Long-term stability of orthopaedic and dental implants depends on the integration of the artificial material into the surrounding bone tissue. The physical and chemical properties of implants, including those made of metals such as titanium, are thought to influence osseointeg-ration. Despite the known importance of this interface, little is known about the factors that promote its formation. In this study, chick embryonic calvarial osteoblasts were cultured in vitro on smooth, rough-textured and porous-coated titanium surfaces and examined for morphology, biosynthesis of extracellular matrix and mineralization as a function of culture time. Scanning electron microscopy revealed that osteoblasts adhered securely to the titanium surface and frequently bridged the uneven surface by means of cellular processes. The osteoblast phenotype was retained in the cell cultures on titanium. In addition, the synthesis of extracellular matrix and subsequent mineralization were both substantially enhanced in the cultures on rough-textured and porous-coated titanium. These results strongly suggest that porous or rough titanium implant surfaces may act like “natural” substrata to permit microscopic tissue/cell ingrowth to improve clinical implant fixation.

Increasing clinical usage and application of biomaterials in tissue substitution and replacement have led to the development of a new field of biomedical research, implantology, that explores the biology involved in the integration of implant surfaces into the surrounding tissues. In orthopaedics and dentistry, the process of “osseointegration” is considered of pivotal importance for establishing an intimate and stable connection between bone and implant surface (Brânemark, 1983). If osseointegration is not successfully achieved, fibrous tissue is generated instead, leading ultimately to loosening and loss of the implant. Despite the known importance of this interface, little is known about the factors that promote its formation. It is generally believed that both the physical and the chemical properties of the implants significantly influence the nature of the tissue response (Crowninshield, 1988).

The implant surface morphology is thought to contribute to adhesion and mechanical interlocking with the surrounding tissue (Thomas et al. 1987). Thus, the degree of smoothness or roughness of the implant surface can influence the response of the tissue to the implant (Hanker and Giammara, 1988; Michaels et al. 1989) and subsequently its stability within the skeleton. Bone ingrowth into a porous surface appears to require interface conditions less stringent than that required for bone ingrowth and bonding to a smooth surface (Albrektsson et al. 1981), perhaps as a result of the increased surface area. Therefore, it is reasonable to postulate that implant fixation may be significantly improved through enhancement of microscopic tissue ingrowth into porous implant surfaces. Metals, plastics and bioceramics are the most common types of materials used as implants. Interestingly, it has been shown that titanium-containing metals are more receptive to direct bone bonding than other alloys, polymers, or bioglass (Albrektsson et al. 1981; Barth et al. 1985; Williams, 1981). Golijanin and Bernard (1988) recently characterized mouse embryonic calvarial mesenchymal cells cultured on Petri dishes coated with 40–60 nm thick layers of titanium, titanium-64, and vitallium. These workers reported that cultures on pure titanium exhibit considerable biocompatibility, good tissue adhesion, the presence of a larger number of bone colonies and lack of corrosion, when compared to those cultured on titanium-64 and vitallium. A direct bone-implant contact mediated by a titanium oxide-glycoprotein layer has been postulated as the reason for the high biocompatibility of titanium surfaces (Gristina, 1987). Studies indicate that bone cells interact with the titanium oxide surface and that collagen filaments approach to within 20 nm of the oxide layer (Albrekts-son, 1985; Kasemo and Lausmaa, 1988).

To test the hypothesis that the physical characteristics of titanium implant surfaces may influence their integration into bone tissue, we have analyzed the behavior of osteoblasts, isolated from chicken embryonic calvaria and cultured on different titanium surfaces in vitro. In culture, these cells have the capacity to synthesize extracellular matrix components characteristic of bone, including collagen type I, and undergo mineralization (Gerstenfeld et al. 1988).

In this study, bone cells were isolated enzymatically from chick embryonic calvaria and grown on disks of titanium alloy with smooth (satin-polished surface), rough-textured surface, or standard porous coating, and on conventional, tissue culture grade polystyrene plastic or glass coverslips as control. Morphological and biochemical analyses showed that the bone cells adhered to and were “integrated” with the titanium substrata, and that synthesis of bone extracellular matrix and subsequent mineralization were both enhanced in cultures plated on rough-textured and porous-coated titanium surfaces.

