Effect of 3D microgroove surface topography onplasma and cellular fibronectin of humangingival fibroblasts

Yingzhen Lai a, Jiang Chen b,*, Tao Zhang c, Dandan Gu d,Chunquan Zhang d, Zuanfang Li e, Shan Lin f, Xiaoming Fu a,Stefan Schultze-Mosgau g

a School of Stomatology, Fujian Medical University, Fuzhou, Fujian 350000, ChinabDepartment of Oral Implantology, Affiliated Stomatological Hospital of Fujian Medical University, Fuzhou,

Fujian 350002, ChinacDepartment of Immunology, Fujian Academy of Medical Science, Fuzhou, Fujian 350000, ChinadMEMS Research Center of Xiamen University, Xiamen, Fujian 361000, Chinae Fujian Academy of Integrative Medicine, Fujian University of Traditional Chinese Medicine, Fuzhou, Fujian 350122,

Chinaf First Affiliated Hospital of Fujian Medical University, Fuzhou, Fujian 350000, ChinagDepartment of Oral and Maxillofacial Surgery/Plastic Surgery, University of Jena, Jena, Thuringia 07747, Germany

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 1

a r t i c l e i n f o

Article history:

Received 3 May 2013

Received in revised form

2 August 2013

Accepted 3 August 2013

Keywords:

Dental implant

Microgroove

Titanium surfaces

Plasma FN

Human gingival fibroblasts

Cellular FN

a b s t r a c t

Objectives: Fibronectin (FN), an extracellular matrix (ECM) glycoprotein, is a key factor in

the compatibility of dental implant materials. Our objective was to determine the optimal

dimensions of microgrooves in the transmucosal part of a dental implant, for optimal

absorption of plasma FN and expression of cellular FN by human gingival fibroblasts

(HGFs).

Methods: Microgroove titanium surfaces were fabricated by photolithography with parallel

grooves: 15 mm, 30 mm, or 60 mm in width and 5 mm or 10 mm in depth. Smooth titanium

surfaces were used as controls. Surface hydrophilicity, plasma FN adsorption and cellular

FN expression by HGFs were measured for both microgroove and control samples.

Results: We found that narrower and deeper microgrooves amplified surface hydrophobic-

ity. A 15-mm wide microgroove was the most hydrophobic surface and a 60-mm wide

microgroove was the most hydrophilic. The latter had more expression of cellular FN than

any other surface, but less absorption of plasma FN than 15-mm wide microgrooves.

Variation in microgroove depth did not appear to effect FN absorption or expression unless

the groove was narrow (�15 or 30 mm). In those instances, the shallower depths resulted in

greater expression of cellular FN.

Conclusions: Our microgrooves improved expression of cellular FN, which functionally

compensated for plasma FN. A microgroove width of 60 mm and depth of 5 or 10 mm appears

to be optimal for the transmucosal part of the dental implant.

# 2013 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: +86 591 83735488; fax: +86 591 83700838.

Available online at www.sciencedirect.com

journal homepage: www.intl.elsevierhealth.com/journals/jden

E-mail address: dentistyz@126.com (J. Chen).

0300-5712/$ – see front matter # 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.jdent.2013.08.004

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 11110

1. Introduction

Soft tissue integration is a prerequisite for implant success

and requires an effective seal between the soft tissue and

implant to protect the underlying bone from microorganism

invasion.1 Peri-implantitis begins at the soft tissue of implant

is a risk factor in implant failure.2 The soft tissue interface

consists of two zones: a slim epithelium and a thicker

connective tissue. Connective tissues have poor mechanical

resistance at the implant interface compared to natural teeth.3

Furthermore, as Jansen et al. state: ‘‘the quality of the

connective tissue in the transitional area is apparently more

important for the long-term prognosis of oral implants than

the epithelial attachment’’.4 Epithelial down-growth and

attachment to the implant can be inhibited by a firm

connection between the underlying connective tissue and

the implant.3 Our study focused on reinforcing the attachment

between connective tissue and titanium implants via well-

defined surface topographies.

New advances in micro-electromechanical systems

(MEMS) allow the fabrication of biological MEMS (BioMEMS)

and have been used in the investigation of cell response to

microgrooved surfaces.5 Microgrooves integrated into the

structure of an implant have advantages compared with the

traditionally smooth implant surface. For instance, epithelial

down-growth could be inhibited by microgroove-induced

contact guidance of connective tissue during growth.6 The

microgroove surface has also affects cell interaction and

behaviour, including modulation of cell adhesion, prolifera-

tion and gene expression.7,8 However, how the specific

microgroove dimensions affect the extracellular matrix

(ECM) has not yet been studied.

The organization and composition of the ECM mediate

cell adhesion by controlling the degree to which cells can

attach. Fibronectin (FN) is a high-molecular weight

(�440 kDa) glycoprotein of the ECM that is a typical marker

of the biocompatibility of new materials for connective

tissue in dental implant studies.9 FN occurs in two forms:

soluble plasma FN in blood, and insoluble cellular FN in ECMs

or on cell surfaces.10 Plasma FN is a major protein

component of blood plasma and is produced in the liver

by hepatocytes.10 Cellular FN is secreted by various cells,

primarily fibroblasts, as a soluble protein dimer and is then

assembled into an insoluble matrix in a complex cell-

mediated process.11 In both forms, FN is involved in tissue

repair. The plasma form of FN is incorporated into fibrin

clots, affecting platelet function and mediating the early

stages of wound healing. Plasma FN also can be bound to the

cell surface and assembled into extracellular fibrils.12

Cellular FN is synthesized and assembled by cells for

attachment to the ECM as they migrate into a clot to repair

damaged tissue.10 Both forms of FN are vital for establishing

and maintaining tissue architecture and for regulating

cellular processes and behaviours, such as adhesion,

spreading, proliferation, migration and differentiation.10–12

Plasma proteins are spontaneously adsorbed onto an

implant surface within seconds and therefore help determine

the bioactivity of implants.13,14 Plasma FN plays an important

role in the interactions of implants with their surrounding

ECMs by enhancing the attachment of cells to implant

materials.15,16 Furthermore, the surface structure and wetta-

bility of implant materials determine the extent of protein

adsorption.17,18 However, observations regarding the effects of

surface wettability on protein have not been consistent.

