Laser-engineered topography: correlation between structuredimensions and cell control

Sabrina Schlie • Elena Fadeeva • Anastasia Koroleva •

Boris N. Chichkov

Received: 27 February 2012 / Accepted: 27 July 2012 / Published online: 10 August 2012

� Springer Science+Business Media, LLC 2012

Abstract Topographical cues have a significant impact on

cell responses and by this means, on the fabrication of

innovative implant materials. However, analysis of cell-

topography interactions in dependence of the surface feature

dimensions is still challenging due to limitations in the fab-

rication technology. Here, we introduce surface structuring

via picosecond laser systems, which enable a fast production

of micro-sized topologies. Changes in the processing

parameters further control the feature sizes of so-called

spikes. Using surfaces with big and small spike-to-spike-

distances for comparisons, we focussed on cell adhesion via

extracellular matrix adsorption and focal adhesion com-

plexes, morphology, localisation and proliferation of fibro-

blasts. The observed cell control was dependent on a

turnover point related to the structure dimensions: only big

spike-to-spike-distances reduced cell behaviour. Therefore,

this technology offers a platform to study cell and tissue

interactions with a defined microenvironment.

1 Introduction

The natural environment of cells consists of a defined

architecture of the extracellular matrix and neighbouring

cells providing anchorage points and biochemical signals

for cell functions and survival. When implants have to

substitute tissue defects, the engineered constructs have to

copy the state of art within the tissue—otherwise tissue

regeneration and reconstruction cannot be guaranteed.

To better represent the geometry, chemistry, and signal-

ing environment, different biological, chemical, and

physical approaches have been made for material function-

alization. In the first place, a biocompatible basis material

has to be selected, which mimics the mechanical, structural,

biochemical and chemical properties the best—but which is

also applicable for functional modifications. Depending on

the biomedical application, metals, ceramics, polymers or

composites are used. However, the critical parameter of the

material is its surface, since it is crucial in controlling

interactions between cells and the substrate [1]. This inter-

action occurs in a specific order: the extracellular matrix

associates with the material surface providing ligands for cell

binding with integrin receptors. Afterwards, the formation of

focal adhesion complexes initiates signaling cascades guid-

ing proliferation, differentiation, survival and others [2].

Finding a material-processing technology, which enables the

systematic study and understanding of parameters that

govern cell behaviour, would be a huge benefit in tissue

engineering.

One of the critical surface parameter is its topography.

Many studies describe that cells recognize the topographical

features of their environment such as the fibres of the

extracellular matrix with 10 to 300 nm in diameter [3]. It has

been demonstrated that structures like groove or pits in

micro- and nanometer scale affect cell responses [4–6].

However, analysis of one type of surface feature with vari-

able dimensions is still challenging. Whereas groove-struc-

tures or pores are easier to be established in distinct scales,

the production of more complicated textures is limited by the

fabrication technology. Additional standards of the tech-

nique refer to manufacturing in a time- and cost-effective

manner. Therefore, the recent proceedings in nanotechnol-

ogy make laser-technologies very attractive for a topo-

graphical functionalization of biomaterials. In general, laser

systems offer a flexible and controllable structuring of

material surfaces. The usage of ultrashort laser-pulses

S. Schlie (&) � E. Fadeeva � A. Koroleva � B. N. Chichkov

Laser Zentrum Hannover e.V., Hollerithallee 8,

30419 Hannover, Germany

e-mail: s.schlie@lzh.de

123

J Mater Sci: Mater Med (2012) 23:2813–2819

DOI 10.1007/s10856-012-4737-9

further enables a better resolution and reduces heat-affected

zones. In previous studies we showed, that femtosecond

lasers can be applied for material ablation to texture the

surface of different metals such as titanium, platinum or

silicon [7, 8]. Laser-structuring of polymers is also possible;

however, ablation debris remaining on the surface may cause

cytotoxic effects, so that microreplication of laser-generated

master samples has to be used alternatively [9, 10]. In

addition we found a correlation between surface structuring

and increase in material wettability without influencing the

chemical composition of the material.

