ROCK signaling in contact guidance of bone-forming cells on anisotropic Ti6Al4V surfaces

Acta Biomaterialia 7 (2011) 1890–1901

Contents lists available at ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat

On the role of RhoA/ROCK signaling in contact guidance of bone-forming cells on anisotropic Ti6Al4V surfaces A. Calzado-Martín a,b, A. Méndez-Vilas b,c, M. Multigner b,d,1, L. Saldaña a,b, J.L. González-Carrasco b,d, M.L. González-Martín b,c, N. Vilaboa a,b,⇑ a

Hospital Universitario La Paz-IdiPAZ, Paseo de la Castellana 261, 28046 Madrid, Spain CIBER de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Spain Departamento de Física Aplicada, Universidad de Extremadura, Avda. de Elvas s/n, 06006 Badajoz, Spain d Centro Nacional de Investigaciones Metalúrgicas, CENIM-CSIC, Avda. Gregorio del Amo 8, 28040 Madrid, Spain b c

a r t i c l e

i n f o

Article history: Received 30 June 2010 Received in revised form 18 November 2010 Accepted 23 November 2010 Available online 27 November 2010 Keywords: Contact guidance Mesenchymal stem cells Osteoblasts Rho GTPase Ti6Al4V

a b s t r a c t Patterned surfaces direct cell spatial dynamics, yielding cells oriented along the surface geometry, in a process known as contact guidance. The Rho family of GTPases controls the assembly of focal adhesions and cytoskeleton dynamics, but its role in modulating bone-cell alignment on patterned surfaces remains unknown. This article describes the interactions of two human cell types involved in osseointegration, specifically mesenchymal stem cells and osteoblasts, with submicron- or nano-scale Ti6Al4V grooved surfaces generated by mechanical abrasion. The surface chemistry of the alloy was not affected by grinding, ensuring that the differences found in cellular responses were exclusively due to changes in topography. Patterned surfaces supported cell growth and stimulated mesenchymal stem cell viability. Anisotropic surfaces promoted cell orientation and elongation along the grates. Both cell types oriented on nanometric surfaces with grooves of 150 nm depth and 2 lm width. The number of aligned cells increased by approximately 30% on submicrometric grooves with sizes of about 1 lm depth and 10 lm width. Cells were treated with drugs that attenuate the activities of the GTPase RhoA and one of its downstream effectors, Rho-associated kinase (ROCK), and contact guidance of treated cells on the grooved surfaces was investigated. The data indicate that the RhoA/ROCK pathway is a key modulator of both mesenchymal stem cell and osteoblast orientation on nanometric surface features. RhoA and its effector participate in the alignment of mesenchymal stem cells on submicrometric grooves, but not of osteoblasts. These findings show that RhoA/ROCK signaling is involved in contact guidance of bonerelated cells on metallic substrates, although to a varying extent depending on the specific cell type and the dimensions of the pattern. Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction The increasing incidence of bone-related diseases has encouraged research and development into a new generation of orthopedic implants that lead to improved osseointegration and long-term implant stability. A promising approach to stimulating the initial interactions of bone cells with the implant is based on the generation of implant surfaces that mimic the characteristics of natural bone. Bone is a highly organized structure that assembles from the nano- to macro-scale to generate a structural network that provides this tissue with its unique mechanical properties. The basic ⇑ Corresponding author at: Hospital Universitario La Paz-IdiPAZ, Paseo de la Castellana 261, 28046 Madrid, Spain. Tel.: +34 912071034; fax: +34 917277524. E-mail address: [email protected] (N. Vilaboa). 1 Present address: Instituto de Magnetismo Aplicado (RENFE-UCM-CSIC), PO Box 155, Las Rozas, 28230 Madrid, Spain.

building blocks of bone are the mineralized collagen fibrils, which are organized as bundles or aligned arrays. In bone tissue, the alignment of cells and the extracellular collagen matrix determine tissue-specific functions [1]. Many methods have been developed to create regular microscale and nanoscale features on the surfaces of biomaterials, but most available studies in the literature are focused on cell substrates made of glass, silicon or polymers [2–6]. Regarding metallic materials used in the manufacture of orthopedic devices, the treatments usually employed to generate topographies with anisotropic features, such as dry etching or photolithography, also result in chemical heterogeneities, which could influence the observed cellular responses [7,8]. However, abrasion techniques using sandpaper have been successfully employed to generate grooved topographies on titanium and its alloys, without any apparent effect on its chemical composition [9,10]. In addition, the process of abrasion is fast and cheap, factors that make it favorable for the orthopedic industry.

1742-7061/$ - see front matter Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2010.11.035

