Osteoblast adhesion on nanophase ceramics

Osteoblast adhesion on nanophase ceramics

Biomaterials 20 (1999) 1221 } 1227 Osteoblast adhesion on nanophase ceramics Thomas J. Webster , Richard W. Siegel, Rena Bizios * Department of Bio...

337KB Sizes 0 Downloads 19 Views

Biomaterials 20 (1999) 1221 } 1227

Osteoblast adhesion on nanophase ceramics Thomas J. Webster , Richard W. Siegel, Rena Bizios * Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, NY 12180-3590, USA Department of Materials Science and Engineering, Rensselaer Polytechnic Institute, Troy, NY 12180-3590, USA Received 7 December 1998; accepted 3 February 1999

Abstract Osteoblast adhesion on nanophase alumina (Al O ) and titania (TiO ) was investigated in vitro. Osteoblast adhesion to nanophase    alumina and titania in the absence of serum from Dulbecco's modi"ed Eagle medium (DMEM) was signi"cantly (P(0.01) less than osteoblast adhesion to alumina and titania in the presence of serum. In the presence of 10% fetal bovine serum in DMEM osteoblast adhesion on nanophase alumina (23 nm grain size) and titania (32 nm grain size) was signi"cantly (P(0.05) greater than on conventional alumina (177 nm grain size) and titania (2.12 lm grain size), respectively, after 1, 2, and 4 h. Further investigation of the dependence of osteoblast adhesion on alumina and titania grain size indicated the presence of a critical grain size for osteoblast adhesion between 49 and 67 nm for alumina and 32 and 56 nm for titania. The present study provides evidence of the ability of nanophase alumina and titania to simulate material characteristics (such as surface grain size) of physiological bone that enhance protein interactions (such as adsorption, con"guration, bioactivity, etc.) and subsequent osteoblast adhesion.  1999 Elsevier Science Ltd. All rights reserved Keywords: Nanophase; Ceramics; Alumina; Orthopaedic/dental; Osteoblast; Adhesion

1. Introduction Traditional ceramics (such as alumina used in the treatment of hand, elbow fractures, edentations, and in anthroplasty) have long been appreciated for their biocompatibility but have often clinically failed due to lack of direct bonding with bone, that is, insu$cient osseointegration [1, 2]. Novel ceramic formulations (such as hydroxyapatite, bioglasses, bioactive glass}ceramics, and calcium phosphate) have been shown to enhance formation of new bone mineralized matrix [3}6]. The mechanical properties (speci"cally, ductility and toughness) of these biosubstitutes, however, are generally not comparable to physiological bone and, consequently, use of these materials in orthopaedic/dental applications have been limited [7]. The extent of osseointegration between bone and a newly implanted material is in#uenced by many factors including a number of host biological and surrounding tissue responses. Properties of the biomaterial surface (such as topography and chemistry) control the type and magnitude of cellular and molecular events at the tissue} implant interface. Design of biomaterials with surface * Corresponding author. Fax: 001 518 276 3035.

properties similar to physiological bone (characterized by surface grain sizes in the nanometer regime [8]) would undoubtedly aid in the formation of new bone at the tissue/biomaterial interface and, therefore, improve orthopaedic/dental implant e$cacy. To date, there have been no successful attempts to design and produce a biocompatible substitute which simulates the surface topography of normal, healthy bone. With the advent of nanostructured materials (materials with grain sizes less than 100 nm in at least one direction [9]) it may now be possible to synthesize materials for orthopaedic and dental applications which simulate the surface properties of physiological bone. The present in vitro study is the "rst of its kind to investigate the potential use of nanomaterials for orthopaedic/dental applications and provides the "rst evidence of signi"cantly enhanced osteoblast (the bone forming cells) adhesion on both nanophase alumina and titania. 2. Materials and methods 2.1. Substrates Alumina (Al O ) and titania (TiO ) samples (circular    disks 10 mm in diameter and 2 mm thick) were prepared

0142-9612/99/$ - see front matter  1999 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 2 - 9 6 1 2 ( 9 8 ) 0 0 0 2 0 - 4

