Applied Surface Science 449 (2018) 152–165

Contents lists available at ScienceDirect

Applied Surface Science

journal homepage: www.elsevier.com/locate/apsusc

Full Length Article

Interaction of osteoblast -TiO2 nanotubes in vitro: The combinatorial

effect of surface topography and other physico-chemical factors
governs the cell fate
Sahely Saha, Ravi Kumar, Krishna Pramanik, Amit Biswas ∗
Department of Biotechnology & Medical Engineering, National Institute of Technology, Rourkela, 769008, Odisha, India

a r t i c l e i n f o a b s t r a c t

Article history: In the context of surface modification of implants by TiO2 nanotubes, the topography of the nanotubes
Received 19 September 2017 has been considered as a sole modulator of cellular response of adherent cells. However, a recent report
Received in revised form has indicated about the existence of other contributing factors. Keeping this contradiction in mind, here
24 December 2017
we have performed a comprehensive study to account the contribution of the various factors such as
Accepted 19 January 2018
Available online 8 February 2018
morphological aspects, physico chemical aspects and the topographical features of TiO2 nanotubes on
the cellular response. Here, the TiO2 nanotubes of varied topography were synthesized on the Ti6Al4V
surface through anodic oxidation. SEM analysis showed that there was a variation in tube diameter, cross-
Keywords:
TiO2 nanotubes sectional length, tube density and wall thickness along with the variation in oxidation time and voltage.
Cell-material interactions An attempt to co-relate the cellular response of osteoblast cells (MG-63) cultured on the nanotubes has
Implant interface revealed that except cell proliferation; cell adhesion, spreading, vinculin distribution (focal adhesion)
Surface modification and differentiation (ALP activity) do not possess any direct relationship with the nanotube topography.
Nano-modified surfaces Further, by comprehending the cellular responses with the variation in the extent of anatase phase (XRD
study), surface roughness and wettability of nanotube surfaces, we established that the variation in
cellular responses is a combinatorial effect of surface topography and other physico-chemical properties
of TiO2 nanotubes. This study will help in developing bioactive implant surfaces modified with TiO2
nanotubes, for clinical application.
© 2018 Elsevier B.V. All rights reserved.

1. Introduction There is a rising amount of data pertaining to the improved biolog-

Titanium and its alloys have been widely used in orthopaedic
implants because of their superior mechanical properties, low
elastic modulus, excellent corrosion resistance and good biocom-
patibility. However, they fail to bond directly to bone tissue in vivo
because of their bio-inert nature. In the past decade, a number
research groups have focused on the surface modification of Tita-
nium based implants to improve the bioactivity of the implant
surface and its subsequent osteointegration. Recently, it has been
observed that the nanostructured TiO2 especially nanotubular TiO2
is remarkably more bioactive in comparison to the bulk Titanium
[1]. TiO2 is a stable oxide of Titanium and often found as a native
layer, that forms spontaneously when Titanium is exposed to ambi-
ent air at room temperature [2]. The passive TiO2 layer itself imparts
biocompatibility and corrosion resistance to the Titanium surface.
∗ Corresponding author.
E-mail address: amitb79@nitrkl.ac.in (A. Biswas).
ical performance of Titanium implants having surfaces modified
with TiO2 nanotubes [3–7]. So far, significant attempts have been
made to optimize the morphology of the TiO2 nanotubes to improve
the biological function of the material [8,9]. Although these reports
individually highlighted the influence of a single process parameter
on nanotube morphology, yet there is a dearth of a comprehensive
report describing the combined effect of oxidation time and applied
voltage on nanotube morphology.
It has long been established that formation of TiO2 nanotubes
enhances the adhesion of cells, in comparison to a bare Ti sur-
face. Das et al. showed the preferential attachment of OPC1 human
osteoblast on nanotube surface compared to that on bare Ti sur-
face [10]. It is well documented that morphological attributes of
the TiO2 nanotubes such as tube diameter and tube wall thickness
have a vital influence on the formation of focal adhesion points
which in turn strongly affects the cell adhesion and proliferation.
Different research groups have also reported the improvement of
initial cell adhesion on small diameter nanotubes (≤30 nm), using
various cell types [3,11,12]. Interestingly, this firmly established

