lable at ScienceDirect

Biomaterials 31 (2010) 4818e4828

Contents lists avai

Biomaterials

journal homepage: www.elsevier .com/locate/biomateria ls

Nonvolatile buffer coating of titanium to prevent its biological agingand for drug delivery

Takeo Suzuki a,b, Katsutoshi Kubo a,b, Norio Hori a, Masahiro Yamada a, Norinaga Kojima a,Yoshihiko Sugita a,b, Hatsuhiko Maeda b, Takahiro Ogawa a,*

a Laboratory for Bone and Implant Sciences (LBIS), The Jane and Jerry Weintraub Center for Reconstructive Biotechnology, Division of Advanced Prosthodontics, Biomaterialsand Hospital Dentistry, UCLA School of Dentistry, 10833 Le Conte Avenue (B3-081 CHS), Box 951668, Los Angeles, CA 90095-1668, USAbDepartment of Pathology, Aichi-Gakuin University, School of Dentistry, Nagoya, Japan

a r t i c l e i n f o

Article history:Received 24 January 2010Accepted 23 February 2010Available online 28 March 2010

Keywords:Bone-titanium integrationHydrophilicityOsseointegrationImplantsHEPESN-acetyl cysteine (NAC)

* Corresponding author. Tel.: þ1 (310) 825 0727; faE-mail address: togawa@dentistry.ucla.edu (T. Oga

0142-9612/$ e see front matter � 2010 Elsevier Ltd.doi:10.1016/j.biomaterials.2010.02.061

a b s t r a c t

The osseointegration capability of titanium decreases over time. This phenomenon, defined as biologicalaging of titanium, is associated with the disappearance of hydrophilicity and the progressive accumu-lation of hydrocarbons on titanium surfaces. The objective of this study was to examine whether coatingof titanium surfaces with 4-(2-Hydroxylethyl)-1-piperazineethanesulfonic acid (HEPES) buffer,a nonvolatile zwitterionic chemical buffering agent, could prevent the time-dependent degradation ofthe bioactivity of titanium. Commercially pure titanium samples, prepared as disks and cylinders, wereacid-etched with H2SO4. A third of the samples were used for experiments immediately after processing(new surfaces), while another third were stored under dark ambient conditions for 3 months (3-month-old surfaces). The remaining third were coated with HEPES after acid-etching and were stored for 3months (HEPES-coated 3-month-old surfaces). The 3-month-old surfaces were hydrophobic, while newand HEPES-coated 3-month-old surfaces were superhydrophilic. Protein adsorption and the number ofosteoblasts attached during an initial culture period were substantially lower for 3-month-old surfacesthan for new and HEPES-coated 3-month-old surfaces. Alkaline phosphatase activity and calciumdeposition in osteoblast cultures were reduced by more than 50% on 3-month-old surfaces compared tonew surfaces, whereas such degradation was not found on HEPES-coated 3-month-old surfaces. Thestrength of in vivo bone-implant integration for 3-month-old implants, evaluated by the push-in test,was 60% lower than that for new implants. The push-in value of HEPES-coated 3-month-old implantswas equivalent to that of new implants. Coating titanium surfaces with HEPES containing an antioxidantamino acid derivative, N-acetyl cysteine (NAC), further enhanced osteoblast attachment to the surfaces,along with the increase level of intracellular glutathione reserves as a result of cellular uptake of NAC.These results suggest that HEPES coating of titanium surfaces maintained their superhydrophilicity for atleast 3 months and resulted in a continuous retention of bioactivity and osteoconductivity similar tofreshly prepared surfaces. This coating technology may be useful for preventing biological aging oftitanium and delivering biological molecules for synergistic enhancement of bone-titanium integration.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Recent reports on time-dependent degradation of the bio-activity of titanium, defined as biological aging of titanium, led tothe first discovery of time-related changes in biological potential ofbiomaterials, and provided significant clinical impacts in the field ofosseous implant therapy [1,2]. In vitro bioactivity, such as thoseresulting in the attraction and proliferation of osteogenic cells, is

x: þ1 (310) 825 6345.wa).

All rights reserved.

reduced by 40e80% on 4-week-old titanium surfaces compared tonewly prepared surfaces [1,2]. In terms of in vivo bone integrationcapability, at an early healing stage in an animal model, thestrength of implant fixation and the percentage of bone coveragearound implants for 4-week-old surfaces were reduced to less than50% of those for new surfaces [1]. It should be noted that thebiomechanical and histological indices of these aged implantsshowed lower values not only in the early stage of healing but alsoin the late stage. This indicated that the aging of titanium implantsresulted in not only delayed but also a compromised level ofbone-implant integration. In fact, in the late stage of healing, bone-implant contact remained less than 60% for 4-week-old implants,

T. Suzuki et al. / Biomaterials 31 (2010) 4818e4828 4819

while it reached over 90% for new implants [1]. Commerciallyavailable orthopedic and dental implants are sold as storablemedical devices. These products invariably age during theirinventory and distribution, as well as storage before use. It isunlikely that these products are used within 4 weeks aftermanufacturing. Therefore, the pivotal implication is that the bio-logical capabilities of commercially available implants may havebeen unpredictably and unavoidably compromised.

In addition, the scientific significance of biological aging oftitanium is crucial for exploring future advancements in this field.Titanium forms a passive, protective oxide layer on its surfaceonce exposed to the atmosphere. Because of the chemical stabilityof this oxidized surface (primarily TiO2), titanium had beencategorized into a group of bioinert materials, which means thatthey do not interact with host cells/tissues. Accordingly, the bio-logical capability of titanium surfaces had been accepted as innateand invariable over time. The necessity for correcting theseassumptions was firmly suggested by the discovery of biologicalaging of titanium. A series of reports have indicated that freshlyprepared titanium surfaces were bioactive, whereas thecommonly available titanium surfaces, including commercialproducts, that have been exposed to the atmosphere long enoughare bioinert [1e3]. These new findings were expected to alter thelong-held assumption of titanium that titanium is a bioinertmaterial. Titanium should be defined as bioactive under thecondition that the surface is new. Moreover, the demonstration ofsubstantial differences in bioactivity of titanium surface atdifferent stages of their aging implied that standardizing the ageof titanium samples is required in future research designs in thefiled or otherwise a careful interpretation is required for theresults. Biological aging is seen to occur in all surface topogra-phies of titanium that were tested, ranging from microtopo-graphical features (e.g., acid-etched surfaces) to a relativelysmooth morphology (e.g., machined surfaces) [1e3]. Currently,there are no effective measures to prevent this phenomenon.

Although the exact mechanisms underlying the biologicalaging of titanium have not been fully elucidated, this phenomenonis associated with disappearance of hydrophilicity and theprogressive contamination of hydrocarbons on titanium surfacesover time [1,2]. For example, naturally induced super-hydrophilicity of freshly prepared acid-etched titanium surfacesonly lasts for a week [1]. The atomic ratio of carbon, which is lessthan 20% on a fresh titanium surface, increases to approximately60% after 4 weeks of exposure to the atmosphere [1,2]. Wehypothesized that coating a newly prepared titanium surface witha nonvolatile liquid could prevent or reduce the time-dependentdegradation in its bioactivity by preserving its hydrophilicity andavoiding excessive chemical contamination from the atmosphere.HEPES s is a nonvolatile zwitterionic buffering agent witha molecular formula of C8H18N2O4S. It is a stable nonreactive agentthat is widely used for cell culture reagents and for storingbiological molecules and cells [4e7].

