79-88, 2006.
71
96 Zhou GS, Su ZY, Cai YR, Liu YK, Dai LC, Tang RK, Zhang M: Different effects of nanophase and
conventional hydroxyapatite thin films on attachment, proliferation and osteogenic differentiation of bone
marrow derived mesenchymal stem cells. Biomed Mater Eng 17(6):387-395, 2007.
97 Bjerre L, Bünger CE, Kassem M, Mygind T: Flow perfusion culture of human mesenchymal stem cells on
silicate-substituted tricalcium phosphate scaffolds. Biomaterials 29(17):2616-2627, 2008.
98 Huang Y, Jin X, Zhang X, Sun T, Tu J, Tang T, Chang J, Dai K: In vitro and in vivo evaluation of akermanite
bioceramics for bone regeneration. Biomaterials 30(28):5041-5048, 2009.
99 Shi X, Wang Y, Ren L, Huang W, Wang DA: A protein/antibiotic releasing poly(lactic-co-glycolic acid)/lecithin
scaffold for bone repair applications. Int J Pharm 373(1-2):85-92, 2009.
100 Sabetrasekh R, Tiainen H, Lyngstadaas SP, Reseland J, Haugen H: A novel ultra-porous titanium dioxide
ceramic with excellent biocompatibility. J Biomater Appl 25(6):559-580, 2011.
101 Sunay O, Can G, Cakir Z, Denek Z, Kozanoglu I, Erbil G, Yilmaz M, Baran Y: Autologous rabbit adipose
tissue-derived mesenchymal stromal cells for the treatment of bone injuries with distraction osteogenesis.
Cytotherapy 15(6):690-702, 2013.
102 Marini F, Luzi E, Fabbri S, Ciuffi S, Sorace S, Tognarini I, Galli G, Zonefrati R, Sbaiz F, Brandi ML:
Osteogenic differentiation of adipose tissue-derived mesenchymal stem cells on nanostructured Ti6AI4V and
Ti13Nb13Zr. Clin Cases Miner Bone Metab 12(3):224-237, 2015.
72
11. Paper I
Journal of Cranio-Maxillo-Facial Surgery 42 (2014) 41–47
Contents lists available at SciVerse ScienceDirect
Journal of Cranio-Maxillo-Facial Surgery
journal homepage: www.jcmfs.com
Carriers in mesenchymal stem cell osteoblast
mineralization – State-of-the-art
Morten Dahl a, Niklas Rye Jørgensen
b, Mette Hørberg
a, Else Marie Pinholt
a, *
a Institute of Odontology, Department of Oral and Maxillofacial Surgery, Faculty of Health Sciences, University of Copenhagen, Nørre Allé 20, 2200
Copenhagen, Denmark b
Research Center for Ageing and Osteoporosis, Departments of Clinical Biochemistry and Medicine, Copenhagen University Hospital Glostrup, Glostrup,
Denmark
A R T I C L E I N F O
Article history:
Paper received 6 September 2012
Accepted 29 January 2013
Keywords:
Mesenchymal stem cells
Tissue scaffolds
Drug carriers
Tissue engineering
Bone regeneration
ABSTRACT
Purpose: Tissue engineering is a new way to regenerate bone tissue, where osteogenic capable cells
combine with an appropriate scaffolding material. Our aim was in a Medline Search to evaluate osteo-
blast mineralization in vitro and in vivo including gene expressing combining mesenchymal stem cells
(MSCs) and five different carriers, titanium, collagen, calcium carbonate, calcium phosphate and poly-
lactic acid-polyglycolic acid copolymer for purpose of a meta– or a descriptive analysis.
Materials and methods: The search included the following MeSH words in different
combinations–mesenchymal stem cells, alkaline phosphatase, bone regeneration, tissue engineering,
drug carriers, tissue scaffolds, titanium, collagen, calcium carbonate, calcium phosphates and
polylactic acid-polyglycolic acid copolymer.
Results: Two out of 80 articles included numerical values and as control, carriers and cells, on mineral-
ization and gene expression. β-tricalcium phosphate (β-TCP) revealed elevated alkaline phosphatase
activity, and calcium-deficient hydroxyapatite a greater gene expression of osteocalcin when seeded with
induced MSCs.
Conclusion: No data are published on titanium used as a carrier in MSC osteoblast mineralization. A meta–
as well as a descriptive analysis includes numerical values of test materials and of control reactions from
carrier and cells, respectively. Only two articles fulfilled these requirements.
© 2013 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights
reserved.
1. Introduction
Autologous bone grafting is the preferred treatment for bone
reconstruction due to transfer of osteoprogenitor cells or osteo-
blasts, osteoconductivity and its osteoinductive capacity of bone
(Goldberg and Stevenson, 1987; Behnia et al., 2012). Harvesting of
autologous bone grafts is associated with morbidity (pain, blood
loss, surgical scars, necrosis), besides the risk of insufficient avail-
able volume of grafting material (Younger and Chapman, 1989;
Martins et al., 2009; Stockmann et al., 2012). However, researchers
seek alternative methods through synthetic or natural biomaterials.
These materials, unfortunately, have an inferior osteogenic poten-
tial compared to autografts. Therefore biomaterials are combined
with osteogenic cells for purpose of improved osteogenesis (Chen
et al., 2005; Seitz et al., 2007; Sutter et al., 2009; Behnia et al.,
2012; Metzler et al., 2012; Stockmann et al., 2012).
* Corresponding author. Tel.: +45 33326611.
E-mail address: [email protected] (E.M. Pinholt).
Demineralized bone (DB) and dentin (DD) induces heterotopic
osteogenesis in rodents (Bang and Urist, 1967; Glowacki and
Mulliken, 1985; Urist, 2002). For clinical practice however, DB and
DD act as bone fillers with minimal osteoinductivity (Pinholt et al.,
1992, 1994). Therefore tissue engineering is an advance within
bone regeneration.
Tissue engineering involves the in vitro seeding of cells onto
scaffolds supporting cell adhesion, migration, proliferation and
differentiation, and defines the three dimensional (3D) shape of the
tissue to be engineered. The carrier should be a scaffold with sur-
face characteristics so mesenchymal stem cells (MSCs) are able to
attach, proliferate and differentiate. The optimal scaffold should be
biocompatible, biodegradable and osteoconductive to generate
new bone formation (Leong et al., 2006; Ben-David et al., 2010).
It is preferable that scaffolds are 3D as a pre-requisite for cells to be
able to proliferate while maintaining their ability to differentiate. The
success of tissue engineering is dependent on oxygen and nutrient
transport to the implanted cells (Laschke et al., 2006; Schumann et al.,
2009; Brady et al., 2011). A great limitation of the 3D scaffolds is that
1010-5182/$ – see front matter © 2013 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jcms.2013.01.047
73
the cells at the interior of the scaffold have decreased nutrient and
oxygen transport and decreased removal of waste products (Stiehler
et al., 2009). To secure a high density of colonizing cells and to pro-
mote neovascularization when implanted in vivo, the scaffolds
should have high porosity, large surface area, mechanical properties
and pore size appropriate for the application, and a highly inter-
connected porous structure (Langer and Vacanti, 1993; Leong et al.,
2003; Rezwan et al., 2006; Heo et al., 2009; Martins et al., 2009;
Schumann et al., 2009). Scaffold porosity may be inversely related to
the mechanical properties of the material. It is important to find a
balance of securing mechanical needs of the tissue to be replaced and
scaffold porosity which allows tissue growth (Salgado et al., 2004;
Hutmacher et al., 2007; Martins et al., 2009).
The ideal engineered cellular bone graft needs to exhibit the
following features e the presence of osteogenic cells to generate
new bone directly, an appropriate extracellular matrix to provide an
osteoconductive scaffold, osteoinductive growth factors to provide
signals to the resident cells and an adequate blood supply to support
cell growth and function (Goldberg, 2000; Lund et al., 2008).
We chose five main carriers to evaluate; titanium, collagen, calc-
ium carbonate, calcium phosphate and polylactic acid-polyglycolic
acid copolymer (PLGA). Titanium is a metal alloy and is bio inert, but
it is not biodegradable (Bjerre et al., 2008). The advantage of using
titanium is its good mechanical strength, corrosion resistance and
biocompatibility (Kang et al., 2010). Recent years have lead to
modifying the surface of titanium chemically for improving its
surface properties (Kang et al., 2010). A hydrophilic surface is
assumed to be advantageous during the early phase of wound
healing and during the cascade of events that occurs during
osseointegration (Sawase et al., 2008). In titanium dioxide, the oxide
surface is hydrophilic and binding structural water and forming OH
and O2 groups. Possibly the creation of a hydroxylated oxide surface
enhances the surface reactivity with the surrounding ions, amino
acids and proteins in the tissue fluids (Sawase et al., 2008).
Type I collagen is the main organic component of the extracel-
lular bone matrix. Type I collagen may affect the development of an
osteoblastic stem cell differentiation and growth positively (Burr,
2002; Reichert et al., 2009; Schofer et al., 2009). A scaffold matrix
composed of type I collagen and hydroxyapatite (HA), providing
support for MSC growth without compromising their osteogenic
differentiation ability, may thus be indicated for bone tissue engi-
neering (Gigante et al., 2008).
Ceramics such as calcium phosphates and calcium carbonates
have been used as matrices for bone regeneration, because of the
inorganic component of bone is composed of the ceramic calcium
hydroxyapatite (Ducheyne and Qiu, 1999; Livingston et al., 2002;
Ben-David et al., 2010).
Especially calcium phosphate ceramics have been used for bone
tissue repair in orthopedic and dental applications, due to their
biocompatibility and osteoconductivity through its ability to induce
HA formation in the physiological environment (Mehlisch et al.,
1990; Warren et al., 2004; Xin et al., 2005; Zhou et al., 2007;
Huang et al., 2009). HA is a bioactive, biocompatible and osteo-
conductive ceramic material (Mehlisch et al., 1990; Bagambisa
et al., 1993; Warren et al., 2004; Zhou et al., 2007; Huang et al.,
2008). HA is used as coating material for surface modification of
some bioinert scaffolds such as titanium to induce bioactivity and
osteoconductivity and as bone tissue-engineering scaffold in which
osteogenic cells could be seeded. Nanophase HA may be better than
conventional HA due to superior biomimetics and osteo-
conductivity (Zhou et al., 2007). However HA has limited clinical
applications because of its brittleness, difficulty of shaping and an
extremely slow degradation rate (Wang, 2003; Huang et al., 2008).
Tricalcium phosphate (TCP) and HA have excellent biocompat- ibility and osteoconductiveness, due to the resemblance to the
mineral phase of bone (Porter et al., 2004; Bjerre et al., 2008).
During physiologic conditions, the resorption of HA is almost non- existent, providing TCP with a biological advantage compared to HA
(Mastrogiacomo et al., 2005; Bjerre et al., 2008). TCP has shown
improved stability when it is doped by silicon substitution (Si-TCP), which ensures an in vivo degradation time of more than 1 year.
Silicate-substituted HA (Si-HA) shows highly organized apatite
crystal structure and superior bone remodeling properties as compared to pure HA (Porter et al., 2004; Bjerre et al., 2008). Cal-
cium phosphate-coated chitosan-based scaffolds with incorporated
lysozymes may possibly enhance bone bonding, osteoconductive and/or osteoinductive capacity (Martins et al., 2009).
β-tricalcium phosphate (β-TCP) is a synthetic calcium phosphate
ceramic used alternatively to an autologous bone graft (LeGeros, 2002; Matsuno et al., 2006, 2008). β-TCP is similar in its molecu- lar
composition to human bone (Huang et al., 2009). However it has low
compressive strength (Miranda et al., 2008). Calcium-deficient hydroxyapatite (CDHA, Ca9(PO4)5(HPO4)OH)
is a ceramic with a high specific surface area (SSA), 20-80 m2/g,
which is very similar to that of natural bone, about 80 m2/g. β-
TCP has a lower SSA, less than 0.5 m2/g. Cells adhere more easily to
high SSA ceramics than to low SSA scaffolds (Kasten et al., 2003,
2006).
A combination of the bioactive ceramic HA and the biodegrad-
able polymer poly(ε-caprolactone) (PCL), a commercial polymer, may take advantage of the properties of both materials. PCL de-
grades slowly in vivo, has a good biocompatibility and formability,
while HA is similar to bone mineral and has greater surface energy and surface activity (Heo et al., 2009).
PLGA is a commonly used synthetic polymer within tissue en-
gineering and drug delivery (Quaglia, 2008; Shi et al., 2009). Polymers derived from D, L-lactic and glycolic acids as PLGA are
biocompatible and biodegradable (Visscher et al., 1985; Fournier et
al., 2003), and its nontoxic hydrolytic degradation products, lactic and glycolic acid, are metabolized in vivo (Gopferich, 1996; Stiehler et
al., 2009). PLGA has good mechanical strength, excellent
processability which can secure flexible structures, as well as a tailored degradation rate (Giteau et al., 2008; Shi et al., 2009).
The purpose of the current study was to evaluate state – of the art of
the mentioned different carriers combined with MSCs in osteoblast
mineralization in vitro and in vivo within mineralization and gene
expression for purposes of a meta- or a descriptive analysis.
2. Materials and methods
A Medline search (Pub Med) was carried out May 2012, and
studies published in English from January 2000 to May 2012 were
included in the review within the inclusion and exclusion criteria,
Table 1.
Table 1
Inclusion and exclusion criteria as well as outcomes measures.
Inclusion criteria Exclusion criteria Outcomes measures
Valid statistics Diffuse statistics Mineralization
After January 2000 Before January 2000 Gene expression
Published in English Published in other languages
Control (carrier No control (carrier without
without cells) cells) presented
Calcium phosphate Other carriers than included
carrier
Titanium carrier Size or volume of carrier
not presented
Collagen carrier Number of cells not presented
Calcium carbonate Cartilage formation
carrier
PLGA carrier
74
The following MeSH words were used in different combinations –
“mesenchymal stem cells”, “alkaline phosphatase”, “bone regen-
eration”, “tissue engineering”, “drug carriers”, “tissue scaffolds”, “ti-
tanium”, “calcium carbonate”, “calcium phosphates”, “collagen” and
“polylactic acid-polyglycolic acid copolymer”. Example of search
(Table 2): Search #12 MeSH words “mesenchymal stem cells”, “alka-
line phosphatase”, “tissue engineering”, and “calcium phosphates”,
results in 28 articles, where 20 articles where presented in previous
search and eight articles were excluded. Titles and abstracts were
screened, and full-text analysis was performed in relevant publica-
tions. All combinations in the MeSH search had “mesenchymal stem
cells” and “alkaline phosphatase” as MeSH words, Table 2.
3. Results
The MeSH search resulted in 80 different articles. Following screening of titles and abstracts by defining the chosen inclusion–
and exclusion criteria, 51 potential publications were found
relevant and full-text analysis was performed. Out of the 51 articles, nine articles (Kasten et al., 2005, 2006; Champa
Jayasuriya and Bhat, 2010; Korda et al., 2010; Weir and Xu, 2010;
Xu et al., 2010; Chen et al., 2011a; Kruger et al., 2011; Barhanpurkar et al., 2012) were included in the present study. 71 studies (Yang et
al., 2003; Ignatius et al., 2004; Liu et al., 2004; Pang et al., 2004;
Takahashi et al., 2004; Yin et al., 2004; Hosseinkhani et al., 2005a; 2005b; Pang et al., 2005; Turhani et al., 2005; Abramovitch-Gottlib
et al., 2006; George et al., 2006; Ku et al., 2006; Moioli et al., 2006; Nuttelman et al., 2006; Shih et al., 2006; Weissenboeck et al.,
2006; Hwang et al., 2007; Rust et al., 2007; Sun et al., 2007; Zhou
et al., 2007; Bjerre et al., 2008; Gigante et al., 2008; Huang et al., 2008; Matsuno et al., 2008; Schofer et al., 2008; Valarmathi et
al., 2008; Vermonden et al., 2008; Diederichs et al., 2009; Heo et
al., 2009; Huang et al., 2009; Kim et al., 2009; Liu et al., 2009; Martins et al., 2009; Mei et al., 2009; Nair et al., 2009; Niu et al.,
2009; Park et al., 2009; Pereira et al., 2009; Schofer et al., 2009;
Schumann et al., 2009; Shi et al., 2009; Stiehler et al., 2009; Wen et al., 2009; Binulal et al., 2010; Breyner et al., 2010; Cordonnier et al.,
2010; Duan and Wang, 2010; Hess et al., 2010; Liu et al., 2010;
Nandakumar et al., 2010; Sittichokechaiwut et al., 2010; Wang et al.,
2010; Zhang et al., 2010; Bjerre et al., 2011; Broese et al., 2011;
Chen et al., 2011b; Fan et al., 2011; Gandolfi et al., 2011; Janicki et
al., 2011; Lee et al., 2011; Li et al., 2011; Liu et al., 2011; Miranda et al., 2011; Sala et al., 2011; Shafiee et al., 2011; Silva et al., 2011;
Zhou et al., 2011; Zou et al., 2011; Tseng et al., 2012; Yang et al.,
2012) did not meet the inclusion criteria and were therefore excluded from this analysis. Among studies included no publications
were present on calcium carbonate and titanium as carriers.
Numerical values of defined outcomes were not presented in
any of the publications. Outcomes in all included articles were exclusively presented as figures and tables. The corresponding
authors of the included nine articles were contacted to obtain nu-
merical values of their studies. Subsequently two (Kasten et al., 2005, 2006) out of the nine articles are presented by numerical
values.
Tables 2 and 3, show outcome measures from the two (Kasten et al., 2005, 2006) included articles with enclosed numerical
values of carriers and cells, separately, tested for mineralization and
gene expression for purposes of control. Kasten et al. (2006) shows elevated alkaline phosphatase (ALP)
activity for β-TCP carrier and even more for β-TCP platelet-rich
plasma (PRP) carrier seeded with MSCs and a greater gene expression of osteocalcin (OC) within the CDHA PRP carrier. Kasten
et al. (2005) shows a higher ALP activity for β-TCP carrier seeded
with MSCs, and a greater gene expression of OC within deminer-alized bone matrix (DBM) carrier seeded with induced mesen-
chymal stem cells.
