Correlation between substratum roughness and wettability,
cell adhesion, and cell migration
´
M. Lampin,1,2 R. Warocquier-Clerout,1 C. Legris,1 M. Degrange,2 M. F. Sigot-Luizard1
1 ´
Laboratoire de Biologie Cellulaire Experimentale, Universite de Technologie de Compiegne, BP 529, 60206 Compiegne,
France
2 ´ ´ ´
Groupe de Recherche Biomateriaux, Faculte de Chirurgie Dentaire, Universite Pairs V, Paris, France
Received 14 March 1996; accepted 13 June 1996
Abstract: Cell adhesion and spreading of chick embryo vas-
cular and corneal explants grown on rough and smooth poly
(methyl methacrylate) (PMMA) were analyzed to test the
cell response specificity to substratum surface properties.
Different degrees of roughness were obtained by sand-
blasting PMMA with alumina grains. Hydrophilic and hy-
drophobic components of the surface free energy (SFE) were
calculated according to Good-van Oss’s model. Contact
angles were determined using a computerized angle meter.
The apolar component of the SFE, csLW, increased with a
slight roughness whereas the basic component, cs2, de-
creased. The acido-basic properties disappeared as
roughness increased. Incubation of PMMA in culture me-
dium, performed to test the influence of the biological envi-
ronment, allowed surface adsorption of medium proteins
INTRODUCTION
Poly(methyl methacrylate) (PMMA) implants are
widely used in surgery as vascular, ophthalmic, or skin
substitutes and bone cements. A range of in vitro and
in vivo biocompatibility tests have been performed to
control their integration into host tissues. Surface prop-
erties of biomaterials play a critical role in the adhesion
process of adjacent cells. Some of these properties have
been well studied in vitro in cell cultures.1 These are
the porosity which affects the endothelial cell behavior
grown in an organotypic culture technique2; the surface
charge which enhances the fibroblast proliferation on
hydroxyethyl methacrylate copolymers3; the surface
topography, a property relevant to titanium implants
which affects human osteoblast differentiation, prolif-
eration, and matrix production4 as well as human gin-
gival fibroblast morphology, orientation, and prolifera-
tion by interfering in the assembly of the focal adhesion
Correspondence to: M. Lampin.
 1997 John Wiley & Sons, Inc. CCC 0021-9304/97/010099-10
´ ` `
which annihilated roughness effect and restored hydrophilic
properties. An organotypic culture assay was carried out in
an attempt to relate the biocompatibility to substratum sur-
face state. Cell migration was calculated from the area of cell
layer. Cellular adhesion was determined by measuring the
kinetic of release of enzymatically dissociated cells. A slight
roughness raised the migration area to an upper extent no
matter which cell type. Enhancement of the cell adhesion
potential was related to the degree of roughness and the
hydrophobicity.  1997 John Wiley & Sons, Inc. J Biomed Mater
Res, 36, 99–108, 1997.
Key words: biomaterial roughness; surface free-energy; cell
adhesion; organotypic culture
sites5; and the wettability which interferes both in hu-
man fibroblast6,7 and endothelial cell attachment8,9 and
in plasma protein adsorption.10–12 Variation of the hy-
drophobicity induced by an ion-etching surface treat-
ment affects both fibroblast attachment and spread-
ing.13 In general, hydrophilic surfaces displayed better
affinity for cells but lower affinity for proteins than
hydrophobic surfaces. Optimal wettability conditions
for human skin fibroblast adhesion have been defined
on physicochemically characterized gradient surfaces.6
Measurement of the wettability of a material, ex-
pressed by the contact angle in the presence of different
liquids, might be a predictive index of cytocompatibil-
ity. Previous work reported an effect of surface
roughness on the contact angle, the increase or de-
crease of which depended on its initial value.14–18 There-
fore, a material surface treatment might enable the
adaptation of its surface free energy to biological re-
quirements.
We proposed studying the influence of the degree
of roughness of PMMA, obtained by sandblasting, both
on surface free energy and on cell adhesion and migra-
tion. Explant cultures of corneal and vascular endothe-
lia from chick embryos provided culture models reli-
able for comparative behavior studies of tissues which
exhibit different specific functions.
100
MATERIALS AND METHODS
Materials
Poly(methyl methacrylate) (Goodfellow, Lille, France)
was used for its compatibility with tissue and its bio-
medical applications as contact lenses or intraocular
implants. PMMA was cut into square samples (1 3 1
cm) 1 mm in thickness. Specimens were sonicated 2 3
10 min in distilled water then 10 s in ethanol before
drying. Smooth surfaces were sandblasted with alu-
mina (Al2O3) grains. Different topographies were gen-
erated according to the sand grain size (50, 125, and
250 em) and the air pressures (0.2–0.4 MPa) with a
sandblaster (Basic Duo Renfert). The spray angle was
908, the abrasive working time was about 10 s and the
jet length about 3 cm. The samples were washed and
dried as described previously. The samples were classi-
fied into five groups: (1) smooth PMMA, i.e., PMMA;
(2) sandblasted PMMA 50 em and 0.4 MPa, i.e., PMMA
50/4; (3) sandblasted PMMA 125 em and 0.4 MPa, i.e.,
PMMA 125/4; (4) sandblasted PMMA 250 em and 0.2
MPa, i.e., PMMA 250/2, and (5) sandblasted PMMA
Figure 1. Scanning mechanical microscopy of (a) PMMA and (b) PMMA 250/4.
LAMPIN ET AL.
250 em and 0.4 MPa, i.e., PMMA 250/4. These poly-
mers were sterilized by dry heat at 1208C for 30 min.
Methods
Roughness measurement
Surface topography was performed using a profi-
lometer, a scanning mechanical microscope (SMM;
Akilog, Besançon, France). It was based on an inductive
sensor equipped with a diamond stylus probe with a
pyramidal shape and a vertex angle of 608. The lateral
resolution was 2 em and the vertical one was 0.01 em.
Scanning operations were performed by a step engine
which moved samples in the X and Y directions. The
system was monitored through a PC computer. The
software provided a dimensional view of the PMMA
surface and characterized its surface in terms of
roughness parameters. Among the roughness parame-
ters, we have selected the Ra parameter (or CLA), which
is the center line average, the Rt parameter, which is
the distance between the highest and the lowest point
ROUGHNESS, WETTABILITY, ORGANOTYPIC CELL CULTURE 101

