Nanoporous hydroxyapatite/sodium titanate bilayer on titanium implants for improved osteointegration

Abstract

Objective

The aim of this study was to improve the strength and quality of the titanium–hydroxyapatite interface in order to prevent long-term failure of the implanted devices originating from coating delamination and to test it in an in-vivo model.

Methods

Ti disks and dental commercial implants were etched in Kroll solution. Thermochemical treatments of the acid-etched titanium were combined with sol–gel hydroxyapatite (HA) coating processes to obtain a nanoporous hydroxyapatite/sodium titanate bilayer. The sodium titanate layer was created by incorporating sodium ions onto the Ti surface during a NaOH alkaline treatment and stabilized using a heat treatment. HA layer was added by dip-coating in a sol–gel solution. The bioactivity was assessed in vitro with murine MC3T3-E1 and human SaOs-2 cells. Functional and histopathological evaluations of the coated Ti implants were performed at 22, 34 and 60 days of implantation in a dog lower mandible model.

Results

Nanoporous hydroxyapatite/sodium titanate bilayer on titanium implants was sensitive neither to crack propagation nor to layer delamination. The in vitro results on murine MC3T3-E1 and human SaOs-2 cells confirm the advantage of this coating regarding the capacity of cell growth and differentiation. Signs of progressive bone incorporation, such as cancellous bone formed in contact with the implant over the existing compact bone, were notable as early as day 22. Overall, osteoconduction and osteointegration mean scores were higher for test implants compared to the controls at 22 and 34 days.

Significance

Nanoporous hydroxyapatite/sodium titanate bilayer improves the in-vivo osteoconduction and osteointegration. It prevents the delamination during the screwing and it could increase HA-coated dental implant stability without adhesive failures. The combination of thermochemical treatments with dip coating is a low-cost strategy.

Introduction

Titanium (Ti) and its alloys have been extensively used in dental implants since the discovery of the osteointegration of Ti in bones . However, the native metallic devices are bioinert and are capsulated by fibrous tissue preventing tight contact to the surrounding host bone tissue . Many improvements have been made to decrease the healing time by increasing the surface biocompatibility and the osteoconductive properties . Some studies have demonstrated that the bone to implant contact and the biomechanical interaction between them can be improved at an early implantation stage when the surface roughness is increased . Acid-etching or grit-blasting/acid-etching are common methods used for commercial implants to obtain roughness values ranging from 1 to 2 μm that, in most cases, lead to acceptable bone integration . Combining the latter with chemical changes by incorporating bioactive cations (Ca, Si, Zn, Sr and Mg) enhances the osteogenic differentiation of stem cells .

Despite the success of these implant surfaces over several decades, further progress is of particular importance in treating the growing number of patients having poor bone quality due to pathologies or in the elderly . The addition of calcium- and phosphorous-based materials, such as calcium phosphate and particularly hydroxyapatite (HA), as coatings has received significant attention due to the similarities of these elements with the basic components of natural bone. Most commercially available bioceramic coatings are plasma-sprayed HA coatings of 20–50 μm thick . They provide a higher osteoconductivity in comparison to uncoated implants . However, a high adhesion strength requires high plasma energy and temperature that in turn causes increased thermal decomposition of HA and limitations, such as residual stresses, may lead to coating delamination. In an attempt to mitigate the plasma-sprayed coating limitations, other physical processes ( e.g . pulsed laser deposition, sputtering coating techniques, ion-beam assisted deposition, electrophoretic deposition), or chemical routes ( e.g . sol–gel method, growth in simulated body fluid, anodic oxidation) have been developed. The main advantages and limitations of these techniques for HA-based coatings are presented elsewhere . Among them, the chemical or sol–gel methods are simple, low cost procedures and they provide a method for growing bioactive layers.

