Surface engineering of titanium by collagen immobilization. Surface characterization and in vitro and in vivo studies
M. Morra, , a, C. Cassinellia, G. Cascardoa, P. Cahalanb, L. Cahalanb, M. Finic and R. Giardinoc
a Nobil Bio Ricerche, Str.S. Rocco 36, Villafranca d'Asti 14018, Italy
b Percadia, Merrimack, NH, USA
c Experimental Surgery Department Research, Institute Codivilla Putti, Rizzoli, Orthopaedic Institute, Bologna, Italy
Received 8 May 2003; accepted 13 May 2003. ; Available online 29 July 2003.
Abstract
Collagen was covalently linked to the surface of Titanium (Ti) by a surface modification process involving deposition of a thin film from hydrocarbon plasma followed by acrylic acid grafting. The composition and properties of surface-modified Ti were investigated by a number of surface sensitive techniques: XPS, ATR-IR, atomic force microscopy and AFM force-separation curves. In vitro tests were performed to check samples cytotoxicity and the behavior of osteoblast-like SaOS-2 cells. In vivo experiments involved 12 weeks implants in rabbit muscle as general biocompatibility assessment and 1-month implants in rabbit bone to evaluate the effect of surface modification on osteointegration rate. Results of XPS measurements show how surface chemistry is affected throughout each step of the surface modification process, finally leading to a complete and homogeneous collagen overlayer on top of the Ti samples. AFM data clearly display the modification of the surface topography and of the surface area of the samples as a consequence of the grafting and coupling process. AFM force-distance curves show that the interfacial structure responds by shrinking or swelling to variations of ionic force of the surrounding aqueous environment, suggesting that the aqueous interface of the biochemically modified Ti samples has enhanced degrees of freedom as compared to the inorganic surface of plain Ti. As to biological evaluations, the biochemically modified Ti samples are safe in terms of cytotoxicity and in vivo biocompatibility assessment. SaOS-2 cells growth rate is lower on collagen modified surfaces, and no significant difference is detected in terms of alkaline phosphatase production as compared to control Ti. Importantly, implants in rabbit femur show a significant increase of bone growth and bone-to-implant contact in the case of the collagen modified samples, confirming that biochemical modifications of Ti surface can enhance the rate of bone healing as compared to plain Ti.
Author Keywords: Titanium; Dental and orthopedic implants; Collagen; Surface modification; Surface analysis; Atomic force microscope; Osteointegration
Article Outline
1. Introduction
Titanium is the material of choice for load-bearing, bone-contacting devices. The orthopedic and dental implants industry is largely based on the favorable interfacial interaction at the bone-titanium interface, and an extensive literature on osteointegration and related issues exists [1, 2, 3, 4 and 5]. From this body of literature, it is clear that osteointegration is a well-established property of titanium implant surfaces and that the current success rate is satisfactory [5]. Nevertheless, surface modification of titanium is still a very active area of research, because definite interest exists in surface treatments that can induce acceleration of normal bone healing phenomena. An accelerated stable fixation between bone and implant would allow early or immediate loading of the device, with significant implications in terms of decreased patient morbidity, patient physicology and health care costs.
In an excellent review on the bone-implant interface, Puleo and Nancy described three different approaches to the surface modification of bone-contacting titanium devices [6]: physicochemical methods, morphological methods and biochemical methods. Among currently available dental implants, morphological methods of surface modification take the lionshare. In particular, the so-called sandblasted-acid attacked (SLA) and the dual-acid-etched (DAE) surfaces are probably the most advanced and interesting examples of attempts to direct interfacial interactions at the bone-titanium interface by the fine tuning of surface topography [7, 8, 9, 10, 11, 12 and 13].
As suggested by Puleo and Nanci [6], biochemical methods of surface modification offer an alternative or adjunct to physicochemical or morphological methods. Biochemical methods are aimed at the control of the tissue-implant interface by the immobilization and/or delivery of proteins, enzymes, or peptides for the purpose of inducing specific cell and tissue responses [6]. They rely on the current understanding of the biology and biochemistry of cellular function and differentiation and on suitable surface modification techniques. Several interesting examples of biochemical modifications of titanium or metal devices have been presented in the literature [14, 15, 16, 17, 18, 19, 20, 21, 22, 23 and 24]. Some of the obtained results confirm, at least at a laboratory level, the promises of this approach [19].
Among the molecules of direct relevance to the biochemical modification of bone-contacting surfaces, collagen is of definite interest: it is among the main extracellular matrix proteins controlling adhesion of cells of direct relevance to orthopedic applications [25], through the amino acid sequence Arg-Gly-Asp (RGD) it contains. It is commonly used in dental surgery as osteogenic and bone filling material [26 and 27]; in this respect, Gungormous and Kaya have recently shown that heterologous type I collagen provides a more rapid regeneration of bone defects [27]. It notoriously exerts a strong pro-coagulant (hemostatic) activity and stimulates platelets in a unique way [28]. This aspect is of definite interest on the light of recent suggestions on the role of early blood-implant interactions in bone healing [13, 29 and 30]. In the spirit of biochemical modification of implant surfaces [6], implants bearing a surface layer of collagen could be exploited to apply directly to the implant site the properties of this interesting biomolecule.
Collagen coatings on titanium and titanium alloys have been investigated as carriers of biomolecules or to improve cell adhesion [20, 23, 31 and 32]. It has been shown that collagen-coated alloys enhance osteoblast spreading and result in a more rapid formation of focal adhesions and their associated stress fibers as compared to uncoated titanium.
We have discussed the use of thin film deposition from glow discharge plasma for the immobilization of biomolecules on titanium surfaces [33]. In this work, the same approach, involving the Ce(IV)-induced surface grafting of acrylic acid [34] on the hydrocarbon film deposited on titanium, is used to immobilize collagen on titanium surfaces. We believe that this approach can have a very significant relevance in the development of a new generation of biochemically surface-modified dental and orthopedic implants, and the aim of this work is to present the general strategy for surface modification and some in vitro and in vivo findings. In particular, in this paper, we present the results of surface/interfacial analysis performed to clarify the nature of the surface modification process and of the coating structure. Then, both in vitro and in vivo results are discussed. In particular, experiments are aimed at the evaluation of the general safety of surface-modified Ti and of the efficacy in the intended application of the surface-modified device. To this end, cytotoxicity testing were performed to show that this surface structure is safe and does not release any cytotoxic compound; in vivo evaluation of the biocompatibility was performed through implants in rabbit muscle [35]. An osteoblast-like cell line (SaOS-2) was used to evaluate the in vitro growth rate and phosphatase alkaline activity on untreated and collagen coupled samples. Finally, 1-month implantation in rabbit femur was performed to check the effect of the coating on the bone healing rate. As recently discussed [24] this animal model and experimental time can give valuable information on bone healing at an early phase. This is directly related to the present goal of researches on surface modification of titanium and osteointegration, that is surface treatments that can induce acceleration of normal bone healing phenomena. As underlined by Puleo and Nancy [6], an accelerated stable fixation between bone and implant would allow early or immediate loading of the device, with significant implications in terms of decreased patient morbidity, patient physicology and health care costs.
2. Experimental
2.1. Chemicals
Titanium foils (0.125 mm thick, 99.7%), acrylic acid, propylene, N-(3-dimethyaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS) were purchased from Sigma Aldrich. For ATR-IR experiments, very thin Ti foils were used (99.98%, 0.025 mm thickness, Sigma Aldrich), to improve contact with the ATR crystal due to the sample flexibility.
Collagen (KNC Semed S collagen powder) was supplied from Kensey Nash (Lot. No. 24903). This is basically type I bovine collagen.
2.2. Surface modification
2.2.1. Plasma deposition
Plasma deposition was performed in a capacitively coupled parallel-plate reactor placed inside a class 10,000 clean room. Samples were located on the water-cooled grounded electrode. Both the reactor and electrodes are made of stainless steel. The reactor volume is about 3 dm3 and the distance between the electrodes 5 cm. The following treatment conditions were used: 1 min air plasma, 50 W, 20 sccm flow rate (cleaning step), followed by 5 min propylene plasma deposition, 50 W, 100 sccm flow rate (flow rate is controlled by a MKS mass flow controller) (deposition step).
