The proliferation and phenotypic expression of human osteoblasts on tantalum metal
David M. Findlay, , a, b, Katie Welldona, Gerald J. Atkinsa, Donald W Howiea, b, Andrew C. W. Zannettinoc and Dennis Bobynd
a Department of Orthopaedics and Trauma, University of Adelaide, Adelaide, South Australia, Australia
b Department of Orthopaedics and Trauma, Royal Adelaide Hospital, Level 4, Bice Building, Adelaide, South Australia, Australia
c Division of Haematology, Institute of Medical and Veterinary Science, Adelaide, South Australia, Australia
d Montreal General Hospital and McGill University, Montreal, Que., Canada
Received 7 July 2003; accepted 7 September 2003. ; Available online 14 November 2003.
Biomaterials
Volume 25, Issue 12 , May 2004, Pages 2215-2227
Abstract
Tantalum (Ta) is increasingly used in orthopaedics, although there is a paucity of information on the interaction of human osteoblasts with this material. We investigated the ability of Ta to support the growth and function of normal human osteoblast-like cells (NHBC). Cell responses to polished and textured Ta discs were compared with responses to other common orthopaedic metals, titanium and cobalt-chromium alloy, and tissue culture plastic.
No consistent differences, that could be attributed to the different metal substrates or to the surface texture, were found in several measured parameters. Attachment of NHBC to each substrate was similar, as was cell morphology, as determined by confocal microscopy. Cell proliferation was slightly faster on plastic than on Ta at 3 days, but by 7 days neither the absolute cell numbers, nor the number of cell divisions, was different between Ta and the other substrates. No consistent, substrate-dependent differences were seen in the expression of a number of mRNA species corresponding to the pro-osteoclastic or the osteogenic activity of osteoblasts. No substrate-dependent differences were seen in the extent of in vitro mineralisation by NHBC. These results indicate that Ta is a good substrate for the attachment, growth and differentiated function of human osteoblasts.
Author Keywords: Tantalum; Human osteoblasts; Cell attachment and proliferation; Gene expression; Mineralisation
Article Outline
1. Introduction
Normal human osteoblast-like cells (NHBC) can be derived from cancellous bone of human donors. NHBC display the expected properties of osteoblasts (OB), in terms of the profile of genes expressed and mineralisation in vitro under appropriate culture conditions [1]. Osteoblasts exist in two functional states; firstly, they are capable of an anabolic, bone-forming activity, which involves the expression of bone matrix proteins, including type-I collagen (COL-1) I, osteocalcin (OCN), and bone sialoprotein (BSP)-1, and the laying down of hydroxyapatite mineral in this organic scaffold [2]. Secondly, cells of the osteoblast lineage perform an essential catabolic role in the bone resorption process by recruiting and promoting the differentiation of osteoclast (OC) precursor cells and activating mature osteoclasts [3], at sites targeted for bone resorption. The differentiation of OC precursor cells is dependent on osteoblastic expression of an OC differentiation factor, termed RANKL, which interacts with a receptor termed RANK on the OC precursor cells. RANKL is a membrane-associated member of the tumour necrosis factor (TNF) ligand family [4]. A natural secreted antagonist of RANKL, termed osteoprotegerin (OPG), binds to RANKL, and inhibits both the formation and the activation of OCs [5]. It is now clear that the local RANKL/OPG ratio determines the strength of the osteoclastogenic signal. Sub-fractionation of NHBC by FACS, on the basis of their expression of the bone marrow stromal precursor marker, STRO-1 yields the less mature STRO-1bright cells with high proliferative potential and the more mature osteogenic STRO-1dull cells [6]. Both cell populations express core binding factor alpha (CBFA)-1/Runx2, a transcription factor essential for expression of the osteoblast) [7 and 8]. A number of additional skeletally active molecules, such as interleukin (IL)-11, have been shown to have pleiotropic roles in osteoblast function, with effects on both the anabolic and catabolic activities of osteoblasts [9 and 10].
It is well recognised that interaction of cells with their substrate is fundamental to the processes of cellular differentiation and function. Thus the extracellular matrix (ECM) has been shown to influence aspects of cellular activity including cell architecture, movement, gene expression and responses of cells to external stimuli [11]. Indeed, we, and others have reported that the ECM components of bone can modulate the differentiation state of osteoblast-like cells in vitro. In particular, we showed that growth of osteoblasts on a type I collagen matrix produced changes consistent with a more differentiated osteoblast phenotype, compared with growth of cells on tissue culture plastic [12]. ECM-mediated effects included changes in cell morphology [12], altered expression of membrane-bound tyrosine phosphatases [13], and changes in gene expression [14 and 15] and signal transduction [16]. The effects of cell substrate on cell behaviour are not limited to biological molecules, but have also been shown with a range of biomaterials [17 and 18]. Relevant to this report, tantalum (Ta) metal has been shown to support tissue ingrowth in a number of orthopaedic implant contexts [19, 20 and 21]. However, osteoblast interaction with tantalum, and in particular its effects on the phenotype and function of human osteoblasts, remain to be investigated in detail.
Ta is an elemental metal that has recently gained interest for a variety of orthopaedic applications. For example, porous "trabecular" Ta constructs have been shown in animal studies to be excellent scaffolds for bone ingrowth and mechanical attachment [19, 20 and 21]. The primary aim of this study was to investigate the interaction of human OB with Ta in a cell culture model. The study hypothesis was that the cellular response to Ta was similar to the response to the traditional biomaterials, titanium and cobalt-chromium (Co-Cr) alloy. Since the microtexture of biomaterials has been claimed to affect cellular response [22 and 23], cell culture discs of each metal were prepared with both polished and textured surfaces.
2. Materials and methods
Ta was compared with the response to the traditional metallic biomaterials, CP titanium (CP Ti) and Co-Cr alloy. Each of the three metals was prepared in the form of solid discs of 22 mm diameter. For each metal, two groups of discs were prepared, one with a smooth surface treatment and one with a microtextured surface treatment. The smooth surface treatment involved progressive polishing to a buff finish using standard commercial techniques. For the Ta discs, the microtextured surface treatment was created by exposing polished discs to the same chemical vapour deposition (CVD) process used in the manufacture of trabecular metal [19]. To establish a microtexture on the polished Ti and Co-Cr discs, they were subjected to heat treatments in a vacuum furnace as would be used for the bonding of porous coatings to an implant substrate. This involved diffusion bonding for Ti and sintering temperatures for Co-Cr. The purpose of the heat treatments was to alter the polished surface of the discs by thermal etching. The metal substrates were all compared with standard tissue culture plastic.
