- Email: [email protected]
ultrafine-grained structures: Interaction with fibroblasts
Acta Biomaterialia 6 (2010) 3339–3348
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat
Biological significance of nanograined/ultrafine-grained structures: Interaction with fibroblasts R.D.K. Misra a,*, W.W. Thein-Han a, T.C. Pesacreta b, M.C. Somani c, L.P. Karjalainen c a
Biomaterials and Biomedical Engineering Research Laboratory, Center for Structural and Functional Materials, University of Louisiana at Lafayette, P.O. Box 44130, Lafayette, LA 70504, USA b Department of Biology and Microscopy Center, University of Louisiana at Lafayette, P.O. Box 42451, Lafayette, LA 70504, USA c Department of Mechanical Engineering, The University of Oulu, P.O. Box 4200, 90014 Oulu, Finland
a r t i c l e
i n f o
Article history: Received 20 August 2009 Received in revised form 12 November 2009 Accepted 22 January 2010 Available online 28 January 2010 Keywords: Nanostructured materials Phase reversion Stainless steel Cellular response Fibroblasts
a b s t r a c t Given the need to develop high strength/weight ratio bioimplants with enhanced cellular response, we describe here a study focused on the processing–structure–functional property relationship in austenitic stainless steel that was processed using an ingenious phase reversion approach to obtain an nanograined/ ultrafine-grained (NG/UFG) structure. The cellular activity between fibroblast and NG/UFG substrate is compared with the coarse-grained (CG) substrate. A comparative investigation of NG/UFG and CG structures illustrated that cell attachment, proliferation, viability, morphology and spread are favorably modulated and significantly different from the conventional CG structure. These observations were further confirmed by expression levels of vinculin and associated actin cytoskeleton. Immunofluorescence studies demonstrated increased vinculin concentrations associated with actin stress fibers in the outer regions of the cells and cellular extensions on NG/UFG substrate. These observations suggest enhanced cell–substrate interaction and activity. The cellular attachment response on NG/UFG substrate is attributed to grain size and hydrophilicity and is related to more open lattice in the positions of high-angle grain boundaries. Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
1. Introduction In the replacement or substitution of hard tissue implants, metallic materials including austenitic stainless steels and titanium alloys are widely used biomaterials as devices for bone fixation, partial/total joint replacement and spring clips for the repair of large aneurysmal defects because they are inherently corrosion resistant and possess the necessary mechanical strength and biocompatibility [1–3]. In recent years, surface modification processes such as heat treatment have been explored to induce nanostructuring on the surface and consequently promote osseointegration and modulation of cellular activity. While ultrafine structures may provide benefits of enhanced cellular attachment, stimulate metabolic activity and upregulate protein formation [4–9], it is important that we consider both the surface and bulk properties of the material that is in direct contact with the bone for long-term stability and an acceptable success rate of prosthetic rehabilitation. Thus, processing of bulk nanostructured materials assumes particular significance because it is a viable and potentially transformative approach to favorably modulate the cellular response of biomaterials and constitutes the motiva* Corresponding author. Tel.: +1 337 482 6430; fax: +1 337 482 1220. E-mail address: [email protected] (R.D.K. Misra).
tion of the study presented here. Furthermore, nanostructured metals with high strength will enable the use of thinner bioimplants because of high strength/weight ratio. Given the above considerations, the present study aims to combine fundamental aspects of materials science and engineering and cellular and molecular biology to favorably modulate the cell–substrate response on nanograined/ultrafine-grained (NG/UFG) austenitic stainless steels, using a fibroblast cell line to investigate the cellular response. Fibroblast cells are relevant because they are cells of connective tissue that synthesize extracellular matrix (ECM) and collagen. Regarding the success of a bioimplant in vivo, surface properties are critical for substrate–tissue interaction [10,11]. The chemistry and morphology of the surface can affect the attachment and subsequent growth behavior of cells on bioimplant material and consequently the compatibility between the host tissue and the implant [12,13]. Assuming that the ability of cells to adhere to a foreign surface is related to compatibility, cell adhesion is an important parameter for understanding biocompatibility [14,15]. It is safe to assume that substrate properties determine cell attachment, orientation, migration and metabolism [15–18]. Other surface parameters that modulate cell response include hydrophilicity, roughness and texture [19]. Higher roughness promotes bone-to-implant contact [20,21] and increases removal torque forces [22–24]. The surface properties determine the degree
1742-7061/$ - see front matter Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2010.01.034
3340
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
of bioimplant integration. Early colonization of the implant surface is likely to promote tissue repair in the peri-implant region, leading to effective integration of the metal implant [25,26]. The adhesion of cells to the substrate through an ECM provides signals that influence their ability to survive, proliferate and express specific developmental phenotypes [27]. The distinct and special properties of materials with sub-micron- to nanometer-sized grains are derived from the high density of grain/interphase boundaries, which are regions of high free energy [28]. Thermomechanical processing (TMP) is one of the primary methods to achieve grain refinement in metals [29–32]. Grain refinement limits imposed by conventional TMP can be overcome by the application of extensive plastic deformation, which leads to the formation of sub-micron or ultrafine structures in metallic materials [33,34]. The laboratory scale methods that have been adopted to obtain NG/UFG materials include equal channel angular pressing [34,35], accumulative roll bonding [36–38], high-pressure torsion (HPT) [39–42], multiple compression [43] and upsetting extrusion [44]. However, the ductility of the UFG materials produced by these methods is low compared to the coarse-grained (CG) materials. In this context, the work presented here is of particular significance because high ductility was obtained in the NG/UFG material. The objective of the study is to describe processing–structure– functional property relationships in austenitic stainless steel that continues to be used for medical device in view of its long history. The NG/UFG austenitic stainless steels were processed by a novel procedure involving controlled phase reversion of strain-induced martensite in a cold-rolled austenitic stainless steel. Multi-pass cold deformation (40–65%) of austenite at room temperature led to strain-induced transformation of austenite (face-centered cubic c) to dislocation-cell-type martensite (bcc a0 ), which upon annealing in the temperature range of 600–850 °C transforms to NG/UFG austenite through a diffusional reversion mechanism, depending on the temperature–time annealing sequence [45–48]. Given that NG/UFG steel is likely to have surface properties different from conventional CG structure, the grain size effects on cellular response of CG and NG/UFG austenitic stainless steels were investigated. A comparison of the degree of cell–substrate interaction was assessed by studying the cellular activity of fibroblast cells cultured on NG/UFG and CG austenitic steel substrates. Considering that proteins in the ECM of cells regulate cell–substrate interactions and influence cell adhesion, proliferation, viability and differentiation [26], fibronectin, actin and vinculin, as well as cytoskeletal organization and focal adhesion contacts, were studied. Thus, an important objective of the study was to examine how cells recognize surfaces and interact with them and each other. If the initial cell–substrate interaction is optimal, then greater quantities of ECM would be released than in less stimulated cells. Thus, we hypothesize that a surface to which cells can more readily attach would promote cell adhesion and proliferation, and lead to effective integration of implants. We examined this hypothesis by testing the differences in cellular response to NG/UFG and CG structures and their relationship to surface properties.
2. Experimental 2.1. Cell substrate materials The experimental material was commercially available 316L stainless steel. Stainless steel strips were obtained from Outokumpu Stainless Oy, Tornio and the nominal chemical composition in wt.% was Fe–0.017C–1.29Mn–17.3Cr–6.5Ni–0.15Mo–0.52Si. To develop NG/UFG structures, stainless steel strips were cold deformed to 40–65% in a laboratory rolling mill using several passes
(generally about 3–15% reduction per pass depending on the available rolling loads, where the contact area determines the rolling load). The reversion annealing was carried out in a Gleeble-1500 simulator in the range of 600–850 °C for periods of 10–100 s to obtain NG/UFG austenitic stainless steel. For experiments described here, the cold reduction was 62% and annealing was carried out at 800 °C for 1–10 s. The heating rate to the holding temperature was 200 °C s1, and following annealing, cooled in an airflow at rates of 200–400 °C s1. 2.2. Characterization of cell substrate materials The grain structure of CG and NG/UFG austenitic stainless steels was examined using light microscopy and transmission electron microscopy (TEM) operating at 120 kV (Hitachi 7600), respectively. It may be noted that the resolution and magnification of light microscope is inadequate to examine the grain structure of NG/ UFG steel, while CG structure can be conveniently examined by light microscope at low magnification, avoiding the need to prepare electron transparent samples for TEM. Thin foils for examining NG/UFG structures by transmission electron microscopy were prepared by punching 3 mm disks from the specimens and twinjet electropolishing with a solution of 10% perchloric acid in acetic acid as electrolyte. The mechanical properties of the cold-rolled and reversion-annealed austenitic stainless steel specimens were determined by tensile testing samples machined to a profile of 25 mm 25 mm with a 20 mm gage length. The structure of CG austenitic stainless steel was examined after electropolishing in a solution of 39% nitric at a current density of 30 mA cm2 and 1.3 V DC for 2–3 min. The crystallographic texture was studied in the surface layer. Samples (14 mm 24 mm) were cut, ground close to the investigated layer, and electropolished. Mo Ka radiation was used in the X-ray diffraction, and complete pole figures were measured in back reflection mode. Pole figures (1 1 0), (2 0 0), (2 1 1) and (3 1 0) were used for ferritic (a0 -martensite) phase and (1 1 1), (2 0 0), (2 2 0) and (3 1 1) for the austenite. The orientated distribution functions were calculated using a series expansion method (Imax = 22) from the pole figure data and plotted in contour lines in the Euler space (Bunge’s notation) [46,48]. Surface roughness of the CG and NG/UFG austenitic stainless steel substrates was examined by atomic force microscopy (AFM Dimension 3000, Digital Instruments, Santa Barbara, CA) in tapping mode. Surface wettability was measured as contact angle of deionized water using a goniometer system equipped with imaging analysis software (Ramé-Hart, Inc., Germany). An autopipetting system was used to ensure a droplet of water with uniform volume (0.5 lL). 2.3. Fibroblast–metal surface interaction Polished NG/UFG and CG stainless steel samples were cleaned in an ultrasonic bath with ethanol, followed by washing with deionized water, then wrapped individually in gauze and sterilized in an autoclave. Cell culture studies were performed using mouse fibroblast cell line (L-cell-L929) obtained from the American Type Cell Culture Collection (Manassas, VA, USA). Eagle’s minimum essential medium (Invitrogen Corporation, USA) supplemented with 10% horse serum, penicillin (100 U ml1) and streptomycin (100 lg ml1) was used to culture fibroblasts. Cells with 80–85% confluence obtained from Tflask cultures by trypsinization were used to seed onto the disk samples. Briefly, cells were washed with phosphate-buffered saline (PBS), incubated with 0.25% trypsin/0.53 mM ethylenediaminetetraacetic acid (EDTA) for 5–7 min to detach the cells from the Petri dish, dispersed in trypsin/EDTA, transferred to a centrifuge tube and centrifuged at 322 g for 5 min. The cell pellet obtained after