1222

T.J. Webster et al. / Biomaterials 20 (1999) 1221 } 1227

by compacting, respectively, nanophase (20, 23, 24, 33, and 67 nm grain size) alumina (c-phase) and nanophase (20, 26, 32, and 56 nm) titania (90 wt% anatase and 10 wt% rutile phases) powders (Nanophase Technologies Corp.) in a tool-steel die via a uniaxial pressing cycle (0.2 to 1 GPa over a 10 min period). Nanophase alumina samples were obtained by sintering (in air at 103C/min) the 23 nm grain size alumina compacts from room temperature to a "nal temperature of either 1000 or 11003C and by maintaining either the 1000 or 11003C temperature, respectively, for 2 h to obtain material formulations with grain sizes less than 100 nm [10]. Conventional alumina samples were obtained by sintering (in air at 103C/min) the 23 nm alumina compacts from room temperature to a "nal temperature of 12003C and by maintaining this temperature for 2 h to obtain materials with grain sizes greater than 100 nm [10]. Nanophase titania samples were obtained by sintering (in air at 103C/min) the 32 nm grain size titania compacts from room temperature to a "nal temperature of either 600 or 8003C and by maintaining the 600 or 8003C temperature, respectively, for 2 h to obtain materials with grain sizes less than 100 nm. Conventional titania samples were obtained by sintering (in air at 103C/min) the 32 nm titania compacts from room temperature to a "nal temperature of 12003C and by maintaining this temperature for 2 h to obtain materials with grain sizes greater than 100 nm. Borosilicate glass coverslips (reference material) were etched in 1 N NaOH and prepared for cell culture experiments according to standard protocols [11]. All samples were degreased, ultrasonically cleaned, sterilized (in a steam autoclave at 1203C for 30 min) according to the standard laboratory procedures [11] and used without modi"cation in experiments with cells. 2.2. Bulk material characterization Alumina and titania grain size was calculated by averaging multiple-point Brunauer, Emmett, and Teller (BET) measurements on the nanophase and conventional ceramic formulations of interest to the present study. Measurements were run in triplicate per substrate type. 2.3. Surface characterization Aqueous wettability of alumina, titania, and borosilicate glass (reference material) substrates were analyzed by contact angle measurements. The wetting angle was calculated using standard procedures [12] with image analysis software (Image Pro). Measurements were run in triplicate per substrate type and repeated at three di!erent times. Nanophase and conventional alumina and titania topography was evaluated by atomic force microscopy

(using an Autoprobe CP Atomic Force Microscope). Image analysis software (Pro-scan version 1.5b) was used to generate micrographs and to quantitatively compare surface roughness and surface area of the materials of interest in the present study. Measurements were run in triplicate per substrate type. 2.4. Cell cultures Osteoblasts were isolated via sequential collagenase digestions of neonatal rat calvaria (6}8 pups per litter, per isolation) according to established protocols [11] and cultured in Dulbecco's modi"ed Eagle medium (DMEM; Gibco), supplemented with 10% fetal bovine serum (Gibco) in a 373C, humidi"ed, 5% CO /95% air  environment. The osteoblastic phenotype of the cells was determined by alkaline phosphatase activity and the formation of mineral deposits in the extracellular matrix. Osteoblasts at passage numbers 2}4 were used in the experiments. 2.5. Cell adhesion Osteoblasts were enzymatically lifted from polystyrene tissue culture #asks using less than 1 ml of low-trypsin EDTA (Sigma) before suspension in DMEM (in the presence or absence of 10% fetal bovine serum). Osteoblasts (3500 cells/cm) in DMEM (in the presence of 10% fetal bovine serum) were seeded per substrate and allowed to adhere in a 373C, humidi"ed, 5% CO /95% air environ ment for 0.5, 1, 2, and 4 h; controls served as similar experimental runs with the exception that osteoblasts were seeded in DMEM without serum. After each prescribed time period, non-adherent osteoblasts were removed by rinsing in phosphate bu!ered saline. Osteoblasts adherent on opaque alumina and titania were "xed with 4% formaldehyde in sodium phosphate bu!er and stained with Hoechst (No. 33342; Sigma); the cell nuclei were, thus, visualized and counted in situ using #uorescence (365 nm excitation; 400 nm emission) microscopy with image analysis software (Image Pro). Osteoblasts adherent on glass were "xed with 4% formaldehyde in sodium phosphate bu!er, stained with Coomassie Blue and counted using light microscopy with image analysis software (Image Pro). Osteoblasts from di!erent isolations of neonatal rat calvaria were used in this study; cells from one and the same batch, however, were used per experimental run. The adhesion experiments were run in triplicate and repeated at three di!erent times per substrate type. Cell density (cells/cm) was determined by averaging the number of adherent cells in "ve random "elds per substrate. Cell adhesion density was analyzed statistically using standard analysis of variance (ANOVA) techniques; statistical signi"cance was considered at P(0.05.

