Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Jan 1.
Published in final edited form as: J Tissue Eng Regen Med. 2009 Jan;3(1):26–36. doi: 10.1002/term.131

Effect of functional end groups of silane self assembled monolayer surfaces on apatite formation, fibronectin adsorption and osteoblast cell function

GK Toworfe a, S Bhattacharyya a, RJ Composto a,b, CS Adams a,c, IM Shapiro a,c, P Ducheyne a,d,*
PMCID: PMC2610238  NIHMSID: NIHMS74043  PMID: 19012271

Abstract

Bioactive glass (BG) can directly bond to living bone without fibrous tissue encapsulation. Key mechanistic steps of BG’s activity are attributed to calcium phosphate formation, surface hydroxylation and fibronectin (FN) adsorption. In the present study, self-assembled monolayers (SAMs) of alkanesilanes with different surface chemistry (OH, NH2, and COOH) were used as a model system to mimic BG’s surface activity. Calcium phosphate (Ca-P) was formed on SAMs by immersion in a solution which simulates the electrolyte content of physiological fluids. FN adsorption kinetics and monolayer coverage was determined on SAMs with or without Ca-P coating. The surface roughness was also examined on these substrates before and after FN adsorption. The effects of FN-adsorbed, Ca-P coated SAMs on the function of MC3T3-E1 were evaluated by cell growth, expression of alkaline phosphatase activity, and actin cytoskeleton formation. We demonstrate that, although the FN monolayer coverage and the rms roughness are similar on −OH and −COOH terminated SAMs with or without Ca-P coating, higher levels of ALP activity, more actin cytoskeleton formation and more cell growth are obtained on −OH and −COOH terminated SAMs with Ca-P coating. In addition, although the FN monolayer coverage is higher on Ca-P coated −NH2 terminated SAMs and SiOx surfaces, higher levels of ALP activity and more cell growth are obtained on Ca-P coated −OH and −COOH terminated SAMs. Thus with same Ca-P coatings, different surface functional groups have different effects on the function of osteoblastic cells. These findings represent new insights into the mechanism of bioactivity of BG and, thereby, may lead to designing superior constructs for bone grafting.

Keywords: self assembled monolayers, calcium phosphate, protein adsorption, cell attachment, proliferation, alkaline phosphatase activity

1. Introduction

The effect of calcium phosphate (Ca-P) surfaces of bioactive ceramics and glasses on the adhesion, proliferation, differentiation, and extracellular matrix formation of cells of the osteoblast lineage has been reported (El-Ghannam et al., 1999; Knabe et al., 2004; Toworfe et al., 2006). The excellent characteristics of Ca-P coated implants include enhanced bone bonding; this is, superiority on both degree and rate of fixation in bone and the presence of more supporting bone on these implant surfaces compared to uncoated ones. This contributes to implant success and rapid bone growth.

Upon implantation, calcium phosphate surfaces are either present or develop on bone biocompatible materials (Ducheyne and Cuckler, 1992; Ducheyne et al., 1990; Li et al., 1996). Immersion experiments, in vitro, can simulate the formation of this calcium phosphate surface. Bioactive glass and ceramics coated with calcium phosphate are reported to maximally influence the rate of bone bonding (Hench, 1991), and this property is very useful in the context of artificial grafts. The proposed sequence of reactions leading to bone bonding of bioactive glass includes the leaching of network modifying ions, concurrent adsorption Ca-P and proteins and gradual maturation of surfaces through a complex set of reactions which are in part cell-mediated. Extracellular protein molecules such as fibronectin and growth factors are adsorbed and trigger cell proliferation and differentiation (Ducheyne and Qiu, 1999).

Previous studies suggest that surface hydroxylation stimulates the calcification of material surface. Hydroxyl groups are found in abundance on BG surfaces. Surface hydroxylation also contributes to the excellent biocompatibility of titanium implants (Kasemo, 1983; Hansson et al., 1983; Albrektsson, 1983; Albrektsson and Hansson, 1986). Since titanium is very reactive, titanium implant surfaces are always covered by a very thin, but persistent oxide film with a thickness of a few nanometers, which is slightly hydroxylated. Calcium, phosphorus and sulfur bind to this hydrated oxide layer (Hanawa and Ota, 1992), leading to the formation of a calcium phosphate layer on the hydrated surface. Other examples of surface hydroxylation stimulating calcification include poly (ethylene oxide-co-butylene terephthalate) (Li et. al, 1997), tyrosine derived polycarbonates (Levene et al., 1991), and sol gel processed silica and titania (Li and de Groot, 1993; Li et al., 1995). All these calcium and phosphorus free materials elicit hydroxyapatite formation on their surface solely by extracting and concentrating the calcium and phosphate ions available in the interstitial fluids, and it has been suggested that this phenomenon is associated with surface hydroxyl groups (Iwasaki et al., 1993; Li and Ducheyne, 1998, Zhu et al., 2004). Moreover Zhu et al. (Zhu et al., 2004) have shown that hydroxyl terminated surface are more favorable for calcification compared to amine terminated surface in stable precursor solution (i.e when pH ≤ 7.4).

Fibronectin (FN) is one of the major components of ECM and it is also secreted by osteoblasts. When adsorbed onto biomaterials, FN plays a crucial role in bone cell attachment, proliferation and differentiation. Therefore, FN adsorption on a calcium phosphate layer is of interest in understanding BG’s bioactivity. It has been documented by Garcia et al. (Garcia et al., 1998) that the formation of a calcium phosphate surface layer with characteristics similar to those of the calcium phosphate mineral phase in bone resulted in enhanced fibronectin-mediated cell adhesion in comparison to control materials and an unreactive form of hydroxyapatite. This effect was not due to differences in adsorbed fibronectin concentration, but arose from differences in fibronectin conformation. Moreover, El-Ghannam et al. (El-Ghannam et al., 1995) documented that the expression of the osteoblast phenotype critically depends on the preparation of the glass surface. Prior to cell seeding, the glass surface was first transformed into a calcium phosphate surface onto which, in a subsequent step, biological molecules were adsorbed from a serum containing tissue culture medium. Western blot analysis revealed that in this second step there was extensive, preferential adsorption of fibronectin (El-Ghannam et al., 1999). Their results showed that neonatal rat calvaria osteoblasts, seeded on porous bioactive glass disks which were conditioned with this two-step treatment, maintained critical features of the osteoblast phenotype such as evidence of high alkaline phosphatase activity (El-Ghannam et al., 1994). In contrast, osteoblasts seeded on bioactive glass templates conditioned with only one of the steps (immersion in either simulated physiological saline or tissue culture medium) alone expressed a low alkaline phosphatase activity.