4. Discussion
In this systematic review numerical values of the outcomes of
mineralization and gene expression of mesenchymal stem cells
combined with the carriers titanium, collagen, calcium carbonate, calcium phosphate or PLGA were only obtained in two out of nine
included articles (Kasten et al., 2005, 2006). Carriers and cells,
separately, were not tested for mineralization and gene expression for purposes of control in the 71 excluded articles (Yang et al.,
2003; Ignatius et al., 2004; Liu et al., 2004; Pang et al., 2004; Takahashi et al., 2004; Yin et al., 2004; Hosseinkhani et al., 2005a,
2005b; Pang et al., 2005; Turhani et al., 2005; Abramovitch-Gottlib
et al., 2006; George et al., 2006; Ku et al., 2006; Moioli et al., 2006; Nuttelman et al., 2006; Shih et al., 2006; Weissenboeck et al.,
2006; Hwang et al., 2007; Rust et al., 2007; Sun et al., 2007;
Zhou et al., 2007; Bjerre et al., 2008; Gigante et al., 2008; Huang et al., 2008; Matsuno et al., 2008; Schofer et al., 2008; Valarmathi
et al., 2008; Vermonden et al., 2008; Diederichs et al., 2009; Heo
et al., 2009; Huang et al., 2009; Kim et al., 2009; Liu et al., 2009; Martins et al., 2009; Mei et al., 2009; Nair et al., 2009; Niu et
al., 2009; Park et al., 2009; Pereira et al., 2009; Schofer et al.,
2009; Schumann et al., 2009; Shi et al., 2009; Stiehler et al., 2009; Wen et al., 2009; Binulal et al., 2010; Breyner et al., 2010;
Cordonnier et al., 2010; Duan and Wang, 2010; Hess et al., 2010;
Liu et al., 2010; Nandakumar et al., 2010; Sittichokechaiwut et al., 2010; Wang et al., 2010; Zhang et al., 2010; Bjerre et al., 2011;
Broese et al., 2011; Chen et al., 2011b; Fan et al., 2011; Gandolfi et
al., 2011; Janicki et al., 2011; Lee et al., 2011; Li et al., 2011; Liu et
Table 2
Results of the MeSH search. Each search includes four MeSH words, comprising “mesenchymal stem cells”, “alkaline phosphatase”, “bone regeneration or tissue engineering”,
and a scaffold. Results represent number of articles.
Search MeSH word MeSH word Results Presented in previous search Included Excluded
#1 Bone regeneration Drug carriers 1 0 0 1
#2 Bone regeneration Tissue scaffolds 15 0 0 15
#3 Bone regeneration Titanium 0 0 0 0
#4 Bone regeneration Calcium carbonate 0 0 0 0
#5 Bone regeneration Calcium phosphates 13 9 2 2
#6 Bone regeneration Collagen 7 2 0 5
#7 Bone regeneration PLGA 2 2 0 0
#8 Tissue engineering Drug carriers 2 1 0 1
#9 Tissue engineering Tissue scaffolds 30 7 0 23
#10 Tissue engineering Titanium 0 0 0 0
#11 Tissue engineering Calcium carbonate 0 0 0 0
#12 Tissue engineering Calcium phosphates 28 20 0 8
#13 Tissue engineering Collagen 34 14 0 20
#14 Tissue engineering PLGA 7 4 0 3
75
Table 3
Outcome measures from the two included studies. Data represent results at four and eight weeks. Calcium-deficient hydroxyapatite (CDHA), platelet-rich plasma (PRP),
β-tricalcium phosphate (β-TCP), demineralized bone matrix (DBM), hydroxyapatite (HA).
Author Method Type of cell Type of carrier Result
ng p-nitrophenol/mg protein
Kasten P et al., 2006 ALP assay Human bone marrow MSCs CDHA:
2 x 10
5 cells/ceramic
-Empty
-With MSC
12.54 and 16.70
74.65 and 68.09
-With induced MSC 94.65 and 126.98
CDHA with PRP:
-Empty 23.05 and 13.24
-With MSC 148.90 and 180.41
-With induced MSC 124.30 and 116.85
β-TCP:
-Empty 19.67 and 18.51
-With MSC 188.91 and 168.32
-With induced MSC 132.19 and 92.88
β-TCP with PRP:
-Empty 26.14 and 19.16
-With MSC 297.09 and 140.68
-With induced MSC 149.59 and 124.55
Kasten P et al., 2005 ALP assay Human bone marrow MSCs CDHA:
2 x 10
5 cells/ceramic
-Empty
-With MSC
12.54 and 16.70
74.65 and 68.09
-With induced MSC 94.65 and 126.98
β-TCP:
-Empty 19.67 and 18.51
-With MSC 188.91 and 168.32
-With induced MSC 132.19 and 92.88
DBM:
-Empty 20.59 and 15.74
-With MSC 22.16 and 10.27
-With induced MSC 8.75 and 12.14
HA:
-Empty 36.65 and 20.65
-With MSC 52.94 and 88.03
-With induced MSC 34.30 and 280.40
Kasten P et al., 2006 ELISA kit, OC Human bone marrow MSCs CDHA:
2 x 10
5 cells/ceramic
-Empty
-With MSC
0.739 and 0
1,072 and 0.227
-With induced MSC 1,258 and 0.703
CDHA with PRP:
-Empty 0.084 and 0
-With MSC 1,577 and 1,183
-with induced MSC 0.885 and 2,672
β-TCP:
-Empty 1,089 and 0.409
-With MSC 1,015 and 0.264
-With induced MSC 0.792 and 0.081
β-TCP with PRP:
-Empty 0.713 and 0
-With MSC 0.729 and 0
-With induced MSC 0.629 and 0.102
Kasten P et al., 2005 ELISA kit, OC Human bone marrow MSCs CDHA:
2 x 10
5 cells/ceramic
-Empty
-With MSC
0.739 and 0
1,072 and 0.227
-With induced MSC 1,258 and 0.703
β-TCP:
-Empty 1,089 and 0.409
-With MSC 1,015 and 0.264
-With induced MSC 0.792 and 0.081
DBM:
-Empty 6,984 and 2,496
-With MSC 6,819 and 2,098
-With induced MSC 7,601 and 6,728
HA:
-Empty 1,075 and 0
-With MSC 0.919 and 0
-With induced MSC 0.735 and 0
al., 2011; Miranda et al., 2011; Sala et al., 2011; Shafiee et al.,
2011; Silva et al., 2011; Zhou et al., 2011; Zou et al., 2011;
Tseng et al., 2012; Yang et al., 2012). Data were subsequently
considered too sparse for a meta-analysis. The numerical value
of an outcome of a carrier combined with mesenchymal stem
cells comprises reactions from both the carrier and from the
cells. Therefore measurements of each of these, carrier and
cells, were deemed necessary for evaluation as basis for a meta-
as well as a descriptive analysis. Since only two experiments
including separate measurements of the carriers, for
purposes of control, were the two published experiments by
Kasten (Kasten et al., 2005, 2006) a descriptive analysis was
performed on these two.
76
Kasten et al. (2006) showed elevated ALP activity for β-TCP PRP
seeded with human bone marrow MSCs and for induced human bone
marrow MSCs compared to the same carrier without any cells. β-TCP
combined with PRP showed also higher numerical values for ALP
activity compared to the carriers CDHA, CDHA combined with PRP
and β-TCP, separately. In the other study, Kasten et al. (2005) also
showed elevated ALP activity for β-TCP seeded with human bone
marrow MSCs or with induced human bone marrow MSCs compared
to the same carriers without any cells. Additionally, numerical values
for outcomes of the carrier β-TCP were increased in comparison with
the values for the carriers CDHA, DBM and HA. As OC is produced by
osteoblasts, it is often used as a biochemical marker for the bone
formation process and subsequently a high serum OC level is rela-
tively well correlated with serum calcium (Parker et al., 2010). In the
publication Kasten et al. (2006), the carrier CDHA combined with PRP
and seeded with human bone marrow MSCs or with induced human
bone marrow MSCs revealed the highest expression of OC compared
to the value of the control and to the other carriers (Kasten et al.,
2006). However, in Kasten et al. (2005) expression of OC was
increased through DBM seeded with induced human bone marrow
MSCs. These findings do not reveal any optimal carrier and titanium
has not been used as a carrier.
5. Conclusion
There are no publications on titanium used as a carrier in MSC
osteoblast mineralization. The numerical values of an outcome of a
carrier combined with MSCs comprise reactions from carriers as
well as from cells, respectively. Measurements of each of these,
carrier and cells, are deemed necessary for evaluation as basis for a
meta- as well as a descriptive analysis. In only two publications,
data as control on a carrier without cells were included, however
results on an optimal carrier were inconclusive. Since in vitro re-
sults are supposed to support clinical applications mechanical
properties are also important to consider, however, there was no
evidence for an outstanding biomaterial.
Conflict of interest statement
All authors disclose any financial and personal relationships
with other people or organizations that could inappropriately in-
fluence our work.
References
Abramovitch-Gottlib L, Geresh S, Vago R: Biofabricated marine hydrozoan: a
bioactive crystalline material promoting ossification of mesenchymal stem
cells. Tissue Eng 12(4): 729–739, 2006
Bagambisa FB, Joos U, Schilli W: Mechanisms and structure of the bond between
bone and hydroxyapatite ceramics. J Biomed Mater Res 27(8): 1047–1055, 1993
Bang G, Urist MR: Bone induction in excavation chambers in matrix of decalcified
dentin. Arch Surg 94(6): 781–789, 1967
Barhanpurkar AP, Gupta N, Srivastava RK, Tomar GB, Naik SP, Joshi SR, et al: IL-3
promotes osteoblast differentiation and bone formation in human mesen-
chymal stem cells. Biochem Biophys Res Commun 418(4): 669–675, 2012
Behnia H, Khojasteh A, Soleimani M, Tehranchi A, Atashi A: Repair of alveolar cleft
defect with mesenchymal stem cells and platelet derived growth factors: a
preliminary report. J Craniomaxillofac Surg 40(1): 2–7, 2012
Ben-David D, Kizhner T, Livne E, Srouji S: A tissue-like construct of human bone
marrow MSCs composite scaffold support in vivo ectopic bone formation. J Tissue
Eng Regen Med 4(1): 30–37, 2010
Binulal NS, Deepthy M, Selvamurugan N, Shalumon KT, Suja S, Mony U, et al: Role of
nanofibrous poly(caprolactone) scaffolds in human mesenchymal stem cell
attachment and spreading for in vitro bone tissue engineering–response to
osteogenic regulators. Tissue Eng Part A 16(2): 393–404, 2010
Bjerre L, Bünger C, Baatrup A, Kassem M, Mygind T: Flow perfusion culture of human
mesenchymal stem cells on coralline hydroxyapatite scaffolds with various pore
sizes. J Biomed Mater Res A 97(3): 251–263, 2011
Bjerre L, Bünger CE, Kassem M, Mygind T: Flow perfusion culture of human
mesenchymal stem cells on silicate-substituted tricalcium phosphate scaffolds.
Biomaterials 29(17): 2616–2627, 2008
Brady MA, Sivananthan S, Mudera V, Liu Q, Wiltfang J, Warnke PH: The primordium
of a biological joint replacement: coupling of two stem cell pathways in biphasic
ultrarapid compressed gel niches. J Craniomaxillofac Surg 39(5): 380–386, 2011
Breyner NM, Hell RC, Carvalho LR, Machado CB, Peixoto Filho IN, Valério P, et al:
Effect of a three-dimensional chitosan porous scaffold on the differentiation of
mesenchymal stem cells into chondrocytes. Cells Tissues Organs 191(2): 119–
128, 2010
Broese M, Toma I, Haasper C, Simon A, Petri M, Budde S, et al: Seeding a human
tendon matrix with bone marrow aspirates compared to previously isolated
hBMSCsdan in vitro study. Technol Health Care 19(6): 469e479, 2011
Burr DB: The contribution of the organic matrix to bone’s material properties. Bone
31(1): 8–11, 2002
Champa Jayasuriya A, Bhat A: Mesenchymal stem cell function on hybrid
organic/inorganic microparticles in vitro. J Tissue Eng Regen Med 4(5): 340–348,
2010
Chen D, Shen H, Shao J, Jiang Y, Lu J, He Y, et al: Superior mineralization and
neovascularization capacity of adult human metaphyseal periosteum-derived cells
for skeletal tissue engineering applications. Int J Mol Med 27(5): 707–713, 2011a
Chen J, Sotome S, Wang J, Orii H, Uemura T, Shinomiya K: Correlation of in vivo bone
formation capability and in vitro differentiation of human bone marrow stromal
cells. J Med Dent Sci 52(1): 27–34, 2005
Chen M, Le DQ, Baatrup A, Nygaard JV, Hein S, Bjerre L, et al: Self-assembled
composite matrix in a hierarchical 3-D scaffold for bone tissue engineering. Acta
Biomater 7(5): 2244–2255, 2011b
Cordonnier T, Layrolle P, Gaillard J, Langonné A, Sensebé L, Rosset P, et al: 3D
environment on human mesenchymal stem cells differentiation for bone tissue
engineering. J Mater Sci Mater Med 21(3): 981–987, 2010
Diederichs S, Röker S, Marten D, Peterbauer A, Scheper T, van Griensven M,
et al: Dynamic cultivation of human mesenchymal stem cells in a rotating bed
bioreactor system based on the Z RP platform. Biotechnol Prog 25(6):
1762–1771, 2009
Duan B, Wang M: Customized Ca-P/PHBV nanocomposite scaffolds for bone tissue
engineering: design, fabrication, surface modification and sustained release of
growth factor. J R Soc Interface 7(Suppl. 5): 615–629, 2010
Ducheyne P, Qiu Q: Bioactive ceramics: the effect of surface reactivity on bone
formation and bone cell function. Biomaterials 20(23–24): 2287–2303,
1999
Fan D, Akkaraju GR, Couch EF, Canham LT, Coffer JL: The role of nanostructured
mesoporous silicon in discriminating in vitro calcification for electrospun
composite tissue engineering scaffolds. Nanoscale 3(2): 354–361, 2011
Fournier E, Passirani C, Montero-Menei CN, Benoit JP: Biocompatibility of
implantable synthetic polymeric drug carriers: focus on brain biocompatibility.
Biomaterials 24(19): 3311–3331, 2003
Gandolfi MG, Shah SN, Feng R, Prati C, Akintoye SO: Biomimetic calcium-silicate
cements support differentiation of human orofacial mesenchymal stem cells.
J Endod 37(8): 1102–1108, 2011
George J, Kuboki Y, Miyata T: Differentiation of mesenchymal stem cells into os-
teoblasts on honeycomb collagen scaffolds. Biotechnol Bioeng 95(3): 404–411,
2006
Gigante A, Manzotti S, Bevilacqua C, Orciani M, Di Primio R, Mattioli-Belmonte M:
Adult mesenchymal stem cells for bone and cartilage engineering: effect of
scaffold materials. Eur J Histochem 52(3): 169–174, 2008
Giteau A, Vernier-Julienne MC, Aubert-Pouessel A, Benoit JP: How to achieve sus-
tained and complete protein release from PLGA-based microparticles? Int J
Pharm 350(1–2): 14–26, 2008
Glowacki J, Mulliken JB: Demineralized bone implants. Clin Plast Surg 12(2): 233–
241, 1985
Goldberg VM: Selection of bone grafts for revision total hip arthroplasty. Clin
Orthop Relat Res 381: 68–76, 2000
Goldberg VM, Stevenson S: Natural history of autografts and allografts. Clin Orthop
Relat Res 225: 7–16, 1987
Gopferich A: Mechanisms of polymer degradation and erosion. Biomaterials 17(2):
103–114, 1996
Heo SJ, Kim SE, Wei J, Kim DH, Hyun YT, Yun HS, et al: In vitro and animal study of novel
nano-hydroxyapatite/poly(epsilon-caprolactone) composite scaffolds fabricated by
layer manufacturing process. Tissue Eng Part A 15(5): 977–989, 2009
Hess R, Douglas T, Myers KA, Rentsch C, Worch H, Shrive NG, et al: Hydrostatic
pressure stimulation of human mesenchymal stem cells seeded on collagen-
based artificial extracellular matrices. J Biomech Eng 132(2): 021001, 2010
Hosseinkhani H, Inatsugu Y, Hiraoka Y, Inoue S, Shimokawa H, Tabata Y:
Impregnation of plasmid DNA into three-dimensional scaffolds and medium
perfusion enhance in vitro DNA expression of mesenchymal stem cells. Tissue
Eng 11(9–10): 1459–1475, 2005a
Hosseinkhani H, Inatsugu Y, Hiraoka Y, Inoue S, Tabata Y: Perfusion culture
enhances osteogenic differentiation of rat mesenchymal stem cells in collagen
sponge reinforced with poly(glycolic acid) fiber. Tissue Eng 11(9–10): 1476–
1488, 2005b
Huang Y, Jin X, Zhang X, Sun T, Tu J, Tang T, et al: In vitro and in vivo evaluation
of akermanite bioceramics for bone regeneration. Biomaterials 30(28): 5041–
5048, 2009
Huang YX, Ren J, Chen C, Ren TB, Zhou XY: Preparation and properties of
poly(lactide-co-glycolide) (PLGA)/nano-hydroxyapatite (NHA) scaffolds by
thermally induced phase separation and rabbit MSCs culture on scaffolds.
J Biomater Appl 22(5): 409–432, 2008
Hutmacher DW, Schantz JT, Lam CX, Tan KC, Lim TC: State of the art and future
directions of scaffold-based bone engineering from a biomaterials perspective. J Tissue
Eng Regen Med 1(4): 245–260, 2007
Hwang NS, Varghese S, Puleo C, Zhang Z, Elisseeff J: Morphogenetic signals from
chondrocytes promote chondrogenic and osteogenic differentiation of mesenchymal
stem cells. J Cell Physiol 212(2): 281–284, 2007
Ignatius A, Blessing H, Liedert A, Kaspar D, Kreja L, Friemert B, et al: Effects of
mechanical strain on human osteoblastic precursor cells in type I collagen
matrices. Orthopade 33(12): 1386–1393, 2004
77
Janicki P, Boeuf S, Steck E, Egemann M, Kasten P, Richter W: Prediction of in vivo
bone forming potency of bone marrow-derived human mesenchymal stem cells.