TABLE I TABLE III

Means and Standard Deviations for SFE Components (mJ/m2) of Smooth and Rough PMMA
Roughness Parameters
Material csLW cs2 cs1 csab cs
Material Rt (em) Ra (em) DA (%)
PMMA 40.6 26 0 0 40.6
PMMA 0.71 6 0.47 0.07 6 0.02 100 6 0.0 PMMA 50/4 49.5 0 0 0 49.5
PMMA 50/4 4.46 6 0.72 0.18 6 0.02 100.21 6 0.03 PMMA 125/4 49.5 0 0 0 49.5
PMMA 125/4 14.64 6 1.58 1.11 6 0.2 102.66 6 0.48 PMMA 250/2 49.5 0 0 0 49.5
PMMA 250/2 24.39 6 3.23 1.79 6 0.20 105.67 6 2.59 PMMA 250/4 48.9 0 0 0 48.9
PMMA 250/4 34.26 6 3.71 3.34 6 0.54 108.81 6 3.34
csLW, apolar component; cs2, basic component; cs1, acid
Values represent the means of 15 measurement. component; csab, acido-basic component; cs, total component
of the SFE calculated according to Good van Oss’s equations.

of the surface irregularities, and the three-dimensional

(3D) developed area (DA) corresponding to the ratio
between the experimental and the theoretical area. This
was used as a functional parameter to surface free
energy (SFE) and adhesion performances. A total of
15 areas were recorded for each group. The acquisition
conditions were: sampling step in X and Y direction:
3 em; number of points; 256 3 256; evaluation area:
0.6 mm2.
Determination of SFE
Contact angles were measured on the substrate by
a sessile drop method using two stages—which moved
the samples and the syringe in the X, Y, and Z direc-
tions—equipped with a video camera (Hitachi Denshi
Ltd., CCD black and white) (500 3 500) connected to
a computer (Kentac). A specific program was devel-
oped for the measurement.19 A liquid drop was care-
fully deposited with a syringe onto previously washed
and dried material. A picture of the drop was recorded
by means of a punch card (Video Blaster), digitized,
and transmitted to a computer screen. The user drew
the baseline and chose two points, one inside and an-
other outside the drop. The software determined suc-
cessively the line equation, the circle equation, and
their intersection point; then the circle tangent was
drawn to calculate the contact angle. Four reference
liquids (distilled water, diiodomethane, glycerol, and
ethylene glycol) were used to determine the apolar,
csLW, and the acido-basic, csab, acidic, cs1, and basic,
cs2, components of SFE by means of Good van Oss’s
equations.20 To prevent erratic results owing to the
absorption of water from the atmosphere, we mea-
sured the contact angle in at least four liquids, includ-
ing water as one of three polar liquids. All experiments
were performed at room temperature.
Organotypic culture technique
The organotypic culture technique21 was performed
to assess the cytocompatibility of PMMA samples. Cor-
neal and aorta fragments from 14-day-old chick em-
bryos were placed on a semisolid nutrient medium
and covered with 1-cm2 PMMA so that the endothe-
lium was in direct contact with the surface of PMMA
under investigation. The nutrient medium was a mix-
ture (v/v) of 1% Bacto agar (Difco) in Gey solution and
medium 199 or MEM Iscove (Boehringer Mannheim,
Germany) respectively used for vascular or corneal
explant cultures, and 10% fetal calf serum (Boehringer
Mannheim, Germany) tricin (0.02M) (Merck Darm-
stadt, Germany) and L-glutamin (2 mM) (Boehringer
Mannheim). A total of 60 PMMA samples were depos-
ited onto 60 explants of each tissue. Cultures were
incubated for 7 days at 378C.
To control the medium protein adsorption, PMMA
samples were deposited onto agar medium and incu-
bated at 378C for 7 days. According to Ueda et al.,22
the presence of proteins was verified by the contact
angle method and SFE was determined as pre-
viously described.
Cytocompatibility assessment
After neutral red staining, the surfaces of the cell
layers surrounding the explants were measured with