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
centrifugation was re-suspended in culture medium and then diluted with culture medium to get the required cell concentration. Subsequently, the sterilized steel disks were placed in a 24-well plate and incubated with cell suspension at 37 °C in a humidified atmosphere with 5% CO2 and 95% air. 2.4. Cell initial attachment, proliferation, viability and morphology Initial cell attachment, density, morphology and viability were investigated both qualitatively and quantitatively. Fibroblast (10,000 cells cm2) were seeded on CG and NG/UFG steel surfaces and incubated for 1, 2 and 4 h at 37 °C in a CO2 incubator (5% CO2 and 95% air) to study cell attachment. Cell-seeded NG/UFG and CG steel surfaces were cultured for up to 7 days at 37 °C in a CO2 incubator. The culture medium was changed every 2 days. The initial fibroblast attachment and viability on different steel surfaces was measured using the MTT (3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide) assay at predetermined times. The assay is based on the reductase activity in mitochondria of living cells. These enzymes cleave the tetrazolium ring, which turns the pale yellow MTT into dark blue formazan, the concentration of which is directly proportional to the number of metabolically active cells. The samples were washed twice with PBS and incubated with fresh culture medium containing MTT (0.5 mg ml1 medium) at 37 °C for 4 h in dark. The unreacted dye was then removed and dimethyl sulfoxide was added to release the intracellular purple formazan product into solution. The absorbance of this solution was quantified by photospectrometry at 570 nm with a plate reader (Bio TEK Instrument, EL307C). Additionally, cell attachment was also measured using the trypan blue dye-exclusion assay. Briefly, cell-seeded disks were trypsinized and the cell pellet was collected as described above in Section 2.3. The cell pellet was suspended in culture medium, incubated with 0.4% trypan blue stain for 5 min at room temperature and the viable cells were counted by a hemocytometer. To further corroborate the MTT assay and to acquire an understanding of grain size effect on initial cell attachment, the cellseeded disks were evaluated using fluorescence microscope after staining with the nucleic acid dye, Hoechst 33342. The cell-seeded disks were washed twice with PBS, incubated (10 lg dye per ml of PBS) for 10 min at 25 °C and viewed under a fluorescence microscope (Nikon, ECLIPSE E 600 FN) with excitation and emission maxima of 346 and 442 nm, respectively; cell nuclei appeared as blue fluorescent spots. A similar procedure was carried out with 24 h cultures to determine cell proliferation. To observe the initial spreading pattern and morphology, fibroblasts were seeded on the samples, grown for 1 and 24 h, fixed with 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4, for 20 min, rinsed with PBS, dehydrated with a graded series of ethanol (25–100%) and critical point dried. The samples were sputtercoated with gold and examined in a JEOL JSM 6300 field emission scanning electron microscope. 2.5. Immunofluorescence and confocal microscopy The studies of cell–substrate interaction were further analyzed using an immunostaining technique. The expression and distribution of vinculin and the organization of actin filaments were studied via immunofluorescence and confocal microscopy. The response of cells with different substrates is expected to result in a differential signal of vinculin-associated actin stress fibers. Fibroblast-seeded CG and NG/UFG steel samples were cultured for 2 days and used for immunostaining. The cell-seeded disks were twice washed with PBS, fixed with 4% paraformaldehyde for 20 min, permeabilized with 0.1% Triton X-100 for 5 min and blocked with 0.1% bovine serum albumin (BSA) for 30 min. Subse-
3341
quently, immunocytochemistry staining was performed for fibronectin, actin and vinculin expression. Fibronectin was immunostained (labeled) with diluted mouse monoclonal antibody against fibronectin (1:800) (Sigma–Aldrich, St. Louis, MO, USA) followed by rabbit–antimouse fluorescein isothiocyanate (FITC)-conjugated secondary antibody at a working dilution of 1:200 (Sigma–Aldrich, St. Louis, MO, USA). Cell nuclei were then stained with 40 ,6-diamidino-2-phenylindole (DAPI; Chemicon International, Inc., Temecula, CA, USA) at a working dilution of 1:1000 for 5 min. The actin cytoskeleton and focal adhesion contacts were stained with a focal adhesion staining kit (Chemicon International, Inc., Temecula, CA, USA). Actin and vinculin double staining was performed by incubating cells in diluted primary antibody (anti-vinculin) in blocking solution (1:200) for 1 h and subsequently labeling them with diluted secondary antibody (1:100) (goat–antimouse FITC conjugate; Chemicon International, Inc., Temecula, CA, USA) for 45 min, with simultaneous incubation with diluted tetramethyl rhodamine isothiocyanate-conjugated phalloidin (1:400) [9]. Finally, cell nuclei were labeled with DAPI as described above. After immunostaining, samples were mounted on a glass slide using anti-fade mounting solution, covered with a coverslip and examined by confocal laser scanning microscopy (Leica DM 6000B and Leica DM 600M). 2.6. Quantification of protein and sodium dodecyl sulphate– polyacrylamide gel electrophoresis (SDS–PAGE) The cell–substrate interaction was quantitatively investigated by measuring the expression levels of proteins of fibroblast on NG/UFG and CG steel surfaces. Cell-seeded disks were used after 2 days to analyze total protein content on each surface. Cells grown on the substrates were lysed with Tris-lysis buffer (1 dilution) (20 mM Tris–HCl, pH 7.5; 150 mM NaCl, 1 mM EDTA and 1% Triton X-100) (Sigma–Aldrich, St. Louis, MO, USA) containing protease inhibitors (Sigma–Aldrich, St. Louis, MO, USA). Quantitative analysis of protein was done via a Mini Bradford protein assay (Sigma–Aldrich, St. Louis, MO, USA). This assay is based on the equilibrium between three forms of Coomassie Blue G dye. Proteins bind to the reagent under acidic conditions and cause a spectral shift from reddish/brown to blue. Spectrophotometric measurement of the resulting level of blue solution was used to determine the concentration of protein. The measurement was done at 595 nm. Standard protein concentration plots were generated using 0–20 lg ml1 BSA and were run parallel with each experiment. Subsequently, reduced proteins were separated by SDS–PAGE (using 10% sodium dodecyl sulphate–polyacrylamide gel). The SDS–PAGE molecular weight standard (SDS–PAGE marker for molecular weight, Sigma–Aldrich, St. Louis, MO, USA) was used to determine the relative sizes of the proteins from the samples. 2.7. Statistical analysis of data Values for each experiment were normalized to control experiments and expressed as the mean of at least three replicates ± SD (standard deviation). Three sets of experiments were carried out for each study. Statistical analysis was performed using a oneway analysis of variance with 95% confidence interval. 3. Results 3.1. Grain structure, mechanical and physical properties, surface roughness and wettability The average grain size of the CG austenitic stainless steel was 52 lm. The yield strength and elongation were 205 MPa and 40%,
3342
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
respectively (Fig. 1a). A TEM micrograph of the NG/UFG austenitic stainless steel (Fig. 1b) shows steel strips with 62% multi-pass cold reduction subjected to reversion annealing at 800 °C in a Gleeble thermo-simulator for 1–10 s. This reversion annealing temperature–time sequence provided an excellent combination of yield strength of 800 MPa and ductility of 45%, and resulted in nearcomplete transformation of martensite to austenite. The grain structure consisted of nanocrystalline (d < 100 nm), ultrafine (100–500 nm) and some sub-micron (500–1000 nm) grains. These optimized parameters are specific to the experiments described here and any generalization to application-oriented ap-
proach will necessitate fine-tuning of the experimental parameters. Lowering the annealing temperatures to about 750 °C allowed adequate holding time, while at higher temperatures (900–1000 °C) the grain size increased to greater than 0.5 lm. These experiments suggested that the annealing temperatures of 800 °C or lower reduced grain growth. Polished surfaces of CG and NG/UFG austenitic stainless steel exhibited almost similar surface roughness and topography. The average roughness of 3 lm 3 lm scan area for CG and NG/UFG austenitic stainless steel was 1.45 ± 0.21 and 1.52 ± 0.29 nm, respectively. In contrast, the surface wettability, measured as the
Fig. 1. (a) Light micrograph of CG and (b) transmission electron micrograph of NG/UFG austenitic stainless steels. The CG steel has an average grain size of 52 lm. NG/UFG steel is characterized by a combination of nanocrystalline grains (d < 100 nm), ultrafine grains (100–500 nm) and some grains of size 500–1000 nm.
1
1.5
1
0.5
0
0.8 0.6 0.4 0.2 0
1
2
1
4
2
4
Time (h)
Time (h) 2.0
Cell density (MTT assay)
b
Coarse-grained steel NG/UFG steel
2
a
Coarse- grained steel NG/UFG steel
Cell density (*10000 cells/cm )
Cell attachment (MTT assay) MTT
2
c
Coarse- grained steel NG/UFG steel 1.5
1.0
0.5
0.0
1
2
5
7
Time (h) Fig. 2. Histograms representing initial cell attachment after a culture time of 1–4 h by (a) MTT assay, (b) trypan blue assay and (c) metabolism of fibroblasts up to 1–7 days using MTT assay on NG/UFG and CG substrates. The fibroblasts showed significantly greater cell attachment and proliferation on the NG/UFG substrate than on the CG material. The higher density of cells on the NG/UFG substrates in comparison to the CG substrate was visible within the first hour of culture. Cell viability was significantly higher on the NG/UFG substrates compared to the CG substrates (p < 0.05).