Nanoporous structures of a bioactive titanate layer were grown in concentrated NaOH and subsequently heat treated in air . The bone’s response to the Ti surface has been studied . The heat treatment temperature played a crucial role in promoting superior cell adhesion. In other reported studies, an apatite layer was grown on sodium titanate after the immersion in simulated body fluid . The sol–gel method was also used to grow thin HA films having a homogeneous chemical composition . In the present study, we combine the alkaline treatment of Ti and the sol–gel coating processes to obtain a nanoporous HA/sodium titanate bilayer . The Ti/HA interface displays a micro/nano topography favouring a firm HA to Ti bond. The biological properties of this novel functional Ti surface are evaluated in vitro by tests with murine MC3T3-E1 preosteoblasts and human SaOs-2 cells. Its efficacy is investigated using an in vivo dog mandible implantation model. Bone to implant contact and torque measurements of the novel functional Ti surface are compared to those of sandblasted commercial implants.

Materials and methods

Fabrication and characterization of hydroxyapatite (HA)/sodium titanate coatings

For the in vitro studies, bars of commercial titanium (Ti Gr4) were cut into disks (18 mm diameter, 1.5 mm height). Commercial Ti implants with and without sand-blasting treatments were purchased from Biotech Dental. To reach a final roughness of approximately 500 nm, the disks were mechanically ground and polished to a mirror-like surface finish using plate series (P120 to P4000 grit, 125–2.5 μm grades) and colloidal silica (0.05 μm grade). Homogeneous polishing levels were controlled by using the differential interference contrast (DIC) mode of an optical microscope (Olympus IX70, with an UMP/anF/50x/0.80 BD objective). For comparison, disks were sand-blasted under the same conditions as commercial implants.

Discs and implants were cleaned in acetone (30 min), 100% ethanol (30 min), and double-distilled water in an ultrasonic cleaning bath. They were firstly subjected to an acid-etching treatment using a Kroll solution’s reagent (0.045 M HF, 0.058 M HNO 3 , 3 min at room temperature) to remove native surface oxides, and then to an alkaline treatment in NaOH (5 M–10 M at 60–80 °C for 1–72 h) to obtain a sodium titanate hydrogel layer on the Ti surface. Few morphological variations were observed. For practical reasons, the alkaline treatment of the commercial implants was carried out in NaOH 10 M for 1 h at 80 °C. The sodium titanate layer was thermally stabilized at 630 °C for 1 h using a stepwise heating rate of 5 °C per min. The Kroll- and alkaline-treated discs were labelled Kr and HT respectively.

Similar to previous studies, the HA layers were deposited on the Ti-treated disks and the commercial implants by dip-coating in a sol–gel solution prepared using 3 M calcium nitrate (Ca(NO 3 ) 2 ) and 1.8 M triethyl phosphite P(OCH 2 CH 3 ) 3 from Sigma–Aldrich . Dipping time was 1 and 3 min for disks and implants, respectively. Following this, the samples were annealed at 600 °C for 20 min and washed in acetone (30 s), in ethanol (30 s) and in double-distilled water in an ultrasonic cleaner. The dip-coated discs were labelled DIP1 or DIP2 according to the number of dip-coatings (one or two). X-ray diffraction measurements (XRD) were performed using a Siemens D5000 X-ray diffraction system in reflection mode with a quartz monochromator (Cu K α1 = 0.154056 nm). The surface morphology was characterized by means of scanning electron microscopy (SEM) using a JEOL 6700 electron microscope equipped with an energy dispersive X-ray spectroscopy analysis (EDS-X). In addition to this, a cross-sectional view of HA/sodium titanate layer was also provided by SEM/EDS-X. Observations were carried out at 3 kV and qualitative analyses of the HA layer composition at 10 kV. The scratch resistance of the HA/sodium titanate layer on Ti was evaluated using a nano-scratch device (MTS Systems Corporation). During the scratch test’s initial phase, the surface topography was scanned by applying a 0.05 mN load and moving the indenter in contact with the surface along the testing zone’s length. Then during the scratch phase, the applied normal (F n ) and tangential loads (F t ) were measured to calculate the apparent coefficient of friction (μ app ) as:

<SPAN role=presentation tabIndex=0 id=MathJax-Element-1-Frame class=MathJax style="POSITION: relative" data-mathml='μapp=FtFn’>μapp=FtFnμapp=FtFn
μ app = F t F n

The indenter’s penetration depth and displacement on the surface were also measured. The load applied during the scratch phase varied linearly from 10 to 100 mN. The scratch length was equal to 500 μm and the scratch speed was 10 μm/s. The 90° conical indenter has a spherical tip with a 5 μm radius. The mean contact pressure of the scratch test (P m ) was measured and the scratch induced damage of the HA/titanate layer was observed using a SEM at a 3 kV accelerating voltage.

Cell cultures and analyses

Prior to any biological use, both faces of the supports were subjected to a UV sterilizing treatment in 12, dry well plaques under a laminar flow cell culture hood, washed in sterile phosphate buffered saline (PBS) prior to preincubation in a standard cell culture media for 2–4 h at 37 °C and finally subjected to a second UV sterilization cycle. After media changes, cells were seeded at the desired concentrations onto the supports and sterile glass cover slides within test and control wells, respectively.

Human osteosarcoma SaOS-2 (ATCC85-HTB) and epithelial SW480 (ATCC ® CCL-228™) cells, murine fibroblast NIH/3T3 (ATCC ® CRL-1658™) and MC3T3-E1 sub-clone 4 preosteoblasts (ATCC ® CRL-2593™) were cultured in standard Dulbecco’s modified Eagle medium (DMEM) growth media at 37 °C in 5% CO 2 humidified atmosphere: according to the recommendation of the supplier for each line. SW480 and NIH/3T3 cells were used for preliminary tests of cell adherence and standard growth, SaOS-2 cells for sandblasted surfaces, and MC3T3-E1 preosteoblast behaviour was followed in both standard and in differentiation inducing media.

Cell growth and differentiation, cytotoxicity tests

An equal number of cells (5 × 10 3 cells/mL) were seeded and then cultured in duplicated series of wells containing either HT-treated, HA coated supports or controls (with or without glass cover-slides). The first medium was first changed after 48 h of culture; allowing cells to recover from handling and to adhere to surfaces, and then every 2 days. To measure growth curves and cytotoxic effects, MC3T3-E1 were seeded at 5 × 10 3 cell/mL in 12 well plates and grown for up to 11 days in standard growth medium.

Cell adhesion and proliferation in the standard medium were measured by counting using phase contrast images taken with a Nikon Eclipse Ti-S (10× objective) inverted microscope. Cell growth was quantified using 5–15 field views (chosen at random) in the middle and around all surfaces of each support and in the control wells. Cell vitality was assessed by using a LDH detection kit protocol for TOX7 kit (Sigma–Aldrich) on cell culture supernatant triplicate aliquots which were collected every 2 days from each sample. Results are reported per cell number in each sample.

Alternatively, osteoblastic differentiation of MC3T3-E1 preosteoblasts was induced after 24 h growth by supplementing the standard growth medium with ascorbic acid (Sigma–Aldrich, 50 μg/mL final concentration) and beta-glycerol phosphate (Sigma–Aldrich, 10 mM final concentration) during 2 weeks with medium changes every 2 days.

Scanning electron microscopy imaging of cell morphology

After 7 days of culture on duplicate supports, gold-palladium coated cells were examined using HITACHI TM1000 SEM (15 kV) captured images after conventionally fixing (4% glutaraldehyde—1% paraformaldehyde in 0.1 M cacodylate buffer pH 7.4), post-fixing (1% osmium tetroxide, 1 h), dehydrating steps and transferring to a 11,120 A model BALZERS UNION critical point dryer (Principality of Liechtenstein).