TABLE II
Contact Angles (degrees) on Smooth and Rough PMMA
Material Water Diiodomethane Glycerol Ethylene Glycol
PMMA 62 6 5 38 6 1.6 64 6 1.5 48 6 1.7
PMMA 50/4 104 6 4.5 15 6 4.3 95 6 4.3 43 6 3.8
PMMA 125/4 99 6 6.6 14 6 5.7 85 6 4.8 49 6 2.5
PMMA 250/2 106 6 3.8 15 6 3.4 98 6 3.2 44 6 3.3
PMMA 250/4 102 6 4.2 18 6 3.7 99 6 5 37 6 3.6
Values are means of 50 measurements 6 standard deviation.
102 LAMPIN ET AL.

TABLE IV
Contact Angles (degrees) on Smooth and Rough PMMA after Adsorption of
Proteins from the Nutritive Medium
Material Water Diiodomethane Glycerol Ethylene Glycol
PMMA 6 6 1.8 40 6 3.4 46 6 4.0 9 6 2.5
PMMA 50/4 7 6 1.8 46 6 2.3 44 6 4.5 10 6 5.7
PMMA 125/4 6 6 2.4 47 6 2.4 44 6 4.4 15 6 4.3
PMMA 250/2 5 6 1.8 44 6 3.0 36 6 3.6 8 6 2.1
PMMA 250/4 7 6 2.7 46 6 2.7 37 6 4.3 13 6 4.0
Values are means of 50 measurements 6 standard deviation.

a stereomicroscope fitted with a camera lucida and
digitizing tablet connected to a microcomputer. The
cells were dissociated by trypsinisation and counted
with a Multisizerw (Coultronics). Cellular density
was calculated.
then fixed in 4% formaldehyde solution in PBS for
30 min. After several rinses in PBS, cell layers were
dehydrated in graded alcohol and observed. The silver
nitrate coloration revealed intercellular junctions in a
monolayer endothelium tissue.

Adhesion test Indirect immunofluorescence assays

The cells were enzymatically detached from the
Cell cultures previously fixed in absolute methanol
PMMA samples by a trypsin-EDTA (0.025% v/v) treat-
at 2208C were incubated for 1 h in rabbit polyclonal
ment to establish the curve of the percentages of re-
anti-human von Willebrand factor (vWF) antibody
leased cells versus time. By integration, we calculated
the area included between the curve and the X axis (Sigma), rinsed with PBS, and finally incubated with
fluorescein-conjugated goat anti-rabbit immunoglobu-
which defined a detachment index representative of
lin antibody (Sigma). The samples were examined with
adhesion. This area was inversely proportional to the
cell adhesion on the biomaterial. The higher the index, a Leitz microscope equipped with epifluorescence illu-
the weaker the cell adhesion. According to the detach- mination.
ment index, three adhesion levels have been delimited:
below 3000, a strong adhesion level; between 3000 and
4500, a medium adhesion level; above 4500, a strong RESULTS
adhesion level.23