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
water contact angle, was 75.14 ± 5.34° for CG and 66.01 ± 2.19° for NG/UFG austenitic stainless steels. Thus, NG/UFG steel is more hydrophilic than CG steel. 3.2. Cellular and molecular response to grain structure Initial cell attachment is one of the indicators of the biological properties of the substrates. Fibroblast attachment after 1 h was appreciably greater on the NG/UFG substrate than on the CG substrate (Figs. 2 and 3). Cell density was higher on NG/UFG in relation to CG at 1, 2 and 4 h of culture time (Fig. 2a and b). Cell metabolism, measured by mitochondrial reduction of MTT, was higher for the NG/UFG steel than for the CG steel, which indicates good conditions for fibroblast growth (Fig. 2c). The cell density on both the NG/UFG and CG substrates increased over time, but was significantly higher on the surface of the NG/UFG steel than on the CG steel, indicating the superior cytocompatibility of NG/UFG. A similar conclusion is derived from fluorescence micrographs of cells stained with nucleic acid-specific dye (Fig. 3). This indicates that the cell attachment to the surface is influenced by the grain structure of the steel substrates, possibly facilitating rapid adsorption of proteins onto the substrate [49]. The differences in the density as observed by fluorescence microscopy after 24 h (Fig. 3c and d) indicates that the enhanced cell count data were not skewed by cell development or surface adaptation over time. The initial cell attachment on NG/UFG substrate was higher than on the CG substrate (Fig. 3a and b). It is pertinent to mention that the cell attachment to substrates is fundamental for subsequent cellular activities, including spreading, division and differentiation.
3343
Scanning electron microscopy revealed remarkable differences in the morphology of the fibroblasts grown on the NG/UFG and CG substrates (Figs. 4 and 5), supporting the cell attachment data. The fibroblasts on the NG/UFG substrate attached and spread widely, with numerous cellular extensions, even during the initial incubation of 1 h (Fig. 4a and b). This was subsequently followed by extensive proliferation and cellular processes such that they appear to attach firmly to the NG/UFG surface, as implied by the observations at 24 h culture (Fig. 5a and b). In contrast, the fibroblast cultured on the CG substrate were spherical, but without any cellular extension at 1 h incubation time (Fig. 4c and d). After 24 h culture, cell proliferation was observed to increase on the CG substrate (Fig. 5c), but less so in comparison to the NG/UFG substrate (Fig. 5a). Similarly, the ECM formation was less significant on the CG substrate (Fig. 5d). The fibroblasts on the NG/UFG surface were characterized by starlike morphology with numerous cytoplasmic extensions (Fig. 5a and b). Frequent overlapping of cells or forming cytoplasmic bridges was observed. Thus, in summary, the extent of cell spreading was remarkably greater on the NG/UFG surface compared to the CG surface. This implies superior cellular attachment and cell–substrate interaction with the NG/UFG substrate. The production of proteins from fibroblast grown on the NG/ UFG and CG substrates were examined to further understand the cellular response. Fibronectin expression by fibroblast was greater on the NG/UFG steel than on the CG steel substrate (Fig. 6), as documented by the distinct network with stronger fluorescence intensity of immunostained fibronectin. There were fewer cells on the CG surface than on the NG/UFG surface. It should be noted, however, that the cells on the NG/UFG surface are clumped and pro-
Fig. 3. Fluorescence micrographs of fibroblasts illustrating initial cell attachment at 1 h and cell density at 24 h on NG/UFG and CG steel surfaces after staining the cell nuclei with Hoechst 33258. The cell nuclei showed blue fluorescence after staining. Cell attachment and density on the NG/UFG austenitic stainless steel (a and c) were higher than on the CG austenitic stainless steel (b and d).
3344
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
Fig. 4. Comparison of fibroblasts attachment morphology at 1 h as imaged by scanning electron microscopy on NG/UFG (a and b) and CG austenitic stainless steel (c and d). At low magnification, cells on the NG/UFG steel exhibited a higher number of attached cells and numerous cellular extensions with thicker extracellular matrices compared to the CG surface. Cells on the CG surface are spherical. High-magnification micrographs (b and d) show the greater extent of spreading on the NG/UFG surface than on the CG surface. The cells on the NG/UFG austenitic stainless steel that show numerous cellular extensions are indicative of the extensive attachment and interaction with the substrate.
duce more fibronectin. The cells on the CG surface are largely single cells, which produce less fibronectin, but where they are clumped they produce just as much fibronectin. The production of fibronectin is therefore most likely related to the clumping of cells. The expression level of vinculin, a protein that forms focal contacts and actin stress fibers, after 2 days of culture also showed a greater expression level at the edges and well-defined stress fibers on the NG/UFG steel than on the CG substrate. An example of this is presented in Fig. 7. The expression level of vinculin on CG steel substrate was lower and actin was less prominent on CG steel (Fig. 7). Thus, the examination of individual proteins that are related to cell attachment also indicated that NG/UFG surface was superior in cellular response. Fig. 8 is a confocal micrograph that combines cytoskeletal elements and nuclei and corroborates the relationship between vinculin focal contacts and actin stress fibers in an individual cell. The vinculin focal contact site interconnects with the actin leading edges and the steel substrate. A higher expression level of vinculin and actin is evident for fibroblasts cultured on the NG/UFG surface as compared to the CG surface. In summary, the confocal micrograph for the NG/UFG substrate demonstrates greater vinculin and actin signals along the periphery of cells and point to a higher degree of cellular and molecular interaction between fibroblasts and the NG/UFG steel substrate. These findings were further corroborated by the expression level of different proteins via SDS– PAGE (Fig. 9a) and by quantifying the total protein content
(Fig. 9b) released from the fibroblasts. Qualitative analysis by Western blot analysis and quantitative analysis by the Mini Bradford protein assay are in agreement with the immunofluorescence results and indicate elevated expression levels of proteins on NG/ UFG surface after 2 days of culture compared to the CG surface (Fig. 9). In both analyses, the total protein content on the surface including the extracellular fibronectin protein and the intracellular proteins vinculin and actin are measured. Thus, we can conclude that immunofluorescence and protein analysis studies are consistent and support the viewpoint that the NG/UFG steel substrate had a superior cellular response as compared to the CG steel substrate.
4. Discussion The observations concerning the biological significance of NG/ UFG and CG structures suggest the determining role of grain structure on cell attachment, morphology, and growth. It seems clear that the greater hydrophilic character and finer grain size of the NG/UFG structure of austenitic stainless steel positively influences the biofunctional property of fibroblast attachment. An NG/UFG surface has a more open lattice in the positions of high-angle boundaries as compared to a CG steel, which provides sites for attachment of cells. The higher degree of ECM on the NG/UFG substrate observed here determines events on the cell surface and communication to intracellular cellular processes by the cytoskel-
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
3345
Fig. 5. Comparison of fibroblast morphology at 24 h, as imaged by scanning electron microscopy on NG/UFG (a and b) and CG austenitic stainless steel (c and d). Fibroblasts grown on NG/UFG and CG surfaces are significantly different. Cells on the NG/UFG surface exhibit wide spreading with thick and numerous cellular extensions. These features are significantly less pronounced on the CG surface.
Fig. 6. Merged fluorescence micrographs representing immunocytochemistry of fibronectin expressed by fibroblasts after incubation for 2 days on NG/UFG (a and b) and CG stainless steel surfaces (c and d). A higher fluorescence intensity and expanded network of fibronectin expression along with a higher cell density is observed after labeling of cell nuclei with DAPI.
3346
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
eton [8,9]. We have studied the important proteins that affect cell– substrate interaction, including fibronectin, vinculin and actin. Fibronectin is involved in cell adhesion and functions as a growth stimulator [8,9,50,51], and is one of the earliest products to be released [50,52–55]. Other components of the ECM contain type 1 collagen and the glycoprotein fibronectin, both of which regulate cell attachment and differentiation [56,57]. Surface properties that affect protein expression, secretion and thus adsorption and cell attachment include hydrophilicity,
ionic bonding, electrostatic and van der Waals interactions [58,59]. In this regard, surface roughness and grain structure are important factors that affect the surface wettability. Roughness is expected to play an important role in cell attachment, proliferation and adsorption of proteins. Generally, rougher surfaces have been observed to promote cell attachment and enzyme activity [58]. From the practical viewpoint, it is believed that the distance between peaks or valleys of the surface should not exceed the ability of the cells to form focal attachment on
Fig. 7. An illustration of organization and assessment of focal contacts and actin stress fibers of fibroblasts after 2 days culture on NG/UFG (a and b) and CG (c and d) stainless steel substrates. The fibroblasts grown on the NG/UFG steel (a) show a larger number of vinculin focal adhesion contacts in relation to CG steel (c). The higher number of focal adhesion points corresponded well with thicker bundles of actin stress fibers on NG/UFG steel (b) than on CG steel (d).
Fig. 8. Confocal micrographs showing combined distribution of cytoskeletal organization and focal adhesion contacts; actin (red), vinculin (green) and nucleus (blue) in fibroblasts cultured for 2 days on NG/UFG (a) and CG (b) stainless steel surfaces. The focal contact sites are where vinculin is linked at the actin leading edges, which connect cells to the steel surface.
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
a
MW
Marker CG
10
b
NG/UFG
3347
205
Fibronectin
116 .