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901

Substrates with anisotropic topographical features induce cells to align and migrate along the direction of the anisotropy, a phenomenon termed ‘‘contact guidance’’ [11]. This process varies among cell types [2,12,13] and depends on the geometry of the substrate [6,12,14,15]. Various cell types, such as fibroblasts, osteoblasts, endothelial and epithelial cells, display contact guidance when cultured on grooved surfaces with lateral dimensions in the nanometer and micrometer ranges [5,7,14,16]. While cell alignment and elongation on biomaterials have been widely observed and documented, the molecular mechanisms governing contact guidance on osteoblastic cells remain poorly understood. Focal adhesion formation and cytoskeleton rearrangements are known to play an important role in these processes [2,17]. The Rho family of small GTPases, including RhoA, Rac1 and Cdc42, are cytosolic proteins that regulate signal transduction pathways by linking plasma membrane receptors to the assembly of focal adhesions and associated structures [18]. They act as molecular switches that are responsive to a range of extracellular signals able to influence cell migration and the spatial dynamics of the cytoskeleton [19]. The mechanism by which the cells detect substratum features and translate them into changes in cell shape is not well understood. Recent evidence indicates that the Rho family of GTPases regulates corneal epithelial cell alignment on anisotropic surfaces [14,20]. However, the role of these proteins in the modulation of contact guidance of bone cells on metallic surfaces has not been explored to date. This report provides information regarding the interactions of human bone-forming cells, specifically mesenchymal stem cells from bone marrow and primary osteoblasts, with a patterned Ti6Al4V alloy. This alloy is the metallic biomaterial most widely used for the fabrication of orthopedic implants due to the advantageous combination of mechanical strength, excellent corrosion resistance, and good biocompatibility. The involvement of the RhoA/ROCK pathway in controlling cell guidance was investigated using two surfaces with parallel grooves of different depths and widths generated by mechanical abrasion. 2. Material and methods 2.1. Fabrication of the samples Discs of 2 mm thickness were removed by electrospark erosion from a hot rolled and annealed (700 °C h–1) bar 21 mm in diameter supplied by Surgival SL (Valencia, Spain). Chemical composition in wt.%, as indicated in the analysis certificate is: 4.2 V, 6.1 Al, 0.01 C, 0.12 O, 0.006 N, bal. Ti. Samples were first abraded by grinding their surface with silicon carbide papers of decreasing grain size to remove the outermost part of the discs, which was modified during cutting, and were finally polished with diamond paste of 1 lm to get a mirror-like finish. At this stage, samples were washed in a jet of warm water before sonication in alcohol. To generate grooves, polished samples were subjected to a process of mechanical abrasion using 320 (coarse) and 2400 (fine) grit silicon carbide paper to obtain two different topographies. Finally, the specimens were cleaned following the procedure described earlier. The 2400 and 320 ground samples will be referred to as G1 and G2, respectively. Polished samples were used as a control, and will be referred to as PL. All experiments were carried out on samples with an area of 1 cm2. Before cell culture experiments, specimens were washed in distilled water and sterilized under ultraviolet light. 2.2. Material characterization Microstructural analysis of the samples was carried out before and after surface modification by scanning electron microscopy

1891

(SEM) using a JEOL JSM6500F (Peabody, MA, USA) equipped with a field emission gun (FEG) and coupled with an energy dispersive X-ray (EDX) system for chemical analysis (Software Rontec EDR288, Berlin, Germany). Images were obtained both by secondary electrons and backscattered electrons in topographic mode. Surface topography analysis was carried out using an Autoprobe CP commercial atomic force microscope (AFM) (Veeco Instruments, CA, USA). Both the cantilever and the tip (Supersharp tips, Mikromasch, Estonia) are made of silicon and coated by a continuous film of Cr (first layer) and Au (second layer), each of 20 nm width. The cantilever force constant is 0.35 N m–1 and the resonance frequency is 28 kHz. The same tip was used throughout the entire study, and its apex has a radius of curvature <50 nm, having a full cone tip angle <30°. Images were taken in dynamic mode at a frequency slightly higher than the resonance frequency of the cantilever. Surface roughness was determined from the computation of two amplitude roughness parameters: the average roughness (Ra) and the root mean square roughness (Rrms), defined elsewhere [21]. Additionally, line profiles of the topography were recorded. From them, the average values of the depth and the width of the grooves were obtained by measuring the lateral and vertical dimensions of the largest grooves in 12 different images per sample. To obtain information on frequency features that contribute to the surface topography, we performed two-dimensional (2-D) Fourier transforms on AFM images to obtain the 2-D power spectra using a fast Fourier transform (FFT) routine. The low-frequency components of the 2-D Fourier transforms were filtered to maintain only the high-frequency components, thus showing the finest details of the topography. Once done, the high-frequency components were retrieved using an inverse FFT routine. The topography of the low-frequency components, corresponding to the large undulations, was obtained by subtracting the topography of the high-frequency components from the original raw topography. Surface directionality was also quantified by AFM topography maps. Texture direction (Std), defined as the angle of the dominating texture in the image, was used to indicate the direction of the predominant lay of the surface and was calculated from the 2-D Fourier spectrum. For images consisting of parallel ridges, the texture direction is parallel to the direction of the ridges, so when the ridges are perpendicular to the X-scan, Std = 0. The texture direction index parameter (Stdi), defined as the average amplitude sum divided by the amplitude sum of the dominating direction, indicates the relative significance of the dominating direction. All quantifications made from the AFM images were performed using the Gwyddion (David Necˇas and Petr Klapetek, Czech Republic) and SPIP (Image Metrology, Lyngby, Denmark) AFM analysis software packages. 2.3. Cell culture and treatments Purified human mesenchymal stem cells derived from bone marrow (hMSCs) (CD105+, CD29+, CD44+, CD14–, CD34–, CD45–) were purchased from Cambrex Bio Science (Verviers, Belgium) and expanded in a defined medium (Cambrex Bio Science) consisting of MSC basal medium and the SingleQuotsÒ growth supplements, containing fetal bovine serum (FBS), L-glutamine and penicillin/streptomycin. Human osteoblast cells (hOBs) were isolated from trabecular bone explants aseptically collected from patients (aged 70–80 years old) undergoing total knee arthroplasty, as previously described [22]. Bone fragments were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Cambrex Bio Science) containing 15% (v/v) heat inactivated FBS, 500 UI ml–1 of penicillin and 0.1 mg ml–1 streptomycin. Each bone sample was processed in a separated primary culture and experiments were performed using independent cultures obtained from different patients. Confluent cultures were subcultured from initial isolates for

1892

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901

subsequent experiments. Patients enrolled in this research signed an informed consent form and all procedures using human tissue designated ‘‘surgical waste’’ were approved by the Human Research Committee of University Hospital La Paz (Date of Approval: 05-16-2009). Cells were maintained at 37 °C under 5% CO2 in a humidified incubator.