An issue here is that HEPES is a highly viscous liquid and cannot be thinly coated onto material surfaces under normal condi-tions. Successful bone-titanium integration requires an intimatejuxtaposition of titanium surfaces and newly formed bone. Asignificantly thick intervention of a nonbiological layer, whichmight prevent osteogenic cells from reaching the titaniumsurfaces, was deemed to be unsuitable. Our experimental strategywas to utilize the superhydrophilic nature of newly preparedtitanium surfaces. Here, we demonstrate that HEPES buffer canspread and form a thin layer exclusively on freshly preparedtitanium surfaces, but not on old titanium surfaces.

The objective of this study was to determine if a thin layer ofcoating with HEPES buffer on freshly prepared titanium surfaces

could prevent the biological aging of titanium. The bioactivityand osseointegration capability of titanium samples after 3months of storage (3-month-old titanium surfaces), with andwithout HEPES coating, as well as those of freshly processedtitanium samples (new titanium surface) were compared. Wealso tested the potential of HEPES coating as a carrier vehicle forbiological molecules to enhance the osteogenic environmentaround titanium surfaces. N-acetyl cysteine (NAC) is an aminoacid derivative and known as an strong antioxidant [8]. NAC iseasily deacetylated into cysteine, which is an importantprecursor of glutathione [8]. The glutathione-mediated redoxcycle is the most important removal system for exogenous andendogenous reactive oxygen species (ROS) [9,10]. Therefore,cellular uptake of NAC results in the increased level of antioxi-dant defense capacity [11]. Also, the NAC-mediated directscavenging effect for polymerization-induced reactive oxygenspecies has been demonstrated in various biomaterials [12e14].In particular, a recent study demonstrated that NAC-incorporatedpoly(methyl methacrylate) (PMMA) resin-based bone cementshowed significant improvement in its osteoconductivity [11].Besides these exogenous oxidants, cells are subjected tosubstantial degrees of oxidative stress by mechanical andchemical stimuli during invasive surgery in vivo and even duringvarious steps in cell culture procedures. To test the feasibility oftitanium surface-mediated drug delivery, we examined the effectof HEPES/NAC coating to further enhance the osteogenicenvironment around titanium surfaces.

2. Materials and methods

2.1. Preparation of HEPES-coated titanium samples

Commercially pure titanium samples prepared as disks (20 mm in diameter,1 mm in thickness) and cylinders (1 mm in diameter, 2 mm in length) were acid-etched with H2SO4 using a previously established protocol [15e17]. The surfacemorphology of the acid-etched surface was examined by scanning electronmicroscopy (SEM) (XL30, Philips, Eindhoven, Netherlands). A third of the sampleswere used for experiments immediately after processing (new surface), whileanother third was stored under dark ambient conditions for 3 months (3-month-oldsurface). The remaining third were coated with HEPES buffer immediately after acidetching, and were stored for 3 months (HEPES-coated 3-month-old surface). HEPESbuffer was prepared as a filter-sterilized 1 M stock solution at pH 7.2. Ten ml and 0.3 mlof HEPES were used to coat titanium disks and cylinders, respectively. To indirectlymeasure the thickness of HEPES coated on titanium surfaces, 3 ml of HEPES wasdropped onto new titanium disks. The thickness of HEPES was calculated by dividingthe volume of HEPES drop (3 ml) by the area of spread. Hydrophilicity of titaniumsurfaces was assessed by measuring the contact angle of a 10 ml ddH2O drop. To testthe potential use of HEPES coating as a carrier vehicle, titanium disks coated withNAC were also prepared. NAC (Sigma-Aldrich, St. Louis, MO) was prepared asa 0.5 mol/L stock solution in HEPES buffer whose pH was adjusted to 7.2. Ten ml ofHEPES/NAC was used to coat titanium disks, and the disks were stored for 3 months(HEPES/NAC-coated 3-month-old surface). Electrostatic potential of HEPES-coatedtitanium surfaces was measured using a coulomb meter (NK-1001, Kasuga Denki,Inc. Tokyo, Japan).

2.2. Protein adsorption

Bovine serum albumin, fraction V (Pierce Biotechnology, Inc., Rockford, IL) wasused as a model protein. A previously established bicinchoninic acid (BCA)-basedcolorimetric detectionwas used [3,15]. The BCA reagent is used to detect Cuþ1, whichis formed when Cuþ2 is reduced by proteins in an alkaline environment [18e20].Three hundred ml of a protein solution (1 mg/ml protein/saline) was spread overa titanium disk using a pipette. After 3 or 24 h of incubation under sterile humidifiedconditions at 37 �C, the solution containing nonadherent proteins was removed andmixed with microbicinchoninic acid (Pierce Biotechnology) at 37 �C for 60 min. Theamount of the removed protein was quantified using a microplate reader (SynergyHT, BioTek Instruments, Winooski, VT) at 562 nm. A color response curve thatproduced a consistent standard of proof (correlation coefficient¼ 1.00) was used forprotein estimation. The total amount of protein initially applied was also quantifiedwith this method. The proportion of protein adsorption was calculated as thepercentage of proteins adsorbed to titanium surfaces relative to the total amount ofproteins initially applied.

T. Suzuki et al. / Biomaterials 31 (2010) 4818e48284820

2.3. Osteoblast cell culture

As established previously [21], bone marrow cells isolated from the femurs of8-week-old male SpragueeDawley rats were placed into alpha-modified Eagle'smedium supplemented with 15% fetal bovine serum, 50 mg/ml ascorbic acid, 10 mM

Na-b-glycerophosphate, 10�8 M dexamethasone, and antibiotic/antimycoticsolution. Cells were incubated in a humidified atmosphere of 95% air and 5% CO2 at37 �C. At 80% confluency, the cells were detached using 0.25% trypsin-1 mM

EDTA-4Na and seeded onto titanium disks at a density of 4�104 cells/cm2. Theculture medium was renewed every 3 days.

2.4. Cell attachment, density, and proliferation assays

Initial attachment of cells was evaluated by measuring the amount of cellsattached to titanium substrates after 3 h and 24 h of incubation. Propagated cellswere also quantified as cell density at days 2 and 5 of culture. Quantifications wereperformed using WST-1-based colorimetry (WST-1, Roche Applied Science,Mannheim, Germany). A culture well was incubated at 37 �C for 4 h with 100 mltetrazolium salt (WST-1) reagent. The amount of formazan product was measuredusing an ELISA reader at 420 nm. The proliferative activity of cells was measured byBrdU incorporation during DNA synthesis. At day 2 of culture, 100 ml of 100 mM BrdUsolution (Roche Applied Science) was added to the culture wells and incubated for10 h. After trypsinizing cells and denaturing DNA, cultures were incubated withanti-BrdU antibody conjugated with peroxidase for 90 min and reacted withtetramethylbenzidine for color development. Absorbance was measured using anELISA reader at 370 nm (Synergy HT, BioTek Instruments, Winooski, VT).