Eur Cell Mater 21: 488–507, 2011
Kang SM, Kong B, Oh E, Choi JS, Choi IS: Osteoconductive conjugation of bone
morphogenetic protein-2 onto titanium/titanium oxide surfaces coated with
non-biofouling poly(poly(ethylene glycol) methacrylate). Colloids Surf B Bio-
interfaces 75(1): 385–389, 2010
Kasten P, Luginbühl R, van Griensven M, Barkhausen T, Krettek C, Bohner M, et al:
Comparison of human bone marrow stromal cells seeded on calcium-deficient
hydroxyapatite, beta-tricalcium phosphate and demineralized bone matrix.
Biomaterials 24(15): 2593–2603, 2003
Kasten P, Vogel J, Luginbühl R, Niemeyer P, Tonak M, Lorenz H, et al: Ectopic bone
formation associated with mesenchymal stem cells in a resorbable calcium
deficient hydroxyapatite carrier. Biomaterials 26(29): 5879–5889, 2005
Kasten P, Vogel J, Lüginbuhl R, Niemeyer P, Weiss S, Schneider S, et al: Influence of
platelet-rich plasma on osteogenic differentiation of mesenchymal stem cells and
ectopic bone formation in calcium phosphate ceramics. Cells Tissues Organs
183(2): 68–79, 2006
Kim J, Hefferan TE, Yaszemski MJ, Lu L: Potentials of hydrogels based on poly(-
ethylene glycol) and sebacic acid as orthopedic tissue engineering scaffolds.
Tissue Eng Part A 15(8): 2299–2307, 2009
Korda M, Hua J, Little NJ, Heidari N, Blunn GW: The effect of mesenchymal stromal
cells on the osseoinduction of impaction grafts. Tissue Eng Part A 16(2): 675–
683, 2010
Kruger EA, Im DD, Bischoff DS, Pereira CT, Huang W, Rudkin GH, et al: In vitro
mineralization of human mesenchymal stem cells on three-dimensional type I
collagen versus PLGA scaffolds: a comparative analysis. Plast Reconstr Surg
127(6): 2301–2311, 2011
Ku CH, Johnson PH, Batten P, Sarathchandra P, Chambers RC, Taylor PM, et al:
Collagen synthesis by mesenchymal stem cells and aortic valve interstitial cells
in response to mechanical stretch. Cardiovasc Res 71(3): 548–556, 2006
Langer R, Vacanti JP: Tissue engineering. Science 260(5110): 920–926, 1993
Laschke MW, Harder Y, Amon M, Martin I, Farhadi J, Ring A, et al: Angiogenesis in
tissue engineering: breathing life into constructed tissue substitutes. Tissue Eng
12(8): 2093–2104, 2006
Lee GS, Park JH, Shin US, Kim HW: Direct deposited porous scaffolds of calcium
phosphate cement with alginate for drug delivery and bone tissue engineering.
Acta Biomater 7(8): 3178–3186, 2011
LeGeros RZ: Properties of osteoconductive biomaterials: calcium phosphates. Clin
Orthop Relat Res 395: 81–98, 2002
Leong KF, Cheah CM, Chua CK: Solid freeform fabrication of three-dimensional
scaffolds for engineering replacement tissues and organs. Biomaterials 24(13):
2363–2378, 2003
Leong NL, Jiang J, Lu HH: Polymer-ceramic composite scaffold induces osteogenic
differentiation of human mesenchymal stem cells. Conf Proc IEEE Eng Med Biol
Soc 1: 2651–2654, 2006
Li Y, Dånmark S, Edlund U, Finne-Wistrand A, He X, Norgård M, et al: Reseratrol-
conjugated poly-ε-caprolactone facilitates in vitro mineralization and in vivo bone
regeneration. Acta Biomater 7(2): 751–758, 2011
Liu B, Li X, Liang G, Liu X: VEGF expression in mesenchymal stem cells promotes
bone formation of tissue-engineered bones. Mol Med Report 4(6): 1121–1126,
2011
Liu X, Li X, Fan Y, Zhang G, Li D, Dong W, et al: Repairing goat tibia segmental bone
defect using scaffold cultured with mesenchymal stem cells. J Biomed Mater Res B
Appl Biomater 94(1): 44–52, 2010
Liu XJ, Ren GH, Liao H, Yu L, Yuan L: Induced differentiation of adult human bone
marrow derived mesenchymal stem cells in vitro toward osteoblasts. Di Yi Jun
Yi Da Xue Xue Bao 24(4): 408–411, 2004
Liu YK, Lu QZ, Pei R, Ji HJ, Zhou GS, Zhao XL, et al: The effect of extracellular calcium
and inorganic phosphate on the growth and osteogenic differentiation of
mesenchymal stem cells in vitro: implication for bone tissue engineering.
Biomed Mater 4(2): 025004, 2009
Livingston T, Ducheyne P, Garino J: In vivo evaluation of a bioactive scaffold for bone
tissue engineering. J Biomed Mater Res 62(1): 1–13, 2002
Lund AW, Bush JA, Plopper GE, Stegemann JP: Osteogenic differentiation of
mesenchymal stem cells in defined protein beads. J Biomed Mater Res B Appl
Biomater 87(1): 213–221, 2008
Martins AM, Pham QP, Malafaya PB, Raphael RM, Kasper FK, Reis RL, et al: Natural
stimulus responsive scaffolds/cells for bone tissue engineering influence of
lysozyme upon scaffold degradation and osteogenic differentiation of cultured
marrow stromal cells induced by CaP coatings. Tissue Eng Part A 15(8): 1953–
1963, 2009
Mastrogiacomo M, Muraglia A, Komlev V, Peyrin F, Rustichelli F, Crovace A, et al:
Tissue engineering of bone: search for a better scaffold. Orthod Craniofac Res
8(4): 277–284, 2005
Matsuno T, Hashimoto Y, Adachi S, Omata K, Yoshitaka Y, Ozeki Y, et al:
Preparation of injectable 3D-formed beta-tricalcium phosphate bead/alginate
composite for bone tissue engineering. Dent Mater J 27(6): 827–834, 2008
Matsuno T, Nakamura T, Kuremoto K, Notazawa S, Nakahara T, Hashimoto Y, et
al: Development of beta-tricalcium phosphate/collagen sponge composite for bone
regeneration. Dent Mater J 25(1): 138–144, 2006
Mehlisch DR, Leider AS, Roberts WE: Histologic evaluation of the
bone/graft interface after mandibular augmentation with
hydroxylapatite/purified fibrillar collagen composite implants. Oral Surg Oral
Med Oral Pathol 70(6): 685–692, 1990
Mei F, Tang JM, Zhong JY, Qi WH, Tang Y, Li F: Potential application of
human amniotic mesenchymal cells for seeding cells in bone tissue
engineering. Beijing Da Xue Xue Bao 41(2): 192–195, 2009
Metzler P, von Wilmowsky C, Zimmermann R, Wiltfang J, Schlegel KA: The effect
of current used bone substitution materials and platelet-rich plasma on periosteal
cells by ectopic site implantation: an in-vivo pilot study. J Craniomaxillofac Surg
40(5): 409–415, 2012
Miranda P, Pajares A, Saiz E, Tomsia AP, Guiberteau F: Mechanical properties of
calcium phosphate scaffolds fabricated by robocasting. J Biomed Mater Res A 85(1):
218–227, 2008
Miranda SC, Silva GA, Hell RC, Martins MD, Alves JB, Goes AM: Three-dimensional
culture of rat BMMSCs in a porous chitosan-gelatin scaffold: a promising as-
sociation for bone tissue engineering in oral reconstruction. Arch Oral Biol
56(1): 1–15, 2011
Moioli EK, Hong L, Guardado J, Clark PA, Mao JJ: Sustained release of TGFbeta3 from
PLGA microspheres and its effect on early osteogenic differentiation of human
mesenchymal stem cells. Tissue Eng 12(3): 537–546, 2006
Nair MB, Varma HK, Menon KV, Shenoy SJ, John A: Reconstruction of goat femur
segmental defects using triphasic ceramic-coated hydroxyapatite in combination
with autologous cells and platelet-rich plasma. Acta Biomater 5(5): 1742–1755,
2009
Nandakumar A, Fernandes H, de Boer J, Moroni L, Habibovic P, van Blitterswijk CA:
Fabrication of bioactive composite scaffolds by electrospinning for bone
regeneration. Macromol Biosci 10(11): 1365–1373, 2010
Niu X, Feng Q, Wang M, Guo X, Zheng Q: Porous nano-HA/collagen/PLLA scaffold
containing chitosan microspheres for controlled delivery of synthetic peptide
derived from BMP-2. J Control Release 134(2): 111–117, 2009
Nuttelman CR, Tripodi MC, Anseth KS: Dexamethasone-functionalized gels induce
osteogenic differentiation of encapsulated hMSCs. J Biomed Mater Res A 76(1):
183–195, 2006
Pang Y, Cui P, Chen W: Incipient establishment of human bone marrow mesen-
chymal stem cells bank. Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi 19(7): 578–
582, 2005
Pang YG, Cui PC, Chen WX, Cao YX: Experimental research using human bone
marrow mesenchymal stem cells as the seed cells for bone and cartilage tissue
engineering. Xi Bao Yu Fen Zi Mian Yi Xue Za Zhi 20(3): 306–309, 2004
Park KH, Kim H, Moon S, Na K: Bone morphogenic protein-2 (BMP-2) loaded
nanoparticles mixed with human mesenchymal stem cell in fibrin hydrogel for bone
tissue engineering. J Biosci Bioeng 108(6): 530–537, 2009
Parker BD, Bauer DC, Ensrud KE, Ix JH: Association of osteocalcin and abdominal
aortic calcification in older women: the study of osteoporotic fractures. Calcif
Tissue Int 86(3): 185–191, 2010
Pereira CT, Huang W, Jarrahy R, Rudkin G, Yamaguchi DT, Miller TA: Human and
mouse osteoprogenitor cells exhibit distinct patterns of osteogenesis in three-
dimensional tissue engineering scaffolds. Plast Reconstr Surg 124(6): 1869–1879,
2009
Pinholt EM, Haanaes HR, Donath K, Bang G: Titanium implant insertion into dog
alveolar ridges augmented by allogenic material. Clin Oral Implants Res 5(4):
213–219, 1994
Pinholt EM, Haanaes HR, Roervik M, Donath K, Bang G: Alveolar ridge
augmentation by osteoinductive materials in goats. Scand J Dent Res 100(6): 361–
365, 1992
Porter AE, Patel N, Skepper JN, Best SM, Bonfield W: Effect of sintered silicate-
substituted hydroxyapatite on remodeling processes at the bone-implant
interface. Biomaterials 25(16): 3303–3314, 2004
Quaglia F: Bioinspired tissue engineering: the great promis of protein delivery
technologies. Int J Pharm 364(2): 281–297, 2008
Reichert JC, Heymer A, Berner A, Eulert J, Nöth U: Fabrication of polycaprolactone
collagen hydrogel constructs seeded with mesenchymal stem cells for bone
regeneration. Biomed Mater 4(6): 065001, 2009
Rezwan K, Chen QZ, Blaker JJ, Boccaccini AR: Biodegradable and bioactive porous
polymer/inorganic composite scaffolds for bone tissue engineering. Biomaterials
27(18): 3413–3431, 2006
Rust PA, Kalsi P, Briggs TW, Cannon SR, Blunn GW: Will mesenchymal stem cells
differentiate into osteoblasts on allograft? Clin Orthop Relat Res 457: 220–226,
2007
Sala A, Hänseler P, Ranga A, Lutolf MP, Vörös J, Ehrbar M, et al: Engineering 3D cell
instructive microenvironments by rational assembly of artificial extracellular
matrices and cell patterning. Integr Biol (Camp) 3(11): 1102–1111, 2011
Salgado AJ, Coutinho OP, Reis RL: Bone tissue engineering: state of the art and
future trends. Macromol Biosci 4(8): 743–765, 2004
Sawase T, Jimbo R, Baba K, Shibata Y, Ikeda T, Atsuta M: Photo-induced hydrophi-
licity enhances initial cell behavior and early bone apposition. Clin Oral Im-
plants Res 19(5): 491–496, 2008
Schofer MD, Boudriot U, Wack C, Leifeld I, Gräbedünkel C, Dersch R, et al: Influence
of nanofibers on the growth and osteogenic differentiation of stem cells: a
comparison of biological collagen nanofibers and synthetic PLLA fibers. J Mater
Sci Mater Med 20(3): 767–774, 2009
Schofer MD, Fuchs-Winkelmann S, Gräbedünkel C, Wack C, Dersch R, Rudisile M, et
al: Influence of poly(L-lactic acid) nanofibers and BMP-2-containing poly(L-
lactic acid) nanofibers on growth and osteogenic differentiation of human
mesenchymal stem cells. Scientific World Journal 8: 1269–1279, 2008
Schumann P, Tavassol F, Lindhorst D, Stuehmer C, Bormann KH, Kampmann A, et al:
Consequences of seeded cell type on vascularization of tissue engineering
constructs in vivo. Microvasc Res 78(2): 180–190, 2009
Seitz S, Ern K, Lamper G, Docheva D, Drosse I, Milz S, et al: Influence of in vitro
cultivation on the integration of cell-matrix constructs after subcutaneous
implantation. Tissue Eng 13(5): 1059–1067, 2007
Shafiee A, Seyedjafari E, Soleimani M, Ahmadbeigi N, Dinarvand P, Ghaemi N:
A comparison between osteogenic differentiation of human unrestricted somatic
stem cells and mesenchymal stem cells from bone marrow and adipose tissue.
Biotechnol Lett 33(6): 1257–1264, 2011
Shi X, Wang Y, Ren L, Huang W, Wang DA: A protein/antibiotic releasing poly(lactic-
co-glycolic acid)/lecithin scaffold for bone repair applications. Int J Pharm 373(1–2):
85–92, 2009
Shih YR, Chen CN, Tsai SW, Wang YJ, Lee OK: Growth of mesenchymal stem cells on
electrospun type I collagen nanofibers. Stem Cells 24(11): 2391–2397, 2006
Silva TS, Primo BT, Silva Júnior AN, Machado DC, Viezzer C, Santos LA: Use of
calcium phosphate cement scaffolds for bone tissue engineering: in vitro
study. Acta Cir Bras 26(1): 7–11, 2011
78
Sittichokechaiwut A, Edwards JH, Scutt AM, Reilly GC: Short bouts of mechanical
loading are as effective as dexamethasone at inducing matrix production by
human bone marrow mesenchymal stem cell. Eur Cell Mater 20: 45–57, 2010
Stiehler M, Bünger C, Baatrup A, Lind M, Kassem M, Mygind T: Effect of dynamic 3-D
culture on proliferation, distribution, and osteogenic differentiation of human
mesenchymal stem cells. J Biomed Mater Res A 89(1): 96–107, 2009
Stockmann P, Park J, von Wilmowsky C, Nkenke E, Felszeghy E, Dehner JF, et al:
Guided bone regeneration in pig calvarial bone defects using autologous
mesenchymal stem/progenitor cells – a comparison of different tissue sources.
J Craniomaxillofac Surg 40(4): 310–320, 2012
Sun H, Qu Z, Guo Y, Zang G, Yang B: In vitro and in vivo effects of rat kidney vascular
endothelial cells on osteogenesis of rat bone marrow mesenchymal stem cells
growing on polylactide-glycoli acid (PLGA) scaffolds. Biomed Eng Online 6: 41, 2007
Sutter W, Stein E, Koehn J, Schmidl C, Lezaic V, Ewers R, et al: Effect of different
biomaterials on the expression pattern of the transcription factor Ets2 in bone-
like constructs. J Craniomaxillofac Surg 37(5): 263–271, 2009
Takahashi K, Igura K, Zhang X, Mitsuru A, Takahashi TA: Effects of osteogenic in-
duction on mesenchymal cells from fetal and maternal parts of human placenta.
Cell Transplant 13(4): 337–341, 2004
Tseng PC, Young TH, Wang TM, Peng HW, Hou SM, Yen ML: Spontaneous osteo-
genesis of MSCs cultured on 3D microcarriers through alteration of cytoskeletal
tension. Biomaterials 33(2): 556–564, 2012
Turhani D, Watzinger E, Weissenbock M, Yerit K, Cvikl B, Thurnher D, et al: Three-
dimensional composites manufactured with human mesenchymal cambial
layer precursor cells as an alternative for sinus floor augmentation: an in vitro
study. Clin Oral Implants Res 16(4): 417–424, 2005
Urist MR: Bone: formation by autoinduction. 1965. Clin Orthop Relat Res 395: 4–10,
2002
Valarmathi MT, Yost MJ, Goodwin RL, Potts JD: The influence of proepicardial cells
on the osteogenic potential of marrow stromal cells in a three-dimensional
tubular scaffold. Biomaterials 29(14): 2203–2216, 2008
Vermonden T, Fedorovich NE, van Geemen D, Alblas J, van Nostrum CF, Dhert
WJ, et al: Photopolymerized thermosensitive hydrogels: synthesis,
degradation, and cytocompatibility. Biomacromolecules 9(3): 919–926, 2008
Visscher GE, Robison RL, Maulding HV, Fong JW, Pearson JE, Argentieri GJ:
Biodegradation of and tissue reaction to 50:50 poly(DL-lactide-co-glycolide)
microcapsules. J Biomed Mater Res 19(3): 349–365, 1985
Wang L, Fan H, Zhang ZY, Lou AJ, Pei GX, Jiang S, et al: Osteogenesis and
angiogenesis of tissue-engineered bone constructed by prevascularized β-
tricalcium phosphate scaffold and mesenchymal stem cells. Biomaterials 31(36):
9452–9461, 2010
Wang M: Developing bioactive composite materials for tissue replacement.