https://doi.org/10.1016/j.apsusc.2018.01.160
0169-4332/© 2018 Elsevier B.V. All rights reserved.
 S. Saha et al. / Applied Surface Science 449 (2018) 152–165
view point has been contradicted in a recent report published in
2015. The authors suggested that nanotopography impaired the
formation of focal contacts and FAK expression and proposed that
this may be due to the reduction of total contact surface available.
Nonetheless, the high proliferation of cells was observed at an early
stage and it was attributed to the existence of some other potential
compensation signalling pathways [13].
A critical analysis revealed that the available literature majorly
focused on the impact of nanotube diameters on cell response, how-
ever, the combined influence of TiO2 nanotube surface topography
related features such as surface roughness, wettability, crystallinity
of the oxide layer, protein adsorption has not been considered at the
time of analysis. These surface characteristics often determine the
process of osteo-integration of an implant by influencing a num-
ber of cellular events such as cell adhesion, migration proliferation,
differentiation and apoptosis [14–17].
This set of literature clearly suggests that a comprehensive anal-
ysis is essential to decipher the relationship between the process
parameters, TiO2 nanotube morphology, surface characteristics of
nanoscale TiO2 layers and cellular responses. Keeping these per-
spectives in mind here we have conducted a detailed study on
anodic oxidation based surface modification of Ti6Al4V to find out
(i) the correlation between process parameters (oxidation time and
voltage) and the morphological attributes and surface character-
istics of TiO2 nanotube formed, and (ii) to decipher the combined
influence of nanotube structure (tube diameter and wall thickness)
and bulk properties of the surface modified with TiO2 nanotubes
layer on bone cell compatibility. Anodic oxidation was chosen as
a method of surface modification as it can produce highly orga-
nized nanostructures in a small time frame. Here, we have oxidized
the Ti6Al4V surfaces by varying the oxidation time and applied
voltage. The phase analysis of the oxide layer was carried out by
XRD. The morphology of the nanotubes (tube diameter, wall thick-
ness, and distribution) was inspected by SEM and further analysed
using image analyser. Surface characterization was performed by
a surface profiler and contact angle measurement. We further
studied the protein adsorption behaviour and biomineralization
in-vitro. Cellular responses were investigated using bone cell line
(MG-63). For this purpose, both quantitative and qualitative cell
adhesion studies were performed by Trypan blue assay and SEM,
focal adhesions were studied by immunostaining of vinculin. Fur-
ther, proliferation of MG63 cells was reported by MTT assay and the
differentiation status of the cells was investigated using Alkaline
Phosphatase assay (ALP).
2. Materials and methods
2.1. Ti6Al4V specimen preparation
Medical grade Ti6Al4V cold-rolled sheet with 0.90 mm thickness
was procured (Classic Alloys, India). It was sectioned into rectan-
gular blocks and polished using Silicon Carbide grit sheets (1/0,
2/0, 3/0, 4/0) to remove stains, scratches and the native oxide layer.
Thereafter, the samples were cloth polished, using alumina suspen-
sion. All samples were degreased and cleaned in ethanol for 30 min,
in an ultrasonic bath (APL digital).
2.2. Fabrication of TiO2 nanotubes
Anodic oxidation of the polished samples was carried out by
using a conventional two-electrode configuration with a DC power
supply (APLAB). Ti6Al4V was used as an anode and a graphite rod as
a cathode. The electrodes were kept apart in the electrolyte at a dis-
tance of 2 cm throughout the study. The electrolyte was composed
of 0.25% ammonium fluoride in 90% vol. ethylene glycol and 10%
153
vol. distilled water, as demonstrated by reference [4]. The fabrica-
tion of TiO2 nanotube was carried out with variation in oxidation
time and voltage around 30 V and 3 h respectively. The selected pro-
cess parameters taken for studies are given in Table 1 along with
the respective group names. After performing anodic oxidation, the
sample was detached from the anode and washed in ethanol in an
ultrasonic bath for 8 min. Further, all oxidized samples were heat
treated in a muffle furnace at 500 ◦ C for 3 h.
2.3. Phase analysis and morphological characterization
X-Ray Diffraction technique (Rigaku, D/MAXULTIMA) was used
to confirm the formation of oxide layer after anodic oxidation and
heat treatment. The surface morphology of anodized samples was
inspected by FESEM (Nova Nano Sem450, FEI).
2.4. Analysis of surface roughness and topography
The surface roughness of both polished and oxidized samples
was measured using a surface profiler (VeecoDektak 150). The
instrument generates data based on a single linear traverse of
a mechanical stylus, over the sample surface. Profiling was per-
formed at three locations spanning the entire oxidized region of
the Ti6Al4V samples. A length of 1000 ␮m was scanned for 80 s
in a single trace. Ra (Average Roughness) value was calculated by
considering the mean of average roughness values obtained after
three traces. Further, the topography of the polished substrate
and TiO2 nanotubes grown under three different voltage condi-
tions were characterized using an atomic force microscope (AFM),
VEECO diInnova, operating in contact mode, with a 0.01–0.025-
 cm antimony-(n)-doped Si tip (Bruker, model RESPA-20). The
average roughness (Ra) was determined using a scanning area of
3 × 3 ␮m with a scan rate of 0.5 Hz. Further, 3D image analysis and
roughness calculation was performed using WSxM 5.0 Develop 9.0
software.
2.5. Wettability study
To evaluate the wettability of oxidized surfaces, Contact angle
meter (Data Physics OCA20) was used. For the assessment of surface
hydrophilicity and hydrophobicity, the contact angle was measured
with respect to ultrapure water, at three different locations on each
sample. The contact angle for each drop was recorded by calculating
the mean of left and right angles. The average of three readings has
been reported. The polished specimen was used as a control.
2.6. Protein adsorption assay
The protein adsorption behaviour of the nanotubular TiO2 sur-
face in comparison to the polished Ti6Al4V specimen was evaluated
using Bradford assay. All the samples used in the study were of
equal size (0.5 × 1 cm2 ). Bovine serum albumin (BSA) and Lysozyme
were used as model proteins for the study. The assay was per-
formed using a method outlined elsewhere, with modifications
[18]. Briefly, samples were placed in 24 well polystyrene plate
and 1 ml aqueous protein solution (1 mg/ml) was added to each
well. Protein adsorption was carried out at 37 ◦ C for 24 h. Sub-
sequently, the supernatant was collected and the absorbance of
residual protein was determined using a spectrophotometer (Sys-
tronix, double beam spectrophotometer 2203) at 595 nm, using
Bradford assay. The corresponding protein concentration was eval-
uated using standard graphs of BSA and Lysozyme respectively. The
amount of protein adsorbed to the sample surface was calculated
by deducting the residual protein concentration from the initial
protein concentration.
154 S. Saha et al. / Applied Surface Science 449 (2018) 152–165
Table 1
Process parameters for conducting anodic oxidation of Ti6Al4V alloy to form TiO2 nanotubes.
Oxidation parameters
Treatment time (h) Voltage (V) Group name Voltage (V)
3 20 TN20 3 30
4 20 TN20 4 30
5 20 TN20 5 30
2.7. In vitro bioactivity in SBF
The in vitro bioactivity of the polished Ti6Al4V substrate as well
as the oxidized samples at different voltages (TN20 5, TN30 5 and
TN40 5) were evaluated in terms of their apatite-forming ability by
immersing them in simulated body fluid (SBF) at 37 ◦ C for a period
of 2 days and 7 days respectively. SBF was prepared according to
the procedure outlined in reference [19]. Briefly, 0.185 g CaCl2 ,
0.4 g KCl, 0.06 g KH2 PO4 , 0.1 g MgCl2 ·6H2 O, 0.1 g MgSO4 ·7H2 O, 8.0 g
NaCl, 0.35 g NaHCO3 , 0.48 g Na2 HPO4 and 1.00 g d-glucose were dis-
solved in 1 l of mili-Q water. The pH was adjusted to 7.2–7.6 with
1 M HCl. Thereafter, the samples were immersed into polystyrene
sterile containers with 50 ml of the as prepared SBF solution. After
the aforementioned time period, the samples were taken out and
washed with distilled water and dried at 37 ◦ C. The distribution,
morphology and structure of the formed apatite were observed
by SEM and the elemental composition was determined by EDX
(Oxford instruments).
2.8. Cell culture
MG63 human osteosarcoma cell lines (NCCS, Pune, India) were
cultured and maintained in a 5% CO2 incubator, at 37 ◦ C. The cell
line was cultured using complete media composed of high glucose
DMEM (Himedia), 10% Fetal bovine serum (Gibco) and 1% antibiotic
solution (Himedia). The cells were subcultured when roughly 70%
confluence was observed. A split ratio of 1:3 was maintained for
subculturing and the third passage was used for the cytocompati-
billity studies.
2.9. Quantitative estimation of cell adhesion
To study the initial adhesion of MG63 cells on polished and oxi-
dized substrates, 1.7 × 104 cells were seeded on each sample and
incubated at 37 ◦ C, in a 5% CO2 incubator. After 4 h of culture, the
samples were transferred to a fresh plate and cells were detached
from the substrate by trypsinization, followed by its neutraliza-
tion with complete media. The cell suspensions were centrifuged