RhoA and ROCK activities were inhibited using 1 lg ml–1 cell permeable C3 transferase (Cytoskeleton Inc., Denver, USA) and 10 lM hydroxyfasudil (HF) (Calbiochem-Merck Biosciences, CA, USA), respectively. Both inhibitors were dissolved in distilled water and stored at 20 °C. HF or C3 transferase was directly applied to the cell culture media to reach their final concentrations. Parallel

Fig. 1. Topographical features of grooved Ti6Al4V surfaces. (A) SEM images obtained by secondary electrons of PL (without etching), and of G1 and G2 samples obtained by backscattered electrons in topographic mode. Bar = 3 lm. (B) AFM topographical profile of the studied surfaces. (C) AFM original topography of G1 surfaces (D) AFM spectral decomposition of G2 surfaces into two different levels. The top image shows the original topography. The intermediate image shows the low-frequency component containing the largest topographical features, and the bottom image is the high-frequency component containing the finest features of the surface. (E) 2-D Fourier spectra obtained from the original topographical AFM images showing the directionality of the samples.

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901 Table 1 Roughness and directionality parameters for the different samples. PL Ra (nm) Rrms (nm)

3.3 ± 0.7 5 ± 1.0

Std Stdi

Meaningless 0.8 ± 0.1

G1

G2

58 ± 7 83 ± 14

200 ± 65 248 ± 68

0° 0.47 ± 0.01

0° 0.26 ± 0.06

Each value represents the mean ± SD of five independent measurements.

Table 2 Average values of depth and width of the grooves.

Depth (lm) Width (lm)

G1

G2

0.15 ± 0.09 2 ± 0.90

0.95 ± 0.30 9 ± 4.00

Each value represents the mean ± SD of 12 independent measurements.

cultures of untreated cells were subjected to the same manipulations as treated cells and used as controls. 2.4. Cell viability Cells were seeded on PL, G1 and G2 surfaces in 24-well plates (7  103 cells per well) and cultured up to 7 days. Cell viability was assessed at days 1, 4 and 7 using the alamarBlue assay (Biosource, Nivelles, Belgium), which incorporates a redox indicator that fluoresces in response to cellular metabolic reduction. After washing with 10 mM phosphate buffered, 140 mM saline (PBS, pH 7.4), cells were incubated in DMEM containing 10% alamarBlue dye for 4 h. After excitation at 530 nm, the fluorescence emitted at 590 nm was quantified using a microplate reader Synergy 4 (BioTek Instruments, Winooski, VT, USA). 2.5. Actin cytoskeleton reorganization, cell orientation and elongation Cells were seeded on polystyrene (PS) or on the metallic surfaces in 24-well plates (1  104 cells per well) and incubated in the presence or absence of C3 transferase or HF using two different experimental conditions. In experimental Condition 1, the drugs were added at the time of cell seeding. In experimental Condition 2, the treatments were applied 24 h after seeding the cells on the samples. In both experimental conditions, cells were incubated for 24 h in the presence or absence of the drugs. The cell monolayer was washed with PBS and attached cells were fixed with 4%

1893

paraformaldehyde in PBS and permeabilized with 0.1% Triton X100 in PBS. Filamentous actin was stained with PBS containing 4  10–7 M phalloidine-TRITC (Sigma, Madrid, Spain) and nuclear DNA staining was performed with PBS containing 3  10–6 M 4,6diamidino-2-phenylindole (DAPI, Sigma). Cells were observed using a fluorescence microscope (Leica AF6000, Wetzlar, Germany) and high resolution fluorescence images were captured. Cell orientation and elongation were determined by measuring a total of 180 well-spread cells per sample, randomly selected from nine representative images obtained from three independent experiments with similar results. Cells were manually outlined using ImageJ v1.34 image analysis software (http://rsbweb.nih.gov/ij), which fits each cell to a fittest ellipse. The lengths of the major and the minor axis of the fittest ellipse represent cell length and width, respectively. The alignment of an individual cell along the grooves was calculated by measuring the angle formed by the major axis of the fittest ellipse and the horizontal. Elongation was quantified as the result of dividing the major axis between the minor axis of the fittest ellipse. 2.6. Scanning electron microscopy Cells were seeded on the metallic surfaces in 24-well plates (1  104 cells per well) and cultured for 24 h. Attached cells were washed with PBS and fixed with 2.5% glutaraldehyde for 1 h at room temperature. Samples were subsequently dehydrated in a graded ethanol series, and critical point dried with CO2 (Quorum Technologies CPD7501, UK). Once dried, the samples were gold sputter-coated on a rotating-tilting stage (Sputter Coater SC510, Bio-Rad, Spain) before examination by scanning electron microscopy (FEI Inspect SEM, Hillsboro, OR, USA). 2.7. Assessment of RhoA and ROCK activities Cells were seeded in 60 mm PS culture plates (8  105 cells per well) in the presence or absence of C3 transferase or HF, and cultured for 2 and 24 h. RhoA activation levels were determined by quantification of GTP-bound form of RhoA using a RhoA G-LISA™ activation kit (Cytoskeleton Inc.). Cell layers were washed exhaustively with PBS, and protein lysates were extracted using the cell lysis buffer supplied with the kit. Cell lysates were clarified by centrifugation and RhoA activity was quantified in aliquots of extracts containing 25 lg of total protein, following the manufacturer’s instructions. For ROCK activity measurements, cell layers were washed exhaustively with PBS and extracted with a cell lysis buffer

Fig. 2. Cell viability on grooved Ti6Al4V surfaces. Cells were cultured for 1, 4 and 7 days on PL ( ), G1 ( ) and G2 ( ) surfaces. The results are expressed as the percentage of the fluorescence measured on PL samples at day 1, which was normalized to 100. Each value represents the mean ± SD of four independent experiments. p < 0.05 compared to PL samples.