Biomaterials 24(13): 2133–2151, 2003
Warren SM, Nacamuli RK, Song HM, Longaker MT: Tissue-engineered bone using
mesenchymal stem cells and a biodegradable scaffold. J Craniofac Surg 15(1): 34–
37, 2004
Weir MD, Xu HH: Culture human mesenchymal stem cells with calcium phosphate
cement scaffolds for bone repair. J Biomed Mater Res B Appl Biomater 93(1):
93–105, 2010
Weissenboeck M, Stein E, Undt G, Ewers R, Lauer G, Turhani D: Particle size of
hydroxyapatite granules calcified from red algae affects the osteogenic poten-
tial of human mesenchymal stem cells in vitro. Cells Tissues Organs 182(2): 79–
88, 2006
Wen F, Magalhães R, Gouk SS, Bhakta G, Lee KH, Hutmacher DW, et al: Vitreous
cryopreservation of nanofibrous tissue-engineered constructs generated using
mesenchymal stromal cells. Tissue Eng Part C Methods 15(1): 105–114, 2009
Xin R, Leng Y, Chen J, Zhang Q: A comparative study of calcium phosphate formation
on bioceramics in vitro and in vivo. Biomaterials 26(33): 6477–6486, 2005
Xu HH, Zhao L, Weir MD: Stem cell-calcium phosphate constructs for bone engi-
neering. J Dent Res 89(12): 1482–1488, 2010
Yang X, Zhang S, Pang X, Fan M: Pro-inflammatory cytokines induce odontogenic
differentiation of dental pulp-derived stem cells. J Cell Biochem 113(2): 669–677,
2012
Yang XB, Green DW, Roach HI, Clarke NM, Anderson HC, Howdle SM, et al: Novel
osteoinductive biomimetic scaffolds stimulate human osteoprogenitor activity-
implications for skeletal repair. Connect Tissue Res 44(Suppl. 1): 312–317, 2003
Yin XX, Chen ZQ, Guo ZQ: Inducing human marrow mesenchymal stem cells into
osteoblasts directionally and identification of their osteogenesis characteristics.
Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi 18(2): 88–91, 2004
Younger EM, Chapman MW: Morbidity at bone graft donor sites. J Orthop Trauma
3(3): 192–195, 1989
Zhang YG, Yang Z, Zhang H, Wang C, Liu M, Guo X, et al: Effect of negative pressure
on human bone marrow mesenchymal stem cells in vitro. Connect Tissue Res
51(1): 14–21, 2010
Zhou GS, Su ZY, Cai YR, Liu YK, Dai LC, Tang RK, et al: Different effects of nanophase
and conventional hydroxyapatite thin films on attachment, proliferation and
osteogenic differentiation of bone marrow derived mesenchymal stem cells.
Biomed Mater Eng 17(6): 387–395, 2007
Zhou Y, Yan Z, Zhang H, Lu W, Liu S, Huang X, et al: Expansion and delivery of
adipose-derived mesenchymal stem cells on three microcarriers for soft tissue
regeneration. Tissue Eng Part A 17(23–24): 2981–2997, 2011
Zou D, Zhang Z, He J, Zhu S, Wang S, Zhang W, et al: Repairing critical-sized calvarial
defects with BMSCs modified by a constitutively active form of hypoxia-
inducible factor-1α and a phosphate cement scaffold. Biomaterials 32(36):
9707–9718, 2011
79
12. Paper II
Journal of Cranio-Maxillo-Facial Surgery 41 (2013) e213–e220
Contents lists available at SciVerse ScienceDirect
Journal of Cranio-Maxillo-Facial Surgery
journal homepage: www.jcmfs.com
Adipose derived mesenchymal stem cells – Their osteogenicity
and osteoblast in vitro mineralization on titanium granule carriers
Morten Dahl a, Susanne Syberg
b, c, Niklas Rye Jørgensen
b, c, Else Marie Pinholt
a, *
a Institute of Odontology, Department of Oral and Maxillofacial Surgery, Faculty of Health Sciences, University of Copenhagen, Nørre Allé 20, 2200 Copenhagen, Denmark
b Research Center for Ageing and Osteoporosis, Department of Clinical Biochemistry, Copenhagen University Hospital Glostrup, Glostrup, Denmark
c Research Center for Ageing and Osteoporosis, Department of Medicine, Copenhagen University Hospital Glostrup, Glostrup, Denmark
A R T I C L E I N F O
Article history:
Paper received 6 September 2012
Accepted 9 January 2013
Keywords:
Adipose mesenchymal stem cells
Titanium
Drug carriers Tissue engineering
Mineralization
ABSTRACT
Purpose: Adipose derived mesenchymal stem cells (ADMSCs) may be osteogenic, may generate neo-
angiogenisis and may be progenitors for differentiated osteoblast mineralization. Titanium granules may
be suitable as carriers for these cells. The aim was to demonstrate the osteogenic potential of ADMSCs
and the effect of porous non-oxidized (Ti) and oxidized titanium (TiO2) granules as carriers for ADMSCs
mineralization in vitro.
Materials and methods: ADMSCs were isolated, cultivated in osteoblast medium and evaluated for
alkaline phosphatase (ALP) assay, RNA isolation, and ALP staining. Osteoblast in vitro mineralization cells
without granules or seeded on Ti or TiO2 granules were evaluated for Alizarin Red assay and RNA iso-
lation for later gene expressing.
Results: ADMSCs express osteoblastic lineage genes, CBFA-1 and stain strongly for ALP. Mineralization
was significantly higher for cells seeded on TiO2 than on Ti granules or pure cells. Expression of ALPL and
RUNX2 was significantly higher for cells seeded on TiO2 granules and expression of COL1α1 for pure cells
was significantly higher than for cells seeded on granules.
Conclusion: ADMSCs have osteogenic potential. Mineralization was significantly high when cells were
seeded on TiO2 granules. TiO2 granules may be used as carriers for adipose derived mesenchymal
osteoblastic cells from laboratory bench to the patient.
© 2013 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights
reserved.
1. Introduction
It is desirable to combine biomaterials with osteogenic cells in
an attempt to promote osteogenesis (Goldberg and Stevenson,
1987; Younger and Chapman, 1989; Chen et al., 2005; Leong et al., 2006; Seitz et al., 2007; Martins et al., 2009; Sutter et al.,
2009; Ben-David et al., 2010; Dahl et al., submitted for
publication; Behnia et al., 2012; Metzler et al., 2012; Stockmann et al., 2012).
Studies have investigated different carriers such as type I col-
lagen, titanium, calcium carbonates, calcium phosphates, hy-droxyapatite, tricalcium phosphate, calcium-deficient
hydroxyapatite and polylactic acid–polyglycolic acid copolymer
with modifications for mineralization and gene expression (Dahl et al., submitted for publication). The numerical values of an
* Corresponding author. Tel.: +45 33326611.
E-mail address: [email protected] (E.M. Pinholt).
outcome of a carrier combined with mesenchymal stem cells
comprise reactions from carriers as well as from cells, respectively.
Measurements of each of these, carrier and cells, are deemed
necessary for evaluation. Unfortunately most published experi-
ments do not use control measurements on carrier without cells
(Dahl et al., submitted for publication).
Available titanium granules are irregular porous granules of
commercially pure titanium. Titanium is biocompatible and non-
toxic, even in large doses. The porous properties of Tigran, a tita-
nium granule product, may lead to ingrowth of newly formed bone
by interlocking them with each other thus creating an uninter-
rupted structure (Bystedt and Rasmusson, 2009). Histology and
scanning electron microscopy from both clinical and experimental
studies have revealed ingrowth of bone into implanted porous ti-
tanium granules resulting in the formation of an integrated mantle
of bone and titanium granules (Alffram et al., 2007; Turner et al.,
2007). Titanium granules provide the necessary initial mechanical
stability for bone ingrowth by keeping the blood clot, thereby ini-
tiating bone development (Masuda et al., 1997; Hong et al., 1999).
1010-5182/$ – see front matter © 2013 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jcms.2013.01.021
80
)
Fig. 1. Scanning Electron Microscopy, magnification x295, of porous titanium granule
(Natix®
).
The osseointegration process of dental titanium implants is
dependent of the surface which mainly comprises titanium oxide
(Sul et al., 2005). Since the available titanium granules are manu-
factured in a non-oxidized, Ti, and an oxidized, TiO2, form, they are
interesting to evaluate as carriers for mesenchymal stem cells
(MSCs).
In light of previous studies a cell population has been identified
in adipose tissue, adipose derived mesenchymal stem cells
(ADMSCs), similar to the population of MSCs from bone marrow
stroma (Zuk et al., 2001, 2002). The osteogenic potential of ADMSCs
is supported by in vitro studies (Zuk et al., 2001). These cells are
able to undergo in vitro differentiation into cell types found in
bone, cartilage, fat and muscle tissue by using induction (Zuk et al.,
2001). Besides that, human ADMSCs are shown to differentiate into
endothelial cells in vitro and to improve postnatal neo-
vascularization in vivo. Furthermore ADMSCs are shown to improve
skin regeneration by increasing neoangiogenesis and collagen
synthesis within a dermal substitute (Cao et al., 2005; Meruane
et al., 2012).
1.1 Hypothesis
Adipose derived mesenchymal stem cells differentiated into
osteoblasts and seeded on titanium granules, oxidized and non-
oxidized, will result in a significantly increased positive outcome
of osteoblast mineralization of the oxidized form.
1.2 Aim
The aim of the current in vitro study was to demonstrate the
osteogenic potential of ADMSCs and to evaluate the effect of porous
titanium granules, oxidized and non-oxidized, as a carrier material
for ADMSCs in osteoblast mineralization.
2. Materials and methods
2.1 Fabrication and characterization of scaffolds
Porous titanium granule material (Natix®) (Tigran Technologies
AB, Malmö, Sweden) (Fig. 1) is commercially pure titanium grade I
bone substitute intended for use in dental intraosseous and oral/
maxillofacial defects for augmentation purposes. Natix® (Tigran
Technologies AB) consists of porous, unalloyed, irregularly shaped
titanium particles, 0.7–1.0 mm in size, with a porosity of about 80%.
Natix® granules are fabricated in a grey non-oxidized, Ti, and a white
oxidized, TiO2, version. The oxide layer that surrounds a grey
granule is only 1 µ m. By heating the grey version of Natix® granules,
the oxide layer increases and becomes white. The white granules are
composed almost entirely of titanium oxide, TiO2 (Tigran, 2010).
The granules are provided sterile for single use by gamma ra-
diation at minimum 25 kGy, ISO 11137-1, and a sterility assurance
level (SAL) of 10-6.
2.2 Murine adipose mesenchymal stem cells
The current study was approved by the Danish Ethical Animal
Research Committee. Balb/cJ mice were euthanized and abdominal
adipose tissue was collected. The tissue was divided into fine par-
ticles and washed in Dulbecco’s Phosphate Buffered Saline w/o
Ca2+ and Mg2+ (PBS) (Lonza, Verviers, Belgium). Next, the tissue was
collagenase treated (7.5 mg collagenase in 10 mL PBS) (SIGMA–
ALDRICH, St. Louis, USA) and after inactivation of the collagenase, 106
cells were plated in 100 mm culture dishes (NUNC, Roskilde,
Denmark) with mouse osteoblast medium (mOB MEM: 435 mL
MEM Earle’s w/o Phenol red (MEM) (Invitrogen, Auckland, New
Zealand), 50 mL Fetal Calf Serum (FCS) heat inactivated (SIGMA–
ALDRICH), 5 mL Penicillin/Streptomycin (P/S) (Invitrogen), 5 mL
glutamax (Invitrogen), 500 µL L-Ascorbic Acid Phosphate Magnesium
Salt n-Hydrate (Wako Chemicals, Virginia, USA) (50 mg/mL) and 5
mL glycerol 2-phosphate disodium salt hydrate (SIGMA–ALDRICH)
(1 M)). Cells were maintained in a humidified atmosphere at 37 oC
and 5% CO2. After two weeks most of the cells were differentiated into
osteoblasts judged by visual inspection in a light microscope.
For demonstrating the potential of ADMSCs to differentiate into
osteoblasts, cells were cultured for eight weeks in mOB MEM
(Invitrogen) and transferred to 6-well plates (NUNC) for alkaline
phosphatase assay and RNA isolation, and onto 2-chamber cell
culture slides (SIGMA–ALDRICH) for alkaline phosphatase staining. The
cells were cultured for an additional 4 weeks.
For the evaluation of the effect of porous titanium granules as
carrier material for ADMSCs, cells were cultured for an initial two
weeks. Thereafter, they were trypsinized and transferred to the
titanium granules.
For mineralization assays 24-well plates (NUNC) were used with
inserts (FALCON, Franklin Lakes, NJ, USA) with a pore size on 3.0
µm. The amount of the carrier, granules in one insert, was 0.20 mL. Four
different combinations of cells and carriers were used: Ti granules
combined with cells (4 x 104 cells per insert) (Ti+) and TiO2 granules
combined with cells (TiO2+ and as controls no cells in two inserts (Ti
granules without cells (Ti-) and TiO2 granules without cells (TiO2-)).
For RNA isolation 6-well plates (NUNC) were used with inserts
(FALCON) with a pore size on 3.0 µm. The amount of the carrier,
granules in one insert, was 1.5 mL. Cells were kept in mOB MEM
(Invitrogen) and maintained at 37 °C and 5% CO2. Fig. 2 shows the
time schedule for the in vitro mineralization experiment.
2.3 RNA isolation
RNA was isolated using the RNeasy Mini kit from QIAGEN ac-
cording the manufacture’s recommendations. The RNA content in the
samples was measured using NanoDrop 2000c (Thermo Scientific,
Søborg, Denmark) and RNA was stored at -80 °C.
For showing adipose mesenchymal stem cells potential in dif-
ferentiating into osteoblast RNA was isolated at week 12.
81
Fig. 2. Time schedule for the in vitro mineralization experiment.
For the evaluation of the effect of porous titanium granules as
carrier material for adipose mesenchymal stem cells RNA was iso-
lated at day 1, 4 and 8 after the first two weeks of culturing.
2.4 cDNA and polymerase chain reaction (HotStarTaq PCR)
Mastermix was produced according to the manufacture’s rec-
ommendation (Omniscript Reverse Transcription, QIAGEN) corre-
sponding to 15 µg RNA. 100 µL Mastermix was added for each tube
containing 5 µg RNA. Hereafter the reverse transcription process was
done using an Eppendorf Mastercycler. To determine the expression of
genes related to osteoblast differentiation, activity and osteogenesis,
primers for the following genes were ordered (Applied Biosystems,
Foster City, USA): CBFA-1 (RUNX2), Osteocalcin (OC), Peroxisome
Proliferator-Activated Receptor-γ1 (PPAR- γ1), Peroxisome Proliferator-
Activated Receptor-γ2 (PPAR-γ2) plus control GAPDH. The PCR
reaction was as follows: 10 min at 95 °C, hereafter 40 cycles with 2
min at 94 °C, 45 s at hybridizing temperature and 60 s at 72 °C, and
finally 10 min at 72 °C.
2.5 Alkaline phosphatase colouring assay
First medium was removed from the chamber slides (SIGMA–
ALDRICH). The cell layer was dried and fixated with 70% ethanol
for 5 min, after which they were incubated in alkaline phosphatase
(ALP) solution (substrate 1:150 mg variamin blausaltz B and sub-
strate 2:75 mg sodium-naphthyl-P dissolved in a buffer, 75 mL 2.1%
amino-methyl-propandiol 0.2 M, 15 mL HCl 0.1 M and 210 mL
distilled H2O) (SIGMA–ALDRICH) for 20 min at 4 °C. Next, the cells
were rinsed in tap water for 5 min and counter-stained in Meyers
haematoxylin (SIGMA–ALDRICH) for 30 s, and again rinsed in tap
water for 10 min where after the coverslip was attached to the slide
and the extent of the staining examined in a light microscope.
2.6 Protein corrected ALP assay
First, cells were lysed to obtain the intracellularly located ALP,
next, assay buffer, 7 mL Sigma 211 Buffer (1.5 M), 1 mL MgCl2 (1 M),
92 mL distilled H2O, was prepared. Substrate solution was prepared
by dissolving p-Nitrophenyl-Phosphat tablet in 10 mL assay buffer.
Sigma p-Nitrophenyl-Phosphat standard solutions were made (0,
25, 50, 75, 100, 125 and 150 nmol/mL) by dilution with assay buffer
and by adding 200 µL NaOH (1.0 M). Assay buffer and substrate
solution was heated in waterbath at 37 °C for 10 min. The cell lysate
was diluted 25 times (20 µL lysate + 480 µL TBS buffer). 20 µL
prediluted lysate, control or standard was transferred into a 96-well
plate (NUNC) in duplicate. Hereafter 80 µL assay buffer (37 °C) and
100 µL substrate was added to each well, and the plate was
incubated at 37 °C for 30 min. The reaction was terminated by
adding 100 µL NaOH (1.0 M). Absorbance was read at 405 nm. To determine protein content, the Pierce BCA Protein Assay
(Thermo Scientific) was used. Cell lysate was thawed. The reaction
solution, BCA “Working Reagent” composed of 25 mL BCA reagent
A and 500 µL reagent B. An albumin standard was prepared with
the following concentrations: 0, 50, 100, 200, 300, 400 and 500
µg/mL by diluting albumin standard with TBS buffer. 50 µL lysate, control or standard was transferred to a 96-well plate (NUNC) in
duplication. 200 µL BCA “Working Reagent” was added to each
well. The reaction was incubated at 37 °C for 60 min. After cooling down the plate to room temperature for 10 min, the absorbance
was read at 562 nm.
2.7 Osteoblast mineralization assay – Alizarin Red assay
The Alizarin Red-S (AR-S) assay (Puchtler et al., 1969; Bonewald
et al., 2003; Wang et al., 2006) was used to determine the amount
of mineralized matrix formed by the osteoblasts. It was performed
on the cells after they had been transferred to the carriers, on day 1,
3, 7, 10, 14, 17 and 21. After the proliferation assay the cells in the
inserts were fixated in 70% ethanol for 1 h. Next, cells were stained
in 1 mL 40 mM AR-S, pH 4.2 (684.6 mg Alizarin Red-S, AR-S,
(SIGMA–ALDRICH) dissolved in 50 mL distilled water, pH adjusted
to 4.2 with NaOH/HCl) for 10 min at room temperature and
destained in 1 mL 10% Cetylpyridinium Chloride (CPC) (SIGMA–
ALDRICH) (55 g CPC dissolved in 550 mL 10 mM Sodium Phos-
phate (Region Hovedstadens Apotek, København, Denmark)) for
15 min at room temperature with rotation (150 rpm). Standard
dilutions and AR-S-extracts (1 mM AR-S standard – 34.2 mg AR-S
dissolved in 10 mL 10% CPC, from where 1 mL was added another
9 mL 10% CPC, fresh solution for each assay-day) were pipetted into
a 96-well plate (NUNC) (200 µL/well in duplicate), and the AR-S
concentration was determined on a plate reader at 562 nm (infin-
ite M200, TECAN).