at 4000 rpm for 5 min and re-suspended in 20 ␮l of PBS. Thereafter,
5 ␮l of Trypan blue was added to each cell suspension. The num-
ber of adherent live cells were counted under a microscope using
a haemocytometer and the Trypan blue stained dead cells were
excluded.
2.10. Cell adhesion and morphology by SEM analysis
MG63 cells were seeded over the experimental specimens and
cultured for 24 h at 37 ◦ C in a 5% CO2 incubator. Thereafter, super-
natant media was discarded and samples were washed with SEM
buffer (pH = 7.4). Further, cells were treated with 2.5% SEM grade
glutaraldehyde and incubated for 2.5 h. After, the incubation period
samples were again washed with SEM buffer followed by a serial
dehydration of cells with increasing concentration of ethanol (30%,
50%, 70%, 90%, 95% and 100%). Finally, the cell seeded samples were
sputter coated with gold for 180 s and morphology of the adhered
cells was observed via Scanning Electron Microscope (FEI Quanta,
250 FEG).
Heat treatment
Group name Voltage (V) Group name
TN30 3 40 TN40 3 500 ◦ C
TN30 4 40 TN40 4 for
TN30 5 40 TN40 5 3h
2.11. Study of the distribution of vinculin and F-actin
To evaluate the morphology and distribution of vinculin, 1 × 104
MG63 cells were seeded on polished Ti6Al4V as well as on all
oxidized substrates. Immunocytochemical staining was performed
after 60 h of culture. The cells were fixed in 4% paraformaldehyde
and incubated at 37 ◦ C for 15 min. The cells were permeabilized
with 0.25% TritonX/PBS (15 min) and washed thrice in PBS. To
block nonspecific binding, the cells were further incubated in 2%
BSA/PBS for 1 h. Thereafter, incubation was carried out in Anti-
Vinculin mouse monoclonal antibody (Abcam, 1:200) at 37 ◦ C for
1hr. After rinsing thrice with PBS, the secondary antibody goat anti-
mouse IgG H&L tagged with DyLight 488 (Abcam, 1:250) was added
for 1 h at room temperature. TRITC phalloidin was added to stain
the actin cytoskeleton and Hoescht was used to stain the nucleus.
The cells were washed repeatedly before observing under Confocal
Microscope (TCS-SP8, Leica, Germany) [20]. Further, to analyse the
spreading behaviour of the cells on each sample type, the elonga-
tion factor, area of each adhered cell and the shape of cells in terms
of roundness were estimated using Image J software. The length of
major and the minor axis of minimum 15 cells over each sample,
were measured. Here, elongation factor is the ratio of the major
axis and minor axis.
2.12. Cell proliferation assay
The cells were seeded on each specimen with a density of
1 × 104 cells. After culturing for 3 and 7 days, the prolifera-
tion of cells in oxidized and polished samples was determined
using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-
mide (MTT) assay. The study was performed to analyse the
influence of various surface characteristics on the proliferation
of MG63 cells, in comparison to that on a bare Ti6Al4V surface.
5 mg/ml MTT reagent was dissolved in DMEM in 1:10 ratio and
added to each sample at the end of the culture period. After an
incubation period of 2.5 h at 37 ◦ C, the MTT reagent was pumped
out and dimethyl sulfoxide was added to solubilize the purple
coloured formazan. Further, the absorbance was recorded using a
spectrophotometer, at 595 nm.
2.13. Alkaline phosphatase assay
For differentiation assay, 0.5 × 104 cells were seeded on all sub-
strates and cultured for 7 days. Thereafter, all cell seeded samples
were washed twice with PBS followed by its treatment with Tri-
tonX (0.1% in PBS) for 20 min. The cell suspension was centrifuged
to remove cell debris and p-nitrophenol phosphate based substrate
(pNPP, Sigma) was added. Incubation was carried out at 37 ◦ C for
2 h. Absorbance was measured at 405 nm using an ELISA plate
reader (Perkin Elmer 2030) and relative ALP levels were reported
[21].
2.14. Statistical analysis
All experiments were performed in triplicate. Statistical dif-
ferences were calculated using ANOVA single factor analysis of
 S. Saha et al. / Applied Surface Science 449 (2018) 152–165
Fig. 1. XRD data of polished Ti6Al4V and substrates anodized at 20 V, 30 V and 40 V after (A) 3 h ((TN(20)3), (TN(30)3), (TN(40)3)), (B) 4 h ((TN(20)4), (TN(30)4), (TN(40)4))
and (C) 5 h ((TN(20)5), (TN(30)5), (TN(40)5)) of oxidation respectively.
variance at p < 0.05. The statistical analysis data was represented
as the mean ± standard deviation (SD).
3. Results
3.1. XRD analysis
Fig. 1(A–C) shows the XRD profile of the polished and the oxi-
dized Ti6Al4V substrates for 3 h, 4 h and 5 h in three different
groups. The peaks were compared with JCPDS data cards 84-1286
(TiO2 , Anatase), 73-1765 (TiO2, Rutile) and 44-1294 (Alpha Tita-
nium) and 44-1288 (Beta Titanium). It was observed that the
polished Ti6Al4V substrate showed characteristic peaks of alpha-Ti
at 35.6◦ , 38.8◦ , 40.8◦ , 53.4◦ , 63.6◦ , 70.9◦ , 77.0◦ 2␪ and beta Ti peak at
38.2◦ 2␪. On the other hand, all the anodized substrates showed the
presence of oxide peaks corresponding to anatase phase of TiO2 ,
at 25.6◦ , 48.1◦ , 54.2◦ 2␪ along with the alpha and beta Ti peaks.
Among all the samples, the maximum intensity of anatase peak
was found in the samples oxidized at 40 V (as evident from the
profile of TN(40)3, TN(40)4 and TN(40)5). It was observed that the
crystallinity of the anatase phase increased with increase in voltage
(Fig. 1C). However, no direct proportionality was observed between
oxidation time and peak intensity of anatase phase. Hence, it may
be concluded from the XRD that the TiO2 formed at higher voltage
is rich in crystalline phase of anatase. No peaks corresponding to
Rutile TiO2 were found in any of the aforementioned samples.
3.2. Morphological characterization of TiO2 nanotubes
It was observed that at 20 V, spiky needle-like structures were
formed when oxidation was carried out for 3 h (Fig. 2A(a, d)).
Disordered, rudimentary nanotubular structures with very small
diameters were formed, only after 4 h of anodization. While,
TN20 3, TN20 4 showed needle-like cross-sectional morphology,
TN20 5 demonstrated the formation of a tubular pattern with highly
rippled tube walls (Fig. 2A(d–f)). The diameter of TN20 4 and TN20 5
was observed to be in the range of 25–55 nm (Fig. 2D). The cross-
sectional lengths of nanostructures in all three samples varied in the
range of 0.25–0.35 ␮m (Fig. 2A(d–f)). Further, Fig. 2B(a–c) shows
the successful formation of nanotubular structures, at 30 V, irre-
spective of the variation in oxidation time. The morphology of
oxide layers, formed under these conditions were found to be sim-
ilar. All nanostructures had roughly circular openings and average
155
diameters in the range of 55–80, 63–80 and 70–90 nm for TN30 3,
TN30 4 and TN30 5 respectively (Fig. 2B(a–c) and D). In addition,
an increase in nanotube length was observed, with an increase
in voltage. All three samples showed a clear tubular morphology
across the cross section and the height of the tubes varied in the
range of 0.60–0.75 ␮m, 0.50–0.70 ␮m and 0.5–0.75 ␮m for TN30 3,
TN30 4 and TN30 5 respectively (Fig. 2B(d–f)). Fig. 2B(c) shows the
accumulation of flaky precipitates over the nanotube entrance. The
presence of these precipitates can be attributed to the formation
of Ti(OH)4 as a result of spontaneous dissolution and ejection of
Ti4+ from the nanopores [3]. Similarly, nanotubes were also formed
when oxidation was carried out at 40 V for 3 h, 4 h, 5 h Fig. 2C(a–f)
respectively. It is evident from Fig. 2C(a) that after oxidation for
3 h at 40 V, tubular morphology was partially attained. The cross-
sectional image shows the successful formation of smooth and
long tubular structure (Fig. 2C(d)). However, the oxidized layer
formed at 40 V after 3 h of anodization appeared as a pore like
sheet instead of a tubular array, as the individual tube openings
were not distinctly visible. On increasing the oxidation time upto
4 h or more, distinct tube walls appeared and a noticeable ring-
like entrance morphology was formed (Fig. 2C(d, f)). Also, the flaky
precipitates of Ti(OH)4 were no longer visible. The tube diame-
ters varied significantly among the nanotubes formed after 3 h, 4 h
and 5 h of oxidation at 40 V (Fig. 2D). The substantial increase in
tube diameter can be attributed to the increased F− attack on TiO2 .
The small diameter pits merged to give rise to tubes with larger
diameters, on increasing the anodization period [22]. Addition-
ally, the average cross-sectional length was observed to be in the
range of 1.0–1.5 ␮m for TN40 3 as shown in Fig. 2C(d). On the other
hand, it is evident in Fig. 2C(e, f) that the tube lengths dropped to
0.65–0.85 ␮m and 0.60–0.78 ␮m on increasing the oxidation time
to 4 h and 5 h respectively. From Fig. 2D, it is possible to say that the
tube diameter is majorly influenced by the oxidation voltage and
it increases with an increase in voltage. Fig. 2E shows the variation
in tube wall thickness among all the samples oxidized at differ-
ent voltage and time interval. From the figure, it is evident that
the wall thickness is lowest at 20 V. However, on further increas-
ing the applied voltage from 30 V to 40 V, no considerable change
in wall thickness was observed. Also, no noticeable change in tube
wall thickness was recorded with variation in time, at a constant
voltage. Fig. 2F represents the distribution of nanotubes with open
pores per ␮m2 . It is apparent that TN20 5 exhibited the presence of
a maximum number of densely packed small diameter nanotubes
156 S. Saha et al. / Applied Surface Science 449 (2018) 152–165