2.1. Surface characterisation of discs by atomic force microscopy (AFM)
The Ta, Ti and Co-Cr cell culture discs were examined with AFM to provide qualitative and quantitative information on the polished and microtextured surfaces.
2.2. Cell culture
Human osteoblast-like (bone) cells (NHBC) from three different individuals were grown from trabecular bone samples obtained at joint replacement surgery, or during bone marrow donation from the iliac crest, and processed for culture as previously described [6]. Briefly, cells were cultured in
-MEM (Flow Laboratories, Irvine, Scotland) with 10% FCS, 0.2
-glutamine and ascorbate 2-phosphate at 37°C/5%CO2 in air in a humidified incubator [6]. All experiments were performed on first passage cells, which were enzymically removed from dishes using collagenase, dispase, and trypsin and plated onto tissue culture plastic or 22 mm diameter metal discs, in 12-well dishes. In order to perform all the experiments it was necessary to use the discs several times, which necessitated washing in non-ionic detergent (Extran MA, Merck Pty Ltd., Kilsyth, Australia) and autoclaving. Preliminary experiments were performed to ensure that washed discs gave results no different from virgin discs.
2.3. Rate and extent of cell attachment
The rate and extent of cell attachment was measured by quantitating attached cells as a fraction of total cells added at 30 min, 1 and 2 h. Each of the six metal surfaces and the tissue culture plastic substrate were divided into quadrants with a wax pen, into each of which was placed 50
l of cell suspension containing 2×104 cells. At each time point, unattached cells were washed off in phosphate buffered saline (PBS) with vigorous pipetting, and remaining cells were then stained with crystal violet, and quantitated by measuring the OD570 of the cell lysates, which correlates closely with cell number, as we have previously described [24].
2.4. Use of confocal microscopy to assess cell morphology
Cells (4×104) were seeded in a 400
l droplet onto each of the metal discs, which were then incubated in a humid chamber at 37°C for 1 h. After cell attachment, discs were flooded with 2 ml of media. As a control, cells were also seeded into 8-chamber plastic slides at an equivalent density to those on the metal discs. Cells were incubated overnight, then washed twice with serum-free
-MEM and fixed with 4% paraformaldehyde for 20 min. The discs were then washed 3 times with PBS before being blocked for 1 h with 1 ml of 2.5% goat serum in PBS. After washing with PBS as above, the cells were permeabilised with 1 ml of 0.5% (v/v) Triton X-100 for 5 min on ice. Cells were washed again and incubated with 150
l of phalloidin-TRITC (Sigma Chemical Co., St. Louis, MO, USA) at 10
g/ml for 1 h at room temperature in the dark. Cells were then washed and incubated with 0.8 mg/ml 4′, 6-diamidine-2′-phenylindole dihydrochloride (DAPI, Roche Diagnostics, Castle Hill, NSW, Australia) in PBS for 15 min at 37°C. After washing the cells again in PBS they were fixed with 1 ml of FACS fix (PBS+10% formalin, 20%
-glucose and 0.2% sodium azide) for 10 min at room temperature, covered with coverslips and sealed with nail polish. Images were captured using the BioRad Radiance confocal miscoscope (Bio-Rad Microscience Ltd., UK) equipped with three lasers, Argon ion 488 nm (14 mW); Green HeNe 543 nm (1.5 mW): Red Diode 637 nm (5 mW) outputs and Olympus IX70 inverted microscope. The objective used was a 40× UPLAPO with NA=1.5 water. Phalloidin was excited with Green HeNe 543 nm laser line and the emission was viewed through a long pass barrier filter (E570P). The image data were stored on a CD for further analysis using Confocal Assistant® software (Todd Clarke Brelje, USA).
2.5. Proliferation rate of cells
Cell proliferation was measured at days 3 and 7, using two different methods: manual cell counting and staining with the cell permeant dye, carboxyfluorescein succinimidyl ester (CFSE), which partitions equally between daughter cells at each cell division [25]. For measurement of cell number by manual cell counting, cells were removed from the various surfaces with trypsin and washed with complete media. An improved Neubauer chamber was used to count cells, in triplicate. Cell proliferation was also measured by irreversibly labelling NHBC with CFSE, as we have previously described [26]. Cells were harvested and resuspended in 1 ml PBS/0.1% BSA, at room temperature. CFSE was then added to the cells and incubated at 37°C for 10 min. Staining was quenched by adding ice-cold media (
-MEM + 10% FCS) and incubating on ice for 10 min. The cells were then washed with media and resuspended at a cell concentration of 2×105 cells/ml. Four hundred microlitres of cell suspension was then pipetted onto the metal discs so that a liquid `bubble' was formed and as much of the surface as possible was covered. The cells were then incubated in a humid chamber at 37°C for 1 h to allow cell attachment to the metal surfaces before wells were flooded with 2 ml of media and further incubated for the times indicated.
Further analysis of the CFSE-labelled NHBC was achieved by staining with the monoclonal antibody STRO-1, as we have previously described [26]. For this, labelled cells were resuspended in 200
l of blocking buffer (10% normal rabbit serum, 5% BSA in PBS, 0.1% sodium azide) containing STRO-1 antibody, or isotype-matched negative control antibody (1A6.12), and were incubated on ice for 1 h. Cells were then washed 3 times with Hank's medium containing 10 m
Hepes and 0.1% w/v BSA (HHF) and incubated with a 1/50 dilution of goat anti-mouse IgM (
-chain specific) phycoerythrin (PE) secondary antibody (Southern Biotechnology Associates, Birmingham, AL) for 45 min on ice. The cells were again washed 3 times with HHF and fixed with 400
l of FACS fix. Fluorescence-activated cell sorting (FACS) analysis, using a FACStarPLUS flow cytometer (Becton Dickinson, Sunnyvale, CA, USA), was used to determine the number of cell doublings that had occurred. Data generated were analysed using ModFit LT 2 software or Winmdi software (Becton Dickinson, Sunnyvale, CA, and USA), as we have described previously [26].
2.6. Expression of osteogenic and osteoclastogenic genes
To determine the expression of mRNA species that correspond to the pro-osteoclastogenic and the osteogenic activity of osteoblasts, cells were lysed using 1 ml per 1×106 cells of TRIzol reagent (Life Technologies Inc., Gaithersburg, MD, USA). Total RNA was prepared from the dissolved cells, as per manufacturer's instructions.