Vinculin
97 66
Actin
45 Lane
1
2
3
Total Protein Content (µg)
(KD)
8
6
4
2
0 CG NG/UFG Type of Stainless Steel
Fig. 9. (a) Expression of proteins from fibroblasts grown on NG/UFG and CG steel after 2 days of culture. Determination of proteins by SDS–PAGE illustrates a higher intensity of different proteins on NG/UFG substrate compared to CG substrate. Bands of proteins observed in the sample lanes are the relative sizes of fibronectin, vinculin and actin. S molecular weight standard marker was loaded in lane 1. (b) Quantitative analysis of proteins by Mini Bradford protein assay showing a higher total protein content in fibroblasts cultured on the NG/UFG substrate in relation to the CG substrate (n = 3).
two or more peaks, or else the cell will recognize the rough surface as a smooth surface [58]. With regard to the grain structure, the wettability can be distinctly different with different grain sizes because of differences in free energy, and this difference is further amplified by the structural arrangements of atoms within the grain [9]. In our case, given that the roughness and constituent phase (austenite) were similar for both the CG and NG/UFG austenitic stainless steel substrates, the hydrophilicity (contact angle: 75.14 ± 5.34° for CG and 66.01 ± 2.19° for NG/UFG) and grain size are relevant factors that contribute to the differences in cell attachment, proliferation, morphology, expression level of vinculin and expression of actin stress fibers. Hydrophilic surfaces are known to enhance activity of alkaline phosphatases and osteocalcin, which promote an osteogenic microenvironment [59,60]. Furthermore, hydrophilic surfaces promote adsorption of fibronectin and albumin [61,62]. The combination of enhanced cell attachment and proliferation and the higher expression levels of fibronectin and vinculin on the cells grown on the NG/UFG steel substrates indicates that this substrate promotes cellular activity and the processes required for strong cell– substrate interaction, and therefore cell adhesion. In summary, during the early stages, the initial attachment of cells is favored on the NG/UFG substrate. Subsequently, the adhesion of cells via ECM influences their ability to proliferate, which is also promoted on the NG/UFG substrate. Interestingly, a similar observation of cellular response was made by Faghihi et al. [9] on ultrafine titanium processed by the HPT method. The degrees of cell attachment and cell proliferation were significantly increased on the HPT-processed titanium surface. The improved cell activity was attributed to the nanostructured feature of titanium substrates, which provided a higher degree of surface wettability. It is unlikely that the strong texture of NG/UFG in comparison to the CG austenitic stainless steel is responsible for the stronger attachment of cells, because the crystallographic orientations were similar [46]. Thus, the modulation of cellular response on the surface of CG and nanostructured metallic substrates depends on grain size but not crystal structure. This study advances our understanding of cell–substrate interaction and provides strong evidence for a positive cellular response related to enhanced cell attachment and proliferation on nanostructured metal substrates. Our observations of strongly pro-
nounced cellular and molecular interaction between cells and NG/UFG material suggest the growth of connective tissues. The study underscores that nanostructured or ultrafine-grained bioimplants provide superior cell–substrate attachment and biocompatibility. 5. Conclusions and future outlook The development of NG/UFG structures for bioimplants is characterized by an enhanced cellular response combined with a high strength/weight ratio. The positive influence of the NG/UFG structure on cellular response was reflected in the improved cell attachment, proliferation, viability, morphology and spread compared to the CG structure. Furthermore, immunocytochemistry analysis through the expression levels of ECM protein, fibronectin, vinculin and associated actin cytoskeleton indicated the superior cytocompatibility of the NG/UFG substrate in relation to the CG substrate. Observations of protein analysis, such as SDS–PAGE and total protein assays, were consistent with cell density measurement and electron microscopy studies, suggesting the enhanced cell–substrate interaction and activity of the NG/UFG structure. The improved cellular response of the NG/UFG substrate is attributed to grain size and hydrophilicity, and a more open lattice in the position of high-angle boundaries. In summary, the processing–structure–biofunctional property study addresses challenges to the engineering of bulk nanostructured materials with specific physical and surface properties for medical devices with improved cellular response and strengthens the foundation of nanostructured materials for biomedical applications. Appendix A. Figures with essential colour discrimination Certain figures in this article, particularly Figures 3, 6, 7, 8 and 9, 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.2010.01.034. References [1] Borsari V, Giavaresi G, Fini M, Torrielli P, Salito A. J Biomed Mater Res B Appl Biomater 2005;75B:359–68. [2] Arenas MA, Tate TJ, Conde A, De Damboreba J. Br Corros J 2000;35:232–6.
3348
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
[3] Willman G. Adv Eng Mater 1999;1:95–105. [4] Webster TJ, Ergun C, Doremus RH, Siegel RW, Bizious R. Biomaterials 2000;67:1803–10. [5] Webster TJ, Siegel RW, Bizios R. Biomaterials 1999;20:1221–7. [6] Thapa A, Webster TJ, Haberstroth KM. J Biomed Mater Res A 2003;67:1374–83. [7] Webster TJ, Smith TA. J Biomed Mater Res A 2005;74A:677–86. [8] Faghihi S, Azari F, Zhilyeav AP, Vali H, Tabrizian M. Adv Mater 2007;19:1069–73. [9] Faghihi S, Azari F, Zhilyeav AP, Vali H, Tabrizian M. Biomaterials 2007;28:3887–95. [10] Hunter A, Archer CW, Walker PS, Blunn GW. Biomaterials 1995;16:287–95. [11] van Recum AF, Shannon CE, Cannon EC, Long KJ, van Kooten TG, Meyle J. Tissue Eng 1996;2:241–53. [12] van Recum AF, van Kooten TG. J Biomater Sci Polym Ed 1995;2:181–98. [13] Chesmel K, Black J. J Biomed Mater Res 1995;29:1089–99. [14] Bagambisa FB, Joos U. Biomaterials 1990;11:50–9. [15] Sailaja GS, Ramesh P, Kumary TV, Varma HK. Acta Biomater 2006;2:651–7. [16] Hott M, Noel B, Assolant DB, Rey C, Marie PJ. J Biomed Mater Res 1997;37:508–16. [17] Meyer U, Buchter A, Wiesmann HP, Joos U, Jones DB. Eur Cells Mater J 2005;9:39–49. [18] Helen HL, El-Amin Saadig F, Scott DK, Cato L. J Biomed Mater Res 2002;64A:465–74. [19] Zhao G, Schwartz Z, Wieland M, Rupp F, Geis-Gerstorfer J, Cochran DL, et al. J Biomed Mater Res 2005;74A:49–58. [20] Goldberg VM, Jinno T. J Long Term Eff Med Implants 1999;9:11–21. [21] Buser D, Schenk R, Steinemann S, Fiorellini J, Fox C, Stich H. J Biomed Mater Res 1991;25:889–902. [22] Li D, Ferguson SJ, Beutler T, Cochran DL, Sittig C, Hirt HP, et al. J Biomed Mater Res 2002;60:325–32. [23] Wennerberg A, Wktessabi A, Albrektsson T, Joansson C, Andersson B. Int J Oral Maxillofac Implants 1997;12:486–94. [24] Klokkevold PR, Nishimura RD, Adachi M, Caputo A. Clin Oral Implants Res 1997;8:442–7. [25] Cooper LF, Masuda T, Yliheikkila PK, Felton DA. Int J Oral Maxillofac Implants 1998;13(2):163–74. [26] Anselme K. Biomaterials 2000;21:667–81. [27] Garcia AJ, Vega MD, Boettiger D. Mol Biol Cell 1999;10:785–98. [28] Gleiter H. Nanostruct Mater 1992;1:1. [29] Misra RDK, Thompson SW, Hylton TA, Boucek AJ. Metall Mater Trans A 2001;35:745. [30] Misra RDK, Tenneti KK, Weatherly GC, Tither G. Metall Mater Trans A 2004;35:341. [31] Misra RDK, Nathani H, Hartmann JE, Siciliano F. Mater Sci Eng A 2005;394: 339. [32] Shanmugam S, Misra RDK, Hartmann JE, Jansto S. Mater Sci Eng A 2006;437:436. [33] Song R, Ponge D, Raabe D, Kaspar R. Acta Mater 2005;53:845.
[34] Humphreys FJ, Prangnell PB, Bowen JR, Gholinia A, Harris C. Philos Trans R Soc Lond A 1999;357:1681. [35] Pitan C, Hashimoto T, Kawazoe M, Nagahora J, Higashi K. Mater Sci Eng A 2000;280:62. [36] Tsuji N, Saito Y, Utsunomiya H, Tanigawa S. Scripta Mater 1999;40:795. [37] Park KT, Lee CS, Shin DH. Rev Adv Mater Sci 2005;10:133. [38] Costa ALM, Reis ACC, Kestens L, Andrade MS. Mater Sci Eng A 2005;406:279. [39] Webster TJ, Siegel RW, Bizios R. Biomaterials 1999;20:1221. [40] Thapa A, Webster TJ, Haberstroh KM. J Biomed Mater Res A 2003;67:1374. [41] Zhilyeav AP, Nurislamova GV, Kim BK, Baro MD, Szpunar JA, Langdon TG. Acta Mater 2003;51:753. [42] Ivanisenko Y, Lojkowski W, Valiev RZ, Fecht HJ. Acta Mater 2003;51:5555. [43] Belyakov A, Sakai T, Miura H, Kaibyshev R. Scripta Mater 2000;42:319. [44] Lianxi H, Yuping L, Erde W, Yang Y. Mater Sci Eng A 2006;422:327. [45] Misra RDK, Nayak S, Mali S, Shah J. Metall Mater Trans A. doi:10.1007/511661009-9920-B. [46] Misra RDK, Ravikumar B, Somani MC, Karjalainen LP. Scripta Mater 2008;59:79. [47] Somani MC, Karjalainen LP, Misra RDK. In: Karjalainen LP, Hertzman S, editors. Proceedings of the 6th european stainless steel conference science and market, Helsinki; 10–13 June 2008. p. 519. [48] Somani MC, Karjalainen LP, Misra RDK, Karjalainen A. Metall Mater Trans A 2009;40:729. [49] Andrade JD, Hlady V. Ann NY Acad Sci 1987;516:158. [50] Chen LB, Murray A, Segal RA, Bushnell A, Walsh ML. Cell 1978;14:377–91. [51] Hynes RO, Bye JM. Cell 1974;3:113–20. [52] MacDonald DE, Rapuano BE, Deo N, Stranick M, Somasundaran P, Boskey AL. Biomaterials 2004;25:3135–46. [53] Meyer U, Szulczewski DH, Moller K, Heide H, Jones DB, Gross U. Cells Mater J 1993;3:129–40. [54] Daculsi G, Pilet P, Cottrel M, Guicheux G. J Biomed Mater Res 1999;47:228–33. [55] Globus RK, Doty SB, Lull JC, Holmuhamedov E, Humphries MJ, Damsky CH. J Cell Sci 1998;111:1385–93. [56] Nordahl J, Widholm SM, Hultenby K, Reinholt F. Calcif Tissue Int 1995;57:442–9. [57] Winnard R, Gerstenfeld L, Toma C, Franceschi R. J Bone Miner Res 1995;10:1969–77. [58] Lincks J, Boyan BD, Blanchard CR, Lohmann CH, Liu Y, Cochran DL, et al. Biomaterials 1998;19:2219–32. [59] Webster TJ, Ergun C, Doremus RH, Siegel RW, Bizios R. J Biomed Mater Res 2000;51:475. [60] Zhao G, Schwartz Z, Wieland M, Rupp F, Geis-Gerstorfer J, Cochran DL, et al. J Biomed Mater Res A 2005;74:49. [61] Sousa SR, Moradass-Ferreira P, Saramago B, Melo LV, Barbosa MA. Langmuir 2004;20:9745. [62] Sousa SR, Moradass-Ferreira P, Barbosa MA. J Mater Sci Mater Med 2005;16:1173.