A few samples were used for evaluation of the thickness of the acrylic acid grafted layer. To this end, after plasma deposition, a mask made by a stainless steel foil with round holes of 40
m diameter was gently pressed on the titanium sample. The masked sample was placed in the plasma reactor and an etching treatment was performed to remove the plasma-deposited layer from the exposed areas (air plasma, 10 min 100 W).
2.2.2. Acrylic acid grafting
After deposition, treated samples were subjected to acrylic acid grafting. In particular, grafting was performed from a 40% aqueous acrylic acid solution. The solution was gently mixed using a stirrer bar for 2 min. At this point, Argon was gently bubbled in the solution for 10 min, then 3% of an acidic solution of ammonium cerium nitrate was added, without discontinuing Ar. The solution was mixed while bubbling for 2 more minutes, then it was gently poured in test tubes, containing the plasma-deposited samples. Grafting was performed for 45 min, at room temperature. At the end of the grafting reaction, samples were rinsed overnight in water. Plasma deposited, acrylic acid grafted samples will be coded AATi in the rest of this paper.
2.2.3. Collagen coupling
AATi samples were further subjected to collagen coupling. Coupling was performed as follows: AATi samples were immersed in a 0.5% collagen, 1% acetic acid aqueous solution. After 2 h, samples were removed from the solution and rinsed several time in 1% aqueous acetic acid, to remove excess adsorbed collagen. After rinsing, samples were immersed in water containing 0.25% EDC and 0.25% NHS, both from Sigma, and kept overnight in this coupling solution. After coupling, all samples were carefully rinsed with 1% aqueous acetic acid and water, then dried under a hood. Collagen coupled samples will be coded ColTi in the rest of the paper.
2.2.4. Surface characterization
XPS analysis was performed with a Perkin Elmer PHI 5500 ESCA system. The instrument is equipped with a monochromatic X-ray source (Al K
anode) operating at 14 kV and 250 W. The diameter of the analysed spot is approximately 400
m, the base pressure 10−8 Pa. The angle between the electron analyser and the sample surface was 45°. Quantification of elements were accomplished using the software and sensitivity factors supplied by the manufacturer.
2.2.5. ATR-IR
ATR-IR spectra were obtained using a Mattson Research Series Spectrometer. Spectra were obtained using a 4 cm−1 resolution and a 45° Ge crystal. Using this arrangement, the analytical sampling depth in the carbonyl region of the spectrum is about 300 nm. Samples were pressed against the crystal using MiniGrip (Thermo Spectra-Tech). Data presented are averaged from 256 spectra.
2.2.6. AFM analysis
AFM analysis was performed using a Nanoscope III AFM by Digital. Measurements were performed both in air and in aqueous solutions. A conventional silicon nitride tip was used in these experiments. The spring constant of the tip, as evaluated by the producer, is 0.06 N/m. Images were obtained in contact mode. The "Height" data type mode was used, that is data correspond to the change in the piezo height needed to keep the cantilever deflection constant. In order to reduce damage of the sample surface by the cantilever tip, force was constantly monitored and minimized using the instrument control software. Images were constantly monitored to detect effects due to the cantilever tip "plowing" the sample surface. In no case such effects were observed.
Roughness data were calculated from the images using the instrument software. Student unpaired t-tests were performed to determine any differences between data, as reported in the relevant section.
Force-distance curves in aqueous were obtained using the fluid cell supplied by Digital. In the force measurement the sample is moved continuously up and down. Deflection of the cantilever and height position of the sample are recorded. The force is obtained by multiplying the deflection of the cantilever times its spring constant. In this case the spring constant was not measured and the value suggested by the producer was used, but this does not affect the meaning of the data. Measurements were repeated using the same cantilever for the whole set of samples, and repeated to confirm that the obtained values were not affected by wear or other unexpected effects on the tip. Differences between the data obtained on the same sample at different times were not significant, and the same general trend among different samples was detected.
To obtain force versus distance curves, the deflection of the cantilever must be subtracted to the piezo scanner position. The point of zero distance was determined from the linear part of the contact line. Force versus scanner position curves were continuously recorded with a typical frequency of about 1 Hz, the scan length was 200 nm. A minimum of 20 curves in five different regions of the each sample were recorded.
2.3. Biological testing
2.3.1. Materials
Samples used for in vitro and intramuscular implants studies are shown in Fig. 1. They were obtained from 2 mm thick Ti plates (99.7%, Aldrich). The size of the samples is 10×3×2 mm. The photograph was taken after toluidine blue staining of the samples. The control Ti (coded Ti in the following), on the left, did not take the dye, while the collagen coated Ti (ColTi in the following) did, and it looks darker in the photograph. Before toluidine blue staining, samples are undistinguishable. Samples were homogeneously coated on all sides.
Fig. 1. Photograph of the samples used for muscle implants and cell studies. Samples are toluidine blue stained to show the homogenety of the coating on ColTi (on the right).
Fixtures for bone implant studies were obtained from cpTi, grade 2. They are shown in Fig. 2, also in this case after toluidine blue staining, to show the homogeneity of the coating (ColTi is on the right). The fixtures diameter is 2 mm. No pre-treatments aimed at the modification of the surface topography of the screws were performed. That is, from a morphological point of view, the investigated surfaces belong to the "machined" class [13].
Fig. 2. Photograph of the fixtures used for bone implants. Samples are toluidine blue stained to show the homogenety of the coating on ColTi (on the left).
2.3.2. Cytotoxicity testing
Cytotoxicity was evaluated by the direct contact method, using the continuous mouse fibroblasts L-929 cell line. Experimental cell culture medium (SIGMA, Milan) was Minimum Essential Eagle's Medium without
-glutamine, 10% fetal bovine serum, streptomycin (100
g/l), penicillin (10 U/ml), 2 mmoles/l
-glutamine in 250-ml plastic culture flask (Falcon™). Cells were cultured at 37°C in a humidified incubator equilibrated with 5% CO2. Fibroblasts were harvested prior to confluence by means of a sterile trypsin-EDTA solution (0.05 trypsin, 0.02 EDTA in normal Phosphate Buffered Saline, pH 7.4) from the culture flasks, resuspended in the experimental cell culture medium, and diluted to 1×105 cells/ml, 3.5 ml of the cells suspension were seeded into 6-Well tissue culture polystyrene plates (9.6 cm2 of growth area; Falcon™), containing the samples. After 3 days of growth at 37°C, 5% CO2 in a humidified incubator, the following evaluations were performed: (1) Cell death and cell morphology. The cell monolayer around the samples was observed by an inverted microscope (DM IL, Leica). The boundary between the samples and the cells monolayer was carefully controlled, to check if release from the samples resulted in cell death or number reduction. Also, the cell morphology was carefully controlled, and compared with results obtained with the negative control (that is, cells grown into the polystyrene well with a gold ingot inside, same size of the experimental samples). The positive control was obtained from an ingot of a Ni-Cu-Al alloy, same size as the experimental sample. (2) Biosynthetic activity. The widely used MTT Test was performed to evaluate cells health through their biosynthetic activity. This test measures the succinate dehydrogenase (SDH) activity of cells after 72 h contact with the samples. SDH is a key enzyme of the Kreb's cycle (that is, the citric acid cycle), its evaluation by biochemical means is commonly used to check cells health.
Briefly, at the end of 72 h of contact, cells were washed with sterile PBS, then the PBS was replaced with 2 ml/well of MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide)-sodium succinate solution. The cells and MTT solution were incubated at 37°C for 3 h in the incubator. During this time, yellow MTT solution is transformed by the cells mitochondrial dehydrogenase into insoluble blue formazan. By measuring the amount of formazan produced it is possible to measure mitochondrial activity, and, as a consequence, cell viability. At the end of the incubation period, the MTT solution was removed and replaced with 2 ml/well of a 6.25% v/v 0.1 mol/l NaOH in dimethylsulfoxide, in order to dissolve formazan. The wells were swirled for 5 min until the purple color was uniform and the adsorbance was evaluated at 560 nm. The adsorbance values obtained were averaged, and the average was expressed as a percentage of the negative control (which, by definition, is 100%).