All these experimental findings appear to imply that the combined effect of surface hydroxylation, calcium phosphate (Ca-P) formation and FN adsorption is important for fully understanding the bioactivity of BG. Due to the complexity of circumstances in which proteins, cells and BG surface come in contact, a model system that mimics bioactive glass is needed. Self-assembled monolayers (SAMs) provide us a well-defined surface chemistry and have been applied to model biomaterial surfaces (Toworfe et al., 2006; Keselowsky et al., 2005, Liu et al., 2006). In a previous study (Toworfe et al., 2006), we have reported on Ca-P formation on SAMs. Three terminal groups of alkanesilane SAMs, including hydroxyl (−OH), carboxyl (−COOH) or amine-terminated (−NH2) groups, were used to induce and nucleate Ca-P surfaces on substrates, since these surfaces then possess varying capabilities for the heterogeneous nucleation and growth of hydroxyapatite (Toworfe et al., 2006). In the present study, we have investigated the behavior of FN adsorption on these Ca-P coated SAM surfaces in comparison with SAMs. Cellular response, such as the early events of osteoblast phenotypic expression, viz. alkaline phosphatase activity, cell growth, and the actin cytoskeleton formation, to these SAMs surfaces with or without Ca-P coating has also been tested. Here, we demonstrate that, although the FN monolayer coverage and the rms roughness are similar on −OH and −COOH terminated SAMs with or without Ca-P coating, higher levels of ALP activity, more actin cytoskeleton formation, and more cell growth are obtained on −OH and −COOH terminated SAMs with Ca-P coating. Obviously, with a Ca-P coating on the biomaterial surface, the function of osteoblastic cells is enhanced. In addition, although the FN monolayer coverage is higher on Ca-P coated −NH2 terminated SAMs and SiOx surfaces, higher levels of ALP activity and more cell growth are obtained on Ca-P coated −OH and −COOH terminated SAMs. With the same Ca-P coating, different surface functional groups have different effects on the function of osteoblastic cells. Surface hydroxyl groups, such as −OH and −COOH, certainly play an important role in bioactivity. These findings enable us to develop new insights into the mechanism of bioactivity of BG and design superior constructs for bone grafting.

2. Materials and Methods

2.1. Preparation of silicon substrates

Silicon wafers (Silicon Quest Int., <1-0-0>, 20–30 Ω-cm), polished on one side, were cut into approximately 15 × 15 mm coupons and cleaned in a highly acidic piranha solution (H2O2:H2SO4) over low heat for 30 min. The substrates were thoroughly rinsed with deionized water (dH2O) and then blown dry with a stream of nitrogen (N2) gas. This process produced a very clean hydrophilic substrate which was then oxidized by exposure to ultraviolet (UV) in low/radio frequency discharge chamber for 10 min to generate an ultrathin surface oxide coating. The silicon oxide surfaces were then functionalized by grafting the SAMs thereon for use as the template for inducing apatite mineralization.

2.2. Functionalization of silicon substrates

3-aminopropyltriethoxysilane (APTES), 3-triethoxysilylpropyl succinic anhydride (TESPSA) and 3-glycidoxypropyltrimethoxysilane (GPTMS) organosilanes (Table 1a) were grafted onto the silicon oxide surfaces. Silane deposition on the silicon substrates was performed under argon in an anhydrous anaerobic glove box (MBraun, Stratham NH: UniLab) environment, by immersing the samples into a 5 – 20% (v:v) solution of the silane solution (Gelest Inc., Morrisville PA) in an organic solvent (Toluene: 99.8% anhydrous, Aldrich) for 16 h. A complete study of the preparation and characterization of these SAMs has been presented (Lee et. al, 2005). The concentrations of the organic solvent and the silanes were optimized for each SAM type, i.e. APTES: 5% toluene, TESPSA: 10% toluene and GPTMS: 20% toluene. After immersion for 16 h, the samples were sonicated serially, in three different solvents (for 20 min each) in order to enhance the organization of the silane monolayers on the surfaces and to remove physisorbed molecules. Toluene, N,N-dimethlyformamide (DMF, Aldrich: 99.9% HPLC grade) and ultrapure water were used for the sonication, in that order. In all cases, the root mean square (rms) roughness of the grafted silane layers was less than 0.3 nm over 1 µm × 1 µm area, as demonstrated in a previous study (Lee et. al, 2005).

Table 1.

(a) The table shows the terminal groups of the organosilanes used in the self-assembled monolayer technique; (b) Ion concentration of the components of the simulated physiological fluid used in the biomimetic synthesis. They are compared to the concentration of ions in human blood plasma [20]

(a)
Label Name Formula Functional group
SiOx Silicon Oxide (control)
APTES 3-Aminopropyltriethoxy silane (C2H5O)3Si.(CH2)3.NH2 NH2
TESPSA 3-Triethoxysilylpropyl succininc anhydride (C2H5O)Si.(CH2)3C4H4O3 −COOH
GPTMS 3-Glycidoxypropyltrimethoxy silane (CH3O)3Si.(CH2)3.OCH2CHOCH2 Epoxide → −OH
(b)
Ion Ca2+ HPO42− Na+ Cl− K+ Mg2+ HCO3− SO42−
Human Blood Plasma 2.5 1.0 142.0 103.0 5.0 1.5 27.0 0.5
Simulated Body Fluid 2.5 2.5 152.0 136.6 5.0 1.5 27.0 0.5
Open in a new tab

2.3. Calcium phosphate surface layer formation

Calcium phosphate precipitates were formed by immersing SiOx and SAMs surfaces in a supersaturated solution, which simulates the electrolyte content of physiological fluids. Specimens were immersed for 1 day and 7 days in a tris buffer solution (pH 7.4 at 37°C), complemented with various concentrations of electrolytes (Table 1b) comparable to the human blood plasma (Li et al., 1995). The phosphate concentration of 2.5 mM was higher, though, than the equilibrium saturation concentration in aqueous solution under physiological conditions (Kokubo et al., 1990; Radin and Ducheyne, 1996). The experimental samples were immersed in 20 mL of solution for 1 and 7 days duration, at 37 °C, with continuous shaking.

2.4. Fibronectin adsorption

Protein adsorption studies were done to investigate the effect of Ca-P coating on protein coverage. Fibronectin (FN) from bovine plasma (Sigma-Aldrich, St. Louis, MO), lyophilized from 0.05M Tris buffered saline, pH 7.5, was fluorescent labeled using Oregon Green (OG) 488 dye (Molecular Probes, Eugene, OR). The protein was first equilibrated to room temperature and then reconstituted with sterile distilled water for 30 min. The protein labeling protocol was followed as described elsewhere (Toworfe et al, 2004) to achieve a stable protein-dye conjugate that is stable at 4 °C for several months without loss of biological activity.

