Biomaterials 22 (2001) 1467}1474

The in#uence of titania/hydroxyapatite composite coatings

on in vitro osteoblasts behaviour

P.A. Ramires*, A. Romito, F. Cosentino, E. Milella

PASTIS-CNRSM, Biomaterials Unit, S.S. 7 Appia Km 714, 72100 Brindisi, Italy

Received 30 April 2000; accepted 3 August 2000

Abstract

The biocompatibility of titania/hydroxyapatite (TiO/HA) composite coatings, at di!erent ratio obtained by sol}gel process, were

investigated studying the behaviour of human MG63 osteoblast-like cells. The biocompatibility was evaluated by means of
cytotoxicity and cytocompatibility tests. Cytotoxicity tests, i.e., neutral red (NR), MTT and kenacid blue (KB) assays, were performed
to assess the in#uence of the material extracts on lysosomes, mitochondria and cell proliferation, respectively. Cell proliferation, some
preliminary indications of cell morphology, alkaline phosphatase activity, collagen and osteocalcin production of MG63 cells,
cultured directly onto TiO/HA substrates, were evaluated. The results showed that these materials have no toxic e!ects. Cell growth

and morphology were similar on all the materials tested; on the contrary, alkaline-phosphatase-speci"c activity and collagen
production of osteoblasts cultured on TiO/HA coatings were signi"cantly higher than uncoated titanium and polystyrene of culture

plate and were in#uenced by chemical composition of the coatings. In particular, TiO/HA coating at 1:1 ratio (w/w) seems to

stimulate more than others the expression of some di!erentiation markers of osteoblastic phenotype. TiO/HA coatings resulted to be

bioactive owing to the presence of hydroxyl groups detected on their surface that promote the calcium and phosphate precipitation
and improve the interactions with osteoblastic cells.

2001 Elsevier Science Ltd. All rights reserved.

Keywords: Titania/hydroxyapatite coatings; Biocompatibility; Osteoblastic cells

1. Introduction

Critical to the long-term success of orthopaedic and

dental implants is the development of a stable direct
connection between bone and surface implant, which
must be structural and functional (osteointegration).

The establishment and maintenance of osteointegra-

tion depend on wound healing tissues, repair and re-
modelling. The tissue response to an implant involves
physical factors, depending on the implant design, sur-
face topography, and chemical factors associated with
the composition and structure of the material.

To improve a direct implant "xation to bone, several

strategies have been developed focusing the attention on
the surface of materials. Chemical modi"cations have
been realised by covalent attachment of an organic
monolayer anchored by a siloxane network [1] and
immobilising speci"c adhesive peptides, like arginine}
glycine}aspartic acid}serine (RGDS) [2]. Surface rough-

* Corresponding author. Fax: #[phone].

ness has been modi"ed by di!erent techniques, because it
has been demonstrated that the osteoblastic cells tend to
attach more rapidly to surfaces with a rougher micro-
topography [3,4] increasing the bone apposition [5,6].
Moreover, the implants have been coated with di!erent
materials like calcium phosphate ceramics, bioactive
glass [7}12], diamond-like carbon and amorphous C}N

"lm [13].

Many studies have been carried out to develop stable

hydroxyapatite coatings for their capacity to establish
a bond with tissues in vivo, and to promote rapid attach-
ment and cell growth in vitro [14}16].

Various techniques, such as plasma spray [17], ion-

beam-assisted deposition [18], radiofrequency magnet-
ron sputtering [19], have been used to produce coatings
on implants.

Some drawbacks have been noticed regarding the

long-term performance of the obtained coatings: coating
resorption, poor mechanical properties, high thickness,
non-homogeneity, lack of adherence [20].

Sol}gel processing represents an alternative approach

for the coating preparation with potential advantages,

0142-9612/01/$ - see front matter

2001 Elsevier Science Ltd. All rights reserved.

PII: S 0 1 4 2 - 9 6 1 2 ( 0 0 ) 0 0 2 6 9 - 6

such as higher purity and homogeneity, lower processing
temperatures, reduced thickness, simple and cheap
method of preparation. Moreover, materials prepared by
sol}gel process have shown to be more bioactive than
those with the same composition but prepared with dif-
ferent methods [21,22].