Out of all manufactured topographies, so-called spike

structures in micrometer scale were analyzed most. Con-

cerning cell interactions, such spikes even allow a selective

cell control: inhibiting fibroblasts, while stimulating neu-

roblastoma cells [9]. Since fibroblasts participate in scar

tissue formation surrounding the implanted material, their

growth inhibition is one major goal in tissue engineering

[11]. Even though the laser technologies permit the fabri-

cation spikes in silicon (Si) with different size dimensions, a

detailed study of fibroblast inhibition in dependence of the

spike sizes has not been performed so far. However, this

knowledge could be a vantage point for any other desired

biomedical application assuming that a selective cell control

depends on the provided topography. Additionally, the

interaction between the presented microtopography of a

biomaterial and extracellular matrix needed for cell attach-

ment is poorly understood. Therefore, this study shall give

an insight in the potential of laser technologies for defined

surface texturing and in a systematic analysis of tissue/cell

responses to a given microenvironment.

2 Materials and methods

2.1 Surface fabrication and characterisation

Spike structures were fabricated by ablation of single-

crystal p–type Silicon (110) samples (Si) in SF6 gas

atmosphere with infrared picosecond laser pulses. In our

experiments we used a picosecond laser system Rapid

(Lumera Laser GmbH, Kaiserslautern, Germany) with

amplifier system miniVan (neoLASE GmbH, Hannover,

Germany). This system delivers 8 ps laser pulses at 1.064

nm and repetition rate up to 640 kHz by output power of

16 W. For structuring, the samples were placed in a vac-

uum chamber evacuated down to a residual pressure of

10 m Torr, and backfilled with 500 Torr SF6. A motorised

translation stage (Kugler GmbH, Salem, Germany) was

used for samples positioning and translation. To fabricate

the spike structure, the silicon surfaces were uniformly

irradiated with 500 laser pulses at laser fluences of

2.6 J/cm2 for small and 12 J/cm2 for big spikes, respectively.

Size dimensions, described by spike-to-spike distances,

were quantified by the use of ImageJ software (http://

rsbweb.nih.gov/ij/). With the help of the line selection tool,

a straight line was placed from one spike tip to another of

scanning electron microscope (SEM) images. Automati-

cally, the line length is given in pixel scale, which can be

converted into lm dimensions. The results were given as

average ± standard derivation of at least 250 different

distances of both topologies. In the following, the struc-

tures are entitled small and big spikes, respectively.

2.2 Fibronectin adsorption

To analyze fibronectin adsorption as a component of the

extracellular matrix in dependence of the spike dimensions,

the enzyme-linked immunosorbent assay (ELISA) was per-

formed. With respect to the control, a coating-concentration

dependency (5, 10, and 20 lg/cm2) was evaluated. For each

sample, the relative fluorescence intensity, which is pro-

portional to the amount of antibody-binding, was determined

as a function of ligand adsorption. All antibodies were solved

in 0.05 % Tween/phosphate buffer saline (PBS). After

30 min fibronectin coating at room temperature and washing

with PBS, the primary antibody-solution was added at 37 �C

for 1 h. Following three washing steps with PBS 5 min each,

the secondary antibody solution (anti-rabbit IgG-alkaline

phosphatase conjugated) was incubated at 37 �C for 1 h.

Before adding the substrate 4-methylumbelliferyl phosphate

for 20 min at room temperature, the samples were washed

three times with PBS 5 min each. Finally, reaction super-

natants were transferred to a black 96-well plate and fluo-

rescence (365 nm excitation/450 nm emission) was read

using multimode microplate reader Mithras LB 940 (Bert-

hold Technologies, Bad Wildbad, Germany). The back-

ground signal referred to treatment specific progression

without ligand-coating and was subtracted from the corre-

sponding intensities. For statistical reasons, the measure-

ment was repeated six times and given as average.

2.3 Cell culture and proliferation

Before use the laser-structured Si samples were sterilized

under UV light for 30 min. Glass slides served as a control.

Fibroblasts were cultivated on the different samples in a cell

incubator at 37 �C and 5 % CO2 atmosphere (Thermo

Electron Corporation, Bonn, Germany) in Dulbecco’s

Modified Eagles Medium (DMEM/F-12, Lonza, Basel,

Switzerland) supplemented with 10 % fetal calf serum and

antibiotics.