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat
Biological significance of nanograined/ultrafine-grained structures: Interaction with fibroblasts R.D.K. Misra a,*, W.W. Thein-Han a, T.C. Pesacreta b, M.C. Somani c, L.P. Karjalainen c a
Biomaterials and Biomedical Engineering Research Laboratory, Center for Structural and Functional Materials, University of Louisiana at Lafayette, P.O. Box 44130, Lafayette, LA 70504, USA b Department of Biology and Microscopy Center, University of Louisiana at Lafayette, P.O. Box 42451, Lafayette, LA 70504, USA c Department of Mechanical Engineering, The University of Oulu, P.O. Box 4200, 90014 Oulu, Finland
a r t i c l e
i n f o
Article history: Received 20 August 2009 Received in revised form 12 November 2009 Accepted 22 January 2010 Available online 28 January 2010 Keywords: Nanostructured materials Phase reversion Stainless steel Cellular response Fibroblasts
a b s t r a c t Given the need to develop high strength/weight ratio bioimplants with enhanced cellular response, we describe here a study focused on the processing–structure–functional property relationship in austenitic stainless steel that was processed using an ingenious phase reversion approach to obtain an nanograined/ ultrafine-grained (NG/UFG) structure. The cellular activity between fibroblast and NG/UFG substrate is compared with the coarse-grained (CG) substrate. A comparative investigation of NG/UFG and CG structures illustrated that cell attachment, proliferation, viability, morphology and spread are favorably modulated and significantly different from the conventional CG structure. These observations were further confirmed by expression levels of vinculin and associated actin cytoskeleton. Immunofluorescence studies demonstrated increased vinculin concentrations associated with actin stress fibers in the outer regions of the cells and cellular extensions on NG/UFG substrate. These observations suggest enhanced cell–substrate interaction and activity. The cellular attachment response on NG/UFG substrate is attributed to grain size and hydrophilicity and is related to more open lattice in the positions of high-angle grain boundaries. Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
1. Introduction In the replacement or substitution of hard tissue implants, metallic materials including austenitic stainless steels and titanium alloys are widely used biomaterials as devices for bone fixation, partial/total joint replacement and spring clips for the repair of large aneurysmal defects because they are inherently corrosion resistant and possess the necessary mechanical strength and biocompatibility [1–3]. In recent years, surface modification processes such as heat treatment have been explored to induce nanostructuring on the surface and consequently promote osseointegration and modulation of cellular activity. While ultrafine structures may provide benefits of enhanced cellular attachment, stimulate metabolic activity and upregulate protein formation [4–9], it is important that we consider both the surface and bulk properties of the material that is in direct contact with the bone for long-term stability and an acceptable success rate of prosthetic rehabilitation. Thus, processing of bulk nanostructured materials assumes particular significance because it is a viable and potentially transformative approach to favorably modulate the cellular response of biomaterials and constitutes the motiva* Corresponding author. Tel.: +1 337 482 6430; fax: +1 337 482 1220. E-mail address: [email protected] (R.D.K. Misra).
tion of the study presented here. Furthermore, nanostructured metals with high strength will enable the use of thinner bioimplants because of high strength/weight ratio. Given the above considerations, the present study aims to combine fundamental aspects of materials science and engineering and cellular and molecular biology to favorably modulate the cell–substrate response on nanograined/ultrafine-grained (NG/UFG) austenitic stainless steels, using a fibroblast cell line to investigate the cellular response. Fibroblast cells are relevant because they are cells of connective tissue that synthesize extracellular matrix (ECM) and collagen. Regarding the success of a bioimplant in vivo, surface properties are critical for substrate–tissue interaction [10,11]. The chemistry and morphology of the surface can affect the attachment and subsequent growth behavior of cells on bioimplant material and consequently the compatibility between the host tissue and the implant [12,13]. Assuming that the ability of cells to adhere to a foreign surface is related to compatibility, cell adhesion is an important parameter for understanding biocompatibility [14,15]. It is safe to assume that substrate properties determine cell attachment, orientation, migration and metabolism [15–18]. Other surface parameters that modulate cell response include hydrophilicity, roughness and texture [19]. Higher roughness promotes bone-to-implant contact [20,21] and increases removal torque forces [22–24]. The surface properties determine the degree
1742-7061/$ - see front matter Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2010.01.034
3340
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
of bioimplant integration. Early colonization of the implant surface is likely to promote tissue repair in the peri-implant region, leading to effective integration of the metal implant [25,26]. The adhesion of cells to the substrate through an ECM provides signals that influence their ability to survive, proliferate and express specific developmental phenotypes [27]. The distinct and special properties of materials with sub-micron- to nanometer-sized grains are derived from the high density of grain/interphase boundaries, which are regions of high free energy [28]. Thermomechanical processing (TMP) is one of the primary methods to achieve grain refinement in metals [29–32]. Grain refinement limits imposed by conventional TMP can be overcome by the application of extensive plastic deformation, which leads to the formation of sub-micron or ultrafine structures in metallic materials [33,34]. The laboratory scale methods that have been adopted to obtain NG/UFG materials include equal channel angular pressing [34,35], accumulative roll bonding [36–38], high-pressure torsion (HPT) [39–42], multiple compression [43] and upsetting extrusion [44]. However, the ductility of the UFG materials produced by these methods is low compared to the coarse-grained (CG) materials. In this context, the work presented here is of particular significance because high ductility was obtained in the NG/UFG material. The objective of the study is to describe processing–structure– functional property relationships in austenitic stainless steel that continues to be used for medical device in view of its long history. The NG/UFG austenitic stainless steels were processed by a novel procedure involving controlled phase reversion of strain-induced martensite in a cold-rolled austenitic stainless steel. Multi-pass cold deformation (40–65%) of austenite at room temperature led to strain-induced transformation of austenite (face-centered cubic c) to dislocation-cell-type martensite (bcc a0 ), which upon annealing in the temperature range of 600–850 °C transforms to NG/UFG austenite through a diffusional reversion mechanism, depending on the temperature–time annealing sequence [45–48]. Given that NG/UFG steel is likely to have surface properties different from conventional CG structure, the grain size effects on cellular response of CG and NG/UFG austenitic stainless steels were investigated. A comparison of the degree of cell–substrate interaction was assessed by studying the cellular activity of fibroblast cells cultured on NG/UFG and CG austenitic steel substrates. Considering that proteins in the ECM of cells regulate cell–substrate interactions and influence cell adhesion, proliferation, viability and differentiation [26], fibronectin, actin and vinculin, as well as cytoskeletal organization and focal adhesion contacts, were studied. Thus, an important objective of the study was to examine how cells recognize surfaces and interact with them and each other. If the initial cell–substrate interaction is optimal, then greater quantities of ECM would be released than in less stimulated cells. Thus, we hypothesize that a surface to which cells can more readily attach would promote cell adhesion and proliferation, and lead to effective integration of implants. We examined this hypothesis by testing the differences in cellular response to NG/UFG and CG structures and their relationship to surface properties.
2. Experimental 2.1. Cell substrate materials The experimental material was commercially available 316L stainless steel. Stainless steel strips were obtained from Outokumpu Stainless Oy, Tornio and the nominal chemical composition in wt.% was Fe–0.017C–1.29Mn–17.3Cr–6.5Ni–0.15Mo–0.52Si. To develop NG/UFG structures, stainless steel strips were cold deformed to 40–65% in a laboratory rolling mill using several passes
(generally about 3–15% reduction per pass depending on the available rolling loads, where the contact area determines the rolling load). The reversion annealing was carried out in a Gleeble-1500 simulator in the range of 600–850 °C for periods of 10–100 s to obtain NG/UFG austenitic stainless steel. For experiments described here, the cold reduction was 62% and annealing was carried out at 800 °C for 1–10 s. The heating rate to the holding temperature was 200 °C s1, and following annealing, cooled in an airflow at rates of 200–400 °C s1. 2.2. Characterization of cell substrate materials The grain structure of CG and NG/UFG austenitic stainless steels was examined using light microscopy and transmission electron microscopy (TEM) operating at 120 kV (Hitachi 7600), respectively. It may be noted that the resolution and magnification of light microscope is inadequate to examine the grain structure of NG/ UFG steel, while CG structure can be conveniently examined by light microscope at low magnification, avoiding the need to prepare electron transparent samples for TEM. Thin foils for examining NG/UFG structures by transmission electron microscopy were prepared by punching 3 mm disks from the specimens and twinjet electropolishing with a solution of 10% perchloric acid in acetic acid as electrolyte. The mechanical properties of the cold-rolled and reversion-annealed austenitic stainless steel specimens were determined by tensile testing samples machined to a profile of 25 mm 25 mm with a 20 mm gage length. The structure of CG austenitic stainless steel was examined after electropolishing in a solution of 39% nitric at a current density of 30 mA cm2 and 1.3 V DC for 2–3 min. The crystallographic texture was studied in the surface layer. Samples (14 mm 24 mm) were cut, ground close to the investigated layer, and electropolished. Mo Ka radiation was used in the X-ray diffraction, and complete pole figures were measured in back reflection mode. Pole figures (1 1 0), (2 0 0), (2 1 1) and (3 1 0) were used for ferritic (a0 -martensite) phase and (1 1 1), (2 0 0), (2 2 0) and (3 1 1) for the austenite. The orientated distribution functions were calculated using a series expansion method (Imax = 22) from the pole figure data and plotted in contour lines in the Euler space (Bunge’s notation) [46,48]. Surface roughness of the CG and NG/UFG austenitic stainless steel substrates was examined by atomic force microscopy (AFM Dimension 3000, Digital Instruments, Santa Barbara, CA) in tapping mode. Surface wettability was measured as contact angle of deionized water using a goniometer system equipped with imaging analysis software (Ramé-Hart, Inc., Germany). An autopipetting system was used to ensure a droplet of water with uniform volume (0.5 lL). 2.3. Fibroblast–metal surface interaction Polished NG/UFG and CG stainless steel samples were cleaned in an ultrasonic bath with ethanol, followed by washing with deionized water, then wrapped individually in gauze and sterilized in an autoclave. Cell culture studies were performed using mouse fibroblast cell line (L-cell-L929) obtained from the American Type Cell Culture Collection (Manassas, VA, USA). Eagle’s minimum essential medium (Invitrogen Corporation, USA) supplemented with 10% horse serum, penicillin (100 U ml1) and streptomycin (100 lg ml1) was used to culture fibroblasts. Cells with 80–85% confluence obtained from Tflask cultures by trypsinization were used to seed onto the disk samples. Briefly, cells were washed with phosphate-buffered saline (PBS), incubated with 0.25% trypsin/0.53 mM ethylenediaminetetraacetic acid (EDTA) for 5–7 min to detach the cells from the Petri dish, dispersed in trypsin/EDTA, transferred to a centrifuge tube and centrifuged at 322 g for 5 min. The cell pellet obtained after