First strand complementary DNA (cDNA) was synthesised from 1
g of total RNA from each sample, using a cDNA synthesis kit, as per manufacturer's instructions (Promega Corp., Madison, WI, USA). cDNA was then amplified by PCR to generate products corresponding to mRNA encoding human gene products listed in Table 1. The 20
l amplification mixture contained 1 U of AmpliTaq Gold DNA polymerase (Perkin Elmer, Norwalk, CT, USA), 100 ng each of the 5′ and 3′ primers, 0.2 m
dNTPs (Pharmacia Biotech, Uppsala, Sweden), 1.5 m
MgCl2, 2
l 10× reaction buffer, and sterile DEPC-H2O. PCR was performed for 23 cycles for GAPDH and 30-35 cycles for other primer pairs, such that all products could be assayed in the exponential phase of the amplification curve, in a thermal cycler (Corbett Research, Melbourne, Vic., Australia). After an initial step at 95°C for 9 min to activate the polymerase, each cycle consisted of 1 min of denaturation at 94°C, 1 min of annealing at the temperatures indicated in Table 1, and 1 min of extension at 72°C. This was followed by an additional extension step at 72°C for 1 min. Human sequence-specific oligonucleotide primers, designed on the basis of published sequences, were obtained from Life Technologies, Gaithersburg, MD, USA. Primer sequences and predicted PCR product sizes are shown in Table 1. Amplification products were resolved by electrophoresis on a 2% w/v agarose gel and post stained with SYBR-1 Gold (Molecular Probes, Eugene, OR, USA). The relative amounts of the PCR products were determined by quantitating the intensity of bands using a FluorImager/Typhoon and ImageQuant software (Molecular Dynamics, Sunnyvale, CA, USA). Amplified products corresponding to OCN, OPG, RANKL, BSP-1, IL-11, TNF-
, COL-1 and CBFA-1 mRNA are represented as a ratio of the respective PCR product/glyceraldehyde 3-phosphate dehydrogenase (GAPDH) PCR product. To show that there were no false-positive results, PCR reactions were carried out on reaction mixtures to which no cDNA was added [27].
Table 1. Human sequence-specific oligonucleotide primers, designed on the basis of published sequences, and predicted PCR product sizes
Key: S, sense sequence; AS, antisense sequence; BP, base pairs.
2.7. Expression of secreted OPG
Secreted OPG levels in culture supernatants, in media harvested from cells grown for 7 days on each substrate in the proliferation experiments, were measured by an enzyme-linked immunoassay (ELISA). Nunc Maxisorp 96-well ELISA plates were coated with 2
g/ml capture antibody (monoclonal anti-human OPG Ab, MAB8051 R&D Systems Inc Minneapolis, MN, USA), sealed and incubated overnight at room temperature. Plates were washed 5 times with wash buffer (phosphate buffered saline (PBS) pH 7.4 with 0.05% Tween 20) and then incubated with 100
l/well blocking buffer (PBS pH 7.4 with 1% BSA) for 1 h. Recombinant human OPG (Peprotech, Rocky Hill, NJ, USA) standards were prepared by serial dilution of a 1 mg/ml stock in media (
-MEM+10% FCS), and these and culture supernatants (diluted 1:10) were incubated on the plate for 2 h at room temperature before washing 3 times in wash buffer. The detection antibody (biotinylated anti-human OPG Ab, Cat. No. BAF805, R&D Systems Inc.), diluted in TBS pH 7.3 containing 0.1% BSA, was then added to each well and incubated for 2 h at room temperature. After washing, streptavidin conjugated to horseradish peroxidase (Pierce, Rockford, IL, USA) diluted in blocking buffer was added to each well and incubated for 20 min at room temperature. Substrate solution was added to each of the wells and allowed to develop for 15 mins. The reaction was stopped using 0.5
sulphuric acid and the OD450nm was measured. Concentrations of the test samples were then calculated using a standard curve.
2.8. Mineralisation in vitro
To determine the ability of NHBC to form mineralised matrix, a method reported previously [16] was modified to accommodate the use of the metal discs. The donor NHBs were cultured in triplicate (8×104 cells/disc) on the various surfaces in
-MEM supplemented with 20% FCS,
-glutamine (2 m
),
-ascorbate 2-phosphate (100
), dexamethasone sodium phosphate (10−8
), KH2PO4 (1.8 m
), and HEPES (10 m
) at 37°C, 5% CO2. Medium was replaced twice weekly and incubation continued for periods of up to 7 weeks before measurement of cell layer-associated Ca2+ levels.
2.9. Measurement of calcium levels
Calcium levels in the cell layer were determined from triplicate cultures at 2, 4, and 7 weeks [6]. The cultures were first washed 3 times with Ca2+- and Mg2+-free PBS and calcium was then extracted using 0.6
HCl. Calcium standards were made using a 2 m
and 1 m
CaCl2 stock to create a standard curve. Standards and supernatants from cells grown on the various substrates (each 25
l) at the different time points were placed in a 96 well plates. Calcium levels were measured in a colorimetric assay using the cresolphthalein method, as per manufacturer's instructions (TRACE Laboratories, Melbourne, VIC, Australia), and the absorbance at 570 nm was measured on a MR7000 microplate reader (Dynatech Laboratories, Guernsey, Channel Islands).
3. Results
3.1. Surface characterisation of discs by AFM
AFM images of the six different cell culture disc surfaces (polished and microtextured for each of the three metals) are illustrated in Figs. 1A-F. Roughness measurements of the AFM information yielded root mean square (RMS) values of the surface topographies, as shown in Table 2. The CVD process resulted in a substantially rougher surface on the Ta discs compared with the polished Ta surface (1.12
m versus 0.29
m). For the Ti and Co-Cr discs, there were smaller differences in overall surface roughness caused by the diffusion bonding and sintering heat treatments.
Fig. 1. Atomic force micrographs. Discs were examined with AFM to provide qualitative and quantitative information on the surface topography of the polished and microtextured metal surfaces. (A) TaP, polished tantalum; (B) TaCVD, chemical vapour deposition Ta; (C) TiP, polished titanium; (D) TiDB, diffusion bonded Ti; (E) CoCrP, polished cobalt chrome; (F) CoCrDB, diffusion bonded cobalt chrome.
Table 2. RMS values for polished versus microtextured surfaces
3.2. Cell attachment to discs
3.2.1. Cell morphology on tantalum
Confocal microscopy showed that the human osteoblast-like cells attached and spread on each of the substrates. Interestingly, the morphology of the cells was not obviously different when adherent to any of the three metals and the surface texture of the metals did not markedly influence cell adherence or shape (Fig. 2). In all cases, cells assumed the elongated, fibroblastic appearance typical of osteoblast-like cells plated onto tissue culture plastic.