Fig. 2. FESEM images of the Ti6Al4V samples oxidized at (A) 20 V, (B) 30 V and (C) 40 V respectively. Each panel shows the top view (a–c) and the cross-sectional view (d–f) of
the nanostructures formed after 3 h, 4 h and 5 h of oxidation. (D) Effect of applied voltage on the average diameter of the nanotubes. (E) Wall thickness of the nanotubes formed
in each sample type. (F) Number of nanotubes formed per ␮m2 , in each oxidized sample. Here N is the number of nanotubes measured for each sample type using Image J
software. * No nanotubular morphology was observed in TN(20)3. ** Nanotubes with fused walls were observed in the top view of TN(40)5. ## Rudimentary nanotubular
structures were formed.

i.e. ≈57 nanotubes/␮m2 (Fig. 2D). On increasing the applied volt- cipitates of Ti(OH)2, as explained earlier in this section. Therefore,
age to 30 V, the diameter and wall thickness of individual nanotube minimum number of nanotubes with open pores were observed
increased, therefore the number of nanotubes per ␮m2 dropped. In in TN30 5 i.e. ≈17 nanotubes/␮m2 . All substrates oxidized at 40 V
the case of TN30 5, most of the tube pores are blocked with pre- showed nearly 39–40 nanotubes/␮m2 .
 S. Saha et al. / Applied Surface Science 449 (2018) 152–165 157

Fig. 3. (A) Measurement of Surface roughness of polished Ti6Al4V substrate (as control) and various oxidized Ti6Al4V samples by surface profiler. (B) AFM images of (a)
Polished Ti6Al4V and substrates oxidized at different voltage (b) TN(20)5, (c) TN(30)5, (d) TN(40)5: for 5 h (C) Graphical representation of water contact angles (◦ ) obtained
by a Contact angle meter. Data shown are mean ± SD from triplicates.

