Behavior of osteoblastic cells cultured on titanium and structured zirconia surfaces.

ral
ssBioMed CentHead & Face Medicine
Open AcceResearch
Behavior of osteoblastic cells cultured on titanium and structured
zirconia surfaces
Rita Depprich1, Michelle Ommerborn*2, Holger Zipprich3,
Christian Naujoks†1, Jörg Handschel†1, Hans-Peter Wiesmann4,
Norbert R Kübler1 and Ulrich Meyer1
Address: 1Department of Cranio- and Maxillofacial Surgery, Heinrich-Heine-University, Düsseldorf, Germany, 2Department of Operative and
Preventive Dentistry and Endodontics, Heinrich-Heine-University, Düsseldorf, Germany, 3Department of Prosthetic Dentistry, Section of Materials
Sciences, Johann Wolfgang Goethe University, Frankfurt, Germany and 4Department of Cranio- and Maxillofacial Surgery, Westphalian Wilhelms-
University, Münster, Germany
Email: Rita Depprich - depprich@med.uni-duesseldorf.de; Michelle Ommerborn* - ommerborn@med.uni-duesseldorf.de;
Holger Zipprich - zipprich@em.uni-frankfurt.de; Christian Naujoks - christian.naujoks@med.uni-duesseldorf.de;
Jörg Handschel - handschel@med.uni-duesseldorf.de; Hans-Peter Wiesmann - HansPeter.Wiesmann@ukmuenster.de;
Norbert R Kübler - kuebler@med.uni-duesseldorf.de; Ulrich Meyer - ulrich.meyer@med.uni-duesseldorf.de
* Corresponding author †Equal contributors
Abstract
Background: Osseointegration is crucial for the long-term success of dental implants and depends
on the tissue reaction at the tissue-implant interface. Mechanical properties and biocompatibility
make zirconia a suitable material for dental implants, although surface processings are still
problematic. The aim of the present study was to compare osteoblast behavior on structured
zirconia and titanium surfaces under standardized conditions.
Methods: The surface characteristics were determined by scanning electron microscopy (SEM).
In primary bovine osteoblasts attachment kinetics, proliferation rate and synthesis of bone-
associated proteins were tested on different surfaces.
Results: The results demonstrated that the proliferation rate of cells was significantly higher on
zirconia surfaces than on titanium surfaces (p < 0.05; Student's t-test). In contrast, attachment and
adhesion strength of the primary cells was significant higher on titanium surfaces (p < 0.05; U test).
No significant differences were found in the synthesis of bone-specific proteins. Ultrastructural
analysis revealed phenotypic features of osteoblast-like cells on both zirconia and titanium surfaces.
Conclusion: The study demonstrates distinct effects of the surface composition on osteoblasts in
culture. Zirconia improves cell proliferation significantly during the first days of culture, but it does
not improve attachment and adhesion strength. Both materials do not differ with respect to protein
synthesis or ultrastructural appearance of osteoblasts. Zirconium oxide may therefore be a suitable
material for dental implants.
Published: 8 December 2008
Head & Face Medicine 2008, 4:29 doi:10.1186/1746-160X-4-29
Received: 27 October 2008
Accepted: 8 December 2008
This article is available from: http://www.head-face-med.com/content/4/1/29
© 2008 Depprich et al; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 9
(page number not for citation purposes)

Head & Face Medicine 2008, 4:29 http://www.head-face-med.com/content/4/1/29
Background
The objective of implantology is to design devices that
induce controlled, guided, and rapid integration into sur-
rounding tissues [1]. Events leading to integration of an
implant, and ultimately to success or failure of the device,
take place largely at the tissue-implant interface, and oste-
oblasts covering the implant surface are the crucial cell
type that regulate the tissue response at the biomaterial
surface [2]. Based on the results of numerous in vitro stud-
ies, it is now well understood that surface morphology
decisively determines the cellular behavior of osteoblasts
[2-4].
Titanium (Ti) and titanium alloys are widely used as
implant materials due to their excellent biocompatibility.
Many surface modifications have been developed to
improve cell reactions on the surface. In addition to exist-
ing titanium implants bearing machined or plasma-
sprayed surfaces, there is a great number of implants on
the market which offer surfaces altererd by grit blasting
and/or acid etching. Zirconia (zirconium dioxide, ZrO2) is
a bio-inert non-resorbable metal oxide that offers
improved mechanical properties compared to other
ceramic biomaterials, i.e. alumina. It has a good chemical
and dimensional stability, and a high strength and tough-
ness [5]. Tetragonal zirconia polycrystals (TZP) are used
for manufacturing femoral heads for total hip replace-
ments since the late 1980s [6]. Because of the tooth-like
colour, the excellent biocompatibility and mechanical
properties, ambitious efforts were made to introduce zir-
conia for applications in dentistry. Successful use of zirco-
nia for treatment of non-vital teeth [7,8], crown and
bridge restorations [9] and ceramic abutments [10] are
reported. Zirconia is also a desirable alternative material
to titanium for the fabrication of dental implants.
