Acta Biomaterialia 3 (2007) 573–585
www.elsevier.com/locate/actabiomat
Surface modifications and cell–materials interactions with anodized Ti
Kakoli Das, Susmita Bose, Amit Bandyopadhyay
*
W.M. Keck Biomedical Materials Research Laboratory, School of Mechanical and Materials Engineering, Washington State
University, Pullman, WA 99164, USA
Received 30 August 2006; received in revised form 21 November 2006; accepted 4 December 2006
Abstract
The objective of this study was to investigate in vitro cell–materials interactions using human osteoblast cells on anodized titanium. Titanium is a bioinert material and therefore becomes encapsulated after implantation into the living body by a fibrous tissue
that isolates it from the surrounding tissues. In this work, a bioactive TiO2 layer was grown on commercially pure titanium substrate
by an anodization process using different electrolyte solutions, namely H3PO4, HF and H2SO4. These electrolytes produced bioactive
TiO2 films with a nonporous structure showing three distinctive surface morphologies. Human osteoblast cell growth behavior was
studied with as-received and anodized surfaces using an osteoprecursor cell line (OPC 1) for 3, 5 and 11 days. When anodized surfaces were compared for cell–materials interaction, it was noticed that each of the surfaces has different surface properties, which led
to variations in cell–materials interactions. Colonization of the cells was noticed with a distinctive cell-to-cell attachment in the HF
anodized surface. Good cellular adherence with extracellular matrix extensions in between the cells was noticed for samples anodized
with H3PO4 electrolyte. The TiO2 layer grown in H2SO4 electrolyte did not show significant cell growth on the surface, and some cell
death was also noticed. Cell adhesions and differentiation were more pronounced with vinculin protein and alkaline phosphatase,
respectively, on anodized surfaces. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium assays also showed an increase in living cell
density and proliferation with anodized surfaces. It was clear that rough surface morphology, high surface energy and low values of
contact angles were important factors for better cell materials interaction. A mineralization study was done in simulated body fluid
with ion concentrations nearly identical to those of human blood plasma to further understand biomimetic apatite deposition behavior. Similar to cell–materials interaction, variations in mineral deposition behavior were also noticed for films grown with different
electrolytes.
Ó 2007 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
Keywords: Anodization; Cell–material interactions; Oxide film
1. Introduction
Titanium and its alloys are widely used for orthopedic
and dental implants for their superior mechanical properties, low modulus, excellent corrosion resistance and good
biocompatibility. Titanium, being bioinert, cannot directly
bond to bone. After implantation into the living body it
becomes encapsulated by a fibrous tissue that isolates it
from the surrounding bone [1]. Studies have demonstrated
that bone tissue can form on the titanium surface with a
*
Corresponding author. Tel.: +1 509 335 4862.
E-mail address: amitband@wsu.edu (A. Bandyopadhyay).
very thin oxide cementum layer in between [2,3]. This phenomenon is known as osseointegration [4]. It takes several
months for titanium implants and bone tissue to reach integration. Hence, there is a growing interest in shortening the
process toward osseointegration and thereby reducing surgical restrictions. Various surface modifications, including
chemical treatment [5–7], thermal treatment [8], electrochemical method [9] and anodization [10], have been
applied to form a bioactive titanium oxide layer on the
metal surface. Evaluation of the above-processed films in
a simulated body fluid (SBF) demonstrated that apatite
deposition depends on chemical composition, surface properties, concentration of the electrolytes and crystalline
phases present in the film [11].
1742-7061/$ - see front matter Ó 2007 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.actbio.2006.12.003
574
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
Several in vitro and in vivo studies have shown that
modified surfaces have a higher early level of cell attachment than the untreated Ti surface. It has been shown that
micro-arc-oxidized and hydrothermally synthesized surfaces showed higher cell adhesion and proliferation rates
than did the untreated Ti surface [12]. Hydrothermal treatment of anodized Ti-substrate with calcium glycerophosphate and calcium acetate as electrolytes also led to
higher protein production than with untreated Ti surface
[13]. It was also shown that a titanium oxide surface had
a greater number of osteoblasts with higher cell activity
than did bare Ti surface [14].
The objective of this study was to understand the bone cell
materials interactions on a commercially pure Ti (cp-Ti) surface that has been modified via anodization in acid electrolytes. A titanium oxide layer on the Ti surface was grown
by an anodization process using H2SO4, HF and H3PO4 electrolytes at an anodizing voltages of 20 V. The biological
properties of these anodized surfaces were evaluated in
SBF for 3, 7, 14 and 21 days, and in vitro with the osteoblast
precursor cell line (OPC1). Apatite precipitation on substrates was observed for all anodized surfaces after incubation in SBF, and each surface gave rise to a different
cellular response due to variations in the surface properties.
2. Materials and methods
spectroscopy (EDS) was performed to qualitatively identify
the composition of the film. Oxide film thicknesses were
measured in SEM from a cross-sectional view of the Tifoils. Three samples were used to give the average thickness
value for each anodized condition.
Glancing angle X-ray diffraction (GAXRD) was conducted using a Bruker/Seimens platform system at University of Wisconsin, Madison. GAXRD studies were carried
out from a sealed Cu tube operating at 40 keV and
20 MPa. Glancing angles of 5° and 10° were used for all
the samples. Each of these samples was scanned in the 2h
range of 20–80°.
2.2. Surface analysis
2.2.1. Surface roughness
The surface roughness of each Ti-control and anodized
sample was measured using a surface profilometer (SPN
Technology, Goleta, CA). The scan was performed on each
sample three times, with a scan distance of 4 mm, at different
places on the sample. For each anodized condition three
samples were used. The stylus radius was 25 lm. The roughness data for 4 mm scan was based on 5000 points on the tip
of the profilometer. From all of these 5000 points, the root
mean square data (r.m.s.) were calculated. The average of
9 r.m.s. values (three from each sample) for each anodized
condition was determined along with the standard deviation.