2.5. Morphology and morphometry of cells

Cell morphology and cytoskeletal arrangement of osteoblasts seeded ontotitanium surfaces were examined using confocal laser scanning microscopy. After3 h of culture, cells were fixed in 10% formalin and stained using fluorescent dyesDAPI (nuclei blue color, Vector, CA) and rhodamine phalloidin (actin filament, redcolor; Molecular Probes, OR). The area, perimeter and Feret's diameter of cells werequantified using an image analyzer (ImageJ, NIH, Bethesda, ML).

2.6. Alkaline phosphatase (ALP) activity

ALP activity of osteoblasts was examined at day 10 using a colorimetry-basedassay. Cultures were rinsed with ddH2O followed by the addition of 250 mlp-Nitrophenylphosphate (LabAssay ATP, Wako Pure Chemicals, Richmond, VA), andthen incubated at 37 �C for 15 min. ALP activity was evaluated as the amount ofnitrophenol released by the enzymatic reaction andmeasured using an ELISA readerat 405 nm.

2.7. Mineralization assay

The mineralization capability of cultured osteoblasts was examined bycolorimetry-based quantification of calcium deposition at day 20. Cultures werewashed with PBS and incubated overnight in 1 ml of 0.5 M HCl solution with gentleshaking. The solution was mixed with o-cresolphthalein complexone in an alkalinemedium (calcium binding and buffer reagent, Sigma, St Louis, MO) to produce a redcalcium-cresolphthalein complexone complex. Color intensity was measured by theabsorbance at 575 nm using an ELISA reader.

2.8. Animal surgery

Eight-week-old male SpragueeDawley rats were anesthetized using 1e2%isoflurane inhalation. Their left legs were shaved and scrubbed using 10% providone-iodine solution. The distal aspect of the left femurs was carefully exposed by skinincision and muscle dissection. The flat surface of the distal femur was selected forimplantation. The implant site was prepared 9 mm from the distal edge of the femurby drilling with a 0.8 mm round burr and was then enlarged using reamers (#ISO090 and 100). Profuse irrigation with sterile isotonic saline solution was performedfor cooling and cleaning. One cylindrical titanium sample was inserted into eachprepared hole in the femur. Surgical sites were then closed in layers. Muscle and skinwere sutured separately with resorbable sutures. The University of California at LosAngeles Chancellor's Animal Research Committee approved this protocol and allexperimentation was performed in accordance with the United States Departmentof Agriculture guidelines of animal research.

2.9. Implant biomechanical push-in test

The implant biomechanical push-in test was used to assess the biomechanicalstrength of boneeimplant integration as described elsewhere [22,23]. Femurscontaining a cylindrical implant were harvested at weeks 2 and 4 of healing andembedded into autopolymerizing resin with the top surface of the implant beinghorizontal. MicroCT was used to confirm that implants were free from cortical bonesupport to the lateral and inferior sides of the implant. The testing machine (Instron

5544 electromechanical testing system, Instron, Canton, MA) equipped with a 2000-N load cell and a pushing rod (0.8 mm diameter) was used to load the implantvertically downward at a crosshead speed of 1 mm/min. The push-in value wasdetermined by measuring the peak of a loadedisplacement curve.

2.10. Cell viability assay

To examine possible cytotoxic effects of HEPES coating on titanium surfaces,viability of cultured osteoblasts was evaluated using a flowcytometry-based assay.Flow cytometric detection of annexin V binding and propidium iodide (PI) staining(Annexin V-FITC Kit, BD Bioscience, San Jose, CA, USA) was performed on the cellsafter incubation for 24 h on the titanium disks with and without HEPES coating. Thismethod is based on the binding properties of annexin V to phosphatidylserine (PS)and on the DNA-intercalating capabilities of PI. After incubation for 24 hours thecells were tested, and the intensity of PI staining (y-axis) was plotted versus FITCintensity (x-axis). Viable cells were observed in the left lower quadrant (3, annexin Vnegative/PI negative), apoptotic cells in the right lower quadrant (4, annexin Vpositive/PI negative), and secondary necrotic cells in the right upper quadrant(2, annexin V positive/PI positive).

2.11. Intracellular glutathione detection

Asmentioned in the Introduction, ifNAC is successfullydelivered intoosteoblasts,the level of intracellular glutathione is expected to increase. Therefore, we quantifiedintracellular glutathione in osteoblasts cultured on HEPES-coated 3-month-old andHEPES/NAC-coated 3-month-old surfaces using a DTNB (5,50-dithiobis(2-nitro-benzoic acid))-based method (Total Glutathione Quantification Kit, Dojindo Molec-ular Technologies, Inc, Gaithersburg, MD). DTNB and glutathione react to generate 2-nitro-5-thiobenzoic acid and glutathione disulfide (GSSH). The glutathione producedfrom GSSH by glutathione reductase again reacts with DTNB. Therefore, the totalglutathione concentration can be determined by measuring the 2-nitro-5-thio-benzoic acid concentration. Twenty-four hours after seeding, the culture was incu-bated with DTNB and glutathione reductase for 10 minutes at 37 �C. Intracellularglutathione concentration measured by the absorbance at 412 nm using an ELISAreader was standardized by the number of cells evaluated by WST-1 assay.

2.12. Statistical analyses

Three samples were used for all biological experiments (n¼ 3) except for cellmorphometry (n¼ 10) and the implant push-in test (n¼ 6). The thickness of HEPEScoating and electrostatic potential were measured on 6 titanium disks. One-wayANOVAwas used to examine the effects of different surface conditions. If necessary,a post-hoc Bonferroni test was used as a multiple comparisons test; p< 0.05 wasconsidered significant.

3. Results

3.1. HEPES coating possible only on newly prepared titaniumsurfaces

Acid-etched titanium surfaces showed uniformmicroroughnessfeatures comprising sharp peaks and valleys with intervals of0.5e2 mm (Fig. 1A). HEPES buffer coating was attempted ontitanium surfaces by dropping 10 ml of HEPES buffer on both newsurfaces (immediately after acid etching) and 3-month-old surfaces(stored for 3 months after acid etching) (Fig. 1B). HEPES spreadimmediately and evenly on new surfaces, resulting in the rapidformation of a very thin layer of coating, whereas it remained asa rounded droplet on 3-month-old surfaces. It was technicallyimpossible to coat 3-month-old surfaces. The thickness of HEPEScoated onto new surfaces was 10.5�1.2 mm. Electrostatic potentialof HEPES-coated titanium surfaces was 0.10� 0.21 nC.

3.2. Preserved superhydrophilicity of titanium by HEPES coating

Three differently conditioned titanium surfaces were tested fortheir hydrophilic properties; (1) new surfaces, (2) 3-month-oldsurfaces, and HEPES-coated 3-month-old surfaces (titaniumsurfaces coated with HEPES immediately after acid etching andthen stored for 3months). The contact anglewith 10 ml ddH2O dropfor new surfaces was 0.0�, indicating their superhydrophilicproperty (<5�), while the contact angle for 3-month-old surfaces

Fig. 1. Surface characteristics and feasibility of HEPES coating on titanium surfaces. (A) Scanning electron microscopy images of acid-etched titanium disks used in this study. (B)Macroscopic photographic images of acid-etched titanium disks before and after HEPES coating. Note that HEPES coating was successful only on new titanium surfaces (imme-diately after acid etching) but not on 3-month-old titanium surfaces (stored for 3 months after acid-etching).