3.3. Surface roughness 3.4. Wettability

The roughness of the samples was evaluated in terms of Ra val-
ues. It is clear, that the roughness of all oxidized samples is greater
than the polished sample (Fig. 3A). However, those oxidized at 20 V
and 30 V (except TN30 5) have comparable roughness values and lie
in the range of ≈64 nm to ≈77 nm. On the other hand, a consider-
able increase in roughness has been observed in samples TN40 4 and
TN40 5 i.e. 120 ± 20 nm and 171 ± 16 nm respectively. This obser-
vation can be correlated with the increase in diameter of TN40 4
and TN40 5 respectively. Further, AFM analysis revealed that differ-
ent surface topographies were obtained for the polished (control)
substrate as well as the substrate oxidized at 20 V, 30 V and 40 V
(Fig. 3B). The Ra was found to be 13 nm, 40.53 nm, 41.64 nm and
48 nm for polished, TN(20)5, TN(30)5 and TN(40)5 respectively. It is
clear, that all oxidized substrates showed higher surface roughness
than the polished surface and the Ra increased with the increase
in applied voltage (also higher nanotube diameter). This trend is
in consistence with the surface roughness values obtained from
surface profiler.
The contact angle values showed that the wettability of the
surface increased with the increase in voltage (Fig. 3C). The max-
imum wettability was reported in the case of TN40 4 and TN40 5,
(measured contact angles were 43.70◦ ± 2.83 and 42.43◦ ± 2.59
respectively) in comparison to the cloth polished surface (contact
angle 89.56◦ ± 1.97). Also, TN40 4 and TN40 5 exhibited maximum
surface roughness among all the samples which could be related to
the drop in contact angle in the same samples.
3.5. Protein adsorption
The protein adsorption study showed that overall the adsorp-
tion of lysozyme was relatively higher than bovine serum
albumin irrespective of the surface topography (Fig. 4). The pol-
ished surface adsorbed 0.37 ± 0.001 mg/ml of lysozyme. Among
the oxidized substrates maximum adsorption of lysozyme was
reported in the case of TN20 3 (0.70 ± 0.014 mg/ml) followed
by TN20 5 and TN20 4 (0.66 ± 0.014 mg/ml and 0.62 ± 0.034 mg/ml
respectively) while minimum adsorption was observed in case
158 S. Saha et al. / Applied Surface Science 449 (2018) 152–165
Fig. 4. Estimation of protein adsorption (mg/ml) in the polished Ti6Al4V substrate
(as control) and oxidized Ti6Al4V samples, using Lysozyme & BSA as model proteins.
Data shown are mean ± SD from triplicates.
of TN40 4 and TN40 5 (0.45 ± 0.007 mg/ml and 0.47 ± 0.00 mg/ml
respectively). On the other hand, when the samples were
incubated with BSA all surfaces adsorbed more than the pol-
ished Ti6Al4V surface, however, roughly similar adsorption
behaviour was observed in all the oxidized samples irrespec-
tive of applied voltage, oxidation time, and nanotube diameter.
The polished substrate showed 0.22 ± 0.013 mg/ml of BSA adsorp-
tion whereas all oxidized substrates adsorbed in the range of
0.28 ± 0.026 mg/ml–0.37 ± 0.034 mg/ml.
3.6. In vitro bioactivity in SBF
The qualitative analysis of mineralization in SBF showed that
after two days of immersion, white agglomerates were sparingly
formed and sparsely scattered on all the surfaces (Fig. 5). On the
other hand, the deposits covered a very large fraction of the surface
after seven days of immersion and the polished surface showed
a lower degree of apatite deposition in comparison to the oxi-
dized samples. TN40 5 exhibited maximum deposition of the apatite
clusters (Fig. 5B(g)). There is a decrease in mineralization on the
substrate oxidized at a lower voltage. The EDS spectrum of TN30 5
and TN40 5 (Fig. 5B(g, h)) showed no peak corresponding to Ti which
indicated the formation of a thick apatite deposit over the sub-
strate. The Ca/P ratio varied between 1.55–1.60 in all the samples
after 7 days of immersion, which further confirmed the formation
of biological apatite [23].
3.7. Quantitative estimation of cell adhesion
Quantitative estimation of cell adhesion (Fig. 6) showed
that number of cells adhered on the polished surface was
0.11 × 104 ± 0.015. The extent of cell adhesion on TN20 3, TN20 4,
TN40 3 and TN40 4 was found to be similar to that on the
polished surface. Among all the samples cell adhesion was max-
imum in TN40 5 (0.21 × 104 ± 0.017 × 104 cells) followed by TN30 5
(0.2 × 104 ± 0.023 × 104 cells). It is evident that at all applied volt-
ages (20 V, 30 V and 40 V), cell adhesion was maximum in the
samples which were oxidized for the maximum time duration (i.e.
TN20 5, TN30 5 and TN40 5). In the previous section, it was noticed that
the same substrates showed increased surface roughness. Thus, the
results indicate that initial adhesion of MG63 cells has a corre-
lation with the surface roughness, in the case of the nanotubular
substrates.
3.8. Cell adhesion and morphology by SEM analysis
The SEM study (Fig. 7) revealed that all the surfaces support
cell adhesion. No distinct difference was observed in the cell mor-
phology among the different groups. Only in the case of polished
Ti6Al4V, cells appeared completely flattened. Detailed investi-
gation revealed that on the oxidized surface cells were able to
establish stable filopodia. Such filopodia were not found in the
cells cultured on the polished surface. A visual inspection gives an
impression that the cells cultured on nanotube diameters ranging
from ∼70 nm to ∼100 nm (Fig. 7(e–j)) have relatively more num-
ber of filopodia extensions. Among the large diameter nanotubes
maximum spread of cells was observed in TN40 4 and TN40 5.
3.9. Study of the distribution of vinculin and F-actin
To compare the pattern of adhesion of MG63 cells cultured over
polished and various nanotubular surfaces after 60 h of culture,
the cells were stained for actin filaments (red), vinculin (green).
The immunocytochemistry of vinculin revealed that the cells cul-
tured on polished surface showed vinculin condensation at distinct
points in the periphery along with diffused cytoplasmic vinculin
(Fig. 8). On the other hand, only cytoplasmic vinculin was evi-
dent in all the nanotubular surfaces. In addition, the distribution
of cytoplasmic vinculin showed a diffused pattern when cells were
cultured over nanotubular surfaces. The coexistence of vinculin and
F-actin was observed only in the cells cultured on the polished
surface.
Further, on comparing the cell area on all the surfaces, it was
found that no significant difference was evident among the cells
cultured on various surfaces. Also, the cells on each substrate
showed a wide deviation in the occupied cell area (Fig. 8B). It is evi-
dent from the confocal images (Fig. 8C) that there was no statistical
significance in the cells cultured on the polished Ti6Al4V surface
and that grown on the surface anodized at 20 V, in terms of cell
elongation. The stress fibers were more prominently formed among
TN20 3, TN20 4 and TN20 5 in comparison to the polished Ti6Al4V sur-
face. All substrates oxidized at 30 V showed nearly the same extent
of cell elongation. A slight increase in cell elongation was observed
in case of cells cultured on TN40 4 in comparison to the other sub-
strates. The increase in cell elongation in the case of TN40 4 was
accompanied by a drop in roundness as evident in Fig. 8D.
3.10. Cell proliferation assay
Fig. 9 presents the cell proliferation index of MG63 cells on
polished and oxidized Ti6Al4V substrates, using MTT assay. It is
clear from the graph that all the oxidized samples show signifi-
cantly higher proliferation of MG63 cells compared to the polished
surface, after 3 and 7 days of culture. All samples oxidized at 20 V
(TN20 3, TN20 4, TN20 5) show a similar pattern of proliferation, irre-
spective of the duration of oxidation. After 7 days of culture, the
nanotubes with larger diameter (TN(30) & TN(40) groups) in gen-
eral showed better proliferation of MG63 cells than the small
diameter nanotubes (TN(20) group). Further, highest proliferation
index was exhibited by the cells grown on TN(40) group after both
3 and 7 days of culture. Roughly two fold increase in cell prolifera-
tion was found among the cells grown on samples oxidized at 40 V,
compared to the polished substrate. This could be attributed to the
larger diameter of TiO2 nanotubes at 40 V which allows better inter-
action of the cells with the culture medium, gasses and nutrients
[24].
 S. Saha et al. / Applied Surface Science 449 (2018) 152–165 159

Fig. 5. The in vitro bioactivity of polished Ti6Al4V and TN(20)5, TN(30)5, TN(40)5 in terms of the apatite-forming ability, after immersion in SBF, using SEM and EDS. (a–d)
and (e–h) represents the distribution of calcium and phosphorus deposits along with the respective EDS spectrum, after (A) two days and (B) seven days of immersion in
SBF, respectively.