Though many previous studies have shown that FN absorp-

tion is enhanced on hydrophobic surfaces,19–21 it has also been

reported that FN adsorption is greater on hydrophilic

surfaces.22,23 Anisotropic wetting is attributed to a liquid

contact line encountering a physical discontinuity (e.g., a solid

edge along a microgroove present on solid surfaces).24 The

microgroove surface leads to anisotropic wetting and

these micro-scale topographical surface structures may

encourage macroscopic changes in droplet wetting behaviour.

What is more, to the best of our knowledge there has been only

a few studies23 on anisotropic wetting and the dimensions of

microgrooves affecting plasma FN absorption.

Cells continuously synthesize their own FN matrix, which

they deposit and organize as a three-dimensional network.

Material-mediated cellular FN reorganization is an important

factor in determining the biocompatibility of a material,25 cell

attachment is required for cellular synthesis of FN.26 Cells

react to micromaterial corrugation, possibly through mem-

brane deformation and stretching.27 Thus, forces can be

generated from just the process of cells recognizing topo-

graphical or other cues.28,29 This is relevant because mechani-

cal forces are necessary for efficient (integrin-mediated)

cellular FN assembly to fibrillar matrices.30–32 Furthermore,

the forces placed upon cells, stimulated by material topogra-

phy, can influence cellular FN assembly.27–32 Since topograph-

ical features alter surface wettability, adsorption of preferred

proteins might result. We hypothesized that the dimensions of

some microgroove surfaces might influence the synthesis and

assembly of cellular FN more effectively than others.

The tissue repair function of FN is ubiquitous and useful for

dental implant biocompatibility studies. The aim of this work

was to investigate the effect of 3D microgroove surfaces on

plasma and cellular FN. Through this effort, we hoped to

identify an optimal set of dimensions for microgroove surfaces

suitable for the transmucosal part of a dental implant. Plasma

FN adsorption on different surface topographies was mea-

sured by immunofluorescence and ELISA. Human gingival

fibroblasts (HGFs) were chosen for this study as they are most

abundantly found in gingival connective tissues and have

been used in numerous in vitro studies of implant integra-

tion.33,34 Cells were cultured on the topographically modified

surfaces; real-time PCR and Western blotting were used to

confirm enhanced cellular FN activity. The structure and

properties of the microgrooves, including topography, groove

dimensions, surface wettability in FN adsorption and synthe-

sis are discussed in detail.

2. Materials and methods

2.1. Fabrication of micro-structured substrates

Microgroove surfaces were fabricated by photolithography

with a micro-structured silicon substrate and an overlying

200-nm thick layer of titanium sputtering. Groove widths of

Fig. 1 – (A) Schematic diagram of each sample surface with the microgroove dimensions. (B) The key steps of the

photolithography.

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 1 1111

60 mm, 30 mm, and 15 mm, and groove depths of 5 mm and

10 mm were fabricated. The combinations of groove width and

depth are hereafter denoted as T15/5, T15/10, T30/5, T30/10,

T60/5, T60/10 (Fig. 1(A)). The cross-sectional shape of each

groove was an inverted trapezoid, with an angle between the

sidewall and groove-ridge of 54.748. Group T0 (i.e. the controls)

meanwhile was a sputter of titanium on a simple planar

silicon substrate.

The key steps of the MEMS wet etching process (Fig. 1(B))

were as follows: a 4-in. silicon wafer was chemically cleaned (III:

4H2SO4 + H2O2; II: NH4OH + H2O2 + 5H2O; I: HNO3 + H2O2 +

5H2O), forming a 1.5-mm silicon dioxide layer on the wafer

(by four micro-controlled diffusion system 4470) (1100 8C, 7 h). A

photoresistant layer (AZ 5214-E Japan) was next coated to a

thickness of approximately 1.8 mm, followed by pre-baking at

96 8C for 12 min. The wafer was exposed to UV light through a

photo-mask containing patterns with 15-mm, 30-mm, or 60-mm

wide lines for 15 s, and developed for 80 s; post-baking followed

at 135 8C for 12 min. The pattern was transferred into the silicon

by etching silicon dioxide. The photo-resistant layer was

removed, silicon etching and silicon dioxide were removed,

and the wafer was cleaned. Lastly, a 200-nm thick sputter of

titanium was deposited on the silicon wafers. Etching time

controlled the depth of the microgrooves; the patterned mask

determined the width of the ridge and groove. The wafers were

then cut to 10 mm � 10 mm and 20 mm � 20 mm chips. Small

chips were placed in the bottom of 24-well plates for

immunostaining, 20 mm � 20 mm chips were placed in the

bottom of 6-well plates for real-time PCR and Western blotting.

In all experiments, the substrates were cleaned by ultrasonica-

tion for 5 min in acetone, soaked in absolute ethylalcohol for

2 h, washed another three times with distilled water and dried

for 20 min in ultraviolet light before the cells were plated.

2.2. Scanning electron microscopy

The surfaces and profiles of the samples were imaged using

a scanning electron microscope (1000�) (Philips-XL30;

Netherlands).

2.3. Contact angle determination

Surface hydrophilicity was determined by measuring liquid

contact angles. A 1-ml drop of distilled water (about 2 mm in

diameter) was placed on the disc perpendicular to the surface

microgrooves and then photographed. The contact angle

between the surface and the drop tangent was measured on

the photograph (Dataphysics OCA20, Germany). The mean

value was calculated from five separate measurements.

2.4. Culture of HGFs

Healthy gingival tissues were obtained from orthodontic

patients who had their premolar teeth removed. The process

was approved by the Ethics Committee of Affiliated Stoma-

tological Hospital of Fujian Medical University. Tissues from

different donors were minced to about 3 mm3, placed in 6-well

plates and covered by a cover slip to prevent tissues from

floating. Tissues were maintained in Dulbecco’s Modified

Eagle’s Medium (DMEM, Gibco, USA) supplemented with 10%

foetal bovine serum (FBS, Hyclone, USA) and antibiotics with

1% (v/v) penicillin–streptomycin solution (Hyclone, USA).

Cultures were maintained in a humidified atmosphere of 5%

(v/v) CO2 at 37 8C. After reaching 80% confluence the HGFs

were digested with 0.25% (w/v) trypsin and 0.02% (w/v) EDTA,

and subcultured at a 1:3 ratio. The cells from passages 3–5

were used in this study.