Titanium has a superior corrosion resistance because of its
characteristic oxide layer, however, accumulation of tita-
nium in the inner organs and lymph nodes after implan-
tation has been reported [11]. Galvanic side effects after
contact with saliva and fluoride were also described [12].
Although allergic reactions to titanium are very rare, cellu-
lar sensitization has been demonstrated [13,14]. The
main disadvantage of the biomaterial titanium is its dark
grayish colour. Unfavorable soft tissue conditions or
retraction of the gingiva may lead to aesthetic impair-
ment, especially when the maxillary incisors are involved
[15]. The clinical use of zirconia is limited, because fabri-
cation of surface modifications is difficult and smooth
implant surfaces are not beneficial for osseointegration,
due to a poor interaction with tissues [1].
Some animal experiments and numerous case reports
might be a suitable implant material [16-19]. However,
data evaluating the role of surface topography on the
response of osteoblasts at zirconia interfaces are rare [20].
Cell reactions on surfaces are strongly dependent on the
culture system that is used [21]. Since most of the widely
used osteosarcoma cell lines do not demonstrate a com-
plete pattern of osteoblastic features in vitro, the use of pri-
mary non-transformed cells seems to be superior for
assessing of osteoblast reactions on biomaterial surfaces
[2]. Therefore, the aim of this study was to compare oste-
oblast behavior on structured zirconia and titanium sur-
faces under standardized conditions using primary bovine
osteoblasts. Attachment kinetics, proliferation rate, and
synthesis of bone-associated proteins on both surfaces
were examined and compared between each other.
Methods
A modified (acid-etched) zirconia implant surface was
compared to an acid-etched titanium surface. Standard
24-well tissue culture plates (polystyrene) were used as
control surface. Zircona disks (12 mm diameter, 1 mm
thick) were made of yttrium-stabilized tetragonal poly-
crystals and titanium disks (13 mm diameter, 1.5 mm
thick) were made of commercially pure titanium. Both
materials were supplied by Konus Dental Implants (Bin-
gen, Germany). To evaluate the surfaces of zirconia and
titanium disks, scanning electron microscopy (SEM) was
performed using a a JEOL 6300F (JEOL, Eching, Ger-
many) high-resolution field emission scanning electron
microscope equipped with a EDX analysis system. The zir-
conia and titanium disks were carefully washed in diluted
water, rinsed thoroughly in 70% ethanol, and ultrasoni-
cally cleaned for 20 min in absolute alcohol. Finally, the
samples were air dried and maintained under sterile con-
ditions after gamma ray sterilization.
Primary osteoblast cell culture
Primary bovine osteoblasts were used in this study. Extrac-
tion and cultivation were performed following the
instructions of Jones et al. [22]. Under sterile conditions
periosteum was removed from the bovine metacarpus.
The periosteum was cultured at 37°C in an atmosphere of
5% CO2 and 100% humidity for 4–5 weeks in high-
growth enhancement medium (High GEM, Flow Labora-
tories, Rickmansworth, UK) containing 10% fetal bovine
serum (FBS, Gibco Laboratories Grand Island, NY, USA).
Media were changed weekly. Osteoblastic differentiation
was tested by detection of osteocalcin/osteonectin and
high alkaline phosphatase activity. When the cells reached
confluence they were harvested (20 min incubation at
37°C with 0.4 g collagenase, 98.8 mg HAM's F10 in 10 ml
HEPES (2-[4-(2-hydroxyethyl)-1-piperazinyl]ethanesul-
fonic acid); repeated washing with phosphate-bufferedPage 2 of 9
(page number not for citation purposes)
demonstrated osseointegration of zirconia implants simi-
lar to that of titanium implants, suggesting that zirconia
saline (PBS); subsequent incubation for 15 min with 300
mg ethylenediaminetetraacetic acid (EDTA)-Na, 200 mg

Head & Face Medicine 2008, 4:29 http://www.head-face-med.com/content/4/1/29
KCl, 8 g NaCl, 1 g NaHCO3, 50 mg NaH2PO4 and 1 g glu-
cose/l) and centrifuged. The pellets were resuspended
with buffer and the cell numbers were counted in a cell
counter (CASY®I Modell TT, Schärfe System, Reutlingen,
Germany).