3.11. Alkaline phosphatase expression 2H2 O ⇔ 2O2 + 4H+ + 4e− (2)

To investigate the differentiation status of MG63 cells over the
polished and oxidized substrate after 7 days of culture, the expres-
sion of differentiation marker ALP was studied. It is evident from
Fig. 10 that compared to the polished surface all oxidized surfaces
showed a lowered ALP expression. The ALP expression dropped
with the increase in applied voltage and minimum activity was
found in the cells cultured on substrates oxidized at 40 V.
4. Discussions
The entire process of anodic oxidation comprises of three major
reactions which includes dissolution of metals from the substrate
to form metallic ions which are soluble in the electrolyte, the reac-
tion of the metal ions with the oxygen ions derived from the water
molecules presents in the electrolyte, results in the formation of
metallic oxide at the substrate electrolyte interface. Finally, the for-
mation of the tubular/porous morphology of oxide layer occurs as a
result of oxide layer deposition and oxide dissolution under precise
treatment conditions [25].
Ti ⇔ Ti2+ + 2e−
2+ 2− −
Ti + 2O ⇔ TiO2 + 2e (3)
TiO2 + 6F − + 4H+ → [TiF6 ]2 + 2H2 O (4)
The electrolyte consists of anions such as oxide, hydroxide and
fluoride ions. On applying electric field, the migration of these
ions is initiated and directed towards the anode. Ionic migration
depends upon the discharge of ions at the oxide-electrolyte sur-
face, their size, charge and concentration in the electrolyte. It has
been reported that fluoride ions migrate faster in comparison to
oxide ions [26,27] and as a result of the rapid migration, fluoride
ions accumulate at the interface between the oxide layer and the
substrate material, in the form of TiF4 [28,29]. The thickness of the
layer depends upon the migration rate and concentration of the ion
in the electrolyte. However, the TiF4 deposited at the top surface of
interporous region are most exposed to the action of electrolytes
and get dissolved more rapidly than the inner layers deposited at
the bottom of the pores [30]. Consequently interstices are created
between nanotubes with varying depths depending upon the rate
of oxide formation and dissolution. Finally, the small diameter pits
merge to give rise to tubes with larger diameters, on increasing the
anodization period [22]. Regonini and Clemens reported that the
(1) variation in oxidation time from 10 min to 120 min at a constant
160 S. Saha et al. / Applied Surface Science 449 (2018) 152–165
Fig. 6. Quantitative estimation of cell adhesion on polished and oxidized substrates
after 4 h of culture. The graph represents mean ± SD. * represents a significant differ-
ence in adhesion behaviour among the cells cultured over polished Ti6Al4V (control)
and oxidized, (p < 0.05).
voltage of 30 V resulted in an increase in the nanotube length from
≈0.8 ␮m to ≈2.9 ␮m [31]. Chernozem et al. reported similar trend at
60 V [32]. On the other hand, numerous studies have also reported
a significant enlargement of nanotube diameter with an increase in
voltage. Yoriya et al. showed that the nanotube diameter increased
from ≈60 nm to ≈118 nm on increasing the voltage from 25 V to
40 V, after 24 h of oxidation [33]. Likewise, another group found a
direct correlation between applied voltage and nanotube diameter
[3]. In the current study, different nanotubular morphologies were
synthesized on the Ti6Al4V surface by varying both oxidation time
and voltage. From the review of previous literature, it was found
Fig. 7. SEM images of MG63 cellular morphology after 24 h of culture on (a) polished Ti6Al4V substrate (control) and the oxidized Ti6Al4V samples (b) TN(20)3, (c) TN(20)4,
(d) TN(20)5, (e) TN(30)3, (f) TN(30)4, (g) TN(30)5, (h) TN(40)3, (i) TN(40)4 and (j) TN(40)5 at 1000×. The insets represent the adhesion of cells on respective substrate at
higher magnification of 20,000×.
that Wang et al. already achieved titanium nanotubes on titanium
at 30 V for 3 h, for implant application [34]. Therefore, in the present
study, the synthesis and optimization of TiO2 nanotube morphol-
ogy, was carried out by selecting the voltage and oxidation time
around the aforementioned conditions.
Earlier, reports have documented that among the morphological
attributes of nanotubes, the tube diameter plays a vital role in the
formation of focal adhesion points which in turn strongly guides
cellular adhesion and proliferation. Few studies have emphasized
the importance of small diameter nanotubes (<30 nm) for promot-
ing cell response [3,11,12], while other studies proved nanotubes
with a larger diameter are more favourable for cell adhesion and
proliferation [24]. However, the present study revealed a contra-
dictory scenario. It is evident from the quantitative cell adhesion
studies that initial adhesion on the polished Ti6Al4V substrate was
higher than many nanotubular surfaces. Also, vinculin immuno-
cytochemistry images revealed that maximum peripheral focal
contact points were found in the polished Ti6Al4V substrate. On
the other hand, the cells grown on nanotubular surfaces showed
the presence of cytoplasmic vinculin but the focal contact points
were impaired by the nanotopography. It is interesting to note that
the nanotubular surfaces encouraged the formation of long filopo-
dia projections (evident from SEM images) which helped the cells
to anchor on the surface, however, the formation of focal adhesion
complexes was inhibited. In addition, regardless of the inhibition of
focal contacts, a significant increase in proliferation of MG63 cells
was observed on all nanotubular surfaces compared to the polished
substrate, after 3 and 7 days of culture. This contradiction in the
results and the existence of a possible compensatory (RhoA/FAK)
pathway has been supported by Zhang et al. in a recent literature
[13]. Hence, it is logical to assume that it is the resultant of various
TiO2 nanotube surface topography and surface chemistry related
features (such as surface roughness, wettability, crystallinity of the
 S. Saha et al. / Applied Surface Science 449 (2018) 152–165 161

Fig. 8. (A) Immunostaining of MG63 cells after 60 h of culture on (a1) polished Ti6Al4V substrate (control) and the oxidized Ti6Al4V samples (b) TN(20)3, (c) TN(20)4, (d)
TN(20)5, (e) TN(30)3, (f) TN(30)4, (g) TN(30)5, (h) TN(40),3 (i) TN(40)4 and (j) TN(40)5. The cytoskeletal arrangement was studied using TRITC phalloidin (Red), the nucleus
was stained with Hoescht (Blue) and vinculin localization was studied using anti-vinculin antibodies (green dots or green-stained area). (a2) Vinculin immunostaining of
cells cultured on the polished substrate. The arrows represent the focal complexes formed at the periphery of cells. (B) The cell area (␮m2 ) represents the spread of cells over
each type of substrate (C) Estimation of elongation factor of the cells cultured on various substrates. It was calculated as the ratio of the major axis and minor axis. (D) The
shape of the cells has been denoted in terms of its roundness. Estimation of cell elongation, area and roundness were carried out using Image J software. (For interpretation
of the references to colour in this figure legend, the reader is referred to the web version of this article.)