2.8 Reverse Transcription Polymerase Chain Reaction (RT-PCR)
PCR OneStep High Capacity cDNA Reverse Transcription Kit (RT)
(Applied Biosystems) was used to prepare 2x RT master mix (per
20 µL reaction) on ice calculated for 5 µg RNA (1 mL 10× RT Buffer,
0.4 mL 25× dNTP Mix (100 mM), 1 mL 10× RT Random Primers,
0.5 mL Multiscribe™ Reverse Transcriptase, 2.1 mL Nuclease-free
H2O). PCR reactions were performed using a PTC-100TM Pro-
grammable Thermal Controller (MJ Research, Inc.). The program,
ABI-RT, was as follows: 10 min at 25 °C, 120 min at 37 °C, 5 s at
85 °C, final extension – hold at 4 °C and cDNA RT tubes (Eppendorf)
were stored at -18 °C.
Isolation of
adipose tissue
from mice.
Start culturing
Day -14
Cells
seeded on
well-plates
Day 0
Day 1 Day 3 Day 4 Day 7 Day 8 Day 10 Day 14 Day 17 Day 21
Mineralization
RNA isolation
Gene expression
Mineralization
RNA isolation
Gene expression
RNA isolation
Gene expression
Mineralization Mineralization Mineralization Mineralization Mineralization
82
AL
P p
-Nit
rophen
ol/
pro
tein
mm
ol/
g
Fig. 3. RT-PCR assay for osteoblastic genes. Adipose der ived mesenchymal
stem cel ls (ADMSC) highly express CBFA-1 (RUNX2) and smal l amounts
of PPAR-γ2 expression . ADMSC show st rong osteoblast phenotype .
2.9 Quantitative Polymerase Chain Reaction (Quantitative PCR)
A comparative CT (relative standard curve) method was selected
for Quantitative PCR performed by OneStep Real Time PCR System
(Applied Biosystems). TaqMan Gene Expression Assays (Applied
Biosystems) of alkaline phosphatase (ALPL), collagen type 1
(COL1α1) and osteoblast transcription factor, runt-related gene 2
(RUNX2) were used as target genes and GAPDH was used as
endogenous control. TaqMan Universal PCR master mix 2x (10 µL)
(Applied Biosystems), sterile water Mini-Plasco (6 µL) (B. Braun
Melsungen AG, Melsungen, Germany) and gene assay on-demand
Expression mix 20x (1 µL) (Applied Biosystems) was used in or-
der to prepare 17 µL master mix. Manually prepared cDNA (3 µL)
and master mix (17 µL) were pipetted to each well of 48-well
MicroAmp Fast Optical Reaction Plate (Applied Biosystems) to
reach a volume of 20 µL in each well. The plates were covered with
MicroAmp 48-Well Optical Adhesive Film PCR Compatible, DNA/
RNA/RNase Free (Applied Biosystems), where after the plate was
spinned for 5 min at 3575 rpm at 20 °C (Heraeus, Labofuge 400 R
Centrifuge, Thermo Scientific) right before measuring.
2.10 Statistical analysis
All results were analyzed either by one-way ANOVA or by stu-
dents T-test for post hoc comparison. *Indicated significance level
for comparison between TiO2 granules and Ti granules, while #
indicated significance level for comparison between TiO2 granules
and cells. Significance levels of 0.05 were used throughout the
study.
3. Results
Numerical values can be found in Supplementary Table.
3.1 Adipose tissue-derived mesenchymal cells show strong osteoblast phenotype
Gene expression of osteoblast-specific genes is shown in Fig. 3.
The ADMSCs highly express genes characteristic of the osteoblastic
lineage including CBFA-1 (RUNX2) and only very small amounts of
the adipocyte-related gene PPAR-γ1 and none PPAR-γ2 expression,
thus none expression of OC, which is the last step before mineral-
ization. In addition, the cells stained strongly for ALP, which is
specific for osteoblastic cells (Fig. 4a).
When comparing the ALP activity of the adipose tissue derived
osteoblasts, ADMSC, with that of mature osteoblasts extracted from
mouse bone, bOB, and osteoblasts differentiated from bone marrow
mesenchymal stem cells, BMSC, we found no significant difference
between the adipose tissue derived osteoblasts and the other two
osteoblastic cell lineages (Fig. 4b), respectively p = 1 and p = 0.947.
However, the bone marrow derived osteoblasts had significantly
lower ALP activity than the bone derived osteoblasts.
As the ADMSCs appeared to have osteoblast phenotype, we
chose to use these in the second part of the study on the carriers.
3.2. Osteoblast mineralization assay – Alizarin Red assay
The quantity of calcium per scaffold was determined using Ali-
zarin Red-S Staining. Fig. 5 shows relative total calcium contents as
absorbance as mean ± SEM (n = 8 for each condition and time
point) after editing from the controls without cells. Formation of
mineralized matrix from pure cells was very slow. It rose towards
day 7, after which it stabilized. However, an initial increase in the
mineralization on cells seeded on both TiO2 and Ti granules was
shown. After day 7 the curve stabilized for cells seeded on Ti
granules, while it continued to rise for cells seeded on TiO2 gran-
ules. The mineralization for cells seeded on TiO2 granules appeared
to peak between day 14 and 21.
Fig. 6 shows relative total calcium contents as absorbance as
mean for pure cells and an empty insert with only medium (Fig. 6a),
cells seeded on Ti granules and an insert with Ti granules and
medium (Fig. 6b), and finally cells seeded on TiO2 granules and an
insert with TiO2 granules and medium (Fig. 6c). Notice the high
value for the controls without any cells.
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0,0
BMSC ADMSC bOB
Fig. 4. a) Light microscopy, magnification x50. ALP stains positively. Osteoblasts are
shown by arrows. b) ALP enzymatic activity. Bone marrow mesenchymal stem cells
(BMSC), adipose derived mesenchymal stem cells (ADMSC), mature bone osteoblast
(bOB).
a
b
83
0,8
1,2
0,6
0,4
0,2
0,0
1,0
0,30
0,35
0,20
0,15
0,10
0,05
0,00
0,25
0,12
0,08
0,06
0,04
0,02
0,00
0,10
0,00
0,05
0,10
0,15
0,20
0,25
0,30
0,35
0,40
0,45
Ab
sorb
an
ce
TiO2
Ti
Cells
###
***
##
**
##
**
Ab
sorb
an
ce
Ab
sorb
an
ce
Ab
sorb
an
ce
Cells
Empty
**
*
**
*** ***
**
** * *
Ti + cells
Ti – cells
TiO2 + cells
TiO2 – cells
**
*
*
*
3.3. Quantitative PCR
Quantitative PCR was performed to evaluate the gene expres-
sion of ALPL, COL1α1 and RUNX2 in each assay (Fig. 7). This figure
shows relative gene (ALPL, COL1α1 and RUNX2) expression for pure
cells, cells seeded on Ti granules and cells seeded on TiO2 granules.
Data presented as mean ± SEM (n = 5–8 for each condition and
time point). Already at day 1, a higher gene expression of ALPL was
noticed for cells seeded on Ti granules and even more for cells
seeded on TiO2 granules, after which expression dropped to a lower
level. The gene expression of ALPL for pure cells was rising slowly
from day 1 towards day 8. The expression of COL1α1 for pure cells
was higher than for cells seeded on granules. The COL1α1 expres-
sion for cells decreased markedly from day 1 to day 8, while the
expression of COL1α1 for cells seeded on Ti and TiO2 granules was
stable. Expression of RUNX2 was significantly higher for cells see-
ded on TiO2 granules at day 1, where after it decreased over time.
The expression of RUNX2 for cells and cells seeded on Ti granules
was stable and low.
4. Discussion
In our study, we isolated MSCs from adipose tissue, and results
showed that these could be differentiated into fully mature and
functional osteogenic osteoblasts as determined both by the
expression of RNA for a number of genes characteristic for the
osteoblastic cell lineage, by the activity of alkaline phosphatase, as
well as by their ability to form mineralized matrix in vitro, and to
the same extent as for bone derived mature osteoblasts. Adipose
tissue as a source of osteogenic cells provides some advantages over
c
10
Days
1 3 7 10 14 17 21
Days
Fig. 5. Relative total calcium contents as absorbance for pure cells (Cells), cells seeded on
Ti granules (Ti) and cells seeded on TiO2 granules (TiO2) after editing from the controls
without cells. Data are presented as mean ± SEM (n = 8 for each condition and time
point). p-level below 0.05 is considered as significant (*,#: p < 0.05; **,##: p < 0.01;
***,###: p < 0.001). Numerical values can be found in Supplementary Table.
Fig. 6. a) Relative total calcium contents as absorbance for pure cells (Cells) and an empty insert with medium (Empty). b) Relative total calcium contents as absorbance for cells
seeded on Ti granules (Ti+) and an insert with Ti granules and medium (Ti-). c) Relative total calcium contents for cells seeded on TiO2 granules (TiO2+) and an insert with TiO2
granules and medium (TiO2-). Data are presented as mean (n = 8 for each condition and time point). Numerical values can be found in Supplementary Table.
1 3 7 14 17 21 1 3 7 10 14 17 21
Days
1 3 7 10 14 17 21
Days
a b
84
6,0
5,0
4,0
3,0
2,0
1,0
0,0
7,0
2,5
2,0
1,5
1,0
0,5
0,0
3,0
2,0
1,5
1,0
0,5
0,0
2,5
Gen
e ex
pre
ssio
n A
LP
L
Gen
e ex
pre
ssio
n R
UN
X2
Gen
e ex
pre
ssio
n C
OL
1a1
**
**
***
***
**
**
**
**
*
**
**
**
Fig. 7. a) Relative ALPL gene expression for pure cells (Cells), cells seeded on Ti granules (Ti) and cells seeded on TiO2 granules (TiO2). Data are presented as mean ± SEM (n = 8 for
each condition and time point, except for cells at day 8, n = 7, and TiO2 at day 4, n = 7, and day 8, n = 5). b) Relative COL1α1 gene expression for pure cells (Cells), cells seeded on Ti
granules (Ti) and cells seeded on TiO2 granules (TiO2). Data are presented as mean ± SEM (n = 8 for each condition and time point). c) Relative RUNX2 gene expression for pure cells
(Cells), cells seeded on Ti granules (Ti) and cells seeded on TiO2 granules (TiO2). Data are presented as mean ± SEM (n = 8 for each condition and time point, except for cells at day 8,
n = 7). p-level below 0.05 is considered as significant (*: p < 0.05; **: p < 0.01; ***: p < 0.001). Numerical values can be found in Supplementary Table.
bone marrow, including almost unlimited access to stem cells, an
easier and safer procedure to obtain the cells, and with less risk of
infection. Seeding of osteogenic cells onto scaffolds is increasingly
being investigated as an alternative to autologous bone grafts for
bone reconstruction (Chen et al., 2005; Seitz et al., 2007). In con-
trast to autologous bone grafts, there are potentially less side effects
and there is theoretically no limitation to the amount of bone tissue
that can be generated. Traditionally, osteoblastic cells have been
derived from MSCs obtained from the bone marrow. However,
obtaining cells from the bone marrow requires bone marrow
aspiration that may result in a limited number of cells and thus
limited osteogenic potential.
We also tested the ability of the cells to produce mineralized
matrix on titanium scaffolds. Overall, cells were able to produce
mineralized matrix both on Ti and TiO2 granules. Interestingly, the
amount of matrix formed was significantly higher on both types of
granules than in the cultures of cells without carrier. This might be
due to better osteoinductive conditions when cells were cultured
on a three dimensional structure such as the granules compared to
the bottom of the well of the culture inserts. Another explanation
could be that the Alizarin Red assay was bound to the granules and
gave a false positive signal, when binding to the surface. Though
there might be some attachment of the colour to the surface of the
granules, the two carriers still induced bone matrix formation from
the cells, as the mineralized matrix formation was determined as
the difference between the culture wells with cells and carrier
versus the carrier alone. This method is in contrast to most other
studies investigating the osteoinductive properties of scaffolds and
carriers. Most studies published have not included controls with
the carrier without cells seeded upon them. Therefore, false posi-
tive measurements might have affected the results in the previ-
ously published studies (Dahl et al., submitted for publication).
Another interesting finding in the current study was that,
though not statistically significant for all values (Fig. 6) the nu-
merical values for mineralization were increased for all days for
TiO2 compared to Ti granules. This might be due to the increased
porosity of the TiO2 granules, thus giving the cells a larger surface
area and thereby promoting bone formation even more. Therefore
our hypothesis was fulfilled.
We chose to culture the cells on the inserts with the subsequent
risk of escape of cells through the net of the inserts instead of
culturing directly on the wells, as performed in similar other pub-
lished experiments (Hosseinkhani et al., 2005; Turhani et al., 2005;
Shih et al., 2006; Weissenböeck et al., 2006; Zhou et al., 2007;
Bjerre et al., 2008; Heo et al., 2009; Huang et al., 2009; Shi et al.,
2009; Stiehler et al., 2009). This makes change of the medium
Days Days
Days
1 4 8 1 4 8
1 4 8
c
a b
85
possible without disturbing the cells and was performed for stan-
dardized practical purposes for later transitional research trials.
Many similar studies have shown large mineralization of the
combined carrier and cells (Hosseinkhani et al., 2005; Turhani et al.,
2005; Shih et al., 2006; Weissenböeck et al., 2006; Zhou et al.,
2007; Bjerre et al., 2008; Heo et al., 2009; Huang et al., 2009;
Martins et al., 2009; Shi et al., 2009; Stiehler et al., 2009). As
mentioned above these publications did not reveal controls of the
carrier itself and hence without control of the pure cells them-
selves. This is in contrast to our study where the results reflect the
numerical values after the controls of pure cells have been sub-
tracted from values of the combined carrier and cells. Subsequently
the numbers presents very small values compared to the absorb-
ance of the combinations.
In the present study we emphasized control and publication of
mineralization and gene expression of studied cells with and
without seeding on a carrier. The purpose of this was to control and
publish the results of the behaviour of the cells in a pure form. This
is in contrast to what has been performed in most other studies. The
numerical value of an outcome of a carrier combined with mes-
enchymal stem cells comprises reactions from carrier as well as
from cells, respectively. Therefore measurements of each of these,
carrier and cells, are deemed necessary for evaluation.
5 . Conclusion
Adipose derived mesenchymal stem cells have many similarities
with BMSCs. They share some surface markers and similar differ-
entiating potential. The osteogenic potential in ADMSCs is made
possible within differentiation into osteoblasts in vitro. Besides
this, ADMSCs are an easily accessible rich resource of MSCs, there is
a likely alternative to other stem cell populations. Furthermore
adipose derived mesenchymal stem cells are shown to improve
neovascularization and skin regeneration by increasing neoangio-
genesis and collagen synthesis.
In our study, we demonstrated that adipose tissue derived
osteoblastic cells upon induction with specialized medium
expressed genes unique to the osteoblast lineage. Furthermore, they
were capable of producing mineralized matrix when cultured on
scaffolds of titanium granules. In addition, there was a marked dif-
ference in the formation of mineralized matrix when cells were
cultured on titanium granules as compared to when they were just
cultured in the culture dish insert. Moreover, the matrix formation in
the cell cultures on TiO2 was significantly higher than in the other
cultures, indicating that TiO2 granules are more osteoinductive than
Ti granules. Thus, TiO2 might be a promising carrier for the trans-
plantation of adipose tissue derived osteoblastic cells. However,
more studies are warranted to document the potential use in vivo.
Conflict of interest statement
All authors disclose any financial and personal relationships
with other people or organizations that could inappropriately in-
fluence our work.
Acknowledgements
This study was supported by the Danish National Research
Foundation (09-067289). Porous Titanium Granule material was
supported by Tigran Technologies AB exclusively.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jcms.2013.01.021.
References
Alffram PA, Bruce L, Bjursten LM, Urban RM, Andersson GB: Implantation of the
femoral stem into a bed of titanium granules using vibration: a pilot study of a new
method for prosthetic fixation in 5 patients followed for up to 15 years. Ups J Med Sci
112(2): 183–189, 2007
Behnia H, Khojasteh A, Soleimani M, Tehranchi A, Atashi A: Repair of alveolar cleft
defect with mesenchymal stem cells and platelet derived growth factors: a preliminary
report. J Craniomaxillofac Surg 40(1): 2–7, 2012
Ben-David D, Kizhner T, Livne E, Srouji S: A tissue-like construct of human bone
marrow MSCs composite scaffold support in vivo ectopic bone formation. J Tissue Eng
Regen Med 4(1): 30–37, 2010
Bjerre L, Bünger C, Kassem M, Mygind T: Flow perfusion culture of human mes-
enchymal stem cells on silicate-substituted tricalcium phosphate scaffolds.