2.3.3. Osteoblast-like cell growth and alkaline phosphatase activity
Osteoblast like SaOS-2 cells were used in the cell growth experiments. Experimental cell culture medium (BIOCHROM KG, Berlin) consisted of Minimum Eagle's Medium without
-glutamine, 10% fetal bovine serum, streptomycin (100
g/l), penicillin (100 U/ml), and 2 mmoles/l
-glutamine in 250-ml plastic culture flask (Corning™). Cells were cultured at 37°C in a humidified incubator equilibrated with 5% CO2. Cells were harvested prior to confluence by means of a sterile trypsin-EDTA solution (0.5 trypsin g/l, 0.2 g/l EDTA in normal Phosphate Buffered Saline, pH 7.4), resuspended in the experimental cell culture medium, and diluted to 1×105 cells/ml. For experiments, 5 ml of the cell suspension were seeded into 6-Well tissue culture polystyrene plates (9.6 cm2 of growth area; Falcon™), containing the samples. Experiments were performed in triplicate.
At selected time intervals (1 day, 4 days and 10 days), the number of cells on the sample was evaluated. Briefly, samples were removed from the wells, rinsed with PBS and placed in another multiwell plate. Cells were harvested from the samples surface by trypsin and counted in a hemocytometer.
Measurement of alkaline phosphatase activity was performed after 10 days growth as follows: samples were rinsed with PBS and transferred to a new multiwell plate. One ml of lysing solution (0.2% Triton X-100, Sigma) was added to each sample. After that, lysates were incubated for 2 h at 37°C with 1 ml of the enzyme substrate mixture solution, containing 1 m
MgCl2×6H2O, l.5 m
2-amino-2-methyl-1-propanol and 20 m
p-nitrophenylphosphate [36]. The reaction was stopped with 1 ml well of 1
NaOH. Enzyme activity was measured by reading the samples at a wavelength of 405 nm, using a Tecan Genios microplate reader. Obtained readings subsequently were compared with values measured on a standard curve, obtained by the measurements of standard solution of p-nitrophenol.
The specific alkaline phosphatase activity was obtained by ratioing the previous data to the total protein concentration, obtained by the Bradford method [37] and using a calibration curve based on bovine serum albumin (Sigma) standard solutions. Samples were read at 595 nm using the same microplate reader described above.
2.3.4. In vivo studies
The entire in vivo study was performed following European and Italian Law on animal experimentation and according to the Animal Welfare Assurance No. #A5424-01 by the National Institute of Health (NIH-Rockville, MD, USA). The experimental protocol was sent to the Italian Ministry of Health.
Eleven adult male rabbits weighting 3.250±0.350 were used (Charles River, Calco, Lecco, Italia). Rabbits were acclimated for 10 days and were healthy and free of lameness before admission into the study. Animals were maintained in single cages at a T of 20±0.5°C, humidity rate of 55±10% and water and food ad libitum (Mucedola, Settimo Milanese, Milano, Italia).
Implants were placed in sterile, after shaving and disinfecting with betadine solution. General anesthesia was induced with an i.m. injection of 44 mg/kg ketamine (Ketavet 100, Farmaceutici Gellini SpA, Aprilia Lt, Italy) and 3 mg/kg xylazine (Rompun Bayer AG, Leverkusen, Germany), and assisted ventilation (O2: 1 l/min; N2O: 0.4 l/min; isofluorane: 2.5-3%). In the postoperative period the animals were checked daily, and wounds dressed until taking out of stitches. Antibiotic therapy (Cefazolin, 100 mg/kg) was administered preoperatively, immediately after surgery and after 24 h. Analgesics (metamizole chloride, 50 mg/kg) were prescribed in the immediate postoperative period. The health of the animals was evaluated during the entire study by a veterinarian. At the end of the experimental protocols the animals were sacrificed by pharmacological euthanasia under general anesthesia with intravenous administration of Tanax (Hoechst, Frankfurt am Main, Germany).
2.3.5. Intramuscular implants
Five rabbits were used for the intramuscular implant of 10 Ti and ColTi plates. Two specimens at each animal side were surgically inserted with the long axis parallel to the muscle fibers and at more than 25 mm from the midline and apart from each other as suggested by International Standard Organization [35]. Control plates (uncoated Ti) were implanted on the left side while experimental plates (ColTi) on the right side. A longitudinal incision was made through the skin of the dorso-lumbar region. The skin was separated from the underlying fascia with blunt dissection. Subsequently, the fascia was dissected and a small incision was made into the belly of the muscle. Implantation sites were created by further separation of the muscle fibers using blunt dissection with rounded scissors. After insertion of the implants, the muscle incision was not sutured to avoid suture material side effect. The skin was sutured with non-adsorbable stitches. After 12 weeks animals were sacrificed and the implant with a sufficient (about 2 cm) unaffected surrounding tissue was removed to evaluate local biological response.
2.3.6. Bone implant
Six rabbits were used for bone implantation. The femur middiaphyses was exposed and two defects with a 1.9-mm diameter were drilled at low speed and under continuous saline irrigation in the cortical bone of the right and left femurs. Control uncoated Ti screws were transversally implanted in the left femurs of all rabbits, while ColTi screws were positioned in the right femurs, up to a total of 12 cortical implants for each material. Finally, the skin was sutured in two layers. Four weeks after surgery, the animals were sacrificed. Femurs were removed, cleaned of soft tissues and prepared for histomorphometry.
2.3.7. Histology and histomorphometry
Soft and hard tissues biopsies were fixed in 4% buffered paraformaldeide for 7 days for undecalcified bone processing. Then, samples were dehydrated in graded series of alcohols until the absolute (50%, 75%, 95% and 100% for 24 h at each concentration). Finally they were embedded in methylmethacrylate. Blocks were sectioned along a plane parallel to the long axis of the implants and a series of section of about 100±20
m, spaced 200
m apart, were obtained with a Leica 1600 diamond saw microtome (Leica 1600, Leica SpA, Milano, Italia). They were thinned to about 10±2
m and 30±5
m (muscle and bone, respectively) by an Exact polisher (Bioptica, Milano, Italy). For each sample, three slices were examined, after Ematoxilin-Eosin stainin (muscle implants) and Fast Green-acid Fucsin (bone implants). Istomorphometry was performed by an Axioscop (Carl Zeiss, GmbH, Jena, Germany) connected by a videocamera through a personal computer equipped with the software Kontron KS 300 (Kontron Electronic, GmbH, Eiching bei Munchen, Germany). As far as muscle implant is concerned, the evaluation involved the presence and thickness of the fibrous capsule, inflammation, degeneration, necrosis and the nature of cells at the tissue-implant interface. By histomorphometry, the following parameters were measured: presence of the fibrous capsule around the implant (length of the perimeter occupied by capsule/total sample perimeter ×100) and the capsule thickness. The latter was measured at 16 different locations along the sample perimeter (five locations along the long side, three along the short side of the sample). A rating scale for tissue inflammation was adopted as described by Wataha and coworkers [38].
To evaluate the bone-biomaterial interface and osteointegration, the following histomorphometric measurements were performed [39, 40, 41 and 42]:
Bone-to-implant contact: fractional, linear extent of bone apposed to the implant surface divided by the total surface perimeter of the implant ×100; bone ingrowth (BI): bone amount in the area delimited by the threaded surface and a line connecting the top of the threads, divided by the total thread area×100. It was calculated for all the threads of each implant and was then compared with its mirror image (mirror area) (MA).
Each measurement was taken semiautomatically at an original magnification of ×40 by an experienced blinded investigator.
Data are reported as average±standard deviation.