To analyze topographical effects on cell proliferation, the

cell density of fibroblasts was determined when cultivated on

small and big Si spikes in comparison to the control. Using the

cell counter Casy TT (Roche, Mannheim, Germany), the

2814 J Mater Sci: Mater Med (2012) 23:2813–2819

123

adherent cells were trypsinized, collected, centrifuged at

800 g for 5 min, and counted after 48 and 96 h. For a better

comparison between the experiments, the cell density was

normalized in percent on the seeding density of 3 9 105 cells/

ml at time 0 h. The results were given as average ± standard

error of mean (s.e.m.) of four independent experiments.

2.4 Cell imaging

After 24 h cultivation time, visualization of fibroblasts

grown on the different spikes followed two procedures:

scanning electron microscopy (SEM) and fluorescence. All

methods focussed on cell alignment, localisation, mor-

phology, and adhesion in dependence of the presented

topography. If not specified, stains were purchased from

Sigma Aldrich (Taufkirchen, Germany).

Concerning SEM-preparation, the samples were washed

with PBS for 5 min, fixed in 3 % glutaraldehyde (in PBS)

for 15 min, and washed three times. Following a second

fixation in 2 % osmiumtetroxid (in PBS) and five washing

steps in water, the samples were dehydrated through series

of ethanol concentration (from [%] 30, 50, 70, 90, 95 and

three times 100, 10 min each). At last the samples were

dried in a critical-point dryer and coated with a 5 nm gold

layer. With the help of an SEM, images of the topographies

were recorded and the cells were analyzed.

Fluorescence staining of the nuclei and actin filaments

was performed simultaneously with antibody staining of

focal adhesion complexes focussing on vinculin and focal

adhesion kinase (FAK) p-Tyr397. After washing with PBS,

fibroblasts were fixed with 4 % paraformaldehyde solved

in PBS at 4 �C for 20 min. To inhibit unspecific antibody-

binding, the cells incubated in a 2 % bovine serum-PBS

solution at 37 �C for 30 min. Following two washing steps

5 min each, the primary antibodies (solved in 0.3 % Tri-

ton-X100 in PBS) were added at 4 �C over night. Before

incubating in the secondary anti-rabbit Alexa Fluor anti-

body solution in 0.3 % Triton-X100 in PBS (Invitrogen,

Darmstadt, Germany) including Hoechst 33342 for nucleus

staining at 37 �C for 1 h, the cells were washed twice

5 min each. Following another washing step, phalloidin-

Atto 488 was used to stain actin filaments. After 20 min

incubation time, the cells were washed and kept in PBS.

Images were recorded using a fluorescence microscope

(Nikon TE 2000-E, Nikon, Dusseldorf, Germany).

3 Results

3.1 Structure dimensions

The generated spike structures were uniformly arranged on

the entire surface. By adjusting the process parameter laser

fluence from 2.6 to 12 J/cm2, different size dimensions

were achieved. A quantification of both topologies

revealed, that the spike-to-spike distances were arranged

between 2.08 lm ± 0.04 (small spikes, Fig. 1a) and

6.74 lm ± 0.85 (big spikes, Fig. 1b).

3.2 Fibronectin adsorption onto the spikes

The adsorption of fibronectin was quantified as a function

of ligand-coating concentrations via ELISA. As shown in

Fig. 2, fibronectin adsorption was improved onto the small

spikes, but significantly reduced on the big spikes when

compared with the control surface. For instance, the rela-

tive fluorescence intensity of 2.505 on the control was

increased to 3.857 on the small, but reduced to 769 on the

big spikes when coated with 10 lg/cm2.

3.3 Spike dimensions and cell behaviour

The SEM-images in Fig. 3 point out, that fibroblasts formed

a monolayer on the unstructured silicon surface. Concerning

the spike structures, differences occurred between the small

(Fig. 3a) and big (Fig. 3b) surface features. On the small

Fig. 1 SEM-images of laser-generated spike structures in Si with different size dimensions: small (a) and big (b)

J Mater Sci: Mater Med (2012) 23:2813–2819 2815

123

spikes, fibroblasts proliferated to a high range, presented a

normal elongated morphology, and orientate on the struc-

tures. On the big spikes, cell growth was reduced and

fibroblasts were more rounded.

