- Email: [email protected]
Bioactivity of pseudowollastonite in human saliva
Journal of Dentistry Journal of Dentistry 27 (1999) 107–113
Bioactivity of pseudowollastonite in human saliva P.N. De Aza1a,*, Z.B. Luklinska b, M.R. Anseau c, F. Guitian a, S. De Aza d a Instituto de Ceramica, Universidad de Santiago de Compostela, Santiago de Compostela, Spain Materials Department/IRC in Biomedical Materials, Queen Mary and Westfield College, University of London, Mile End Road, London E1 4NS, UK c St Bartholomews and the Royal London School of Medicine and Dentistry, Department of Biomaterials in Relation to Dentistry, Medical Science Building, Queen Mary and Westfield College, University of London, Mile End Road, London E1 4NS, UK d Instituto de Ceramica y Vidrio (CSIC), Arganda del Rey, Madrid, Spain b
Accepted 30 April 1998
Abstract Objectives: Pseudowollastonite (CaO·SiO 2) was found to be bioactive in a simulated body fluid environment. In the present study, ‘in vitro’ bioactivity of pseudowollastonite was further assessed in human parotid saliva. The main objective was to compare behaviour of the material in a natural medium of high protein content (human parotid saliva) with its behaviour in an acellular protein-free solution (simulated body fluid). Methods: Samples of polycrystalline pseudowollastonite were immersed for one month in human parotid saliva at 37⬚C. Changes in ionic concentrations in the human parotid saliva and the pH right at the interface of pseudowollastonite/human parotid saliva were determined. The products of the interfacial reactions were studied by thin-film X-ray diffraction, scanning and transmission electron microscopy. Results: The results confirmed formation of a hydroxyapatite-like layer on the surface of the material, and also suggested that the mechanism of hydroxyapatite-like layer formation in saliva was similar to that showed in simulated body fluid. Conclusions: The hydroxyapatite-like layer formed at the interface was found to be compact, continuous and composed of many small crystallites with ultrastructure similar to that of natural cortical bone and dentine. The study also concluded that the high pH conditions (10.32) existing right at the pseudowollastonite/human parotid saliva interface promoted hydroxyapatite-like precipitation. At this stage of the study, similarities of the material behaviour in saliva and acellular simulated body fluid suggest that the pseudowollastonite could be of interest in specific periodontal applications for bone restorative purposes. 䉷 1999 Elsevier Science Ltd. All rights reserved. Keywords: Wollastonite; Bioactivity; Human parotid saliva; In vitro
1. Introduction Since the discovery of bioglasses by Hench et al. [1] in the early 1970s, various types of glasses and glass-ceramics have been used as bone replacement [2–4]. Some of these materials have been tested clinically as middle-ear implants [5–7] for the reconstruction of the iliac crest [8], and as vertebral prostheses [9,10] or dental implants [11]. Recently, the chain silicate minerals wollastonite (CaSiO 3) [12–14] and diopside (CaMgSi 2O 6) [15,16] * Corresponding author. Tel.: +34-981-563100 ext. 13544; fax: +34-981564242; e-mail: [email protected] 1 On leave at: St. Bartholomews and the Royal London School of Medicine and Dentistry, Department of Biomaterials in Relation to Dentistry, Medical Science Building, Queen Mary and Westfield College, University of London, Mile End Road, London E1 4NS, UK. This work was presented at the 3rd International Workshop on Interfaces held at Santiago de Compostela (Spain), September 16–19, 1996.
have been prepared synthetically for use as bioactive ceramic materials. Calcium phosphate ceramics, bioglasses and glassceramics have received much attention as potential bone implant materials, because they can form a bond with bone tissue [17–19]. It is well known that chemical reactions occurring on the ceramic surfaces play an important role in the bonding mechanism [20–22]. The ceramic surface initially reacts with the surrounding physiological fluid; therefore, the nature of the solids formed on the surface is determined by the chemistry of the ceramics and the constituents of the body fluid. De Aza et al. [12–14] showed that a hydroxyapatite-like (HA) layer formed on the surface of pseudowollastonite ceramic pellets exposed to acellular simulated body fluid [23]. The reaction starts with an ionic exchange of one Ca from the pseudowollastonite network for 2H 3O þ from the simulated body fluid, which progressively transforms the
0300-5712/99/$ - see front matter 䉷 1999 Elsevier Science Ltd. All rights reserved. PII: S0 30 0 -5 7 12 ( 98 ) 00 0 29 - 3
108
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113
compare its behaviour with the one in simulated body fluid. In view of the results some dental applications for pseudowollastonite in periodontics and restorative dentistry could be envisaged. In the study, changes in ionic concentration of the human parotid saliva and pH right at the pseudowollastonite/human parotid saliva (psW/HS) interface were measured. It was important to record these parameters as they are indicative of chemical changes taking place between the two media. The products of the interfacial reactions were characterised by thin-film X-ray diffraction, scanning and transmission electron microscopy (SEM and TEM).