Cell viability Effect of surface roughness on SFE

The cell viability assessment was based on a trypan Topographies of smooth and rough PMMA samples
blue exclusion test as described by Roux et al.24 were observed by SMM. Smooth PMMA shown in
Figure 1(a) displayed surface irregularities ,1 em,
Scanning electron microscopy (SEM)
whereas on PMMA 250/4 notches exceeded 1 em [Fig.
Cell layers were fixed in 3% glutaraldehyde in 1(b)]. Table I shows that the roughness parameters Rt
Rembaum buffer (pH 7.4), dehydrated in graded alco- and Ra were proportional to pressures and granulari-
hols, critical-point dried with CO2 (Polaron Instrument, ties of alumina grains.
Inc.), and examined using a Jeol (Model JSM 840) SEM Contact angles of water, diiodomethane, ethylene
at an accelerating voltage of 15 kV. The extracellular glycol, and glycerol on smooth and rough PMMA are
matrix (ECM) was also observed by means of SEM. summarized in Table II. Table III shows the dispersive
The cell layers were rinsed in phosphate-buffered sa- and acido-basic components of the SFE. It appears that
line (PBS), then in 20 mM glycin buffer (pH 9.6) con-
taining phenylmethylsulfonyl fluoride (PMSF) 1 mM TABLE V
(Sigma). They were immersed in 1% Nonidet P40 SFE Components (mJ/m2) of Smooth and Rough PMMA
(Sigma) in glycin buffer, pH 9.6, for 10–15 min. Then after Adsorption of Proteins from the Nutritive Medium
the substrates were washed with 10 mM Tris-HCl Material csLW cs2 cs1 csab cs
buffer, pH 7.5, with PBS and fixed in 3% glutaralde-
hyde in Rembaum buffer, pH 7.4. PMMA 39.6 69.1 0.05 3.72 43.3
PMMA 50/4 36.5 68.7 0.24 8.12 44.6
Silver staining PMMA 125/4 35.9 69.9 0.22 7.84 43.7
PMMA 250/2 37.5 65.1 0.47 11.1 48.6
PMMA 250/4 36.4 65.9 0.48 11.2 47.6
Cell layers were exposed to 0.15% AgNO3 solution in
distilled water for 30 s, rinsed with 5% glucose solution, Abbreviations are as in Table III.
ROUGHNESS, WETTABILITY, ORGANOTYPIC CELL CULTURE 103

tion areas reached 19 and 8 mm2 for vascular and cor-

neal cells, respectively. Numbers of cells liberated from
explants grown on PMMA samples did not vary sig-
nificantly with roughness. In vascular cell cultures, we
calculated 25,000–30,000 cells/explant and 10,000–
15,000 cells/explant in corneal cell cultures. Percent-
ages of viable cells were not affected by the roughness.
Higher values (85–90%) were found in vascular than
corneal cultures (82%).

Effect of surface roughness on the

phenotypic expression

Immunostaining of cell layers with an anti-vWF anti-

body revealed the presence of cytoplasmic fluorescent
dots of vWF antigen in both vascular explant cultures

Figure 2. Migration area of vascular and corneal cells on

smooth and rough PMMA. Values are means of 60 measure-
ments 6 standard deviation. The percentage of cell viability
is indicated on each histogram.

roughness increased the water contact angle and the

apolar component of the surface free energy, whereas
the basic one declined to zero. Only a slight degree of
roughness might suppress the acido-basic properties
of the polymer. Thus, roughness enhanced the hydro-
phobicity of PMMA.
Since PMMA was immersed into medium during
the explant culture, proteins from the serum diluted
in the nutrient medium might be adsorbed on the syn-
thetic material. The adsorption of protein modified
both contact angles and components of the SFE, as
shown respectively in Tables IV and V. The water con-
tact angle decreased from 1008 to 78 after 24 h incuba-
tion and did not significantly vary with roughness.
The diiodomethane contact angle increased from 158
to 468. The dispersive and basic components slightly
decreased from 40 and 68 mJ/m2 to 36 and 66 mJ/
m2, respectively and the SFE from 43 to 47 mJ/m2.
Adsorption of proteins modified the roughness effect
by smoothing surface irregularities. The PMMA sam-
ples exhibited equivalent surface energies independent
of the degree of roughness.