Spectrophotometer readings were made to quantitate the labeled protein by using a dilution factor of 1, correction factor 0.12 (obtained from the product information) and molar extinction (absorption) coefficient of FN, 12.8 cm−1mg−1 (Kishore et al., 1997; Gao et al., 2002) as parameters. Increasing concentrations of fibronectin, 0, 0.5, 2.5 and 10 µg/mL, prepared from a stock solution of the labeled protein, were adsorbed on the Ca-P coated surfaces in 12 well tissue culture plates (Corning Inc., Costar NY), for 1 h at 37 °C. 1 h incubation time was selected since the adsorption isotherms show nearly complete saturation after 45 min incubation, reaching complete saturation after 1 h and remains as such after longer incubations times up to 22 h (Toworfe et al., 2004). To dissociate loosely bound protein and to prevent rebinding, substrates were washed three times with phosphate buffered saline (PBS) prior to fluorescent reading in a micro-plate reader (Spectrafluor Plus, TECAN) at 485 nm excitation and 535 nm emission wavelengths. Near saturation is shown by the development of semi plateau adsorption isotherms after 1 h incubation in the FN solutions. The standard curve created (straight line with R2 = 0.9999) from the fluorescent data of relative fluorescent intensity (RFI) vs. FN bulk concentrations was based on using sufficiently diluted solutions to ensure linearity of the fluorescence to the bulk concentration (Model and Healy, 2000). This standard curve correlated the RFI units to the quantity of FN adsorbed and enabled conversion of RFI units to FN surface densities. Prior to cell assays, 10 µg/mL of unlabelled FN was pre-coated on all surfaces.

2.5. Cell culture

A murine osteoblast cell line (MC3T3-E1) was used in this study. This cell line expresses several osteblastic markers, in vitro, and attaches to a variety of ECM proteins including vitronectin, osteopontin, and collagen (Garcia et al., 1998). The cells were maintained in complete medium, Dulbecco’s modification Eagle’s medium, DMEM (Gibco, BRL, Grand Island, NY) supplemented with 10% fetal bovine serum, FBS (Gibco), 2mM L-glutamine, 100 U/mL penicillin (Gibco) and 50 µg/mL streptomycin (Gibco). Cells were incubated in a humidified 5 % CO2 balanced-air atmosphere at 37 °C. The cells were plated in tissue culture dishes; the media was changed every other day and passaged once every week after reaching confluence.

2.6. Attachment and Proliferation Assay

Cells were seeded onto the engineered substrates (functionalized SAM surfaces coated with Ca-P; precoated with FN) at a density of 104 cells/cm2, in 12 well plates for 1 h, 24 h and 7 days. After each incubation period, the cells were released by trypsinization (Trypsin, Gibco), centrifuged at 18,000 g for 5 min after aspirating the culture medium. The cells were rinsed with phosphate buffered saline, PBS (Sigma-Aldrich, St. Louis, MO), soaked in 1% Triton X-100 at 4°C for 30 min and counted with the aid of hemacytometer. The percentage of cells was calculated as the number of cells remained on substrates divided by the number of seeded cells. Two runs of the assay were performed and in each experiment, all the substrates were presented in triplicate.

2.7. Alkaline phosphatase activity

Alkaline phosphate (ALP) activity was determined by means of the enzymatic conversion of 4 (para)-nitrophenylphosphate (pNPP) to 4-nitrophenol (pNP) and measuring the absorbance of pNP at 430 nm. ALP was evaluated by plating the MC3T3-E1 cells on the different engineered substrates in a 12 well plate at a density of 105 cells/well. The cells were cultured in DMEM and fed every other day with the medium supplemented with 50µL/mL of ascorbic acid and 5 mM β-glycerophosphate. After 7 days, cell layers were trypsinized in 60 µL trypsin (10×), added to 5 mL DMEM and centrifuged for 2 min at 10,000 g at room temperature. The cells were resuspended in 1 mL isotonic NaCl. The cell suspension (in isotonic NaCl) was centrifuged at 10,000 g for 2 min at room temperature. The supernatant was discarded and the cells resuspended in 0.9 % NaCl in 0.2 % Triton X-100 and centrifuged for 10,000 g at room temperature. 100 µL of the supernatant was mixed with 100 µL of pNPP, in 96-well plate. The plate was incubated at 37 °C for 30 min and the reaction was stopped by adding 0.3M NaOH. Absorbance was measured before and after 30 min at 430 nm and ALP activity in the samples was calculated. The results were expressed in micromoles of p-nitrophenol per min per 105 cells. Serial dilutions of 4-nitrophenol was made for the standard curve. Samples and standards were assayed in triplicate.

2.8. Cytoskeletal analysis (Actin staining)

Analysis of the cytoskeleton organization was achieved by selective labeling of F-actin of the MC3T3-E1 cells seeded on the engineered substrates. Cells were plated on the substrates in DMEM supplemented with 10% FBS and 1% penicillin/streptomycin, at 37 °C in 5 % CO2 incubator, for 36 h. Substrates were washed with PBS, fixed with 1.4 % paraformaldehyde (in PBS) for 10 min at room temperature, permeabilized in 0.1 % buffered Triton X-100 solution for 15 min and then washed twice with PBS. Substrates were stained for 2 h in 20 µM of rhodamine-conjugated phalloidin 548 (Molecular Probes, Eugene, OG) at 4 °C. Phalloidin conjugates were used because the derivatives selectively stained F-actin and thus provided convenient labels for identifying actin in fixed cell preparations (Wang et al., 1982; Huang et al., 1992). Samples were washed, air-dried, mounted in an aqueous medium with anti-fading agents (Biomeda, Foster City, CA) and examined using an Olympus Fluoview IX70 confocal laser-scanning microscope with a 20× objective and a 10× optical zoom. The microscope was used to scan extent of cell spreading and cytoskeleton formation on the substrates. After each cell is digitized, the average intensity per pixel per cell is determined using Fluoview software to determine cytoskeletal formation during attachment and spreading.

2.9. Statistics

The statistical significance of differences among the experimental parameters were analysed using one way and two-way ANOVA, where appropriate, with Scheffe’s test at a level of 0.05. The differences among the cell incubation times were also statistically analyzed at the same level.