Biomaterials 29(17): 2616–2627, 2008
Bonewald LF, Harris SE, Rosser J, Dallas MR, Dallas SL, Camacho NP, et al: Von Kossa
staining alone is not sufficient to confirm that mineralization in vitro represents
bone formation. Calcif Tissue Int 72: 537–547, 2003
Bystedt H, Rasmusson L: Porous titanium granules used as osteoconductive material
for sinus floor augmentation: a clinical pilot study. Clin Implant Dent Relat Res
11(2): 101–105, 2009
Cao Y, Sun Z, Liao L, Meng Y, Han Q, Zhao RC: Human adipose tissue-derived stem cells
differentiate into endothelial cells in vitro and improve postnatal neovascularization in
vivo. Biochem Biophys Res Commun 332(2): 370–379, 2005
Chen J, Sotome S, Wang J, Orii H, Uemura T, Shinomiya K: Correlation of in vivo bone
formation capability and in vitro differentiation of human bone marrow stromal cells. J
Med Dent Sci 52(1): 27–34, 2005
Dahl M, Jørgensen NR, Hørberg M, Pinholt EM: Carriers in mesenchymal stem cell
osteoblast mineralization – state-of-the-art. J Craniomaxillofac Surg, accepted for
publication
Goldberg VM, Stevenson S: Natural history of autografts and allografts. Clin Orthop
Relat Res 225: 7–16, 1987
Heo SJ, Kim SE, Wei J, Kim DH, Hyun YT, Yun HS, et al: In vitro and animal study of
novel nano-hydroxyapatite/poly(epsilon-caprolactone) composite scaffolds fabricated
by layer manufacturing process. Tissue Eng 15(5): 977–989, 2009
Hong J, Andersson J, Ekdahl KN, Elgue E, Axén N, Larsson R, et al: Titanium is a
highly thrombogenic biomaterial: possible implications for osteogenesis. Thromb
Haemost 82(1): 58–64, 1999
Hosseinkhani H, Inatsugu Y, Hiraoka Y, Inoue S, Tabata Y: Perfusion culture en-
hances osteogenic differentiation of rat mesenchymal stem cells in collagen sponge
reinforced with poly (glycolic acid) fiber. Tissue Eng 11(9–10): 1476–1488, 2005
Huang Y, Jin X, Zhang X, Sun H, Tu J, Tang T, et al: In vitro and in vivo evaluation of
akermanite bioceramics for bone regeneration. Biomaterials 30(28): 5041–5048,
2009
Leong NL, Jiang J, Lu HH: Polymer-ceramic composite scaffold induces osteogenic
differentiation of human mesenchymal stem cells. Conf Proc IEEE Eng Med Biol Soc
1: 2651–2654, 2006
Martins AM, Pham QP, Malafaya PB, Raphael RM, Kasper FK, Reis RL, et al: Natural
stimulus responsive scaffolds/cells for bone tissue engineering: influence of
lysozyme upon scaffold degradation and osteogenic differentiation of cultured
marrow stromal cells induced by CaP coatings. Tissue Eng Part A 15(8): 1953–
1963, 2009
Masuda T, Salvi GE, Offenbacher S, Felton DA, Cooper LF: Cell and matrix reactions at
titanium implants in surgically prepared rat tibiae. Int J Oral Maxillofac Implants
12(4): 472–485, 1997
Meruane M, Rojas M, Marcelain K, Chile S: The use of adipose tissue-derived stem
cells within a dermal substitute improves skin regeneration by increasing
neoangiogenesis and collagen synthesis. Plast Reconstr Surg 130(1): 53–63,
2012
Metzler P, von Wilmowsky C, Zimmermann R, Wiltfang J, Schlegel KA: The effect of
current used bone substitution materials and platelet-rich plasma on periosteal cells by
ectopic site implantation: an in-vivo pilot study. J Craniomaxillofac Surg 40(5): 409–
415, 2012
Puchtler H, Meloan SN, Terry MS: On the history and mechanism of alizarin and
alizarin red S stains for calcium. J Histochem Cytochem 17(2): 110–124, 1969
Seitz S, Ern K, Lamper G, Docheva D, Drosse I, Milz S, et al: Influence of in vitro
cultivation on the integration of cell matrix constructs after subcutaneous im-
plantation. Tissue Eng 13(5): 1059–1067, 2007
Shi X, Wang Y, Ren L, Huang W, Wang DA: A protein/antibiotic releasing poly(lactic-co-
glycolic acid)/lecithin scaffold for bone repair applications. Int J Pharm 373(1–2): 85–92,
2009
Shih YR, Chen CN, Tsai SW, Wang YJ, Lee OK: Growth of mesenchymal stem cells on
electrospun type I collagen nanofibers. Stem Cells 24(11): 2391–2397, 2006
Stiehler M, Bünger C, Baatrup A, Lind M, Kassem M, Mygind T: Effect of dynamic 3-D
culture on proliferation, distribution, and osteogenic differentiation of human
mesenchymal stem cells. J Biomed Mater Res 89(1): 96–107, 2009
Stockmann P, Park J, von Wilmowsky C, Nkenke E, Felszeghy E, Dehner JF, et al:
Guided bone regeneration in pig calvarial bone defects using autologous mes-
enchymal stem/progenitor cells – a comparison of different tissue sources. J
Craniomaxillofac Surg 40(4): 310–320, 2012
Sul YT, Johansson C, Wennerberg A, Cho LR, Chang BS, Albrektsson T: Optimum
surface properties of oxidized implants for reinforcement of osseointegration: surface
86
chemistry, oxide thickness, porosity, roughness, and crystal structure. Int J Oral
Maxillofac Implants 20(3): 349–359, 2005.
Sutter W, Stein E, Koehn J, Schmidl C, Lezaic V, Ewers R, et al: Effect of
different biomaterials on the expression pattern of the transcription factor
Ets2 in bone–like constructs. J Craniomaxillofac Surg 37(5): 263–271, 2009
Informative letter. Malmö, Sweden: Tigran Technologies AB, 2010
Turhani D, Watzinger E, Weissenböeck M, Yerit K, Cvikl B, Thurnher D, et al:
Three-dimensional composites manufactured with human mesenchymal
cambial layer precursor cells as an alternative for sinus floor augmentation:
an in vitro study. Clin Oral Implants Res 16(4): 417–424, 2005
Turner TM, Urban RM, Hall DJ, Andersson GB: Bone ingrowth through porous
titanium granulate around a femoral stem. Ups J Med Sci 112(2): 191–197,
2007
Wang YH, Liu Y, Maye P, Rowe DW: Examination of mineralized nodule formation
in living osteoblastic cultures using fluorescent dyes. Biotechnol Prog 22:
1697–1701, 2006
Weissenböeck M, Stein E, Undt G, Ewers R, Lauer G, Turhani D: Particle size
of hydroxyapatite granules calcified from red algae affects the osteogenic
potential of human mesenchymal stem cells in vitro. Cells Tissues Organs
182(2): 79–88, 2006
Younger EM, Chapman MW: Morbidity at bone graft donor sites. J Orthop
Trauma 3(3): 192–195, 1989
Zhou GS, Su Zy, Cai YR, Liu YK, Dai LC, Tang RK, et al: Different effects of
nanophase and conventional hydroxyapatite thin films on attachment,
proliferation and osteogenic differentiation of bone marrow derived
mesenchymal stem cells. Biomed Mater Eng 17(6): 387–395, 2007
Zuk PA, Zhu M, Ashjian P, De Urgate DA, Huang JI, Mizuno H, et al: Human
adipose tissue is a source of multipotent stem cells. Mol Biol Cell 13(12): 4279–
4295, 2002
Zuk PA, Zhu M, Mizuno H, Huang J, Futrell JW, Katz AJ, et al: Multilineage cells
from human adipose tissue: implications for cell-based therapies. Tissue Eng
7(2): 211–228, 2001
87
13. Manuscript III
Adipose derived mesenchymal stem cells seeded on oxidized titanium
granules increases osteoblast-covered surface in vivo compared with
titanium granules
Morten Dahl DDS,* Ellen-Margrethe Hauge MD, PhD,§ Peter Schwarz MD, DMSci,# Niklas Rye
Jørgensen MD, PhD, DMSci,‡ and Else Marie Pinholt DDS, MSci, dr. odont║
* PhD student, Department of Endocrinology, Clinical Biochemistry and Research Centre for Ageing and Osteoporosis Rigshospitalet, Copenhagen,
Denmark, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark.
§ Professor, Department of Rheumatology, Aarhus University Hospital, Aarhus, Denmark, Department of Clinical Medicine, Aarhus University,
Aarhus, Denmark.
# Professor, Department of Endocrinology and Research Centre for Ageing and Osteoporosis Rigshospitalet, Copenhagen, Denmark.
‡ Professor, Department of Clinical Biochemistry Rigshospitalet, Copenhagen, Denmark and Institute of Clinical Research, University of Southern
Denmark, Odense, Denmark.
║ Professor, University of Southern Denmark, Institute of Regional Health Research, Odense, Denmark.
Address correspondence and reprint requests to Morten Dahl: Department of Endocrinology, Clinical Biochemistry and Research Centre for Ageing and Osteoporosis Rigshospitalet, Copenhagen,
Denmark, Faculty of Health Sciences, University of Copenhagen, Nordre Ringvej 57, 2600 Glostrup, Denmark; telephone: +45 29254048, email: [email protected]
ABSTRACT
Purpose: Adipose derived mesenchymal stem cells
(ADMSCs) express osteoblastic lineage genes and
produce mineralized matrix. Titanium granules are
suitable carriers. The purpose was to evaluate the early
in vivo effect of ADMSCs seeded on titanium granules.
Materials and methods: ADMSCs were seeded on Ti
or TiO2 granules or without granules and implanted
included controls subcutaneously in twenty-four
Balb/cJ mice. After two weeks, pocket contents were
collected for analysis of bone cells-covered granule
surface (GS) and stained with toluidine blue and basic
fuchsin for quantitative histology.
Results: A higher osteoblast-covered surface ratio
(Ob.S/GS) (p=0.002) and osteoclast-covered surface
ratio (Oc.S/GS) (p=0.036) was found between TiO2 vs.
Ti granules with ADMSCs. A higher Ob.S/GS
(p=0.002) and Oc.S/GS (p=0.003) was found between
TiO2 vs. Ti granules without ADMSCs. A higher
number of osteoblasts covering GS (N.Ob/GS) was
found between TiO2 vs. Ti granules with ADMSCs
(p=0.001) and without ADMSCs (p=0.003). A higher
number of osteoclasts covering GS (N.Oc/GS) was
found between TiO2 vs. Ti granules with ADMSCs
(p=0.023) and without ADMSCs (p=0.009).
Conclusion: A higher osteoblast-covered surface in
TiO2 compared with Ti granules suggests that TiO2 is a
more osteoinductive carrier. Osteogenic ADMSCs
seeded on TiO2 granules may have potential in bone
reconstruction.
Key words: Adipose mesenchymal stem cells, mice,
drug carriers, tissue engineering, osteoblasts.
88
1. INTRODUCTION
Cranio-maxillofacial surgery is essential in
traumatic head injuries, guided bone regeneration in
implant dentistry and reconstructive surgery. Most
often, the surgery involves bone reconstruction
comprising autologous bone grafting (Marx, 2005;
Montespan et al., 2014), but an increasing body of
evidence has shown that synthetic and natural
biomaterials combined with osteogenic cells offer
advantages compared to traditional bone reconstruction
techniques including autologous bone grafting.
Autologous bone grafting implies the transfer of
osteoprogenitor cells or osteoblasts and in this way
works osteoconductive, osteogenic and osteoinductive
(Goldberg et Stevenson, 1987, Behnia et al., 2012).
The combination of synthetic and natural biomaterials
with osteogenic cells has the potential to promote
osteogenesis (Chen et al., 2005, Seitz et al., 2007,
Nino-Fong et al., 2013, Schubert et al., 2013, Sunay et
al., 2013, Kim et al., 2014, Ma et al., 2014, Mirsaidi et
al., 2014, Chen et al., 2015, Dosier et al., 2015, Wu et
al., 2015). Osseous tissue engineering involves the in
vitro seeding of cells onto carriers supporting cell
adhesion, migration, proliferation and differentiation,
and defines the 3D shape of the tissue to be engineered
(Leong et al., 2006, Ben-David et al., 2010). It is
important to find a balance between mechanical needs
of the tissue to be replaced and carrier porosity, which
allows tissue growth (Ilizarov, 1989, Salgado et al.,
2004, Hutmacher et al., 2007, Martins et al., 2009).
The optimal carrier should be biocompatible,
biodegradable and osteoconductive to generate new
bone formation (Leong et al., 2006, Ben-David et al.,
2010). The engineered cell-containing bone graft
should exhibit the following features: the presence of
osteogenic cells facilitating immediate formation of
new bone; an appropriate extracellular matrix to
provide an osteoconductive carrier; osteoinductive
growth factors to provide signals to the resident cells;
and an adequate blood supply to support cell growth
and function (Goldberg, 2000, Lund et al., 2008,
Tiainen et al., 2010).
In adipose tissue, a cell population similar to the
population of mesenchymal stem cells from bone
marrow-derived stroma cells (BMSCs) was identified,
called adipose-derived mesenchymal stem cells
(ADMSCs), (Zuk et al., 2001, Zuk et al., 2002). The
osteogenic potential of ADMSCs is supported by in
vitro and in vivo studies (Zuk et al., 2001, Dahl et al.,
2013, Schubert et al., 2013, Sunay et al., 2013, Yue et
al., 2013, Lu et al., 2014, Ma et al., 2014, Mirsaidi et
al., 2014, Chen et al., 2015). In vitro, these cells can
differentiate into cell types found in bone, cartilage, fat,
and muscle tissue (Zuk et al., 2001). We have
previously shown that ADMSCs stained positive for
alkaline phosphatase (ALP), revealed high expression
of genes characteristic of the osteoblastic lineage
(Cbfa-1) and only low expression of the adipocyte-
related gene PPAR-γ1 (Dahl et al., 2013). Human
ADMSCs can differentiate into endothelial cells in
vitro and improve postnatal neovascularization in vivo.
(Cao et al., 2005). Furthermore, a recent study showed
increased osteogenesis and angiogenesis in a co-culture
of ADMSCs and BMSCs both in vitro and in vivo
(Kim et al., 2014). Adipose tissue is an easily
accessible rich resource of osteogenic MSCs. ADMSCs
are able to expand to clinical scales, still having the
potential to undergo osteogenic differentiation
independently of donor age and bone quality. Finally,
there is a relevant lower morbidity when harvesting
ADMSCs compared to BMSCs (Sunay et al., 2013, Lu
et al., 2014, Mirsaidi et al., 2014). Therefore, ADMSCs
may be a more promising osteogenic resource in the
future.
Many different carriers have been studied
within gene expression and mineralization including
type I collagen, titanium, calcium carbonates, calcium
phosphates, hydroxyapatite, tricalcium phosphate,
89
calcium-deficient hydroxyapatite, demineralized bone
matrix, and polylactic acid-polyglycolic acid
copolymer with modifications. In our previous
literature study, we found only two studies (Kasten et
al., 2005, Kasten et al., 2006) including control
measurements of the carrier without osteogenic cells
(Dahl et al., 2014). The search for the best carrier
material for bone reconstruction is still going on.
However, titanium granules have received some
attention, and as a carrier they may provide mechanical
stability for initiation of bone development and
ingrowth by interlocking the granules with each other,
creating an uninterrupted structure (Masuda et al.,
1997, Hong et al., 1999, Alffram et al., 2007, Turner et
al., 2007, Bystedt et al., 2009). We have previously
shown that ADMSCs seeded on irregular titanium
granules in vitro differentiate into osteoblasts and
produce mineralized matrix (Dahl et al., 2013). There
was a significantly higher matrix formation on the
heated oxidized titanium granules (TiO2) when
compared to the non-heated non-oxidized (Ti) analogue
(Dahl et al., 2013).
The aim of the study was therefore to
investigate the early in vivo effect of ADMSCs seeded
on Ti and TiO2 granules implanted subcutaneously in
mice. The primary outcome was the occurrence of bone
cells on the granule surface. Our hypothesis is that
ADMSCs seeded on porous Ti and TiO2 granules will
result in a significantly increased osteoblast-covered
surface with consequent more osteoid formation on
TiO2 granules in vivo.
2. MATERIALS AND METHODS
2.1. Carriers
Unalloyed, irregular shaped commercially pure
titanium particles, diameter 0.7-1.0 mm, 80% porosity,
in a Ti and a TiO2 form were used as carrier material
(Natix®) (Tigran Technologies AB, Malmö, Sweden)
(Dahl et al., 2013).
2.2. Adipose derived mesenchymal stem cells
The current study was approved by the Danish
Animal Experiments Inspectorate, 2010/561-184L.
ADMSCs were obtained from eight, eight-week-old
Balb/cJ mice. They were euthanized and abdominal
adipose tissue was collected, divided into small pieces
and washed in Dulbecco’s Phosphate Saline w/o Ca2+
and Mg2+ (PBS) (Lonza, Verviers, Belgium). The
tissue was then collagenase-treated (7.5 mg collagenase
in 10 mL PBS) (SIGMA-ALDRICH, St. Louis, USA).
After inactivation of the collagenase, ADMSCs were
isolated and cultured in a mouse osteoblast medium
(mOB MEM: 435 mL MEM Earle’s w/o Phenol red
(MEM) (Invitrogen, Auckland, New Zealand), 50 mL
Fetal Calf Serum heat inactivated (SIGMA-
ALDRICH), 5 mL Penicillin/Streptomycin
(Invitrogen), 5 mL glutamax (Invitrogen), 500 µL L-
Ascorbic Acid Phosphate Magnesium Salt n-Hydrate
(Wako Chemicals, Virginia, USA) (50 mg/mL) and 5
mL glycerol 2-phosphate disodium salt hydrate
(SIGMA-ALDRICH) (1 M)) according to a previously
published protocol (Dahl et al., 2013). The cells were
maintained in a humidified atmosphere at 37°C and 5%
CO2. After two weeks, most of the cells were
differentiated into osteoblasts assessed by visual
inspection in a light microscope and based on previous
characterization of the cells (Dahl et al., 2013). The
cells were then seeded on either porous Ti and TiO2
granules or without granules for one week before they
were ready for transplantation into the animals (Figure
1). Osteoblast phenotype was confirmed in Alizarin
Red Assay before implantation and results were similar
to our in vitro study (Dahl et al., 2013).
90
Figure 1: Time line for the early in vivo effect of adipose derived mesenchymal stem cells implanted subcutaneously in mice. Osteoblast phenotype
was confirmed on day -4 and -1 before implantation with Alizarin Red S (AR-S) assay.
2.3. Study design
Twenty-four 12-week-old Balb/cJ mice (Figure
2a) were anesthetized using inhaled Isofluran (1-2%)
(Baxter A/S, Allerød, Denmark). Each mouse was
prepared for surgical intervention with application of
ophthalmic ointment (Neutral Ophtha) (Ophtha A/S,
Gentofte, Denmark), subcutaneously prophylactic
antibiotic 0.05 mL Norostrept vet (200000 ie/mL
benzylpenicillinprocain, 200 mg/mL
dihydrostreptomycin) (Scanvet, Fredensborg,
Denmark), shaving and disinfection with 2.5% iodine
alcohol. Four subcutaneous pockets on the back of each
mouse were created through skin incisions. The
subcutaneous pockets were filled in random order with
1) mOB MEM without ADMSCs, 2) mOB MEM with
ADMSCs, 3) titanium granules without pre-seeded
ADMSCs (either Ti- (12 mice) or TiO2- (12 mice)),
and 4) titanium granules with pre-seeded ADMSCs
(either Ti+ (12 mice) or TiO2+ (12 mice)), and closed
with Prolene 4.0 (Figure 2b). All mice received
postoperatively analgesic 0.04 mL Rimadyl vet (50
mg/mL carprofen) (Pfizer Aps, Ballerup, Denmark)
and 1 mL sterile saline (sodium chloride isotonic SAD
9 mg/mL) (Amgros I/S, Copenhagen, Denmark)
subcutaneously. All mice were kept single caged for
four hours at temperature 26°C hereafter at temperature
22°C. All mice had a natural ingredient diet ad libitum
and free access to water.