T. Suzuki et al. / Biomaterials 31 (2010) 4818e4828 4821

was 88.4�, indicating a hydrophobic property (side view imagesand histogram in Fig. 2). In contrast, HEPES-coated 3-month-oldsurfaces showed a contact angle of 0.0�, indicating that theymaintained superhydrophilic properties after 3 months of storage.

3.3. Preservation of high protein adsorption on HEPES-coatedtitanium surfaces

Protein adsorptionwas examined on the three different titaniumsurfaces using serum albumin as a model protein (Fig. 3A). Theamount of albumin adsorbed on 3-months-old surfaceswas only 20%of that fornewsurfaces after3 hof incubation (Bonferroni,p< 0.001).In contrast, HEPES-coated 3-months-old surfaces showed anequivalent amount of albumin adsorption to that of new surfaces(p> 0.05). After 24 h of incubation, the age-related decrease was

New Ti

3-m-old Ti

HEPES-coated 3-m-old Ti

Co

ntact an

gle o

f w

ater

Fig. 2. Hydrophilic status of three different acid-etched titanium surfaces: new surfaces (months after acid etching), and HEPES-coated 3-month-old surfaces (surfaces coated with Hview images of 10 ml H2O droplets pipetted onto these titanium surfaces (left images). Meaa contact angle of 0� .

substantial for 3-months-old surfaces (p< 0.001), but this declinewasnotobserved forHEPES-coated3-months-old surfaces (p> 0.05).

3.4. Preserved capability of osteoblast attachment on HEPES-coatedtitanium surfaces

The number of osteoblasts attached to 3-months-old surfaceswas approximately 50% of that for new surfaces after 3 h incubation(p< 0.01; Fig. 3B). There was no difference between new surfacesand HEPES-coated 3-months-old surfaces (p> 0.05). The number ofcells attached even after 24 h was significantly decreased on 3-months-old surfaces compared to the new surfaces (p< 0.05).Significantly greater numbers of cells were attached to HEPES-coated 3-months-old surfaces than to new surfaces after 24 h(p< 0.05).

20

40

60

0

(d

eg

ree)

80

100

New

3-m

-old

i

HEPES-c

oate

d

3-m

-old

surfaces immediately after acid-etching), 3-month-old surfaces (surfaces stored for 3EPES immediately after acid etching and then stored for 3 months). Photographic side-n� SD contact angle of 10 ml H2O is presented in histograms (n¼ 3). Arrows indicate

0.6

0

0.4

0.2

Cell attach

men

t (W

ST

-1)

60

0

40

20

3 h

New

24 h

3-m-old

50

30

10

70A

lb

um

in

ad

so

rp

tio

n (%

)

BA

3 h 6 h 24 h

HEPES-coated 3-m-old

New

3-m-old

HEPES-coated 3-m-old

Fig. 3. Initial bioactivity of three different titanium surfaces: new, 3-month-old, and HEPES-coated 3-month-old surfaces. (A) Albumin adsorption to titanium surfaces during 2 and24 h of incubation. (B) The number of rat bone marrow-derived osteoblasts attached to titanium surfaces after incubation for 3, 6, and 24 h. Data are mean� SD (n¼ 3) for panels Aand B.

T. Suzuki et al. / Biomaterials 31 (2010) 4818e48284822

3.5. High osteoblast affinity is maintained on HEPES-coatedtitanium surfaces

Confocal microscopic images showed that osteoblasts attached tonew surfaces and HEPES-coated 3-month-old surfaces were clearlylarger than those on 3-month-old surfaces without HEPES at 3 h ofincubation (top images in Fig. 4). The cells on new surfaces andHEPES-coated 3-month-old surfaces started to show cytoskeletalarrangements as depicted by the initiation of actin fiber formation.Stretching of cell projections was also initiated, as shown by defor-mations of cell shape, whereas most cells were rounded on 3-month-old surfaces. Cytomorphometric parameters confirmed theseobservations. The three parameters were significantly greater fornew surfaces and HEPES-coated 3-month-old surfaces than for 3-month-old surfaces (p< 0.01), whereas there were no differencesbetween new surfaces and HEPES-coated 3-month-old surfaces(p> 0.05). Low magnification images (lower images) confirmed thatthe number of cells attached during this initial incubationperiodwasconsiderably greater on new surfaces and HEPES-coated 3-month-old surfaces than on 3-month-old surfaces.

3.6. Prevention of titanium age-related degradation of osteoblastproliferation and function by HEPES coating

The cell density on 3-month-old surfaces at days 2 and 5 ofculture was significantly lower than that on new surfaces (p< 0.01;Fig. 5A). The cell density on HEPES-coated 3-month-old surfaceswas comparable to that on new surfaces on both culture days(p> 0.05). The cell-based BrdU incorporation on 3-month-oldsurfaces was significantly lower than that on new surfaces andHEPES-coated 3-month-old surfaces (p< 0.05; Fig. 5B), while nodifference was found between new surfaces and HEPES-coated 3-month-old surfaces (p> 0.05). These results confirmed that the rateof proliferation was significantly reduced on 3-month-old surfaces.In contrast, such titanium age-related reduction was eliminated bycoating the 3-month-old surfaces with HEPES. ALP activity andcalcium deposition were substantially decreased (by more than50%) on 3-month-old surfaces compared to new surfaces (p< 0.01;Fig. 6A and B). This decrease was not seen on HEPES-coated 3-month-old surfaces (p> 0.05).

3.7. HEPES coating to overcome titanium age-related decrease inbone-implant integration capability

The push-in value during the early healing stage at week 2 wasreduced to less than 50% (from 43.0 N to 17.5 N) for 3-month-old

titanium implants compared to new implants (p< 0.01; Fig. 7).The push-in value for HEPES-coated 3-month-old implants wascomparable to that of the new implants (p> 0.05). During thelate healing stage at week 4, the decreased push-in valuefor 3-month-old implants remained significant compared tothat for new implants and HEPES-coated 3-month-old implants(p< 0.05). New implants and HEPES-coated 3-month-oldimplants did not significantly differ in their push-in value(p> 0.05).

3.8. Cell viability on HEPES-coated titanium

Flow cytometric analyses performed for osteoblasts incubatedfor 24 hours on three different titanium disks (new, 3-month-old,and HEPES-coated 3-month-old surfaces) showed that thepercentages of viable cells, secondary necrotic cells, and apoptoticcells did not differ among the three groups (p> 0.05; Fig. 8). Thisindicated that the HEPES coating did not adversely affect theviability of the cells or induce any cytotoxic effects that could leadto cell death.