In our laboratories, we have prepared a composite

coating constituted of a titania (TiO) matrix encapsulat-

ing hydroxyapatite (HA) deposited onto titanium sub-
strate by sol}gel process in order to obtain a thin, rough
and adherent "lm onto titanium [23]. Surface analysis of
these coatings has shown the presence of hydroxyl
groups on their surface that could promote the nuclea-
tion of calcium and phosphate.

This paper reports the evaluation of the bioactivity

and biocompatibility of TiO/HA coatings prepared

at di!erent ratio (1:1, 2:1, 1:2). The biocompatibility
has been evaluated by means of cytotoxicity and
cytocompatibility tests. Level I cytotoxicity tests, i.e.,
neutral red (NR), MTT and kenacid blue (KB) assays,
have been performed using mouse 3T3 "broblasts
and human MG63 osteoblasts, to assess the in#uence of
the material extracts on lysosomes, mitochondria and
cell proliferation, respectively (indirect tests). The
cytocompatibility has been determined by studying the
behaviour of

MG63

osteoblasts cultured

directly

onto TiO/HA coatings (direct tests). Cell proliferation

and some preliminary indications of morphology, as well
as the expression of some biochemical parameters of
osteoblastic phenotype (alkaline phosphatase activity,
collagen

and

osteocalcin

production),

have

been

monitored.

2. Materials and methods

2.1. Materials andsample preparation

Commercially pure titanium (Goodfellow, Germany),

cut into pieces 20

;10 mm size, was used as substrate,

ultrasonically rinsed in acetone for 20 min, in 70%
ethanol solution for 20 min and then in distilled water for
15 min.

Titania (TiO) sol was prepared by mixing titanium

isopropoxide, acetyl acetone, nitric acid, n-propane alco-
hol and distilled water (solution A).

Hydroxyapatite (HA) powders (Fin Ceramica, Italy)

were added to anhydrous ethanol in the same ratio
(solution B).

Solutions A and B were mixed in di!erent ratio

(w/w): 2:1 (TiO/HA 0.5), 1:1 (TiO/HA 1) and 1:2 (TiO/

HA 2).

The titanium coatings were obtained by the dip tech-

nique at a speed of 15 cm/min and sinterized at 5003C for
30 min. The dip-coating process was repeated four times
for every sample.

2.2. Cell cultures

MG63 osteoblast-like cells (ATCC, USA), originally

isolated from a human osteosarcoma, and mouse "bro-
blast 3T3 cell line (ATCC, USA) were cultured in Dul-
becco Modi"ed Eagle's Medium (DMEM, Biowhittaker,
Belgium), containing penicillin/streptomycin (100/100 U),
amphotericin B (2.5

g/ml) and gentamycin (100 g/ml),

supplemented with 10% foetal calf serum (Mascia Bru-
nelli, Italy), and kept at 373C in an atmosphere of 5%
CO and 99% humidity. Media were changed every

3 days.

2.3. Biocompatibility studies

The toxicity level deriving from the use of chemical

solvents for the coatings preparation was investi-
gated through indirect tests. Taking into account that the
procedure for the coatings preparation is the same and
that the "nal response is not in#uenced by the di!erent
percentage of components, for the cytotoxicity evalu-
ation only the TiO/HA 1:1 sample was taken as an

example.

In cytotoxicity tests, the viability and metabolic activ-

ity of cells, previously incubated with material extracts,
was checked by the NR uptake assay, the KB binding
method, and the 3-(4,5-dimethilthiazol-2-yl)-2,5-diphenyl-
tetrazolium bromide (MTT) assay. NR is taken up into
the lysosomes by viable cells, MTT into the mitochon-
dria. The KB binding assay measures the total cell pro-
tein content of the test wells.

Extracts for indirect tests were obtained from materials

under standardised conditions (ISO 10993-5). The mate-
rials were immersed in a complete culture medium for
120 h, at 373C without agitation. The ratio between the
sample surface and the volume of the extraction vehicle
was 2 cm

/ml. A negative control was represented by the

extraction vehicle with no material and the positive con-
trol by the same medium containing a 0.1% phenol
solution. The pure extracts (100%) and the 50 and 20%
dilutions (prepared using the complete culture medium),
were added to cells, seeded in 96 multiwell plates 24 h
before. After 3 days of incubation, viability and metabolic
activity of cells were tested by the MTT, NR and KB
assays, respectively [24,25]. The dyes accumulated with-
in the cells were liberated and absorbance was measured
with a microplate spectrophotometer (Multiskan EX,
Labsystems, Finland).