Fig. 2. Confocal microscopy of human osteoblast-like cells plated onto the substrates indicated. Cells on the various substrates were stained with phalloidin and DAPI to enable fluorescent imaging of the actin cytoskeleton or nuclei, respectively. Microphotographs are of typical fields and were obtained, as described in the Methods, at 40× magnification. Images of cells from one donor show normal cell morphology and indicate that the nuclei were not apoptotic. (A) TaP, polished tantalum; (B) TaCVD, chemical vapour deposition Ta; (C) TiP, polished titanium; (D) TiDB, diffusion bonded Ti; (E) CoCrP, polished cobalt chrome; (F) CoCrDB, diffusion bonded cobalt chrome; (G) Pl, tissue culture polystyrene plastic. Cells from two other donors were morphologically similar.
3.2.2. Cell attachment kinetics
Experiments were performed using NHBC from three donors. The rate of attachment to each of the substrates was tested, with observations made at 30 min, 1 and 2 h. The results for cells from one donor show that cell attachment to each of the substrates occurred rapidly, with approximately 58%, 70% and 80% of the input cells attached to tissue culture plastic by 30 min, 1 and 2 h, respectively (Fig. 3). Attachment to each of the other substrates was at least as fast as on plastic and this was true for cells from each of the donors tested. By 2 h, no consistent differences were seen between the number of cells adherent to any of the substrates, and there was no evidence that attachment rate or extent was dependent on the surface characteristics of the metal substrates. The results therefore indicate that, in the serum replete conditions used for these experiments, human osteoblast-like cells have a high avidity for Ta, which is similar to that for Ti or CoCr, as well as tissue culture plastic.
Fig. 3. Attachment of human osteoblast-like cells to tantalum and other substrates. Cells (2×104) from one donor were plated onto each of the substrates and allowed to adhere for 30, 60 or 120 min. Cell attachment at each time was quantitated as described in the Methods. Results are means±SEM of three determinations. The results indicate similar rates of attachment of cells onto each of the substrates, with approximately 80% of cells adherent after 120 min. TiP, polished titanium; TiDB, diffusion bonded Ti; CoCrP, polished cobalt chrome; CoCrDB, diffusion bonded cobalt chrome; TaP, polished tantalum; TaCVD, chemical vapour deposition Ta; Pl, tissue culture polystyrene plastic. Results obtained for cells from two other donors gave qualitatively similar results.
3.3. Cell proliferation
Cell proliferation was followed for 7 days, quantitating both cell number and number of cell doublings at day 3 and day 7. Fig. 4 shows proliferation, in terms of the number of cells on each of the substrates, for cells derived from one donor (Fig. 4A) and pooled data from experiments with 3 donors ( Fig. 4B). Fig. 4 also indicates the cell number in the presence of colchicine, a cytoskeletal disruptor that prevents cell division, which indicates the input cell number. It can be seen that cell number increased approximately 4-fold over a 7 day culture, and this was not different for cells on any of the substrates. In addition, there was no evidence that proliferation rate was dependent on the surface characteristics of the metal substrates. The results therefore indicated that, in the serum replete conditions used for these experiments, human osteoblast-like cells can attach and grow on polished and microtextured Ta, as well as Ti and CoCr, and can do so to a similar extent as on tissue culture plastic.
Fig. 4. Proliferation of human osteoblast-like cells to tantalum and other substrates. Cells (8×104) were plated onto each of the substrates and incubated on the various substrates for 3 days or 7 days. Relative cell numbers at each time were quantitated as described in the Methods. Results are means ±SEM of three determinations and are shown for cells from one donor (A) and for the pooled data of cells from three donors (B). TiP, polished titanium; TiDB, diffusion bonded Ti; CoCrP, polished cobalt chrome; CoCrDB, diffusion bonded cobalt chrome; TaP, polished tantalum; TaCVD, chemical vapour deposition Ta; Pl, tissue culture polystyrene plastic; Col, colchicine added to cells on plastic to represent non-proliferating cells.
In order to more sensitively determine the effects of the different substrates on cell growth, we adapted a system that enabled the simultaneous tracking of cell division and changes in the expression of the pre-osteoblast cell surface marker, STRO-1, as we have recently published [26]. The cell permeant fluorescent dye carboxyfluorescein succinimidyl ester (CFSE) allows the number of divisions of a labelled cell population to be tracked by flow cytometry, as daughter cells each receive half of the CFSE of the parent cell [25]. A significant advantage of using this technique is that cell growth, in relation to changes in phenotype, can be monitored in specific sub-populations, whereas more conventional techniques measure only the average number of cell divisions. Cells were simultaneously stained for STRO-1 expression to determine whether the substrates might influence the degree of maturation of human osteoblasts and whether there might be substrate dependent changes in proliferation rate within sub-populations of these cells.
Fig. 5 shows the result of an experiment using cells derived from one donor, of the data derived by CFSE/STRO-1 staining, after growth on plastic, polished Ta or Ta CVD for 7 days. At day 3, cells were distributed between populations that had undergone 0, 1 or 2 divisions (not shown) and by day 7 cells had undergone 0-5 populations (Fig. 5). It was found that Ta metal in either form delayed cell division slightly at both day 3 and 7. There were no striking differences in the number of cell divisions, based on STRO-1 status. Thus, it appeared that, while Ta reduced slightly the entry of cells into cell cycle, it did not influence the distribution of cells between less mature (STRO-1+) and more mature (STRO-1−) populations. The percentage of cells that had undergone each number of doublings was quantitated and the data from three donors was pooled. The pooled data for day 3 is shown in Fig. 6A and for day 7 in Fig. 6B and indicate a slightly greater percentage of cells in the parental (no divisions) pool for cells on Ta than for cells on plastic, and a correspondingly greater number of cells on plastic having divided once. Although seen consistently, these differences for Ta, compared with plastic, did not reach significance. No important differences were found when the data for cells on each of the substrates was plotted (not shown). Thus, neither the absolute cell numbers, nor the number of cell divisions, was significantly different on any of the substrates by day 7 of culture.
Fig. 5. Cell division of human osteoblast-like cells on TaP and TaCVD, compared with tissue culture plastic. Cells (2×106) were stained with CFSE, as described in the Methods, and incubated on the various substrates for 7 days. Cells were then stained with the antibody, STRO-1 and FACS analysis was performed to determine the percentages of STRO-1+ and STRO-1− cells, respectively, that remained undivided (
) or had undergone 1, 2, 3, 4 or 5 divisions (histograms from right to left) in each panel. The pooled data for cells from three donors grown on Ta P, Ta CVD and Pl is presented in Fig. 6B. TaP, polished tantalum; TaCVD, chemical vapour deposition Ta; Pl, tissue culture polystyrene plastic.