3. Results

3.1. Calcium phosphate surface layer

We have previously determined the structure and composition of calcium phosphate surface layers by means of their surface wettability, nanoscale thickness using ellipsometry, XPS analysis, FTIR and SEM (El-Ghannam et al., 2004; Lee et al., 2005; Toworfe et al., 2006). The nanoscale thickness of the grafted SAMs on the silicon oxide surfaces were approximately 1 nm, with their root mean square roughness values comparable to that of the silicon oxide substrate. The water contact angle on the SAMs was ranked as: θAPTES > θTESPSA > θGPTMS-OH > θSiO2, in order of increasing wettability. XPS analysis of the SAMs surfaces confirmed the presence of characteristic surface functionalized groups, demonstrating the stability of the model substrates under conditions relevant for calcium hydroxyapatite coating and biomolecular studies. The Ca-P coatings on the SAMs showed an increased roughness and became less homogeneous with time. At all time points, the thickness was greatest on the −OH terminated surface. FTIR spectra (Fig. 1a) showed well-defined phosphate (P-O) and carbonate (C-O) peaks characteristic of carbonated hydroxyapatite, with stronger P-O bands on the −OH-terminated surface in comparison to the SiOx (control), −NH2- and −COOH-terminated surfaces. SEM images and energy dispersive X-ray (EDX) analyses of Ca-P precipitates on −OH terminated Ca-P coated surface (such as shown in Fig. 1b after immersing for 7 days) showed micro-sized precipitates in a two layer coverage. Rutherford backscattering was used to determine the depth distribution of Ca, P and O on the Ca-P coated surfaces (Fig. 1c). At the same time, it revealed the relative roughness, continuity and stoichiometry of the Ca-P layer.

Figure 1.

Figure 1

Open in a new tab

Figure 1a: FTIR spectra (% Transmittance) of a representative SAM (i.e., −OH terminated) surface with Ca-P coating after 7 days immersion in the physiological fluid. The figure shows the prominence of P-O and C-O absorption bands characteristic of a carbonated hydroxyapatite coated surface.

Figure 1b: Scanning electron micrograph of characteristic Ca-P coated SAM substrates. Inserts are EDX patterns showing the intensities of the elements (Ca, P and O) in Ca-P coating.

Figure 1c: Rutherford backscattering spectrum on Ca-P coated SAM surfaces. Spectrum was obtained at 3 MeV at normal incidence. The arrows indicate the surface channels and depths of Ca, P and O of the Ca-P coating. The Si signal detected, is attributed to the SiOx surface either under the Ca-P, Si1 or exposed SiOx.

3.2. AFM on fibronectin adsorbed surfaces

AFM was used to image the surface topography. In our previous study (Toworfe et al., 2006), AFM revealed that Ca-P morphology varied with SAM type, reflecting the different Ca-P thickness resulting from the different nucleation capabilities of the SAMs. In this study, the morphology of adsorbed FN on both SAM with and without Ca-P coating was revealed by AFM. Fig. 2 shows representative AFM scans (10 µm × 10 µm) of adsorbed FN on both an −OH SAM surface (Fig. 2a) and an −OH surface with a Ca-P coating (7 day coating duration) (Fig. 2b). Apparently, the surface roughness is much higher on the Ca-P coated SAM surfaces. The rms roughness on these Ca-P coated surfaces did not change significantly after FN adsorption (124.6 nm rms). This is because the FN molecules tend to align themselves in the pores and grooves of the porous Ca-P surfaces.

Figure 2.

Figure 2

Open in a new tab

Atomic force microscopy images showing representative surface profiles (Z scale) of protein coverage on −OH terminated SAM (a) without Ca-P coating; (b) 7 days Ca-P coated substrate.

3.3. Surface density of FN on Ca-P coated SAMs

Figure 3a shows the surface density (Γ) of FN on Ca-P coated SAMs, as a function of the FN concentration in solution. As it can be seen in this figure, the coverage of FN on these Ca-P coated SAMs is approaching to the saturation value when increasing FN concentration. Figure 3b shows the Γ plots after incubating the SAMs and Ca-P coated SAMs substrates in 10 µg/mL of FN for 1 h. The surface density of FN ranged between 299 – 344 ng/cm2 on the SAM surfaces, while the surface density was slightly higher (344 – 357 ng/cm2) on the Ca-P coated SAM surfaces.

Figure 3.

Figure 3

Open in a new tab

(a) Figure shows the surface density of fibronectin, on the Ca-P coated substrates. Increasing protein concentrations of 0, 0.5, 2.5 and 10 µg/mL were adsorbed onto the surfaces for 1 h under physiological conditions; (b) the figure shows the surface density of FN on the substrates (including the protein coverage on substrates without Ca-P coating).

3.4. Cell attachment

The attachment and proliferation of MC3T3-E1 cells maintained in DMEM culture medium was determined by plating 104 cells/cm2 on the engineered substrates for 1 h and 24 h, under physiological conditions. Fig. 4a shows cell number on 1-day Ca-P coated surfaces (partially coved by Ca-P) after 1 h and 24 h incubation. The % cells were evaluated in comparison to the number of cell/cm2 initially seeded on the substrates. On all the surfaces, about 30 – 40 % of cells remained attached after 1h and 24 h incubation. There is no significant difference between each type of surfaces.

Figure 4.

Figure 4

Open in a new tab

Ca-P coated SiOx and SAM surfaces preadsorbed with FN, were incubated in cell culture for 1 h (initial) and 24 h. This figure shows % of cellular attachment on the (a) 1 day Ca-P coated substrates; (b) 7 days Ca-P coated substrates.

Considering cell number on 7-day Ca-P coated surfaces (fully covered by Ca-P), a much higher cell number was achieved on the −OH (50 %) and the −COOH surfaces (45 %) than that on the SiOx and −NH2 surfaces (about 30 %) after one hour of incubation (Figure 4b). After 24 hour of incubation, even a higher number was achieved on the −COOH and −OH surfaces (70 %).

In all cases, more cells remained on the −COOH and −OH surfaces than those on the SiOx and −NH2 surfaces. The cell number was increased with incubation time. In addition, more cells remained on 7-day Ca-P coated surfaces than those on 1-day Ca-P coated surfaces (Fig. 4a and 4b),

3.5. Cell proliferation

The proliferation of osteoblast cells cultured for 7 days was evaluated on the three different FN preadsorbed SAM surfaces: (i) without apatite coating, (ii) 1 day Ca-P coated and (iii) 7 days Ca-P coated (Fig. 5). The cells barely proliferated on the surfaces without apatite coating as depicted. On the 1-day Ca-P coated surfaces, however, proliferation was least on the SiOx followed by a slight increase on the −NH2 surface. Cells proliferated more on the −COOH and −OH surfaces, especially on the −COOH surface. On the 7-day Ca-P coated substrates, cell proliferation on the SiOx surface was comparatively low and similar in magnitude to the 1 day Ca-P coated surface. However, cell proliferation significantly increased by 4 to 6 folds on the −OH and −COOH surfaces respectively.