Effect of surface roughness on migration area

Assessment of migration area (Fig. 2) indicated simi-

lar tendencies for both cell types despite different
ranges. Vascular tissues showed higher responses than
corneal ones on both smooth and rough PMMA sam-
ples. The migration area increased twofold in vascular Figure 3. vWF staining of (a,b) vascular and (c,d) corneal
cell cultures and threefold in corneal cell cultures cells grown on (a,c) PMMA and (b,d) PMMA 250/4 (original
grown on rough PMMA compared to smooth. Migra- magnification 31250).
104
Figure 4. Silver-stained culture of (a,b) vascular and (c,d) corneal tissue on (a,c) PMMA (original magnification 3500) and
(b,d) PMMA 250/4 (original magnification 31250).
performed on smooth and rough PMMA [Fig. 3(a,b)].
Corneal cell cultures did not express the vWF antigen,
and therefore no fluorescence could have been detected
[Fig. 3(c,d)]. However, the presence of endothelial cells
was revealed by silver staining. Both cell layers from
vascular [Fig. 4(a,b)] and corneal [Fig. 4(c,d)] explant
cultures exhibited a staining of the intercellular junc-
tions which is endothelial cell-specific. SEM of cell lay-
ers issued from both explant cultures showed the pres-
ence of apposed hexagonal cells on smooth PMMA
[Fig. 5(a,b)]. On PMMA 250/4 [Fig. 5(c,d)], cell layers
exhibited a slightly different aspect characterized by
irisated cell boundaries. Similar observations were
made on both rough PMMAs.
Effect of surface roughness on cell adhesion
Regarding cell adhesion, both cell types had similar
behavior (Fig. 6). Although all of the materials were
situated in the zone of medium adhesion, PMMA
reached the limit of weak adhesion around 4000,
whereas PMMA 250/4 approached the limit of a strong
adhesion zone. Thus, cell adhesion potential seemed
to increase with roughness. The ECM involved in a
cell adhesion process had been observed by SEM. In
LAMPIN ET AL.
both vascular and corneal cultures, a fibrous network
covered the PMMA samples, but the ECM was thicker
and more abundant on PMMA 250/4 [Fig. 7(a,c)] than
on PMMA [Fig. 7(b–d)]. It contained fibronectin as
well as collagen types I and III as revealed by immuno-
staining (not shown).
DISCUSSION
Modification of the surface topography of PMMA
by sandblasting was responsible for changes of SFE
components and of biological parameters of endothe-
lial cell cultures. Few investigators have related the
influence of surface roughness on wettability.13–17 We
found that surface roughness increased the dispersive
component csLW of SFE to an upper level independent
of the degree of roughness. In a nonpolar liquid such as
diiodomethane, adhesion strengths may have resulted
essentially from dispersive energy, a component of
which was calculated from the equation csLW 5 cl(1 1
´
cos u)2/4, derived from Young,25 Dupre,26 and Fowkes.27
When cos u tended toward 1, cs tended toward clLW,
LW
an intrinsic parameter of the dispersive liquid used.
ROUGHNESS, WETTABILITY, ORGANOTYPIC CELL CULTURE
Figure 5. Scanning electron micrographs of (a,c) vascular and (b,d) corneal cell layer grown on (a,b) PMMA and (c,d)
PMMA 250/4.
In diiodomethane, csLW might reach 50.8 J/m2, the clLW
value provided by Good’s report.28
The contact angle of sessile droplets also shifted to
constant values on rough PMMA independent of the
roughness but dependent on its initial value. In water
and in glycerol, the contact angle increased from 628–
648, the initial values, to a maximum of 1068. In contrast,
in diiodomethane and in ethylene glycol, the initial
contact angle smaller than 508 decreased with
roughness. These quantitative results differed some-
what from those reported by Busscher et al.14 Although
the effect of roughness on SFE was similar to our obser-
vations when the initial contact angles were above 868
or below 608, surface roughening had no influence be-
tween 608 and 868. These differences might be due to
105
a variation of scale of surface roughening. The stylus
surface roughness Ra of sandblasted PMMA varied
from 0.07 to 3.34 em, whereas the Ra of the polymers
studied by Busscher et al. was smaller than 1 em. More
recently, Keisler et al.18 studied the influence of steel
roughness on contact angle hysteresis. In their experi-
mental device, the scale of roughness, spreading from
1 to 2 em, did not influence wettability. This agrees
with our results, which showed that PMMA 125/4 and
250/2 had an equivalent SFE.