MTT assay: the medium was replaced with 50

l/well

of MTT solution (1 mg/ml in culture medium without
phenol red). After 4 h of incubation at 373C the solution
was removed, 100

l/well of DMSO were added and after

10 min of slow shaking the absorbance was read at
540 nm.

NR assay: the medium was replaced with 100

l/well of

NR solution (0.01% in culture medium) and the plates

1468

P.A. Ramires et al. / Biomaterials 22 (2001) 1467}1474

were incubated for 3 h at 373C. Then the cells were rinsed
with PBS and 100

l/well of stain extraction solution (1%

glacial acetic acid, 50% ethanol, 49% distilled water)
were added. The plates were shaken for 10 min and the
absorbance was read at 540 nm.

KB assay: the cells were "xed with glutaraldehyde

(2.5% in PBS) and then stained with KB dye (0.4 mg/ml)
for 20 min. Then the wells were rinsed twice with a
washing solution (ethanol 0.1%, acetic acid 0.05%, distil-
led

water

0.85%).

The

stain

extraction

solution

(potassium acetate 1

M

in ethanol) was added and the

plates shaken for 20 min. The absorbance was read at
620 nm.

For the direct contact method, titanium substrates

coated with titania/hydroxyapatite (TiO/HA 0.5, 1 and

2) were located in a 6-multiwell plate and then seeded at
a cell density of 2.0

;10 cells/cm, in the complete cul-

ture medium containing sodium ascorbate (50

g/ml) and

-glycerolphosphate (10 m

M

). The uncoated titanium (Ti)

and the polystyrene standard culture plate (ps) were used
as controls. On every sample 200

l of a cell suspension

was applied with great care and the cells were allowed to
attach for 2 h to the underlying substrate, then 3.5 ml of
culture medium was also added carefully. The amount of
cells seeded on polystyrene control (1.9

;10 cells) was

such as to obtain the same cell density used for the
coatings.

After 10 days of incubation, cell proliferation was

evaluated by MTT assay, whereas the biochemical ana-
lyses were performed on cell lysates.

Cells were released from the culture surface by the

addition of trypsin/EDTA solution for 10 min. The
reaction was stopped by the addition of DMEM contain-
ing 10% serum. A second trypsinisation was performed
to ensure the complete cell detachment from the substra-
te. After centrifugation, cell pellet was resuspended in
1 ml of sterile distilled water, soni"ed for 10 s and sub-
jected to three freezing cycles at !803C for 20 min fol-
lowed by thawing at room temperature for 30 min to lyse
the cells.

Three replicates have been performed for each experi-

ment and the values of biochemical parameters obtained
by the assays next described were normalised per unit of
area of 1 cm

.

2.4. Alkaline phosphatase activity

Alkaline phosphatase (E.C. 3.1.3.1.) activity was evalu-

ated on the transformation of p-nitrophenylphosphate
into p-nitrophenol at 373C and pH 10.2 and the activity
calculated. The substrate, 100

l of 20 m

M

p-nitrophenyl-

phosphate in appropriate bu!er (Sigma, USA), was ad-
ded to 100

l of cell lysate. The mixture was incubated at

373C for 30 min. The enzyme kinetic has been evaluated
in the time, monitoring the absorbance change per min-
ute at 410 nm that corresponds to the p-nitrophenol

production. This value was compared with the values of
a standard series. The speci"c activity was calculated
referring to protein concentration of lysates. Protein con-
tent was determined by a commercially available
colorimetric assay (BioRad, Germany).

2.5. Collagen assay

Collagen assay is based on the binding of a dye, sirius

red F3BA (BDH, UK), to the triple helical collagen "bril
[26]. 50

l/well of cell lysate was dispersed into 96 multi-

well plates, incubated at 373C for 16 h in humidi"ed
atmosphere and then for 24 h in dried atmosphere. The
staining was performed with 100

l/well of 0.1% sirius

red F3BA in saturated picric acid for 1 h at room temper-
ature. The plates were rinsed "ve times with 200

l of

10 m

M

HCl, then washed with 0.1

M

NaOH and absorb-

ance was read at 540 nm. The optical density values were
compared to the standard curve of type I collagen
(Sigma, USA) samples.