Figure 5.

Figure 5

Open in a new tab

Proliferation of MC3T3-E1 cells were evaluated on 1 day and 7 days Ca-P coated SiOx and SAM surfaces preadsorbed with FN. The cells were trypsinized and counted using hemacytometer after they were allowed to proliferate for 7 days in culture.

3.6. Alkaline phosphatase activity

The level of alkaline phosphatase activity (ALP) on the three different FN preadsorbed SAM surfaces: (i) without apatite coating, (ii) 1 day Ca-P coated and (iii) 7 days Ca-P coated is shown in Fig. 6a–c. Minimum ALP activity of ~ 25 U (where U is µmol p-NP/105 cell/min) in magnitude, was exhibited by the non-apatite coated SAMs and SiOx (control) surfaces that were incubated in cell culture for 7 days (Fig. 6a). However, ALP activity on the 1-day Ca-P coated samples were, significantly higher than that on the non-coated substrates (Fig. 6b). Among those substrates, the −COOH surface exhibited the highest activity of ~ 125 U, which was approximately 5 folds increase, while the SiOx, control surface exhibited the least (~ 48 U). On the 7-day Ca-P coated surfaces (Fig. 6c), ALP activity was most enhanced on the −OH surface, while least activity was evaluated on the −NH2 surface. Up to 4 times higher ALP activity was measured on the −OH compared to the −NH2 surface.

Figure 6.

Figure 6

Open in a new tab

Figure shows alkaline phosphatase (ALP) expression by MC3T3-E1 cells after 7 days in culture. ALP expression was normalized for cell number. (a) shows ALP activity evaluated on SiOx and SAMs without Ca-P coating; while (b) indicates ALP activity on SiOx and SAMs coated with Ca-P precipitates (1 day); and (c) ALP activity on substrates coated with Ca-P (7 days). All the substrates were precoated with FN.

3.7. Formation of actin cytoskeleton

Visual examination of confocal microscopy images shows that MC3T3-E1 cells seeded on the FN preadsorbed Ca-P coated substrates exhibited comparatively higher fluorescence intensity than on the non-coated ones (Fig.7). An increase in cytoskeleton organization of cells after 36 h seeding on 7 day Ca-P coated −OH surface (Fig. 7b, right) compared to −OH surface without Ca-P (Fig. 7a, left) was observed. The fluorescence intensity of cells seeded on the biomineralized surface show full actin formation and development after 36 h of cell attachment. Fully spread cells displayed a robust cytoskeleton organization and actin stress fibers. Results show that cytoskeleton organization of MC3T3-E1 cells is enhanced on Ca-P coated surfaces.

Figure 7.

Figure 7

Open in a new tab

Confocal microscopy images show the comparison, qualitative, between spreading of MC3T3-E1 cells on (a) non-coated and (b) 7 days Ca-P coated −OH terminated substrates after 36 h incubation in cell medium. Spreading of the cells on Ca-P coated surface depicts enhanced actin filament formation and cytoskeleton organization, while cell spreading was inhibited on the non-coated Ca-P surface.

4. Discussion

The results presented in this study consistently demonstrate that the activity and function of osteoblast-like cells and biomolecules are extensively influenced by Ca-P coatings. It is known that osteoblastic cell differentiation may be dependent on the type of osteoblast cell line used. In this study, we used the murine osteoblast cell line (MC3T3-E1) which expresses several osteoblast markers and attaches to a variety of ECM proteins including vitronectin, osteopontin and collagen. The Ca-P coated surfaces exhibited different morphologies that gave rise to the different characteristics and thicknesses of the biomineralized surface. Studying the effectiveness of these morphologies for FN adsorption in comparison to the same SAMs and SiOx surfaces without Ca-P (control surfaces) showed that the morphology and the roughness were not significantly different after FN adsorption on the surfaces. This can be ascribed to the monolayer coverage of FN on the SAMs having a nanoscale dimension that does not alter the topography of the Ca-P coated surfaces. The average rms roughness after protein adsorption on the SAMs surfaces without Ca-P coating is, however, higher on the hydroxylated and carboxylated surfaces (Lee et al., 2005). The protein adsorbed SAM surfaces appeared to be uniform on larger scans of 10 µm × 10 µm but non-homogeneous at submicron range.

The FN protein surface coverage on the Ca-P coated surfaces was estimated using a protocol and technique previously determined (Toworfe et al., 2004). In 10 µg/mL FN concentration, The FN surface density data was approaching the saturation value on the SAMs and SiOx surfaces without and with Ca-P coating. The crucial effect, therefore, of the SAM surface groups was not to play a dominant role in the protein adsorption on the Ca-P coated surfaces. The surface density of adsorbed FN was around 350 ng/cm2 on both the Ca-P coated and non-coated SAM surfaces incubated in 10 µg/mL FN concentration. The slightly higher FN coverage obtained on the non-coated −NH2 surface compared to the other non-coated SAM substrates, can be attributed to the presence of amino groups which promotes the adsorption of fibronectin due to hydrophobic interactions (Advincula et al., 2005). The adsorbed mass of fibronectin on all the silanes and the Ca-P coated surfaces was congruent with values reported by other groups (Dimilla et al., 1992; Lhoest et al., 1998). The higher FN densities obtained on surfaces with higher roughness indicate that surface topography together with higher surface area could be partly responsible for the protein surface density, in addition to the bioactivities of the Ca-P surface. The structural rearrangement of the protein molecules during the slow adsorption process as the serum protein binds to the functionalized and mineralized surfaces could also contribute to the extent of protein coverage (do Serro et al., 1999).