Immunohistochemistry

Cells were fixed (3.7% paraformaldehyde) and permeabilized (0.1% Triton X-100). After 30 min blockage with 0.1% bovine serum albumin, polyclonal primary antibodies (anti-vinculin sc-7648, anti-osteopontin sc-20788, anti-osteocalcin sc-83294, Santa Cruz Biotechnology) were subsequently applied having dilutions ranging from 1:500 to 1:2000. The goat or rabbit fluorescein isothiocyanate (FITC)-conjugated secondary antibodies (sc-2012, Santa Cruz Biotechnology) were used having a 1:2000 dilution, and rhodamine-labelled Phalloidin (#77418-1EA, Sigma–Aldrich) having a 0.005 mg/mL final concentration. Samples were mounted on glass slides using Vectashield DAPI (4′,6-diamidino-2-phenylindole) containing mounting medium (#H-1200, Clinisciences). A Nikon Eclipse Ti-S inverted microscope, equipped for fluorescence with a Neofluar 100× oil objective and a Confocal Zeiss LSM 780 were used for image documentation. Experiments were replicated.

Real time polymerase chain reaction (qPCR)

Nucleic acids were prepared by using GE Healthcare Illustra Triple Prep kit (#28-9425-44 Fisher scientific) and protocol. Their concentration and purity were determined by optical density (OD) lecture using a Nanodrop 2000/2000c spectrophotometer (Thermo Fisher). First strand cDNA was prepared on 4 ng RNA from each sample using an iScript cDNA Synthesis kit (#170-8891, BioRad). The qPCR tests were performed in a CFX Connect RT system automat (BioRad), using SsoAdvanced SYBR green Supermix products following instructions (#172-5261, BioRad), with 300 nM final of appropriate primer-pairs (Invitrogen) and 40 amplification cycles at 55 °C in 20 μL total reaction volume. Despite the known expression variability of the housekeeping genes during cell life, glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and Actin B genes were amplified as internal controls . Gene expression levels are presented as relative fold changes with respect to the glass slides and normalized to the equivalent cell number on the basis of DNA concentrations.

In vivo animal studies

Eight Golden Retriever dogs (13–21 months old) were used in this investigation. The study was approved by the Institutional Animal Care and Use Committee and followed recommendation of the Guide for the Care and Use of Laboratory Animals. Surgical procedures (anaesthesia and intubation), pain control, standards of living, and appropriate methods of euthanasia were performed and recorded according to IMM Recherche’s SOPs.

All animals were submitted to dental extraction 3 months before device implantation. The dental implants were inserted in the dog mandible after healing of the extraction area. The implants were placed in orthotopic positions by using standard surgical methods.

Forty-eight (48) devices were implanted: 24 commercial implants coated with HA layer (labelled ‘test’) and 24 uncoated commercial implants (labelled ‘control’). At 3 weeks (D22), 4 weeks (D34) and 8 weeks (D60), respectively, two animals (6 test and 6 control implants each) were euthanized for histological analyses. In addition to this, two animals (6 test and 6 control implants) were euthanized at 8 weeks (D60) for screw loosening torque force-tests.

The osteodensimetry evaluation was performed by a CT-scan examination using a 64 brilliance Philips. The osteodensimetry measurements were performed using Osirix 32-bits software on the sequence named “OS” at this specific window: NF = 300, LF = 4500. A ROI (region of interest >1 mm 2 ) was chosen on specific sites. A CT-scan was performed on all the animals at 1, 2, 4 and 8 weeks. The screw loosening torque force was measured using a torque gauge (Centor Easy™, Andilog Tech). For histological evaluation, mandibular bone with implants were sampled at the site of implantation, routinely fixed in 4% buffered formalin and embedded in poly(methyl methacrylate) (PMMA) resin. Fifty micrometer-thick mesio-distal sections were prepared using the EXAKT ® grinding system (EXAKT Technologies, Inc.) and stained with hematoxylin and eosin (H&E). Bone to implant contact (BIC) values was measured on mesio-distal H&E sections.