2.6. Osteocalcin measurement

The production of osteocalcin by cultures in the cell

medium was measured by a commercially available en-
zymeimmunoassay (EIA, BioResearch Ireland Diagnos-
tics, Ireland) employing highly speci"c monoclonal
antibodies and peroxidase-labelled osteocalcin.

2.7. Calcium determination

The calcium concentration in the culture medium was

evaluated with a commercially available colorimetric as-
say (Sigma, USA).

2.8. Scanning electron microscopy

Samples were "xed with 2.5% glutaraldehyde in phos-

phate bu!er and dehydrated through a graded series of
acetone. Then the specimens were critical point dried
(Bal-Tech, Liechtenstein), sputter-coated with gold and
examined with a scanning electron microscope (Philips
XL 40 LaB6) at 5 kV acceleration voltage, equipped with
an energy dispersive spectrometer (EDS) EDAX DX4i.

2.9. Statistical analysis

Results were expressed as mean values$standard de-

viation for each group of samples. After the assessment of
signi"cant di!erences by one-way variance analysis
(ANOVA), di!erences among groups were established
with t-test analysis by a two population comparison.
Statistical signi"cance was considered at a probability
p(0.05.

P.A. Ramires et al. / Biomaterials 22 (2001) 1467}1474

1469

Fig. 1. Cytotoxicity evaluation of TiO/HA 1 coating using human

MG63 osteoblast-like cells at di!erent extract dilutions by means of
MTT, NR and KB assays. The absorbance values are proportional to
the number of viable cells. The negative control (control!) is repre-
sented by the extraction vehicle with no material and the positive
control (control#) by the same medium containing a 0.1% phenol
solution. Di!erence from control: **p(0.01; ***p(0.001. Di!erence
from control#: 3p(0.001.

Fig. 3. Osteoblast alkaline-phosphatase-speci"c activity after 10 days
of culture on TiO/HA coatings and controls, uncoated titanium (Ti)

and polystyrene (ps). The results are presented as ALP-speci"c activity,
expressed as nmol/min/mg protein, normalised per unit of area of 1 cm

and reported as the mean$SD. Di!erence from Ti: *p(0.05;
**p(0.01; ***p(0.001. Di!erence from ps: 3p(0.05.

Fig. 2. Evaluation of cell viability estimated by the MTT assay of
MG63 osteoblastic cells cultured on TiO/HA coatings after 10 days of

incubation, compared with controls represented by uncoated titanium
(Ti) and polystyrene (ps). The absorbance values, normalised per unit of
area of 1 cm

and reported as the mean$SD, are proportional to the

number of viable cells. The means are not signi"cantly di!erent.

3. Results

In the present work, the cytotoxicity tests of TiO/HA

samples were performed using both human MG63 os-
teoblast-like cells and mouse "broblast 3T3 cell line. The
results of neutral red, MTT and kenacid blue assays are
shown in Fig. 1. The absorbance values obtained for

material extracts resulted signi"cantly higher (p(0.001)
with respect to the positive control (phenol solution) and
comparable to the negative control (medium without
extracts). The extracts did not induce a cell viability
reduction and an inhibition of cell growth resulting to
have no toxic e!ects. No di!erence among the dilutions
of extracts was detected. However, the high concentrated
extract solution (100%) a!ected mitochondrial and ly-
sosomal activity. The results obtained from the cyto-
toxicity tests performed on both cell types were in good
agreement.

In the direct contact method, MG63 cells were seeded

directly onto materials and their viability and prolifer-
ation were assessed after 10 days of incubation by MTT
test (Fig. 2). A titanium substrate was used as control

1470

P.A. Ramires et al. / Biomaterials 22 (2001) 1467}1474

Fig. 5. Osteocalcin production by MG63 osteoblast-like cells during
culture on TiO/HA coatings and controls, uncoated titanium (Ti)

and polystyrene (ps), assayed using an ELISA kit described in Section 2.
The results are normalised per unit of area of 1 cm

and reported

as the

mean$SD. Di!erence

from

Ti:

*p(0.05. Di!erence

from ps: 3p(0.05.

