9386
Langmuir 2007, 23, 9386-9392
Hydroxyapatite Grown on a Native Extracellular Matrix: Initial Interactions with Human Fibroblasts Emilia Pecheva,*,† Lilyana Pramatarova,† and George Altankov‡ Institute of Solid State Physics, Bulgarian Academy of Sciences, 72, Tzarigradsko Chaussee BouleVard, 1784 Sofia, Bulgaria, ICREA-Institute for Bioengineering of Catalunya, Parc Cientı´fic de Barcelona, Josep Samitier 1-5, 08028 Barcelona, Spain, and Institute of Biophysics, Bulgarian Academy of Sciences, Acad. G. BoncheV str., bl. 21, 1113 Sofia, Bulgaria ReceiVed February 14, 2007. In Final Form: June 25, 2007 Proteins are known to modulate the physical properties of minerals, and thus we anticipate that they will strongly influence the structure and the biological properties of biomimetically prepared carbonate-containing hydroxyapatite. This study was designed to learn more about the main morphological characteristics of hydroxyapatite layer grown on different substrates coated with an extracellular matrix, a biological matrix that was produced by cultured osteoblastlike cells. The hydroxyapatite growth was carried out in a simulated body fluid, a solution that resembles the human blood plasma. It was found that the extracellular matrix may serve as a template for the mineralization of biomimetic hydroxyapatite on the surface of materials like stainless steel, silicon, and silica glass, leading to the formation of a homogeneous layer. The latter was consisting of nanometer-sized hydroxyapatite crystals grouped in particles with regular sphere shape and with a significantly higher average diameter in comparison to samples without extracellular matrix coating. Subsequent in vitro studies with living fibroblasts showed that the cellular behavior depended on the type of underlying substrate used for the hydroxyapatite growth, as well as on the immersion time of the samples in the simulated body fluid. Increasing the thickness of the hydroxyapatite layer altered visibly the cellular response, and the fibroblasts developed stellate morphology on the samples with a hydroxyapatite-extracellular matrix coating. Preadsorption with fibronectin significantly improved the initial cell adhesion and spreading to all surfaces. Thus, such an approach may contribute to the development of surfaces with better tissue compatibility.
1. Introduction The mineralization of biological tissues is a well-regulated process that occurs within a matrix of organic macromolecules,1-3 for example, extracellular matrix (ECM). ECM is an organic product synthesized by cells that has diverse functions depending on the cell type and tissue origin. In addition to its conventional role in providing a scaffold for building tissues, the ECM acts as a directional highway for cellular movement and provides instructional information for regulating cell growth and terminal cell differentiation.4,5 A specific role of ECM in bony tissue is to provide an oriented template for mineral deposition and to assist the formation of nuclei by lowering the activation energy necessary for crystal nucleation.6-8 ECM macromolecules stabilize and contribute to the organization of the mineralized tissue. On the other hand, cells deposit various fibrillar and nonfibrillar macromolecules in their ECM, thus forming a composite structure consisting of matrix proteins, such as collagen, fibronectin (FN), laminin, etc., which are embedded in the pool of proteoglycans.9 Once synthesized, the adhesive proteins are * Corresponding author: tel +359 2/979 50 00 (internal 265); fax +359 2/975 36 32; e-mail
[email protected]. † Institute of Solid State Physics, Bulgarian Academy of Sciences. ‡ Parc Cientı´fic de Barcelona and Institute of Biophysics, Bulgarian Academy of Sciences. (1) Lowenstam, H.; Weiner, S. On biomineralization; Oxford University Press: Oxford, U.K., 1989. (2) Addadi, L.; Weiner, S. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 4110. (3) den Braber, E. Microtextured surfaces. Thesis, Katholieke Universiteit Nijmegen, Nijmegen, The Netherlands, 1996. (4) Watt, F. M. EMBO J. 2002, 21, 3919. (5) Danen, E. H.; Yamada, K. M. J. Cell. Physiol. 2001, 189, 1. (6) Berman, A.; Addadi, L.; Kvick, A.; Leiserowitz, L.; Nelson, M.; Weiner, S. Science 1990, 250, 664. (7) Tong, H.; Hu, J.; Ma, W.; Zhong, G.; Yao, S.; Cao, N. Biomaterials 2002, 23, 2593. (8) Mann, S. J. Chem. Soc., Dalton Trans. 1993, 1.