Calvarial cells

Bone cells were isolated by sequential trypsin-collagenase digestion of day-16 chick embryonic calvaria using the procedure of Wong and Cohn (1974); cells were collected from the second and third digests. The calvarial cells were grown in 6-well culture plates (see below) at an initial plating density of 1.5×10s cells/well. The cells were maintained, for up to 6 days, in Dulbecco’s Modified Eagle’s Medium (DMEM), supplemented with 10% fetal calf serum (Hyc-lone), 50 μg ml-1 ascorbic acid, 5 mM βglycerophosphate, and penicillin/streptomycin. The culture medium was changed every two days during culture.

Culture surfaces

In control cultures, calvarial osteoblasts were plated directly onto tissue culture polystyrene plastic (Corning 6-well tissue culture dish) or onto glass coverslips (22 mm diameter, No. 1, A.H. Thomas). For the titanium cultures, disks of titanium alloy (containing 6% aluminum and 4% vanadium, TÍ-6A1-4V; 32 mm diameter) were placed into the wells, and cells were then seeded. Three types of titanium alloy surfaces were used (Fig. 1): 1) smooth, “satin-grade” surface obtained by industrial polishing of the titanium disk; 2) rough-textured surface, produced by sandblasting; and 3) porous-coated surface, produced by plasma spraying the rough-textured surface with TI-6A1-4V powder (particle size, 75–250 μM) to result in the highly uneven porous coating (100–1000 am pore size). These disks were produced and kindly provided by Biomet, Inc. (Warsaw, Indiana) using procedures identical to those employed in the production of orthopaedic implants.

Fig. 1.

Surface view of titanium alloy surfaces. The titanium surfaces tested in this study included smooth (prepared by commercial satin-grade polishing), rough-textured (prepared by sand-blasting), and porous-coated (prepared by plasma spraying with TÍ-6A1-4V alloy). These surfaces are shown respectively from left to right (À), and at higher magnification in (B), (C) and (D). (A) Top view of the disks, 32 mm in diameter. (B-D) Scanning electron microscopic view of the surfaces. Bar, 100μM.

Fig. 1.

Surface view of titanium alloy surfaces. The titanium surfaces tested in this study included smooth (prepared by commercial satin-grade polishing), rough-textured (prepared by sand-blasting), and porous-coated (prepared by plasma spraying with TÍ-6A1-4V alloy). These surfaces are shown respectively from left to right (À), and at higher magnification in (B), (C) and (D). (A) Top view of the disks, 32 mm in diameter. (B-D) Scanning electron microscopic view of the surfaces. Bar, 100μM.

The disks were routinely cleaned by sonicating in sodium dodecyl sulfate (SDS), and sterilized by autoclaving.

Estimation of cell number

To estimate cell number in the cultures, it was discovered that cell detachment by standard trypsin-EDTA dissociation was ineffective for the titanium surfaces. Thus to compare cell number between cultures, relative cellular protein content and level of the cytosolic enzyme, lactate dehydrogenase (LDH) were measured. Cultures were first washed extensively with Hank’s Balanced Salt Solution (HBSS), then lysed in either 1% SDS for protein determination (see below) or 1% Triton X-100 in phosphate-buffered saline for LDH assay (Sigma Diagnostic Kit 500). Repeated initial analyses showed no significant difference between cultures on tissue culture polystyrene and the titanium surfaces with respect to these parameters.

Scanning electron microscopy (SEM)

At various time points, cultures were fixed in glutaraldehyde (25%) in cacodylate buffer (0.13 M, pH 7.4) (Tuan et al. 1991). After critical point drying, the specimens were observed using a JEOL 35C scanning electron microscope.

Light microscopy

Cells cultured on plastic or glass coverslips were observed directly by phase-contrast microscopy using an Olympus BH-2 microscope and photographed using Kodak Panatomic-X film. Cultures on titanium that were stained with Alizarin Red (see below) were photographed using Kodak Ektachrome film.

Estimation and characterization of collagen biosynthesis

For quantitative and qualitative analyses of collagen synthesis, cells were metabolically radiolabelled with [,4C]pro-line and [3H]leucine as follows. At day 5 of culture, cells were fed with medium containing 0.5 μCi ml-1 [14C]proline and 2 uCi ml-1 [3H]leucine (ICN) for 24 h in the presence of 50;μg mT β-aminoproprionitrile (Tuan and Lynch, 1983). To harvest, cells were fixed with 10% trichloroacetic acid for 10 min, washed, and solubilized with 2% SDS in 6 M urea. Incorporation of both isotopes was determined in a 50 βl aliquot by liquid scintillation counting in a Beckman L1800 Liquid Scintillation Counter equipped with dual isotope counting program. Relative collagen synthesis was estimated based on the DPM ratio of proline to leucine incorporation.