T.J. Webster et al. / Biomaterials 20 (1999) 1221 } 1227

1223

3. Results and discussion 3.1. Bulk material characterization As determined through multiple-point BET measurements, the average grain size diameter of the as-pressed, alumina compacts did not change when sintered at 10003C; in contrast, the average grain size of the aspressed 23 nm alumina compacts increased to 49 and 177 nm (a value in the range of conventional material formulations) when sintered at 1100 and 12003C, respectively. Similarly, the average grain size diameter of the aspressed, titania compacts did not change when sintered at 6003C; in contrast, the average grain size of the aspressed 32 nm titania compacts increased to 97 nm and 2.12 lm (a value in the range of conventional material formulations) when sintered at 800 and 12003C, respectively. 3.2. Surface characterization Contact angles for the 23 nm grain size (nanophase) alumina were signi"cantly (P(0.01) lower than for the 177 nm grain size (conventional) alumina (Table 1). Similarly, contact angles for the 32 and 97 nm grain size (nanophase) titania were signi"cantly (P(0.01) lower than for the 2.12 lm grain size (conventional) titania (Table 2). The decrease in contact angle measurements Table 1 Surface wettability of nanophase and conventional alumina. Aqueous wettability values of alumina and borosilicate glass (reference material) substrates was calculated using image analysis software (Image Pro) Material

Contact angle (deg)

Borosilicate glass (reference material) Alumina grain size 177 nm (Conventional) 49 nm (Nanophase) 23 nm (Nanophase)

17.3$1.1 18.6$0.9 10.8$1.3 6.4$0.7

Values are mean$SEM; n"3. P(0.01 (compared to 177 nm (conventional) grain size aluminia).

Table 2 Surface wettability of nanophase and conventional titania. Aqueous wettability values of titania and borosilicate glass (reference material) substrates was calculated using image analysis software (Image Pro) Material

Contact angle (deg)

Borosilicate glass (reference material) Titania grain size 2.12 lm (Conventional) 97 nm (Nanophase) 32 nm (Nanophase)

17.3$1.1 26.8$2.8 18.1$3.2 2.2$0.1

Values are mean$SEM; n"3. P(0.01 (compared to 2.12 lm (conventional) grain size titania).

Fig. 1. Atomic force micrographs of alumina. Representative nanophase (a) and conventional (b) alumina topography as evaluated by atomic force microscopy. Image analysis software (Pro-scan version 1.5b) was used to quantitatively compare surface roughness and surface area.

corresponds to an increase in surface aqueous wettability and, thus, an increase in hydrophilicity and surface reactivity for the nanophase alumina and titania formulations. Atomic force microscopy data provided evidence that, compared to the 177 nm (conventional) grain size alumina, the 23 nm grain size (nanophase) alumina possesses higher surface roughness (17 versus 20 nm , respectively) and larger surface area (1.15 lm versus 1.73 lm/ lm projected area, respectively; Fig. 1). Similarly, compared to the 2.12 lm grain size titania, the 32 nm (nanophase) titania exhibited higher surface roughness (16 versus 32 nm, respectively) and larger surface area (1.07 lm versus 1.45 lm/lm projected area, respectively; Fig. 2). 3.3. Osteoblast adhesion on nanophase ceramics Compared to cell adhesion in the presence of serum in the cell culture medium, osteoblast adhesion to alumina

1224

T.J. Webster et al. / Biomaterials 20 (1999) 1221 } 1227

Fig. 2. Atomic force micrographs of titania. Representative nanophase (a) and conventional (b) titania topography as evaluated by atomic force microscopy. Image analysis software (Pro-scan version 1.5b) was used to quantitatively compare surface roughness and surface area.

and titania in the absence of serum was signi"cantly (P(0.01) less for all time periods tested in the present study (data not shown). In the presence of 10% fetal bovine serum, osteoblast adhesion on nanophase, 23 nm grain size alumina and on 32 nm grain size titania, was signi"cantly (P(0.01) greater than on borosilicate glass (reference substrate) after 0.5, 1, 2, and 4 h (Figs. 3 and 4). More importantly, compared to osteoblast adhesion on conventional (177 nm grain size) alumina and on conventional (2.12 lm grain size) titania, osteoblast adhesion on nanophase (23 nm grain size) alumina and titania (32 nm grain size) was signi"cantly (P(0.05) greater after 1, 2, and 4 h, respectively; in fact, compared to conventional alumina and titania, osteoblast adhesion increased by 46 and 30% on nanophase 23 nm grain size alumina and on nanophase 32 nm grain size titania, respectively, after 4 h. Furthermore, osteoblast adhesion was signi"cantly (P(0.01) greater on the 20 nm grain size nanophase alumina than on the 177 nm grain size conventional alumina after 4 h (Fig. 5); in fact, osteoblast adhesion was signi"cantly (P(0.05) greater on alumina with grain sizes in the range of 23}49 nm than on alumina with grain sizes in the range of 67 and 177 nm. Moreover, the average osteoblast adhesion on alumina with grain sizes less than 49 nm was 52% greater than on alumina with grain sizes greater than 67 nm after 4 h (3125 versus 2050 cells/cm). Similarly, osteoblast adhesion was significantly (P(0.05) greater on titania with grain sizes in the range of 20}32 nm than on titania with grain sizes in the range of 56}2.12 lm (Fig. 6). The average osteoblast adhesion on titania with grain sizes less than 32 nm was 24% greater than on titania with grain sizes greater than 56 nm after 4 h (2589 versus 1961 cells/cm). These