spatially organized by the cells at different levels of complexity, ranging from the three-dimensional fibrillar network of the connective tissues to the sophisticated structure of basement membranes.9 The minerals give the bones strength, while the ECM and particularly collagen fibers provide resiliency. An important function of the ECM proteins is to modulate the cellular interaction that is essential for hard-tissue regeneration. Therefore, the composites of biomimetic hydroxyapatite (HA) and proteins resemble biological mineral substances more closely than do purely inorganic ones, and they are therefore considered a promising material for bone repair. The formation of organicinorganic composites has been extensively studied.10-13 However, to the best of our knowledge, subsequent deposition of HA on a predeposited matrix of proteins, such as a native ECM produced by osteoblast-like cells, has not been performed. From the point of view of materials used for the HA deposition, studies are usually concentrated on one type of material (i.e., metal for example) and the effects that different types of material surfaces (metals, semiconductors, insulators) have on the HA deposition is not considered. On the other hand, proteins are known to modulate the physical properties of mineral substances, and thus it is expected that they will strongly influence the structure and the biological properties of biomimetically prepared carbonatecontaining HA. The initial cellular interaction with biomaterials might be approximated with the process of cell adhesion. The assessment (9) Ayad, S.; Boot-Handford, R. P.; Humphries, M. J.; Kadler, K. E.; Shutteleworth, C. A. The Extracellular Matrix Facts Book; Academic Press/ Harcourt Brace & Co. Publishers: New York, 1994; p 617. (10) Pellenc, D.; Berry, H.; Gallet, O. J. Colloid. Interface Sci. 2006, 298, 132. (11) Zhai, Y.; Cui, F. J. Cryst. Growth 2006, 291, 202. (12) Chang, S.; Chen, H.; Liu, J.; Wood, D.; Bentley, P.; Clarkson, B. Calcif. Tissue Int. 2006, 78, 55. (13) Yunyu, H.; Zhang, Ch.; Zhang, Sh.; Xiong, Zh.; Xu, J. J. Biomed. Mater. Res. 2003, 67A, 591.
10.1021/la700435c CCC: $37.00 © 2007 American Chemical Society Published on Web 08/07/2007
Hydroxyapatite Grown on NatiVe Extracellular Matrix
of material biocompatibility relies heavily on the analysis of macroscopic cellular responses to the given material, which are particularly well defined with the fibroblast cell model.14 Cell adhesion is the first cellular event that takes place on the biomaterials interface,15-19 and numerous in vitro experiments have shown that it depends on the physicochemical properties of the surface, such as wettability, surface chemistry, surface charge, roughness, etc.8,20-23 Therefore, surface modifications are frequently applied to tailor the initial tissue-biomaterial interaction.24-28 Cell adhesion is strongly dependent on the absorbed proteins,29 particularly on adhesive ECM proteins such as collagen, FN, and laminin, which play a fundamental role due to their ability to influence the functional behavior of adhering cells. Within these proteins, FN plays a substantial role as it is the main soluble adhesive protein in biological fluids.30 Cell adhesion and spreading are particularly pronounced in fibroblasts, which are predominant in the connective tissue and are important in mechanisms of wounding and healing.30 Our study was divided into two parts: (1) growth and characterization of biomimetic HA layer on different material surfaces, including stainless steel (SS), silicon (S), and silica glass (SG), whose surfaces were coated with a native ECM; and (2) in vitro study, comprising the biological characterization of the above surfaces by means of initial cellular interaction and overall cell morphology utilizing human fibroblasts model, as well as the particular effect of FN preabsorption. 