To characterize collagen type synthesis, the labelled cell extract was analyzed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) using an 8% separating gel (Laemmli and Favre, 1973; Tuan and Lynch, 1983). The gels were imaged by direct autoradiography to detect 14C radioactivity, and by fluorography to detect both 14C and 3H radioactivity. Collagen type was estimated based on the relative radiolabelling intensities, as measured by densitometry, of the collagen al α1 and α2 chains (Tuan and Lynch, 1983).

Estimation and characterization of calcification

For quantitative estimation of calcification, the osteoblast cultures were 1) radiolabelled on day 5 of culture with 1 μCi ml-1 of [45Ca]Cl2 (ICN) for 24 h (to monitor the amount of calcification occuring in one day), or 2) stained with Alizarin Red (Sigma) (to measure the total amount of mineral produced during the 6 days of culture). After 45Ca-radiolabel-ling, the cultures were harvested by washing twice with HBSS to remove free radioactivity and lysed in 2% SDS for another 24 h. The amount of incorporated 45Ca (counts min-1 μg-1 protein) was measured by liquid scintillation counting. For Alizarin Red staining, the cells were washed twice with warm HBSS and fixed with 4% paraformaldehyde for 10 min. Cells were then stained with Alizarin Red for 2 h and soaked in 100% ethanol for 10 min to remove excess dye. Cells were airdried, solubilized with 2% SDS, and absorbance of the eluant at 609 nm was measured. All 45Ca counts min-1 and Alizarin Red absorbance at 609 nm were calculated by subtracting the background values from cell-free substrata that were incubated under identical conditions.

Calcification in the cultures was also visualized by the following methods. The 45Ca-labelled cultures, after being air-dried, were imaged autoradiographically by secured placement onto Kodak XAR film. In addition, the Alizarin Red-stained cultures were visually examined and color-photographed.

Alkaline phosphatase assay

The enzyme alkaline phosphatase (ALP) is prominently associated with osteoblast maturation and calcification (Marks and Popoff, 1988) and was used here as a marker. On day 6, the cells were lysed with 0.5% Triton X-100 for 15 min at 4°C. Samples were sonicated and the supernatant after centrifugation was assayed for ALP at 30°C, using p- nitrophenol phosphate as a substrate (Sigma Bulletin No. 104). The activity was expressed as units mg-1 protein.

Protein determination

Estimation of protein content was carried out using the Pierce BCA Protein Assay Kit (Pierce Chemicals).

The goal of this study was to examine how titanium surfaces similar to those commonly used for orthopaedic endoprosthetic implants might influence the biological activities of bone-derived cells. The analyses included 1) morphology, 2) extracellular matrix synthesis and 3) calcification, and were primarily focused on the rough-textured and porous-coated surfaces, as compared to conventional tissue culture surfaces (plastic and glass). To examine whether the changes were due to surface chemistry or surface topography, some analyses were also carried out using cultures plated onto polished titanium disks.

Cellular morphology

The cultures were examined by phase-contrast light microscopy (plastiq/glass) and SEM (plastic/giassμita-nium) at various time points during the 6-day culture period. As detailed in Figs 25, there did not appear to be major morphological differences between the cells maintained on the various surfaces. The following general profile was seen for all the cultures.

Fig. 2.

Morphology of chick embryonic calvarial cells plated on various surfaces after 42 h in culture. The surfaces were: (A,B) glass coverslip, (C) rough-textured titanium and (D) porous-coated titanium. Note the formation of cell rosettes on all culture surfaces, particularly on the glass coverslip, with cuboidal cells in the center and more elongated cells in the periphery. Bar, 125 μm (A), 100 μm (B), or 50 μm (C and D).

Fig. 2.

Morphology of chick embryonic calvarial cells plated on various surfaces after 42 h in culture. The surfaces were: (A,B) glass coverslip, (C) rough-textured titanium and (D) porous-coated titanium. Note the formation of cell rosettes on all culture surfaces, particularly on the glass coverslip, with cuboidal cells in the center and more elongated cells in the periphery. Bar, 125 μm (A), 100 μm (B), or 50 μm (C and D).

Shortly after plating, within 1–2 days, the calvarial cells formed distinct rosette-like aggregates. On glass coverslips, these aggregates consisted of more cuboidal cells in the center, whereas more elongated cells were seen in the periphery’ (Fig. 2A,B). On the titanium surfaces, similar aggregates were also seen (Fig. 2C,D), although it was obvious that the uneven terrain appeared to limit the size and compromise the symmetry of these aggregates.