Fig. 3. Time course of osteoblast adhesion on alumina. Rat calvarial osteoblasts (3500 cell/cm) in Dulbecco's modi"ed Eagle medium containing 10% fetal bovine serum were seeded on the following substrates: (䊐) borosilicate glass (reference substrate); (k) 77 nm grain size alumina (conventional); (䊏) 49 nm grain size alumina (nanophase); (Z) 23 nm grain size alumina (nanophase). Cell adhesion under standard cell culture conditions (373C, humidi"ed, 5% CO /95% air environment) was determined at 0.5, 1, 2, and 4 h. Values are mean$SEM; n"3; **P(0.01;  *P(0.05 (compared to reference substrate).

T.J. Webster et al. / Biomaterials 20 (1999) 1221 } 1227

1225

Fig. 4. Time course of osteoblast adhesion on titania. Rat calvarial osteoblasts (3500 cell/cm) in Dulbecco's modi"ed Eagle medium containing 10% fetal bovine serum were seeded on the following substrates: (䊐) borosilicate glass (reference substrate); (k) 2.12 lm grain size titania (conventional); (䊏) 97 nm grain size titania (nanophase); (Z) 32 nm grain size titania (nanophase). Cell adhesion under standard cell culture conditions (373C, humidi"ed, 5% CO /95% air environment) was determined at 0.5, 1, 2, and 4 h. Values are mean$SEM; n"3; **P(0.01 (compared to reference substrate). 

Fig. 5. Osteoblast adhesion on alumina of various grain sizes. Rat calvarial osteoblasts (3500 cell/cm) in Dulbecco's modi"ed Eagle medium, containing 10% fetal bovine serum, were allowed to adhere on the surfaces of various grain size alumina and borosilicate glass (reference material) under standard cell culture conditions (373C, humidi"ed, 5% CO /95% air environment) for 4 h. Values are mean$SEM; n"3; **P(0.01;  *P(0.05 (compared to 177 nm grain size alumina); SP(0.05 (compared to reference substrate).

results imply that there may be a critical grain size (between 49 and 67 nm for alumina and between 32 and 56 nm for titania) that plays a crucial role in mediating osteoblast adhesion to nanophase ceramics. A possible explanation for the observed increase in osteoblast adhesion with decreasing alumina and titania grain size could be directly related to the greater surface area exhibited by the nanomaterial formulations (for example, compared to the 177 nm grain size alumina, the 23 nm grain size alumina has approximately 50% more surface area for cell adhesion and compared to the 2.12 lm grain size titania, the 32 nm grain size titania has approximately 35% more surface area for cell adhesion); an increase in nanomaterial surface area would, undoubtedly, correlate to an increase in osteoblast adhe-

sion. However, due to the presence of a critical grain size for osteoblast adhesion on alumina and on titania, the present results indicate that an increase in surface area is not the only factor contributing to enhanced osteoblast adhesion. Moreover, since osteoblast adhesion was signi"cantly less on either nanophase alumina or titania in the absence of serum in the cell culture media, the present data imply that nanophase ceramics may promote interactions (such as adsorption, con"guration, bioactivity, etc.), of select serum protein(s) which, subsequently, enhance osteoblast adhesion. Since proteins mediate adhesion of anchorage-dependent cells, and thus in#uence subsequent cellular functions (such as cell proliferation, deposition of calcium-containing mineral deposits, etc.), the mechanism(s) of protein interactions with nanophase

1226

T.J. Webster et al. / Biomaterials 20 (1999) 1221 } 1227

Fig. 6. Osteoblast adhesion on titania of various grain sizes. Rat calvarial osteoblasts (3500 cell/cm) in Dulbecco's modi"ed Eagle medium, containing 10% fetal bovine serum, were allowed to adhere on the surfaces of various grain size alumina and borosilicate glass (reference material) under standard cell culture conditions (373C, humidi"ed, 5% CO /95% air environment) for 4 h. Values are mean$SEM; n"3; *P(0.01  (compared to reference substrate).