2.3.8. Statistical analysis
Statistical analysis was performed using SPSS v.10.1 software (SPSS/PC Inc., Chicago, IL). Data are reported as mean±SD at a significance level of p<0.05. After having verified normal distribution and homogeneity of variances, paired Student t-test was done to compare data between Ti and ColTi screws.
3. Results
3.1. XPS analysis
XPS analysis was used to measure the surface composition of coated samples. Results (at %) are reported in Table 1. Considering the untreated sample, the detected value is in good agreement with literature on titanium surfaces in general and, in particular, on surfaces of titanium dental implants [43, 44 and 45]. Briefly, the surface of titanium is covered by a thin (about 4 nm thick) oxide layer, so that the maximum theoretical concentration of Ti on pure titanium is 33%, the rest being oxygen (the most stable oxide is TiO2). Surface contamination from ubiquitous hydrocarbons, together with some residual from the machining steps, introduces a significant amount of carbon and further decreases the concentration of Ti. A huge literature exists on this subject, the value of about 15% Ti detected in this case is indicative of a comparatively clean surface [46].
Table 1. Surface composition (at%) of the different samples, as detected by XPS analysis
The deposition from hydrocarbon plasma masks almost completely the signal from underlying Ti. Some oxygen is included in the overall stoichiometry, as expected for this kind of coatings [33]. Also the acrylic acid grafting step yields figures in agreement with the expected stoichiometry (sample AATi).
The detected surface composition after collagen coupling (sample ColTi) is in good agreement with reference values for collagen, as shown in the table. No signal from the underlying titanium is detected, showing that the surface grafted hydrogel layer and the coupled collagen layer homogeneously cover the substrate. The thickness of the surface layer is obviously bigger than the XPS sampling depth, as expected and as discussed in the section on AFM data.
Also the detailed analysis of C1s peak yielded results in agreement with the expected chemistry. In particular, after acrylic acid grafting a strong component at about 4.5 eV from the main hydrocarbon component is observed (acrylic acid carboxyl groups). After collagen coupling, the main peak is still due to C-C, C-H bonds, while the component at higher binding energy shifts to values typical of amide carbon, as expected from the proteinaceous nature of collagen.
XPS analysis was performed also on samples subjected to
-rays or ethylene oxide sterilization cycles. In no cases significant effects on the surface composition and C1s peak shape were detected.
3.2. ATR-IR
Fig. 3 shows ATR-IR spectra of AATi titanium samples. The quality of the spectra is not very high, due to the difficulty in obtaining good IR spectra of very thin films on IR-opaque metal samples. Spectra were obtained after soaking the samples at two different pH: 3 and 7. In the former case, it is possible to see the peaks due to protonated acrylic acid, in particular the stretching of protonated carboxyl at about 1700 cm−1. After soaking at a higher pH, the protonated peak disappears and it is replaced by the ionized carboxyl peak at about 1560 cm−1. The modification of the nature of the carboxyl group upon soaking at different pH shows that, despite immobilization induced by surface grafting, the polyacrylic acid layer keeps its acid-base activity.
Fig. 3. ATR-IR spectra of AATi samples. The upper spectrum was obtained after soaking the sample in aqueous solution at pH 3, the lower spectrum after soaking at pH 7.
After collagen coupling, bands due to carboxyl groups were replaced by the typical amide bands, as expected.
3.3. AFM
The first information we sought from AFM measurements is an evaluation of the thickness of the grafted layer. As described in the Experimental section, a few patterned samples were obtained by selectively etching the plasma deposited hydrocarbon layer from 40
m wide areas, by air plasma etching through a metallic mask. The effectiveness of the process was confirmed by toluidine blue staining of grafted, etched samples. The unmasked areas did not stain, while the grafted layer in the masked portion of the sample was readily stained by toluidine. Fig. 4 shows a section analysis through one of the holes obtained. Some debris can be seen in hole contour or inside the hole. They could be just due to small pieces of homopolymerized acrylic acid, not completely removed by the cleaning routine. Actually, the genuinely grafted surface, that is the area outside and away from the hole, is much more flat and homogeneous. The figure displays the height profile along the black line shown in the lower left portion of the figure. The arrows markers are used as a reference for the measurements. Section and calculations were performed using the software of the AFM. This measurement allows the evaluation of the thickness of the grafted layer, that is the vertical distance between the two arrows. As shown in the figure, the vertical distance is about 117 nm. Twenty measurements in different areas of the pattern were performed. Calculated thickness average and standard deviation are: 112±16 nm.
Fig. 4. Section analysis of an AATi sample, showing the heigth profile along a line that crosses the hole obtained by masking and etching the plasma-deposited surface. The vertical distance between the arrows is shown in the lower right part of the figure.
As far as surface morphology is concerned, Fig. 5 shows representative 5×5
m fields of view of the different surfaces in air. The vertical scale is 100 nm. In particular, Fig. 5a refers to the Ti sample. The features which are detected are likely due to the processing steps. No significant modification of the overall morphology was detected after plasma deposition, as expected by the conformal nature of plasma-deposited coatings. Fig. 5b shows a typical 5×5
m field of view of the AATi sample surface, the vertical scale is the same as in Fig. 5a. Clearly, the surface morphology is completely different from that of the Ti sample, it is possible to see the "bumped" structure typical of surface-grafted hydrogel surfaces. Main bumps are several hundred nm in size, with smaller bumps on them. The process of collagen coupling leads to a significant modification of the surface morphology, as shown in Fig. 5c. This aspect will be discussed later on. Here it is important to underline that, due to its excellent vertical resolution, the AFM can clearly detect differences between the morphology of the samples surface.
Fig. 5. AFM images showing the surface morphology of: Ti (a); AA Ti (b); ColTi (c). Samples in air.
Previous data were obtained in air. Since the working environment of these samples is aqueous, it is of interest to evaluate the surface morphology when the samples are placed under water or in aqueous media. Fig. 6 shows the surface morphology of the grafted (AATi) and the grafted and collagen coupled sample (ColTi), both of them placed in a 0.001
NaCl solution. In both cases a significant modification of the morphology, as compared to the "dry" state (Figs. 5b and c) is observed. The AATi surface looks smoother, as compared to the morphology in air: bumps can be clearly observed, but, overall, they are not so prominent as in Fig. 5b. An even more marked effect is seen in the case of the collagen coated surface (cf. Fig. 6b to Fig. 5c). Table 2 shows the effect of the environment on the surface roughness, as calculated by the instrument software, of AATi and ColTi samples.
Fig. 6. AFM images showing the surface morphology of: AA Ti (a); ColTi (b). Samples in 0.001
NaCl.
Table 2. Roughness values in air
Notes: Rms (Rq): standard deviation of the Z values (i.e. the values measured in the vertical direction).
Mean roughness (Ra): mean value of the surface relative to the center plane.
Data show mean and standard deviation of five measurements. All differences between different samples in the same environment and identical samples in different environments are statistically significant (p<0.05).
While the morphological analysis can clearly distinguish between the AA grafted and the AA grafted-collagen coupled surface, it is of interest to check whether force-separation curves can offer clues on forces at the aqueous interface. Fig. 7 shows force-separation curves obtained in 0.001
NaCl, using a conventional silicone nitride tip. In the force measurement, the sample is moved continuously up and down, and the AFM tip is brought into contact with the sample. The force exerted by the sample surface on the tip and propagated through the medium is recorded. Fig. 7 shows that the approaching of the tip to the AA-grafted surface is dominated by a monotonically increasing repulsive force. This force is due to the compression of the soft, water swollen hydrogel layer by the cantilever tip. When the tip is retracted, a sawtooth behavior is observed, which reflects the disentanglement of the tip from the "net" of hydrogel polymeric molecules adsorbed to or entangled with the silicon nitride tip [47].
Fig. 7. AFM force-separation curves of: (a) AATi, 0.001
NaCl; (b) ColTi, 0.001
NaCl; (c) ColTi, 0.1
NaCl.