1894

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901

(10 mM Tris–HCl, pH 7.5, 150 mM NaCl, 1 mM 2-glycerophosphate, 1% Triton X-100, 1 mM EDTA, 1 mM EGTA, 1 mM Na3VO4) supplemented with a complete protease inhibitors cocktail (Roche, Barcelona, Spain). Cell lysates were clarified by centrifugation and active ROCK levels were quantified in aliquots of the extracts containing 50 lg of total protein using a Rho Kinase Assay (Cell Biolabs Inc., San Diego, CA, USA), according to the manufacturer’s instructions. 2.8. Statistical analysis The SPSS (version 9.0; SPSS Inc., Chicago, IL, USA) package was used for statistical analyses. Data are given as mean ± standard deviation (SD). Quantitative data were tested using two-sided Kruskal–Wallis and Mann–Whitney U rank-sum tests, and p < 0.05 was considered statistically significant. Post hoc comparisons were analyzed by the Mann–Whitney U test, adjusting the

p-value with the Bonferroni correction for multiple comparisons, and the level of significance was set to p < 0.015. 3. Results 3.1. Material characterization SEM examination of non-etched PL samples using secondary electron images (SEIs) revealed a biphasic structure consisting of discontinuous bright zones (b-phase) into a dark matrix (a-phase) (Fig. 1A). After grinding, backscattered electron images (BEIs) revealed that the most relevant microstructural feature corresponds to the formation of grooves that are aligned along the grinding direction. It should be noticed that a and b phases could not be revealed in G1 and G2 samples since, due to topography, the signal masks the compositional contrast. Surface contamination with abrasive particles was ruled out after a detailed examination.

Fig. 3. Cell orientation on grooved Ti6Al4V surfaces. (A) Actin (red) and nuclei (blue) staining of cells cultured for 24 h on PL, G1 and G2 surfaces. Representative images of three independent experiments yield similar results. Bar = 250 lm. (B) Relative frequency histograms showing the distribution of cell orientation. Cells were cultured for 24 h on PL ( ), G1 ( ) and G2 ( ) samples. The groove direction is 90°.

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901

Surface roughness and directionality parameters of the studied surfaces are summarized in Table 1. G1 surfaces displayed values of Ra and Rrms below 100 nm, falling in the nanometer range, while G2 surfaces fit in the submicrometric range, since their roughness values are below 500 nm. Topographical line profiles were extracted from the raw AFM images for measuring the depth and width of the grooves (Fig. 1B). The average values are summarized in Table 2. As expected, G2 surfaces exhibited wider and deeper grooves than G1 samples. The analysis of the frequency features on G1 surfaces was not efficient, since this method is only effective with symmetrical systems, so only the original topography has been shown (Fig. 1C). In the case of G2 surfaces, two different levels of topography can be clearly differentiated: one forming the large grooves and another one forming the small grooves, likely caused by irregularities in the surface of the abrasive sandpapers (Fig. 1D). Both the small and large grooves show a high degree of directionality. It should be noted that the small irregularities detected on G2 surfaces are similar to the original topography of G1 samples. Fig. 1E shows the 2-D Fourier spectra of the topographical images. For G2 surfaces, a strong directionality along the X-axis was detected, corresponding to accurate directionality on the Yaxis in the real space. For G1 surfaces, a precise disc towards 0° is still observed, indicating a dominant direction of 90° in the real

1895

image. No asymmetry was observed for the PL surface, indicating no directionality. The two grooved surfaces displayed Std values of 0°, confirming that groove directionality was successfully achieved (Table 1). Surface directionality defined by Stdi was higher for G2 than for G1 samples, probably due to the clumpy structures observed on G1 samples, resulting in a more discontinuous directionality compared to G2 surfaces (Table 1). 3.2. Cell viability The number of viable cells increased over time on all tested substrates regardless of the cell type (Fig. 2). While no differences were detected in hOBs viability on anisotropic surfaces, cell viability increased in hMSCs cultured on grooved samples at day 7, compared to PL samples. 3.3. Cell orientation and elongation on grooved surfaces Once it was determined that G1 and G2 surfaces were biocompatible, we analyzed cell guidance on these samples. Fluorescence images revealed that actin cytoskeleton was organized in well-defined stress fibers distributed throughout the cell body on the three studied surfaces (Fig. 3A). On patterned surfaces, cells exhibited actin stress fibers aligned parallel to the grinding direction. Both

Fig. 4. Cell elongation on grooved Ti6Al4V surfaces. (A) SEM images of cells cultured for 24 h on PL, G1 and G2 surfaces. Bar = 20 lm. (B) Relative frequency histograms showing the distribution of cell elongation. Cells were cultured for 24 h on PL ( ), G1 ( ) and G2 ( ) samples. Insets of the statistical analysis of the data are on the right side of the plots. p < 0.015.

1896

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901

Fig. 5. Effect of C3 transferase on RhoA activity and cytoskeleton reorganization. (A) RhoA activity in cells seeded on PS in the presence ( ) or absence ( ) of C3 transferase, and cultured for 2 and 24 h. The results are expressed as the percentage of activity measured on untreated hMSCs cultured for 2 h, which was normalized to 100. Each value represents the mean ± SD of three independent experiments. ⁄p < 0.05 compared to untreated cells, #p < 0.05 compared to hMSCs. Actin (red) and nuclei (blue) staining of cells seeded on PS (B), PL (C), G1 (D) or G2 (E) surfaces in the absence or presence of C3 transferase, and cultured for 24 h. Representative images of three independent experiments yielding similar results. Bar = 100 lm.

Fig. 6. Involvement of RhoA activity in cell orientation on grooved Ti6Al4V surfaces. Cells were cultured on grooved samples untreated (black) or treated with C3 transferase (grey) under two different experimental conditions: Condition 1: C3 transferase was added at the time of seeding; Condition 2: C3 transferase was added 24 h after seeding the cells on the samples. In both experimental conditions, cells were cultured for 24 h in the presence or the absence of C3 transferase. Cells untreated and cultured on PL samples (dashed line) are shown as a control. Relative frequency histograms show the distribution of cell orientation. The groove direction is 90°. p values show the differences between the orientation distributions of untreated and treated cells cultured on the grooved surfaces.