Hachom et al.,29 in our laboratory, also studied the
influence of roughness of sandblasted titanium on the
SFE. Their results differed in that increase of the disper-
sive component was proportional to the degree of
roughness. Increases of both basic and acid compo-
106
nents were still more striking. We think that sand-
blasting of metals might be responsible for surface
oxidation of the metals which altered their physico-
chemical properties.
Surface free energy of smooth PMMA, approxi-
mately 40 mJ/m2, was characterized by predominance
of the basic component. An interpretation was that
compounds of the atmospheric
˚ environment deposited
a layer of about 4–100 A thickness, mainly constituted
of carbon dioxide and water.30 It might not be excluded
that the basic properties of PMMA arose from the
chemical nature of the polymer. Nevertheless, the loss
of the acido-basic properties of PMMA, even in the
mildest sandblasting conditions, could not be ex-
plained. Residual deposition of alumina grains, not
completely eliminated by ultrasonication, might be a
non-negligible factor of disturbance. However, contact
angle measurements showed that incubation in nutri-
ent culture medium strongly modified wettability and
SFE. An early adsorption of proteins occurred onto the
material surface and reduced the roughness effect by
smoothing the surface irregularities. The nature of the
adsorbed proteins and their conformation were depen-
dent on the surface properties of synthetic polymer
and on the affinity of proteins for substrata.10,12,31 It
seemed that hydrophilic materials promoted a weak
reversible adsorption of proteins, whereas hydropho-
bic ones induced an irreversible adsorption of high-
molecular-weight proteins. Plasma fibrinogen10 and
fibronectin12 are among those proteins which had a
high affinity for hydrophobic surfaces. Fibronectin
plays an important role in the adhesion process of
mammalian cells through specific interaction with cel-
lular membrane receptors. This binding activates a sig-
naling cascade which generates cell migration, prolifer-
ation, or differentiation. In vivo after implantation or
in vitro in cell culture, tissues and cells might react
differently as they encounter rough or smooth surfaces
maintained in biological fluids able to coat them with
specifically active proteins. The explant culture method
we used tried to reproduce the in vivo biological condi-
tions of implantation by maintaining the integrity of
the tissues. Although the number of cells released from
corneal explant cultures was lower than the number
of cells grown from vascular explants, independent of
the culture substrata, roughness enhanced the migra-
tory potential of both cells but did not affect their
proliferative activities. Cell adhesion increased with
roughness, whereas migration fluctuated around a
higher mean value on rough samples. The presence of
a greater amount of EMC fibrous proteins on rough
PMMA, revealed by SEM, further explained this result.
Roughness might trigger subconfluent cells to secrete
extracellular proteins which allowed better anchorage
of cells to their substratum. Another possibility was
that the deposition of culture medium fibronectin, en-
hanced by the increased hydrophobicity of rough
LAMPIN ET AL.
Figure 6. Effect of roughness on cell adhesion. Cell adhe-
sion assessed by the detachment index is plotted versus sur-
face roughness expressed in terms of developed area.
PMMA, accumulated in greater amounts, or in differ-
ent conformations, so that to strengthen cell adhesion.
Hachom et al.29 reported comparable behavior of hu-
man gingival explants grown in organotypic cultures
on rough titanium.
Several in vitro studies focused on early events oc-
curring during the process of cell attachment and
spreading onto various surfaces. Focal adhesion points
appeared in human gingival fibroblasts after seeding
on electropolished titanium.5 Filamentous actin net-
works developed during cell spread. Fibroblasts did
not spread as efficiently on sandblasted titanium and
had aberrant morphologies. They developed no focal
adhesions. Comparative absence of focal contacts and
diffuse actin were also observed in human skin fibro-
blasts seeded on hydrophobic octadecyl–treated glass.7
Prior coating of the surface with fibronectin improved
cell adhesion and allowed the organization of actin
stress fibers and the characterization of b1 integrin con-
comitant with an increase of phosphorylation activi-