Fig. 6. Cell division of human osteoblast-like cells on TaP and TaCVD, compared with tissue culture plastic. Cells (2×106) were stained with CFSE, as described in the Methods, and incubated on the various substrates for (A) 3 or (B) 7 days. Cells were then stained with STRO-1 antibody and FACS analysis was performed to determine the percentages of STRO-1+ and STRO-1− cells, respectively, that remained undivided (parental) or had undergone 1 (D1), 2 (D2), etc., divisions. (A) Means±SD of the pooled quantitated data for cells from three donors, at day 3. (B) Means±SD of the pooled quantitated data for cells from three donors, at day 7. TaP, polished tantalum; TaCVD, chemical vapour deposition Ta; Pl, tissue culture polystyrene plastic.
3.4. Expression of osteogenic and osteoclastogenic genes
The expression of the following genes was investigated using RT/PCR: The osteoblast transcription factor CBFA1; the osteoblast extracellular matrix proteins, COL-1 and OCN; the osteoblast cytokine products IL-11, TNF-
, RANKL, and OPG, the natural antagonist of RANKL. These genes were expressed differently by cells of different donors plated on different substrates for 3 or 7 days. However, there were no consistent, substrate-dependent differences in the expression of these genes. In particular, gene expression of cells on the Ta substrates was very similar to that seen on tissue culture plastic. Data for cells derived from one donor, plated on the different substrates for 7 days, are shown in Fig. 7A-H. OPG protein levels in the media collected from cells plated on the different substrates for 7 days were determined by ELISA. There were no consistent or significant differences between substrates for this parameter, which was typically between 0.5 and 0.8 ng/ml (data not shown).
Fig. 7. Gene expression of human osteoblast-like cells on Ta and other substrates. Cells were grown for 7 days on each of the substrates, before extraction of RNA, as described in the Methods. RNA was then subjected to semiquantitative RT/PCR analysis to determine the expression of the indicated mRNA species, relative to cells cultured on plastic. Results were quantitated by assigning the PCR product obtained from cells on plastic the value of 1, and expressing all other values relative to 1. TiP, polished titanium; TiDB, diffusion bonded Ti; CoCrP, polished cobalt chrome; CoCrDB, diffusion bonded cobalt chrome; TaP, polished tantalum; TaCVD, chemical vapour deposition Ta; Pl, tissue culture polystyrene plastic.
3.5. In vitro mineralisation potential of human OB on metal substrates
NHBC from the three donors tested were found to mineralise at very different rates, as shown in Fig. 8A, which indicates the rate of accumulation of cell layer calcium over 7 weeks, for cells grown on plastic. The mineralisation data for cells derived from one donor, which accumulated most calcium in the cell layer, are shown in Fig. 8B. Small differences in the extent of mineralisation at 7 weeks were seen for cells on the different substrates and growth on the two forms of Ta resulted in the greatest accumulation, although this difference did not achieve significance.
Fig. 8. Mineralisation of the cell layer of human osteoblast-like cells on Ta and other substrates. Cells (8×104) from each of the donors were grown on each substrate for 7 weeks, as described in the Methods. Cell layer calcium was measured at the times indicated. (A) The time course of mineralisation for cells from each donor grown on plastic. (B) Time course of mineralisation for cells from one donor on each of the substrates. Results are means±SEM of three determinations. TiP, polished titanium; TiDB, diffusion bonded Ti; CoCrP, polished cobalt chrome; CoCrDB, diffusion bonded cobalt chrome; TaP, polished tantalum; TaCVD, chemical vapour deposition Ta; Pl, tissue culture polystyrene plastic.
4. Discussion
Tantalum (Ta) is a metal that has recently gained interest for a variety of orthopaedic implant applications because of its formulation in a porous trabecular-like structure that is well suited for mechanical attachment by the ingrowth of bone [19 and 20]. The porous form of Ta reportedly offers several advantages over conventional porous implant materials by its uniformity and structural continuity, strength to weight ratio, low structural stiffness, high porosity, and high coefficient of friction [28]. Although animal studies have indicated good bioactivity of Ta, similar to other common orthopaedic metals [29], the interaction of osteoblasts with Ta, and in particular the effects of Ta on the phenotype and function of human osteoblasts, has not previously been investigated. This would seem important in view of the very high porosity and surface area of the new porous Ta implant biomaterial.
Many studies have used clonal osteoblast-like cell lines to study the effect of biomaterials on osteoblasts, whereas our approach was to perform studies in primary human osteoblasts derived from multiple human donors. The former approach ensures less variability in the results; however, it is our experience that human osteoblasts from different donors do show individual differences and we considered that experiments designed to test the biocompatibility of biomaterials should take this variability into account. Although there was clearly donor-to-donor variability, the results of this study indicate that Ta is a suitable material for human osteoblast growth and differentiated function. This is supported by the initial cell attachment to Ta, and cell morphology on Ta, being not different from tissue culture plastic, which represents the experimental "gold standard" for investigation of osteoblast biology in vitro. In addition, similar growth rate of cells on Ta as on the other substrates tested, and similar expression of a number of genes associated with osteoblastic function, was observed. In the present study, no important substrate dependent differences were recorded, although various donor-dependent differences were seen, particularly at the level of expression of some mRNA species. Investigation of osteoblast behaviour on Ta CVD was important because this microtextured form of the metal is used in porous implant applications [19]. Interestingly, cells behaved similarly on this surface as on polished Ta or plastic, suggesting that Ta CVD should allow faithful expression of the osteoblast phenotype when implanted surgically.