Fig. 4. Collagen production by MG63 cells during culture on TiO/HA

coatings and controls, uncoated titanium (Ti) and polystyrene (ps), deter-
mined by the sirius red staining described in Section 2. The results are
normalised per unit of area of 1 cm

and reported as the mean$SD.

Di!erence from Ti: *p(0.05; **p(0.01. Di!erence from ps: 33p(0.01;

333

p(0.001. Di!erence from TiO/HA 0.5 and 1: AA p(0.01.

Fig. 6. Scanning electron microscopy images of MG63 osteoblast-like cells grown for 10 days on titanium (a), TiO/HA 0.5 (b), TiO/HA 1 (c) and

TiO/HA 2 (d) coatings. The osteoblasts with dorsal ru%es showed close to each other, connected by "lopodia and disposed in multilayers. Some

round cells with numerous blebs on their membrane were present. On the TiO/HA 2 (d) substrate, cells in mitotic division were observed.

Magni"cation 2000

;.

whereas polystyrene standard culture plate was used for
comparison. Statistical analysis revealed that there were
no signi"cant di!erences between the proliferation of
osteoblastic cells on the various materials.

Di!erences on cell behaviour were found considering

biochemical parameters of cell di!erentiation.

Alkaline-phosphatase-speci"c activity of osteoblasts

seeded on TiO/HA coatings was signi"cantly higher

than the controls (Fig. 3).

P.A. Ramires et al. / Biomaterials 22 (2001) 1467}1474

1471

Fig. 7. SEM images of TiO/HA 1 (a) and 0.5 (b) coatings after 7 days

of incubation in complete culture medium. The TiO/HA 1 sample

surface showed a needle-like structure constituted by calcium and
phosphate as resulted by EDS analysis, not observed in the TiO/HA

0.5 sample. Magni"cation 6000

;.

The collagen synthesis was a!ected by coatings com-

position (Fig. 4). Osteoblast cells grown on TiO/HA 0.5

and 1 samples showed a signi"cant increase in the col-
lagen production compared to uncoated titanium and
polystyrene. On the contrary TiO/HA 2 sample was

similar to titanium control.

The cells grown on TiO/HA 1 substrate showed also

a signi"cant increase in osteocalcin production com-
pared to controls (Fig. 5).

Scanning electron microscopy (Fig. 6) showed the os-

teoblasts with dorsal ru%es as close to each other, con-
nected by "lopodia and disposed in multilayers. Some
round cells with numerous blebs on their membrane
indicating a poor cell condition were observed on
TiO/HA coatings and on titanium control. On the

TiO/HA 2 substrate, cells in mitotic division were ob-

served. The osteoblasts cultured on TiO/HA 0.5 sample

showed a di!erent morphology having a round shape
and no dorsal ru%es.

At higher magni"cation, the SEM image (Fig. 7a) and

the EDS analysis of TiO/HA 1 coating surface after

7 days of incubation showed a needle-like structure con-
stituted by calcium and phosphate not observed in the
TiO/HA 0.5 sample (Fig. 7b). Similar needle-like crys-

tallites were found also in TiO/HA 2 coating. The evalu-

ation of the calcium concentration in the culture medium
revealed a decrease of this ion in the medium containing
titanium (7.43$1 mg/dl), TiO/HA 0.5 (7.1$0.6 mg/dl),

TiO/HA 1 (5.18$0.67mg/dl), and TiO/HA 2 (5.09$

0.46 mg/dl) coatings with respect to polystyrene control
(9.52$0.27 mg/dl). The statistically signi"cant (p(0.05)
decrease of calcium in the medium containing TiO/HA

1 and 2 coating compared with TiO/HA 0.5 could be

ascribed to its precipitation and could be the origin of the
needle-like structure formation on the TiO/HA 1 and

2 coating surface.