2. Experimental Section Materials. Substrates (8 × 8 mm) were prepared from AISI 316 stainless steel (SS; foils from Goodfellow, England), n-type silicon (S; As-doped wafers from Sil Chem, Freiberg, Germany) with (100) crystal orientation, and “Herasil” silica glass (SG; plates from Heraeus Quartzglas GmbH, Hanau, Germany). As-received samples were subjected to standard polishing and cleaning procedures and were further named controls. Polishing and cleaning procedures of SS substrates included metallographic polishing with SiC papers, ultrasonic rinsing in alcohol and acetone, and drying in air. S substrates were chemically cleaned in a mixture of H2SO4 and H2O2 (piranha solution) at 100 °C, cooled down to room temperature, washed by deionized water, and dried in air. Doubly polished SG substrates were ultrasonically cleaned in alcohol and acetone, washed by deionized water, and dried in air. The chemical treatment of the S substrates before ECM deposition yielded hydrophilization of their (14) Baxter, L.; Frauchiger, V.; Textor, M.; ap Gwynn, I.; Richards, R. Eur. Cells Mater. 2002, 4, 1. (15) Wan, H.; Williams, R.; Doherty, P.; Williams, D. J. Mater. Sci.: Mater. Med. 1997, 8, 45. (16) Lee, J.; Jung, H.; Kang, I.; Lee, H. Biomaterials 1994, 15, 705. (17) Schakenraad, J.; Busscher, H.; Wildevuur, C.; Arends, J. J. Biomed. Mater. Res. 1986, 20, 773. (18) Horbett, T.; Waldburger, J.; Ratner, B.; Hoffman, A. J. Biomed. Mater. Res. 1988, 22, 383. (19) Cote, M.; Doillon, C. Biomaterials 1992, 13, 612. (20) Suggs, L.; Shire, M.; Garcia, C.; Anderson, J.; Mikos, A. J. Biomed. Mater. Res. 1999, 46, 22. (21) Jenney, A.; Anderson, J. J. Biomed. Mater. Res. 1999, 44, 206. (22) Chuang, W.; Young, T.; Yao, C.; Chiu, W. Biomaterials 1999, 20, 1479. (23) Kasemo, B.; Lausmaa, J. CRC Crit. ReV. Biocompat. 1986, 2, 335. (24) Ikada, Y. Biomaterials 1994, 15, 725. (25) Dekker, A.; Reitsma, K.; Beugeling, T.; Bantjes, A.; Feijen, J.; Van Aken, W. Biomaterials 1991, 12, 130. (26) Dewez, J.; Doren, A.; Schneider, Y.; Rouxhet, P. Biomaterials 1999, 20, 547. (27) Ricci, J.; Spivak, J.; Blumenthal, N.; Alexander, H. In The bone and biomaterial interface; Davies, J., Ed.; University of Toronto Press: Toronto, Ontario, Canada, 1991. (28) Hercules, D. Crit. ReV. Surf. Chem. 1992, 1, 243. (29) Kasemo, B.; Lausmaa, J. In The bone and biomaterial interface; Davies, J., Ed.; University of Toronto Press: Toronto, Ontario, Canada, 1991. (30) MacDonald, D.; Markovic, B.; Allen, M.; Somasundaran, P.; Boskey, A. J. Biomed. Mater. Res. 1998, 41, 120.
Langmuir, Vol. 23, No. 18, 2007 9387 surface. Thus, the three types of surfaces prepared for an ECM coating were hydrophilic. ECM Preparation. The osteoblast-like cell line SAOS-2 (DSMZ GmbH, osteogenic sarcoma cell line) was chosen for this work. It is a highly differentiated, stable, and nontransfected cell line, which behaves rather closely like normal osteoblasts. Cells were seeded at a concentration of 2 × 105 cells/mL (1 × 105 cells/cm2) in standard R-modified minimal essential medium (R-MEM, Biochromm) supplemented with 15% fetal bovine serum (FBS, Biochrom), kept in a humidified environment of 95% air and 5% CO2 at 37 °C.31 After confluency, the medium was changed with a differentiation R-MEM, now supplemented with 15% FBS, 50 µg/mL ascorbic acid, and 891 µg/mL β-glycerophosphate. After 4 days the samples were exposed to 1000 µL of 15 mM NH4OH for 6 min for selective removal of the cells from the substrates. Subsequently, the samples were washed 5 times with 1000 µL of phosphate-buffered saline. Thus, the resulting surfaces of SS, S, and SG were coated with a native collagen-containing ECM with a thickness of about 0.5 µm, determined by transmission electron microscopic (TEM) cross section of the surfaces (not given here). The matrix surface was stratified with small molecules and molecule aggregates directly attached to the surface (noncollagenous protein