Figure 2: a) Study design: Twenty-four mice allocated in two groups. Four subcutaneous pockets in each mouse were filled in random order with 1)
mouse osteoblast medium (mOB MEM) without adipose derived mesenchymal stem cells (ADMSCs), 2) mOB MEM with ADMSCs, 3) titanium
granules without pre-seeded ADMSCs (either Ti- (12 mice) or TiO2- (12 mice)), and 4) titanium granules with pre-seeded ADMSCs (either Ti+ (12
mice) or TiO2+ (12 mice)). b) Surgical intervention: TiO2+ granules inserted in one of the subcutaneous pockets.
91
After two weeks of observation all mice were
euthanized by cervical dislocation. All four samples
from each animal were sampled including surrounding
musculature and fixated in absolute ethanol 20:1.
Preparation for non-decalcified specimens were
performed at Sahlgrensska Institute, Department for
Biomaterials, University of Gothenburg by sequential
dehydration and embedded in methylmetacrylate
(Erben, 1997). Approximately 5-9 decalcified sections,
250 µm thick, perpendicular to the normal anatomy
were produced from each embedded specimen by
means of a Leiden saw (KDG-95, Meprotech,
Heerhugowaard, The Netherlands), and for hereafter
grinding and polishing to a final thickness of 30-60 µm
(Donath, 1993, Luzi et al., 2009). All sections were
surface stained with basic fuchsin 0.3% and toluidine
blue 1%.
2.4. Quantitative histology
All sections were evaluated using a light
microscope (Nikon Eclipse 80i, Nikon, Tokyo, Japan)
equipped with a digital camera (Olympus DP72,
Olympus, Tokyo, Japan) connected to a PC running
Visiopharm Integrator System – NewCAST version
5.3.0 interactive stereology software system
(Visiopharm, Hørsholm, Denmark), magnification
×464. The following in-plane parameters were
determined: Granule surface (GS) to granule volume
(GV) ratio (GS/GV), the fraction of granule surface
covered with fibrosis (Fb.S/GS), osteoclasts (Oc.S/GS),
osteoblasts (Ob.S/GS) or not evaluable surface
(NE.S/GS), and the profile number per granule surface
of osteoblasts (N.Ob/GS) and osteoclasts (N.Oc/GS)
(Dempster et al., 2013).
Region of interest (ROI) was defined as the
tissue surrounding the carrier material (Figure 3a). A
meander sampling was performed (fraction 100% with
random orientation). The aim was to achieve around
200 points or intersections per specimen for the
parameter being studied. Area per point was 5936 µm2
and area per length was 396.5 µm were superimposed
to the histological section in the microscope. Point-grid
was used for GV estimation (all black/dark areas were
defined as carrier material), while line-grid was used
for GS estimation (total intersections with the granule
surface). An estimation of the in-plane granule surface
density (GS/GV) was calculated as 2×GS/(GV×dgrid),
where dgrid is the distance between the lines (74.85 µm)
(Vesterby et al., 1987). All granule surface
intersections were defined in one of four categories
(Figure 3b): Fibrosis (Fb.S), osteoclasts (Oc.S),
osteoblasts (Ob.S) or not evaluable (NE.S) (Figure 3b).
Fibrosis was defined as collagen with small spindle
shaped fibroblasts (Figure 3d), while large
multinucleated cells were defined as osteoclasts
(Figure 3d), and finally groups of connected mono-
nucleated cuboidal cells were defined as osteoblasts
(Figure 3c). NE was defined as blurred areas either
because of excessive staining or dusty areas because of
grinding and polishing the granules.
92
Figure 3: Cross-section of specimen containing porous oxidized titanium granules with pre-seeded adipose derived mesenchymal stem cells (TiO2+),
stained with basic fuchsin 0.3% and toluidine blue 1%. a) Region of interest is defined as the tissue surrounding the titanium granules. b) Intersections
defined as fibrosis (Fb.S), osteoclasts (Oc.S), osteoblasts (Ob.S), and not evaluable (NE.S). Magnification ×464. c) Osteoblast-covered (Ob) TiO2+
surface. d) An osteoclast (Oc) and fibrosis-covered (Fb) TiO2+ surface.
All measurements were carried out by the same
evaluator (MD), and group allocation was blinded to
the evaluator. The coefficient of variation (CV) was
CVGV=4.0%, CVFb.S=7.5%, which was fully
acceptable. CVOc.S=55.6%, CVOb.S=73.1%, and
CVNE.S=27.7% was relatively high, why profile number
of osteoblasts and osteoclasts on the granule surface
were analyzed (Sterio, 1984). Point-grid and line-grid
was found relatively acceptable due to the few and
scattered bone cells in the sections.
2.5. Statistics:
Data are provided as median and interquartile
range, as data is not normally distributed. Comparisons
between groups were performed using Kruskal-Wallis
one-way analysis of variance for non-parametric data
to test whether the medians of the four groups were
equal. If we found significant differences between the
groups a post hoc Mann-Whitney U Test was
performed for pairwise comparison of groups.
Significance levels of 0.05 were used
throughout the study. All statistical tests were
performed using IBM SPSS Statistics 22.
93
3. RESULTS
In sections from the samples where only mOB
MEM with and without ADMSCs were implanted we
did not find any bone cells or site effects from the
surgical intervention. Statistical tests therefore only
include the four groups Ti-, TiO2-, Ti+, and TiO2+.
3.1. Surface and volume estimation of the carrier
In order to determine the volume of porous Ti
and TiO2 granules implanted subcutaneously, GS/GV
was estimated from histomorphometric analysis
meander sampling (Table 1). There was no statistically
significant difference in GS/GV across the four groups
(p=0.330), and approximately the same amount of
granule material was placed in each subcutaneous
pocket was observed.
Without ADMSCs With ADMSCs
Ti- n=12
TiO2- n=12
P-value Ti- vs. TiO2-
Ti+ n=12
TiO2+ n=11
P-value Ti+ vs. TiO2+
P-value Across all groups#
GS/GV 0.021 [0.020-0.237]
0.019 [0.017-0.022]
--- 0.018 [0.015-0.021]
0.018 [0.015-0.019]
--- 0.330
Ob.S/GS 0.000 [0.000-0.000]
0.0021 [0.0002-0.0056]
0.002 0.000 [0.000-0.000]
0.0028 [0.000-0.0055]
0.002 <0.001
Oc.S/GS 0.071 [0.061-0.089]
0.124 [0.084-0.155]
0.003 0.079 [0.053-0.094]
0.090 [0.084-0.144]
0.036 0.006
Fb.S/GS 0.91 [0.88-0.92]
0.85 [0.83-0.87]
0.001 0.88 [0.86-0.91]
0.83 [0.80-0.89]
0.042 0.002
NE.S/GS 0.015 [0.003-0.037]
0.012 [0.002-0.051]
--- 0.045 [0.006-0.060]
0.0239 [0.009-0.050]
--- 0.495
N.Ob/GS 0.000 [0.000-0.000]
0.013 [0.002-0.038]
0.003 0.000 [0.000-0.000]
0.022 [0.004-0.072]
0.001 <0.001
N.Oc/GS 0.131 [0.107-0.159]
0.258 [0.144-0.325]
0.009 0.152 [0.086-0.165]
0.182 [0.165-0.277]
0.023 0.004
Table 1: Granule surface to volume ratio (GS/GV), fractional surface ratio within osteoblasts (Ob.S/GS), osteoclasts (Oc.S/GS), fibrosis (Fb.S/GS),
and not evaluable (NE.S/GS) for porous titanium granules with (Ti+ and TiO2+) and without (Ti- and TiO2-) adipose derived mesenchymal stem cells
(ADMSCs). Number of osteoblasts and osteoclasts per granule surface (N.Ob/GS, N.Oc/GS) for porous Ti+, TiO2+, Ti-, and TiO2-. No significant
difference could be found between TiO2- vs. TiO2+ or Ti- vs. Ti+. # Kruskal-Wallis one-way analysis of variance.
3.2. Fractional surfaces
Next, we wanted to determine whether the
titanium surface as well as the pre-seeding of ADMSCs
on the granules affected the fraction of surfaces
covered by osteoblasts. A post hoc Mann-Whitney U
Test showed that Ob.S/GS were significantly increased
in TiO2+ compared with Ti+ (p=0.002) and in TiO2-
compared with Ti- (p=0.002). In contrast, no
significant difference could be found between TiO2- vs.
TiO2+ and Ti- vs. Ti+ within Ob.S/GS (p=0.877 and
p=0.952, respectively) (Figure 4, Table 1).
Additionally, we investigated the fraction of
surfaces covered by osteoclasts. A post hoc Mann-
Whitney U Test showed that Oc.S/GS were
significantly increased in TiO2+ compared with Ti+
(p=0.036). At the same time Oc.S/GS were
significantly increased in TiO2- compared with Ti-
(p=0.003). In contrast, no significant difference could
be found between TiO2- vs. TiO2+ and Ti- vs. Ti+
within Oc.S/GS (p=0.667 and p=0.603, respectively)
(Figure 4, Table 1).
Finally, we wanted to determine whether the
titanium surface as well as the pre-seeding of ADMSCs
on the granules affected the fraction of surfaces
covered by fibrosis. A post hoc Mann-Whitney U Test
showed Fb.S/GS was significantly higher in Ti+
compared with TiO2+ (p=0.042) and in Ti- compared
with TiO2- (p=0.001). In contrast, there was no
significant difference between TiO2- compared with
94
TiO2+ (p=0.758) and between Ti- compared with Ti+
(p=0.094) (Figure 4, Table 1).
3.3. Number of osteoblast and osteoclast profiles
In order to determine whether the titanium
surface as well as pre-seeding of ADMSCs on the
granules affected the number of osteoblasts covering
the surface, a post hoc Mann-Whitney U Test showed a
statistically significantly higher appearance of
N.Ob/GS in TiO2+ compared with Ti+ (p=0.001) and
in TiO2- compared with Ti- (p=0.003). In contrast no
statistically significantly difference in appearance of
N.Ob/GS was found in TiO2- compared with TiO2+
(p=0.665) or in Ti- compared with Ti+ (p=0.616)
(Figure 5, Table 1).
Additionally, we wanted to determine whether
the titanium surface as well as pre-seeding of ADMSCs
on the granules affected the number of osteoclasts
covering the surface. A post hoc Mann-Whitney U Test
showed a statistically significant higher appearance of
N.Oc/GS in TiO2+ compared with Ti+ (p=0.023) and
in TiO2- compared with Ti- (p=0.009). In contrast no
statistically significantly difference in appearance of
N.Oc/GS was found in TiO2- compared with TiO2+
(p=0.356) or in Ti- compared with Ti+ (p=0.686)
(Figure 5, Table 1).
Figure 4: Fractional surface ratio of a) osteoblasts (Ob.S/GS), b) osteoclasts (Oc.S/GS), and c) fibrosis (Fb.S/GS) for porous titanium granules with
(Ti+ and TiO2+) and without (Ti- and TiO2-) adipose derived mesenchymal stem cells.
95
Figure 5: Number of a) osteoblasts per granule surface (N.Ob/GS) and b) osteoclasts per granule surface (N.Oc/GS) for porous titanium granules
with (Ti+ and TiO2+) and without (Ti- and TiO2-) adipose derived mesenchymal stem cells.
4. DISCUSSION
In this study, we investigated the early in vivo
effect of ADMSCs seeded on porous Ti and TiO2
granules as carrier material and implanted
subcutaneously in mice. The main finding was a
significantly higher fractional surface covered with
bone cells (osteoblasts and osteoclasts) for porous TiO2
compared to Ti granules both with and without seeding
of ADMSCs. At the same time a significantly higher
appearance of N.Ob/GS on porous TiO2 granules was
detected compared to Ti granules. In line with our
hypothesis, these results suggest that porous TiO2
granules are more osteoinductive as a carrier.
Porous titanium granules have been used for
augmentation of the maxillary sinus in oral implant
surgery to stabilize implants and act as an
osteoconductive matrix with promising results
(Lyngstadaas et al., 2015).
When comparing proliferation activity of human
mesenchymal stem cells on porous Ti and TiO2
granules the rate was similar on day 1 whereas it was
significantly higher in TiO2 granules on day 3
(Sabetrasekh et al., 2011). When analyzing the
proliferation rate normalized to surface area of the
granules, it was significantly higher on day 1 for
porous Ti granules and at day 3 there was no
significant difference between Ti and TiO2 granules
(Sabetrasekh et al., 2011). These results may be
explained by porous TiO2 granules having larger pores
and consequently higher surface-to-volume ratio than
porous Ti granules (Sabetrasekh et al., 2011).
In our study, we found more osteoblasts
covering the surface of the TiO2 granules compared to
Ti granules. These findings indicate that the oxidation
of the granules enhances the seeding efficiency of
osteoblasts on the granules. In addition, the oxidized
titanium may result in a better recruitment of
osteoblasts from the bloodstream.
Different studies have showed osteogenic
potential of ADMSCs differentiated into osteoblasts in
combination with different carriers. Supronowicz et al.
discovered increased bone matrix formation in vitro
and in vivo when combining ADMSCs with
demineralized bone matrix (Supronowicz et al., 2011).
Another study found improved osteogenic regeneration
in rabbits when using osteoblast-differentiated
ADMSCs compared with controls and undifferentiated
ADMSCs (Sunay et al., 2013).
96
Marini et al. compared ADMSCs seeded on
either nanostructured titanium or polystyrene and
concluded that although not superior to polystyrene,
nanostructured titanium still functioned as a basis for
ADMSCs to differentiate into osteoblasts and produce
bone matrix (Marini et al., 2015). However, this study
did not include control measurements of the
nanostructure titanium carrier without the ADMSCs to
evaluate the actual effect of adding the stem cells.
In our study, we did control measurements of
the carriers without adding the ADMSCs. Surprisingly
we did not find any effect of adding the ADMSCs,
neither with the Ti or the TiO2 granules. A possible
explanation is that the overall amount of bone cells on
the carrier material was rather low. Another reason
may be that the carrier itself plays the most important
role in recruiting and maintaining the osteoblasts on the
surface and therefore the addition of ADMSCs does
not contribute significant changes.
We anticipated finding osteoid formation within
two weeks of ectopic in vivo implantation because of
the high metabolic index in mice (Coulson, 1983).
When Supronowicz et al. found bone matrix formation
within 2 weeks, they used demineralized bone matrix
with ADMSCs implanted intramuscular in rats
(Supronowicz et al., 2011). This setup may offer more
osteoinductive factors both from the increased
vascularization in muscle and from the demineralized
bone matrix used as carrier, explaining why they were
able to show bone formation. In our animal model,
prolonging the in vivo implantation to 8 weeks might
have revealed mineralized matrix formation.
In summary, ADMSCs shows great potential for
differentiating into osteoblasts for use in bone
regeneration with different carriers. In addition,
ADMSCs are rather accessible, safe to harvest,
obtainable in large numbers, and with a high capacity
for proliferation (Sunay et al., 2013, Lu et al., 2014,
Mirsaidi et al., 2014).
Strength of this study is the systematic use of
control measurements both with osteoblast medium
alone and with ADMSCs combined with osteoblast
medium. In this way we are able to differentiate
between the effect of the carrier and the ADMSCs.
Another strength is the blinded evaluation of the cells
on the surface of the carriers.
Limitation of the study is that the volume of
carrier material with ADMSCs was only approximately
the same in each implanted specimen. This increases
the possible variation between the samples. To create
an environment with more osteoinductive factors, a
bone defect model would have been advantageous to
our subcutaneous model.
Another limitation is the sampling bias present
when we count the osteoblast and osteoclast profiles on
the carrier surface. Solely because osteoclasts are
considerably larger cells than osteoblasts the number of
osteoclast profiles will tend to be overestimated
(Sterio, 1984). Since we are not comparing the number
of osteoblasts and osteoclasts this does not have any
impact on our results.
Finally, we have a relatively high coefficient of
variation, and intersections with osteoblasts and
osteoclasts are considered as a rare event. However,
further increasing of test-lines will not change this
substantially.
Our findings emphasize how important it is to
measure the carrier with and without cells to have a
control measurement.
5. CONCLUSION
Our study demonstrates a significantly higher
fractional surface of bone cells (osteoblasts and
osteoclasts) compared to granule material on porous
TiO2 granules compared to Ti granules when
transplanted both with and without ADMSCs. This
suggests that porous TiO2 granules act as a more
osteoinductive carrier than Ti granules. At the same
time TiO2 granules keeps the mechanical stability for
97
ensuring regeneration of bone. This was further
supported by the higher number of osteoblasts on
porous TiO2 granules per granule material compared to
Ti granules.
Thus, based on our findings, TiO2 granules may
be a promising carrier for the transplantation of
ADMSCs. However, more studies are warranted to
document the potential use in vivo for a longer period.
6. CONFLICT OF INTEREST
Tigran Technologies AB, Malmø, Sweden
provided Porous Titanium Granule material free of
charge and a grant on DKr. 25,000 for using
synchrotron µCT radiation evaluation. All authors
disclose that they have no relevant conflict of interest
in relation to the study.
7. ACKNOWLEDGEMENT
This study was supported by the Danish
National Research Foundation (09-067289). PhD
salary was obtained from the UCPH Excellence 2016
program – CoNeXT. Porous Titanium Granule material
was kindly provided by Tigran Technologies AB,
Malmø, Sweden. Photos during operation were taken
by Camilla Albeck Neldam. Blinding of sections was
performed by Maria Ellegaard Larsen.
8. REFERENCES
Alffram PA, Bruce L, Bjursten LM, Urban RM,
Andersson GB: Implantation of the femoral stem into a
bed of titanium granules using vibration: a pilot study
of a new method for prosthetic fixation in 5 patients
followed for up to 15 years. Ups J Med Sci 112(2):183-
189, 2007.
Behnia H, Khojasteh A, Soleimani M, Tehranchi A,
Atashi A. Repair of alveolar cleft defect with
mesenchymal stem cells and platelet derived growth
factors: A preliminary report. J Craniomaxillofac Surg
40(1):2-7, 2012.