The results reported for the present studies are largely consistent with other studies of osteoblasts on orthopaedic implant metals. Similar results to those reported here for Ti and CoCr were found when responses of neonatal rat calvarial osteoblasts to a variety of orthopaedic implant materials were compared in vitro. Cell attachment, proliferation, and collagen synthesis were similar on 316L stainless steel, Ti-6Al-4V, Co-Cr-Mo, PMMA, borosilicate glass, and tissue culture polystyrene substrates, while hydroxyapatite altered cell adhesion and growth [30]. Our results are also consistent with those reported for the growth of rat calvarial osteoblasts on two forms of titanium (commercially pure Ti and Ti-6Al-4 V alloy), compared with tissue culture plastic [31]. In that work, phenotypic expression of the bone markers, alkaline phosphatase and osteocalcin were not different between the substrates, nor was the degree of cell layer mineralisation achieved in the cultures. Spyrou et al. [32] found that the human SaOS-2 osteosarcoma cells, described as osteoblast-like, expressed a number of cytokines in a substrate-dependent manner when plated onto several variations of Ti and Ti alloys. In contrast, OPG expression was very similar on each of the substrates and on tissue culture plastic. Somewhat different results from ours were reported from a recent study of the human fetal osteoblast cell line, 1.19, which displayed lower growth rates on CoCrMo and stainless steel than on tissue culture plastic, whereas cells on titanium grew at an identical rate to those on plastic [33]. Measures of cell differentiation showed only small differences in this model, although osteocalcin production was significantly greater on titanium than on plastic, after particular times in culture. As in the present study, the differences in cell behaviour did not relate to the surface roughness of the substrates. This might have been expected, with the possible exception of the Ta CVD substrate, because all the surfaces were relatively smooth whether polished or heat-treated. Previous in vitro studies have shown that osteoblasts are more responsive to irregular surfaces than smooth surfaces (RMS <0.5
m) but only if the RMS roughness is greater than about 1
m [34, 35, 36, 37, 38 and 39]. The Ta CVD surface possessed a surface roughness that was just at this threshold.
A further explanation for the lack of difference in cell behaviour in the present report is that, despite differences in chemistry and topography, the substrates were conditioned, both by the serum proteins used in the media and by the cells themselves producing extracellular matrix proteins that in turn govern the osteoblastic differentiation [11, 12, 13, 14, 15 and 40]. This suggestion is perhaps supported by a report that compared the attachment and proliferation of rat fibroblasts and UMR 106.01 osteoblast-like cells to biomaterials, including Ti 318 alloy and CoCrMo alloy. This study found significant substrate-dependent effects on the growth of fibroblasts, so that growth on Ti and CoCr was reduced by more than one third. In contrast, the osteoblast-like cells were essentially unaffected by any of the substrates. In addition, although major effects on morphology and behaviour of cells, including osteoblasts, due to the surface roughness of the substrate, have been reported [41 and 42], these effects are clearly dependent on the degree of roughness and, again, osteoblasts may be more tolerant of surface texture than other cells. For example, cell morphology, orientation, proliferation and adhesion of human gingival epithelial cells in primary or secondary culture were shown to be dependent on the texture of the titanium surface, whereas, in the same study, no such differences were observed for maxillar osteoblast-like cells [43].
The findings of the present study relate to human osteoblast-like cells plated onto solid Ta and other metals commonly used in orthopaedic devices. As indicated above, Ta has the unique property, compared with other orthopaedic metals, of being suitable for the production of porous trabecular-like structures using CVD techniques. It will be of great interest to perform similar studies on these three-dimensional porous Ta materials, now being used in orthopaedic applications, since evidence to date suggests excellent replication of in vivo cell behaviour when cells of various types are cultured in this material (for example, [44]). It will also be important to consider the effect of particulate forms of Ta on human osteoblasts and other bone cells. To date, there is evidence that high concentrations of Ta particulates can decrease the viability of osteoblast-like cells in culture [45]. We [44 and 46], and others [45] have obtained abundant evidence that wear particles generated from prosthetic materials, and the dissolved ions of some metals, have important bioactivities. Importantly, some of these effects lead to bone erosion around metal implants, and eventually a loss of fixation in the bone [47 and 48].
To conclude, this study found minor differences between cells from different donors. However, attachment of cells to Ta was similar to that on plastic and the other orthopedic substrates, as was cell morphology. The cell proliferation rate was slightly faster on plastic than on Ta but no consistent, substrate-dependent differences were seen in gene expression in the human osteoblasts. No substrate-dependent differences were seen in the extent of in vitro mineralisation by NHBC. These results indicate that solid Ta is a good substrate for the attachment, growth and differentiated function of human osteoblasts.
Acknowledgements
This work was supported by a grant from Zimmer Inc. GJA was supported by the National Health and Medical Research Council of Australia.
References
1. S. Gronthos, S.E. Graves, S. Ohta and P.J. Simmons, The STRO-1+ fraction of adult human bone marrow contains the osteogenic precursors. Blood 84 (1994), pp. 4164-4173. Abstract-MEDLINE | Abstract-Elsevier BIOBASE | Abstract-EMBASE
2. Lian JB, Stein GS. Osteoblast biology. In: Marcus R, Feldman D, Kelsey J, editors. Osteoporosis. San Diego: Academic Press; 1996. p. 23-59.