layer), and collagen was further embedded in the surface network, as seen by its characteristic periodicity.31 Most of the materials; surface was covered by the noncollagenous protein layer due to the small population density of the overlying collagen fibers. Preformed binding sites having cell active structure and density were thus expected. The ECM deposited on titanium and silicon substrates studied by Pham et al.32 revealed no significant effect of the underlying substrate on the ECM composition and deposition process. Investigation with energydispersive X-ray spectrometry (EDX) and Auger electron spectroscopy (AES) revealed that only N, O, and C were found after ECM deposition.32 Immersion in Simulated Body Fluid. To evaluate the ability of the ECM-coated samples (hereafter designated as ECM/SS, ECM/ S, and ECM/SG) to induce HA growth, prolonged immersion in 500 mL of simulated body fluid (SBF) for 4 days was carried out. The SBF was prepared on the basis of the SBF of Kokubo et al.33 but with 1.5 times higher Ca and P concentrations to ensure faster layer precipitation. Reagent-grade chemicals (NaCl, NaHCO3, KCl, K2HPO4‚3H2O, MgCl2‚6H2O, Na2SO4‚10H2O, and CaCl2‚2H2O) were dissolved in distilled water and buffered at pH 7.4 with tris(hydroxymethyl)aminomethane or hydrochloric acid. The ion concentrations were maintained by refreshment of the solution every 24 h up to the end of the immersion procedure. These samples were named as HA/ECM/SS, HA/ECM/S, and HA/ECM/SG. Another group of samples was subjected to a kinetic study by immersion of control samples in the SBF for shorter times, 4 or 24 h, in order to evaluate the effect of the HA thickness on the cell behavior. Refreshment of the solution was not applied for this sample group. Cell Adhesion and Spreading. Human dermal fibroblasts were suspended in a serum-free medium and left to adhere for 2 h at 37 °C under an atmosphere of 95% air and 5% CO2 onto four groups of samples: (1) controls (SS, S, and SG); (2) ECM-coated controls (ECM/SS, ECM/S, and ECM/SG); (3) samples from the second group, covered with a rough HA layer after 4 days of immersion in the SBF (HA/ECM/SS, HA/ECM/S, and HA/ECM/SG); and (4) control samples coated with HA after 4 or 24 h of immersion in the SBF. Samples from the four groups were placed in 12-well tissue culture plates and studied plain or after being coated with FN (20 µm/mL for 30 min). The overall cell morphology and spreading of fibroblasts were visualized via fluorescein diacetate in living cells and observed by a fluorescent laser scanning confocal microscope (LSCM; LSM 510, Zeiss, Germany) at a magnification of 10×. (31) Pham, M. T.; Maitz, M. F.; Reuther, H.; Muecklich, A.; Prokert, F.; Steiner, G. J. Biomed. Mater. Res. A 2004, 71, 16. (32) Pham, M. T.; Reuther, H.; Maitz, M. F. J. Biomed. Mater. Res. A 2003, 66, 310. (33) Kokubo, T.; Kushitani, H.; Sakka, S.; Kitsugi, T.; Yamamuro, T. J. Biomed. Mater. Res. 1990, 24, 721.
9388 Langmuir, Vol. 23, No. 18, 2007
PecheVa et al.
Figure 1. SEM image of the HA layer grown on (a) ECM-coated stainless steel samples and (b) control stainless steel sample; (c) image showing the areas for taking the EDX spectra; (d) EDX spectra from the HA grown on silicon (left) and on stainless steel (right) samples. Table 1. Particle Diameter, Grain Size, Elemental Concentrations, and Ca:P Ratio of HA Layers Grown on Different Materials samplea HA/ECM/SS HA/ECM/S HA/ECM/SG HA/SS, 4 h HA/S, 4 h HA/SG, 4 h HA/SS, 24 h HA/S, 24 h HA/SG, 24 h
HA particle diameterb (µm) HA grain sizec (nm) Ca (at. %) P (at. %) Na (at. %) Mg (at. %) Cl (at. %) K (at. %) 5.0 4.8 4.4
boned
46 ( 10 36 ( 7 23 ( 5
20-40
13.2 11.2 10.1 10.9 3.3 3.3 13.7 5.6 4.5
10.0 8.1 7.2 6.9 2.2 2.4 8.6 3.6 3.0
35.5
17.1
1.8 3.0 0.9
1.3 3.7 0.9
0.1
0.1 0.1
0.8 0.5 0.5 0.4 0.1 0.1 0.8 0.2 0.2
0.1-9.0
0.3-1.0
0.1-9.0
0.1-9.0
0.1 0.1
Ca:P 1.32 1.38 1.40 1.58 1.50 1.38 1.59 1.56 1.50