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
centrifugation was re-suspended in culture medium and then diluted with culture medium to get the required cell concentration. Subsequently, the sterilized steel disks were placed in a 24-well plate and incubated with cell suspension at 37 °C in a humidified atmosphere with 5% CO2 and 95% air. 2.4. Cell initial attachment, proliferation, viability and morphology Initial cell attachment, density, morphology and viability were investigated both qualitatively and quantitatively. Fibroblast (10,000 cells cm2) were seeded on CG and NG/UFG steel surfaces and incubated for 1, 2 and 4 h at 37 °C in a CO2 incubator (5% CO2 and 95% air) to study cell attachment. Cell-seeded NG/UFG and CG steel surfaces were cultured for up to 7 days at 37 °C in a CO2 incubator. The culture medium was changed every 2 days. The initial fibroblast attachment and viability on different steel surfaces was measured using the MTT (3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide) assay at predetermined times. The assay is based on the reductase activity in mitochondria of living cells. These enzymes cleave the tetrazolium ring, which turns the pale yellow MTT into dark blue formazan, the concentration of which is directly proportional to the number of metabolically active cells. The samples were washed twice with PBS and incubated with fresh culture medium containing MTT (0.5 mg ml1 medium) at 37 °C for 4 h in dark. The unreacted dye was then removed and dimethyl sulfoxide was added to release the intracellular purple formazan product into solution. The absorbance of this solution was quantified by photospectrometry at 570 nm with a plate reader (Bio TEK Instrument, EL307C). Additionally, cell attachment was also measured using the trypan blue dye-exclusion assay. Briefly, cell-seeded disks were trypsinized and the cell pellet was collected as described above in Section 2.3. The cell pellet was suspended in culture medium, incubated with 0.4% trypan blue stain for 5 min at room temperature and the viable cells were counted by a hemocytometer. To further corroborate the MTT assay and to acquire an understanding of grain size effect on initial cell attachment, the cellseeded disks were evaluated using fluorescence microscope after staining with the nucleic acid dye, Hoechst 33342. The cell-seeded disks were washed twice with PBS, incubated (10 lg dye per ml of PBS) for 10 min at 25 °C and viewed under a fluorescence microscope (Nikon, ECLIPSE E 600 FN) with excitation and emission maxima of 346 and 442 nm, respectively; cell nuclei appeared as blue fluorescent spots. A similar procedure was carried out with 24 h cultures to determine cell proliferation. To observe the initial spreading pattern and morphology, fibroblasts were seeded on the samples, grown for 1 and 24 h, fixed with 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4, for 20 min, rinsed with PBS, dehydrated with a graded series of ethanol (25–100%) and critical point dried. The samples were sputtercoated with gold and examined in a JEOL JSM 6300 field emission scanning electron microscope. 2.5. Immunofluorescence and confocal microscopy The studies of cell–substrate interaction were further analyzed using an immunostaining technique. The expression and distribution of vinculin and the organization of actin filaments were studied via immunofluorescence and confocal microscopy. The response of cells with different substrates is expected to result in a differential signal of vinculin-associated actin stress fibers. Fibroblast-seeded CG and NG/UFG steel samples were cultured for 2 days and used for immunostaining. The cell-seeded disks were twice washed with PBS, fixed with 4% paraformaldehyde for 20 min, permeabilized with 0.1% Triton X-100 for 5 min and blocked with 0.1% bovine serum albumin (BSA) for 30 min. Subse-
3341
quently, immunocytochemistry staining was performed for fibronectin, actin and vinculin expression. Fibronectin was immunostained (labeled) with diluted mouse monoclonal antibody against fibronectin (1:800) (Sigma–Aldrich, St. Louis, MO, USA) followed by rabbit–antimouse fluorescein isothiocyanate (FITC)-conjugated secondary antibody at a working dilution of 1:200 (Sigma–Aldrich, St. Louis, MO, USA). Cell nuclei were then stained with 40 ,6-diamidino-2-phenylindole (DAPI; Chemicon International, Inc., Temecula, CA, USA) at a working dilution of 1:1000 for 5 min. The actin cytoskeleton and focal adhesion contacts were stained with a focal adhesion staining kit (Chemicon International, Inc., Temecula, CA, USA). Actin and vinculin double staining was performed by incubating cells in diluted primary antibody (anti-vinculin) in blocking solution (1:200) for 1 h and subsequently labeling them with diluted secondary antibody (1:100) (goat–antimouse FITC conjugate; Chemicon International, Inc., Temecula, CA, USA) for 45 min, with simultaneous incubation with diluted tetramethyl rhodamine isothiocyanate-conjugated phalloidin (1:400) [9]. Finally, cell nuclei were labeled with DAPI as described above. After immunostaining, samples were mounted on a glass slide using anti-fade mounting solution, covered with a coverslip and examined by confocal laser scanning microscopy (Leica DM 6000B and Leica DM 600M). 2.6. Quantification of protein and sodium dodecyl sulphate– polyacrylamide gel electrophoresis (SDS–PAGE) The cell–substrate interaction was quantitatively investigated by measuring the expression levels of proteins of fibroblast on NG/UFG and CG steel surfaces. Cell-seeded disks were used after 2 days to analyze total protein content on each surface. Cells grown on the substrates were lysed with Tris-lysis buffer (1 dilution) (20 mM Tris–HCl, pH 7.5; 150 mM NaCl, 1 mM EDTA and 1% Triton X-100) (Sigma–Aldrich, St. Louis, MO, USA) containing protease inhibitors (Sigma–Aldrich, St. Louis, MO, USA). Quantitative analysis of protein was done via a Mini Bradford protein assay (Sigma–Aldrich, St. Louis, MO, USA). This assay is based on the equilibrium between three forms of Coomassie Blue G dye. Proteins bind to the reagent under acidic conditions and cause a spectral shift from reddish/brown to blue. Spectrophotometric measurement of the resulting level of blue solution was used to determine the concentration of protein. The measurement was done at 595 nm. Standard protein concentration plots were generated using 0–20 lg ml1 BSA and were run parallel with each experiment. Subsequently, reduced proteins were separated by SDS–PAGE (using 10% sodium dodecyl sulphate–polyacrylamide gel). The SDS–PAGE molecular weight standard (SDS–PAGE marker for molecular weight, Sigma–Aldrich, St. Louis, MO, USA) was used to determine the relative sizes of the proteins from the samples. 2.7. Statistical analysis of data Values for each experiment were normalized to control experiments and expressed as the mean of at least three replicates ± SD (standard deviation). Three sets of experiments were carried out for each study. Statistical analysis was performed using a oneway analysis of variance with 95% confidence interval. 3. Results 3.1. Grain structure, mechanical and physical properties, surface roughness and wettability The average grain size of the CG austenitic stainless steel was 52 lm. The yield strength and elongation were 205 MPa and 40%,
3342
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
respectively (Fig. 1a). A TEM micrograph of the NG/UFG austenitic stainless steel (Fig. 1b) shows steel strips with 62% multi-pass cold reduction subjected to reversion annealing at 800 °C in a Gleeble thermo-simulator for 1–10 s. This reversion annealing temperature–time sequence provided an excellent combination of yield strength of 800 MPa and ductility of 45%, and resulted in nearcomplete transformation of martensite to austenite. The grain structure consisted of nanocrystalline (d < 100 nm), ultrafine (100–500 nm) and some sub-micron (500–1000 nm) grains. These optimized parameters are specific to the experiments described here and any generalization to application-oriented ap-
proach will necessitate fine-tuning of the experimental parameters. Lowering the annealing temperatures to about 750 °C allowed adequate holding time, while at higher temperatures (900–1000 °C) the grain size increased to greater than 0.5 lm. These experiments suggested that the annealing temperatures of 800 °C or lower reduced grain growth. Polished surfaces of CG and NG/UFG austenitic stainless steel exhibited almost similar surface roughness and topography. The average roughness of 3 lm 3 lm scan area for CG and NG/UFG austenitic stainless steel was 1.45 ± 0.21 and 1.52 ± 0.29 nm, respectively. In contrast, the surface wettability, measured as the
Fig. 1. (a) Light micrograph of CG and (b) transmission electron micrograph of NG/UFG austenitic stainless steels. The CG steel has an average grain size of 52 lm. NG/UFG steel is characterized by a combination of nanocrystalline grains (d < 100 nm), ultrafine grains (100–500 nm) and some grains of size 500–1000 nm.
1
1.5
1
0.5
0
0.8 0.6 0.4 0.2 0
1
2
1
4
2
4
Time (h)
Time (h) 2.0
Cell density (MTT assay)
b
Coarse-grained steel NG/UFG steel
2
a
Coarse- grained steel NG/UFG steel
Cell density (*10000 cells/cm )
Cell attachment (MTT assay) MTT
2
c
Coarse- grained steel NG/UFG steel 1.5
1.0
0.5
0.0
1
2
5
7
Time (h) Fig. 2. Histograms representing initial cell attachment after a culture time of 1–4 h by (a) MTT assay, (b) trypan blue assay and (c) metabolism of fibroblasts up to 1–7 days using MTT assay on NG/UFG and CG substrates. The fibroblasts showed significantly greater cell attachment and proliferation on the NG/UFG substrate than on the CG material. The higher density of cells on the NG/UFG substrates in comparison to the CG substrate was visible within the first hour of culture. Cell viability was significantly higher on the NG/UFG substrates compared to the CG substrates (p < 0.05).