The calvarial cells proliferated and formed almost confluent cultures by day 5 to 6 (Fig. 3A-E). Again, the morphology of cells plated on plastic or glass coverslips (Fig. 3A) was similar to that of cells plated on titanium surfaces (Fig. 3B-E). In both the rough-textured and porous-coated titanium cultures, the cells spread to cover essentially all available surface regions. In particular, cells were often seen to form bridges between domains of the uneven surface (arrows, Fig. 3D).

Fig. 3.

Morphology of chick embryonic calvarial cells plated on various surfaces after 5 days in culture. The surfaces were: (A) glass coverslip, (B,C) rough-textured titanium and (D,E) porous-coated titanium. Note the extent of cellular confluency on all culture surfaces. On titanium, cells appeared to extend actively onto all available surfaces; at high magnification (C,E), cellular processes were seen to bridge the uneven surfaces (arrows). Bar, 125 μm (A), 100 μm (B and D), or 50 μm (C and E).

Fig. 3.

Morphology of chick embryonic calvarial cells plated on various surfaces after 5 days in culture. The surfaces were: (A) glass coverslip, (B,C) rough-textured titanium and (D,E) porous-coated titanium. Note the extent of cellular confluency on all culture surfaces. On titanium, cells appeared to extend actively onto all available surfaces; at high magnification (C,E), cellular processes were seen to bridge the uneven surfaces (arrows). Bar, 125 μm (A), 100 μm (B and D), or 50 μm (C and E).

On day 6 to 7, matrix calcification was seen in all cultures (Fig. 4A-D). When examined by light microscopy, cultures plated on plastic revealed abundant refractile extracellular materials that gathered as clumps around areas of high cell density. As observed by SEM (Fig. 4B), this material consisted of extracellular particles of varying sizes (/on dimensions) attached to the cell surface. Similar structures were also detected in the titanium cultures (Fig. 4C,D). Because of the high cell density and the uneven surface of the titanium substrata, it was difficult to decipher whether these particulate structures, which often displayed a fuzzy, powdery appearance (Fig. 4D), also formed in the more indented areas of the surface. The calcified nature of this material was indicated by analyses described below, as well as supported by our previous electron probe elemental analysis (Groessner-Schreiber et al. 1991), that revealed a hydroxyapatite-like composition.

Fig. 4.

Morphology of chick embryonic calvarial cells plated on various surfaces after 5 days in culture. The surfaces were: (A,B) glass coverslip, (C) rough-textured titanium and (D) porous-coated titanium. Note the presence of refractile particles adherent to the culture (A), and the apparent globular, extracellular and matrix-associated particles adherent to the cultured cells as shown by SEM (B, C, and D). Bar, 400 μm (A), or 20 μm (B, C, and D).

Fig. 4.

Morphology of chick embryonic calvarial cells plated on various surfaces after 5 days in culture. The surfaces were: (A,B) glass coverslip, (C) rough-textured titanium and (D) porous-coated titanium. Note the presence of refractile particles adherent to the culture (A), and the apparent globular, extracellular and matrix-associated particles adherent to the cultured cells as shown by SEM (B, C, and D). Bar, 400 μm (A), or 20 μm (B, C, and D).

Collagenous extracellular matrix biosynthesis

The cultures were compared with respect to the biosynthesis of the collagenous extracellular matrix on day 6 of culture as described in Materials and methods. Specifically, the biosynthesis of extracellular matrix was analyzed with respect to the formation of collagen based on [14C]proline incorporation, and general protein synthesis as evaluated by [3H]leucine labeling. The [14C]proline:[3H]leucine ratio represented a normalized index for estimating relative collagen biosynthesis. The data in Fig. 5A show that the ratio of [MC]proline to [3H]leucine incorporation was significantly higher in cells grown on titanium disks compared to cells cultured on plastic as control, suggesting that cells on titanium disks synthesized more collagen.

Fig. 5.

Characterization of collagenous matrix biosynthesis in calvarial cell cultures plated on various surfaces (plastic, rough-textured titanium and porous-coated titanium). (A) Estimation of collagen biosynthesis based on the relative incorporation of [14C]proline:[3H]leucine in the cultures (see Materials and methods). The proline:leucine ratio was significantly higher in cultures plated on titanium surfaces (P <0.001) compared to the control, plastic surface. Mean values for rough versus porous titanium disks showed no significant difference. Values represent mean ± S.D. for 9-wells or titanium disks. (B) Characterization of collagen type synthesis by SDS-PAGE (see Materials and methods). The gels were imaged autoradiographically (right panel) to detect 14C radioactivity, and fluorographically (left panel) to detect both 14C and 3H radioactivity. Equal amounts of protein (left) or 3H radiolabel (right) were loaded. In both panels, collagen a chains represented the most predominant labeled proteins, with an α1:β2 ratio of approximately 2, consistent with collagen type I synthesis. C, control (polystyrene plastic); P, porous-coated titanium; R, rough-textured titanium.