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901

hOBs and hMSCs were randomly oriented after culturing for 24 h on PL surfaces. Quantification of cell orientation distribution revealed that both cell types aligned in the direction of the grooves, and cell alignment increased with increasing size of the grooves (p < 0.015) (Fig. 3B). Cells cultured on PL samples were well spread and showed typical polygonal morphologies. When cultured on grooved surfaces, both cell types adopted a stretched shape that was more noticeable on G2 surfaces (Fig. 4A). The analysis of the distribution of cell orientation showed that hOBs elongation increased on both grooved surfaces, whereas hMSCs only elongated when cultured on G2 samples (p < 0.015) (Fig. 4B). 3.4. Involvement of RhoA and ROCK in cell orientation and elongation on grooved surfaces We first investigated the activation status of RhoA and ROCK in hMSCs and hOBs cultured on PS. After seeding the cells, RhoA activity of both cell types decreased with culture time from 2 to 24 h, with activity levels higher in hMSCs than in hOBs (Fig. 5A). When the cells were plated on PS in the presence of C3 transferase, which ADP-ribosylates RhoA [23], active RhoA levels decreased in both cell types, with a more pronounced effect after 24 h of treatment (p < 0.05) (Fig. 5A). RhoA inhibition resulted in cell morphology changes characterized by disassembly of actin filaments and higher cell surface area (Fig. 5A). These changes were also observed when cells were seeded in the presence of C3 transferase on nongrooved and grooved surfaces (Fig. 5B–E). The plots in Fig. 6 reveal

1897

that C3 transferase completely suppressed cell orientation on G1 samples in both cell types (p < 0.015). In terms of orientation, the cells were indistinguishable from cells cultured on polished surfaces both when the inhibitor was added at the time of cell seeding and 24 h after. On G2 samples, RhoA inhibition decreased hMSCs (p < 0.015) but not hOBs (p > 0.015) orientation in both experimental conditions (Fig. 6). There were not significant differences between the orientation of hMSCs treated with C3 transferase at the time of seeding or 24 h after. Similar to that observed for RhoA, ROCK activity decreased in cells seeded on PS along incubation time (Fig. 7A). Both cell types expressed similar levels of active ROCK 2 h after seeding the cells, although the levels were higher in hMSCs than in hOBs after culturing for 24 h. Both hMSCs and hOBs experienced a 30–40% decrease in ROCK activity levels after treatment for 2 h with HF, a specific ROCK inhibitor. Although the inhibitory effect of HF on ROCK activity was transient (Fig. 7A), likely due to the short halflife of HF [24], the treatment of cells at the time that they were plated on all tested surfaces led to the disorganization of the actin network and increased the cell area (Fig. 7B–E). HF suppressed cell orientation on G1 surfaces in hMSCs and hOBs, both when the inhibitor was added at the time of cell seeding and 24 h after (p < 0.015) (Fig. 8). On G2 surfaces, ROCK inhibition had no effect on hOBs (p > 0.015), and only affected the initial orientation of freshly seeded hMSCs (p < 0.015) (Fig. 8). Cell elongation was analyzed in cells treated with C3 transferase or HF at the time of cell seeding on the grooved surfaces. Upon RhoA attenuation, cell elongation was significantly decreased on

Fig. 7. Effect of HF on ROCK activity and cytoskeleton reorganization. (A) ROCK activity in cells seeded on PS in the presence ( ) or absence ( ) of HF, and cultured for 2 and 24 h. The results are expressed as the percentage of the activity measured on untreated hMSCs cultured for 2 h, which was normalized to 100. Each value represents the mean ± SD of three independent experiments. ⁄p < 0.05 compared to untreated cells, #p < 0.05 compared to hMSCs. Actin (red) and nuclei (blue) staining of cells seeded on PS (B), PL (C), G1 (D) or G2 (E) surfaces in the absence or presence of HF, and cultured for 24 h. Representative images of three independent experiments yielding similar results. Bar = 100 lm.

1898

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901

Fig. 8. Involvement of ROCK activity in cell orientation and elongation on grooved Ti6Al4V surfaces. Cells were cultured on grooved samples untreated (black) or treated with HF (grey) under two different experimental conditions: Condition 1: HF was added at the time of seeding; Condition 2: HF was added 24 h after seeding the cells on the samples. In both experimental conditions, cells were cultured for 24 h in the presence or the absence of HF. Cells untreated cultured on PL samples (dashed line) are shown as control. Relative frequency histograms show the distribution of cell orientation. The groove direction is 90°. p values show the differences between the orientation distributions of untreated and treated cells cultured on the grooved surfaces.

both grated surfaces (p < 0.015) (Fig. 9A) while it was unaffected by HF (Fig. 9B).

4. Discussion The main aim of this study was to gain new insights into the interactions occurring between bone-forming cells and patterned Ti6Al4V surfaces. Nano and submicron grooves were generated on the alloy by mechanical abrasion, a very simple, cheap, and effective technique that does not alter the chemistry of the surface. Hence, the differences found in cellular responses must be exclusively due to the changes in topographical features. Osteoblasts arise from mesenchymal precursors, but the behavior of these two cell types on biomaterial surfaces can be different [5,9,25]. Previous studies showed that mesenchymal stem cell proliferation increased with roughness values ranging from 320 to 874 nm when cultured on Ti6Al4V grooved surfaces generated by mechanical abrasion [25], while osteoblast growth on similar surfaces was unaffected by a wider range of grooves featuring from 70 nm to 1.2 lm [9]. We found that compared to flat substrates, nano- and submicron-grooved Ti6Al4V surfaces enhanced hMSCs viability while not affecting hOBs, suggesting that precursors in an undifferentiated state are more sensitive to surface patterning than mature osteoblasts. These data support the idea that surface patterning might improve implant colonization by precursor cells, thus facilitating its osseointegration. Both nano and submicrometric Ti6Al4V surfaces promote hMSCs and hOBs orientation in the direction of the grooves. In terms of degree of cell orientation, both cell types behave similarly; they were able to orient on G1 surfaces with 150 nm depth and 2 lm width grooves, and the number of aligned cells increased by approximately 30% when the size of the grooves reached 1 lm depth and 10 lm width. This reliance of cell alignment on the