2. Materials and methods Fig. 1. SEM image of the pseudowollastonite unexposed pellet.
pseudowollastonite crystals into an amorphous silica phase. As the reaction proceeds, the calcium concentration and the pH right at the pseudowollastonite/human saliva interface increase, creating the appropriate conditions for partial dissolution of the amorphous silica and subsequent HA-like phase precipitation. This type of HA-like layer has not been observed for non-bioactive materials, indicating that the HA-like layer plays an essential role in forming the tight chemical bond between the bioactive material and the bone [24–26]. Previous study found that bulk pseudowollastonite was bioactive in simulated body fluid environment [12–14] and therefore could be considered as hard tissue replacement material. The main purpose of the present investigation was to study the reactivity of polycrystalline pseudowollastonite ceramic pellets in human parotid saliva and
The starting material was polycrystalline pseudowollastonite ceramic (a-CaO·SiO 2) powder of 2–3 mm average particle size, synthesized at 1500⬚C for 2 h from an stoichiometric mixture of calcium carbonate (Merk R.G.) and silica (Aldrich R.G.). Pseudowollastonite bars were obtained by cold isostatic pressing at 200 MPa, followed by sintering at 1400⬚C for 2 h with a heating rate of 5⬚C/min. The bars were cut into pellets measuring 5 mm in diameter and 2 mm in thickness. Fifteen pellets of pseudowollastonite in total, prepared from the same batch of sintered bars, were divided into three groups and placed in three polyethylene bottles containing 40 ml of a stimulated human parotid saliva each, at the human body temperature (37⬚C). Several crystals of thymol (98% purity) were added to each bottle in order to avoid the formation of bacteria. The stimulated human parotid saliva was collected from one of the authors (healthy
Fig. 2. Thin film X-ray diffraction results of the pseudowollastonite pellet after one month soaking in human parotid saliva at different depth levels. (A ¼ inside the pseudowollastonite; B ¼ pseudowollastonite/human parotid saliva interface; X ¼ HA-like; o ¼ pseudowollastonite.)
109
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113 Table 1 Changes in pH versus time at the interface of the pseudowollastonite/human saliva Time (min)
0
1
3
5
10
15
60
1 day
1 week
1 month
pH ( ⫾ 0.02)
8.00
9.10
9.73
10.28
10.30
10.32
10.32
10.32
10.32
10.32
and not under medication), using a Lashley cup [27] (purpose built at the Royal London Hospital, Department of Pediatric Dentistry, London, England based on Lashleys original paper and redesigned by Dr M.P. Hector) during five clinical sessions. The sessions were organized at the same time of the day after a very light lunch taken 1 h before the saliva collection. Each session lasted about 3 h, in order to collect the same amount of stimulated saliva. 1 vol.% solution of citric acid was used as the stimulus (1–2 ml/ 5 min). It was checked by chemical analysis that the collected saliva was a typical sample of human parotid saliva composition. The immersion period of the pellets in saliva was 1 month. This period was chosen based on the results from previous ‘in vitro’ experiments performed in the simulated body fluid [12,14]. One batch of pellets was used for pH measurements at certain time intervals. The second batch was used in thin film X-ray diffraction experiments and SEM examination, while the third was used in the TEM study. Changes in ionic concentrations of the human parotid saliva were determined using Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES, VARIAN Liberty 200, Varian Australia Pty Limited). An ionsensitive field-effect transistor (ISFET-Meter) of Si 3N 4 type [28,29] (purpose built at the Centro Nacional de Microelectronica CNN-CSIC, Universidad Autonoma de Barcelona. 08193 Bellaterra, Barcelona, Spain) was used to measure the pH exactly at the pseudowollastonite/ human parotid saliva interface. The structural changes of the pseudowollastonite ceramic were first analysed by thin-film X-ray diffraction, which enables analysis of up to a 1 mm thick layer, depending on the incident X-ray beam angle. The morphology of the cross-sections of the specimens was examined by SEM at
10 keV using a JEOL, JSM 6300 (Jeol Ltd, Tokyo, Japan) fitted with energy-dispersive and wavelength-dispersive X-ray spectrometers (EDS and WDS, Oxford Instruments UK, Ltd, High Wycombe, HP12 3S, England). The crosssections studied in the SEM were previously polished to a 1 mm finish using diamond paste, gently cleaned in an ultrasonic bath and carbon coated. X-ray elemental maps of Ca, Si and P of the cross-sections were also obtained. The product of the reaction was separately examined on a transmission electron microscope, JEOL JEM 2010 at 200 keV. (Jeol Ltd, Tokyo, Japan). These specimens were
Fig. 3. Hydroxyapatite-like layer formed on the surface of pseudowollastonite after 1 month immersion in human parotid saliva.