Fig. 5.

Characterization of collagenous matrix biosynthesis in calvarial cell cultures plated on various surfaces (plastic, rough-textured titanium and porous-coated titanium). (A) Estimation of collagen biosynthesis based on the relative incorporation of [14C]proline:[3H]leucine in the cultures (see Materials and methods). The proline:leucine ratio was significantly higher in cultures plated on titanium surfaces (P <0.001) compared to the control, plastic surface. Mean values for rough versus porous titanium disks showed no significant difference. Values represent mean ± S.D. for 9-wells or titanium disks. (B) Characterization of collagen type synthesis by SDS-PAGE (see Materials and methods). The gels were imaged autoradiographically (right panel) to detect 14C radioactivity, and fluorographically (left panel) to detect both 14C and 3H radioactivity. Equal amounts of protein (left) or 3H radiolabel (right) were loaded. In both panels, collagen a chains represented the most predominant labeled proteins, with an α1:β2 ratio of approximately 2, consistent with collagen type I synthesis. C, control (polystyrene plastic); P, porous-coated titanium; R, rough-textured titanium.

Rough-textured and porous-coated titanium surfaces yielded similar results. On the other hand, with regard to the type of collagen being synthesized, no apparent difference was detected between cells on plastic and on titanium disks; both appeared to synthesize primarily type I collagen (Fig. 5B), with the ratio being approximately 2.

Matrix calcification

To investigate further if mineralization, a seminal feature of osteoblasts, was influenced by the surface morphology of titanium disks, the cultures were 1) labeled with [45Ca]Cl2 for 24 h to quantify the amount of mineral formed within one day, 2) stained with Alizarin Red to monitor the amount of stainable mineral produced during the entire culture period, or 3) assayed for ALP activity. The results shown in Fig. 6 indicate that both rough-textured and porous-coated titanium surfaces significantly enhanced mineralization in calvarial osteoblast cultures compared to conventional tissue culture plastic surfaces. Incorporation of 45Ca, Alizarin Red staining, and ALP activity (Fig. 6A-C) were all significantly higher in cells grown on titanium disks. Interestingly, the pattern of mineralization also differed considerably between the various cultures. As shown in the autoradiograms in Fig. 7A, the control culture maintained on glass coverslip showed spotty and sparse accumulation of 45Ca, whereas intense, tightly packed foci of calcification were clearly evident in the titanium cultures. In particular, the culture on the porous-coated titanium surface showed more evenly distributed calcification compared to that on the rough-textured surface. The autoradiographic observations agreed essentially with those obtained by Alizarin Red staining (Fig. 7B), which were more difficult to visualize clearly because of the opaque nature of the titanium disks.

Fig. 6.

Characterization of calcification by calvarial osteoblasts plated on plastic or titanium. Cells were plated at an initial plating density of 1.5 ×105 per well or disk and maintained for 6 days. The cultures were (A) labeled with 1 AzCi ml-1 of [45Ca]Cl2 for 24 h to quantify the mineral formed in one day, (B) stained with Alizarin Red to monitor the stainable mineral produced during the entire culture period, or (C) assayed for alkaline phosphatase (ALP) activity. Protein was estimated using the Pierce BCA reagent. (A) 45Ca incorporation. Cells were harvested at day 6, and 45Ca incorporation determined as described in Materials and methods and expressed as a percentage of control (mean ± S.E.M. of 12 wells). 45Ca incorporation by cells grown on titanium disks (both rough and porous) was significantly higher (**, P<0.001) compared to cells grown on plastic. (Note: to examine whether titanium itself might contribute to the difference in 45Ca incorporation, parallel analyses were carried out with cells cultured on polished, smooth (commercial satin-grade) titanium discs, that revealed no significant difference compared with the controls on plastic (data not shown)). (B) Alizarin Red staining. On day 6, cells were stained with Alizarin Red, washed and dye content determined by absorbance at 609 nm (see Materials and methods). Calcification in cells grown on plastic was significantly lower than in cells grown on porous (*, P<0.05) or rough titanium disks (**, P<0.001; porous versus rough: P<0.001). Results represent the mean ± S.D. of 12 wells or disks. (C) ALP activity. On day 6, the cultures were extracted and assayed for ALP as described in Materials and methods. The results showed that ALP activities (units mg-1 protein) in cells grown on titanium disks were significantly higher than those in cells cultured on plastic (**, P<0.001 for porous and rough surfaces versus plastic; for porous compared to rough titanium disks, mean values showed no significant difference). Numbers represent mean ± S.D. for 9 wells or titanium disks.