The differences in cellular attachment and proliferation on the bioengineered substrates we have studied may have been influenced mostly by roughness and surface composition of Ca-P precipitates together with the SAM surface groups. It must be pointed out that FN is one of the major components of ECM and is also secreted by osteoblasts. When absorbed onto biomaterials, FN plays a crucial role in bone cell attachment, proliferation and differentiation. El- Ghannam et al. (El-Ghannam et al., 1999) and Gercia et al. (Garcia et al., 1998) documented this effect for the calcium phosphate layers that forms on bioactive glass. This is also supported by findings that FN binding on porous calcium phosphate ceramics supports the osteoblastic differentiation of mesenchymal stem cells (Kotobuki et al, 2005). The increase in MC3T3-E1 cell number on the 7-day Ca-P coated surfaces was very significant, although the properties of adhesive proteins (FN) adsorbed onto a material may directly influence cellular function. This effect of surface roughness due to Ca-P coating has been reported with rat (Lincks et al., 1998) and human osteoblast-like cells (Gross et al., 1997; Chou et al., 1999), and also with human osteosarcoma cells (Ong et al., 1997). It was supposed that cells encounter a limited number of adhesion sites on the relatively smooth surfaces, such as on the 1-day Ca-P coated surfaces, which does not stabilize the membrane extensions. On the other hand, the rougher substrates, like the Ca-P coated −OH surfaces, make available a greater number of attachment sites which result in the stabilization of actin filaments and the association of the cytoskeletal elements with focal contacts (Stanford et al., 1994). This reasoning is also valid for the formation of actin filaments of cell cytoskeleton.

Alkaline phosphatase activity is an indicator of the differentiated osteoblast phenotype. We observed a dramatic increase in ALP activity over the 7 day cell culture on the 7-day Ca-P coated substrates, compared to the minimum ALP activity evident on the SiOx and SAM surfaces without Ca-P coating. Clearly ALP activity was greatly enhanced on the −OH SAM coated surface due to the abundant presence of hydroxyapatite. This corroborates our initial assertion that bone formation is promoted on HA surfaces, since ALP activity is associated with the matrix maturation leading to bone matrix formation (Gerstenfeld et al., 1987; Siffert, 1951). In addition, increased ALP is accompanied by increased mineralization. It is also interesting to note the similarity in trend of ALP activity and cell proliferation on the substrates without Ca-P and with 1-day Ca-P coating. Surface groups SAM (particularly −COOH and −OH) might have substantial effect on the roughness and composition of Ca-P layers formed.

Cell spreading and shape are important modulators of cellular function. Cell spreading appears to be greatly enhanced on the Ca-P coated substrates. The early stages of the fundamental biological process of cell spreading have been related to surface wetting by a liquid drop (Frisch and Thoumine, 2002). We observed in this study that under physiological conditions MC3T3-E1 cells transform from rounded morphology, characteristic of initial attachment (that occurs within a few seconds to minutes) to a discoid shape after 36 h. The cell body shows flattening and its plasma membrane spreads over the substratum, while the organization of actin into microfilament bundles also occurs. Changes in cell shape are accompanied by changes in cell function. In MC3T3-E1 cells, inhibition of cell spreading induces apoptosis (Grigoriou et al., 2005) and proliferation can only occur in cells that are spread out. The presence of hydrophilic −COOH and −OH groups of SAM might have played a crucial role to this wetting phenomenon.

5. Conclusions

In conclusion, we found that biological molecules, such as adhesive proteins and MC3T3-E1 cells, exhibit differences in attachment and spreading on materials with different surface characteristics. We have also showed that alkaline phosphate expression can be influenced by substrate surface composition, i.e., the presence of a Ca-P coating on a material. Ca-P thus enhances attachment and proliferation of MC3T3-E1 cells, in vitro. Moreover, the presence of SAMs surface groups play a crucial role in the final morphology and composition of Ca-P layers, which in turn influence its biological activity.

Table 2.

The table outlines a summary of experimental methods used for the surface functionalization of silicon substrates, biomineralization and biomolecular adsorption.

Substate Preparation Method
a) SiOx Cleaning and oxidation of Silicon oxide
b) SiOx +SAMs SAMs – APTES, TESPSA and GPTMS grafted onto to SiOx
c) SiOx +SAMs+Ca-P Substrate in b) coated with Ca-P (by immersion in SPF)
d) SiOx +SAMs+Ca-P+Fn Substrate in c) incubated in Fn solutions (in PBS)
e) SiOx +SAMs+Fn Substrate in b) +Fn (10µg/mL)
Open in a new tab

Acknowledgements

The authors would like to express their gratitude to Doug Yates, Jim Ferris and Kevin Macke of Penn Regional Materials Characterization Facility, University of Pennsylvania, for their technical assistance and Shula Radin of the Center for Bioactive Materials and Tissue Engineering, University of Pennsylvania who assisted with the FTIR analysis; Adam Zahn and Bradley Snyder (both of the Department of Orthopaedic Surgery, Thomas Jefferson University) for alkaline phosphate bioassay and cell culture work. This Research was supported by NIH Grant: ROl-DE-13009.