The obtained differences in the growth behaviour pat-

tern were supported by the detailed estimation of prolif-

eration (Fig. 4). After 96 h cultivation time, the cells

reached 546 % on the control surface. This rate was

comparable on the small spikes with a value of 489 %. In

accordance with Fig. 3b, on the big spikes fibroblasts

reduced their proliferation significantly to 394 %.

Different staining procedures were performed to analyze

cell morphology and cell attachment (Fig. 5). Indepen-

dently from the surface, the shape of the nuclei (blue) was

not negatively affected. On the small spikes, actin filaments

(green) were well organized, and the fibroblasts formed

many cellular extensions (Fig. 5a). On the contrary, fibro-

blasts formed less and shorter extensions and were more

rounded in the presence of the big spikes (Fig. 5c). Con-

cerning vinculin (red), it was less detectable on the big

spikes than on the small spikes (Fig. 5a, c). Furthermore,

fibroblasts formed less FAK (red) on the big spikes

(Fig. 5d) than on the small spikes (Fig. 5b).

4 Discussion

Since cells are very sensitive to the chemical and topo-

graphical pattern of their environment, there is a demand to

understand biomaterial-cell interactions for tissue engi-

neering applications. However, to capture the degree of

complexity which is responsible for cell guidance, a bio-

material with just one modifiable property has to be engi-

neered—in turn, the specific role of this certain parameter

can be identified. Due to the geometric architecture of the

tissue and defined scale of the extracellular matrix and

signaling molecules, focussing on the parameter ‘topogra-

phy’ is very promising.

Concerning surfaces structuring the fabrication pro-

cesses are very diverse and include photolithography,

electron beam lithography, laser holography, polymer

demixing, electrochemical porous etching, plasma treat-

ment and others [12, 13]. However, a systematic analysis of

cell responses to defined patterned materials requires a

technique, which is not limited by the resolution, repro-

ducibility, variety of texturing, production time and costs.

The usage of laser-technologies was shown to be a very

reliable method. Via material ablation different topologies

can be generated without changes in its biocompatibility or

causing side-effects like mechanical damages of the sub-

strate or heat-affected zones [7, 8]. Changes in the micro-

fabrication parameters such as laser fluence enable a

control over surface feature dimensions [9]. To overcome

the drawbacks of commonly used femtosecond lasers

related to the production time, picosecond lasers were

Fig. 2 Fibronectin adsorption onto small and big Si-spike structures

in comparison to the control quantified by ELISA as a function of

ligand-coating concentrations. The results are given as average of

relative fluorescence intensity of six independent measurements

Fig. 3 SEM-images of fibroblasts cultivated on a small and b big Si-spikes

2816 J Mater Sci: Mater Med (2012) 23:2813–2819

123

applied for surface structuring in this study. As shown in

Fig. 1, microsized spike structures in silicon were pro-

duced. In comparison the femtosecond laser-generated

silicon spikes in [9], we indeed confirmed a reduction of

production time up to 50 times. These advantages make

picosecond laser very attractive for surface structuring.

Whether the spike structures in silicon manipulate cell/

tissue responses, was analyzed with respect to two different

spike features named ‘small’ (Fig. 1a) and ‘big’ spikes

(Fig. 1b). The parameter of interest spike-to-spike distance

was arranged between 2.08 lm ± 0.04 and 6.74 lm ±

0.85, respectively.

Before cells can attach to implanted materials, it has to

be surrounded by the extracellular matrix, which in turn

provides anchoring points for cell binding [2]. In this study,

the adsorption of fibronectin was investigated. We found

that the adsorption is improved on the small spikes and

reduced on the big spikes (Fig. 2). An explanation of this

observation is difficult, since an increase of absorption on

both surface features should have been more plausible—

correlating with an increase of the entire surface area after

structuring. It can be assumed that the topography affects

the conformation of fibronectin, which was shown to be

very sensitive to the substrate [14, 15]. That might be the

reason why our performed ELISA detected differences.

To follow the question how cells respond to both spike

topographies, we concentrated on adhesion via analyzing

focal adhesion complexes, localisation, morphology via the

Fig. 4 Proliferation profiles of fibroblasts in dependence of small and

big Si-spikes in comparison to the control over 96 h cultivation time.