Fig. 6.

Characterization of calcification by calvarial osteoblasts plated on plastic or titanium. Cells were plated at an initial plating density of 1.5 ×105 per well or disk and maintained for 6 days. The cultures were (A) labeled with 1 AzCi ml-1 of [45Ca]Cl2 for 24 h to quantify the mineral formed in one day, (B) stained with Alizarin Red to monitor the stainable mineral produced during the entire culture period, or (C) assayed for alkaline phosphatase (ALP) activity. Protein was estimated using the Pierce BCA reagent. (A) 45Ca incorporation. Cells were harvested at day 6, and 45Ca incorporation determined as described in Materials and methods and expressed as a percentage of control (mean ± S.E.M. of 12 wells). 45Ca incorporation by cells grown on titanium disks (both rough and porous) was significantly higher (**, P<0.001) compared to cells grown on plastic. (Note: to examine whether titanium itself might contribute to the difference in 45Ca incorporation, parallel analyses were carried out with cells cultured on polished, smooth (commercial satin-grade) titanium discs, that revealed no significant difference compared with the controls on plastic (data not shown)). (B) Alizarin Red staining. On day 6, cells were stained with Alizarin Red, washed and dye content determined by absorbance at 609 nm (see Materials and methods). Calcification in cells grown on plastic was significantly lower than in cells grown on porous (*, P<0.05) or rough titanium disks (**, P<0.001; porous versus rough: P<0.001). Results represent the mean ± S.D. of 12 wells or disks. (C) ALP activity. On day 6, the cultures were extracted and assayed for ALP as described in Materials and methods. The results showed that ALP activities (units mg-1 protein) in cells grown on titanium disks were significantly higher than those in cells cultured on plastic (**, P<0.001 for porous and rough surfaces versus plastic; for porous compared to rough titanium disks, mean values showed no significant difference). Numbers represent mean ± S.D. for 9 wells or titanium disks.

Fig. 7.

Visualization of calcification in calvarial osteoblast cultures plated on glass coverslip, rough-textured titanium and porous-coated titanium. (A) Autoradiography of 45Ca incorporation in day 6 cultures. From left to right: (a) glass coverslip, (b) rough-textured titanium and (c) porous-coated titanium. Note the more intense and different patterns of 45Ca accumulation by cells plated on titanium. (B) Alizarin Red staining. Top row, porous-coated titanium; bottom row, rough-textured titanium; left to right, culture days 2, 4 and 6. Note the dark red staining for mineral, similar in pattern to those seen by autoradiography.

Fig. 7.

Visualization of calcification in calvarial osteoblast cultures plated on glass coverslip, rough-textured titanium and porous-coated titanium. (A) Autoradiography of 45Ca incorporation in day 6 cultures. From left to right: (a) glass coverslip, (b) rough-textured titanium and (c) porous-coated titanium. Note the more intense and different patterns of 45Ca accumulation by cells plated on titanium. (B) Alizarin Red staining. Top row, porous-coated titanium; bottom row, rough-textured titanium; left to right, culture days 2, 4 and 6. Note the dark red staining for mineral, similar in pattern to those seen by autoradiography.

The primary goal of this investigation was to test the hypothesis that the surface physical characteristics of titanium implants determine their integration into bony tissues. The results obtained here clearly show that bone cells adhere securely onto titanium surfaces, and that rough-textured and porous-coated titanium surfaces enhance both the synthesis and mineralization of the extracellular matrix. These findings strongly suggest that porous or rough titanium implant surfaces may act like “natural” substrata to permit microscopie tissue/cell ingrowth to improve clinical implant fixation.