oxide layer and protein adsorption) that triggers a compensatory

pathway which rescues the cell proliferation irrespective of the
impairment of focal complexes on the nanotubular surfaces. There-
fore, in our consecutive section, we have discussed the relationship
between the cellular response and other physicochemical proper-
ties. However, the investigation and discussion of the molecular
pathway that rescues the proliferation of cells, on the nanotubes
synthesized at a higher voltage, is beyond the scope of the current
study.
Immediately after the implantation of a biomaterial, blood
(plasma) proteins come in contact with the biomaterial surface
which is followed by rapid adsorption of proteins. The adsorption
of plasma proteins further influences the cellular interaction with
the implant [35]. Specifically, in the case of bone-to-implant attach-
ment, plasma proteins such as fibronectin enhance chemotaxis and
provide cues for focal adhesion of bone cells through integrin recep-
tors [36]. Therefore, the protein adsorption behaviour of the as
Fig. 9. Cell proliferation index using MTT assay. MG63 cells cultured on polished
fabricated nanostructured surfaces was studied. Fig. 4 shows that
(control) and oxidized Ti6Al4V surface for 3 and 7 days respectively. The graph repre-
the adsorption of Lysozyme was greater than adsorption of BSA, on sents mean ± SD. * & # represents a significant difference between polished Ti6Al4V
all surfaces. This could be attributed to the net positive and net neg- (control) and various oxidized surfaces on Day 3 and Day 7 respectively, (p < 0.05).
ative charges on lysozyme and BSA respectively, at neutral pH [37].
162 S. Saha et al. / Applied Surface Science 449 (2018) 152–165
Fig. 10. Relative ALP activity of MG63 cells after 7 days of culture on polished
(control) and oxidized substrates. The graph represents mean ± SD. ** represents
a significant difference in ALP activities between the cells cultured over polished
Ti6Al4V (control) and substrates oxidized at 30 V and 40 V after Day 7, (p < 0.05).
It is already established that the TiO2 nanotube surface has a small
negative charge owing to the terminal hydroxyl groups present
in TiO2 [37]. Thus, larger adsorption of lysozyme can be justified
due to its stronger electrostatic interaction with TiO2 surface. Fur-
ther, higher adsorption of lysozyme was observed on the surfaces
oxidized at 20 V in comparison to the TN40 4 and TN40 5 (Average
diameters have been presented in Fig. 2D). In a study by Oh et al. it
was observed that in an aqueous solution, albumin protein folded
into small sized aggregates which anchored on the mouth of small
diameter nanotubes with greater ease, rather than on larger diam-
eter tubes with bigger open space leading to lower adsorption of
proteins on the surface with a larger diameter [11]. A similar trend
was observed in the current study when lysozyme was used as a
model protein. Although the concentration of proteins adsorbed to
the larger diameter nanotubes is lower, yet the unfolded confor-
mation of the protein is likely to assist the cellular interaction as
the binding sites in the non- aggregated proteins would be more
easily available for the cell based epitopes [38]. However, further
experimental evidence is essential to validate the rationale.
Further, the surface roughness is important and in many cases,
the high surface roughness of a material leads to enhanced cell
adhesion, proliferation and ALP activity. Earlier researchers have
reported that surface roughness governs the initial adhesion of
cells through a complex biological system which includes pro-
tein adsorption, surface receptor binding and contact guidance
phenomenon [39–41]. In this study, an increase in the average
roughness Ra was reported on increasing the oxidation time to
5 h (TN20 5, TN30 5, TN40 5), keeping the applied voltage constant.
Also, the cell adhesion (Fig. 6) increased with the increase in surface
roughness among the nanotubular surfaces oxidized at a constant
applied voltage. This direct correlation between surface roughness
and cell adhesion has also been reported by other research groups
[40,42].
Surface hydrophilicity is another deciding factor in determining
cell adhesion and proliferation. Hydrophilicity of the TiO2 surface
is related to the density of surface hydroxyl groups. These surface
hydroxyl groups interact with water molecules through hydro-
gen bonds, resulting in enhanced hydrophilicity of the surface. The
anatase phase of TiO2 is associated with hydroxyl groups. Thus,
increase in anatase content (ref. Fig. 1) increases the hydroxyl con-
centration resulting in enhanced wettability at 40 V [43]. According
to the observations in the current study, an increase in wettability is
accompanied by a rise in proliferation index, at 40 V. Similar obser-
vations have been documented in other recent studies. According to
Moravec et al., these favourable influence of crystal structure on cell
colonization and cell phenotypic maturation could be attributed to
the appropriate orientation of hydroxyl groups. Thus, it is reason-
able to infer that changes in surface chemistry affect the surface
hydrophilicity which impacts the cellular response [44]. Although
larger diameter nanotubes showed better proliferation of MG63
cells than the small diameter nanotubes yet, another contradiction
has emerged in this study. It is noteworthy that the average diam-
eter of TN40 3 (51.08 nm) is lower than that of TN30 3 (68.50 nm),
however, the proliferation of cells on TN40 3 is higher than that on
TN30 3. Furthermore, in alkaline phosphatase assay, it was observed
that the cells cultured on nanotube surface expressed lower levels
of ALP in comparison to the polished surface.
On correlating the results obtained from MTT assay and ALP
assay it was revealed that the substrates with higher cell prolifer-
ative potential induce lower expression of differentiation marker
and vice versa. It is already established that proliferation and dif-
ferentiation generally maintains an inverse relationship [45–47].
Therefore, it can be inferred that the cells cultured on the substrates
oxidized at a higher voltage (40 V) stays in the proliferative phase
rather than undergoing differentiation. However, it is important to
mention that MG63 is a matured osteoblast with limited differen-
tiation potential, therefore, this result is a mere indication of the
differentiation potential of TiO2 nanotube surfaces.
Additionally, it is known, that surface chemistry also influences
the apatite forming ability of the surface [48]. The in vitro bioactiv-
ity study indicates the ability of a substrate to form apatite when it
comes in contact with a physiological fluid. The apatite layer assists
the implant material in bonding strongly with the living bone tis-
sue and ensures successful integration of the implant at the site of
injury [49]. Li et al. established that the presence of abundant OH−
groups as well as negative surface charge induces apatite formation
in a biological fluid [50]. Fig. 5 suggests that TN40 5 showed the max-
imum apatite forming ability. It is clear from the XRD profile that
TN40 5 has a higher intensity of anatase phase than TN20 5, TN30 5 and
the polished surface. The increased concentration of OH− groups
attract more Ca+2 from SBF, rendering a positive charge to the
surface. Subsequently, negatively charged phosphate reacts with
calcium, gradually resulting in the enhanced formation of apatite
[37,51].
The role of the physicochemical properties on the overall cell
response has been summarized in Table 2. Based on these obser-
vations, it is rational to suggest that cell response is guided by the
combinatorial impact of surface chemistry and surface topography
related features and contrary to the popular belief, it cannot be
directed by a single morphological feature of the nanotubes.
5. Conclusions
The present study successfully elaborated the influence of
anodization parameters such as oxidation time and applied volt-
age over the formation TiO2 nanotube morphology and topography.
Through this study, the focus has been drawn towards a major
dichotomy that exists in the understanding of cellular response
towards TiO2 nanotubes. The contradiction has been addressed at
two levels. Firstly, it has been proved that a single morphologi-
cal aspect of the nanotubes is not sufficient to direct a particular
cell response. Secondly, it has been pointed out that the promotion
of cellular proliferation is not entirely reliant on the formation of
focal complexes. Further, it was noted that cell adhesion increased
with the roughness of the nanotubular surface. An improvement
in the proliferation of MG63 cells was observed, with the increase
in anatase content and wettability. Although, the nanotubes sup-
ported the proliferation of cells yet the expression of differentiation
marker (ALP) was poor. Keeping this limitation in mind, it is ratio-
Table 2
Effect of various physico-chemical properties on cell response.

Cell response

Physicochemical Number of cells Cell morphology and Focal complex Cell elongation Cell area Cell proliferation Cell differentiation
properties adhered filopodia formation

Intensity of anatase No direct relationship Increase in anatase No direct relationship Slight increase in No significant Increase in anatase content Drop in relative ALP

S. Saha et al. / Applied Surface Science 449 (2018) 152–165

phase observed content results in more observed elongation with increase in difference among results in enhanced expression with increase in
number of filopodia anatase content various samples proliferation of cells anatase content
Tube density No direct relationship No direct relationship No direct relationship No direct relationship Increase in cell area No direct relationship No direct relationship
observed observed observed observed with increase in tube observed observed
density
Tube diameter No direct relationship Appearance of more No focal complex on Slight increase in Lowered cell area at No direct relationship No direct relationship
observed no. of filopodia in nanotubular surface. elongation with highest tube diameter observed observed
samples with at higher Diffused cytoplasmic increase in diameter observed. (High range
tube diameter vinculin over all of deviation in all
surfaces. samples)
Surface roughness More adhesion of cells No significant No direct relationship No direct relationship No direct relationship Surfaces with higher The smooth polished
on nanotubes with difference in cell observed observed observed roughness showed Ti6Al4V surface showed
greater roughness morphology among relatively better higher ALP expression
various substrates proliferation of cells. compared to the rough
nanotubular surface.
No direct relationship No direct relationship
could be established could be established
among various
nanotubular surfaces
Wettability No direct relationship Appearance of No direct relationship Slight increase in No direct relationship Higher proliferation of cells No direct relationship
observed filopodia with Increase observed elongation with increase in observed in samples with greater could be established
in wettability wettability wettability
Protein adsorption No direct relationship No direct relationship No direct relationship Among the nanotubes an No direct relationship Among the nanotubes, No direct relationship
could be established could be established could be established inverse trend was observed observed surfaces with lower could be established
between protein lysozyme adsorption
adsorption and cell showed better proliferation
elongation