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
water contact angle, was 75.14 ± 5.34° for CG and 66.01 ± 2.19° for NG/UFG austenitic stainless steels. Thus, NG/UFG steel is more hydrophilic than CG steel. 3.2. Cellular and molecular response to grain structure Initial cell attachment is one of the indicators of the biological properties of the substrates. Fibroblast attachment after 1 h was appreciably greater on the NG/UFG substrate than on the CG substrate (Figs. 2 and 3). Cell density was higher on NG/UFG in relation to CG at 1, 2 and 4 h of culture time (Fig. 2a and b). Cell metabolism, measured by mitochondrial reduction of MTT, was higher for the NG/UFG steel than for the CG steel, which indicates good conditions for fibroblast growth (Fig. 2c). The cell density on both the NG/UFG and CG substrates increased over time, but was significantly higher on the surface of the NG/UFG steel than on the CG steel, indicating the superior cytocompatibility of NG/UFG. A similar conclusion is derived from fluorescence micrographs of cells stained with nucleic acid-specific dye (Fig. 3). This indicates that the cell attachment to the surface is influenced by the grain structure of the steel substrates, possibly facilitating rapid adsorption of proteins onto the substrate [49]. The differences in the density as observed by fluorescence microscopy after 24 h (Fig. 3c and d) indicates that the enhanced cell count data were not skewed by cell development or surface adaptation over time. The initial cell attachment on NG/UFG substrate was higher than on the CG substrate (Fig. 3a and b). It is pertinent to mention that the cell attachment to substrates is fundamental for subsequent cellular activities, including spreading, division and differentiation.
3343
Scanning electron microscopy revealed remarkable differences in the morphology of the fibroblasts grown on the NG/UFG and CG substrates (Figs. 4 and 5), supporting the cell attachment data. The fibroblasts on the NG/UFG substrate attached and spread widely, with numerous cellular extensions, even during the initial incubation of 1 h (Fig. 4a and b). This was subsequently followed by extensive proliferation and cellular processes such that they appear to attach firmly to the NG/UFG surface, as implied by the observations at 24 h culture (Fig. 5a and b). In contrast, the fibroblast cultured on the CG substrate were spherical, but without any cellular extension at 1 h incubation time (Fig. 4c and d). After 24 h culture, cell proliferation was observed to increase on the CG substrate (Fig. 5c), but less so in comparison to the NG/UFG substrate (Fig. 5a). Similarly, the ECM formation was less significant on the CG substrate (Fig. 5d). The fibroblasts on the NG/UFG surface were characterized by starlike morphology with numerous cytoplasmic extensions (Fig. 5a and b). Frequent overlapping of cells or forming cytoplasmic bridges was observed. Thus, in summary, the extent of cell spreading was remarkably greater on the NG/UFG surface compared to the CG surface. This implies superior cellular attachment and cell–substrate interaction with the NG/UFG substrate. The production of proteins from fibroblast grown on the NG/ UFG and CG substrates were examined to further understand the cellular response. Fibronectin expression by fibroblast was greater on the NG/UFG steel than on the CG steel substrate (Fig. 6), as documented by the distinct network with stronger fluorescence intensity of immunostained fibronectin. There were fewer cells on the CG surface than on the NG/UFG surface. It should be noted, however, that the cells on the NG/UFG surface are clumped and pro-
Fig. 3. Fluorescence micrographs of fibroblasts illustrating initial cell attachment at 1 h and cell density at 24 h on NG/UFG and CG steel surfaces after staining the cell nuclei with Hoechst 33258. The cell nuclei showed blue fluorescence after staining. Cell attachment and density on the NG/UFG austenitic stainless steel (a and c) were higher than on the CG austenitic stainless steel (b and d).
3344
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
Fig. 4. Comparison of fibroblasts attachment morphology at 1 h as imaged by scanning electron microscopy on NG/UFG (a and b) and CG austenitic stainless steel (c and d). At low magnification, cells on the NG/UFG steel exhibited a higher number of attached cells and numerous cellular extensions with thicker extracellular matrices compared to the CG surface. Cells on the CG surface are spherical. High-magnification micrographs (b and d) show the greater extent of spreading on the NG/UFG surface than on the CG surface. The cells on the NG/UFG austenitic stainless steel that show numerous cellular extensions are indicative of the extensive attachment and interaction with the substrate.
duce more fibronectin. The cells on the CG surface are largely single cells, which produce less fibronectin, but where they are clumped they produce just as much fibronectin. The production of fibronectin is therefore most likely related to the clumping of cells. The expression level of vinculin, a protein that forms focal contacts and actin stress fibers, after 2 days of culture also showed a greater expression level at the edges and well-defined stress fibers on the NG/UFG steel than on the CG substrate. An example of this is presented in Fig. 7. The expression level of vinculin on CG steel substrate was lower and actin was less prominent on CG steel (Fig. 7). Thus, the examination of individual proteins that are related to cell attachment also indicated that NG/UFG surface was superior in cellular response. Fig. 8 is a confocal micrograph that combines cytoskeletal elements and nuclei and corroborates the relationship between vinculin focal contacts and actin stress fibers in an individual cell. The vinculin focal contact site interconnects with the actin leading edges and the steel substrate. A higher expression level of vinculin and actin is evident for fibroblasts cultured on the NG/UFG surface as compared to the CG surface. In summary, the confocal micrograph for the NG/UFG substrate demonstrates greater vinculin and actin signals along the periphery of cells and point to a higher degree of cellular and molecular interaction between fibroblasts and the NG/UFG steel substrate. These findings were further corroborated by the expression level of different proteins via SDS– PAGE (Fig. 9a) and by quantifying the total protein content
(Fig. 9b) released from the fibroblasts. Qualitative analysis by Western blot analysis and quantitative analysis by the Mini Bradford protein assay are in agreement with the immunofluorescence results and indicate elevated expression levels of proteins on NG/ UFG surface after 2 days of culture compared to the CG surface (Fig. 9). In both analyses, the total protein content on the surface including the extracellular fibronectin protein and the intracellular proteins vinculin and actin are measured. Thus, we can conclude that immunofluorescence and protein analysis studies are consistent and support the viewpoint that the NG/UFG steel substrate had a superior cellular response as compared to the CG steel substrate.
4. Discussion The observations concerning the biological significance of NG/ UFG and CG structures suggest the determining role of grain structure on cell attachment, morphology, and growth. It seems clear that the greater hydrophilic character and finer grain size of the NG/UFG structure of austenitic stainless steel positively influences the biofunctional property of fibroblast attachment. An NG/UFG surface has a more open lattice in the positions of high-angle boundaries as compared to a CG steel, which provides sites for attachment of cells. The higher degree of ECM on the NG/UFG substrate observed here determines events on the cell surface and communication to intracellular cellular processes by the cytoskel-
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
3345
Fig. 5. Comparison of fibroblast morphology at 24 h, as imaged by scanning electron microscopy on NG/UFG (a and b) and CG austenitic stainless steel (c and d). Fibroblasts grown on NG/UFG and CG surfaces are significantly different. Cells on the NG/UFG surface exhibit wide spreading with thick and numerous cellular extensions. These features are significantly less pronounced on the CG surface.
Fig. 6. Merged fluorescence micrographs representing immunocytochemistry of fibronectin expressed by fibroblasts after incubation for 2 days on NG/UFG (a and b) and CG stainless steel surfaces (c and d). A higher fluorescence intensity and expanded network of fibronectin expression along with a higher cell density is observed after labeling of cell nuclei with DAPI.
3346
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
eton [8,9]. We have studied the important proteins that affect cell– substrate interaction, including fibronectin, vinculin and actin. Fibronectin is involved in cell adhesion and functions as a growth stimulator [8,9,50,51], and is one of the earliest products to be released [50,52–55]. Other components of the ECM contain type 1 collagen and the glycoprotein fibronectin, both of which regulate cell attachment and differentiation [56,57]. Surface properties that affect protein expression, secretion and thus adsorption and cell attachment include hydrophilicity,
ionic bonding, electrostatic and van der Waals interactions [58,59]. In this regard, surface roughness and grain structure are important factors that affect the surface wettability. Roughness is expected to play an important role in cell attachment, proliferation and adsorption of proteins. Generally, rougher surfaces have been observed to promote cell attachment and enzyme activity [58]. From the practical viewpoint, it is believed that the distance between peaks or valleys of the surface should not exceed the ability of the cells to form focal attachment on
Fig. 7. An illustration of organization and assessment of focal contacts and actin stress fibers of fibroblasts after 2 days culture on NG/UFG (a and b) and CG (c and d) stainless steel substrates. The fibroblasts grown on the NG/UFG steel (a) show a larger number of vinculin focal adhesion contacts in relation to CG steel (c). The higher number of focal adhesion points corresponded well with thicker bundles of actin stress fibers on NG/UFG steel (b) than on CG steel (d).
Fig. 8. Confocal micrographs showing combined distribution of cytoskeletal organization and focal adhesion contacts; actin (red), vinculin (green) and nucleus (blue) in fibroblasts cultured for 2 days on NG/UFG (a) and CG (b) stainless steel surfaces. The focal contact sites are where vinculin is linked at the actin leading edges, which connect cells to the steel surface.
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
a
MW
Marker CG
10
b
NG/UFG
3347
205
Fibronectin
116 .
Vinculin
97 66
Actin
45 Lane
1
2
3
Total Protein Content (µg)
(KD)
8
6
4
2
0 CG NG/UFG Type of Stainless Steel
Fig. 9. (a) Expression of proteins from fibroblasts grown on NG/UFG and CG steel after 2 days of culture. Determination of proteins by SDS–PAGE illustrates a higher intensity of different proteins on NG/UFG substrate compared to CG substrate. Bands of proteins observed in the sample lanes are the relative sizes of fibronectin, vinculin and actin. S molecular weight standard marker was loaded in lane 1. (b) Quantitative analysis of proteins by Mini Bradford protein assay showing a higher total protein content in fibroblasts cultured on the NG/UFG substrate in relation to the CG substrate (n = 3).