3.9. Enhanced bioactivity after HEPES/NAC coating

For testing the potential use of HEPES coating as a drugdelivery vehicle, we examined the hydrophilic properties andinitial osteoblast behaviors on HEPES-coated 3-month-oldtitanium surfaces that contained NAC. We also sought for evidenceshowing successful transfer of NAC into the cells. Super-hydrophilicity was maintained after 3 months by HEPES coating,regardless of NAC (Fig. 9A). High magnification confocalmicroscopic images (top images in Fig. 9B) and cytomorphometryparameters (histograms in Fig. 9B) at 3 h of incubation showedthat cells were well spread with already initiated cytoskeletaldevelopment on both HEPES-coated 3-month-old surfaces andHEPES/NAC-coated 3-month-old surfaces. However, the lowmagnification images revealed a clear contrast in the numbers ofcells attached to the surfaces during this 3-h incubation. Morecells were found on HEPES/NAC-coated 3-month-old surfaces thanon HEPES-coated 3-month-old surfaces. A quantitative analysisconfirmed that the number of cells attached after 24 h wassignificantly greater for HEPES/NAC-coated 3-month-old surfacesthan for 3-month-old surfaces with HEPES alone (p< 0.05;Fig. 9C). The level of intracellular glutathione per cell considerablyincreased in osteoblasts cultured on HEPES/NAC-coated 3-month-old surfaces compared to HEPES-coated 3-month-old surfaces(p< 0.01; Fig. 9D).

200 µm

New 3-m-old

HEPES-coated

3-m-old

2000

0

1000

100

0

200

40

0

80

20

Feret’s d

iam

eter (µm

)

60

300

HEPES-coated 3-m-oldNew 3-m-old

Perim

eter (µm

)

Area (µm

2)

20 µm

Fig. 4. Initial spread and cytoskeletal arrangement of osteoblasts on three different titanium surfaces: new, 3-month-old, and HEPES-coated 3-month-old surfaces. Representativeconfocal microscopic images of cells stained with DAPI for nuclei (green) and rhodamine phalloidin for actin filaments (red), along with cytomorphometric evaluations, arepresented. Data are mean� SD (n¼ 10).

T. Suzuki et al. / Biomaterials 31 (2010) 4818e4828 4823

4. Discussion

The present study has explored a means to overcome biologicalaging of titanium by demonstrating that HEPES-coated titaniumsurfaces maintained high bioactivity and osseointegration capa-bility similar to newly prepared titanium surfaces, even after 3months of storage. In contrast, titanium surfaces without HEPEScoating had reduced biological capability by approximately 50%during this period of time, as shown by the significant reductions ofin vitro ALP and mineralization activities as well as in vivoosseointegration capability.

It was an unexpected advantage to discover the property ofnewly prepared titanium surfaces to be coated with a thin layer ofHEPES. HEPES is a buffering agent with high viscosity and it waspossible to coat HEPES onto titanium surfaces only when thesurfaces were fresh and superhydrophilic. As shown in the image inFig. 1, coating old hydrophobic titanium surfaces was technicallyimpossible. The measurement of the thickness of HEPES coatingshowed an approximately 10% of coefficient of variation. Thisindicated that the coating can be consistent and reproduced as longas new titanium surfaces are used. Also, HEPESwas prepared in thisstudy as a filter-sterilized stock solution, which has proved the

2.0

0

3.0

1.0

Brd

U in

co

rp

oratio

n/cell

(arb

itrary u

nit)

0

0.8

0.4

D 2 D 5

1.0

0.6

0.2

BA

Cell d

en

sity (W

ST

-1)

1.2

New

3-m-old

HEPES-coated 3-m-old

New

3-m-old

D 2

HEPES-coated 3-m-old

Fig. 5. Osteoblast proliferation on three different titanium surfaces: new, 3-month-old, and HEPES-coated 3-month-old surfaces. (A) Cell density at days 2 and 5 of osteoblastcultures. (B) Cell proliferative activity evaluated by BrdU incorporation per cell at day 2 of culture. Data are mean� SD (n¼ 3).

T. Suzuki et al. / Biomaterials 31 (2010) 4818e48284824

feasibility of this method without biological contamination at leastat an experimental level. We believe that these warrant the futureexploration of this method to be applicable at a manufacture levelof implant products.

Although the reason underlying the HEPES-mediated preven-tion of biological aging was not fully identified in this study, it wasnoted that the HEPES-coated surface maintained superhydro-philicity for 3 months. This state naturally disappears on baretitanium surfaces, as shown in Fig. 2 and our previous studies [1,2].The role of surface hydrophilicity of biomaterials for determiningtheir bioactivity is contentious. It is not a universal principle thathydrophilicity of the surface is proportional to the bioactivity of thematerial. For instance, poly(lactide-co-beta-benzyl malolactonate)with improved hydrophilicity promotes the attachment and

0.1

0

0.15

0.05

0.2

D 10

To

tal alkalin

e p

ho

sp

hatase

activity (arb

itrary

u

nit)

New

3-m-old

HEPES-coated 3-m-old

A

Fig. 6. Osteoblast phenotypic function on three different titanium surfaces: new, 3-month-oquantified. (B) Total calcium deposition at day 20. Data are mean� SD (n¼ 3).

proliferation of NIH3T3 fibroblasts [24], whereas polyhydroxy-alkanoates also with improved hydrophilicity reduces theirproliferation [25]. Biofilms coated with calcium phosphate thatexhibits enhanced hydrophilicity increase osteoblast proliferation[26]. In contrast, more hydrophobic polymer scaffold materials areeffective for promoting bone regeneration [27]. A series of ourprevious studies demonstrated that the maintenance of hydro-philicity on titanium surfaces appeared to have a positive effect onincreasing attachment, spread, and proliferation of osteogenic cells;however, we found no evidence to show its critical role indetermining their bioactivity and osteoconductivity [1,2,15].

In addition to the change in water wettability from hydro-phobicity to superhydrophilicity, HEPES coating must have alteredthe surface chemistry of the titanium surface. Titanium surfaces,

0

15

10

D 20

To

tal calciu

m d

ep

ositio

n

(m

g/d

L)

5

New

3-m-old

B

HEPES-coated 3-m-old

ld, and HEPES-coated 3-month-old surfaces. (A) ALP activity at day 10 colorimetrically

20

50

0

30

10

Pu

sh

-in

valu

e (N

)

W 2

40

60

W 4

New

3-m-old

HEPES-coated 3-m-old

Fig. 7. The effects of different age of titanium and HEPES coating on the strength ofbone-implant integration evaluated by biomechanical push-in test at weeks 2 and 4 ofhealing. Titanium implants with three different titanium surfaces were compared:new, 3-month-old, and HEPES-coated 3-month-old surfaces. Data are shown as themean� SD (n¼ 6).