References

  • 1.Advincula M, Fan XW, Lemons J, et al. Surface modification of surface sol-gel derived titanium oxide films by self-assembled monolayers (SAMs) and non-specific protein adsorption studies. Colloids and Surfaces B-Biointerfaces. 2005;42:29–43. doi: 10.1016/j.colsurfb.2004.12.009. [DOI] [PubMed] [Google Scholar]
  • 2.Albrektsson T, Branemark PI, Hansson HA, et al. The Interface Zone of Inorganic Implants in vivo - Titanium Implants in Bone. Annals of Biomedical Engineering. 1983;11:1–27. [Google Scholar]
  • 3.Albrektsson T, Hansson HA. Ultrastructural Characterization of the Interface between Bone and Sputtered Titanium or Stainless-Steel Surfaces. Biomaterials. 1986;7:201–205. doi: 10.1016/0142-9612(86)90103-1. An. [DOI] [PubMed] [Google Scholar]
  • 4.Chou L, Marek B, Wagner WR. Effects of hydroxylapatite coating crystallinity on biosolubility, cell attachment efficiency and proliferation in vitro. Biomaterials. 1999;20:977–985. doi: 10.1016/s0142-9612(98)00254-3. [DOI] [PubMed] [Google Scholar]
  • 5.Dimilla PA, Albelda SM, Quinn JA. Adsorption and Elution of Extracellular-Matrix Proteins on Nontissue Culture Polystyrene Petri Dishes. Journal of Colloid and Interface Science. 1992;153:212–225. [Google Scholar]
  • 6.do Serro A, Fernandes AC, Saramago BDV, et al. Bovine serum albumin adsorption on titania surfaces and its relation to wettability aspects. Journal of Biomedical Materials Research. 1999;46:376–381. doi: 10.1002/(sici)1097-4636(19990905)46:3<376::aid-jbm10>3.0.co;2-t. [DOI] [PubMed] [Google Scholar]
  • 7.Ducheyne P, Cuckler JM. Bioactive ceramic prosthetic coatings. Clinical Orthopedics and Related Research. 1992;276:102–114. [PubMed] [Google Scholar]
  • 8.Ducheyne P, Radin S, Heughbaert M, Heughbaert JC. Calcium phosphate ceramic coating on porous titanium: Effect of structure and composition on electrophoretic deposition, vacuum sintering and in vitro dissolution. Biomaterials. 1990;11:244–254. doi: 10.1016/0142-9612(90)90005-b. [DOI] [PubMed] [Google Scholar]
  • 9.Ducheyne P, Qiu Q. Bioactive ceramics: the effect of surface reactivity on bone formation and bone cell function. Biomaterials. 1999;20:2287–2303. doi: 10.1016/s0142-9612(99)00181-7. [DOI] [PubMed] [Google Scholar]
  • 10.El-Ghannam A, Ducheyne P, Shapiro IM. Bioactive material template for in vitro synthesis of bone. J Biomed Mater Res. 1995;29:359–370. doi: 10.1002/jbm.820290311. [DOI] [PubMed] [Google Scholar]
  • 11.El-Ghannam A, Ducheyne P, Shapiro IM. Effect of serum proteins on osteoblast adhesion to surface-modified bioactive glass and hydroxyapatite. Journal of Orthopaedic Research. 1999;17:340–345. doi: 10.1002/jor.1100170307. [DOI] [PubMed] [Google Scholar]
  • 12.El-Ghannam A, Ducheyne P, Shapiro I. Bioactive glass templates for the synthesis of bone-like tissue in vitro. In: Mikos AG, Murphy RM, Bernstein H, et al., editors. Biomaterials for cell and drug delivery. Pittsburgh: Materials Research Society; 1994. pp. 257–262. [Google Scholar]
  • 13.El-Ghannam AR, Ducheyne P, Risbud M, et al. Model surfaces engineered with nanoscale roughness and RGD tripeptides promote osteoblast activity. Journal of Biomedical Materials Research Part A. 2004;68A:615–627. doi: 10.1002/jbm.a.20051. [DOI] [PubMed] [Google Scholar]
  • 14.Frisch T, Thoumine O. Predicting the kinetics of cell spreading. Journal of Biomechanics. 2002;35:1137–1141. doi: 10.1016/s0021-9290(02)00075-1. [DOI] [PubMed] [Google Scholar]
  • 15.Gao BC, Saba TM, Tsan MF. Role of αvβ3-integrin in TNF-α-induced endothelial cell migration. American Journal of Physiology-Cell Physiology. 2002;283:C1196–C1205. doi: 10.1152/ajpcell.00064.2002. [DOI] [PubMed] [Google Scholar]
  • 16.Garcia AJ, Ducheyne P, Boettiger D. Effect of surface reaction stage on fibronectin-mediated adhesion of osteoblast-like cells to bioactive glass. Journal of Biomedical Materials Research. 1998;40:48–56. doi: 10.1002/(sici)1097-4636(199804)40:1<48::aid-jbm6>3.0.co;2-r. [DOI] [PubMed] [Google Scholar]
  • 17.Gerstenfeld LC, Chipman SD, Glowacki J, et al. Expression of Differentiated Function by Mineralizing Cultures of Chicken Osteoblasts. Developmental Biology. 1987;122:49–60. doi: 10.1016/0012-1606(87)90331-9. [DOI] [PubMed] [Google Scholar]
  • 18.Grigoriou V, Shapiro IM, Cavalcanti-Adam EA, et al. Apoptosis and survival of osteoblast-like cells are regulated by surface attachment. Journal of Biological Chemistry. 2005;280:1733–1739. doi: 10.1074/jbc.M402550200. [DOI] [PubMed] [Google Scholar]
  • 19.Gross KA, Berndt CC, Goldschlag DD, et al. In vitro changes of hydroxyapatite coatings. International Journal of Oral & Maxillofacial Implants. 1997;12:589–597. [PubMed] [Google Scholar]
  • 20.Hanawa T, Ota M. Characterization of surface film formed on titanium in electrolyte using XPS. Appl. Surface Sci. 1992;55:269–276. [Google Scholar]
  • 21.Hansson HA, Albrektsson T, Branemark PI. Structural Aspects of the Interface between Tissue and Titanium Implants. Journal of Prosthetic Dentistry. 1983;50:108–113. doi: 10.1016/0022-3913(83)90175-0. [DOI] [PubMed] [Google Scholar]
  • 22.Hench LL. Bioceramics - from Concept to Clinic. Journal of the American Ceramic Society. 1991;74:1487–1510. [Google Scholar]
  • 23.Hench LL. Bioactive ceramics. In: Ducheyne P, Lemons JE, editors. Bioceramics: Material Characteristics Versus In Vivo Behavior. New York: The New York Academy of Sciences; 1988. pp. 54–71. [Google Scholar]
  • 24.Huang ZJ, Haugland RP, You WM. Phallotoxin and Actin Binding Assay by Fluorescence Enhancement. Analytical Biochemistry. 1992;200:199–204. doi: 10.1016/0003-2697(92)90299-m. [DOI] [PubMed] [Google Scholar]
  • 25.Iwasaki M, Nakata K, Nakahara H, et al. Transforming Growth-Factor-Beta-1 Stimulates Chondrogenesis and Inhibits Osteogenesis in High-Density Culture of Periosteum-Derived Cells. Endocrinology. 1993;132:1603–1608. doi: 10.1210/endo.132.4.8462458. [DOI] [PubMed] [Google Scholar]
  • 26.Knabe C, Berger G, Gildenhaar R, Meyer J, Howlett CR, Markoviv B, Zreiqat H. Effect of rapidly resorbable calcium phosphates and a calcium phosphate bone cement on the expression of bone related genes and proteins in vivo. Journal of Biomedical Materials Research. 2004;69(A):154–154. doi: 10.1002/jbm.a.20131. [DOI] [PubMed] [Google Scholar]