Cell proliferation
Cell proliferation was measured after 1, 3 and 5 days,
respectively. Cells were marked with fluorescent dye
(Vybrant® CM DiI, Molecular Probes, Netherlands) and
10.000/cm2 osteoblasts were seeded into 24-well plates
on the zirconia/titanium disks or the well plate. The
experiments were repeated at least three times. Osteob-
lasts were fixed in methanol and stained with methylene
blue and azure blue according to the method described by
Richardson. Morphometric evaluation of cells was per-
formed by means of light microscopy. To determine the
cell number digital photos were taken under standardized
conditions and counted using the software program Anal-
ysis 3.0 (Olympus Soft Imaging System, Münster, Ger-
many).
Cell detachment
To determine cell adhesion on the surface of the different
materials, 60.000/cm2 primary osteoblasts were seeded
into 24-well plates on the zirconia/titanium disks or the
well plate. After incubation for 24 hrs at 37°C, 500 μl of a
trypsin-containing solution (0.25% diluted 1:2 in PBS)
was added and 400 μl aliquots of the cell suspension were
taken after a contact time of 5, 15, 25, and 35 min. Cell
numbers were determined by the use of a cell counter. As
control, the remaining of the 500 μl was removed from
the wells and 500 μl trypsin (0.25% solution, non-
diluted) was added to detach the remaining cells. After 5
min contact time and washing with PBS, aliquots of the
cell suspension (400 μl) were taken and the cell number
counted.
Immunocytochemistry
To test for osteoblastic differentiation, expression of colla-
gen I, osteocalcin and osteonectin was assessed by means
of immunocytochemistry. 60.000 osteoblasts/cm2 were
seeded into 24-well plates on the zirconia/titanium disks
and into 6-well plates on polystytol. After incubation for
7, 14, or 28 days at 37°C in an atmosphere of 5% CO2 in
the High GEM medium, primary antibodies were used
according to the manufacturers' instructions: rabbit poly-
clonal anti-collagen I (Biotrend, Cologne, Germany),
Mouse monoclonal anti-osteocalcin (TaKaRa Bio,
MoBiTec, Goettingen, Germany) and rabbit polyclonal
anti-osteonectin (SPARC; Chemicon Millipore GmbH,
Schwalbach, Germany). Alexa Flour 488-labelled second-
ary antibodies were purchased from MoBiTec (Goettin-
ized conditions using a fluorescence microscope and
processed using the software program Analysis 3.0.
Scanning electron microscopy (SEM)
Cell morphology was investigated after 2 hrs, 4 hrs and 7
days. Primary osteoblasts were seeded at a density of
15.000/cm2 on zirconia/titanium disks and for control on
smooth titanium disks and incubated for 2 hrs or 4 hrs at
37°C in an atmosphere of 5% CO2 in the High GEM
medium. To investigate confluent cells after 7 days,
40.000/cm2 osteoblasts were seeded on the zirconia/tita-
nium disks and incubated under the same conditions.
Cells were fixed in 2.5% glutaraldehyde for 3 hrs and then
washed with PBS. After sputtering with gold (Bal-tec Ag,
Balzers, Liechtenstein) the samples were investigated
using the scanning electron microscope JEOL 6300F
(JEOL, Eching, Germany).
Statistical analysis
Statistical analyses were performed using Student's t-tests
and Mann-Whitney U tests. A p < 0.05 was considered sig-
nificant. Experiments were repeated three-fold.
Results
Surface topography
Scanning electron microscopy demonstrated noticeable
differences between zirconia and titanium surfaces by
SEM revealed (Figure 1). The titanium surface was rough
and contained many pores and grooves of different size
which were regularly distributed over the whole surface.
In contrast, the zirconia surface appeared smooth with
only a few pores.
Energy-dispersion X-ray analysis
Energy-dispersion X-ray analysis confirmed the character-
istic element composition of commercial pure titanium
and zirconium dioxide. Titanium disks were composed of
the elements titanium and oxygen but also traces of sili-
cium and carbon were detected. Zirconia consisted of zir-
conium (Zr) and oxygen (O), but also hafnium (Hf) was
found frequently associated with ZrO2.
Cell proliferation
Cell proliferation was assessed on the different surfaces.
We found an increase in cell number on all surfaces over
the observation period (Figure 2). At day 1 cell prolifera-
tion was significantly higher on zirconia surfaces as com-
pared to polystyrene controll surfaces (p = 0.000) but was
similar to titanium surfaces (p = 0.158). At day 3 cell
growth was significantly higher on the zirconia surfaces
than on polystyrene (p = 0.037) and titanium surfaces (p
= 0.002). At day 5 cell proliferation was continued to be
significantly higher on zirconia surfaces than on titaniumPage 3 of 9
(page number not for citation purposes)
gen, Germany) and used according to the manufacturers'
instructions. Digital images were taken under standard-
(p = 0.001) or polystyrene surfaces (p = 0.001).