Collagen coupling bears on significant modifications not only at a morphological level, as seen in Fig. 5 and Fig. 6, but also to the interfacial field of forces. In particular, Fig. 7b shows the force distance curve obtained on the ColTi surface. Again, upon approaching, a repulsive force is detected as expected from the hydrophilic nature of this surface. However, the range of the repulsive force decreases significantly after collagen coupling, that is, there is no longer a soft, compressible surface overlayer. In other terms, the collagen interface with water is much more "rigid" than the AATi one. A further modification is observed upon increasing the ionic strength of the solution ( Fig. 7c, obtained in 0.1 NaCl). The range of interaction at the ColTi interface is further reduced, and the curve looks like a typical "hard wall" interaction on rigid materials.
3.4. Cytotoxicity testing
Direct contact between L-929 cells and Ti and ColTi samples did not evoke any adverse effect at the optical microscope level. No cell death or effects on cell morphology were observed, as demonstrated by Fig. 8, which shows the boundary between the ColTi sample and the cell monolayer. Cell density and shape did not differ from that observed in the case of Ti samples or of the negative control. Only in the case of the positive control a significant cell death was observed, as expected.
Fig. 8. Optical microscope photograph showing the results of direct contact cytotoxicity testing of ColTi samples. The L-929 cell monolayer at contact with the sample (whose shadow can be seen on the left, diagonal) shows no adverse effect. Identical results were obtained on Ti and the negative control samples.
Results of MTT tests are reported in Table 3, expressed as % of the negative control, as described in the Experimental section. Only in the case of the positive control Cu-Al-Ni alloy the SDH activity dropped to 29% of negative control activity, while both Ti and ColTi samples show the lack of adverse effects, in agreement with microscope observation.
Table 3. Results of MTT tests
Note: The difference between Ti and ColTi is not significant. The difference between the positive control and both tested samples is statistically significant (p>0.001).
3.5. Osteoblast-like cells growth and alkaline phosphatase activity
The number of SaOS-2 cells on the samples surface as a function of time is shown in Fig. 9. At short time (1 day), no significant difference is detected between the figures measured on the two samples. On the other hand, on the long term, the number of cells, and hence the cell growth rate, is greater on Ti samples than on ColTi ones, as shown by 4 and 10 days data. It must be underlined that this is a genuine effect due to the surface structure; the number of cells on the bottom of the wells containing the ColTi samples was not different from that measured on the bottom of the wells containing the Ti samples.
Fig. 9. Results of SaOS-2 cell growth experiments. The difference between the number of cells on Ti and ColTi surfaces after 4 and 10 days is statistically significant (p>0.001).
Alkaline phosphatase measurements did not show any statistically significant difference between the two samples. The absolute reading was higher in the case of Ti surfaces, but this was obviously due to the higher number of cells, as shown in Fig. 9. When specific activity was calculated by ratioing the measured absorbance to the total amount of proteins, no significant difference was detected. The calculated figures are 4.1±1.2 and 4.3±1.6
p-nitrophenol/mg protein/minute for Ti and ColTi, respectively, in general agreement with the expected order of magnitude [36].
3.6. In vivo experiments
All animals tolerated well surgery and survived without any clinical complication until the final experimental time. Radiography performed at sacrifice confirmed that the samples were well located at the implant sites. At macroscopic level, during animal autopsy, no signs of inflammation, gross infection or tissue damages were observed at each examined experimental site around soft and hard tissue implants.
3.7. Muscle implants
After 12 weeks implants, histological observation did not reveal any remaining inflammatory cell or material debris at the implant site (inflammation score: 0) [38]. Both Ti and ColTi samples were encapsulated by fibrous capsule of varied thickness, as shown in Fig. 10. The observed capsules consisted of dense collagen fibres oriented parallel to the implant interface. There was no evidence of foreign body giant cells, lymphocytes, histiocytes, macrophages, polymorphonuclear leukocytes and monocytes. The mean extension of the capsule was 93.3%±3.1% (Ti) and 98.3%±1.6% (ColTi). The mean thickness of the capsule was 22.6±13
m (Ti) and 32±12.0
m (ColTi). No significant differences between the parameters measured on the two samples were detected.
Fig. 10. Results of intramuscular implant experiments: (a) Ti. A thin fibrous capsule is visible around the material (about 20
m of thickness) in the presence of muscle, adipose tissue and vessels (on the right side) (H&E, 5×). (b) ColTi. A homogeneous fibrous capsule of about 20
m thickness is visible. The dorsal muscle has a normal histological aspect. (H&E, 5×).
3.8. Bone implants
The histological appearance of the implanted screws and bone, after 4 weeks implant, is shown in Fig. 11 and Fig. 12. Newly formed bone was detected around the different screws. Both reticular and lamellar bone was seen inside the screw threads of each screw and on the periosteal cortical surfaces. Direct apposition of bone was observed to both Ti and ColTi screw surfaces, even if in some cases the osteointegration process is still in progress at 30 days in the case of uncoated Ti. Percentages of bone-to-implant contact and bone ingrowth in the threaded screw area and mirror area are shown in Table 4. These results clearly demonstrate that ColTi screws show a higher percent bone contact with the surrounding bone than uncoated Ti. A significant increase of bone to implant contact and bone ingrowth was observed in ColTi versus uncoated Ti (+23.8%, p<0.01; +7.6%, p<0.01, respectively). As expected, no significant differences were observed outside the screw threads in the mirror area (MA) of each group thus demonstrating the homogeneity of the implanted bone. The BI/MA rate, on the contrary, showed a significant increase in the collagen-coated Ti group in comparison with the uncoated Ti one (+6.7%, p<0.005).
Fig. 11. Results of bone implant experiments: (a) Ti. Osteointegration processes are in progress and bone is growing inside the screw threads at 1 month (Fast Green and Acid Fuchsin, 1,25×). (b) ColTi. Bone is directly apposed to the material surface 1 month after surgery. Periosteal bone growth on the external site of the implant is also observable (Fast Green and Acid Fuchsin, 1,25×).
Fig. 12. Results of bone implant experiments: (a) Ti. A detail of Fig. 6a at higher magnification showing that bone is growing in direct contact with the Ti surface (Fast Green and Acid Fuchsin, 10×). (b) ColTi. A detail of Fig. 6b at higher magnification. Neoformed bone inside the screw threads. Bone directly apposed to the material surface (Fast Green and Acid Fuchsin, 5×).
Table 4. Histomorphometric results of experimemntal (ColTi) and control (Ti) screws implanted in rabbit femoral diaphysis at 4 weeks (Mean±SD, n=5 replicates; Paired Student t-test)
4. Discussion
Titanium (Ti) is the material of choice in many medical devices applications. While its good biocompatibility and the excellent strength to weight ratio are difficult to challenge, it is often felt that the fine tuning of interfacial properties of Ti could lead to a new generation of devices endowed with improved performances, as recently reviewed by Puleo and Nancy in the case of bone contacting devices [6]. A fundamental requirement along this line is the development of a reliable surface modification technique, which could be used to immobilize biochemical molecules on the Ti surface in a well controlled and characterized way and which could be scaled-up. We have shown that thin film deposition form hydrocarbon plasma can be a suitable starting point to turn the titanium surface into an organic surface [33], prone to undergo the typical reactions of polymer surfaces. Among them, Ce(IV) promoted grafting of acrylic acid allows to prepare in a reliable and controlled way a carboxyl group-rich surface that can be subjected to further coupling reactions [34].
In this work, plasma deposition-acrylic acid grafting were used to couple collagen to the titanium surface. Surface characterization shows that a collagen layer, thicker than the XPS sampling depth (Table 1), homogeneously covers the substrate surface. Moreover, the surface structure readily responds to the interfacial aqueous environment, as shown by ATR-IR spectra of the grafted hydrogel layer ( Fig. 3) and by force-separation curves at different ionic strength ( Fig. 7).