4. Discussion

The results of the cytotoxicity tests suggested that the

extracts obtained by TiO/HA samples do not a!ect cell

viability and proliferation, as shown by the total protein
content (KB assay), even if the more concentrated extract
solution seems to exert some damaging e!ects on
mitochondrial and lysosomal function (MTT and NR
assay, respectively). However, the absorbance values ob-
tained for diluted material extracts resulted signi"cantly
higher with respect to the positive control and compara-
ble to the negative control, indicating the absence of toxic
e!ects. The di!erent behaviour between concentrated
and diluted material extracts can be explained consider-
ing the decrease of calcium concentration detected in the
undiluted extracts. In fact, the pure extracts (100%) are
obtained after 5 days of sample incubation when a signi"-
cant decrease of this ion in the culture medium and the

calcium presence on the substrate surface were detected
as suggested by EDS analysis. Since the extract dilutions
of 50 and 20% are prepared using the fresh culture
medium, a partial restoration of ionic concentrations
occurs. Probably, this di!erent composition of the cul-
ture medium for undiluted and diluted extracts could be
responsible for the di!erent cell growth rate.

Some cytotoxicity tests reported in literature [27]

were performed preparing material extracts with a
ratio [surface area/extraction volume] of 1 cm

/ml. In

this study we used a ratio of 2 cm

/ml (undiluted extracts)

and the 50 and 20% dilutions did not produce toxic
e!ects.

When the osteoblastic cells were seeded directly onto

TiO/HA-coated titanium, the cells proliferation occur-

red equally well on all substrates, revealing no signi"cant
di!erences. SEM analysis showed a comparable cell mor-
phology on all tested samples except for TiO/HA 0.5

substrate, where round cells without ru%ed membranes
were observed.

1472

P.A. Ramires et al. / Biomaterials 22 (2001) 1467}1474

On the contrary, di!erences of cell behaviour were

evident considering the biochemical parameters of cell
di!erentiation. Cells grown on TiO/HA coatings

showed an elevated alkaline phosphatase activity and
collagen production, in comparison with cultures grown
on uncoated titanium and polystyrene. All coated sam-
ples exhibited the same cell behaviour, stimulating the
expression of some peculiar biochemical markers of os-
teoblasts di!erentiation. Moreover, the osteoblastic cells
grown on TiO/HA 1 coating showed a signi"cant in-

crease of osteocalcin production. This calcium-binding
polypeptide represents the peculiar marker of "nal di!er-
entiation of osteoblasts and plays a fundamental role in
bone remodelling.

The better performances of these coatings with res-

pect to uncoated titanium and polystyrene can be
explained with the characteristics of the chemical com-
position and the deposition technique that a!ect physical
properties.

In fact, as several authors have demonstrated [28,29],

the TiO coatings deposited by sol}gel process are bioac-

tive, improving the calcium and phosphate precipitation
onto the surface.

The presence of hydroxyl groups, such as Ti}OH,

detected on the coatings surface [23], could promote the
interactions with bone cells by providing the site for
calcium and phosphate nucleation [30,31].

The detected decrease of calcium concentration in the

medium containing TiO/HA coatings, the appearance

of a "ne structure with a needle-like morphology on the
coatings surface containing calcium and phosphorous, as
detected by EDS analysis, could be ascribed to a precipi-
tation of this ion. Similar needle-type crystallites with Ca
and P have been described in literature [32,33].

Further investigations are in progress in order to

deepen these results and to yield a complete interpreta-
tion of these events.

5. Conclusion

The study of biocompatibility of titania/hydroxyapa-

tite coatings of titanium, prepared by sol}gel process,
suggested that these materials were not cytotoxic and
allowed the di!erentiation of cells, stimulating the ex-
pression of

some

peculiar

osteoblast biochemical

markers: alkaline phosphatase activity, collagen and os-
teocalcin production. In particular the TiO/HA 1 coat-

ing seems to stimulate more than others samples the
increase of some di!erentiation markers of osteoblastic
phenotype.

The good performances of these coatings can be ex-

plained with the characteristics of the chemical composi-
tion and the deposition technique. The TiO/HA

coatings resulted to be bioactive thanks to the presence
of hydroxyl groups detected on their surface that pro-

mote the calcium and phosphate precipitation and im-
prove the interactions with osteoblastic cells.

Therefore, these biomaterials are very promising and

a better understanding of cell/biomaterial interactions
and mechanisms can help in the development of more
e!ective orthopaedic and dental implants.