2.5. Immunofluorescence staining of adsorbed plasma FN

Samples were cut to 10 mm � 10 mm and placed in separate

wells in 24-well plate prepared for immunofluorescence

staining of adsorbed proteins. Human plasma FN (Sigma, F

2006) was reconstituted to a concentration of 20 mg/ml in

PBS. The low concentration value was taken with the

fundamental interest of understanding the adsorption

behaviour and morphology of plasma FN as in previous

studies.35 The surfaces of experimental samples were

immersed in the prepared plasma FN solution (1 ml) for

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 11112

2 h at room temperature with gentle rotation. Additional

samples were also immersed in PBS solution, without

protein, to be used as control surfaces. After incubation,

the solutions were removed and the samples were carefully

washed three times with PBS to eliminate any unbound

protein. The patterned and control surfaces were removed

from PBS and incubated in 2% bovine serum albumin (BSA,

Sigma, A2058) solution for 30 min at room temperature, in

order to block non-specific antibody binding. The immuno-

fluorescence procedure was performed with mouse anti-

human FN (1:50, Sigma, F 0791) for 1 h. Following a PBS wash,

samples were incubated for 30 min in anti-mouse FITC (1:32,

Sigma, F 0257). The sample was then washed with PBS.

Finally, each sample surface was mounted with fluorescence

mounting medium (DAKO, S3023). The protein adsorbed on

each surface was observed with a fluorescence microscope

(OLYMPUS BX43). Plasma FN was visualized as green dots.

Colour fluorescence micrographs were converted to black-

and-white bitmaps with ‘Image J’ (from NIH and available at

http://rsb.info.nih.gov/ij/); its operation and use were de-

scribed previously.36,37 The white spots represented plasma

FN. Image J’s ‘‘Analyze particles’’ was used to count the FN

spots in each bitmap. Seven digital images of each sample

group were captured and analysed for adhesion in the

smooth and microgroove areas.

2.6. Plasma FN adsorption by enzyme-linkedimmunosorbent assay (ELISA)

Plasma FN was suspended in PBS to a concentration of 200 mg/

ml (close to the physiological concentration of FN in plasma)20

to determine the relative adsorption of plasma FN on each

group using ELISA absorbance.36 The procedure for preparing

samples pre-coated with plasma FN was as previously

described (Section 2.5), the only difference being the specific

concentration of FN in solution between the two tests.

Negative controls without FN were also assayed. After

blocking with 2% BSA for 30 min, samples were then incubated

with mouse anti-human FN (1:50, Sigma F 0791) as the primary

antibody for 1 h and anti-mouse-alkaline phosphatase anti-

body (1:10,000, Sigma A-1293) as the secondary antibody for

30 min. Unbound antibodies were rinsed away with two PBS

washes, using p-nitrophenyl-phosphate (pNPP, SigmaN2765)

as substrate for 30 min. The reaction was stopped with 3 M

NaOH. Finally, the reaction solution was transferred to a 96-

well plate and the absorbances of each well were recorded

with a microplate reader at 405 nm. Corrected absorbances

were obtained by subtracting the negative control absor-

bances from the experimental values. All experiments were

performed in triplicate.

2.7. Morphology of cellular FN by immunofluorescence

Samples were introduced to separate wells in a 24-well plate.

HGFs were seeded onto the samples at a density of 5 � 103 cells/

well for one day. Cells were then fixed with 4% formaldehyde for

30 min and permeabilized for 5 min with 0.1% Triton X-100 in

PBS. A solution of 1% BSA in PBS was added for 30 min before

samples were subjected to the primary antibody, mouse anti-

human FN (1:50, Sigma, F 0791), for another 60 min at 37 8C.

Following a PBS wash, cells were stained with the secondary

antibody, anti-mouse-FITC (1:32, Sigma, F 0257), for 30 min,

then rhodamine-labelled with Phalloidin (1:400, Cytoskeleton,

Cat. # PHDR1) for 30 min to view the cytoskeletal actin. Finally,

6-diamidino-2-phenylindole (DAPI) (1:1000, sigma, D9542) was

added for nuclear fluorescence at 1:1000 for 3–5 min. Samples

were washed three times in PBS and mounted with fluorescence

mounting medium (DAKO, S3023) before analysis with by

confocal laser scanning microscopy (CLSM, LSM710, Carl Zeiss,

Germany). We used a laser combiner featuring a diode laser

(405 nm; blue fluorescence) to visualize the cell nucleus, an

argon-ion laser (488 nm; green fluorescence) for cellular FN, and

a helium-neon laser (543 nm; red fluorescence) for cytoskeletal

actin. For smooth controls (T0) we collected 2D images. For

microgroove groups, optical sections were gathered in 1-mm

steps perpendicular to the microscope optical axis. Assembling

a stack of these 3D images from successive focal planes created

the full image.

2.8. Quantification of mRNA levels of cellular FN by real-time PCR

Samples were cut down to 20 mm � 20 mm and were prepared

for real-time PCR. Six of each sample type were introduced onto

a 6-well plate. HGFs were seeded onto the sample surfaces

simultaneously at a density of 3 � 105 cells/well and incubated

for three days until confluent. Total RNA was extracted using

Trizol (Invitrogen). The RNA concentration was determined

using a NanoDrop 1000 (Thermo, USA). Total RNA (1 mg) was

then reverse transcribed with a cDNA Reverse Transcription Kit

(TaKaRa PrimeScript1 RT reagent Kit DRR037A). TaKaRa real-

time PCR primers for human FN (GenBank Accession

NM_212482) included a forward primer: (AGGAAGCCGAGGTTT-

TAACTG) and a reverse primer: (AGGACGCTCATAAGTGT-

CACC), primers for human b-actin (GenBank Accession

NM_001101) included a forward primer (CATGTACGTTGCT

ATCCAGGC) and a reverse primer (CTCCTTAATGTCACGCAC-

GAT). For each experimental condition, reverse-transcribed

cDNA was amplified using SYBR1 Premix Ex TaqTM II (Tli

RNaseH Plus; TaKaRa Code:DRR820A) for each gene on an ABI

Prism 7500 real-time PCR cycler (Applied Biosystems). Data

were analysed by the Ct method; values for FN were normalized

to b-actin for each sample. Data are the means � SD of three

independent experiments.

2.9. Quantification of protein levels of cellular FN byWestern blotting

Six of each sample type (20 mm � 20 mm) were introduced

onto a 6-well plate. HGFs were seeded at a density of

3 � 105 cells/well and incubated for three days until confluent.