dimensions of the grooves has been widely described in other cell types [5,6,12]. For instance, calf osteoblast alignment on grooved surfaces was enhanced by almost 20% with increasing depth of grooves from 50 to 150 nm [5], and rat fibroblasts also showed a similar orientation trend with increasing pitch width [12]. An interesting finding of the present study was that the small irregularities found in G2 surfaces were in a similar range as the topographical features of G1 samples. Although cell orientation may be influenced by the ‘‘nanometric’’ features detected in both surfaces, the differences found between G1 and G2 samples must be due to the large undulations present in G2 samples. However, the smaller topographic components may account for the enhanced hMSCs viability detected on both types of samples. Various factors influence contact guidance on patterned substrates, such as the specific distribution of extracellular matrix proteins on the substrate, but as yet very little is known about the mechanisms controlling this phenomenon [17]. Metallic grooved surfaces give rise to anisotropic distributions of surface charges [26] that might modulate the formation of focal contacts and direct cell orientation [13,27]. A number of studies show that cells align on patterned surfaces as a consequence of the accumulation of focal adhesions on the ridges, which results in the reorganization of actin stress fibers parallel to the grooves [2,28,29]. SEM images of hMSCs and hOB on grooved Ti6Al4V showed that cellular extensions were mainly attached on the top of the grooves, and actin staining revealed stress fibers parallel to the surface features, suggesting that focal adhesions are likely occurring on the ridges. Indeed, we previously observed that the active form of focal adhesion kinase (FAK), a protein that accumulates in the focal contacts, is mainly localized in the cellular extensions of osteoblastic cells [22,30]. Bearing in mind that cytoskeleton remodeling is critical to contact guidance, we hypothesized that the GTPase RhoA might be involved in regulating bone-forming cells alignment on the studied surfaces, since this protein is involved in the assembly of

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901

1899

Fig. 9. Involvement of RhoA and ROCK activities in cell elongation on grooved Ti6Al4V surfaces. Relative frequency histograms show the distribution of cell elongation. Cells were seeded in the absence (black) or in the presence of C3 transferase (A) or HF (B) (grey) and cultured for 24 h on grooved samples.

focal adhesions and actin polymerization [31,32]. Rajnicek et al. reported that alignment of corneal epithelial cells on grooved surfaces is mediated by RhoA [14,20], but to our knowledge there is no available data addressing whether this small GTPase participates in the orientation of bone-forming cells. ROCK is one of the main downstream effectors of RhoA [32], which prompted us to assess its role in contact guidance. The first approach was to compare RhoA and ROCK activities of both cell types when cultured in standard tissue plastic. RhoA activity was higher in hMSCs than in hOB, which may be related to their different maturation states. In this regard, RhoA inactivation in osteoblastic cells has been shown to increase bone sialoprotein expression, suggesting that this GTPase could be involved in the acquisition of the osteoblastic phenotype [33]. It is interesting to note that when both cell types were incubated in standard tissue culture plastic, RhoA and ROCK activities decreased with time in culture. These observations are consistent with earlier reports demonstrating an increase in Rho activity at the time new adhesions are being formed, while in stably adhered

cells that display less dynamic adhesions, Rho signaling is downregulated to prevent an excessive formation of focal adhesions [34,35]. The involvement of the RhoA/ROCK pathway in cell orientation on the studied surfaces was investigated using C3 transferase and HF, which attenuate RhoA and ROCK activities, respectively. Two different experimental conditions were set up to discern whether the RhoA/ROCK pathway is involved in the initial orientation of freshly plated cells on grooved surfaces, and whether it participates in the maintenance of the alignment once the cells are oriented on the substrates. Previous studies reported that RhoA inhibition led to changes in mesenchymal stem cells and fibroblast morphologies [36,37]. Accordingly, under the two experimental conditions we employed, C3 transferase and HF led to disorganization of the actin network and increased cell spreading in non-grooved and grooved surfaces. We found that RhoA modulates hMSCs and hOBs orientation on nanogrooved titanium alloy surfaces, and the alignment process on these substrates is conducted through its effector ROCK. On these nanometric

1900

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901

features, the RhoA/ROCK pathway controls both the initial cell orientation on the grooves as well as the aligned status of attached cells. Interestingly, we observed that neither RhoA nor ROCK inhibition affected hOBs orientation on submicrometric grooves. However, treatment with C3 transferase decreased the orientation of both freshly plated cells and of adhered and oriented hMSCs. The findings that RhoA inhibition could not completely suppress hMSCs orientation on submicrometric surfaces nor affect osteoblast orientation indicate that other pathways must regulate contact guidance in these cells. In fact, the GTPase Rac1 and the signaling adaptor protein RACK1 participate in the alignment of other cell types [38–40]. ROCK inhibition only affected hMSCs orientation on submicrometric surfaces when the inhibitor was added at the time of cell seeding, suggesting that the effects RhoA exerts through ROCK in these cells occur in the earliest stages of cell attachment. Finally, we observed that during the initial adhesion of osteoblasts and osteoprogenitor cells to nano- and submicrongrooved surfaces, cell elongation was controlled by RhoA, but not by ROCK. RhoA has been shown to induce microtubule-dependent shape changes through its target mDia in different cell types such as HeLa or 3T3 [41,42]. In this regard, cell elongation on grooved Ti6Al4V samples might be controlled not only by the actin network but also by microtubule dynamics. Titanium-based implants having random microrough surfaces foster stable osseointegration, as demonstrated by numerous studies [43,44]. The clinical success of these topographies is explained, at least in part, by the effects of microroughness on specific boneforming cell functions, which lead to the fast mineralization of the extracellular matrix at the bone/implant interface, improving fixation and stability through an increase in interlocking surface area. Recent results obtained using silicon wafers and polystyrene indicate that, in addition to inducing osteoblastic alignment, nanogrooved patterning promote in vitro calcification of the extracellular matrix along the grates [45,46]. However, information regarding to the in vivo performance of metallic implants having nanopatterned surfaces is not available yet. Although an ‘‘ideal’’ surface topography for dental or orthopaedic devices is not defined in the marketplace or in the literature, it can be envisioned that a future generation of implantable biomaterials will combine macro-, micro- and nano-topographical features, their suitability for clinical settings being dictated by application-specific requirements. In addition to enhanced osteoprogenitor proliferation, potential benefits of nanometric and submicrometric grooved Ti6Al4V surfaces include organized orientation of bone-forming cells. Further investigations will determine whether anisotropic roughening of the alloy by mechanical abrasion could promote in vivo formation of a stronger interface between the implant surface and bone tissue, which would benefit the long-term performance of the devices. Finally, the results of this study contribute to the understanding of the complex molecular mechanisms involved in contact guidance of bone cells on grooved metallic substrates. Although we detected that RhoA/ROCK signaling participates in the control of contact guidance of bone cells on these surfaces, additional regulatory pathways underlie this phenomenon. Acknowledgments This work was supported by grants MAT2009-14695-C04-0102-04 from the Ministerio de Ciencia e Innovación (Spain), grants from Fundación Mutua Madrileña (Spain) and grant PRI08A124 from Consejería de Economía, Comercio e Innovación de la Junta de Extremadura (Spain). A. Calzado-Martín is supported by Fundación para la Investigación Biomédica del Hospital Universitario La Paz (Madrid, Spain). N. Vilaboa is supported by Program I3SNS from Fondo de Investigaciones Sanitarias (Spain). The authors are greatly indebted to Alba Boré (Unidad de Investigación, Hospital