80

0

Viab

le cells (%

)

40

60

20

10

0

Seco

nd

ary n

ectro

tic cells (%

)

5

New Ti 3-m

An

Pro

pid

iu

m io

did

e

New 3-m-old

1000

.1

.1 1000

4

21

1000

.1

.1

Fig. 8. Viability of osteoblasts 24 hours after seeding onto three different titanium surfaces(top) and percentages of viable cells (quadrant 3 on the top images), secondary necrotic cells

T. Suzuki et al. / Biomaterials 31 (2010) 4818e4828 4825

virtually composed of a thin layer of titanium dioxide, are known toabsorb hydrocarbons from the atmosphere. This chemicalcontamination is considered unavoidable and accumulates overtime, which has been demonstrated to be one of the reasons for thenatural decrease in the bioactivity of titanium [1,2]. The HEPEScoating was probably effective for protecting the titanium surfacefrom such atmosphere-originated carbon contamination. However,HEPES is also an organic chemical reagent and any chemicaldeposition from HEPES per se onto titanium surfaces needs furtherinvestigation. Another function of this type of bi-ionic bufferingagent is to maintain pH [4e7], which may have helped to improvecellular behavior and reactions in the in vitro culture environmentand possibly to control inflammatory reactions in vivo whenimplants were surgically inserted. By testing these hypotheses andexploring other unknown possibilities, mechanisms underlying theeffective preservation of bioactivity of HEPES-coated titaniumsurfaces require further investigation.

We also tested HEPES coating as a possible effective deliveryvehicle for biological molecules in an effort to enhance the bio-logical environment around titanium surfaces. Incorporation ofNAC into HEPES did not compromise the superhydrophilic nature ofthe surface after 3 months of storage. In addition, HEPES/NAC-coated titanium surfaces further enhanced the attachment ofosteoblasts, while maintaining the advantage of expedited cellspread observed on HEPES-coated surfaces. In addition to its role asa direct oxidant scavenger, NAC enhances a glutathione redox cyclewhich is themost important oxidant removal system in cells. NAC is

10

0

Ap

op

to

tic cells (%

)

5

-old Ti

HEPES-coated

3-m-old Ti

nexin V

HEPES-coated 3-m-old

1000

4

211000

.1

.1 1000

4

21

: new, 3-month-old, and HEPES-coated 3-month-old surfaces. Flow cytometric images(quadrant 2), and apoptotic cells (quadrant 4). Data are shown as the mean� SD (n¼ 3).

HEPES/NAC-coated

3-m-old Ti

HEPES-coated

3-m-old Ti

2000

0

1000

Area (µm

2)

100

0

200

Perim

eter (µm

)

40

0

80

20

Feret’s d

iam

eter (µm

)

60

0

3000.3

0

0.2

0.1

In

tracellu

lar g

lu

tath

io

ne /cell

(arb

itrary u

nit)

C

0.4HEPES-coated 3-m-old

HEPES/NAC-coated 3-m-old

24 h

20

40

60

0

Co

ntact an

gle o

f w

ater

(d

eg

ree)

80

100

BA

20 µm

200 µm

0.6

0

0.4

0.2

Cell attach

me

nt (W

ST

-1)

0.7

24 h

D

HEPES

-coated

3-m-old

HEPES/NAC

-coated

3-m-old

HEPES-coated 3-m-old

HEPES/NAC-coated

3-m-old

HEPES-coated 3-m-old

HEPES/NAC-coated

3-m-old

HEPES-coated 3-m-old HEPES/NAC-coated 3-m-old

Fig. 9. Effects of N-acetyl cysteine (NAC) incorporation into HEPES on initial behaviors and antioxidant capacity of osteoblasts as an exemplar experiment for titanium-mediateddrug delivery. Osteoblasts were cultured on HEPES-coated 3-month-old surfaces with or without NAC in HEPES. (A) Hydrophilic status of HEPES-coated 3-month-old and HEPES/NAC-coated 3month-old surfaces. Photographic side-view images of 10 ml H2O droplets pipetted onto these titanium surfaces (top images). Mean� SD contact angle of 10 ml H2O(histograms; n¼ 3). Arrows indicate a contact angle of 0� . (B) Representative confocal microscopic images of cells cultured on HEPES-coated 3-month-old and HEPES/NAC-coated3month-old surfaces for 3 h, stained with DAPI for nuclei (green) and rhodamine phalloidin for actin filaments (red), along with cytomorphometric evaluations, are presented. Dataare mean� SD (n¼ 10). (C) The number of cells attached to these surfaces 24 h after seeding. Data are mean� SD (n¼ 3). (D) Intracellular glutathione per cell evaluated at 24 h ofosteoblastic culture. Data are mean� SD (n¼ 3).

T. Suzuki et al. / Biomaterials 31 (2010) 4818e48284826

easily deacetylated into cysteine, which is an important precursorof glutathione [8], and promotes the cellular glutathione system[28,29]. The use of NAC with various free radical-producingbiomaterials has been shown to be effective for improving theirbiological capability, as well as removing or alleviating theircytotoxicity and tissue toxicity. As mentioned in the Introductionsection, a NAC-incorporated PMMA resin-based bone cementmaterial was successfully detoxified and possessed osteo-conductivity [11]. During standard cell culture procedures,including trypsin treatment and addition of new culture medium,cells are assumed to encounter a variety of reactive oxygen species(ROS) from these exogenous chemicals. In addition, cell detach-ment and seeding procedures involve physical stress to cells thatgenerate endogenous ROS. NAC may have directly scavenged theseROS and also increased the antioxidant capacity of cells byincreasing their intracellular glutathione reserves [11,13]. In otherwords, synergistic functions of NAC in oxidant removal andcytoprotection may have resulted in the additional increase in cellattachment. In fact, the increased level of intracellular glutathione(Fig. 9D) indicated that NAC coated on titanium surfaces wastransferred into the cells and supplied the glutathione precursors,

cysteine molecules. Glutathione is a cytoplasmic tripeptide ofL-glutamyl-L-cysteinyl glycine. Because glutamic acid and glycineare plentiful in cells, the availability of cysteine is known to controlthe rate of glutathione synthesis. Future in vivo studies usingHEPES/NAC-coated titanium implants would be of great interest.Furthermore, other bioenhancing molecules, such as cytokines orgrowth factors, could be effectively delivered around titaniumimplants in vivo using HEPES coating as a carrier vehicle.

Titanium implants are widely used as anchors in reconstructiveand restorative surgeries for osteoporotic fractures, diseased joints,and edentulous jaws. Despite the growing demand for such implantprocedures in a universally aging society, several issues directlyrelated to the biological capability of titanium remain unresolved.With orthopedic implants, the treatment outcome includes a highpercentage of revision surgery ranging from 5 to 40% (average, 25%)[30e34]. The average functional lifetime of the currently usedorthopedic titanium implants is only 10e25 years [35]. For dentalimplants, the protracted healing time required before fabricatingprosthetic teeth and low bone metabolic potential of the host bonelimit this application [23,36e38]. Rapid, firm establishment ofbone-titanium implant integration has been a persistent challenge

T. Suzuki et al. / Biomaterials 31 (2010) 4818e4828 4827

in these fields [39e41]. A crucial biological question remainedunaddressed: Why does bone tissue not form completely aroundtitanium implants? The implant area eventually covered by bone(bone-titanium contact percentage) remained at 45�16% [42], or50e65% [43], which were far below the ideal 100%. Most implantsfail because of incomplete establishment, or early/late destructivechanges at the bone-implant interface [44e46]. The recentdiscovery of biological aging of titanium has provided an answer tothis question, albeit partially. Simply put, the use of new surfacescan considerably improve the bone-implant contact as high as 90%,as opposed to 60% obtained by 4-week-old implants with anidentical surface topography. This demonstrates the in vivo signif-icance of the biological aging of titanium and implies the clinicalsignificance of the phenomenon. Orthopedic and dental implantproducts are sold as storable medical devices. These products age inan unavoidable manner during their inventory phase and distri-bution, as well as during storage before use. At present there is noeffective means to prevent this phenomenon. Although themechanisms need to be identified, the effect of HEPES coating forovercoming biological aging of titanium, at least during 3 monthsas tested in the present study, may be a new surface technology fortitanium implants.