Materials and methods

Fabrication and characterization of hydroxyapatite (HA)/sodium titanate coatings

For the in vitro studies, bars of commercial titanium (Ti Gr4) were cut into disks (18 mm diameter, 1.5 mm height). Commercial Ti implants with and without sand-blasting treatments were purchased from Biotech Dental. To reach a final roughness of approximately 500 nm, the disks were mechanically ground and polished to a mirror-like surface finish using plate series (P120 to P4000 grit, 125–2.5 μm grades) and colloidal silica (0.05 μm grade). Homogeneous polishing levels were controlled by using the differential interference contrast (DIC) mode of an optical microscope (Olympus IX70, with an UMP/anF/50x/0.80 BD objective). For comparison, disks were sand-blasted under the same conditions as commercial implants.

Discs and implants were cleaned in acetone (30 min), 100% ethanol (30 min), and double-distilled water in an ultrasonic cleaning bath. They were firstly subjected to an acid-etching treatment using a Kroll solution’s reagent (0.045 M HF, 0.058 M HNO 3 , 3 min at room temperature) to remove native surface oxides, and then to an alkaline treatment in NaOH (5 M–10 M at 60–80 °C for 1–72 h) to obtain a sodium titanate hydrogel layer on the Ti surface. Few morphological variations were observed. For practical reasons, the alkaline treatment of the commercial implants was carried out in NaOH 10 M for 1 h at 80 °C. The sodium titanate layer was thermally stabilized at 630 °C for 1 h using a stepwise heating rate of 5 °C per min. The Kroll- and alkaline-treated discs were labelled Kr and HT respectively.

Similar to previous studies, the HA layers were deposited on the Ti-treated disks and the commercial implants by dip-coating in a sol–gel solution prepared using 3 M calcium nitrate (Ca(NO 3 ) 2 ) and 1.8 M triethyl phosphite P(OCH 2 CH 3 ) 3 from Sigma–Aldrich . Dipping time was 1 and 3 min for disks and implants, respectively. Following this, the samples were annealed at 600 °C for 20 min and washed in acetone (30 s), in ethanol (30 s) and in double-distilled water in an ultrasonic cleaner. The dip-coated discs were labelled DIP1 or DIP2 according to the number of dip-coatings (one or two). X-ray diffraction measurements (XRD) were performed using a Siemens D5000 X-ray diffraction system in reflection mode with a quartz monochromator (Cu K α1 = 0.154056 nm). The surface morphology was characterized by means of scanning electron microscopy (SEM) using a JEOL 6700 electron microscope equipped with an energy dispersive X-ray spectroscopy analysis (EDS-X). In addition to this, a cross-sectional view of HA/sodium titanate layer was also provided by SEM/EDS-X. Observations were carried out at 3 kV and qualitative analyses of the HA layer composition at 10 kV. The scratch resistance of the HA/sodium titanate layer on Ti was evaluated using a nano-scratch device (MTS Systems Corporation). During the scratch test’s initial phase, the surface topography was scanned by applying a 0.05 mN load and moving the indenter in contact with the surface along the testing zone’s length. Then during the scratch phase, the applied normal (F n ) and tangential loads (F t ) were measured to calculate the apparent coefficient of friction (μ app ) as:

<SPAN role=presentation tabIndex=0 id=MathJax-Element-2-Frame class=MathJax style="POSITION: relative" data-mathml='μapp=FtFn’>μapp=FtFnμapp=FtFn
μ app = F t F n
Only gold members can continue reading. Log In or Register to continue

Nov 22, 2017 | Posted by in Dental Materials | Comments Off on Nanoporous hydroxyapatite/sodium titanate bilayer on titanium implants for improved osteointegration
Premium Wordpress Themes by UFO Themes