Acknowledgements

The authors would like to thank Dr. M. Re for SEM

analysis. The research was carried out in the frame of the
Innovation Project P4 funded by the Ministry for the
University and Scienti"c and Technological Research
(MURST).

References

[1] Sukenik CN, Balachander N, Culp LA, Lewandowska K, Merritt

K. Modulation of cell adhesion by modi"cation of titanium surfa-
ces with covalently attached self-assembled monolayers. J Biomed
Mater Res 1990;24:1307}23.

[2] Dee KC, Rueger DC, Andersen TT, Bizios R. Conditions which

promote mineralization at the bone-implant interface: a model
in vitro study. Biomaterials 1996;17:209}15.

[3] Lincks J, Boyan BD, Blanchard CR, Lohmann CH, Liu Y, Coch-

ran DL, Dean DD, Schwartz Z. Response of MG63 osteoblast-
like cells to titanium and titanium alloy is dependent on surface
roughness and composition. Biomaterials 1998;19:2219}32.

[4] Martin JY, Schwartz Z, Hummert TW, SchraubDM, Simpson J,

Lankford J, Dean DD, Cochran DL, Boyan BD. E!ect of tita-
nium surface roughness on proliferation, di!erentiation, and pro-
tein synthesis of human osteoblast-like cells (MG63). J Biomed
Mater Res 1995;29:389}401.

[5] Jinno T, Goldberg VM, Davy D, Stevenson S. Osseointegration

of surface-blasted implants made of titanium alloy and cobalt-
chromium alloy in a rabbit intramedullary model. J Biomed
Mater Res 1998;42:20}9.

[6] Vercaigne S, Wolke JGC, Naert I, Jansen JA. The e!ect of

titanium plasma-sprayed implants on trabecular bone healing in
the goat. Biomaterials 1998;19:1093}9.

[7] Oliva A, Salerno A, Locardi B, Riccio V, Della Ragione F, Iardino

P, Zappia V. Behaviour of human osteoblasts cultured on bioac-
tive glass coatings. Biomaterials 1998;19:1019}25.

[8] Wheeler DL, Stokes KE, Hoellrich RG, Chamberland DL,

McLoughlin SW. E!ect of bioacitve glass particle size on osseous
regeneration of cancellous defects. J Biomed Mater Res 1998;
41:527}33.

[9] Barth E, Johansson C, Albrektsson T. Histologic comparison of

ceramic and titanium implants in cats. Int Oral Max Imp
1990;5:227}31.

[10] Hayashi K, Matsuguchi N, Uenoyama K, Kanemaru T, Sugioka

Y. Evaluation of metal implants coated with several types of
ceramics as biomaterials. J Biomed Mater Res 1989;23:1247}59.

[11] Vrouwenvelder WCA, Groot CG, de Groot K. Histological and

biochemical evaluation of osteoblasts cultured on bioactive glass,
hydroxylapatite, titanium alloy, and stainless steel. J Biomed
Mater Res 1993;27:465}75.

[12] Matsuda T, Davies JE. The in vitro response of osteoblasts to

bioactive glass. Biomaterials 1987;8:275}84.

[13] Du C, Su XW, Cui FZ, Zhu XD. Morphological behaviour of

osteoblasts on diamond-like carbon coating and amorphous C}N

"lm in organ culture. Biomaterials 1998;19:651}8.

P.A. Ramires et al. / Biomaterials 22 (2001) 1467}1474

1473

[14] De Lange GL, De Putter C, De Wijs FLJA. Histological and

ultrastructural appearance of the hydroxyapatite-bone interface.
J Biomed Mater Res 1990;24:829}45.

[15] Faucheux C, Bareille R, Rouais F, AmeHdeHe J, LiebendoKrfer A,

Dard M. Biocompatibility testing of a bovine hydroxy-
apatite ceramic material with the use of osteo-progenitor cells
isolated from human bone marrow. J Mater Sci Mater Med
1994;5:635}9.

[16] Bagambisa FB, Stom J, Schilli W. The interaction of osteogenic

cells with hydroxyapatite implant materials in vitro and in vivo.
JOMI 1990;5:217}26.