3. T.J. Martin and K.W. Ng, Mechanisms by which cells of the osteoblast lineage control osteoclast formation and activity. J Cell Biochem 56 (1994), pp. 357-366. Abstract-MEDLINE | Abstract-EMBASE
4. L.C. Hofbauer, S. Khosla, C.R. Dunstan, D.L. Lacey, W.J. Boyle and B.L. Riggs, The roles of osteoprotegerin and osteoprotegerin ligand in the paracrine regulation of bone resorption. J Bone Miner Res 15 (2000), pp. 2-12. Abstract-MEDLINE | Abstract-EMBASE
5. H. Yasuda, N. Shima, N. Nakagawa, S.I. Mochizuki, K. Yano, N. Fujise, Y. Sato, M. Goto, K. Yamaguchi, M. Kuriyama, T. Kanno, A. Murakami, E. Tsuda, T. Morinaga and K. Higashio, Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): a mechanism by which OPG/OCIF inhibits osteoclastogenesis in vitro. Endocrinology 139 (1998), pp. 1329-1337. Abstract-MEDLINE | Abstract-EMBASE | Full Text via CrossRef
6. S. Gronthos, A.C. Zannettino, S.E. Graves, S. Ohta, S.J. Hay and P.J. Simmons, Differential cell surface expression of the STRO-1 and alkaline phosphatase antigens on discrete developmental stages in primary cultures of human bone cells. J Bone Miner Res 14 (1999), pp. 47-56. Abstract-MEDLINE | Abstract-EMBASE
7. T. Komori, H. Yagi, S. Nomura, A. Yamaguchi, K. Sasaki, K. Deguchi, Y. Shimizu, R.T. Bronson, Y.H. Gao, M. Inada, M. Sato, R. Okamoto, Y. Kitamura, S. Yoshiki and T. Kishimoto, Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89 (1997), pp. 755-764. Abstract
8. F. Otto, A.P. Thornell, T. Crompton, A. Denzel, K.C. Gilmour, I.R. Rosewell, G.W. Stamp, R.S. Beddington, S. Mundlos, B.R. Olsen, P.B. Selby and M.J. Owen, Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 89 (1997), pp. 765-771. Abstract
9. Y. Takeuchi, S. Watanabe, G. Ishii, S. Takeda, K. Nakayama, S. Fukumoto, Y. Kaneta, D. Inoue, T. Matsumoto, K. Harigaya and T. Fujita, Interleukin-11 as a stimulatory factor for bone formation prevents bone loss with advancing age in mice. J Biol Chem 277 (2002), pp. 49011-49018. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Elsevier BIOBASE | Full Text via CrossRef
10. S.G. Shaughnessy, K.J. Walton, P. Deschamps, M. Butcher and S.M. Beaudin, Neutralization of interleukin-11 activity decreases osteoclast formation and increases cancellous bone volume in ovariectomized mice. Cytokine 20 (2002), pp. 78-85. Abstract | Abstract + References | PDF (146 K)
11. J.C. Adams and F.M. Watt, Regulation of development and differentiation by the extracellular matrix. Development 117 (1993), pp. 1183-1198. Abstract-MEDLINE | Abstract-EMBASE
12. K. Traianedes, K.W. Ng, T.J. Martin and D.M. Findlay, Cell substratum modulates responses of preosteoblasts to retinoic acid. J Cell Physiol 157 (1993), pp. 243-252. Abstract-MEDLINE | Abstract-EMBASE
13. M.C. Southey, D.M. Findlay and B.E. Kemp, Regulation of membrane-associated tyrosine phosphatases in UMR 106.06 osteoblast-like cells. Biochem J 305 (1995), pp. 485-490. Abstract-MEDLINE | Abstract-Elsevier BIOBASE | Abstract-EMBASE
14. K. Traianedes, D.M. Findlay, T.J. Martin and M.T. Gillespie, Modulation of the signal recognition particle 54-kDa subunit (SRP54) in rat preosteoblasts by the extracellular matrix. J Biol Chem 270 (1995), pp. 20891-20894. Abstract-MEDLINE | Abstract-Elsevier BIOBASE | Abstract-EMBASE
15. K. Traianedes, T.J. Martin and D.M. Findlay, Regulation of osteopontin expression by type I collagen in preosteoblastic UMR201 cells. Connect Tissue Res 34 (1996), pp. 63-74. Abstract-MEDLINE
16. S. Celic, P.J. Chilco, J.D. Zajac, T.J. Martin and D.M. Findlay, A type I collagen substrate increases PTH/PTHrP receptor mRNA expression and suppresses PTHrP mRNA expression in UMR106-06 osteoblast-like cells. J Endocrinol 150 (1996), pp. 299-308. Abstract-Elsevier BIOBASE | Abstract-MEDLINE
17. H. Zreiqat and C.R. Howlett, Titanium substrata composition influences osteoblastic phenotype: in vitro study. J Biomed Mater Res 47 (1999), pp. 360-366. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex | Full Text via CrossRef
18. H. Zreiqat, P. Evans and C.R. Howlett, Effect of surface chemical modification of bioceramic on phenotype of human bone-derived cells. J Biomed Mater Res 44 (1999), pp. 389-396. Abstract-Compendex | Abstract-MEDLINE | Abstract-EMBASE | Full Text via CrossRef
19. J.D. Bobyn, G.J. Stackpool, S.A. Hacking, M. Tanzer and J.J. Krygier, Characteristics of bone ingrowth and interface mechanics of a new porous tantalum biomaterial. J Bone Jt Surg Br 81 (1999), pp. 907-914. Abstract-EMBASE | Abstract-MEDLINE | Full Text via CrossRef
20. J.D. Bobyn, K.K. Toh, S.A. Hacking, M. Tanzer and J.J. Krygier, Tissue response to porous tantalum acetabular cups: a canine model. J Arthroplasty 14 (1999), pp. 347-354. Abstract-EMBASE | Abstract-MEDLINE
21. S.A. Hacking, J.D. Bobyn, K. Toh, M. Tanzer and J.J. Krygier, Fibrous tissue ingrowth and attachment to porous tantalum. J Biomed Mater Res 52 (2000), pp. 631-638. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex | Full Text via CrossRef
22. M.E. Hogan, D.H. Degaetano and K.L. Klomparens, Effects of substrate texture and curvature on the morphology of cultured cells. J Electron Microsc Technol 18 (1991), pp. 106-116. Abstract-MEDLINE | Abstract-EMBASE
23. S.A. Hacking, M. Tanzer, E.J. Harvey, J.J. Krygier and J.D. Bobyn, Relative contributions of chemistry and topography to the osseointegration of hydroxyapatite coatings. Clin Orthop 405 (2002), pp. 24-38. Abstract-MEDLINE | Abstract-EMBASE | Full Text via CrossRef
24. A. Evdokiou, S. Bouralexis, G.J. Atkins, F. Chai, S. Hay, M. Clayer and D.M. Findlay, Chemotherapeutic agents sensitize osteogenic sarcoma cells, but not normal human bone cells, to Apo2L/TRAIL-induced apoptosis. Int J Cancer 99 (2002), pp. 491-504. Abstract-EMBASE | Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Full Text via CrossRef
25. Lyons AB, Hasbold J, Hodgkin PD. Flow cytometric analysis of cell division history using dilution of carboxyfluorescein diacetate succinimidyl ester, a stably integrated fluorescent probe. In: Methods in cell biology, vol. 63. New York: Academic Press; 2001.
26. Atkins GJ, Kostakis P, Pan B, Farrugia A, Gronthos S, Evdokiou A, Harrison K, Findaly DM. Zannettino ACW: RANKL expression is related to the differentiation state of human osteoblasts. J Bone Miner Res 2003;89:206-14.