The results are given as average ± s.e.m. of four independent

measurements, normalized in percent on the seeding cell density of

3 9 105 cells/ml at t = 0 h

Fig. 5 Fluorescence images of

fibroblasts cultivated on (a,

b) small and (c, d) big Si-spikes

showing the nucleus (blue),

actin filaments (green, a, c),

vinculin (red, a, c) and FAK

p-Tyr397 (red, b, d) (Color

figure online)

J Mater Sci: Mater Med (2012) 23:2813–2819 2817

123

cytoskeleton and proliferation. As illustrated in Figs. 3, 4,

5, the behaviour of fibroblasts could selectively be trig-

gered: small spikes did not negatively affect fibroblasts,

whereas their behaviour was significantly reduced on big

spikes. Similarly to our previous work in [9], we could

repeat that spikes in silicon control fibroblasts. In this

context, the used laser technology for structuring—either

femtosecond or picosecond lasers—did not cause a dif-

ference. However, cell control depends on the spike-to-

spike distances. Only a distance of 6.74 lm enabled posi-

tive cell responses, whereas negative responses occurred on

4.8 lm [9] and 2.08 lm. Therefore, we conclude that cell

control correlates with a turnover point of the topography.

Concerning fibroblasts it has to be arranged between 6.74

and 4.8 lm spike-to-spike distances. In addition, this

turnover point has to be cell specific, since neuroblastoma

cells responded well to 4.8 lm dimensions [9].

How cells are able to detect this turnover point, which

results in cell control, is still unsolved. Theoretically, it

depends on different factors: cell adhesion, cell morphol-

ogy, intercellular contacts or influences on cell cycle pro-

gression. Due to the complex signaling cascades that guide

cell behaviour and functions, probably several aspects

correlate with each other [2]. As a consequence of topo-

graphically-induced changes in the amount of fibronectin

adsorption and/or its conformation (Fig. 2), cell binding

has to be controlled. This assumption is in accordance with

similar effects on the formation of focal adhesion com-

plexes (Fig. 5). The stained component FAK p-Tyr397 is

only formed when ligands like fibronectin from the ECM

are bounded to integrin receptors; the phosphorylated

subtype further indicates the interaction with Src tyrosine

kinase, which is crucial for cell behaviour and further

complex stabilization [16, 17]. According to the outside-in

signaling cascades of cell attachment, a reduced associa-

tion of fibronectin caused a reduction of cell binding to the

substrate followed by a downgrade of signaling needed for

cell behaviour. In this connection, the spacing of integrin

receptors correlating with the spike-to-spike distances may

also play an important role, even though this factor is rather

dependent from nanoscaled topologies [18]. However,

since changes in the signaling cascade were not analyzed in

this study, we cannot be sure, whether this explanation and

relationship between the ECM and cell functions is the key

parameter for cell control by the topography. In case the

outside-in signaling cascades are stimulated correctly—

regardless fibronectin and FAK results (Figs. 2, 5), the

turnover point may then depend on the cell shape. On the

small spikes, fibroblasts represent an abnormal morphology

with respect to the formation of cellular extensions and the

organization of the cytoskeleton such as vinculin and actin

(Fig. 5). The thereby induced intramolecular forces may

affect other cell behaviour pattern [19]. Furthermore,

rounded cell shapes have an influence on cell–cell contacts,

which were reduced selectively (Fig. 3), and by this means

on the intercellular communication, which was shown

to be essential for cell cycle progression [20]. However,

an understanding of the parameters responsible of the

observed cell control in dependence of a microenviron-

ment, is now realizable by the presented laser technology

for material structuring in the future.

5 Conclusions

Picosecond laser-systems offer the fast fabrication of

defined surface features with controllable size dimensions.

Such dimensions were shown to represent a turnover point,

which controls the adhesion, morphology and proliferation

of cells. The used structures can potentially be applied

for fibroblast inhibition. Therefore, this technology offers

a wide platform to study material–cell-interactions in

dependence of the surface topography and includes appli-

cations in biomedicine and tissue engineering.

Acknowledgments This work was partly supported by Cluster of

Excellence Rebirth ‘‘From Regenerative Biology to Reconstructive

Therapy’’ and BMBF-project REMEDIS. The authors thank Prof.

Dr. H. Kuster, head of the Institute of Biophysics (Leibniz University

Hannover, Germany) for granting the use of the microplate reader and

fluorescence microscope.