Several other investigations have also focused on cellular interactions with metallic surfaces. Itakura et al. (1988) observed that crystal aluminum oxide and titanium disks with smooth surfaces (roughness=0.3 /jm) did not inhibit ALP activity and calcium deposition by 16-day cultures of osteogenic MC3T3-E1 cells when compared to the results obtained with Thermanox plastic cover slips. These workers presented cellular metabolism leading to mineralization as the most sensitive indicator of osteocompatibility. In addition, Goldring et al. (1990) observed an increase in type I collagen synthesis in human dermal and bone cells grown on titanium and titanium-containing alloy compared to tissue culture plastic. In similar studies using osteoblasts from neonatal rat calvaria cultured on orthopaedic implant materials (such as Co-Cr-Mo, Ti-6A1-4V), synthesis of collagenous proteins per cell was comparable on all materials (Puleo et al. 1990). Golijanin and Bernard (1988) have also analyzed mouse embryonic calvarial cells cultured on titanium-coated Petri dishes and reported excellent cellular adhesion. Thus, these studies clearly indicate that titanium surfaces function as excellent substrata for the growth and proliferation of bone cells. Our study, showing that both the synthesis and mineralization of the bone extracellular matrix are enhanced upon culturing on rough-textured and porous-coated (but not on smooth) titanium surfaces, indicate further that the physical properties of the titanium surface contribute additionally to the “natural” functioning of the bone cells. Taken together, these observations strongly suggest that the physical properties of the titanium surface may need to be modified (rough or porous instead of smooth) in order to achieve higher levels of calcium deposition or ALP activity. Our previous study (Groessner-Schreiber and Tuan, 1991) and the results presented here, taken together, indicate that porous-coated or rough-textured titanium surfaces are indeed biocompatible with osteoblasts or bone-derived cells in maintaining biosynthesis of the extracellular matrix. Interestingly, it has been reported that in vivo implantations of identically structured and sized cobaltchrome (Co-Cr) alloy and titanium porous-coated cylinders into human cancellous bone result in little ingrowth into the former and excellent ingrowth into the latter (Hofmann et al. 1990). On the other hand, it has been shown that both titanium and Co-Cr alloys appear to allow closer and stronger tissue cell binding than synthetic polymers or bioglass, perhaps due to the relatively high surface energy values that promote tenacious binding of intermediary glycoproteins and colonizing cells (Barth et al. 1985). As the initial stage of collagen bonding is thought to be critical to the development of a bone bond (Davies and Matsuda, 1988), by stimulating collagen biosynthesis, the titanium surface may thus simulate a “natural environment” for bone cells, resulting in enhancement of cellular adhesion and proliferation, and the continued elaboration of additional extracellular matrix.

In conclusion, our results indicate that the physical characteristics of both rough-textured and porous-coated titanium surfaces substantially enhance the mineralization process by bone cells. This effect is most likely to be of mechanistic significance in the stability of titanium-based orthopaedic and dental implants.

The authors thank David Kreitzer for excellent technical assistance, Jennie Platt for preparation of the manuscript and Biomet, Inc. for supplying the titanium disks. This work is supported in part by grants from the NIH (HD 15822, HD 21355), March of Dimes (1-1146); USDA (88-37200-3746 and 90-37200-5265), the Orthopaedic Research and Education Foundation, and the Deutsche Forschungsgemeinschaft (Fellowship to B.G.-S.)