Concerning, in particular, AFM data, the force upon approach on surfaces bearing "soft" polymer overlayers can be fitted by a simple exponential, as discussed by Butt and coworkers [47]:
F=A exp(−D/ |
(1) |
where D is the tip-sample distance. The amplitude A and the typical decay length
are the fitting parameters. It can be shown [47] that:
|
(2) |
where L0 is the equilibrium thickness of the grafted polymer brush, and
A=50kB TRL0Γ3/2, |
(3) |
where kB is the Boltzmann constant, T the temperature, R is the radius of curvature of the tip and Γ is the surface density of grafting sites. Eqs. (2) and (3) are strictly valid only under a number of assumptions, not all of them completely fulfilled in the present case (for instance, the grafted hydrogel is not an ideal polymer brush). However, taking into account the above caveat, it is of some interest to evaluate figures arising from the experimental curves. In particular, the repulsive force detected upon approaching in Fig. 5a yields an excellent exponential fit (r2=0.990), with calculated A=2.5 n
and
=20.1 nm. Using Eqs. (2) and (3), it is possible to calculate an equilibrium thickness of the polymer brush of 126.5 nm and a grafting density of 7.3×1016 sites/m2, or a distance between grafting sites of 3.7 nm. These figures, albeit subjected to the caveat described above, are very reasonable and in good agreement with the thickness measured in Fig. 4.
Collagen coupling results in a significant increase of the stiffness of the aqueous interface, as shown by the marked reduction of the range of repulsive interaction upon approach. Actually, the exponential fit of the approaching curve shown in Fig. 7b yields
=3.8 nm, to be compared with the
=20.1 nm of the AATi interface. Several reasons can account for this behavior: first of all, neutral water or saline is not a good solvent for collagen, which is actually soluble at acidic pH. Then, collagen is intrinsically stiff. It is of interest to observe that a further significant reduction of the range of interaction at aqueous-ColTi interfaces is obtained upon increase of the ionic strength. This result reflects the collapse of the surface structure upon the reduction of the range of electrostatic forces, as a consequence of the increased ionic strength (the Debye length, which is a measure of the range of electrostatic forces, is about 1 nm in 0.1 NaCl and about 10 nm in 0.001 NaCl). Actually, this behavior is closely mirrored by staining experiments using toluidine blue: AATi or ColTi samples readily uptake the cationic toluidine blue dye. The stained samples are stable in water or in low ionic strength solutions, while they readily release all of the dye upon immersion in high ionic strength solutions and collapse of the surface layers. Dye uptake is readily restored if the same samples are stained again after washing in water or low ionic strength solutions.
Incidentally, the staining test offers a quick tool for the evaluation of the stability of the interfacial layer. Samples can be stained to check integrity and homogeneity, de-stained in high ionic strength solutions, then subjected to, for instance, overnight extraction in a relevant medium, and stained again to check coating integrity. Stability tests performed in this way show that the coating withstands overnight extraction in aqueous surfactants.
As to the biological evaluations we performed, as far as the safety of the surface-modified material is concerned, cytotoxicity tests show the lack of any adverse effect, or the lack of release of unexpected chemicals and by-products of the coupling reaction (Fig. 8, Table 3). Also in vivo general biocompatibility assessment through the muscle implant test [35] shows the absence of inflammatory cells and tissue necrosis at 12 weeks, suggesting the good compatibility of the experimental material ColTi. More in particular, the development of an avascular fibrous capsule around implants secondary to the foreign body reaction is a well-recognised chronic inflammatory response [48]. The usual occurrence when a "non-absorbable" material is implanted consists of transient acute inflammation (less than a week), followed by a slowly developing encapsulation by fibrous connective tissue, with or without a giant cell response, requiring 1 month or longer to develop. The capsule surrounding an implant in species other than rodents is static and permanent [48]. The movement of a metallic implant is traumatic to surrounding soft tissue [48]. The present findings show a fibrous capsule developing at 12 weeks around ColTi samples, which did not differ significantly from the one detected around the control material (Ti). These results are in disagreement with other studies reporting that no fibrous capsule develops in the soft tissue mantles that covered the Ti plates when implanted in the tibia of rabbits [49]. However, when a metallic material is implanted in soft tissues having completely different mechanical properties a fibrous capsule is expected, since the influence of material mechanical properties may be superimposed upon the ungoing biological healing processes. Mechanical factors and edge effects may modify the response to a biomaterial such as implant motion and micromotion that can lead to variations in the fibrous capsule thickness and composition [50 and 51]. The end stage healing response to biomaterials and medical devices is generally a process of fibrous encapsulation that is reported as the minimal response [50]. Because of this, results of the present study and the absence of inflammatory cells and tissue necrosis at 12 weeks show the good compatibility of the experimental materials.
As far as the efficacy of ColTi samples in bone contacting applications is concerned, in vitro studies on osteoblast-like cells growth show a decrease of the growth rate on ColTi as compared to Ti and no significant difference in alkaline phosphatase activity. Geisller and coworkers have shown that collagen I coatings on Ti alloys do indeed enhance primary osteoblast adhesion and a number of related parameters [20]. Mizuno and coworkers have shown that bone marrow cells cultured with type I collagen matrix gels show high alkaline phosphatase activity, collagen synthesis, and form mineralized tissues; furthermore, cells expressed osteocalcin and bone sialoprotein genes [52]. The disagreement between these reports and the present findings, in particular the lack of any detectable effect on alkaline phosphatase activity, can likely be accounted for by the different cell models. In the present study, a continuous, osteoblast-like cell line was investigated, while in the quoted paper [20 and 52] either primary osteoblasts or bone marrow cells were used. Moreover, Windeler and coworkers have shown that among commonly used osteosarcoma cell lines alkaline phosphatase activity can be markedly different on the same substratum [53].
The most important finding arising from the present study is the clear evidence that the biochemical modification of the Ti surface by collagen can indeed affect the in vivo response of bone and increase the healing rate at a fairly early healing phase, at least in the present animal model. As shown in Fig. 11 and Fig. 12 and in Table 4, the collagen coating on titanium significantly accelerates bone-to-implant contact and bone ingrowth. In our measurements, the percentage of bone ingrowth inside the screw threads was compared to that of the mirror thread areas. An inside/outside ratio of less than 1 indicates an impaired bone area in the thread due to adverse reactions to the implanted material or bone resorption secondary to other reasons [39 and 40]. The ratio between bone ingrowth and mirror area is close to 1 for both the tested material, and a significant increase in the amount of bone inside the screw threads of the coated material was observed. In a recent report, Schliephake and coworkers used a collagen coating on Ti as a substrate for covalently linked RGD peptides [23]. Collagen-coated Ti without RGD was one of the controls. No significant differences, in an in vivo model involving dogs, were detected between the effects on osteointegration of uncoated Ti and collagen coated Ti, in disagreement with our present findings. Several reasons can account for this disagreement. Among them, the most important are likely the different animal model and, most of all, the completely different way of applying the collagen layer to the Ti surface.
Concerning the reasons that lead to the accelerated bone healing at ColTi surfaces in the present experiments, for the time being we can just submit some hypothesis. The first one involves the notorious role of collagen in differentiation and behavior of bone cells, bone mineralization and bone healing [20, 27 and 52]. Even if our simple cell model failed to show enhanced alkaline phosphatase activity on ColTi surfaces, this could just be due to the cell line used and/or to the intrinsic limitation of the in vitro environment. It is important to underline that heterologous type I collagen has been shown to enhance healing of bone defects [27]; our results could just mirror the same events, which, in the present case, go on at the three-dimensional interface described in this paper. A second hypothesis could be related to the platelet-activating effect of collagen, and the release of growth factor from activated platelets, as nicely described by Davies and coworkers [13]. In a recent work, Goransson and coworkers used a similar rabbit model and experimental time to check the effect of implants bearing a surface-linked 100 nm-thick layer of blood plasma clot [24]. No differences were detected between coated implants and uncoated ones. Several reasons were underlined in the quoted paper to account for the lack of efficacy. Among them, it was observed that, for a truly amplified wound healing response, the presence of blood components including blood cells carrying their content of growth factors and other stimulatory molecules are necessary. The "philosophy" behind the present treatment, that is the use of a very thin layer of covalently bound collagen, is exactly that of stimulating and activating platelets in order to promote the cascade of healing events [29 and 30]. Ongoing studies are aimed at the clarification of these aspects. For the time being it is important to conclude that this study shows that the biochemical modification of Ti surfaces can affect bone healing rate in vivo; and that the increase of the efficacy of the implant device is obtained by a reliable and well-characterized surface modification process. While, presently, surface modification of dental and orthopaedic implants is largely obtained by topography modifications, we believe that the present and similar ongoing works [6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 and 24] are paving the way to a true biochemical revolution of dental and orthopaedic implant devices surfaces.