2.1. Preparation and characterization of anodized Ti
A cp-Ti (99.6 at.%) sheet was used as the starting material. Each of the samples was 12 mm in diameter and
0.5 mm in thickness. Samples were abraded by silicon carbide paper in successive grades from 600 to 1200 grit (Leco
Corporation, MI), then ultrasonically cleaned in distilled
water and dried at room temperature. Final polishing
was performed using a cotton polishing cloth with a 1 lm
alumina suspension. Polished samples were used as controls for the experiments. For the anodization process,
the electrolytic cell consisted of a three-neck round bottomed flask. A cp-Ti anode was suspended from the centre
neck of the cell, a platinum cathode was suspended from
one of the necks by a platinum wire and a thermometer
was placed through the other neck. A DC power supply
(Hewlett Packard 0–60 V/0–50 A, 1000 W) was used to
vary the applied voltage. Three different solutions were
used as electrolyte: 1 (N) sulfuric acid (H2SO4), 1 (N) phosphoric acid (H3PO4) and 0.25 (N) hydrofluoric acid (HF).
All anodization experiments were performed at 20 V. Temperature of the electrolyte was maintained at 30 °C. The
temperature rise due to exothermic anodic reaction was
controlled by using large amount of electrolyte and changing the solution after every 1 h. TiO2 anodized film was
characterized using a scanning electron microscope
(SEM; Hitachi’s-570) and a field emission scanning electron microscope (FESEM; FEI, SIRION, OR) fitted for
energy-dispersive X-ray analysis. The operating voltage
for both SEM and FESEM was 20 kV. Energy-dispersive
2.2.2. Contact angle measurement
Contact angles were measured using the sessile drop
method with a face contact angle set-up equipped with a
microscope and a camera. A 0.5–1.0 ll droplet of distilled
water and McCoys 5A solution at pH 7.4 (cell culture medium) was suspended from the tip of the microliter syringe.
The syringe tip was advanced toward the disk surface until
the droplets made contact with the disk surface. Images
were collected with the camera and the contact angle
between the drop and the substrate was measured from
the magnified image. Three samples for each anodized conditions were used to collect the contact angle data in water
and in cell media. The apolar liquid diiodomethane and
two polar liquids, formamide and glycerol, were used in
the following equation to calculate the surface energy:
LW 1=2
1=2
cL ð1 þ cos hÞ ¼ 2 cLW
þ 2 cþ
S cL
S cL
þ 1=2
þ 2 c
ð1Þ
S cL
In Eq. (1), h is the contact angle of liquid L and solid S,
cLW is the apolar component of the surface energy, c+ is
the Lewis acid component (electron acceptor) and c is
the Lewis base component (electron donor) [15].
2.3. Cell–materials interaction studies
2.3.1. Cell morphology
For the cell–materials interaction studies, human osteoblasts were used. Cells were derived from an osteoblastic
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
precursor cell line (OPC1) established from human fetal
bone tissue [16]. Cells were plated at a density of
105 cm2 in 100 mm tissue culture plates and were cultured
in McCoy’s 5A medium (with L-glutamine, without phenol
red and sodium bicarbonate). Then 5% fetal calf serum, 5%
bovine calf serum, 2.2 g l1 sodium carbonate, 100 mg l1
streptomycin and 8 lg ml1 Fungizone (GibcoTM Labortories, Grand Island, NY) were added to the medium. Cells
were maintained at 37 °C under an atmosphere of 5%
CO2 and 95% air. Anodized samples were sterilized by
autoclaving (Amerex Instruments Ltd., CA) for 20 min at
121 °C before cell culture. OPC1 cells were seeded onto
the sterilized anodized titania surface. Culture medium
was changed every 2–3 days for the duration of experiment.
Samples were then removed from culture at 3, 7 and
11 days of incubation to study cell–materials interactions
under SEM.
For SEM observation, anodized cell cultured samples
were placed in 0.1 M phosphate-buffered saline (PBS) and
rinsed quickly. Samples were subsequently fixed with 2%
paraformaldehyde/2% glutaraldehyde in 0.1 M cacodylate
overnight at 4 °C. Following three rinses in 0.1 M cacodylate, each sample was post-fixed in 2% osmium tetroxide
(OsO4) for 2 h at room temperature. The fixed sample
was then again rinsed three times in 0.1 M cacodylate
and dehydrated in an ethanol series (30%, 50%, 70%,
95% and 100% three times). Samples were then critical
point dried using acetone and hexamethyldisilazane. Samples were mounted in aluminum stubs, gold coated (Technis Hummer, San Jose, CA) and observed under SEM.
2.3.2. Cell proliferation using MTT assay
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium
bromide (MTT) assay was used to evaluate cell proliferation on anodized samples. Duplicate samples per group
were evaluated. An MTT (Sigma, St. Louis, MO) solution
of 5 mg ml1 was prepared by dissolving MTT in PBS,
then filter sterilized. MTT was diluted with 50 ll of the
solution in 450 ll of serum-free phenol red-free Dulbeco’s
minimum essential medium. A 500 ll quantity of the above
diluted solution was used in each sample in 12-well plates
to form formazan by mitochondrial dehydrogenases. After
2 h incubation at 37 °C, 500 ll of the solubilization solution (10% Triton X-100, 0.1 N HCl and isopropanol) were
added in each well plate to dissolve the formazen crystals.
Then 100 ll of the solution was transferred to a new 96well plate and five data points were obtained from each
sample. The optical density of the solution in each well
was measured at a wavelength of 570 nm using a Microplate reader (Cambridge Tech., Inc., EIA). Experimental
results are presented as the mean of 10 values with standard deviation.