alumina and titania need to be further investigated and elucidated. Furthermore, since an increase in osteoblast adhesion was observed for both nanophase alumina and titania, the present study provides evidence that enhanced adhesion is independent of the speci"c chemistry of the material surface and dependent only on the optimal surface topography of nanophase ceramics. The ability of nanophase ceramics to simulate material properties (such as surface topography) of physiological bone (characterized by surface grain sizes in the nanometer regime [8]) constitute design parameters that promise improved orthopaedic/dental implant e$cacy. To date, there has been no diligent, persistent and successful attempts to design and produce a biocompatible substitute which simulates the surface topography of normal, healthy bone. The present study is the "rst of its kind to investigate the potential use of nanomaterials for orthopaedic/dental applications and provides the "rst evidence of enhanced osteoblast adhesion on nanophase alumina and titania. Since for anchorage-dependent cells (like osteoblasts), adhesion is the prerequisite to subsequent cell functions (e.g., proliferation, synthesis of extracellular matrix proteins, formation of mineral deposits, etc.), enhanced other cell responses could also be expected from cell interaction with nanophase ceramics. The results of the present study, thus, provide evidence that nanophase ceramics could promote osseointegration which is critical for the clinical success of orthopaedic/dental implants.

Acknowledgements The authors would like to thank Dr. Sandra Schujman (Department of Physics, Rensselaer Polytechnic Insti-

tute) for assistance with BET measurements; Dr. Abigail Synder-Keller (Wadsworth Center, New York State Department of Health), Carol Charniga and Dr. Himmelberg (Division of Neurosurgery Research Laboratory, Albany Medical College) for neonatal rat calvaria; Mr. Brian Frank and Dr. Georges Belfort (Department of Chemical Engineering, Rensselaer Polytechnic Institute) for assistance with atomic force microscopy; Dr. Natacha DePaola (Department of Biomedical Engineering, Rensselaer Polytechnic Institute) for assistance with #uorescence microscopy and permission to use image analysis software; Nanophase Technologies for the nanophase alumina and titania powder; and Ms. Stacey McManus (participant in the Undergraduate Research Program of Rensselaer Polytechnic Institute) for assistance with preparation of nanophase alumina samples.

References [1] Hayashi H, Uchida A, Hamada H, Yoshikawa H, Shinto Y, Ono K. Alumina ceramic prostheses for bone reactions to aluminum oxide dental implants in man: a case report. Arch Orthop Trauma Surg 1992;112:1}4. [2] Toni A, Lewis CG, Sudanese A et al. Bone demineralization induced by cementless alumina-coated femoral stems. J Arthroplasty 1994;9:435}44. [3] Davies JE, Matsuda T. Extracellular matrix production by osteoblasts on bioactive substrate in vitro. Scanning Microscopy 1988;2:1445}52. [4] Sautier JM, Nefussi JR, Forest N. Ultrasonic study of bone formation on synthetic hydroxyapatite in osteoblast cultures. Cell Mater 1991;1:209}17. [5] Sautier JM, Kokubo T, Ohtsuki T et al. Bioactive glass}ceramic containing crystalline apatite and wollastonite initiates biomineralization in bone cell cultures. Calcif Tissue Int 1994; 55:458}66. [6] Matsuda T, Davies JE. The in vitro response of osteoblasts to bioactive glass. Biomaterials 1987;8:275}85.

T.J. Webster et al. / Biomaterials 20 (1999) 1221 } 1227 [7] Shareef MY, Messer PF, Van Noort R. Fabrication characterization and fracture study of a machinable hydroxyapatite ceramic. Biomaterials 1993;14:69}75. [8] Kaplan FS, Hayes WC, Keaven TM, Boskey A, Einhorn TA, Iannotti JP. In: Simon SR, editor. Orthopaedic basic science. Columbus OH: American Academy of Orthopaedic Surgeons, 1994:127. [9] Siegel RW. Creating nanophase materials. Scienti"c American 1996;275:42}7.

1227

[10] Wu SJ, De Jonghe LC, Rahaman MN. Sintering of nanophase c-Al O powder. J Am Ceram Soc 1996;79:2207}11.   [11] Puleo DA, Holleran LA, Doremus RH, Bizios R. Osteoblast responses to orthopaedic implant materials in vitro. J Biomed Mater Res 1991;25:711}23. [12] Keller JC, Stanford CM, Wightman JP, Draughn RA, Zaharias R. Characterizations of titanium implant surfaces. III. J Biomed Mater Res 1994;28:939}46.