27. G.J. Atkins, D.R. Haynes, S.E. Graves, A. Evdokiou, S. Hay, S. Bouralexis and D.M. Findlay, Expression of osteoclast differentiation signals by stromal elements of giant cell tumors. J Bone Miner Res 15 (2000), pp. 640-649. Abstract-MEDLINE | Abstract-EMBASE
28. R. Cohen, A porous tantalum trabecular metal: basic science. Am J Orthop 31 (2002), pp. 216-217. Abstract-MEDLINE
29. H. Matsuno, A. Yokoyama, F. Watari, M. Uo and T. Kawasaki, Biocompatibility and osteogenesis of refractory metal implants, titanium, hafnium, niobium, tantalum and rhenium. Biomaterials 22 (2001), pp. 1253-1262. SummaryPlus | Full Text + Links | PDF (821 K)
30. D.A. Puleo, L.A. Holleran, R.H. Doremus and R. Bizios, Osteoblast responses to orthopedic implant materials in vitro. J Biomed Mater Res 25 (1991), pp. 711-723. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex
31. J.C. Keller, C.M. Stanford, J.P. Wightman, R.A. Draughn and R. Zaharias, Characterizations of titanium implant surfaces. III. J Biomed Mater Res 28 (1994), pp. 939-946. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex
32. P. Spyrou, S. Papaioannou, G. Hampson, K. Brady, R.M. Palmer and F. McDonald, Cytokine release by osteoblast-like cells cultured on implant discs of varying alloy compositions. Clin Oral Implants Res 13 (2002), pp. 623-630. Abstract-MEDLINE | Full Text via CrossRef
33. C. Hendrich, U. Noth, U. Stahl, F. Merklein, C.P. Rader, N. Schutze, R. Thull, R.S. Tuan and J. Eulert, Testing of skeletal implant surfaces with human fetal osteoblasts. Clin Orthop 394 (2002), pp. 278-289. Abstract-EMBASE | Abstract-MEDLINE | Full Text via CrossRef
34. M. Bigerelle, K. Anselme, B. Noel, I. Ruderman, P. Hardouin and A. Iost, Improvement in the morphology of Ti-based surfaces: a new process to increase in vitro human osteoblast response. Biomaterials 23 (2002), pp. 1563-1577. SummaryPlus | Full Text + Links | PDF (1279 K)
35. K.T. Bowers, J.C. Keller, B.A. Randolph, D.G. Wick and C.M. Michaels, Optimization of surface micromorphology for enhanced osteoblast responses in vitro. Int J Oral Maxillofac Implants 7 (1992), pp. 302-310. Abstract-MEDLINE
36. B. Groessner-Schreiber and R.S. Tuan, Enhanced extracellular matrix production and mineralization by osteoblasts cultured on titanium surfaces in vitro. J Cell Sci 101 (1992), pp. 209-217. Abstract-MEDLINE | Abstract-EMBASE
37. J.C. Keller, Tissue compatibility to different surfaces of dental implants: in vitro studies. Implant Dent 7 (1998), pp. 331-337. Abstract-MEDLINE
38. K. Kieswetter, Z. Schwartz, T.W. Hummert, D.L. Cochran, J. Simpson, D.D. Dean and B.D. Boyan, Surface roughness modulates the local production of growth factors and cytokines by osteoblast-like MG-63 cells. J Biomed Mater Res 32 (1996), pp. 55-63. Abstract-MEDLINE | Full Text via CrossRef
39. Z. Schwartz, C.H. Lohmann, M. Sisk, D.L. Cochran, V.L. Sylvia, J. Simpson, D.D. Dean and B.D. Boyan, Local factor production by MG63 osteoblast-like cells in response to surface roughness and 1,25-(OH)2D3 is mediated via protein kinase C- and protein kinase A-dependent pathways. Biomaterials 22 (2001), pp. 731-741. SummaryPlus | Full Text + Links | PDF (463 K)
40. Gehron Robey P, Boskey AL. The biochemistry of bone. In: Marcus R, Feldman D, Kelsey J, editors. Osteoporosis. San Diego: Academic Press; 1996. p. 23-59.
41. C. Schmidt, D. Kaspar, M.R. Sarkar, L.E. Claes and A.A. Ignatius, A scanning electron microscopy study of human osteoblast morphology on five orthopedic metals. J Biomed Mater Res 63 (2002), pp. 252-261. Abstract-EMBASE | Abstract-MEDLINE | Full Text via CrossRef
42. K. Anselme, M. Bigerelle, B. Noel, A. Iost and P. Hardouin, Effect of grooved titanium substratum on human osteoblastic cell growth. J Biomed Mater Res 60 (2002), pp. 529-540. Abstract-EMBASE | Abstract-MEDLINE | Full Text via CrossRef
43. G. Lauer, M. Wiedmann-Al-Ahmad, J.E. Otten, U. Hubner, R. Schmelzeisen and W. Schilli, The titanium surface texture effects adherence and growth of human gingival keratinocytes and human maxillar osteoblast-like cells in vitro. Biomaterials 22 (2001), pp. 2799-2809. SummaryPlus | Full Text + Links | PDF (669 K)
44. S.D. Rogers, D.W. Howie, S.E. Graves, M.J. Pearcy and D.R. Haynes, In vitro human monocyte response to wear particles of titanium alloy containing vanadium or niobium. J Bone Jt Surg Br 79 (1997), pp. 311-315. Abstract-MEDLINE | Abstract-EMBASE | Full Text via CrossRef
45. T. Sakai, S. Takeda and M. Nakamura, The effects of particulate metals on cell viability of osteoblast-like cells in vitro. Dent Mater J 21 (2002), pp. 133-146. Abstract-MEDLINE
46. D.R. Haynes, S.J. Hay, S.D. Rogers, S. Ohta, D.W. Howie and S.E. Graves, Regulation of bone cells by particle-activated mononuclear phagocytes. J Bone Jt Surg Br 79 (1997), pp. 988-994. Abstract-MEDLINE | Abstract-EMBASE | Full Text via CrossRef
47. H.G. Willert, G.H. Buchhorn and T. Hess, The significance of wear and material fatigue in loosening of hip prostheses. Orthopade 18 (1989), pp. 350-369. Abstract-MEDLINE
48. D.W. Howie, Tissue response in relation to type of wear particles around failed hip arthroplasties. J Arthroplasty 5 (1990), pp. 337-348. Abstract-MEDLINE | Abstract-EMBASE
Corresponding author. Department of Orthopaedics and Trauma, University of Adelaide, Royal Adelaide Hospital, Level 4, Bice Building, , Adelaide, South Australia, , Australia. Tel.: +618-8222-5621; fax: +618-8232-3065
Wyszukiwarka
Podobne podstrony:
The Structure and Heat Treatment of Low Carbon Steel
Jack London The Son of the Wolf and Other Tales of the North
Measurements of the temperature dependent changes of the photometrical and electrical parameters of
The electrochemical and mechanical behavior of passivated an
Mencej The christian and prechristian concept of the master of the wolves
The Luftwaffe And Its War Of Attrition
The Past and Present Ends of History (Hegel, Fukuyama, Kojeve) Krasnodebski
The Ancient and Primitive Rite of Memphis Misraim by Frater Bogomilius 33° 90° 97° IX° XII°
więcej podobnych podstron