Samples with attached cells were removed to a new plate,

washed with ice cold PBS three times and subsequently

scraped in RIPA buffer (RIPA Lysis Buffer, Strong, P0013B,

Beyotime, China) containing a 1% Halt Protease Inhibitor

Cocktail (Thermo, USA). Protein concentration was deter-

mined by BCA Protein Assay Kit (P0012, Beyotime, China).

Equal amounts of proteins were applied to 10% polyacryl-

amide SDS gels (SDS-PAGE), separated electrophoretically

and blotted using PVDF membranes. For detection of FN

Fig. 2 – (A) High magnification SEM images of the right side show microgroove and ridge width (1000T). (B) High

magnification SEM images of the profile show microgroove depth (1000T). (C) Photographs of water droplets on the smooth

and patterned surfaces (100T).

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 1 1113

expression, the membrane was incubated with mouse anti-

human FN (1:300, F7387, Sigma) overnight, washed three times

in TBS-T, incubated subsequently with anti-mouse-peroxi-

dase (1:80,000, A2304, Sigma) for an hour, then washed again

and developed with ECL reagent. GAPDH detected with Anti-

GAPDH (1:20,000, G8795, Sigma) was used as the loading

control. Blotting was simultaneously and independently

repeated three times in HGFs.

2.10. Statistical analysis

The statistical analysis took into consideration statistical

pitfalls common in dental research.38 One-way analysis of

variance (ANOVA) in SPSS 17.0 (SPSS Inc., Chicago, USA) was

used to compare the mean values of the data between the

groups of T0, T15/5, T15/10, T30/5, T30/10, and T60/5,T60/10.

Followed by post hoc statistical tests using an SNK-q test for

Fig. 3 – Multiple-comparison results of contact angles on

different substrates with various surface topographies

(one-way ANOVA, ***p < 0.001, **p < 0.01, mean W SD,

N = 5).

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 11114

each pair of samples compared ( p-values below 0.05 consid-

ered significant).

3. Results

3.1. Topographical characterization and static contactangles

Microgrooves were arranged in parallel (albeit with different

dimensions) on the experimental surfaces (Fig. 2(A)). The

depth of each microgroove group is show in Fig. 2(B). Drop

images were captured by a video camera in the direction

perpendicular to the surface microgroove (Fig. 2(C)). Multiple

comparisons made by one-way ANOVA (Fig. 3) revealed that

sample T60 showed the smallest water contact angle (WCA)

compared with that of other surface samples of smaller widths

( p < 0.001). Furthermore, T60/5 had the greatest surface

hydrophilicity, as demonstrated by its small contact angle.

However, T15/10 had the greatest surface hydrophobicity,

even more than T0 (unmodified control). From these results, it

seems that with narrowing groove width and deepening of the

groove, the greatest surface hydrophobicity was achieved.

3.2. Effects of microgroove titanium surfaces onadsorption of plasma FN by immunofluorescence staining andELISA

Fluorescence staining was used to identify the extent of

plasma FN adhesion to the surfaces of samples (Fig. 4), FN was

detected as green fluorescence (shown in monochrome to

better display contrast). The T15/5 and T15/10 sample sets

showed the greatest degrees of FN adhesion, especially

compared to the control (T0), and the most hydrophilic

microgroove surface (T60). The highest level of fluorescence

was on the T15 sample ( p < 0.001), followed in descending

degrees of FN aggregation by the T30, T60 and T0 samples

(Fig. 5(A)). Different microgroove depths, of the same width,

did not affect plasma FN absorption. These observations were

confirmed using ELISA (Fig. 5(B)).

3.3. Influence of microgroove titanium surfaces on cellularFN and fibroblast morphology

Fibroblasts on the microgroove experimental surfaces (T15/5,

T15/10, T30/5, T30/10, T60/5, and T60/10) had contact guidance

parallel to the microgrooves, whereas the cells on T0 were

oriented randomly (Fig. 6). Most of the cells on T15/5 and T15/

10 were found on the ridge surfaces. As width of the grooves

increased, cells were located not only on the ridge surfaces,

but also into the microgrooves themselves. Newly synthesized

cellular FN was found exclusively in cell aggregates and

existed both on cell surfaces and within the ECM (Fig. 6).

Cellular FN coating on the cells was arranged just like the

cytoskeletal actin, with the extracellular FN scattered along

the groove. Based on 3D images (from microgroove top to

bottom), T15/5, T60/5, and T60/10 samples expressed more

cellular FN than other groups.

3.4. HGF-related cellular expression of FN gene

The relative changes in mRNA expression of FN in HGFs after

3-day culture are expressed as the ratio of the mRNA levels of a

reference gene, b-actin, followed by a standardization of the

threshold cycle (Ct) expressions on the control surface (T0) as

1. In this study, our results demonstrate how the presence and

dimension of microgrooves and depths can affect the up-

regulation and down-regulation of gene expression of FN in

HGFs. T60/5 and T60/10 induced significantly higher mRNA

expression levels of FN than any other surface (Fig. 7(A)). T15/5

and T30/5 exhibited intermediate expression levels that were

somewhat higher than for the controls and T15/10. Taken

together, these results suggest that wider microgrooves (i.e.

T60), even with different depths (5 mm and 10 mm), resulted in

higher cellular FN expression than any other width. Depth

seemed important for the more narrow microgrooves (i.e. T15)

– the shallower depth (5 mm) resulted in higher cellular FN RNA

expression than the deeper grooves (10 mm). FN RNA expres-

sion levels for the T30/5 and T30/10 surfaces did not change

significantly. All microgroove surfaces yielded greater FN

expression than the smooth surface.

3.5. HGF-related cellular FN protein expression

The quantity of FN for the T60/5 and T60/10 surfaces was

greater than that of any other surface (Fig. 7(B)). FN expression

on the T15/5 and T30/5 surfaces was significantly greater than

on T15/10, T30/10 and T0. Results of protein expression had a

trend similar to results of real-time PCR.

4. Discussion

4.1. Design and characterization of microgroove titaniumsurfaces

As our microgroove materials were fabricated with photoli-

thography in one direction, they can be defined as anisotropic

Fig. 4 – (A) Fluorescence micrographs of plasma FN adsorption on the smooth and patterned surfaces (200T). Green

fluorescence spots represent plasma FN. (B) Fluorescence micrographs converted to black-and-white images; white spots

indicate plasma FN. Control micrograph also shown.