Universitario La Paz and CIBER-BBN), to the SEM laboratory (National Museum of Science and Technology, MNCT, Madrid) for excellent technical support, to the clinical staff of the Orthopaedic Department for providing us with bone samples, and to the Section of Biostatistics (Hospital Universitario La Paz) for help with statistical analysis.

Appendix A. Figures with essential colour discrimination Certain figures in this article, particularly Figures 1, 3, 5, and 7, are difficult to interpret in black and white. The full colour images can be found in the on-line version, at doi:10.1016/j.actbio.2010. 11.035. References [1] Weiner S, Traub W, Wagner HD. Lamellar bone: structure-function relations. J Struct Biol 1999;126:241–55. [2] Teixeira AI, Abrams GA, Bertics PJ, Murphy CJ, Nealey PF. Epithelial contact guidance on well-defined micro- and nanostructured substrates. J Cell Sci 2003;116:1881–92. [3] Bettinger CJ, Orrick B, Misra A, Langer R, Borenstein JT. Microfabrication of poly (glycerol-sebacate) for contact guidance applications. Biomaterials 2006;27:2558–65. [4] Grew JC, Ricci JL, Alexander H. Connective-tissue responses to defined biomaterial surfaces. II. Behavior of rat and mouse fibroblasts cultured on microgrooved substrates. J Biomed Mater Res A 2008;85:326–35. [5] Lenhert S, Meier MB, Meyer U, Chi L, Wiesmann HP. Osteoblast alignment, elongation and migration on grooved polystyrene surfaces patterned by Langmuir–Blodgett lithography. Biomaterials 2005;26:563–70. [6] Crouch AS, Miller D, Luebke KJ, Hu W. Correlation of anisotropic cell behaviors with topographic aspect ratio. Biomaterials 2009;30:1560–7. [7] Lu J, Rao MP, MacDonald NC, Khang D, Webster TJ. Improved endothelial cell adhesion and proliferation on patterned titanium surfaces with rationally designed, micrometer to nanometer features. Acta Biomat 2008;4:192–201. [8] Zhao JM, Tsuru K, Hayakawa S, Osaka A. Modification of Ti implant surface for cell proliferation and cell alignment. J Biomed Mat Res B: App Biomat 2007;84:988–93. [9] Anselme K, Bigerelle M, Noël B, Iost A, Hardouin P. Effect of grooved titanium substratum on human osteoblastic cell growth. J Biomed Mater Res 2002;60:529–40. [10] Ball M, Grant DM, Lo WJ, Scotchford CA. The effect of different surface morphology and roughness on osteoblast-like cells. J Biomed Mater Res A 2008;86:637–47. [11] Weiss P, Garber B. Shape and movement of mesenchyme cells as functions of the physical structure of the medium: contributions to a quantitative morphology. Proc Nat Acad Sci USA 1952;38:264–80. [12] Loesberg WA, te Riet J, van Delft FC, Schön P, Figdor CG, Speller S, et al. The threshold at which substrate nanogroove dimensions may influence fibroblast alignment and adhesion. Biomaterials 2007;28:3944–51. [13] Monsees TK, Barth K, Tippelt S, Heidel K, Gorbunov A, Pompe W, et al. Effects of different titanium alloys and nanosize surface patterning on adhesion, differentiation, and orientation of osteoblast-like cells. Cells Tissues Organs 2005;180:81–95. [14] Rajnicek AM, Foubister LE, McCaig CD. Alignment of corneal and lens epithelial cells by co-operative effects of substratum topography and DC electric fields. Biomaterials 2008;29:2082–95. [15] Karuri NW, Liliensiek S, Teixeira AI, Abrams G, Campbell S, Nealey PF, et al. Biological length scale topography enhances cell-substratum adhesion of human corneal epithelial cells. J Cell Sci 2004;117:3153–64. [16] Ricci JL, Grew JC, Alexander H. Connective-tissue responses to defined biomaterial surfaces. I. Growth of rat fibroblast and bone marrow cell colonies on microgrooved substrates. J Biomed Mater Res A 2008;85:313–25. [17] Zhou F, Yuan L, Huang H, et al. Phenomenon of ‘‘contact guidance’’ on the surface with nano-micro-groove-like pattern and cell physiological effects. Chinese Sci Bull 2009;54:3200–5. [18] Etienne-Manneville S, Hall A. Rho GTPases in cell biology. Nature 2002;420:629–35. [19] Ridley AJ, Schwartz MA, Burridge K, Firtel RA, Ginsberg MH, Borisy G, et al. Cell migration: integrating signals from front to back. Science 2003;302:1704–9. [20] Rajnicek AM, Foubister LE, McCaig CD. Prioritising guidance cues: directional migration induced by substratum contours and electrical gradients is controlled by a rho/cdc42 switch. Dev Biol 2007;312:448–60. [21] Me´ndez-Vilas A, Donoso MG, Gonza´lez-Carrasco JL, Gonza´lez-Marti´n ML. Looking at the micro-topography of polished and blasted Ti-based biomaterials using atomic force microscopy and contact angle goniometry. Colloid Surface B 2006;52:157–66. [22] Saldaña L, Vilaboa N, Vallés G, González-Cabrero J, Munuera L. Osteoblast response to thermally oxidized Ti6Al4V alloy. J Biomed Mater Res A 2005;73:97–107.