2.3.3. Immunochemistry and confocal microscopy
Cells cultured on anodized samples for pre-specified
number of days were fixed in 4% paraformaldehyde in
0.1 M phosphate buffer and were kept for 24 h at 4 °C
575
for future use. Those samples were rinsed in Triton for
10 min and blocked with TBST/BSA (Tris-hydroxymethyl
aminomethane (Tris)-buffered saline with 1% bovine serum
albumin, 250 mM NaCl, pH 8.3) for 1 h. Primary antibody
alkaline phosphatase (ALP) (Sigma, St. Louis, MO) or
Vinculin (Sigma–Aldrich, MO) was added at a 1:100 dilution and incubated at room temperature overnight. ALP
was used for cell differentiation, while vinculin was used
for cell attachment. The following day, samples were rinsed
with TBST/BSA three times for 10 min each. The secondary antibody, Oregon green goat anti-mouse antibody
(Molecular Probes, Eugene, OR), was added at a 1:100
dilution and incubated at room temperature for 1 h. Samples were then mounted on coverslips with Vectashield
mounting medium (Vector Labs, Burlingame, CA) with
propidium iodide (PI) and observed in confocal scanning
laser microscopy (BioRad 1024 RMC).
2.4. Mineralization study in simulated body fluids
Bioactivity of anodized titanium samples were evaluated
by immersion in simulated body fluids (SBF), which has a
similar ionic composition to human blood plasma. The
solution was prepared by dissolving NaCl, KCl, NaHCO3,
MgCl2 Æ 6H2O, CaCl2 Æ 2H2O, Na2SO4 Æ 10H2O and
K2HPO4 into distilled water and buffered at pH 7.35 with
tris-hydroxymethyl aminomethane (Tris) and 1 N HCl at
37 °C [17]. Samples were immersed in the glass vial containing 10 ml of SBF solutions and were kept under static conditions inside a biological thermostat at 37 °C for 3, 7, 14
and 21 days. All experiments were performed in duplicate
by running two independent glass vials simultaneously.
After exposure, samples were washed with distilled water
and then dried at 150 °C for 24 h.
3. Results
3.1. Anodization and surface characterization
Three different electrolytes. (a) 1 N H2SO4, (b) 1 N
H3PO4 and (c) 0.25 N HF were used for anodization of
the Ti-substrates that led to three different surface morphologies. Fig. 1 shows representative scanning electron
micrographs of anodized and Ti-control surfaces. The Ticontrol, being metal, shows a smooth surface, as shown
in Fig. 1(a). All the anodic surfaces were nonporous, with
patterned micrographs. The H2SO4 anodized surface was
nonporous but a few bigger pores were noticed in some
regions of the undulated topography, as shown in
Fig. 1(b). The H3PO4 anodized surface was rough with a
flowery pattern, as shown in Fig. 1(c), while the HF anodized surface shows rough granulated debris pattern, and is
shown in Fig. 1(d). The average coating thickness of the
film was between 200 and 400 nm. Samples anodized for
60 min showed complete coverage of TiO2 on Ti for all
electrolytes, whereas samples anodized for <60 min were
partially covered with the oxide film. Fig. 2 shows the
576
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
Fig. 1. FESEM micrographs showing morphology of (a) Ti surface, anodized oxides surfaces prepared by anodization in (b) H2SO4, (c) H3PO4 and
(d) HF electrolytes.
200
Table 1
GAXRD results analysis for different phases in the anodized and heattreated sample
T
160
Intensity
T
120
T
80
A
A
A
(c)
40
T
T
(b)
Sample condition
Phases
H2SO4
H2SO4 – 400 °C
H2SO4 – 600 °C
H3PO4
H3PO4 – 400 °C
H3PO4 – 600 °C
HF
Amorphous/anatase
Anatase
Antase + rutile
Amorphous/anatase
Anatase
Anatase + rutile
Anatase + rutile
(a)
0
30
40
50
60
70
80
2-Theta
Fig. 2. GAXRD results on oxide film anodized in (a) H2SO4, (b) H3PO4
and (c) HF electrolytes. A – denotes anatase, T – titanium phases.
GAXRD results on the anodized films. Most of the oxide
films show some anatase phase in an amorphous background, which is also given in Table 1. The oxide films were
uniform and adherent.
3.2. Cell–materials interactions
3.2.1. Cell morphology
OPC1 cell attachments on sample surfaces were analyzed for cell–materials interactions using a scanning electron microscope. Fig. 3 shows the SEM of cell
morphologies on Ti-control and anodized surfaces after
3 days of cell culture. Cells on Ti-control in Fig. 3(a) show
an elongated, flattened morphology. Cells on the H2SO4
anodized surface show a rounded morphology, along with
some cell death, as shown in Fig. 3(b). The H3PO4 surface
shows good cell attachment in Fig. 3(c), but cells are confined to one region and do not extend like the TiO2 surface
from the HF electrolyte. The HF anodized surface is
entirely covered with OPC1 cell layers, and filopodia extensions were observed coming out from the cell to grasp the
granulated pattern rough surface for anchorage, as shown
in Fig. 3(d). To further understand the osteoconductivity of
the anodized surfaces, cells were cultured for 11 days and
are shown in Fig. 4. Fig. 4(a) shows cells in the Ti-control
surface, in which a flattened, elongated cell morphology
with many filopodia extensions from the cell to the substrate and small calcified nodules after 11 days of cell culture can be seen. It can be seen in Fig. 4(b) that cells
minimized the surface area of attachment in the H2SO4
anodized surface. However, cell spreading and prolifera-
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
577
Fig. 3. SEM micrograph illustrating OPC1 cell adhesion after 3 days of culture on (a) control-Ti and oxide surfaces anodized in (b) H2SO4, (c) H3PO4 and
(d) HF electrolytes.
tion are excellent for the H3PO4 anodized surface shown in
Fig. 4(c). An abundant amount of extracellular matrix
(ECM) can be seen between the neighboring cells on the
anodized H3PO4 surface, forming a three-dimensional fibril
network. Colonization of the cells was noticed, with cells
developing a double net-like layer structure in the HF
anodized surface after 11 days of cell culture; these are
shown in Fig. 4(d). Distinctive cell-to-cell attachment was
noticed, with small calcified nodules as an early sign of cell
differentiation after 11 days in both the HF and H3PO4
anodized surfaces.
anodized surfaces there was no difference (p* < 0.001) in
cell spreading but there was a significant difference between
cells on the H2SO4 (p** < 0.01) and Ti-control at 11 days of
culture.