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 1 1115

Fig. 5 – (A) Multiple-comparison results of immunofluorescence staining of plasma FN on different substrates with various

surface topographies (one-way ANOVA, ***p < 0.001, **p < 0.01, mean W SD, N = 7). (B) Plasma FN adsorption by OD at 405 nm

(one-way ANOVA, ***p < 0.001, **p < 0.01, mean W SD, N = 3).

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 11116

surfaces. SEM images also supported this classification

(Fig. 2(A) and (B)). The anisotropic surface texture seemed to

influence the growth and proliferation of cells, leading to

contact guidance, which depended upon the specific micro-

pattern and size of geometrical elements. Previously, it was

shown that cells experience contact guidance on grooves, but

the extent to which they react varies by cell type.39 Gingival

fibroblasts are among the major cell types found in peri-

implant connective tissues, so the microgroove dimension

used here were of suitable size. When the grooves or ridges are

larger than the cell size, cell orientation is only marginally

affected. With the reduction of the groove/ridge width to less

than the width of the cells, effects on orientation become more

pronounced.40 The mean size of the HGF was about 110 mm

(measured from 30 cells). We decided that the reasonable

dimensions of grooves would be 15 mm, 30 mm, or 60 mm in

width. However, previous studies showed that the depth of the

groove also plays a role in the cell reaction.41 Furthermore, as

groove depth increased (1 mm, 5 mm and 10 mm) contact

guidance increased.40 Thus, in this study, a reasonable depth

of 5 mm and 10 mm was selected to ensure that contact

guidance was seen.

4.2. Anisotropic wetting characteristics on microgroovesurfaces

Unlike isotropic wetting, where wetting properties are

identical in all directions, water contact angles on the

anisotropic microgrooved titanium substrate were measured

in the direction perpendicular to the microgrooves. The

apparent contact angles are different when observed per-

pendicular or parallel to the direction of the grooves.42 The

contact angles measured from the direction parallel to the

grooves were larger than those measured from the perpen-

dicular direction, in a study by Zhao et al.43 Thus far,

experimental studies on anisotropic wetting have been

conducted using microliter-volume drops.44 In our study,

we decided that it was reasonable to assume that the 1-ml

droplet size was millimetre scaled and much larger than the

dimension of surface asperities. Our results using titanium

surfaces are consistent with previous studies,45 decreasing

groove widths and spacing, or increasing groove depth

amplified the anisotropy for the contact angle (equilibrium

contact angle). These trends are likely due to an increasing

energy barrier. It should be pointed out that many studies

demonstrate that surface topographic structure43,46 and

chemical composition47 can result in anisotropic wettability.

This wettability can be further attributed to the difference in

the energy barrier between different directions due to

geometrical and chemical heterogeneity. In our study,

geometrical parameters such as groove width, depth and

spacing played an important role in unique anisotropic

wetting behaviour of the surface. For microgrooved titanium

surfaces, decreasing groove width and increasing depth

amplified the surface hydrophobicity.

4.3. Adsorption of plasma FN on microgroove surface

Plasma FN circulates in the blood and functions during early

wound-healing responses. Various techniques have been

developed for their study in protein–material interactions or

to quantify protein adsorption. The most straightforward

method to quantify protein is by labelling with fluorescent

probes.48 Another method to follow adsorption is verified by

measuring the ELISA by light absorbance.15,36,49 In our study,

we used fluorescent probes and ELISA measurements to

determine the extent of FN absorption on different micro-

groove surfaces. T15/5 and T15/10 had the greatest surface

hydrophobicity, but absorbed the highest plasma FN. These

data showed that hydrophobic microgroove surfaces with

narrower width might be considered more protein-adsorbent

than hydrophilic surfaces with larger groove widths. Surface

chemistry, topography and wettability influence the changes

in protein absorption.18,50,51 Strong hydrophobic forces found

near these surfaces are in direct contrast to the repulsive

solvation forces arising from strongly bound water at the

hydrophilic surface.52–54 Interactions between proteins and

surfaces include van der Waal’s and electrostatic interactions,

hydrogen bonding, and hydrophobic interactions.54 Hydro-

phobic interactions are likely to be the principal binding forces

between proteins and surfaces.55 Water molecules play a

significant role in protein adsorption, as proteins are generally

hydrophilic in character. When the surface is hydrophobic, the

water dipoles near it have to be ordered. Upon protein

adsorption, these ordered molecules are released to the bulk

solution increasing the entropy of the system, thereby

Fig. 6 – Immunofluorescence staining of HGFs, CLSM overlay of triple stain with DAPI (blue), cytoskeleton-actin stress fibres

(red) and FN (green)(400T). (A) 2D image of the smooth surface (T0). Cellular FN was found as a coating on cell surfaces and

in the ECM. FN coating on the cell surface assembled into a complex fibrillar network that resembled a web throughout the

cell (red arrows); extracellular FN assembled near the cell (white arrows). (B–G) 3D microgroove samples are shown as

overlay pictures (from top to bottom) for each microgroove type. (b–g) 3-D microgroove samples are shown as a series of

pictures in 1 mm axial steps (from top to bottom) for each microgroove type.

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 1 1117

decreasing free energy, and enhancing adhesion on hydro-

phobic surfaces.18,54,55

4.4. Cellular FN assembly on microgroove surface

All microgroove surfaces showed successful contact guidance

for HGFs to migrate through the groove. Cellular FN was then

synthesized and assembled by the cells as they migrated.

Fibroblasts depend on classical focal adhesions (FAs) to the

substrate, which are mediated by integrin a5b1. On the

intracellular side this interacts with the actin cytoskeleton;

on the extracellular side, ligands that anchor to integrin

include FN. FAs function as both adhesion and signal

transductors.56 Cell-generated contractile forces generated

from the stiffness or topography of the cell attachment to a

substrate is of significance in our study.27–29,57 Within the

microgroove environment, cells are affected by signals from

the ECM and neighbouring cells.58 Microgrooves lead to cell

elongation; the mechanical stresses stemming from the

microgrooves result in mechano-sensitivity signals through

Fig. 7 – (A) Relative gene expression of FN in HGFs on different surfaces after day 3 of incubation. The results are expressed

as a ratio of the quantified gene expression levels to the reference protein, b-actin, followed by a standardization of the

expressions on the control surface; T0 as reference (one-way ANOVA, ***p < 0.001, **p < 0.01, *p < 0.05, mean W SD, N = 3). (B)

Result from Western blotting for the expression of cellular FN after 3 days of culture on each surface; comparison of protein

quantity on each surface. The results are expressed as a ratio of the quantified protein expression levels to the reference

protein, GAPDH, followed by a standardization of the expressions on the control surface (T0), as reference (one-way

ANOVA, ***p < 0.001, **p < 0.01, *p < 0.01, mean W SD, N = 3).