References

1. Lee J, Cuddihy MJ, Kotov NA. Three-dimensional cell culture

matrices: state of art. Tissue Eng Part B. 2008;14(1):61–86.

2. Giancotti FG, Ruoslathi E. Integrin signalling. Science. 1999;

285:1028–32.

3. Stevens MM, George JH. Exploring and engineering the cell

surface interface. Science. 2005;310(18):1135–8.

4. Flemming RG, Murphy CJ, Abrams GA, Goodman SL, Nealey

PF. Effects of synthetic micro- and nano-structured surfaces on

cell behavior. Biomaterials. 1999;20:573–88.

5. Wilkinson CDW, Riehle M, Wood M, Gallagher J, Curtis ASG.

The use of materials patterned on a nano- and micro-metric scale

in cellular engineering. Mater Sci Eng C. 2002;19:263–9.

6. Bettinger CJ, Langer R, Borenstein JT. Engineering substrate

micro- and nanotopography to control cell funtion. Angew Chem

Int Ed Engl. 2009;48(30):5406–14.

7. Fadeeva E, Schlie S, Koch J, Chichkov BN, Vorobyev AY, Guo

C. Femtosecond laser-induced surface structures on platinum and

their effects on surface wettability and fibroblast cell prolifera-

tion. Cont Angle Wettability Adhes. 2009;6:163–71.

8. Schlie S, Fadeeva E, Ovsianikov A, Koch J, Ngezahayo A,

Chichkov BN. Laser-based nanoengineering for biomedical

applications. Photonic Nanostruct Fundam Appl. 2011;9:159–62.

9. Schlie S, Fadeeva E, Koch J, Ngezahayo A, Chichkov BN.

Femtosecond laser fabricated spike structures for selective con-

trol of cellular behavior. J Biomater Appl. 2010;25(3):217–33.

10. Koroleva A, Schlie S, Fadeeva E, Gittard SD, Ovsianikov A,

Koch J, Narayan RJ, Chichkov BN. Microreplication of laser-

fabricated surfaces and three-dimensional structures. J Opt. 2010.

doi:10.1088/2040-8978/12/12/124009.

2818 J Mater Sci: Mater Med (2012) 23:2813–2819

123

11. Anderson JM. Biological responses to biomaterials. Annu Rev

Mater Res. 2001;31:81–110.

12. Berry CC, Dalby MJ, McCloy D, Affrossman S. The fibroblast

response to tubes exhibiting internal nanotopography. Biomate-

rials. 2005;26:4985–92.

13. Choi C-H, Hagvall SH, Wu BM, Dunn JCY, Beygui RE, Kim

C-J. Cell interaction with three-dimensional sharp-tip nanoto-

pography. Biomaterials. 2007;28:1672–9.

14. Faucheux N, Tzoneva R, Nagel M-D, Groth T. The dependence

of fibrillar adhesions in human fibroblasts on substratum chem-

istry. Biomaterials. 2006;27:234–45.

15. Baujard-Lamotte L, Noinville S, Goubard F, Marque P, Pauthe E.

Kinetics of conformational changes of fibronectin adsorbed onto

model surfaces. Colloids Surf B. 2008;63:129–37.

16. Lo SH. Focal adhesions: what’s new inside. Dev Biol. 2006;

294:280–91.

17. Wozniak MA, Modzelewska K, Kwong L, Keely PJ. Focal

adhesion regulation of cell behavior. Biochim Biophys Acta.

2004;1692:103–19.

18. Selhuber-Unkel C, Erdmann T, Lopez-Garcıa M, Kessler H,

Schwarz US, Spatz JP. Cell adhesion strength his controlled by

intermolecular spacing of adhesion receptors. Biophys J. 2010;

98:543–51.

19. Ingber DE, Tensegrity I. Cell structure and hierarchical systems

biology. J Cell Sci. 2003;116:1157–73.

20. Schlie S, Mazur K, Bintig W, Ngezahayo A. Cell cycle dependent

regulation of gap junction coupling and apoptosis in GFSHR-17

granulosa cells. J Biomed Sci Eng. 2010;3:884–91.

J Mater Sci: Mater Med (2012) 23:2813–2819 2819

123