0.25
Day-5
Day-11
Day-16
P*
Optical density
0.2
3.2.2. Cell proliferation using the MTT assay
The MTT assay was used to determine the OPC1 cell
proliferation on control-Ti and different anodized surfaces.
Fig. 5 shows a comparison of cell densities on the different
anodized surfaces for 5, 11 and 16 days. For 5 days of culture, a slightly higher value in optical density was noticed
in the H3PO4 and HF anodized surfaces in comparison
with the Ti-control and H2SO4 anodized surfaces. After
11 days of culture a significant difference in optical density
was noticed. Cell density for the anodic oxide increased in
the order H3PO4 < HF < H2SO4 < Ti- control. A slight
increase in density was noticed with increasing culture time
to 16 days for each anodic condition when compared with
11 days. However, the number of cells on the H3PO4 and
HF anodic surfaces was always higher than those of the
H2SO4 and Ti-control surfaces. For the H3PO4 and HF
P*
P**
0.15
0.1
0.05
0
H3PO4
HF
H2SO4
Ti-Control
Sample conditions
Fig. 5. Optical density measured after culture for 5, 11 and 16 days at a
wavelength of 570 nm by reader. There were significant difference in
optical density after 11 days in the different anodized conditions samples.
P* < 0.001 for H3PO4, HF and P** < 0.01 for H2SO4 compared with Ticontrol at 11 days of culture.
Fig. 4. SEM micrographs showing the cellular adhesion and proliferation on (a) control Ti, oxide surfaces anodized in (b) H2SO4, (c) H3PO4 and (d) HF
electrolytes after 11 days of incubation.
578
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
3.2.3. Immunochemistry and confocal microscopy
Vinculin aids in the assembly of focal contacts by crosslinking and recruiting other proteins to form adhesive plagues. Vinculin also acts as an adhesion molecule between
the cells and the substratum. Antibody bound to vinculin
expressed green fluorescence and nuclei stained with PI in
the mounting medium expressed red fluorescence. Fig. 6
shows the vinculin study for 5 days at two different magnifications. Vinculin expression shows that a lower number
of cells attach to one another and to the substratum in
the Ti-control and H2SO4 anodized samples, as shown in
Fig. 6(a) and (b), than in the H3PO4 and HF surfaces,
shown in Fig. 6(c) and (d), respectively. However, based
on our data, there was no difference in the distribution of
focal adhesion sites in these samples.
Alkaline phosphatase is an early marker for osteoblast
differentiation and is thought to play a major role in bone
formation [18]. Fig. 7(a)–(d) show ALP activity for samples
after 5 days in culture. Little difference was found in the Ticontrol and H2SO4 anodized surfaces shown in Fig. 7(a)
and (b), respectively. Fig. 7(c) and (d) shows that the
ALP enzymatic activity was higher in the H3PO4 and HF
anodized surfaces. With the increase in cell culture time
from 5 to 11 days, the ALP activity increased progressively
in all the anodized oxide surfaces, as shown in Fig. 7(f)–(h)
for 11 days of culture. At day 11, all the anodized surfaces
show better cell differentiation activity compared with the
Ti-control (Fig. 7(e)).
3.3. Apatite-forming ability of the anodized film
To compare their biocompatibility, the anodized samples and the control Ti were immersed in SBF for 3, 7,
14 and 21 days. Fig. 8(a)–(d) shows surface micrographs
of the samples after immersion in SBF for 21 days. It can
be seen that anodized surfaces have a low level of ability
for Ca-apatite deposition in SBF. No apparent changes
appeared on the polished surface of Ti-control even after
21 days. It has been reported that heat treating anodized
samples at a higher temperature results in more apatite
deposition [19]. Therefore, in the present study the influence of temperature on the surface TiO2 layer for apatite
deposition was investigated by heat treating the anodized
samples at 400 and 600 °C for 2 h. It was found that heat
treatment resulted in anatase and rutile titanium oxide
phases with different surface morphologies compared with
Fig. 6. Immunolocalization of adhesive molecule vinculin in OPC1 cells for 5 days culture on (a) Ti-control, (b) H2SO4, (c) H3PO4, (d) HF surfaces; green
fluorescence indicating antibody bound to vinculin, red fluorescence indicating antibody bound to DNA (nucleus). (For interpretation of the references in
color in this figure legend, the reader is referred to the web version of this article.)
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
579
Fig. 7(a–d). Confocal micrographs of ALP expression for 5 days in OPC1 cells cultured on (a) Ti-control, (b) H2SO4, (c) H3PO4 and (d) HF for 5 days.
Green fluorescence indicating antibody bound to ALP, red fluorescence indicating antibody bound to DNA (nucleus). (For interpretation of the references
in color in this figure legend, the reader is referred to the web version of this article.)
the anodized surface, as shown in Fig. 9(a)–(f). Heat-treated samples showed significant Ca-apatite deposition in
SBF after 7 days, as shown in Fig. 10(a)–(f). From EDS
analysis in Fig. 11(a), it was observed that the precipitated
particles on the surface of anodic film were Ca, P and Mg
compounds, and these are reported in Table 2. Infrared
results in Fig. 11(b) show the presence of OH, PO2
4
and CO2
3 groups.