two or more peaks, or else the cell will recognize the rough surface as a smooth surface [58]. With regard to the grain structure, the wettability can be distinctly different with different grain sizes because of differences in free energy, and this difference is further amplified by the structural arrangements of atoms within the grain [9]. In our case, given that the roughness and constituent phase (austenite) were similar for both the CG and NG/UFG austenitic stainless steel substrates, the hydrophilicity (contact angle: 75.14 ± 5.34° for CG and 66.01 ± 2.19° for NG/UFG) and grain size are relevant factors that contribute to the differences in cell attachment, proliferation, morphology, expression level of vinculin and expression of actin stress fibers. Hydrophilic surfaces are known to enhance activity of alkaline phosphatases and osteocalcin, which promote an osteogenic microenvironment [59,60]. Furthermore, hydrophilic surfaces promote adsorption of fibronectin and albumin [61,62]. The combination of enhanced cell attachment and proliferation and the higher expression levels of fibronectin and vinculin on the cells grown on the NG/UFG steel substrates indicates that this substrate promotes cellular activity and the processes required for strong cell– substrate interaction, and therefore cell adhesion. In summary, during the early stages, the initial attachment of cells is favored on the NG/UFG substrate. Subsequently, the adhesion of cells via ECM influences their ability to proliferate, which is also promoted on the NG/UFG substrate. Interestingly, a similar observation of cellular response was made by Faghihi et al. [9] on ultrafine titanium processed by the HPT method. The degrees of cell attachment and cell proliferation were significantly increased on the HPT-processed titanium surface. The improved cell activity was attributed to the nanostructured feature of titanium substrates, which provided a higher degree of surface wettability. It is unlikely that the strong texture of NG/UFG in comparison to the CG austenitic stainless steel is responsible for the stronger attachment of cells, because the crystallographic orientations were similar [46]. Thus, the modulation of cellular response on the surface of CG and nanostructured metallic substrates depends on grain size but not crystal structure. This study advances our understanding of cell–substrate interaction and provides strong evidence for a positive cellular response related to enhanced cell attachment and proliferation on nanostructured metal substrates. Our observations of strongly pro-
nounced cellular and molecular interaction between cells and NG/UFG material suggest the growth of connective tissues. The study underscores that nanostructured or ultrafine-grained bioimplants provide superior cell–substrate attachment and biocompatibility. 5. Conclusions and future outlook The development of NG/UFG structures for bioimplants is characterized by an enhanced cellular response combined with a high strength/weight ratio. The positive influence of the NG/UFG structure on cellular response was reflected in the improved cell attachment, proliferation, viability, morphology and spread compared to the CG structure. Furthermore, immunocytochemistry analysis through the expression levels of ECM protein, fibronectin, vinculin and associated actin cytoskeleton indicated the superior cytocompatibility of the NG/UFG substrate in relation to the CG substrate. Observations of protein analysis, such as SDS–PAGE and total protein assays, were consistent with cell density measurement and electron microscopy studies, suggesting the enhanced cell–substrate interaction and activity of the NG/UFG structure. The improved cellular response of the NG/UFG substrate is attributed to grain size and hydrophilicity, and a more open lattice in the position of high-angle boundaries. In summary, the processing–structure–biofunctional property study addresses challenges to the engineering of bulk nanostructured materials with specific physical and surface properties for medical devices with improved cellular response and strengthens the foundation of nanostructured materials for biomedical applications. Appendix A. Figures with essential colour discrimination Certain figures in this article, particularly Figures 3, 6, 7, 8 and 9, 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.2010.01.034. References [1] Borsari V, Giavaresi G, Fini M, Torrielli P, Salito A. J Biomed Mater Res B Appl Biomater 2005;75B:359–68. [2] Arenas MA, Tate TJ, Conde A, De Damboreba J. Br Corros J 2000;35:232–6.
3348
R.D.K. Misra et al. / Acta Biomaterialia 6 (2010) 3339–3348
[3] Willman G. Adv Eng Mater 1999;1:95–105. [4] Webster TJ, Ergun C, Doremus RH, Siegel RW, Bizious R. Biomaterials 2000;67:1803–10. [5] Webster TJ, Siegel RW, Bizios R. Biomaterials 1999;20:1221–7. [6] Thapa A, Webster TJ, Haberstroth KM. J Biomed Mater Res A 2003;67:1374–83. [7] Webster TJ, Smith TA. J Biomed Mater Res A 2005;74A:677–86. [8] Faghihi S, Azari F, Zhilyeav AP, Vali H, Tabrizian M. Adv Mater 2007;19:1069–73. [9] Faghihi S, Azari F, Zhilyeav AP, Vali H, Tabrizian M. Biomaterials 2007;28:3887–95. [10] Hunter A, Archer CW, Walker PS, Blunn GW. Biomaterials 1995;16:287–95. [11] van Recum AF, Shannon CE, Cannon EC, Long KJ, van Kooten TG, Meyle J. Tissue Eng 1996;2:241–53. [12] van Recum AF, van Kooten TG. J Biomater Sci Polym Ed 1995;2:181–98. [13] Chesmel K, Black J. J Biomed Mater Res 1995;29:1089–99. [14] Bagambisa FB, Joos U. Biomaterials 1990;11:50–9. [15] Sailaja GS, Ramesh P, Kumary TV, Varma HK. Acta Biomater 2006;2:651–7. [16] Hott M, Noel B, Assolant DB, Rey C, Marie PJ. J Biomed Mater Res 1997;37:508–16. [17] Meyer U, Buchter A, Wiesmann HP, Joos U, Jones DB. Eur Cells Mater J 2005;9:39–49. [18] Helen HL, El-Amin Saadig F, Scott DK, Cato L. J Biomed Mater Res 2002;64A:465–74. [19] Zhao G, Schwartz Z, Wieland M, Rupp F, Geis-Gerstorfer J, Cochran DL, et al. J Biomed Mater Res 2005;74A:49–58. [20] Goldberg VM, Jinno T. J Long Term Eff Med Implants 1999;9:11–21. [21] Buser D, Schenk R, Steinemann S, Fiorellini J, Fox C, Stich H. J Biomed Mater Res 1991;25:889–902. [22] Li D, Ferguson SJ, Beutler T, Cochran DL, Sittig C, Hirt HP, et al. J Biomed Mater Res 2002;60:325–32. [23] Wennerberg A, Wktessabi A, Albrektsson T, Joansson C, Andersson B. Int J Oral Maxillofac Implants 1997;12:486–94. [24] Klokkevold PR, Nishimura RD, Adachi M, Caputo A. Clin Oral Implants Res 1997;8:442–7. [25] Cooper LF, Masuda T, Yliheikkila PK, Felton DA. Int J Oral Maxillofac Implants 1998;13(2):163–74. [26] Anselme K. Biomaterials 2000;21:667–81. [27] Garcia AJ, Vega MD, Boettiger D. Mol Biol Cell 1999;10:785–98. [28] Gleiter H. Nanostruct Mater 1992;1:1. [29] Misra RDK, Thompson SW, Hylton TA, Boucek AJ. Metall Mater Trans A 2001;35:745. [30] Misra RDK, Tenneti KK, Weatherly GC, Tither G. Metall Mater Trans A 2004;35:341. [31] Misra RDK, Nathani H, Hartmann JE, Siciliano F. Mater Sci Eng A 2005;394: 339. [32] Shanmugam S, Misra RDK, Hartmann JE, Jansto S. Mater Sci Eng A 2006;437:436. [33] Song R, Ponge D, Raabe D, Kaspar R. Acta Mater 2005;53:845.
[34] Humphreys FJ, Prangnell PB, Bowen JR, Gholinia A, Harris C. Philos Trans R Soc Lond A 1999;357:1681. [35] Pitan C, Hashimoto T, Kawazoe M, Nagahora J, Higashi K. Mater Sci Eng A 2000;280:62. [36] Tsuji N, Saito Y, Utsunomiya H, Tanigawa S. Scripta Mater 1999;40:795. [37] Park KT, Lee CS, Shin DH. Rev Adv Mater Sci 2005;10:133. [38] Costa ALM, Reis ACC, Kestens L, Andrade MS. Mater Sci Eng A 2005;406:279. [39] Webster TJ, Siegel RW, Bizios R. Biomaterials 1999;20:1221. [40] Thapa A, Webster TJ, Haberstroh KM. J Biomed Mater Res A 2003;67:1374. [41] Zhilyeav AP, Nurislamova GV, Kim BK, Baro MD, Szpunar JA, Langdon TG. Acta Mater 2003;51:753. [42] Ivanisenko Y, Lojkowski W, Valiev RZ, Fecht HJ. Acta Mater 2003;51:5555. [43] Belyakov A, Sakai T, Miura H, Kaibyshev R. Scripta Mater 2000;42:319. [44] Lianxi H, Yuping L, Erde W, Yang Y. Mater Sci Eng A 2006;422:327. [45] Misra RDK, Nayak S, Mali S, Shah J. Metall Mater Trans A. doi:10.1007/511661009-9920-B. [46] Misra RDK, Ravikumar B, Somani MC, Karjalainen LP. Scripta Mater 2008;59:79. [47] Somani MC, Karjalainen LP, Misra RDK. In: Karjalainen LP, Hertzman S, editors. Proceedings of the 6th european stainless steel conference science and market, Helsinki; 10–13 June 2008. p. 519. [48] Somani MC, Karjalainen LP, Misra RDK, Karjalainen A. Metall Mater Trans A 2009;40:729. [49] Andrade JD, Hlady V. Ann NY Acad Sci 1987;516:158. [50] Chen LB, Murray A, Segal RA, Bushnell A, Walsh ML. Cell 1978;14:377–91. [51] Hynes RO, Bye JM. Cell 1974;3:113–20. [52] MacDonald DE, Rapuano BE, Deo N, Stranick M, Somasundaran P, Boskey AL. Biomaterials 2004;25:3135–46. [53] Meyer U, Szulczewski DH, Moller K, Heide H, Jones DB, Gross U. Cells Mater J 1993;3:129–40. [54] Daculsi G, Pilet P, Cottrel M, Guicheux G. J Biomed Mater Res 1999;47:228–33. [55] Globus RK, Doty SB, Lull JC, Holmuhamedov E, Humphries MJ, Damsky CH. J Cell Sci 1998;111:1385–93. [56] Nordahl J, Widholm SM, Hultenby K, Reinholt F. Calcif Tissue Int 1995;57:442–9. [57] Winnard R, Gerstenfeld L, Toma C, Franceschi R. J Bone Miner Res 1995;10:1969–77. [58] Lincks J, Boyan BD, Blanchard CR, Lohmann CH, Liu Y, Cochran DL, et al. Biomaterials 1998;19:2219–32. [59] Webster TJ, Ergun C, Doremus RH, Siegel RW, Bizios R. J Biomed Mater Res 2000;51:475. [60] Zhao G, Schwartz Z, Wieland M, Rupp F, Geis-Gerstorfer J, Cochran DL, et al. J Biomed Mater Res A 2005;74:49. [61] Sousa SR, Moradass-Ferreira P, Saramago B, Melo LV, Barbosa MA. Langmuir 2004;20:9745. [62] Sousa SR, Moradass-Ferreira P, Barbosa MA. J Mater Sci Mater Med 2005;16:1173.