5. Conclusions
In summary, the data discussed in this paper offer the following picture of the biochemically modified Ti surface and its aqueous interface: the inorganic Ti surface is replaced by a homogeneous (in the lateral direction) collagen coating; the interfacial, three-dimensional, structure responds to the aqueous environment by expanding or collapsing as a function of the ionic strength. In practice, the immobilized biomolecule and the overall interfacial structure increase the information contact of the surface and the degrees of freedom as compared to the inorganic surface of plain Ti. Ti modified by collagen coupling according to the present process is safe, both at an in vitro and an in vivo level. The growth rate of SaOS-2 cells in vitro is lower on collagen-coated surfaces than on unmodified Ti, and no difference is detected in alkaline phosphatase activity. Most importantly, in vivo experiments in rabbit femur show that a significant increase of bone to implant contact and bone ingrowth is observed on ColTi versus uncoated Ti fixtures in early phases of bone healing, suggesting that this kind of surface modification could accelerate the time to loading of load-bearing, bone-contacting devices.
References
1. R. Adell, U. Leckholm, B. Rockler and P.I. Branemark, A 15-year study of osseointegrated implants in the treatment of edentulous jaw. Int J Oral Surg 10 (1981), pp. 387-416. Abstract-MEDLINE | Abstract-EMBASE
2. T. Albrektsson, The response of bone to titanium implants. CRC Crit Rev Biocomput 1 (1984), pp. 53-84.
3. H.A. Hansson, T. Albrektsson and P.I. Branemark, Structural aspects of the interface between tissue and titanium implants. J Prosthet Dent 50 (1983), pp. 108-111.
6. D.A. Puleo and A. Nanci, Understanding and controlling the bone-implant interface. Biomaterials 20 (1999), pp. 2311-2321. SummaryPlus | Full Text + Links | PDF (972 K)
7. L.F. Cooper, A role for surface topography in creating and maintaining bone at titanium endosseous implants. J Prosthet Dent 84 (2000), pp. 522-534. Abstract | PDF (182 K)
8. J.Y. Martin, Z. Schwartz, T.W. Hummert, D.M. Schraub, J. Simpson, J. Lankford, Jr, D.D. Dean, D.L. Cochran and B.D. Boyan, Effect of titanium surface roughness on proliferation, differentiation and protein synthesis of human osteoblast-like cells (MG63). J Biomed Mater Res 29 (1995), pp. 389-401. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex
9. B.D. Boyan, R. Batzer, K. Kieswetter, Y. Liu, D.L. Cocharan, S. Szmuckler-Moncler, D.D. Dean and Z. Schwartz, Titanium surface roughness alters responsiveness of MG-63 osteoblast-like cells to 1a,25-(OH)2D3. J Biomed Mater Res 39 (1998), pp. 77-85. Abstract-Compendex | Abstract-MEDLINE | Full Text via CrossRef
10. K. Kieswetter, Z. Schwartz, T.W. Hummert, D.L. Cochran, J. Simpson, D.D. Dean and B.D. Boyan, Surface roughness modulate 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
11. B.D. Boyan and Z. Schwartz, Modulation of osteogenesis via implant surface design. In: J.E. Davies, Editor, Bone engineering, em squared, Toronto (2000), pp. 232-239.
12. R.J. Lazzara, Bone response to dual acid etched and machined titanium implant surfaces. In: J.E. Davies, Editor, , em squared, Bone engineering. Toronto (2000), pp. 381-390.
13. J.Y. Park, C.H. Gemmell and J.E. Davies, Platelet interactions with titanium: modulation of platelet activity by surface topography. Biomaterials 22 (2000), pp. 2671-2682.
14. D.A. Puleo, Activity of enzyme immobilized on silanized Co-Cr-Mo. J Biomed Mater Res 29 (1995), pp. 2951-2957.
15. Sumner DR, Turner TM, Purchio AF, Gombotz WR, Urban RM, Galante JO. Enhancement of bone ingrowth by transforming growth factor-
. J Bone Jt Surg 1195;77A:1135-47.
16. D.A. Puleo, Biochemical surface modification of Co-Cr-Mo. Biomaterials 17 (1996), pp. 217-222. SummaryPlus | Full Text + Links | PDF (684 K)
17. D.A. Puleo, Retention of enzymatic activity immobilized on silanized Co-Cr-Mo and Ti-6Al-V. J Biomed Mater Res 37 (1997), pp. 222-228. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex | Full Text via CrossRef
18. A. El-Ghannam, L. Starr and J. Jones, Laminin-5 coating enhances epithelial cell attachment, spreading, and hemidesmosome assembly on Ti-6Al-4 V implant materials in vitro. J Biomed Mater Res 41 (1998), pp. 30-40. Abstract-MEDLINE | Abstract-Compendex | Abstract-EMBASE | Full Text via CrossRef
19. D.M. Ferris, G.D. Moodie, P.M. Dimond, C.W.D. Gioranni, M.G. Ehrlich and R.F. Valentini, RGD-coated titanium implants stimulate increased bone formation in vivo. Biomaterials 20 (1999), pp. 2323-2331. SummaryPlus | Full Text + Links | PDF (420 K)
20. U. Geissler, U. Hempel, C. Wolf, D. Scharnweber, H. Worch and K.W. Wenzel, Collagen type I coating of Ti6Al4 V promotes adhesion of osteoblasts. J Biomed Mater Res 51 (2000), pp. 752-760. Abstract-MEDLINE | Abstract-Compendex
21. D.A. Puleo, R.A. Kissling and M.S. Sheu, A technique to immobilize bioactive proteins, including bone morphogenetic protein-4 (BMP-4), on titanium alloy. Biomaterials 23 (2002), pp. 2079-2087. SummaryPlus | Full Text + Links | PDF (188 K)
22. K. Matsumura, S.H. Hyon, N. Nakajima, C. Peng, H. Iwata and S. Tsutsumi, Adhesion between poly(ethylene-co-vinyl alcohol) (EVA) and titanium. J Biomed Mater Res 60 (2002), pp. 309-315. Abstract-MEDLINE | Abstract-EMBASE | Full Text via CrossRef
23. H. Schliephake, D. Scharnweber, M. Dard, S. Rossler, A. Sewing, J. Meyer and D. Hoogestraat, Effect of RGD peptide coating of titanium implants on periimplant bone formation in the alveolar crest. An experimental pilot study in dogs. Clin Oral Implants Res 13 (2002), pp. 312-319. Abstract-MEDLINE | Full Text via CrossRef
24. A. Goransson, E. Jansson, P. Tengvall and A. Wennerberg, Bone formation after 4 weeks around blood-plasma-modified titanium implants with varying surface topographies: an in vivo study. Biomaterials 24 (2003), pp. 197-205. SummaryPlus | Full Text + Links | PDF (930 K)
25. R.G. Lebaron and K.A. Athanasiou, Extracellular matrix cell adhesion peptides: functional applications in orthopedic materials. Tissue Eng 6 (2000), pp. 85-104.
26. K. Panduranga Rao, Recent developments of collagen based biomaterials for medical applications and drug delivery systems. J Biomater Sci Polym Ed 7 (1995), pp. 623-645.
27. M. Gungormus and O. Kaya, Evaluation of the effect of heterologous type I collagen on healing of bone defects. J Oral Maxillofac Surg 60 (2002), pp. 541-545.