Head & Face Medicine 2008, 4:29 http://www.head-face-med.com/content/4/1/29
Cell detachment
Results revealed that at every time of the assessment fewer
cells were detached from titanium surfaces compared to
zirconia or polystyrol surfaces. The number of detached
cells from titanium surfaces remained constant at a low
level over the whole period of investigation. In contrast,
detached cells from zirconia surfaces doubled from 5 to
15 min, but remained constant thereafter. A minor
Scanning electron micrographs of a zirconia disk (left) showing occasionally pores on the smooth surface and a titanium disk (right) with rough surface nd frequent pores and grooves of different size (2 kV, magnificati n 500-fold)Fi ure 1
Scanning electron micrographs of a zirconia disk (left) showing occasionally pores on the smooth surface and a titanium disk
(right) with rough surface and frequent pores and grooves of different size (2 kV, magnification 500-fold).
zirconia titanium
Cell proliferation rates of osteoblasts on differently coated surfaces at day 1, 3 and 5, respectivelyFigure 2
Cell proliferation rates of osteoblasts on differently coated surfaces at day 1, 3 and 5, respectively. Increase in cell number was
detected on all surfaces over the observation period. Significantly higher cell proliferation was observed on zirconia surfaces on
day 1, 3 and 5 compared to titanium and polystyrene surfaces. Statistical differences (p < 0.05) as calculated by Student's t-tests
cell proliferation - day 1
0
20
40
60
80
100
120
140
160
180
200
1
n
u
m
be
r
o
f o
st
eo
bl
as
t
polystyrene titanium zirconia
cell proliferation - day 3
0
100
200
300
400
500
1
n
u
m
be
r
o
f o
st
eo
bl
as
ts
polystyrene titanium zirconia
cell proliferation - day 5
0
100
200
300
400
500
600
700
1
n
u
m
be
r
o
f o
st
eo
bl
as
ts
polystyrene titanium zirconia
**
**
*** *
**Page 4 of 9
(page number not for citation purposes)
are marked with arrows.

Head & Face Medicine 2008, 4:29 http://www.head-face-med.com/content/4/1/29
increase of detached cells was found in the polystyrol con-
trol group, and after 35 min the number of detached cells
had quadrupled. Statistical analysis confirmed significant
higher cell detachment rates from zirconia surfaces as
compared to titanium surfaces after 5 min (p = 0.047), 15
min (p = 0.009) and 25 min (p = 0.009) but not after 35
min (p = 0.1). Differences between zirconia and control
group were not significant (p < 0.05) at any time of assess-
ment.
Immunocytochemical analysis
After 7 days expression of collagen I, osteocalcin and
osteonectin were evident on all different surfaces exam-
ined. Cells were uniformly distributed throughout the
material surface and positive immunolabeling was
detected on zirconia, titanium and polystyrol surfaces.
Lower expression of osteocalcin compared to collagen I
and osteonectin was observed on all different surfaces
(Figure 3). After 14 days of culture, up-regulated expres-
sion of reticular collagen I expression was evident espe-
cially on the titanium and zirconia surfaces, whereas
osteocalcin and osteonectin expression showed no detect-
able differences on the investigated surfaces. Expression of
characteristic bone derived proteins was still detectable
after 28 days on all samples and showed no significant
differences between titanium, zirconia and polystyrol sur-
faces except of a minimally denser accumulation of colla-
gen I found on zirconia surfaces as compared to titanium
surfaces (Figure 4).
Scanning electron microscopy (SEM)
The SEM analysis performed on osteoblast-seeded sam-
ples after 2 hrs showed typically flat polygonal cells regu-
larly distributed on the titanium and on the zirconia
surfaces. Development of radiate cell filopodia was appar-
ent. After 4 hrs of culture, cell morphology on both sur-
faces showed no significant differences and was similar to
that after 2 hrs. Cell filopodia exploring the surface could
be demonstrated in fixed cells. After 7 days a mosaic-
shaped confluent cell layer had formed on zircona and
titanium surfaces (Figure 5). No ultrastructural signs of
apoptotic fibroblast-shaped cells were detected. Signifi-
cant differences could not be found.
Immunocytochemical analysis of characteristic bone derived proteinsFigure 3
Immunocytochemical analysis of characteristic bone derived proteins. After 7 days extracellular expression of collagen I and
osteonectin is evident on all different surfaces examined. Scattered expression of osteocalcin is demonstrated (magnification
osteonectin osteocalcin collagen I
titanium
20x 20x20x
20x
zirconia
20x 20x
20x
polystyrene
20x20xPage 5 of 9
(page number not for citation purposes)
20-fold).