Albrektsson
,
T.
(
1985
).
The response of bone to titanium implants
.
CRC Crit. Rev. Biocompat
.
1
,
53
84
.
Albrektsson
,
T.
,
Brinemark
,
P-L
,
Hansson
,
H-A.
and
Lindstrom
,
J.
(
1981
).
Osseointegrated titanium implants: Requirements for ensuring a long-lasting, direct bone-to-implant anchorage in man
.
Acta Orthop. Scand
.
52
,
155
170
.
Barth
,
E.
,
Roenningen
,
H.
and
Solheim
,
L. F.
(
1985
).
Comparison of ceramic and titanium implants in cats
.
Acta Orthop. Scand
.
56
,
491495
.
Brânemark
,
P-I.
(
1983
).
Osseointegration and its experimental background
.
J. Prosthet. Dent
.
50
,
399
410
.
Crowninshield
,
R.
(
1988
).
An overview of prosthetic materials for fixation
.
Clin. Orthop. Rel. Res
.
235
,
166
172
.
Davies
,
J. E.
and
Matsuda
,
T.
(
1988
).
Extracellular matrix production by osteoblasts on bioactive substrata in vitro
.
Scann. Microsc
.
2
,
1445
1452
.
Gerstenfeld
,
L. C.
,
Chipman
,
S. D.
,
Kelly
,
C.
,
Hodgens
,
K. J.
,
Lee
,
D. D.
and
Landis
,
W. J.
(
1988
).
Collagen expression, ultrastructural assembly, and mineralization in cultures of chicken embryo osteoblasts
.
J. Cell Biol
.
106
,
979
989
.
Goldring
,
S. R.
,
Flannery
,
M. S.
,
Petrison
,
K. K
,,
Evins
,
A. E.
and
Jasty
,
M. J.
(
1990
).
Evaluation of connective tissue cell responses to orthopaedic implant materials
.
Connect. Tiss. Res
.
24
,
77
81
.
Golijanin
,
L.
and
Bernard
,
G.
(
1988
).
Biocompatibility of implant metals in bone tissue culture
.
J. Dent. Res
.
67
,
367
.
Gristina
,
A. G.
(
1987
).
Biomaterial-centered infection: Microbial adhesion versus tissue integration
.
Science
237
,
1588
1595
.
Groessner-Schreiber
,
B.
,
Kreitzer
,
D.
and
Tuan
,
R. S.
(
1991
).
Bone cell response to hydroxyapatite-coated titanium surfaces in vitro
.
Sem. Arthropl
.
2
,
260
267
.
Groessner-Schreiber
,
B.
and
Tuan
,
R. S.
(
1991
).
Titanium surfaces enhance mineralization in calvarial osteoblasts in vitro
.
Trans. Orthop. Res. Soc
.
16
,
34
.
Hanker
,
J. S.
and
Glammara
,
B. L.
(
1988
).
Biomaterials and biomedical devices
.
Science
242
,
885
892
.
Hofmann
,
A. A.
,
Bachus
,
K. N.
and
Bloebaum
,
R. D.
(
1990
).
In vivo implantation of identically structured and sized titanium and cobalt chrome alloy porous coated cylinders implanted into human cancellous bone
.
Trans. Orthop. Res. Soc
.
15
,
204
.
Itakura
,
Y.
,
Kosugi
,
A.
,
Sudo
,
H.
,
Yamamoto
,
S.
and
Kumegawa
,
M. J.
(
1988
).
Development of a new system for evaluating the biocompatibility of implant materials using an osteogenic cell line (MC3T3-E1)
.
Biomed. Mater. Res
.
22
,
613
622
.
Kasemo
,
B.
and
Lausmaa
,
J.
(
1988
).
Biomaterial and implant surfaces: A surface science approach
.
Int. J. Oral Maxillofac. Imp
.
3
,
247
259
.
Laemmll
,
U.
and
Favre
,
M.
(
1973
).
Maturation of the head of bacteriophage T4. DNA packaging events
.
J. Mol. Biol
.
80
,
575
599
.
Marks
,
S. C.
and
Popoff
,
S. N.
(
1988
).
Bone cell biology: The regulation of development, structure, and function in the skeleton
.
Am. J. Anat
.
183
,
1
44
.
Michaels
,
C. M
,,
Keller
,
J. C.
,
Stanford
,
C. M.
,
Solursh
,
M.
and
MacKenzie
,
I. C.
(
1989
).
In vitro connective tissue cell attachment to cp Ti
.
J. Dent. Res
.
68
,
276
.
Puleo
,
D. A.
,
Molieran
,
L. A.
,
Bizios
,
R.
and
Martino
,
L. J.
(
1990
).
Collagen synthesis by osteoblasts cultured on orthopaedic implant materials
.
Fed. Proc. Fed. Amer. Socs Exp. Biol
.
4
,
A1049
.
Thomas
,
J. A.
,
Kay
,
J. F.
,
Cook
,
S. D.
and
Jarcho
,
M.
(
1987
).
The effect of surface macrotexture and hydroxyapatite coating on the mechanical strengths and histological profiles of titanium implant materials
.
J. Biomed. Mater. Res
.
21
,
1395
1413
.
Tuan
,
R. S.
and
Lynch
,
M.
(
1983
).
Effect of experimentally induced Osteoblast cultures on titanium surfaces 217 calcium deficiency of the developmental expression of collagen types in chick embryonic skeleton
.
Develop. Biol
.
100
,
374
386
.
Tuan
,
R. S.
,
Turchi
,
D. M.
and
Kreitzer
,
D. S.
(
1991
).
Polylysine stimulation of ectopic cartilage formation
.
Cells Mater
.
1
,
157
170
.
Williams
,
D. F.
(
1981
).
Titanium and titanium alloys
.
In Biocompatibility of Orthopaedic Implants, vol. 1
(ed.
D.
Williams
), pp.
9
44
.
Boca Raton
:
CRC Press Inc
.
Wong
,
G. L.
and
Cohn
,
D. V.
(
1974
).
Separation of parathyroid hormone and calcitonin sensitive cells from nonresponsive bone cells
.
Nature
252
,
712
715
.