28. J.W.M. Heemskerk, W.M.J. Wust, M.A.H. Feijge, C.M.P. Reutelingsperger and T. Lindhout, Collagen but not fibrinogen surfaces induce bleb formation, exposure of phosphatidylserine, and procoagulant activity of adherent platelets: evidence for regulation by protein tyrosine kinase-dependent Ca2+ responses. Blood 90 (1997), pp. 2615-2625. Abstract-MEDLINE | Abstract-EMBASE
29. J.E. Davies and M.M. Housseini, Histodynamics of endosseous wound healing. In: J.E. Davies, Editor, Bone engineering, em squared, Toronoto (2000), pp. 1-14.
30. C.H. Gemmell and J.Y. Park, Initial blood interactions with endosseous implant materials. In: J.E. Davies, Editor, , em squared, Bone engineering. Toronoto (2000), pp. 108-117.
31. D.A. Puleo, Release and retention of biomolecules in collagen deposited on orthopedic biomaterials. Artif Cells Blood Substit Immobil Biotechnol 27 (1990), pp. 65-75.
32. C. Roehlecke, M. Witt, M. Kasper, E. Schulze, C. Wolf, A. Hofer and R.W. Funk, Synergistic effect of titanium alloy and collagen type I on cell adhesion, proliferation and differentiation of osteoblast-like cells. Cells Tissues Organs 168 (2001), pp. 178-187. Abstract-MEDLINE | Abstract-Elsevier BIOBASE | Abstract-EMBASE | Full Text via CrossRef
33. M. Morra and C. Cassinelli, Organic surface chemistry on titanium surfaces via thin film deposition. J Biomed Mater Res 37 (1997), pp. 198-206. Abstract-EMBASE | Abstract-MEDLINE | Abstract-Compendex | Full Text via CrossRef
34. J.K. Yun, K. DeFife, E. Colton, S. Stack, A. Azeez, L. Cahalan, M. Verhoeven, P. Cahalan and J.M. Anderson, Human monocyte/macrophage adhesion and cytokine production on surface-modified poly(tetrafluoroethylene/hexafluoropropylene) polymers with and without protein preadsorption. J Biomed Mater Res 29 (1995), pp. 257-268. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex
35. International Standard ISO 10993-6: biological evaluation of medical devices-Part 6: tests for local effects after implantation.
36. J.Y. Martin, Z. Schwartz, T.W. Hummert, D.M. Schraub, J. Simpson, J. Lankford, Jr, D.D. Dean, D.L. Cochran and B.D. Boyan, Effect of titanium surface roughness on proliferation, differentiation, and protein synthesis of human osteoblast-like cells (MG63). J Biomed Mater Res 29 (1995), pp. 389-401. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex
37. Freshney RI. Culture of animal cells, a manual of basic techniques, 4th ed. New York: Wiley; 2000. p. 315.
38. J.C. Wataha, N. O'Dell, B.B. Singh, M. Ghazi, G.M. Whitford and P.E. Lockwood, Relating nickel-induced tissue inflammation to nickel release in vivo. J Biomed Mater Res (Appl Biomater) 58 (2001), pp. 537-544. Abstract-MEDLINE | Abstract-EMBASE | Full Text via CrossRef
39. T. Albrektsson and C. Johansson, Quantified bone tissue reactions to various metallic materials with reference to the so-called osseointegration concept. In: J.E. Davies, Editor, , Toronto Press, The bone biomaterial interface. Toronto (1991), pp. 357-363.
40. Y.H. An, Methods of evaluation in orthopaedic animal research. In: Y.H. An and R.J. Friedman, Editors, Animal models in orthopaedic research, CRC Press, Boca Raton, FL (1999), pp. 85-114.
41. A. Moroni, L. Orienti, S. Stea and M. Visentin, Improvement of the bone-pin interface with hydroxyapatite coating: an in vivo long-term experimental study. J Orthop Trauma 10 (1996), pp. 236-242. Abstract-MEDLINE | Full Text via CrossRef
42. B. Sanden, C. Olerud, C. Johansson and S. Larsson, Improved bone-screw interface with hydroxyapatite coating. An in vivo study of loaded pedicle screws in sheep. Spine 26 (2001), pp. 2673-2678. Abstract-EMBASE | Abstract-MEDLINE | Full Text via CrossRef
43. B. Kasemo and J. Lausmaa, Biomaterial and implant surfaces: on the role of cleanliness, contamination and preparation procedures. J Biomed Mater Res 22 (1988), pp. 145-158. Abstract-MEDLINE | Abstract-EMBASE
44. D.C. Smith, R.M. Pilliar, N.S. McIntyre and J.B. Metson, Dental implant materials. II: preparative procedures and surface spectroscopies studies. J Biomed Mater Res 25 (1991), pp. 1069-1084. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex
45. M. Wieland, C. Sittig, D.M. Brunette, M. Textor and N.D. Spencer, Measurement and evaluation of the chemical composition and topography of titanium implant surfaces. In: J.E. Davies, Editor, Bone engineering, em squared, Toronto (2000), pp. 163-181.
46. M. Morra, C. Cassinelli, G. Bruzzone, A. Carpi, G. Di Santi, R. Giardino and M. Fini, Surface chemistry effects of topography modification of titanium dental implants surfaces: 1. Surface analysis. Int J Oral Maxillofacial Implants 18 (2003), pp. 40-45. Abstract-MEDLINE
47. H.J. Butt, M. Kappl, H. Mueller, R. Raiteri, W. Meyer and J. Ruhe, Steric forces measured with the atomic force microscope at various temperatures. Langmuir 15 (1999), pp. 2559-2565. Abstract-Compendex | Full Text via CrossRef
48. S.C. Woodward and T.N. Salthouse, The tissue response to implants and its evaluation by light microscopy. In: F. vonRecum, Editor, Handbook of biomaterials evaluation, Macmillian Publishing Company, New York (1986), pp. 364-378.
49. Pohler OEM. Unalloyed titanium for implants in bone surgery. Injury Int Care Injured 2000;31:S-D7-13.
50. Anderson JM. Soft tissue response. In: Black J, Hastings G, editors. Handbook of biomaterial properties. London: Chapman & Hall, 1998. p. 490-8.
51. A. Rosenberg, N. Danielsen and L.M. Bjursten, Reactive capsule formation around soft-tissue implants is related to cell necrosis. J Biomed Mater Res 46 (1999), pp. 458-464.
52. M. Mizuno, R. Fujisawa and Y. Kuboki, Type I collagen-induced osteoblastic differentiation of bone-marrow cells mediated by collagen-alpha2beta1 integrin interaction. J Cell Physiol 184 (2000), pp. 207-213. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Elsevier BIOBASE | Full Text via CrossRef
53. A.S. Windeler, L. Bonewald, A.G. Khare, B. Boyan and G.R. Mundy, The influence of sputtered bone substitutes on cell growth and phenotypic expression. In: J.E. Davies, Editor, The bone biomaterial interface, University of Toronoto Press, Toronto (1991), pp. 205-213.
Wyszukiwarka
Podobne podstrony:
70 1003 1019 Influence of Surface Engineering on the Performance of Tool Steels for Die Casting
Improved biological performance of Ti implants due to surfac
Manovich, Lev The Engineering of Vision from Constructivism to Computers
65 935 946 Laser Surface Treatment of The High Nitrogen Steel X30CrMoN15 1
7 77 93 Heat and Surface Treatment of Hot Works for Optimum Performance
16 Changes in sea surface temperature of the South Baltic Sea (1854 2005)
Best Available Techniques for the Surface Treatment of metals and plastics
Corrosion behavior and surface characterization of titanium
Structure and properties of Ti
Laser surface modification of hydroxyapatite and glass
Warren Murphy Destroyer 103 Engines of Destruction
Effects of machining parameters on surface integrity of hard machined surfaces
The corrosion behaviour of Ti
The Engines of Dawn Paul Cook
więcej podobnych podstron