It was recently discovered that ultraviolet (UV) light treatmentof titanium, with an acquisition of superhydrophilicity, enableda nearly 100% of bone-implant contact, and remarkably acceleratedand enhanced the establishment of bone-titanium implant inte-gration [2,15]. The effect has proven feasible for different surfacetypes of titanium. However, there remains a crucial problem for thistechnology. Biological aging also applies to this surface, as thesuperhydrophilic surface induced by UV light does not last long dueto the natural release of energy and progressive carbon contami-nation over time. Such adverse transformation (i.e., a significantreduction in bioactivity) takes place in 1e2 weeks [15]. Thisproblem prevents using this UV technology at the manufacturelevel, as it is assumed that the UV-treated titanium surfaces losetheir biological advantages during circulation or storage periods.Given the effective maintenance of superhydrophilicity, as well ashigh bioactivity and osseointegration capability, we believe HEPES-coating could also be an effective measure for preserving UV-functionalized titanium surfaces. Future studies should be plannedto address this important challenge. Finally, it should be added thatthe procedure of HEPES coating is simple, safe, and has a lower cost,and can also be used as a carrier vehicle to deliver biologicalmolecules for anticipated synergistic effects that lead to improvedperi-implant osteogenesis.

5. Conclusions

This study tested the potential utility of HEPES coating on tita-nium surfaces to overcome the biological aging of titanium. HEPES,a highly viscous nonvolatile buffering agent, could be coated ontonewly prepared superhydrophilic titanium surfaces, but not onhydrophobic titanium surfaces that were old. Three differentlyconditioned titanium surfaces were tested for their hydrophilicproperties and biological capabilities: (1) new surfaces (immediatelyafter acid etching), (2) 3-month-old surfaces (stored 3 months afteracid etching) and HEPES-coated 3-month-old surfaces (titaniumsurfaces coated with HEPES immediately after acid etching and thenstored for 3 months). New surfaces and 3-month-old surfaces withHEPES were superhydrophilic, while 3-month-old surfaces withoutHEPES were hydrophobic. Albumin adsorption capacity of 3-month-old surfaces was only 20e40% of that for new surfaces. The numberof osteoblasts attached to 3-month-old surfaces within 24 hwas substantially lower than that for new surfaces. The spreadand cytoskeletal development of osteoblasts were delayed on

3-month-old surfaces. The number of proliferating cells, alkalinephosphatase activity, calcium deposition in osteoblastic cultures,and the strength of boneeimplant integration in vivo, for 3-month-old titanium surfaces, were reduced to 50% of those for new tita-nium surfaces. In contrast, HEPES-coated 3-month-old titaniumsurfaces exhibited in vitro bioactivity and in vivo osseointegrationcapability equivalent to those of new titanium surfaces. The HEPES-coated 3-month-old titanium surfaces that also contained an anti-oxidant amino acid derivative, NAC, further enhanced osteoblastattachment. Thus, we have demonstrated that HEPES coating ofnewly prepared titanium surfaces can maintain their high bioac-tivity and osteoconductivity for at least 3 months; otherwise, thebioactivity of titanium surfaces naturally reduced to less than 50%.This indicates the potential usefulness of HEPES coating asa preventive technology for the biological aging of titanium. HEPEScoating was also able to deliver biomolecules into cells at the locallevel on titanium surfaces, establishing an expectation for a tech-nological synergy between the anti-aging effect and titanium-mediated drug delivery.

Acknowledgements

This work was supported by JAMSEA. This study was conductedin a facility constructed with support from the Research FacilitiesImprovement Program, grant No. C06RR014529, of the NationalCenter for Research Resources, National Institute of Health.

Appendix

Figures with essential colour discrimination. Several of thefigures in this article have parts that are difficult to interpret inblack and white. The full colour images can be found in the on-lineversion, at doi:10.1016/j.biomaterials.2010.02.061.

References

[1] Att W, Hori N, Takeuchi M, Ouyang J, Yang Y, Anpo M, et al. Time-dependentdegradation of titanium osteoconductivity: an implication of biological agingof implant materials. Biomaterials 2009;30:5352e63.

[2] Att W, Hori N, Iwasa F, Yamada M, Ueno T, Ogawa T. The effect of UV-photo-functionalization on the time-related bioactivity of titanium and chromium-cobalt alloys. Biomaterials 2009;30:4268e76.

[3] Hori N, Att W, Ueno T, Sato N, Yamada M, Saruwatari L, et al. Age-dependentdegradation of the protein adsorption capacity of titanium. J Dent Res2009;88:663e7.

[4] Hammerman H, Ramadan R, Hir J, Moscovitz M. Beneficial effect of HEPESbuffer in repeated coronary reperfusion. Coron Artery Dis 2000;11:179e82.

[5] Kasraian K, Kuzniar A, Earley D, Kamicker BJ, Wilson G, Manion T, et al. Sus-tained in vivo activity of recombinant bovine granulocyte colony stimulatingfactor (rbG-CSF) using HEPES buffer. Pharm Dev Technol 2001;6:441e7.

[6] Itagaki A, Kimura G. Tes and HEPES buffers in mammalian cell cultures andviral studies: problem of carbon dioxide requirement. Exp Cell Res 1974;83:351e61.

[7] Chagnon A, Corbeil M. Use of an organic buffer (HEPES) in human lympho-cytoid cell line cultures. In Vitro 1973;8:283e7.

[8] Lean JM, Davies JT, Fuller K, Jagger CJ, Kirstein B, Partington GA, et al. A crucialrole for thiol antioxidants in estrogen-deficiency bone loss. J Clin Invest2003;112:915e23.

[9] Meister A, Anderson ME. Glutathione. Annu Rev Biochem 1983;52:711e60.[10] Stanislawski L, Lefeuvre M, Bourd K, Soheili-Majd E, Goldberg M, Perianin A.

TEGDMA-induced toxicity in human fibroblasts is associated with early anddrastic glutathione depletion with subsequent production of oxygen reactivespecies. J Biomed Mater Res A 2003;66:476e82.

[11] Tsukimura N, Yamada M, Aita H, Hori N, Yoshino F, Chang-Il Lee M, et al.N-acetyl cysteine (NAC)-mediated detoxification and functionalization of poly(methyl methacrylate) bone cement. Biomaterials 2009;30:3378e89.

[12] Yamada M, Kojima N, Paranjpe A, Att W, Aita H, Jewett A, et al. N-acetylcysteine (NAC)-assisted detoxification of PMMA resin. J Dent Res 2008;87:372e7.

[13] Yamada M, Ogawa T. Chemodynamics underlying N-acetyl cysteine-mediatedbone cement monomer detoxification. Acta Biomater 2009;5:2963e73.