4. Discussions
The different kinetic steps of growth of an anodic film
reported by Vermileya [20] are as follows: nucleation of
the anodic film; growth of crystals; possible formation of
a continuous film; thickening of the film; limitation of the
film thickness. Deplancke et al. described that there is
always presence of natural continuous oxide film before
anodization [21]. Therefore, the last two steps were changed as: thickening of the natural oxide film; oxygen evolution and dissolution of the film in sulfuric acid. In general,
oxide film begins inhibiting the active dissolution of titanium according to the following reaction:
Ti þ 2H2 O ! TiO2 þ 4Hþ þ 4e
It is generally known that titanium oxide growth involves
field-assisted migration of ions through the oxide films
[22,23]. A slower dissolution rate also creates a saturation
effect on the film thickness.
Surface properties of implant material have a crucial
role for cell adhesion and cell proliferation. The different
morphology for the three different anodized surfaces
leads to different cell–materials interactions. When a biomaterial is implanted into a body, its surface is immediately covered with blood and serum proteins [24]. Cell
adhesive proteins such as fibronectin and vitronectin,
found in high concentrations in the blood, can provide
attachment sites for cells. Materials that adsorb more
attachment proteins can provide more sites for osteoblast
precursor bonding to the implant, which then lead to faster bone in-growth and implant stabilization. Surface
topography plays an important role as it gives focal
adhesion points for the proteins to get attached and thus
help in the cell adhesion process [25,26]. The ‘‘flowery
effect’’ in the H3PO4 and rough patterned HF anodized
surfaces give more focal points than the smooth metallic
580
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
Fig. 7(e–h). Confocal micrographs of ALP expression for 11 days in OPC1 cells cultured on (a) Ti-control, (b) H2SO4, (c) H3PO4 and (d) HF for 11 days.
Green fluorescence indicating antibody bound to ALP, red fluorescence indicating antibody bound to DNA (nucleus). (For interpretation of the references
in color in this figure legend, the reader is referred to the web version of this article.)
Ti-control surface. The three anodized surfaces are nonporous, but only two of them gave better cellular attachment, the exception being the H2SO4 surface. This led us
to investigate other factors that control cell–materials
interactions.
Factors which also play a crucial role in the cell attachment process are surface roughness, contact angles or wettability and surface energy. It is generally accepted that
high roughness led to better cell attachment. Table 3 shows
the average r.m.s. roughness values for different surfaces.
There is almost an order of magnitude difference in roughness for the Ti-control and anodized surfaces. As-received
and polished titanium samples show a smooth surface with
an r.m.s roughness value of 0.0074 lm. In the literature it
has also been noted that cellular adhesion, proliferation
and detachment strength are sensitive to surface roughness.
A roughness of 4.68 lm on the surface of HA disc-shaped
pellets showed a higher cellular adhesion than a roughness
of 0.73 lm [27]. Lampin et al. also showed that the degree
of roughness affects the surface energy and hence the cell
adhesion process [28].
The contact angle on the anodized and Ti-control surfaces was measured using cell media and distilled water,
as shown in Fig. 12. It was seen that the contact angle
decreased significantly in the H3PO4 surface, improving
the wettability of the surface. A contact angle higher than
20° shows the hydrophobic nature of the surface. Since the
cell media is water based, the cellular attachment is poor in
any hydrophobic surface. The H3PO4 surface with
improved wettability enhanced the cell attachment process.
Webb et al. also noted the highest level of cell attachment
on moderately hydrophilic surfaces [29]. A high contact
angle with a hydrophobic nature led to poor cell attachment in the H2SO4 surface. A low contact angle also led
to high surface energy, which is another factor that contributes to better cell attachment.
To better understand the cell–materials interaction, the
effect on surface energies of the anodized surfaces was calculated using three different well-characterized liquids. The
contact angles for each of the liquids on the samples are
listed in Table 4. The contact angles in diiodomethane
and formamide were 1° and 5° for the H3PO4 anodized
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
581
Fig. 8. FESEM photographs demonstrating the deposition of precipitates in SBF for 21 days on (a) Ti-control, oxidized titanium surface anodized in
(b) H2SO4, (c) H3PO4 and (d) HF electrolytes.
Fig. 9. FESEM micrograph showing the surface morphology of TiO2 layer prepared by heat treatment at (a) 400 °C, (b) 600 °C anodized H2SO4 surface,
(c) 400 °C, (d) 600 °C anodized in H3PO4 and (e) 400 °C, (f) 600 °C of the anodized HF surface.
surface, and a very high surface energy value, 101.27 mJ
m2, was obtained in comparison with the Ti-control. A
very low surface energy value of 55.60 mJ m2 leads to
poor cellular attachment in the H2SO4 surface. Cell attachment and proliferation were inferior in the Ti-control surface than in the anodized surfaces, indicating that surface
582
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
Fig. 10. Mineralization study for 7 days on the anodized surface H2SO4 followed by heat treatment at (a) 400 °C, (b) 600 °C, anodized in H3PO4 and heat
treated at (c) 400 °C, (d) 600 °C for anodized HF surface.
Fig. 11. (a) EDS analysis for the apatite deposition on anodized surface in SBF. Presence of calcium to phosphorous ratio: Ca/P = 1.36. (b) IR data for
H3PO4 oxide surface immersed in SBF for 21 days.
Table 2
EDS results presents in the table shows the amount of different element
present in wt.% and at.%
Element
wt.%
at.%
CK
OK
Mg K
PK
Ca K
Ti K
10.04
20.16
0.52
1.60
2.10
65.58
23.21
34.88
0.65
1.43
1.94
37.89
energy plays an important role on cell–materials interaction. Other studies also showed that a better cell attachment takes place on surfaces with higher energy [12].