A. Calzado-Martín et al. / Acta Biomaterialia 7 (2011) 1890–1901 [23] Vogelsgesang M, Pautsch A, Aktories K. C3 exoenzymes, novel insights into structure and action of Rho-ADP-ribosylating toxins. NS Arch Pharmacol 2007;374:347–60. [24] Liao JK, Seto M, Noma K. Rho kinase (ROCK) inhibitors. J Cardiovasc Pharmacol 2007;50:17–24. [25] Deligianni DD, Katsala N, Ladas S, Sotiropoulou D, Amedee J, Missirlis YF. Effect of surface roughness of the titanium alloy Ti-6Al-4V on human bone marrow cell response and on protein adsorption. Biomaterials 2001;22:1241–51. [26] Tipler PA. Física. Barcelona, Spain: Ed Reverté; 1978. [27] Könönen M, Hormia M, Kivilahti J, Hautaniemi J, Thesleff I. Effect of surface processing on the attachment, orientation, and proliferation of human gingival fibroblasts on titanium. J Biomed Mater Res 1992;26:1325–41. [28] Biggs MJP, Richards RG, McFarlane S, Wilkinson CDW, Oreffo ROC, Dalby MJ. Adhesion formation of primary human osteoblasts and the functional response of mesenchymal stem cells to 330 nm deep microgrooves. J R Soc Interface 2008;5:1231–42. [29] Fujita S, Ohshima M, Iwata H. Time-lapse observation of cell alignment on nanogrooved patterns. J R Soc Interface 2009;6:S269–77. [30] Saldaña L, Vilaboa N. Effects of micrometric titanium particles on osteoblast attachment and cytoskeleton architecture. Acta Biomater 2010;6:1649–60. [31] Gerecht S, Bettinger CJ, Zhang Z, Borenstein JT, Vunjak-Novakovic G, Langer R. The effect of actin disrupting agents on contact guidance of human embryonic stem cells. Biomaterials 2007;28:4068–77. [32] Jaffe AB, Hall A. Rho GTPases: biochemistry and biology. Annu Rev Cell Dev Biol 2005;21:247–69. [33] Chaplet M, Detry C, Deroanne C, Fisher LW, Castronovo V, Bellahcéne A. Zoledronic acid up-regulates bone sialoprotein expression in osteoblastic cells through Rho GTPase inhibition. Biochem J 2004;15:591–8. [34] Ren XD, Kiosses WB, Schwartz MA. Regulation of the small GTP-binding protein Rho by cell adhesion and the cytoskeleton. EMBO J 1999;18:578–85.

1901

[35] Ren XD, Kiosses WB, Sieg DJ, Otey CA, Schlaepfer DD, Schwartz MA. Focal adhesion kinase suppresses Rho activity to promote focal adhesion turnover. J Cell Sci 2000;113:3673–8. [36] Jaganathan BG, Ruester B, Dressel L, Stein S, Grez M, Seifried E, et al. Rho inhibition induces migration of mesenchymal stromal cells. Stem Cells 2007;25:1966–74. [37] Hall A, Nobes CD. Rho GTPases: molecular switches that control the organization and dynamics of the actin cytoskeleton. Phil Trans R Soc Lond B 2000;355:965–70. [38] Dalby MJ, Hart A, Yarwood SJ. The effect of the RACK1 signalling protein on the regulation of cell adhesion and cell contact guidance on nanometric grooves. Biomaterials 2008;29:282–9. [39] Nobes CD, Hall A. Rho GTPases control polarity, protrusion, and adhesion during cell movement. J Cell Biol 1999;144:1235–44. [40] Allen WE, Jones GE, Pollard JW, Ridley AJ. Rho, Rac and Cdc42 regulate actin organization and cell adhesion in macrophages. J Cell Sci 1997;110:707–20. [41] Ishizaki T et al. Coordination of microtubules and actin cytoskeleton by the Rho effector mDial. Nat Cell Biol 2001;3:8–14. [42] Palazzo AF, Cook TA, Alberts AS, Gundersen GG. MDia mediates Rho-regulated formation and orientation of stable microtubules. Nat Cell Biol 2001;3:723–9. [43] Geetha M, Singh AK, Asokamani R, Gogia AK. Ti based biomaterials, the ultimate choice for orthopaedic implants - A review. Prog Mater Sci 2009;54:397–425. [44] Ehrenfest DMD, Coelho PG, Kang BS, et al. Classification of osseointegrated implant surfaces: materials, chemistry and topography. Trends Biotechnol 2010;28:198–206. [45] Lamers E, Walboomers XF, Domanski M, te Riet J, van Delft FC, Luttge R, et al. The influence of nanoscale grooved substrates on osteoblast behavior and extracellular matrix deposition. Biomaterials 2010;31:3307–16. [46] Zhu B, Lu Q, Yin J, Hu J, Wang Z. Alignment of osteoblast-like cells and cellproduced collagen matrix induced by nanogrooves. Tissue Eng 2005;11:825–34.