Ben-David D, Kizhner T, Livne E, Srouji S: A tissue-
like construct of human bone marrow MSCs composite
scaffold support in vivo ectopic bone formation. J
Tissue Eng Regen Med 4(1):30-37, 2010.
Bystedt H, Rasmusson L: Porous titanium granules
used as osteoconductive material for sinus floor
augmentation: a clinical pilot study. Clin Implant Dent
Relat Res 11(2):101-105, 2009.
Cao Y, Sun Z, Liao L, Meng Y, Han Q, Zhao RC:
Human adipose tissue-derived stem cells differentiate
into endothelial cells in vitro and improve
neovascularization in vivo. Biochem Biophys Res
Commun 332(2):370-379, 2005.
Chen C, Garber L, Smoak M, Fargason C, Scherr T,
Blackburn C, Bacchus S, Lopez MJ, Pojman JA, Del
Piero F, Hayes DJ: In vitro and in vivo characterization
of pentaerythritol triaccrylate-co-trimethylolpropane
nanocomposite scaffolds as potential bone augments
and grafts. Tissue Eng Part A 21(1-2):320-331, 2015.
Chen J, Sotome S, Wang J, Orii H, Uemura T,
Shinomiya K: Correlation of in vivo bone formation
capability and in vitro differentiation of human bone
marrow stromal cells. J Med Dent Sci 52(1):27-34,
2005.
Coulson RA: Relationship between fluid flow and O2
demand in tissues in vivo and in vitro. Perspect Biol
Med 27(1):121-126, 1983.
Dahl M, Jørgensen NR, Hørberg M, Pinholt EM:
Carriers in mesenchymal stem cell osteoblast
mineralization – state-of-the-art. J Craniomaxillofac
Surg 42(1):41-47, 2014.
Dahl M, Syberg S, Jørgensen NR, Pinholt EM:
Adipose derived mesenchymal stem cells – their
osteogenicity and osteoblast in vitro mineralization on
titanium granule carriers. J Craniomaxillofac Surg
41(8):213-220, 2013.
Dempster DW, Compston JE, Drezner MK, Glorieux
FH, Kanis JA, Malluche H, Meunier PJ, Ott SM,
Recker RR, Parfitt M: Standardized nomenclature,
symbols, and units for bone histomorphometry: a 2012
update of the report of the ASBMR histomorphometry
98
nomenclature committee. J Bone Miner Res 28(1):2-
17, 2013.
Donath K: Preparation of histologic sections by the
cutting grinding technique for hard tissue and other
material not suitable to be sectioned by routine
methods. EXACT-Kulzer-Publication, Norderstedt, 1-
15, 1993.
Dosier CR, Uhrig BA, Willett NJ, Krishnan L, Li MT,
Stevens HY, Schwartz Z, Boyan BD, Guldberg RE:
Effect of cell origin and timing of delivery for stem
cell-based bone tissue engineering using biologically
functionalized hydrogels. Tissue Eng Part A 21(1-
2):156-165, 2015.
Erben RG: Embedding of bone samples in
methylmethacylate: an improved method suitable for
bone histomorphometry, histochemistry, and
immunohistochemistry. J Histochem Cytochem
45(2):307-313, 1997.
Goldberg VM: Selection of bone grafts for revision
total hip arthroplasty. Clin Orthop Relat Res (381):68-
76, 2000.
Goldberg VM, Stevenson S. Natural history of
autografts and allografts. Clin Orthop Relat Res
(225):7-16, 1987.
Hong J, Andersson J, Ekdahl KN, Elgue E, Axén N,
Larsson R, Nilsson B: Titanium is a highly
thrombogenic biomaterial: possible implications for
osteogenesis. Thromb Haemost 82(1):58-64, 1999.
Hutmacher DW, Schantz JT, Lam CX, Tan KC, Lim
TC: State of the art and future directions of scaffold-
based bone engineering from a biomaterials
perspective. J Tissue Eng Regen Med 1(4):245-260,
2007.
Ilizarov GA: The tension-stress effect on the genesis
and growth of tissues: Part II. The influence of the rate
and frequency of distraction. Clin Orthop Relat Res
239:263-285, 1989.
Kasten P, Vogel J, Luginbühl R, Niemeyer P, Tonak
M, Lorenz H, Helbig L, Weiss S, Fellenberg J, Leo A,
Simank HG, Richter W: Ectopic bone formation
associated with mesenchymal stem cells in a resorbable
calcium deficient hydroxyapatite carrier. Biomaterials
26(29):5879-5889, 2005.
Kasten P, Vogel J, Lüginbuhl R, Niemeyer P, Weiss S,
Schneider S, Kramer M, Leo A, Richter W: Influence
of platelet-rich plasma on osteogenic differentiation of
mesenchymal stem cells and ectopic bone formation in
calcium phosphate ceramics. Cells Tissues Organs
183(2):68-79, 2006.
Kim KI, Park S, Im GI: Osteogenic differentiation and
angiogenesis with cocultured adipose-derived stromal
cells and bone marrow stromal cells. Biomaterials
35(17):4792-4804, 2014.
Leong NL, Jiang J, Lu HH: Polymer-ceramic
composite scaffold induces osteogenic differentiation
of human mesenchymal stem cells. Conf Proc IEEE
Eng Med Biol Soc 1:2651-2654, 2006.
Lu T, Xiong H, Wang K, Wang S, Ma Y, Guan W:
Isolation and characterization of adipose-derived
mesenchymal stem cells (ADSCs) from cattle. Appl
Biochem Biotechnol 174(2):719-728, 2014.
Lund AW, Bush JA, Plopper GE, Stegemann JP:
Osteogenic differentiation of mesenchymal stem cells
in defined protein beads. J Biomed Mater Res B Appl
Biomater 87(1):213-221, 2008.
Luzi C, Verna C, Melsen B: Immediate loading of
orthodontic mini-implants: a histomorphometric
evaluation of tissue reaction. Eur J Orthod 31(1):21-29,
2009.
Lyngstadaas SP, Verket A, Pinholt EM, Mertens C,
Haanæs HR, Wall G, Wallström M, Rasmusson L:
Titanium granules for augmentation of the maxillary
sinus – a multicenter study. Clin Implant Dent Relat
Res Suppl 2:e594-600, 2015.
Ma J, Both SK, Ji W, Yang F, Prins HJ, Helder MN,
Pan J, Cui FZ, Jansen JA, van den Beucken JJ: Adipose
tissue-derived mesenchymal stem cells as
monocultures or cocultures with human umbilical vein
99
endothelial cells: performance in vitro and in rat cranial
defects. J Biomed Mater Res A 102(4):1026-1036,
2014.
Marini F, Luzi E, Fabbri S, Ciuffi S, Sorace S,
Tognarini I, Galli G, Zonefrati R, Sbaiz F, Brandi ML:
Osteogenic differentiation of adipose tissue-derived
mesenchymal stem cells on nanostructured Ti6AI4V
and Ti13Nb13Zr. Clin Cases Miner Bone Metab
12(3):224-237, 2015.
Martins AM, Pham QP, Malafaya PB, Raphael RM,
Kasper FK, Reis RL, Mikos AG: Natural stimulus
responsive scaffolds/cells for bone tissue engineering
influence of lysozyme upon scaffold degradation and
osteogenic differentiation of cultured marrow stromal
cells induced by CaP coatings. Tissue Eng Part A
15(8):1953-1963, 2009.
Marx RE: Bone harvest from the posterior ilium. Atlas
Oral Maxillofac Surg Clin North Am 13(2):109-118,
2005.
Masuda T, Salvi GE, Offenbacher S, Felton DA,
Cooper LF: Cell and matrix reactions at titanium
implants in surgically prepared rat tibiae. Int J Oral
Maxillofac Implants 12(4):472-485, 1997.
Mirsaidi A, Genelin K, Vetsch JR, Stanger S, Theiss
F, Lindtner RA, von Rechenberg B, Blauth M, Müller
R, Kuhn GA, Hofmann Boss S, Ebner HL, Richards
PJ: Therapeutic potential of adipose-derived stromal
cells in age-related osteoporosis. Biomaterials
35(26):7326-7335, 2014.
Montespan F, Deschaseaux F, Sensébé L, Carosella
ED, Rouas-Freiss N: Osteodifferentiated mesenchymal
stem cells from bone marrow and adipose tissue
express HLA-G and display immunomodulatory
properties in HLA-mismatched settings: implications in
bone repair therapy. J Immunol Res 2014. Epub 2014
Apr 30.
Nino-Fong R, McDuffee LA, Esparza Gonzales BP,
Kumar MR, Merschrod S EF, Poduska KM: Scaffold
effects on osteogenic differentiation of equine
mesenchymal stem cells: an in vitro comparative study.
Macromol Biosci 13(3):348-355, 2013.
Sabetrasekh R, Tiainen H, Lyngstadaas SP, Reseland
J, Haugen H: A novel ultra-porous titanium dioxide
ceramic with excellent biocompatibility. J Biomater
Appl 25(6):559-580, 2011.
Salgado AJ, Coutinho OP, Reis RL: Bone tissue
engineering: state of the art and future trends.
Macromol Biosci 4(8):743-765, 2004.
Schubert T, Lafont S, Beaurin G, Grisay G, Behets C,
Gianello P, Dufrane D: Critical size bone defect
reconstruction by an autologous 3D osteogenic-like
tissue derived from differentiated adipose MSCs.
Biomaterials 34(18):4428-4438, 2013.
Seitz S, Ern K, Lamper G, Docheva D, Drosse I, Milz
S, Mutschler W, Schieker M: Influence of in vitro
cultivation on the integration of cell-matrix constructs
after subcutaneous implantation. Tissue Eng
13(5):1059-1067, 2007.
Sterio DC: The unbiased estimation of number and
sizes of arbitrary particles using the dissector. J
Microsc 134(Pt 2):127-136, 1984.
Sunay O, Can G, Cakir Z, Denek Z, Kozanoglu I, Erbil
G, Yilmaz M, Baran Y: Autologous rabbit adipose
tissue-derived mesenchymal stromal cells for the
treatment of bone injuries with distraction
osteogenesis. Cytotherapy 15(6):690-702, 2013.
Supronowicz P, Gill E, Trujillo A, Thula T,
Zhukauskas R, Ramos T, Cobb RR: Human adipose-
derived side population stem cells cultured on
demineralized bone matrix for bone tissue engineering.
Tissue Eng Part A 17(5-6):789-798, 2011.
Tiainen H, Lyngstadaas SP, Ellingsen JE, Haugen HJ:
Ultra-porous titanium oxide scaffold with high
compressive strength. J Mater Sci Mater Med
21(10):2783-2792, 2010.
Turner TM, Urban RM, Hall DJ, Andersson GB: Bone
ingrowth through porous titanium granulate around a
femoral stem. Ups J Med Sci 112(2):191-197, 2007.
100
Vesterby A, Kragstrup J, Gundersen HJG, Melsen F:
Unbiased stereologic estimation of surface density in
bone using vertical sections. Bone 8:13-17, 1987.
Wu W, Le AV, Mendez JJ, Chang J, Niklason LE,
Steinbacher DM: Osteogenic performance of donor-
matched human adipose and bone marrow
mesenchymal stem cells under dynamic culture. Tissue
Eng Part A 21(9-10):1621-1632, 2015.
Yue Y, Yang X, Wei X, Chen J, Fu N, Fu Y, Ba K, Li
G, Yao Y, Liang C, Zhang J, Cai X, Wang M:
Osteogenic differentiation of adipose-derived stem
cells prompted by low-intensity pulsed ultrasound. Cell
Prolif 46(3):320-327, 2013.
Zuk PA, Zhu M, Ashjian P, De Urgate DA, Huang JI,
Mizuno H, Alfonso ZC, Fraser JK, Benhaim P,
Hedrick MH: Human adipose tissue is a source of
multipotent stem cells. Mol Biol Cell 13(12):4279-
4295, 2002.
Zuk PA, Zhu M, Mizuno H, Huang J, Futrell JW, Katz
AJ, Benhaim P, Lorenz HP, Hedrick MH: Multilineage
cells from human adipose tissue: implications for cell-
based therapies. Tissue Eng 7(2):211-228, 2001.
101
14. Supplementary data
14.1 Osteogenic potential of ADMSCs
ALP activity:
n Mean Std. deviation Std. error 95% confidence
interval for mean
– lower bound
95% confidence
interval for mean
– upper bound
BMSC 29 0.4104 0.2080 0.0386 0.3313 0.4895
ADMSC 8 0.4906 0.1525 0.0539 0.3630 0.6181
bOB 29 0.5652 0.2072 0.0385 0.4863 0.6440
ANOVA:
Sum of squares df Mean square F Sig.
Between groups 0.347 2 0.174 4.247 0.019
Within groups 2.577 63 0.041
Total 2.924 65
Multiple comparisons, dependent variable, Bonferroni correction:
Mean
difference
Std. error Sig. 95% confidence
interval for mean
– lower bound
95% confidence
interval for mean
– upper bound
BMSC ADMSC
bOB
-0.0802
-0.1548
0.0808
0.0531
0.974
0.015
-0.2788
-0.2854
0.1185
-0.0241
ADMSC BMSC
bOB
0.0802
-0.0746
0.0808
0.0808
0.974
1.000
-0.1185
-0.2733
0.2788
0.1240
bOB BMSC
ADMSC
0.1548
0.0746
0.0531
0.0808
0.015
1.000
0.0241
-0.1240
0.2854
0.2733
102
14.2 Total calcium contents for medium vs. ADMSCs, Ti- vs. Ti+, and
TiO2- vs. TiO2+
Numerical values for total calcium contents as absorbance for mOB MEM with (ADMSCs)
and without ADMSCs (Medium), porous Ti granules with (Ti+) and without ADMSCs (Ti-), and
porous TiO2 granules with (TiO2+) and without ADMSCs (TiO2-). Data are presented as
mean±SEM (n=8 for each condition and time point):
Day 1 3 7 10 14 17 21
Medium 0.0435 0.0435 0.0435 0.0435 0.0435 0.0435 0.0435
SEM 2.62E-18 2.62E-18 2.62E-18 2.62E-18 2.62E-18 2.62E-18 2.62E-18
ADMSCs 0.0483 0.0533 0.0787 0.0564 0.0710 0.0665 0.0522
SEM 0.000999 0.004828 0.020488 0.004629 0.018936 0.005665 0.001354
T-test
p-value
**
0.002
0.081
0.130
*
0.027
0.189
**
0.005
***
0.0004
Ti- 0.1165 0.1406 0.1341 0.1467 0.1422 0.1365 0.1193
SEM 0.003011 0.008060 0.002920 0.008095 0.006016 0.004305 0.002190
Ti+ 0.1408 0.2353 0.2091 0.1888 0.1895 0.1843 0.2002
SEM 0.004680 0.050516 0.016959 0.015569 0.014754 0.010351 0.040434
T-test
p-value
***
0.0006
0.105
**
0.003
*
0.031
*
0.015
**
0.002
0.086
TiO2- 0.4314 0.3748 0.6082 0.5278 0.5043 0.6230 0.5916
SEM 0.033612 0.014972 0.019107 0.020207 0.026602 0.035240 0.035852
TiO2+ 0.4650 0.4943 0.7059 0.7161 0.7781 0.9382 0.8485
SEM 0.021641 0.039742 0.046952 0.045358 0.098466 0.121907 0.131908
T-test
p-value
0.415
*
0.020
0.085
**
0.004
*
0.028
*
0.037
0.097
103
14.3 Total calcium content
Numerical values for total calcium content as absorbance for the difference from the above-
mentioned groups (ADMSCs, Ti, and TiO2). Data are presented as mean±SEM (n=8 for each
condition and time point):
Day 1 3 7 10 14 17 21
ADMSCs 0.0048 0.0098 0.0352 0.0129 0.0275 0.0230 0.0087
SEM 0.0010 0.0048 0.0205 0.0046 0.0189 0.0057 0.0014
Ti 0.0243 0.0947 0.0751 0.0421 0.0473 0.0478 0.0809
SEM 0.0052 0.0522 0.0169 0.0213 0.0143 0.0102 0.0393
TiO2 0.0336 0.1195 0.0977 0.1883 0.2738 0.3152 0.2569
SEM 0.0211 0.0381 0.0393 0.0342 0.0854 0.1077 0.1205
ANOVA
p-value
0.274
0.119
0.284
***
5.7E-05
**
0.004
**
0.006
0.069
104
14.4 Relative ALPL, COL1α1, and RUNX2 gene expression
Numerical values for relative ALPL, COL1α1, and RUNX2 gene expression for ADMSCs,
porous Ti granules with pre-seeded ADMSCs (Ti+), and porous TiO2 granules with pre-seeded
ADMSCs (TiO2+):
ALPL Day 1 4 8
ADMSCs 1.04 0.88 1.52 SEM 0.07 0.05 0.44
Ti+ 1.59 1.20 1.21 SEM 0.26 0.10 0.08
TiO2+ 1.90 1.05 1.19 SEM 0.18 0.07 0.22
T-test (ADMSCs/Ti+)
p-value
0.073
**
0.0098
0.520
T-test (ADMSCs/TiO2+)
p-value
**
0.001
0.057
0.111
T-test (Ti+/TiO2+)
p-value
0.336
0.239
0.895
COL1α1 Day 1 4 8
ADMSCs 5.60 2.12 1.69 SEM 0.78 0.42 0.43
Ti+ 1.02 0.45 0.56 SEM 0.31 0.07 0.07
TiO2+ 1.35 0.17 0.14 SEM 0.36 0.02 0.01
T-test (ADMSCs/Ti+)
p-value
**
0.0004
**
0.005
*
0.035
T-test (ADMSCs/TiO2+)
p-value
***
0.0002
**
0.002
**
0.009
T-test (Ti+/TiO2+)
p-value
0.499
**
0.003
***
0.0006
RUNX2 Day 1 4 8
ADMSCs 1.41 1.00 1.17 SEM 0.13 0.17 0.14
Ti+ 1.35 1.06 1.08 SEM 0.23 0.08 0.08
TiO2+ 2.48 1.32 0.93 SEM 0.27 0.08 0.08
T-test (ADMSCs/Ti+)
p-value
0.832
0.742
0.594
T-test (ADMSCs/TiO2+)
p-value
**
0.003
0.112
0.160
T-test (Ti+/TiO2+)
p-value
**
0.007
0.051
0.209