ties. The number of human osteoblast-like cells (MG63)
was also reduced on rough titanium, and their prolifer-
ation was inversely related to surface roughness.4 Cell
differentiation and matrix production were also altered
by roughness. The effect of surface state on the cyto-
compatibility of human osteoblasts and fibroblasts was
extended to cobalt-chromium–based alloys.32 If the os-
teoblasts were more sensitive than the fibroblasts, it
turned out that a slightly rough surface obtained by
microbead blasting was more favorable than polished
surface to cell attachment, spreading, and proliferation.
These observations pointed out an effect of the mate-
rial surface on cell response which depended on the
degree of roughness. Although expression of cell phe-
notype might be modified, cell sensitivity varied with
ROUGHNESS, WETTABILITY, ORGANOTYPIC CELL CULTURE
Figure 7. Scanning electron micrographs of ECM of (a,c) vascular and (b,d) corneal cells grown on (a,b) PMMA and
(c,d) PMMA 250/4.
the cells’ origin. The organotypic culture techniques
used in our study appeared to be less sensitive, for
two reasons: First, sandblasting of hydrophobic
PMMA tested generated roughness which increased
the water contact angles about 408, whereas chemical
treatment of glass shifted it up 708; second, we mea-
sured adhesion as a long-term phenomenon, which
resulted from the interaction between cells and surface
materials conditioned by the environmental medium
and coated with extracellular proteins. Further investi-
gation will try to identify and quantify cell and me-
dium proteins deposited onto substrata as a function
of their surface properties and incubation time.
107
CONCLUSION
From a physicochemical point of view, the initial
apolar component of the SFE of PMMA strongly in-
creased with roughness, whereas the basic component
annulated. Proteins adsorption on PMMA slightly de-
creased the apolar component that reduced the hydro-
phobicity.
From a biological point of view, vascular and corneal
explant cultures from 14-day-old chick embryos dis-
played a similar behavior. A rise in the PMMA
roughness promoted cell tissue adhesion as a result of
an enhanced hydrophobicity which favored the ad-
108
sorption of adhesive proteins. Increase of cell migration
did not seem to depend on the degree of roughness.
The authors thank J. L. Duval for his help with the scanning
electron microscopy study.
References
1. D. M. Brunette, ‘‘The effect of implant surface topogra-
phy on the behavior of cells,’’ Int. J. Oral Maxillofac.
Implants, 3, 231–246 (1988).
2. M. F. Sigot-Luizard, M. Sigot, R. Guidoin, M. King,
W. W. Von Maltzman, R. Kowligi, and R. C. Eberhart,
‘‘A novel microporous polyurethane blood conduit: Bio-
compatibility assessment of the UTA arterial prothesis
by an organo-typic culture technique,’’ J. Invest. Surg.,
6, 251–271 (1993).
3. S. Hattori, J. D. Andrade, J. B. JR Hibbs, D. E. Gregonis,
and R. N. King, ‘‘Fibroblast cell proliferation on charged
hydroxyethyl methacrylate copolymers,’’ J. Colloid Inter-
face Sci., 104, 72–78 (1985).
4. J. Y. Martin, Z. Schwartz, T. W. Hummert, D. M.
Schraub, J. Simpson, J. J. R. Lankford, D. D. Dean, D. L.
Cochran, and B. D. Boyan, ‘‘Effect of titanium surface
roughness on proliferation, and protein synthesis of
human osteoblast-like cells (MG63),’’ J. Biomed. Mater.
Res., 29, 389–401 (1995).
¨ ¨
5. M. Kononen, M. Hormia, J. Kivilahti, J. Hautaniemi,
and I. Thesleff, ‘‘Effect of surface processing on the
attachment, orientation, and proliferation of human gin-
gival fibroblasts on titanium,’’ J. Biomed. Mater. Res., 26,
1325–1341 (1992).
6. T. G. Ruardy, J. M. Schakenraad, H. C. van der Mei, and
H. J. Busscher, ‘‘Adhesion and spreading of human skin
fibroblasts on physicochemically characterized gradient
surfaces,’’ J. Biomed. Mater. Res., 29, 1415–1423 (1995).
7. T. Groth and G. Altankov, ‘‘Fibroblasts spreading and
proliferation on hydrophilic and hydrophobic surfaces
is related to tyrosine phosphorylation in focal contacts,’’
J. Biomater. Sci. Polym. Edn., 7, 297–305 (1995).
8. H. Yasuda, B. S. Yamanashi, and D. P. Devito, ‘‘The rate
of adhesion of melanoma cells onto nonionic polymer
surfaces,’’ J. Biomed Mater. Res. 12, 701–706 (1978).
9. P. B. van Wachem, A. H. Hogt, T. Beugeling, J. Feijen,
A. Bantjes, J. P. Detmer, and W. G. van Aken, ‘‘Adhesion
of cultured human endothelial cells onto methacrylate
polymers with varying surface wettability and charge,’’