Redey et al. showed that the low-energy strata in carbon-
Table 3
Surface roughness on the anodized and control Ti-surface
Surfaces
Average r.m.s. values (lm)
Standard deviation (lm)
Ti-control
H2SO4
HF
H3PO4
0.007414
0.025
0.0830
0.0157
0.000101
0.00145
0.00415
0.00075
ate apatite led to poorer human trabecular osteoblastic cell
attachment and spreading than the high-energy strata
hydroxyapatite surface [30]. Ti-materials interaction with
osteoblast cells depend not only on the oxide surface but
also on the surface properties of the film. A patterned
microstructural topography, improved wettability, high
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
Contact Angle (deg.)
80
water
60
cell media
40
20
0
Ti-Control
H2SO4
H3PO4
HF
Sample
Fig. 12. Contact angle of liquid droplets measured on the anodized and
Ti-control samples.
surface energy and surface roughness lead to more cell
attachment and proliferation than with the control Ti
surface.
Osteoblasts are attachment-dependent cells, i.e. they
must attach first to the surface then spread and mineralize
their extracellular matrix [31]. The attachment phase is an
initial stage involving physicochemical linkage between
cells and the attaching surface. Adhesion of cells is affected
by the attaching surface morphology, surface roughness,
surface wettability and surface energy [31–33]. With
increasing roughness, the cell attachment also increases,
as shown in the 5 day cell culture results with vinculin protein expression in Fig. 6. A higher intensity of adhesion plague was noticed in Ti6Al4V surfaces with degrees of
roughness of grade 4000, 1200, 80 shows higher intensity
of adhesion plague than the mirror-finish smooth surfaces
as reported by Linez-Bataillon et al. [34]. The surface
roughness affected the surface morphology, which in turn
affected the surface wettability. In the present case, cells
were attached more to the hydrophilic H3PO4 anodized
surface (Fig. 6(c)) than in the hydrophobic Ti-control surface (Fig. 6(a)).
From the MTT assay study, we observed that cell proliferation increased significantly when osteoblasts were cultured for 11 and 16 days. The proliferation of cells
usually follows a standard pattern: (1) the initial lag phase:
a period of adaptation during which cells replace elements
of the cell surface; (2) the log phase: a phase of exponential
increase in cell number; and (3) the plateau phase: cell division is balanced by cell loss and bone cell phenotypic markers are expressed [35,36]. Cell cultures were consistent and
started to follow a log phase after 5 days of culture, and
there was a high yield of proliferative activity in osteoblast
583
cells after 11 days. It was also noticed that cell proliferation
was affected by surface topography and roughness. A significant difference in cell density was noticed on anodized
surfaces and the flat-polished Ti-control surface. The difference in cell density was statistically significant after 11 days
in the H3PO4 and HF anodized surfaces compared with the
Ti-control.
ALP is an early marker for osteoblast differentiation
and it is generally accepted that differentiation increases
with surface roughness [37–39]. Boyan and co-workers
found that, with increase in surface roughness of the Ti
substrate, the ALP activity of MG 63 cells was higher after
7 days [37]. Similar results were observed in our 5 day
results, which showed higher ALP expressions in rougher
H3PO4 and HF anodized surfaces than the smoothpolished Ti-control and H2SO4 anodized surfaces in
Fig. 7(a–d). ALP also plays a role in mineralization. With
an increase in culture time to 11 days there is an enhancement in ALP expression in all the anodized surfaces, indicating the beneficial effect of the anodization of the surface
on ECM mineralization.
A mineralization study of anodized samples showed the
presence of a Ca-compound on the surface. It has been suggested that apatite, once nucleated, will grow spontaneously in the body environment [40]. There have been
several reported models for apatite deposition on the
TiO2-surface [5,42,43]. A TiO2 layer forms Ti–OH groups
in an aqueous solution and these groups serve as nucleation
sites for apatite crystals and favor deposition. Reaction of
the Ti–OH groups with aqueous solution changes the surface charge of the Ti surface. The negatively charged OH
interacts with positively charged Ca2+ ions and thus reveals
a positive charge on the surface. The surface then interacts
with phosphate groups and forms amorphous calcium
phosphate [41]. These calcium phosphates eventually crystallize to bone-like apatite. The high ability for apatite
deposition on heat-treated samples led us to study the influence of the surface properties on the kinetics of nucleation
of calcium-apatite. Surface properties were characterized
by contact angle measurement on two different surfaces –
one with anatase phase heat treated at 400 °C and other
with anatase/rutile phase heat treated at 600 °C. The
observed contact angles of various liquids are much larger
on rutile/anatase surfaces than on the just anatase surface
(Table 5). In nucleation, the tendency for a crystal to form
depends on two thermodynamic factors: (a) Gibbs free
energy and (b) the activation energy barrier to diffusion
Table 4
Surface energy of the samples
Surfaces
Formamide
Glycerol
Diiodomethane
Surface energy (mJ m2)
Ti-control
H2SO4
HF
H3PO4
18 ± 1.15
21 ± 1
11 ± 3.05
5 ± 0.57
40 ± 2.04
41 ± 1.15
39 ± 4.61
31 ± 1.73
37 ± 0.57
35 ± 0.57
23 ± 2.08
1 ± 0.577
73.68 ± 0.20
55.60 ± 0.53
77.25 ± 1.09
101.27 ± 1.08
584
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
Table 5
Contact angles (deg) of various liquids on anatase and rutile surfaces
Liquid
Anatase
Anatase/rutile
Diiodomethane
Glycerol
Formamide
Water
13 ± 0.82
24 ± 1.70
10 ± 1.63
58 ± 2.08
30 ± 1.64
28 ± 3.11
12 ± 1.71
60 ± 2
Table 6
Surface energy components and parameters of anatase and anatase/rutile
Sample
cLW
c+
c
DG
Anatase
Anatase/rutile
49.5
44.22
57.57
4.57
51.97
2.53
705.6
151.42
from bulk solution to substrate. The energy barrier, caused
by Lewis acid–base interactions across the interface, is
reduced due to the high value of the Lewis base (c+) parameter in the anatase surface (Table 6) [44]. According to a
physical model, it is reported that an increase in adhesion
energy promotes nucleation on solid substrates [45]. The
calculated free energy of adhesion of water on the anatase
surface is DG = 705.6 mJ m2. The low free energy on the
anatase surface facilitates faster nucleation of apatite than
the high free energy (DG = 151.42 mJ m2) of the anatase/rutile surface.