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 11118

FAs. The feed back of ‘inside-out’ signalling regulates FN-

matrix secretion and assembly.59–61 Mechanical stresses

produced by cytoskeletal elements provide a mechanism for

regulating matrix dynamics and are required for FN assem-

bly,30–32 a mechanism of bio-mechanotransduction especial-

ly,57,60 We can infer that the quantity of cellular FN expression

in response to different mechanical stresses of the cell

may come from the different dimensions of a microgroove.

From the CLSM we found that T60, still smaller than HGF,

successfully guided the cells to attach along the microgroove

ridge. The 60-mm width is enough to guide cell attachment and

proliferation in the groove. The cells on this substrate may

have received much higher mechanical stresses than on other

grooves including protruding and bending stress from the

groove ridge during cell attachment. These stresses cause

groove sidewall and cell–cell stresses as the cells migrate into

the groove; similar mechanical stresses have been described

before.62,63 However, the T15 groove width was too narrow for

cells to migrate into the groove to set up a full 3-D

environment. In these circumstances, most mechanical

stresses came only from the ridge protrusion and a lack of

crushing stress from the profile of the groove and the

intercellular extrusion. Although the width of 15 mm was

much smaller than the cell, this surface size received much

more protrusion than the surface with a microgroove width of

30 mm. However T30 was between T15 and T60 in terms of how

much a cell could migrate into the microgroove (Fig. 6). So the

mechanical stresses for this group may have been less than

that for the T60 attaching cells.

Cells on the T60 surface received the highest mechanical

stresses, and cellular FN expression was greater than on any

other surface (Fig. 7). For cells attaching to the T60 surface, the

depth of the groove did not make much difference in cellular

expression of FN. However, the narrower microgrooves (T15

and T30) with a shallower depth (5 mm) had greater expression

of cellular FN than the 10 mm depths (Fig. 7(B)). We can infer that

a smaller ridge platform with deeper grooves might result in

poor adhesion for the cell and poor formation of FAs. The

corresponding reduction of signal transmission of mechanical

stresses would lead to lower expression of cellular FN. It would

seem reasonable to suggest that the effects on cellular

expression of FN in response to the topography of the substrate

surface, is a response to the forces induced upon them and is

scalable depending on width and depth of the groove.

4.5. Plasma FN and cellular FN have differentmicrogroove reactions

We found that the most hydrophobic samples (T15/5 and T15/

10) absorbed the greatest amount of plasma FN but relatively

low cellular FN. T60/5 and T60/10 had the opposite tendency –

a greater tendency to express cellular FN but a lower

absorbency of plasma FN. Although the two FN forms are

not identical, they appear to be interchangeable in some

respects. Plasma FN assembles into preexisting or newly

assembled cellular FN matrices.10,26 Plasma and cellular FN

could potentially perform similar, but not necessarily

the same, functions. Plasma FN is known to enhance the

attachment of cells15,16; cellular FN is a basic requirements for

cell attachment. Cellular FN-deficient fibroblasts have defects

in adhesions and mechanotransduction.64 There are also

studies showing that conditional plasma FN knockout mice

have normal wound healing and hemostasis, suggesting

possible compensation by cellular isoforms of FN.65,66

Moreover, the observation that wider grooves are effective

in stimulating cell proliferation in vitro.8

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 1 1119

Our data also indicate that the 60-mm grooves were best at

inducing cells to secrete cellular FN. This greater secretion

may compensate for the decreasing absorption of plasma FN,

since it has been shown that cellular FN function can

compensate for plasma FN.

5. Conclusion

Groove widths of 60 mm, 30 mm, and 15 mm, and groove depths

of 5 mm and 10 mm were fabricated. Narrower and deeper

microgrooves amplified surface hydrophobicity. These results

have implications for the design and application of anisotropic

wetting surfaces.

The effects of microgroove width on plasma FN and cellular

FN were quite different: wide microgrooves were favourable

for adhesion of cellular FN while narrower grooves were

favourable for plasma FN. Microgroove depth did not affect

plasma FN absorption. While groove depth did not effect

cellular FN expression to wide microgrooves (T60), narrow

microgrooves of 5-mm depth (T15/5 and T30/5) resulted in

higher cellular FN adhesion than deeper grooves of the same

width. In general, forces generated from the stiffness or

topography of the cells during attachment to a substrate is of

some considerable biological significance. These parameters

need to be considered during the design of 3D biomaterials as

they provide a means to influence the ‘‘mechanical stress

effect’’ on cells and therefore their physiological and bio-

chemical behaviour.

Due to the adaptability and functionality of FN, our results

pertain to the design of implant surfaces. It seems that the

microgroove width of 60 mm with a depth of either 5 mm or

10 mm would be the best choice for the transmucosal part of

the dental implant.

Acknowledgments

This work was supported by the Science and Technology

Project of Fujian Province (2010H6009) and Provincial

Development and Reform Commission (NDRC) project

([2010]1002). We gratefully thank the staff of the MEMS

Research Center of Xiamen University for fabricating

smooth and microgroove samples used in this study. We

also thank Mr. Tao Zhang and Mr. Zuanfang Li for their

technical assistance.

Appendix A. Supplementary data

Supplementary material related to this article can be

found, in the online version, at http://dx.doi.org/10.1016/

j.jdent.2013.08.004.

r e f e r e n c e s

1. Welander M, Abrahamsson I, Berglundh T. The mucosalbarrier at implant abutments of different materials. ClinicalOral Implants Research 2008;19:635–41.

2. Al-Radha AS, Pal A, Pettemerides AP, Jenkinson HF.Molecular analysis of microbiota associated with peri-implant diseases. Journal of Dentistry 2012;40:989–98.

3. Rompen TE, Domken O, Degidi M, Pontes AE, Piattelli A. Theeffect of material characteristics, of surface topography andof implant components and connections on soft tissueintegration: a literature review. Clinical Oral Implants Research2006;2:55–67.

4. Jansen JA, den Braber ET, Walboomers XF, de Ruijter JE. Softtissue and epithelial models. Advances in Dental Research1999;13:57–66.