Biomaterials, 8, 323–328 (1987).
10. D. R. Absolom, W. Zingg, and A. W. Neumann, ‘‘Pro-
tein adsorption to polymer particles: Role of surface
properties,’’ J. Biomed. Mater. Res., 21, 161–171 (1987).
11. S. Kothari, P. V. Hatton, and C. W. I. Danglas, ‘‘Protein
adsorption to titanic surface,’’ J. Mater. Sci., 6, 695–
698 (1995).
12. G. Altankov and T. Groth, ‘‘Reorganization of substra-
tum on hydrophilic and hydrophobic materials is re-
lated to biocompatibility,’’ J. Mater. Sci., 5, 732–737
(1994).
13. H. J. Busscher, I. Stokroos, J. G. Golverdingen, and J. M.
Schakenraad, ‘‘Adhesion and spreading of human fi-
broblast on superhydrophobic FEP-Teflon,’’ Cells Ma-
ter., 1, 243–249 (1991).
14. H. J. Busscher, A. W. J. van Pelt, P. De Boer, H. P. De
Jong, and J. Arends. ‘‘The effect of surface roughening
of polymers on measured contact angles of liquids,’’
Colloids Surf., 9, 319–331 (1984).
LAMPIN ET AL.
15. M. Quirynen, M. Marechal, H. J. Busscher, A. H. Weer-
kamp, P. L. Darius, and D. van Steenberghe, ‘‘The in-
fluence of surface free energy and surface roughness
on early plaque formation. An in vivo study in man,’’
J. Clin. Periodont., 17, 138–144 (1990).
16. J. L. Charrier, C. Hachom, M. F. Sigot-Luizard, and M.
´
Degrange, ‘‘Rugosite, energie de surface et culture or-
ganotypique,’’ J. Biomat. Dent., 8, 111–118 (1993).
17. R. D. Schulze, W. Possart, H. Kamusewitz, and C.
Bischof, ‘‘Young’s equilibrium contact angle on rough
solid surfaces, part I. An empirical determination,’’ J.
Adhes. Sci. Technol., 1, 39–48 (1989).
18. C. Keisler and J. L. Lataillade, ‘‘The effect of surface
roughness characteristics on wettability and on the
mechanical properties of adhesive joints loaded at
high strain rates,’’ J. Adhes. Sci. Technol., 9, 395–411
(1995).
19. M. Lampin, V. Petit, and M. Degrange, ‘‘Un nouveau
dispositif informatique de mesure de l’angle de contact:
` ´
Application a la cinetique de mouillage de biopoly-
`
meres,’’ J. Biomater. Dent., 10, 67–76 (1995).
20. C. J. van Oss, R. J. Good, and M. K. Chaudhury, ‘‘The
role of van der Waal’s and hydrogen binds in hydropho-
bic interactions between biopolymers and low energy
surfaces,’’ J. Colloid. Interface Sci., 111, 378–390 (1986).
21. M. F. Sigot-Luizard, M. Lanfranchi, J. L. Duval, S. Ben
Slimane, M. Sigot, R. Guidoin, and M. King, ‘‘The cyto-
compatibility of compound polyester-protein surfaces
using an in vitro technique,’’ In Vitro, 22, 234–240 (1986).
22. T. Ueda, K. Ishihara, and N. Nakabayashi, ‘‘Adsorption-
desorption of proteins on phospholipid polymer sur-
faces evaluated by dynamic contact angle measure-
ment,’’ J. Biomed. Mater. Res., 29, 381–387 (1995).
23. J. L. Duval, M. Letort, and M. F. Sigot-Luizard, ‘‘Com-
parative assessment of cell substrate static adhesion us-
ing an in vitro organ culture method and computerized
analysis system,’’ Biomaterials, 9, 155–161 (1988).
24. H. Roux, J. L. Duval, M. F. Sigot-Luizard, and M. Sigot,
‘‘Assessment of the cytocompatibility of biomaterials
by analysis of cellular viability in a Coulterw Multisizer
multichannel analyser,’’ in Biomaterial–Tissue Interfaces,
P. J. Doherty et al. (eds.), Elsevier Science, New York,
1992, pp. 81–87.
25. T. Young, Phil. Trans. R. Soc. Lond., 95, 65 (1805).
´ ´ ´
26. A. Dupre, Theorie Mechanique de la Chaleur, Gauthier
Villars, Paris, 1869, p. 369.
27. F. M. Fowkes, ‘‘Quantitative characterization of the
acid-base properties of solvants, polymers and or-
ganic,’’ J. Adhes. Sci. Technol., 4, 669–691 (1990).
28. R. J. Good, ‘‘Contact angle, wetting and adhesion: A
critical review,’’ J. Adhes. Sci. Technol., 6, 1269–1302
(1992).
29. G. C. Hachom, J. L. Charrier, M. Degrange, and M. F.
Sigot-Luizard, ‘‘Influence of material surface state on
the expression of specific proteins in organotypic cul-
ture of human gingival tissue,’’ in 11th European Confer-
ence on Biomaterials, Pise, Italy, ESB, 1994, pp. 353–356.
´
30. J. Cognard, ‘‘La couche atmospherique: Approache de
´
la surface reelle des solides,’’ J. Chimie Phys., 84, 357–
362 (1987).
´
31. V. H. Perez-Luna, T. A. Horbett, and B. D. Ratner, ‘‘De-
veloping correlations between fibrinogen adsorption
and surface properties using multivariate statistics,’’ J.
Biomed. Mater. Res., 28, 1111–1126 (1994).
32. A. Naji and M. F. Harmand, ‘‘Study of the effect of the
surface state on the cytocompatibility of a Co-Cr alloy
using human osteoblasts and fibroblasts,’’ J. Biomed.
Mater. Res., 24, 861–871 (1990).