5. Conclusions
Titanium oxide nonporous films were grown on Ti surfaces via anodization for biomedical applications in three
different acidic solutions. A difference in microstructure
changes the cellular behavior from a rounded cell morphology with minimized surface attachment in the H2SO4 anodized surface to cell spreading and abundant extracellular
matrix in the H3PO4 and HF anodized surfaces. Cell adhesion using the vinculin adhesive molecule and cell proliferation using MTT assays showed that the best adherence
and proliferation were in the HF and H3PO4 anodized
oxide surfaces compared with the H2SO4 and Ti-control
surfaces. ALP expression on anodized surfaces shows pronounced cell differentiation. It was seen that higher surface
roughness, low contact angle, improved wettability and
high surface energy led to better cell attachment and proliferation in two of the anodized surfaces. It was also
observed that with the emergence of the rutile phase the
free energy of the surface increases, leading to a decrease
in apatite deposition on the oxide surfaces.
Acknowledgements
The authors would like to thanks Ms. Jessica Moore
and Professor Howard L. Hosick of the School of Biological Sciences and Molecular Biosciences, WSU. We would
also like to acknowledge Dr. Weichang Xue, a postdoctorate fellow in our laboratory, for his helpful assistance. We
acknowledge the financial support from the Office of Naval
Research (N00014-01-04-0644 and N00014-01-05-0583).
References
[1] Nishiguchi S, Nakamura T, Kobayashi M, Kim HM, Miyaji F,
Kokubo T. The effect of heat treatment on bone-bonding ability of
alkali-treated titanium. 20, 1999. p. 491–500.
[2] Sennerby L, Thomsen P, Ericson LE. Ultrastructure of the bone–
titanium interface in rabbits. Mater Med 1992;3:262–71.
[3] Branemark P-I, Hansson BO, Adell R, Breine U, Lindström U,
Hallen O, Ohman A. Osseointegrated implants in the treatment of the
edentulous jaw. Experience from a 10 year period. Scand J Plast
Reconstruct Surg 1977;11:1–132.
[4] Branemark P-I. Osseointegration and its experimental background. J
Prosthet Dent 1983;50:399–410.
[5] Jonášová I, Müller FA, Helebrant A, Strnad J, Greil P. Biomimetic
apatite formation on chemically treated titanium. Biomaterial
2004;25:1187–94.
[6] Xiao F, Tsuru K, Hayakawa S, Osaka A. In vitro apatite deposition
on titania film derived from chemical treatment of Ti substrates with
an oxysulfate solution containing hydrogen peroxide at low temperature. Thin Solid Films 2003;441:271–6.
[7] Wu J-M, Hayakawa S, Tsuru K, Osaka A. Porous titania films
prepared from interactions of titanium with hydrogen peroxide
solution. Scripta Mater 2002;46:101–6.
[8] Fujibayashi S, Neo M, Kim H-M, Kokubo T, Nakamura T.
Osteoinduction of porous bioactive titanium metal. Biomaterials
2004;25:443–50.
[9] Sul Y-T, Johansson CB, Jeong Y, Albrektsson T. The electrochemical
oxide growth behavior on titanium in acid and alkaline electrolytes.
Med Eng Phys 2001;23:329–46.
[10] Yang B, Uchida M, Kim H-M, Zhang X, Kokubo T. Preparation of
bioactive titanium metal via anodic oxidation treatment. Biomaterials
2004;25:1003–10.
[11] Rohanizadeh R, Al-Sadeq M, LeGeros RZ. Preparation of different
forms of titanium oxide on titanium surface: effects on apatite
deposition. J Biomed Mater Res 2004;71A:343–52.
[12] Zhang YM, Bataillon-Linez P, Huang P, Zhao YM, Han Y, Traisnel
M, Xu KM, Hilderbrand HF. Surface analyses of micro-arc oxidized
and hydrothermally treated titanium and effect on osteoblast behavior. J Biomed Mater Res 2004;68A:383–91.
[13] Rodriguez R, Kim K, Joo Ong L. In vitro osteoblast response to
anodized titanium and anodized titanium followed by hydrothermal
treatment. J Biomed Mater Res 2003;65A:352–8.
[14] Feng B, Weng J, Yang BC, Qu SX, Zhang XD. Characterization of
surface oxide films on titanium and adhesion of osteoblast. Biomaterials 2003;24:4663–70.
[15] van Oss CJ, Giese Jr RF, Giese RJ. Reevaluation of the surface
tension components and parameters of polyacetylene from contact
angles of liquids. Langmuir 1990;6:1711–3.
[16] Winn SR, Randolph G, Uludag H, Wong SC, Hair GA, Hollinger
JO. Establishing an immortalized human osteoprecursor cell line:
OPC1. J Bone Miner Res 1999;14:1721–33.
[17] Kokubo T, Kushitani H, Sakka S, Kitsugi T, Yamamuro T. Solutions
able to reproduce in vivo surface-structure changes in bioactive glassceramics A-W. J Biomed Mater Res 1990;24:721–34.
[18] Laintinen M, Halttunen T, Jortikka L, Teronen O, Sorsa T,
Lindholm TS. The role of transforming growth factor-beta on
retarded osteoblastic differentiation in vitro. Life Sci 1999;64:847–58.