163
164 S. Saha et al. / Applied Surface Science 449 (2018) 152–165
nal to suggest that functionalization of nanotubes with a suitable
osteogenic molecule would improve both proliferative and dif-
ferentiation potential of the TiO2 nanotubes. In conclusion, it is
expected that this study would be helpful in the future to develop
appropriate surface patterning and functionalization of TiO2 nano-
tubes for a new generation of bio-implants.
Acknowledgements
This work was supported by Department of Biotechnology,
Government of India, New Delhi [BT/01/COE/09/13DT]; Centre of
Excellence in orthopaedic tissue engineering and rehabilitation
[F.No.5-6/2013-TS VII] and MHRD, Government of India, New Delhi.
References
[1] K.S. Brammer, C.J. Frandsen, S. Jin, TiO2 nanotubes for bone regeneration,
Trends Biotechnol. 30 (2012) 315–322, http://dx.doi.org/10.1016/j.tibtech.
2012.02.005.
[2] E. Gemelli, N.H.a. Camargo, Oxidation kinetics of commercially pure titanium,
Matéria (Rio Janeiro) 12 (2007) 525–531, http://dx.doi.org/10.1590/S1517-
70762007000300014.
[3] M.-Y. Lan, C.-P. Liu, H.-H. Huang, J.-K. Chang, S.-W. Lee, Diameter-sensitive
biocompatibility of anodic TiO2 nanotubes treated with supercritical CO2
fluid, Nanoscale Res. Lett. 8 (2013) 150, http://dx.doi.org/10.1186/1556-276X-
8-150.
[4] Y. Wang, C. Wen, P. Hodgson, Y. Li, Biocompatibility of TiO2 nanotubes with
different topographies, J. Biomed. Mater. Res.-Part A 102 (2014) 743–751,
http://dx.doi.org/10.1002/jbm.a.34738.
[5] M.P. Neupane, I.S. Park, T.S. Bae, H.K. Yi, F. Watari, M.H. Lee, Biocompatibility
of TiO 2 nanotubes fabricated on Ti using different surfactant additives in
electrolyte, Mater. Chem. Phys. 134 (2012) 536–541, http://dx.doi.org/10.
1016/j.matchemphys.2012.03.029.
[6] F. Nasirpouri, I. Yousefi, E. Moslehifard, J. Khalil-Allafi, Tuning surface
morphology and crystallinity of anodic TiO2 nanotubes and their response to
biomimetic bone growth for implant applications, Surf. Coat. Technol. 315
(2017) 163–171, http://dx.doi.org/10.1016/j.surfcoat.2017.02.006.
[7] D. Khudhair, H. Amani Hamedani, J. Gaburro, S. Shafei, S. Nahavandi, H.
Garmestani, A. Bhatti, Enhancement of electro-chemical properties of TiO2
nanotubes for biological interfacing, Mater. Sci. Eng. C 77 (2017) 111–120,
http://dx.doi.org/10.1016/j.msec.2017.03.112.
[8] M. Mansoorianfar, M. Tavoosi, R. Mozafarinia, A. Ghasemi, A.
Doostmohammadi, Preparation and characterization of TiO2 nanotube arrays
on Ti6Al4V surface for enhancement of cell treatment, Surf. Coat. Technol. 321
(2017) 409–415, http://dx.doi.org/10.1016/j.surfcoat.2017.05.016.
[9] L. Xu, Kai ge Lv, W. Yu, Effect of TiO2 nanotube layers thickness on
periodontal ligament cells, Dentistry 6 (2016), http://dx.doi.org/10.4172/
2161-1122.1000369.
[10] K. Das, S. Bose, A. Bandyopadhyay, TiO2 nanotubes on Ti: influence of
nanoscale morphology on bone cell-materials interaction, J. Biomed. Mater.
Res. Part A 90A (2009) 225–237, http://dx.doi.org/10.1002/jbm.a.32088.
[11] S. Oh, K.S. Brammer, Y.S. Li, D. Teng, A.J. Engler, S. Chien, S. Jin, Stem cell fate
dictated solely by altered nanotube dimension, Proc. Natl. Acad. Sci. U. S. A.
106 (2009) 2130–2135, http://dx.doi.org/10.1073/pnas.0813200106.
[12] J. Park, S. Bauer, K. Von Der Mark, P. Schmuki, Nanosize and vitality: TiO2
nanotube diameter directs cell fate, Nano Lett. 7 (2007) 1686–1691, http://dx.
doi.org/10.1021/nl070678d.
[13] H. Zhang, S. Yang, N. Masako, D.J. Lee, L.F. Cooper, C.-C. Ko, Proliferation of
preosteoblasts on TiO2 nanotubes is FAK/RhoA related, RSC Adv. 5 (2015)
38117–38124, http://dx.doi.org/10.1039/C4RA16803H.
[14] A. Zareidoost, M. Yousefpour, B. Ghaseme, A. Amanzadeh, The relationship of
surface roughness and cell response of chemical surface modification of
titanium, J. Mater. Sci. Mater. Med. 23 (2013) 1479–1488, http://dx.doi.org/
10.1007/s10856-012-4611-9.
[15] M.-H. Kim, Y. Sawada, M. Taya, M. Kino-Oka, Influence of surface topography
on the human epithelial cell response to micropatterned substrates with
convex and concave architectures, J. Biol. Eng. 8 (2014) 13, http://dx.doi.org/
10.1186/1754-1611-8-13.
[16] K. Metavarayuth, P. Sitasuwan, X. Zhao, Y. Lin, Q. Wang, Influence of surface
topographical cues on the differentiation of mesenchymal stem cells in vitro,
ACS Biomater. Sci. Eng. 2 (2016) 142–151, http://dx.doi.org/10.1021/
acsbiomaterials.5b00377.
[17] L. Feller, Y. Jadwat, R.A.G. Khammissa, R. Meyerov, I. Schechter, J. Lemmer,
Cellular responses evoked by different surface characteristics of intraosseous
titanium implants, BioMed Res. Int. 2015 (2015), http://dx.doi.org/10.1155/
2015/171945.
[18] S. Kulanthaivel, B. Roy, T. Agarwal, S. Giri, K. Pramanik, K. Pal, S.S. Ray, T.K.
Maiti, I. Banerjee, Cobalt doped proangiogenic hydroxyapatite for bone tissue
engineering application, Mater. Sci. Eng. C 58 (2016) 648–658, http://dx.doi.
org/10.1016/j.msec.2015.08.052.
[19] L. Mohan, P. Dilli Babu, P. Kumar, C. Anandan, Influence of zirconium doping
on the growth of apatite and corrosion behavior of DLC-coated titanium alloy
Ti-13Nb-13Zr, Surf. Interface Anal. 45 (2013) 1785–1791, http://dx.doi.org/10.
1002/sia.5323.
[20] P. Gupta, S.N. Gautham Hari Narayana, U. Kasiviswanathan, T. Agarwal, K.
Senthilguru, D. Mukhopadhyay, K. Pal, S. Giri, T.K. Maiti, I. Banerjee, Substrate
stiffness does affect the fate of human keratinocytes, RSC Adv. 6 (2016)
3539–3551, http://dx.doi.org/10.1039/C5RA19947F.
[21] F. Pati, H. Kalita, B. Adhikari, S. Dhara, Osteoblastic cellular responses on
ionically crosslinked chitosan-tripolyphosphate fibrous 3-D mesh scaffolds, J.
Biomed. Mater. Res. Part A 101A (2013) 2526–2537, http://dx.doi.org/10.
1002/jbm.a.34559.
[22] S. Yoriya, Electrolyte factors influencing separated pore growth of anodic TiO2
nanotube arrays, Int. J. Electrochem. Sci. 11 (2016) 9088–9099, http://dx.doi.
org/10.20964/2016.11.69.
[23] A. Ogose, T. Hotta, H. Kawashima, N. Kondo, W. Gu, T. Kamura, N. Endo,
Comparison of hydroxyapatite and beta tricalcium phosphate as bone
substitutes after excision of bone tumors, J. Biomed. Mater. Res.-Part B Appl.
Biomater. 72 (2005) 94–101, http://dx.doi.org/10.1002/jbm.b.30136.
[24] R. Zhang, H. Wu, J. Ni, C. Zhao, Y. Chen, C. Zheng, X. Zhang, Guided
proliferation and bone-forming functionality on highly ordered large
diameter TiO2 nanotube arrays, Mater. Sci. Eng. C 53 (2015) 272–279, http://
dx.doi.org/10.1016/j.msec.2015.04.046.
[25] F. Keller, M.S. Hunter, D.L. Robinson, Structural features of oxide coatings on
aluminum, J. Electrochem. Soc. 100 (1953) 411, http://dx.doi.org/10.1149/1.
2781142.
[26] K. Shimizu, K. Kobayashi, The migration of fluoride ions in growing anodic
oxide films on tantalum, J. Electrochem. Soc. 144 (1997) 418–423.
[27] H. Habazaki, K. Fushimi, K. Shimizu, P. Skeldon, G.E. Thompson, Fast migration
of fluoride ions in growing anodic titanium oxide, Electrochem. Commun. 9
(2007) 1222–1227, http://dx.doi.org/10.1016/j.elecom.2006.12.023.
[28] S.P. Albu, A. Ghicov, S. Aldabergenova, P. Drechsel, D. LeClere, G.E. Thompson,
J.M. Macak, P. Schmuki, Formation of double-walled TiO2 nanotubes and
robust anatase membranes, Adv. Mater. 20 (2008) 4135–4139, http://dx.doi.
org/10.1002/adma.200801189.
[29] S. Berger, R. Hahn, P. Roy, P. Schmuki, Self-organized TiO2 nanotubes: factors
affecting their morphology and properties, Phys. Status Solidi Basic Res. 247
(2010) 2424–2435, http://dx.doi.org/10.1002/pssb.201046373.
[30] Z. Su, W. Zhou, F. Jiang, M. Hong, Anodic formation of nanoporous and
nanotubular metal oxides, J. Mater. Chem. 22 (2012) 535–544, http://dx.doi.
org/10.1039/C1JM13338A.
[31] D. Regonini, F.J. Clemens, Anodized TiO2 nanotubes: effect of anodizing time
on film length, morphology and photoelectrochemical properties, Mater. Lett.
142 (2015) 97–101, http://dx.doi.org/10.1016/j.matlet.2014.11.145.
[32] R.V. Chernozem, M.A. Surmeneva, R.A. Surmenev, Influence of anodization
time and voltage on the parameters of TiO2 nanotubes, IOP Conf. Ser. Mater.
Sci. Eng. 116 (2016) 12025, http://dx.doi.org/10.1088/1757-899X/116/1/
012025.
[33] S. Yoriya, W. Kittimeteeworakul, N. Punprasert, Effect of anodization
parameters on morphologies of TiO 2 nanotube arrays and their surface
properties, J. Chem. Chem. Eng. 6 (2012) 686–691, http://dx.doi.org/10.1016/j.
jmst.2015.07.012.
[34] Y. Wang, C. Wen, P. Hodgson, Y. Li, Biocompatibility of TiO2 nanotubes with
different topographies, J. Biomed. Mater. Res.-Part A 102 (2014) 743–751,
http://dx.doi.org/10.1002/jbm.a.34738.
[35] S. Cei, D. Karapetsa, E. Aleo, F. Graziani, Protein adsorption on a laser-modified
titanium implant surface, Implant Dent. (2015) 1–2, http://dx.doi.org/10.
1097/id.0000000000000214.
[36] R. Jimbo, M. Ivarsson, A. Koskela, Y.-T. Sul, C.B. Johansson, Protein adsorption
to surface chemistry and crystal structure modification of titanium surfaces, J.
Oral Maxillofac. Res. 1 (2010) e3, http://dx.doi.org/10.5037/jomr.2010.1303.
[37] K.C. Popat, M. Eltgroth, T.J. LaTempa, C.A. Grimes, T.A. Desai, Titania
nanotubes: a novel platform for drug-eluting coatings for medical implants?
Small 3 (2007) 1878–1881, http://dx.doi.org/10.1002/smll.200700412.
[38] W.Q. Yu, X.Q. Jiang, F.Q. Zhang, L. Xu, The effect of anatase TiO2 nanotube
layers on MC3T3-E1 preosteoblast adhesion, proliferation, and differentiation,
J. Biomed. Mater. Res.-Part A 94 (2010) 1012–1022, http://dx.doi.org/10.1002/
jbm.a.32687.
[39] D. Deligianni, N. Katsala, S. Ladas, D. Sotiropoulou, J. Amedee, Y. Missirlis,
Effect of surface roughness of the titanium alloy Ti-6Al-4V on human bone
marrow cell response and on protein adsorption, Biomaterials 22 (2001)
1241–1251, http://dx.doi.org/10.1016/S0142-9612(00)00274-X.
[40] H.H. Huang, C.T. Ho, T.H. Lee, T.L. Lee, K.K. Liao, F.L. Chen, Effect of surface
roughness of ground titanium on initial cell adhesion, Biomol. Eng. 21 (2004)
93–97, http://dx.doi.org/10.1016/j.bioeng.2004.05.001.
[41] J.I. Rosales-Leal, M.A. Rodríguez-Valverde, G. Mazzaglia, P.J.
Ramón-Torregrosa, L. Díaz-Rodríguez, O. García-Martínez, M.
Vallecillo-Capilla, C. Ruiz, M.A. Cabrerizo-Vílchez, Effect of roughness,
wettability and morphology of engineered titanium surfaces on
osteoblast-like cell adhesion, Colloids Surf. A Physicochem. Eng. Asp. 365
(2010) 222–229, http://dx.doi.org/10.1016/j.colsurfa.2009.12.017.
[42] J.S.P. da Silva, S.C. Amico, A.O.N. Rodrigues, C.A.G. Barboza, C. Alves, A.T. Croci,
Osteoblastlike cell adhesion on titanium surfaces modified by plasma
nitriding, Int. J. Oral Maxillofac. Implants 26 (2011) 237–244 http://www.
ncbi.nlm.nih.gov/pubmed/21483875.
 S. Saha et al. / Applied Surface Science 449 (2018) 152–165
[43] L. Yang, M. Zhang, S. Shi, J. Lv, X. Song, G. He, Z. Sun, Effect of annealing
temperature on wettability of TiO(2) nanotube array films, Nanoscale Res.
Lett. 9 (2014) 621, http://dx.doi.org/10.1186/1556-276X-9-621.
[44] H. Moravec, M. Vandrovcova, K. Chotova, J. Fojt, E. Pruchova, L. Joska, L.
Bacakova, Cell interaction with modified nanotubes formed on titanium alloy
Ti-6Al-4V, Mater. Sci. Eng. C 65 (2016) 313–322, http://dx.doi.org/10.1016/j.
msec.2016.04.037.
[45] S. Ruijtenberg, S. van den Heuvel, Coordinating cell proliferation and
differentiation: antagonism between cell cycle regulators and cell
type-specific gene expression, Cell Cycle 15 (2016) 196–212, http://dx.doi.
org/10.1080/15384101.2015.1120925.
[46] T.A. Owen, M. Aronow, V. Shalhoub, L.M. Barone, L. Wilming, M.S. Tassinari,
M.B. Kennedy, S. Pockwinse, J.B. Lian, G.S. Stein, Progressive development of
the rat osteoblast phenotype in vitro: reciprocal relationships in expression of
genes associated with osteoblast proliferation and differentiation during
formation of the bone extracellular matrix, J. Cell. Physiol. 143 (1990)
420–430, http://dx.doi.org/10.1002/jcp.1041430304.
[47] G.S. STEIN, J.B. LIAN, Molecular mechanisms mediating
proliferation/differentiation interrelationships during progressive
165
development of the osteoblast phenotype, Endocr. Rev. 14 (1993) 424–442,
http://dx.doi.org/10.1210/edrv-14-4-424.
[48] C.Y. Guo, J.P. Matinlinna, A.T.H. Tang, Effects of surface charges on dental
implants: past, present, and future, Int. J. Biomater. 2012 (2012) 1–5, http://
dx.doi.org/10.1155/2012/381535.
[49] H.M. Kim, F. Miyaji, T. Kokubo, T. Nakamura, Bonding strength of bonelike
apatite layer to Ti metal substrate, J. Biomed. Mater. Res. 38 (1997) 121–127,
http://dx.doi.org/10.1002/(SICI)1097-4636(199722)38:2<121:AID-JBM6>3.0.
CO;2-S.
[50] P. Li, C. Ohtsuki, T. Kokubo, K. Nakanishi, N. Soga, K. de Groot, The role of
hydrated silica, titania, and alumina in inducing apatite on implants, J.
Biomed. Mater. Res. 28 (1994) 7–15, http://dx.doi.org/10.1002/jbm.
820280103.
[51] D.K. Pattanayak, T. Kawai, T. Matsushita, H. Takadama, T. Nakamura, T.
Kokubo, Effect of HCl concentrations on apatite-forming ability of NaOH-HCl-
and heat-treated titanium metal, J. Mater. Sci. Mater. Med. 20 (2009)
2401–2411, http://dx.doi.org/10.1007/s10856-009-3815-0.