[14] Paranjpe A, Sung EC, Cacalano NA, Hume WR, Jewett A. N-acetyl cysteineprotects pulp cells from resin toxins in vivo. J Dent Res 2008;87:537e41.

T. Suzuki et al. / Biomaterials 31 (2010) 4818e48284828

[15] Aita H, Hori N, Takeuchi M, Suzuki T, Yamada M, Anpo M, et al. The effect ofultraviolet functionalization of titanium on integration with bone. Biomate-rials 2009;30:1015e25.

[16] Iwasa F, Hori N, Ueno T, Minamikawa H, Yamada M, Ogawa T. Enhancement ofosteoblast adhesion to UV photofunctionalized titanium via an electrostaticmechanism. Biomaterials 2010;31:2717e27.

[17] Ueno T, Yamada M, Suzuki T, Minamikawa H, Sato N, Hori N, et al. Enhance-ment of bone-titanium integration profile with UV-photofunctionalized tita-nium in a gap healing model. Biomaterials 2010;31:1546e57.

[18] Yang G, Obiakor H, Sinha RK, Newman BA, Hood BL, Conrads TP, et al.Activation-induced deaminase cloning, localization, and protein extractionfrom young VH-mutant rabbit appendix. Proc Natl Acad Sci U S A 2005;102:17083e8.

[19] Gembitsky DS, Lawlor K, Jacovina A, Yaneva M, Tempst P. A prototype antibodymicroarray platform to monitor changes in protein tyrosine phosphorylation.Mol Cell Proteomics 2004;3:1102e18.

[20] Paratcha G, Ledda F, Ibanez CF. The neural cell adhesion molecule NCAM is analternative signaling receptor for GDNF family ligands. Cell 2003;113:867e79.

[21] Saruwatari L, Aita H, Butz F, Nakamura HK, Ouyang J, Yang Y, et al. Osteoblastsgenerate harder, stiffer, and more delamination-resistant mineralized tissueon titanium than on polystyrene, associated with distinct tissue micro- andultrastructure. J Bone Miner Res 2005;20:2002e16.

[22] Ogawa T, Ozawa S, Shih JH, Ryu KH, Sukotjo C, Yang JM, et al. Biomechanicalevaluation of osseous implants having different surface topographies in rats.J Dent Res 2000;79:1857e63.

[23] Ozawa S, Ogawa T, Iida K, Sukotjo C, Hasegawa H, Nishimura RD, et al.Ovariectomy hinders the early stage of bone-implant integration: histo-morphometric, biomechanical, and molecular analyses. Bone 2002;30:137e43.

[24] He B, Wan Y, Bei J, Wang S. Synthesis and cell affinity of functionalized poly(L-lactide-co-beta-malic acid)withhighmolecularweight. Biomaterials2004;25:5239e47.

[25] Wang YW, Wu Q, Chen GQ. Reduced mouse fibroblast cell growth by increasedhydrophilicity of microbial polyhydroxyalkanoates via hyaluronan coating.Biomaterials 2003;24:4621e9.

[26] Ber S, Torun Kose G, Hasirci V. Bone tissue engineering on patterned collagenfilms: an in vitro study. Biomaterials 2005;26:1977e86.

[27] Jansen EJ, Sladek RE, Bahar H, Yaffe A, Gijbels MJ, Kuijer R, et al. Hydropho-bicity as a design criterion for polymer scaffolds in bone tissue engineering.Biomaterials 2005;26:4423e31.

[28] Gillissen A, Bartling A, Schoen S, Schultze-Werninghaus G. Antioxidant func-tion of ambroxol in mononuclear and polymorphonuclear cells in vitro. Lung1997;175:235e42.

[29] Gillissen A, Nowak D. Characterization of N-acetylcysteine and ambroxol inanti-oxidant therapy. Respir Med 1998;92:609e23.

[30] Espehaug B, Furnes O, Havelin LI, Engesaeter LB, Vollset SE. The type of cementand failure of total hip replacements. J Bone Joint Surg Br 2002;84:832e8.

[31] Hudson JI, Kenzora JE, Hebel JR, Gardner JF, Scherlis L, Epstein RS, et al. Eight-year outcome associated with clinical options in the management of femoralneck fractures. Clin Orthop Relat Res; 1998:59e66.

[32] Lu-Yao GL, Keller RB, Littenberg B, Wennberg JE. Outcomes after displacedfractures of the femoral neck. A meta-analysis of one hundred and six pub-lished reports. J Bone Joint Surg Am 1994;76:15e25.

[33] Tidermark J, Ponzer S, Svensson O, Soderqvist A, Tornkvist H. Internal fixationcompared with total hip replacement for displaced femoral neck fractures inthe elderly. A randomised, controlled trial. J Bone Joint Surg Br 2003;85:380e8.

[34] Ravikumar KJ, Marsh G. Internal fixation versus hemiarthroplasty versus totalhip arthroplasty for displaced subcapital fractures of femur e 13 year resultsof a prospective randomised study. Injury 2000;31:793e7.

[35] Martinez de Aragon JS, Keisu KS. 21-year results of the uncemented fullytextured lord hip prosthesis. Clin Orthop Relat Res 2007;454:133e8.

[36] van Steenberghe D, Jacobs R, Desnyder M, Maffei G, Quirynen M. The relativeimpact of local and endogenous patient-related factors on implant failure upto the abutment stage. Clin Oral Implants Res 2002;13:617e22.

[37] Nevins ML, Karimbux NY, Weber HP, Giannobile WV, Fiorellini JP. Woundhealing around endosseous implants in experimental diabetes. Int J OralMaxillofac Implants 1998;13:620e9.

[38] Zhang H, Lewis CG, Aronow MS, Gronowicz GA. The effects of patient age onhuman osteoblasts' response to Ti-6Al-4V implants in vitro. J Orthop Res2004;22:30e8.

[39] LeGeros RZ, Craig RG. Strategies to affect bone remodeling: osteointegration.J Bone Miner Res 1993;8(Suppl 2):S583e96.

[40] Puleo DA, Nanci A. Understanding and controlling the bone-implant interface.Biomaterials 1999;20:2311e21.

[41] Pilliar RM. Cementless implant fixation e toward improved reliability. OrthopClin North Am 2005;36:113e9.

[42] Weinlaender M, Kenney EB, Lekovic V, Beumer 3rd J, Moy PK, Lewis S. His-tomorphometry of bone apposition around three types of endosseous dentalimplants. Int J Oral Maxillofac Implants 1992;7:491e6.

[43] Ogawa T, Nishimura I. Different bone integration profiles of turned and acid-etched implants associated with modulated expression of extracellular matrixgenes. Int J Oral Maxillofac Implants 2003;18:200e10.

[44] Moy PK, Medina D, Shetty V, Aghaloo TL. Dental implant failure rates andassociated risk factors. Int J Oral Maxillofac Implants 2005;20:569e77.

[45] Esposito M, Hirsch JM, Lekholm U, Thomsen P. Failure patterns of fourosseointegrated oral implant systems. J Mater Sci Mater Med 1997;8:843e7.

[46] Chuang SK, Wei LJ, Douglass CW, Dodson TB. Risk factors for dental implantfailure: a strategy for the analysis of clustered failure-time observations. J DentRes 2002;81:572e7.