5. Park TH, Shuler ML. Integration of cell culture andmicrofabrication technology. Biotechnology Progress2003;19:243–53.

6. Gehrke TA, Walboomers XF, Jansen JA. Influence oftransforming growth factor-beta3 on fibrous capsuleformation around microgrooved subcutaneous implantsin vivo. Tissue Engineering 2000;6:505–17.

7. Kokubu E, Hamilton DW, Inoue T, Brunette DM. Modulationof human gingival fibroblast adhesion, morphology,tyrosine phosphorylation, and ERK 1/2 localization onpolished, grooved and SLA substratum topographies. Journalof Biomedical Materials and Research A 2009;91:663–70.

8. Kim SY, Oh N, Lee MH, Kim SE, Leesungbok R, Lee SW.Surface microgrooves and acid etching on titaniumsubstrata alter various cell behaviors of cultured humangingival fibroblasts. Clinical Oral Implants Research2009;20:262–72.

9. Pivodova V, Frankova J, Ulrichova J. Osteoblast and gingivalfibroblast markers in dental implant studies. BiomedicalPapers of the Medical Faculty of the University Palacky OlomoucCzech Republic 2011;155:109–16.

10. To WS, Midwood KS. Plasma and cellular fibronectin:distinct and independent functions during tissue repair.Fibrogenesis and Tissue Repair 2011;4:1–17.

11. Winters BS, Raj BK, Robinson EE, Foty RA, Corbett SA. Three-dimensional culture regulates Raf-1 expression to modulatefibronectin matrix assembly. Molecular Biology of the Cell2006;17:3386–96.

12. Peters DM, Portz LM, Fullenwider J, Mosher DF. Co-assemblyof plasma and cellular fibronectins into fibrils in humanfibroblast cultures. Journal of Cell Biology 1990;111:249–56.

13. Molino PJ, Higgins MJ, Innis PC, Kapsa RM, Wallace GG.Fibronectin and bovine serum albumin adsorption andconformational dynamics on inherently conductingpolymers: a QCM-D study. Langmuir 2012;28:8433–45.

14. Zhu XD, Zhang HJ, Fan HS, Li W, Zhang XD. Effect of phasecomposition and microstructure of calcium phosphateceramic particles on protein adsorption. Acta Biomaterialia2010;6:1536–41.

15. Tsai WB, Ting YC, Yang JY, Lai JY, Liu HL. Fibronectinmodulates the morphology of osteoblast-like cells (MG-63)on nano-grooved substrates. Journal of Materials ScienceMaterials in Medicine 2009;20:1367–78.

16. Pendegrass CJ, El-Husseiny M, Blunn GW. The developmentof fibronectin-functionalised hydroxyapatite coatings toimprove dermal fibroblast attachment in vitro. Journal ofBone and Joint Surgery British Volume 2012;94:564–9.

17. Vogler EA. Protein adsorption in three dimensions.Biomaterials 2012;33:1201–37.

18. Sethuraman A, Han M, Kane RS, Belfort G. Effect of surfacewettability on the adhesion of proteins. Langmuir2004;20:7779–88.

19. Fischer M, Sperling C, Werner C. Synergistic effect ofhydrophobic and anionic surface groups triggers bloodcoagulation in vitro. Journal of Materials Science Materials inMedicine 2010;21:931–7.

20. Vadillo-Rodrıguez V, Pacha-Olivenza MA, Gonzalez-MartınML, Bruque JM, Gallardo-Moreno AM. Adsorption behavior

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 11120

of human plasma fibronectin on hydrophobic andhydrophilic Ti6Al4V substrata and its influence on bacterialadhesion and detachment. Journal of Biomedical Materials andResearch A 2013;101:1397–404.

21. Campillo-Fernandez AJ, Unger RE, Peters K, Halstenberg S.Analysis of the biological response of endothelial andfibroblast cells cultured on synthetic scaffolds with varioushydrophilic/hydrophobic ratios: influence of fibronectinadsorption and conformation. Tissue Engineering Part A2009;15:1331–41.

22. Wei J, Igarashi T, Okumori N, Igarashi T, Maetani T, Liu B,et al. Influence of surface wettability on competitive proteinadsorption and initial attachment of osteoblasts. BiomedicalMaterials 2009;4:1–7.

23. Lee SW, Lee MH, Oh N, Park JA, Leesungbok R, Ahn SJ.Correlation between surface hydrophilicity and osteoblasticdifferentiation on microgrooved titanium substrata. Journalof Oral Implantology 2012;38:11–9.

24. Bliznyuk O, Jansen HP, Kooij ES, Zandvliet HJ, Poelsema B.Smart design of stripe-patterned gradient surfaces tocontrol droplet motion. Langmuir 2011;27:11238–45.

25. Iuliano DJ, Saavedra SS, Truskey GA. Effect of theconformation and orientation of adsorbed fibronectin onendothelial cell spreading and the strength of adhesion.Journal of Biomedical Materials Research 1993;27:1103–13.

26. Ruoslahti E. Fibronectin in cell adhesion and invasion.Cancer and Metastasis Reviews 1984;3:43–51.

27. Lim JY, Donahue HJ. Cell sensing and response to micro-and nanostructured surfaces produced by chemical andtopographic patterning. Tissue Engineering 2007;13:1879–91.

28. Ventre M, Causa F, Netti PA. Determinants of cell-materialcrosstalk at the interface: towards engineering of cellinstructive materials. Journal of the Royal Society Interface2012;9:2017–32.

29. Reilly GC, Engler AJ. Intrinsic extracellular matrix propertiesregulate stem cell differentiation. Journal of Biomechanics2010;43:55–62.

30. Smith ML, Gourdon D, Little WC, Kubow KE, Eguiluz RA,Luna-Morris S, et al. Force-induced unfolding of fibronectinin the extracellular matrix of living cells. PLoS Biology2007;5:2243–54.

31. Lemmon CA, Chen CS, Romer LH. Cell traction forces directfibronectin matrix assembly. Biophysical Journal 2009;96:729–38.

32. Singh P, Carraher C, Schwarzbauer JE. Assembly offibronectin extracellular matrix. Annual Review of Cell andDevelopmental Biology 2010;26:397–419.

33. Abdulmajeed AA, Walboomers XF, Massera J, Kokkari AK,Vallittu PK, Narhi TO. Blood and fibroblast responses tothermoset BisGMA-TEGDMA/glass fiber-reinforcedcomposite implants in vitro. Clinical Oral Implants Research2013;3:1–9.