[19] Wang X-X, Yan W, Hayakawa S, Tsuru K, Osaka A. Apatite
deposition on thermally and anodically oxidized titanium surfaces in
a simulated body fluid. Biomaterials 2003;24:4631–7.
[20] Vermileya DA. Anodic films. Adv Electrochem Electrochem Eng
1963;3:211–86.
[21] Delplancke J-L, Degrez M, Fontana A, Winand R. Self-colour
anodizing of titanium. Surf Technol 1982;16:153–62.
K. Das et al. / Acta Biomaterialia 3 (2007) 573–585
[22] Yahalom J, Zahavi J. Electrolytic breakdown crystallization of
anodic oxide films on A1, Ta and Ti. Electrochim Acta
1970;15:1429–35.
[23] Ammar IA, Kamal I. Kinetics of anodic oxide-film growth on
titanium-I. Electrochim Acta 1971;16:1539–53.
[24] Puleo DA, Nanci A. Understanding and controlling the bone–
implant interface. Biomaterials 1999;20:2311–21.
[25] Hormia M, Kononen M. Immunolocalization of fibronectin and
vitronectin receptors in human gingival fibroblasts spreading on
titanium surfaces. J Periodontal Res 1994;29:146–52.
[26] Degasne I, Basle MF, Demais V, Hure G, Lesourd M, Grolleau B,
et al. Effects of roughness, fibronectin and vitronectin on attachment,
spreading and proliferation of human osteoblast-like cells (Saos-2) on
titanium surface. Calcified Tissue Int 1999;64:499–507.
[27] Deligianni DD, Katsala ND, Koutsoukos PG, Missirlis YF. Effect of
surface roughness of hydroxyapatite on human bone marrow cell
adhesion, proliferation, differentiation and detachment strength.
Biomaterials 2001;22:87–96.
[28] Lampin M, Warocquier-Clerout R, Degrange M, Sigot-Luizard
MF. Correlation between substratum roughness and wettability,
cell adhesion, and cell migration. J Biomed Mater Res
1997;36:99–108.
[29] Webb K, Hlady V, Tresco PA. Relative importance of surface
wettability and charged functional groups on NIH3T3 fibroblast
attachment, spreading and cytoskeletal organization. J Biomed Mater
Res 1998;41:422–30.
[30] Redey SA, Nardin M, Bernache-Assolant D, Rey C, Delannoy P,
Sedel L, Marie PJ. Behavior of human osteoblastic cells on
stoichiometric hydroxyapatite and type A carbonate apatite: role of
surface energy. J Biomed Mater Res 2000;50:353–64.
[31] Anselme K. Osteoblast adhesion on biomaterials. Biomaterials
2000;21:667–81.
[32] Boyan BD, Lohmann Ch, Dean DD, Sylvia VL, Cochran DL,
Schwartz Z. Mechanism involved in osteoblast response to implant
surface morphology. Annu Rev Mater Res 2001;31:357–71.
[33] Zhu X, Chen J, Scheideler L, Reichl R, Geis-Gerstorfer J. Effects of
topography and composition of titanium surface oxides on osteoblast
responses. Biomaterial 2004;25:4087–103.
585
[34] Linez-Bataillon P, Monchau F, Bigerelle M, Hildebrand
HF. In vitro MC3T3 osteoblast adhesion with respect to
surface roughness of Ti6Al4V substrates. Biomol Eng
2002;19:133–41.
[35] Stein GS, Lian JB. Molecular mechanism mediating proliferation/
differentiation interrelationships during progressive development of
the osteoblast phenotype. Endocr Rev 1993;1:424–42.
[36] Stein GS, Lian JB, Owen TA. Relationship of cell growth to the
regulation of tissue-specific gene expression during osteoblast differentiation. FASEB J 1990;4:3111–23.
[37] Martin JY, Schwartz Z, Hummart TW, Schraub DM, Simpson J,
Lankford J, et al. Effect of titanium surface-roughness on proliferation, differentiation, and protein-synthesis of human osteoblast-like
cells (MG63). J Biomed Mater Res 1995;29:389–401.
[38] Kieswetter K, Schwartz Z, Hummert TW, Cochran DL, Simpson J,
Dean DD, et al. Surface roughness modulates the local production of
growth factors and cytokines by osteoblast-like MG-63 cells. J
Biomed Mater Res 1996;32:55–63.
[39] Boyan BD, Batzer R, Kieswetter K, Liu Y, Cochran DL, SzmucklerMoncler S, et al. Titanium surface roughness alters responsiveness of
MG63 osteoblast-like cells to 1alpha, 25-(OH)2D3. J Biomed Mater
Res 1998;39:77–85.
[40] Li P, Ohtsuki C, Kokube K, Nakanishi N, Soga N, Yamamuro T,
et al. A role of hydrated silica, titania and alumina in forming
biologically active bone-like apatite on implant. J Biomed Mater Res
1994;7:28–30.
[41] Brunette DM, Tengvall P, Textor M, Thomsen P. Titanium in
medicine: material science, surface science, engineering, biological
responses, and medical applications. New York: Springer; 2001.
[42] Li PJ, Ohtsuki C, Kokubo T, Nakanishi K, Soga N, Degroot K. The
role of hydrated silica, titania and alumina in inducing apatite on
implants. J Biomed Mater Res 1994;28:7–15.
[43] Kim H-M, Miyaji F, Kokubo T, Nakamura T. Preparation of
bioactive Ti and its alloys via simple chemical surface treatment. J
Biomed Mater Res 1996;32:409–17.
[44] Oss van CJ. Interfacial forces in aqueous media. New York: Dekker;
1994.
[45] Walton AG. Nucleation. New York: Dekker; 1969, p. 225.