  • 27.Kasemo B. Biocompatibility of Titanium Implants - Surface Science Aspects. Journal of Prosthetic Dentistry. 1983;49:832–837. doi: 10.1016/0022-3913(83)90359-1. [DOI] [PubMed] [Google Scholar]
  • 28.Keselowsky BG, Collard DM, Garcia AJ. Integrin binding specificity regulates biomaterial surface chemistry effects on cell differentiation. Proceedings of the National Academy of Sciences of the United States of America. 2005;102:5953–5957. doi: 10.1073/pnas.0407356102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Kishore R, Samuel M, Khan MY, et al. Interaction of the NH2-terminal domain of fibronectin with heparin - Role of the Omega-loops of the type I modules. Journal of Biological Chemistry. 1997;272:17078–17085. doi: 10.1074/jbc.272.27.17078. [DOI] [PubMed] [Google Scholar]
  • 30.Kokubo T. Bioactive Glass-Ceramics - Properties and Applications. Biomaterials. 1991;12:155–163. doi: 10.1016/0142-9612(91)90194-f. [DOI] [PubMed] [Google Scholar]
  • 31.Kokubo T, Kushitani H, Sakka S, et al. Solutions Able to Reproduce In vivo Surface-Structure Changes in Bioactive Glass-Ceramic A-W. Journal of Biomedical Materials Research. 1990;24:721–734. doi: 10.1002/jbm.820240607. [DOI] [PubMed] [Google Scholar]
  • 32.Kotobuki N, Ioku K, Kawagoe D, et al. Observation of osteogenic differentiation cascade of living mesenchymal stem cells on transparent hydroxyapatite ceramics. Biomaterials. 2005;26:779–785. doi: 10.1016/j.biomaterials.2004.03.020. [DOI] [PubMed] [Google Scholar]
  • 33.Levene H, Phuvanartnuruks V, Abramson S, et al. Chelation of calcium ions by some tyrosine-derived polymers may be related to improved bone biocompatibility. Transactions of the Society for Biomaterials; 25th Annual Meeting of the Society for Biomaterials; 1999. p. 13. [Google Scholar]
  • 34.Lhoest JB, Detrait E, van den Bosch de Aguilar P, et al. Fibronectin adsorption, conformation, and orientation on polystyrene substrates studied by radiolabeling, XPS, and ToF SIMS. Journal of Biomedical Materials Research. 1998;41:95–103. doi: 10.1002/(sici)1097-4636(199807)41:1<95::aid-jbm12>3.0.co;2-g. [DOI] [PubMed] [Google Scholar]
  • 35.Li P, Bakker D, vanBlitterswijk CA. The bone-bonding polymer Polyactive® 80/20 induces hydroxycarbonate apatite formation in vitro. Journal of Biomedical Materials Research. 1997;34:79–86. doi: 10.1002/(sici)1097-4636(199701)34:1<79::aid-jbm11>3.0.co;2-k. [DOI] [PubMed] [Google Scholar]
  • 36.Li PJ, de Groot K. Calcium-Phosphate Formation within Sol-Gel Prepared Titania in vitro and in vivo. Journal of Biomedical Materials Research. 1993;27:1495–1500. doi: 10.1002/jbm.820271205. [DOI] [PubMed] [Google Scholar]
  • 37.Li PJ, Ye X, Kangasniemi I, et al. In-Vivo Calcium-Phosphate Formation Induced by Sol-Gel-Prepared Silica. Journal of Biomedical Materials Research. 1995;29:325–328. doi: 10.1002/jbm.820290307. [DOI] [PubMed] [Google Scholar]
  • 38.Li PJ, Ducheyne P. Quasi-biological apatite film induced by titanium in a simulated body fluid. Journal of Biomedical Materials Research. 1998;41:341–348. doi: 10.1002/(sici)1097-4636(19980905)41:3<341::aid-jbm1>3.0.co;2-c. [DOI] [PubMed] [Google Scholar]
  • 39.Li T, Lee J, Kobayashi T, Aoki H. Hydroxyapatite coating by dipping method and bone bonding strength. Journal of Materials Science: Materials in Medicine. 1996;7:355–357. [Google Scholar]
  • 40.Lincks J, Boyan BD, Blanchard CR, et al. Response of MG63 osteoblast-like cells to titanium and titanium alloy is dependent on surface roughness and composition. Biomaterials. 1998;19:2219–2232. doi: 10.1016/s0142-9612(98)00144-6. [DOI] [PubMed] [Google Scholar]
  • 41.Liu DP, Majewski P, O'Neill BK, et al. The optimal SAM surface functional group for producing a biomimetic HA coating on Ti. Journal of Biomedical Materials Research Part A. 2006;77A:763–772. doi: 10.1002/jbm.a.30641. [DOI] [PubMed] [Google Scholar]
  • 42.Model MA, Healy KE. Quantification of the surface density of a fluorescent label with the optical microscope. Journal of Biomedical Materials Research. 2000;50:90–96. doi: 10.1002/(sici)1097-4636(200004)50:1<90::aid-jbm13>3.0.co;2-3. [DOI] [PubMed] [Google Scholar]
  • 43.Ong JL, Carns DL, Cardenas HL, et al. Surface roughness of titanium on bone morphogenetic protein-2 treated cells in vitro. Implant Dent. 1997;6:19–24. doi: 10.1097/00008505-199700610-00004. [DOI] [PubMed] [Google Scholar]
  • 44.Radin SR, Ducheyne P. The Effect of Calcium-Phosphate Ceramic Composition and Structure on in vitro Behavior .2. Precipitation. Journal of Biomedical Materials Research. 1993;27:35–45. doi: 10.1002/jbm.820270106. [DOI] [PubMed] [Google Scholar]
  • 45.Siffert RS. The role of ALPase in osteogenesis. J Exp Med. 1951;93:415–425. doi: 10.1084/jem.93.5.415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Stanford CM, Keller JC, Solursh M. Bone Cell Expression on Titanium Surfaces Is Altered by Sterilization Treatments. Journal of Dental Research. 1994;73:1061–1071. doi: 10.1177/00220345940730050801. [DOI] [PubMed] [Google Scholar]
  • 47.Toworfe GK, Composto RJ, Shapiro IM, et al. Nucleation and growth of calcium phosphate on amine-, carboxyl- and hydroxyl-silane self-assembled monolayers. Biomaterials. 2006;27:631–642. doi: 10.1016/j.biomaterials.2005.06.017. [DOI] [PubMed] [Google Scholar]
  • 48.Toworfe GK, Composto RJ, Adams CS, et al. Fibronectin adsorption on surface-activated poly(dimethylsiloxane) and its effect on cellular function. Journal of Biomedical Materials Research Part A. 2004;71A:449–461. doi: 10.1002/jbm.a.30164. [DOI] [PubMed] [Google Scholar]
  • 49.Wang K, Feramisco JR, Ash JF. Fluorescent Localization of Contractile-Proteins in Tissue-Culture Cells. Methods in Enzymology. 1982;85:514–562. doi: 10.1016/0076-6879(82)85050-7. [DOI] [PubMed] [Google Scholar]
  • 50.Zhu PX, Masuda Y, Koumoto K. The effect of surface charge on hydroxyapatite nucleation. Biomaterials. 2004;25:3915–3921. doi: 10.1016/j.biomaterials.2003.10.022. [DOI] [PubMed] [Google Scholar]

RESOURCES