34. Bruckmann C, Walboomers XF, Matsuzaka K, Jansen JA.Periodontal ligament and gingival fibroblast adhesion todentin-like textured surfaces. Biomaterials 2005;26:339–46.

35. Sousa SR, Bras MM, Moradas-Ferreira P, Barbosa MA.Dynamics of fibronectin adsorption on TiO2 surfaces.Langmuir 2007;23:7046–54.

36. Puckett SD, Taylor E, Raimondo T, Webster TJ. Therelationship between the nanostructure of titaniumsurfaces and bacterial attachment. Biomaterials 2010;31:706–13.

37. Pool M, Thiemann J, Bar-Or A, Fournier AE. NeuriteTracer: anovel Image J plug in for automated quantification ofneurite outgrowth. Journal of Neuroscience Methods2008;168:134–9.

38. Hannigan A, Lynch CD. Statistical methodology in oral anddental research: pitfalls and recommendations. Journal ofDentistry 2013;41:385–92.

39. Hasirci V, Pepe-Mooney BJ. Understanding the cell behavioron nano-/micro-patterned surfaces. Nanomedicine2012;7:1375–89.

40. Clark P, Connolly P, Curtis AS, Dow JA, Wilkinson CD.Topographical control of cell behavior: I. Simple step cues.Development 1987;99:439–48.

41. Kaiser JP, Reinmann A, Bruinink A. The effect of topographiccharacteristics on cell migration velocity. Biomaterials2006;27:5230–41.

42. Chen Y, He B, Lee J, Patankar NA. Anisotropy in the wettingof rough surfaces. Journal of Colloid and Interface Science2005;281:458–64.

43. Zhao Y, Lu Q, Li M, Li X. Anisotropic wetting characteristicson submicrometer-scale periodic grooved surface. Langmuir2007;23:6212–7.

44. Yang J, Rose FR, Gadegaard N, Alexander MR. Effect ofsessile drop volume on the wetting anisotropy observed ongrooved surfaces. Langmuir 2009;25:2567–71.

45. Li W, Fang G, Li Y, Qiao G. Anisotropic wetting behaviorarising from superhydrophobic surfaces: parallel groovedstructure. Journal of Physical Chemistry B 2008;112:7234–43.

46. Bukowsky C, Torres JM, Vogt BD. Slip-stick wetting and largecontact angle hysteresis on wrinkled surfaces. Journal ofColloid and Interface Science 2011;354:825–31.

47. Neuhaus S, Spencer ND, Padeste C. Anisotropic wetting ofmicrostructured surfaces as a function of surface chemistry.ACS Applied Materials &Interfaces 2012;4:123–30.

48. Pankov R, Momchilova A. Fluorescent labeling techniquesfor investigation of fibronectin fibrillogenesis (labelingfibronectin fibrillogenesis). Methods in Molecular Biology2009;522:261–74.

49. Salim M, O’Sullivan B, McArthur SL, Wright PC.Characterization of fibrinogen adsorption onto glassmicrocapillary surfaces by ELISA. Lab on a Chip 2007;7:64–70.

50. Roach P, Farrar D, Perry CC. Interpretation of proteinadsorption: surface-induced conformational changes.Journal of the American Chemical Society 2005;127:8168–73.

51. Glassford S, Chan KL, Byrne B, Kazarian SG. Chemicalimaging of protein adsorption and crystallization on awettability gradient surface. Langmuir 2012;28:3174–9.

52. Noh H, Vogler EA. Volumetric interpretation of proteinadsorption mass and energy balance for albumin adsorptionto particulate adsorbents with incrementally increasinghydrophilicity. Biomaterials 2006;27:5801–12.

53. Xu LC, Siedlecki CA. Effects of surface wettability andcontact time on protein adhesion to biomaterial surfaces.Biomaterials 2007;28:3273–83.

54. Xu LC, Siedlecki CA. Atomic force microscopy studies of theinitial interactions between fibrinogen and surfaces.Langmuir 2009;25:3675–81.

55. dos Santos EA, Farina M, Soares GA, Anselme K. Surfaceenergy of hydroxyapatite and beta-tricalcium phosphateceramics driving serum protein adsorption and osteoblastadhesion. Journal of Materials Science Materials in Medicine2008;19:2307–16.

56. Goldmann WH. Mechanotransduction and focal adhesions.Cell Biology International 2012;36:649–52.

57. Chen CS. Mechanotransduction – a field pulling together?Journal of Cell Science 2008;121:3285–92.

58. Harjanto D, Zaman MH. Modeling extracellular matrixreorganization in 3D environments. PLoS ONE 2013;8:1–11.

59. De R, Zemel A, Safran SA. Theoretical concepts and modelsof cellular mechanosensing. Methods in Cell Biology2010;98:143–75.

60. Daley WP, Peters SB, Larsen M. Extracellular matrixdynamics in development and regenerative medicine.Journal of Cell Science 2008;121:255–64.

61. Geiger B, Bershadsky A, Pankov R, Yamada KM.Transmembrane crosstalk between the extracellular matrix

j o u r n a l o f d e n t i s t r y 4 1 ( 2 0 1 3 ) 1 1 0 9 – 1 1 2 1 1121

– cytoskeleton crosstalk. Nature Reviews Molecular Cell Biology2001;2:793–805.

62. Walboomers XF, Jansen JA. Cell and tissue behavior onmicro-grooved surfaces. Odontology 2001;89:2–11.

63. Seo CH, Furukawa K, Montagne K, Jeong H, Ushida T. Theeffect of substrate microtopography on focal adhesionmaturation and actin organization via the RhoA/ROCKpathway. Biomaterials 2011;32:9568–75.

64. Lutz R, Sakai T, Chiquet M. Pericellular fibronectin isrequired for RhoA-dependent responses to cyclic strain infibroblasts. Journal of Cell Science 2010;123:1511–21.

65. Rohwedder I, Montanez E, Beckmann K, Bengtsson E,Duner P, Nilsson J, et al. Plasma fibronectindeficiency impedes atherosclerosis progression andfibrous cap formation. EMBO Molecular Medicine 2012;4:564–76.

66. Sakai T, Johnson KJ, Murozono M, Sakai K, Magnuson MA,Wieloch T, et al. Plasma fibronectin supports neuronalsurvival and reduces brain injury following transientfocal cerebral ischemia but is not essential for skin-wound healing and hemostasis. Nature Medicine2001;7:324–30.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具