[17] De Groot K, Geesink RGT, Klein CPAT, Serekion P. Plasma-

sprayed coatings of hydroxylapatite. J Biomed Mater Res 1987;
21:1375}81.

[18] Cui FZ, Luo ZS, Feng Q. Highly adhesive hydroxyapatite coat-

ings on titanium alloy formed by ion beam assisted deposition.
J Mater Sci Mater Med 1997;8:403}5.

[19] Hulsho! JEG, van Dijk K, van der Waerden JPCM, Wolke JGC,

Ginsel LA, Jansen JA. Biological evaluation of the e!ect of mag-
netron sputtered Ca/P coatings on osteoblast-like cells in vitro.
J Biomed Mater Res 1995;29:967}75.

[20] Kangasniemi IMO, Verheyen CCPM, van der Velde EA, de

Groot K. In vivo tensile testing of #uoroapatite and hydroxylapa-
tite plasma-sprayed coatings. J Biomed Mater Res 1994;
28:563}72.

[21] Haddow DB, James PF, van Noot R. Characterization of sol}gel

surfaces for biomedical applications. J Mater Sci Mater Med
1996;7:255}60.

[22] Peltola T, Patsi M, Rahiala H, Kangasniemi I, Yli-Urpo A.

Calcium phosphate induction by sol}gel derived titania coatings
on titanium substrate in vivo. J Biomed Mater Res 1998;
48(3):504}10.

[23] Cosentino F, Licciulli A, Massaro C, Milella E. Prepa-

ration

and

characterization

of

titania/hydroxyapatite

composite coatings obtained by sol}gel process. Biomaterials,
in press.

[24] Ramires PA, Milella E. Biocompatibility of poly(vinyl alcohol)-

hyaluronic acid and poly(vinyl alcohol)-gellan membranes
crosslinked by glutaraldehyde vapours. J Mater Sci Mater Med,
accepted for publication.

[25] Sbarbati Del Guerra R, Lelli L, Tonelli C, Trombetta T,

Cascone MG, Taveri M, Narducci P, Giusti P. In vitro biocom-
patibility of #uorinated polyurethanes. J Mater Sci Mater Med
1994;5:452}6.

[26] Walsh BJ, Thornton SC, Penny R, Breit S. Microplate reader-

based quantitation of collagens. Anal Biochem 1992;203:187}90.

[27] MacNair R, Rodgers EH, MacDonald C, Wykman A, Goldie I,

Grant MH. The response of primary rat and human osteoblasts
and an immortalized rat osteoblast cell line to orthopaedic mate-
rials: comparative sensitivity of several toxicity indices. J Mater
Sci Mater Med 1997;8:105}11.

[28] Li P, De Groot K, Kokubo T. Bioactive Ca(PO)(OH)-TiO

composite coating prepared by sol}gel process. J Sol-Gel Sci
Technol 1996;7:27}34.

[29] Jokinen M, PaKtsi M, Rahiala H, Peltola T, Ritala M, Rosenholm

JB. In#uence of sol}gel-derived TiO and TiO-SiO "lms depos-

ited by dip-coating method. J Biomed Mater Res 1998;
42:295}302.

[30] Li P, Ohtsuki C, Kokubo T, Nakanish K, Soga N, de Groot K.

A role of hydrated silica, titania and alumina in forming biolo-
gically active bone-like apatite on implant. J Biomed Mater Res
1994;28:7}15.

[31] Li P M, de Groot K, Kokubo T. Bonelike hydroxyapatite induc-

tion by sol}gel derived titania coating on a titanium substrate.
J Am Ceram Soc 1994;77:1307}15.

[32] Izquierdo-Barba I, Salinas A J, Vallet-Regm` M. In vitro calcium

phosphate layer formation on sol}gel glasses of the CaO-SiO

system. J Biomed Mater Res 1999;47:243}50.

[33] Perez-Pariente J, Balas F, Roman J, Salinas A J, Vallet-Regm` M.

In#uence of composition and surface characteristics on the
in vitro bioactivity of SiO}CaO}PO}MgO sol}gel glasses.

J Biomed Mater Res 1999;47:170}5.

1474

P.A. Ramires et al. / Biomaterials 22 (2001) 1467}1474