Fig. 4. X-ray maps of: Si, Ca and P, and SEM image of a cross-section of the pseudowollastonite pellet after soaking in human parotid saliva.
110
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113
Fig. 5. TEM micrograph of the morphology of the hydroxyapatite-like phase.
prepared by careful removal of the reaction layer from the surfaces of the pellets using a razor blade, and dispersing the powder on the surface of methanol in a Petri dish. The powder specimens were then collected on carbon coated TEM copper grids of 200 mesh and carbon coated. Electron beam transparent particles were chosen for TEM examination by selected area diffraction, low and high magnification imaging, and also energy dispersive X-ray spectroscopy (EDS).
3. Results SEM observation of pseudowollastonite control pellets (as sintered before immersion in saliva) showed that they were composed of nearly equiaxial grains, with an average Table 2 Concentration of Ca þþ, Si 4þ, HPO4¼ in human parotid saliva before and after immersion of pseudowollastonite pellets for 1 month mg/l
Ca þþ
Si 4þ
HPO4¼
Saliva control solution Saliva after 1 month of pseudowollastonite immersion
64.13 44.96
35.07 36.84
441.7 421.7
grain size of about 21.3 mm. The microstructure of a pseudowollastonite pellet is shown in Fig. 1. About 5% of pores were present in the structure and these were mainly found at the grain boundaries [13]. The average pore size was measured to be about 7.8 mm. A small amount of glassy phase was also observed, usually located at the triple points. SiO 2/CaO molar ratio of the glassy phase was 0.73, as determined by wavelength dispersive spectrometer technique (WDS) [13]. The pH, recorded exactly at the interface (not in the surrounding solution) of the specimens with the saliva, was found to increase sharply from 8.00 to 10.32 ⫾ 0.02 in the first 15 min of immersion. It then remained unchanged at 10.32 ⫾ 0.02 over 1 month (Table 1). Thin-film X-ray diffraction spectra collected at different depth levels below the surface of the samples exposed to saliva showed that the layer formed on the surface of the pellets is indeed composed of an HA-like phase (Fig. 2). Changes in the elemental ionic concentrations of human parotid saliva after 1 month of immersion are shown in Table 2. The composition of the original human parotid saliva solution is also enclosed for comparison. The chemical analysis of the human parotid saliva used in the study agreed with the composition of the one reported in the literature [30]. It was found that silicon ion concentration in human parotid saliva slightly increased over the exposure time indicating partial dissolution of the pseudowollastonite crystals. On the other hand, phosphorous and calcium ion concentrations decreased more significantly because of the precipitation of an HA-like phase on the surfaces of the pellets. The HA-like phase precipitated on the pseudowollastonite surface after 1 month exposure to saliva showed a characteristic globular morphology. This is clearly presented in Fig. 3. The HA-like globules formed a compact and continuous layer over the surface of the pellet. Fig. 4 shows the overall microstructure of the crosssection of pseudowollastonite pellet after 1 month in saliva. The elemental X-ray maps of silicon, calcium and phosphorus are also included in the figure. The outside layer indicated a well-textured calcium phosphate phase of 5–7 mm average thickness in direct contact with the pseudowollastonite substrate. This calcium phosphate phase was of HA-like type as previously identified by thin-film X-ray diffraction. In some regions between the pseudowollastonite and HA-like phase, a silica phase was detected. TEM was used to examine the ultrastructure of the surface product formed after the exposure of pseudowollastonite sample to saliva for one month. The characteristic feather-like morphology of the HA-like globular product is shown in Fig. 5. Two separate agglomerates seen in the figure are composed of crystallites radiating from the centre outwards. High-resolution lattice imaging was performed on the HA-like crystals formed at the surface of pseudowollastonite. The individual crystals were found
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113
111
Fig. 6. (a) High-resolution TEM image of hydroxyapatite-like crystals formed during exposure to human parotid saliva after one month; (b) EDS analysis of the area; (c) selected-area diffraction pattern of the region.
to grow in close contact forming a continuous phase. (002) lattice plane images with 0.344 nm spacing were well resolved in many areas (Fig. 6) and these appeared to be defect free. EDS analysis of the area confirmed the presence of calcium and phosphorus in the layer [Fig. 6(b)]. When the specimen was appropriately oriented, the selected area diffraction pattern often displayed (111) and (002) arcs corresponding to the 0.388 and 0.344 nm lattice spacings, respectively, indicating the preferential orientation of the HA-like crystals in the layer [Fig. 6(c)]. Sporadically, amorphous silica particles were also found. Fig. 7(a) and (b) shows, respectively, the low magnification image of the silica particles and their amorphous selected area diffraction pattern. EDS analysis of the particles confirmed their composition, as indicated in Fig. 7(c). Selected area diffraction and EDS analyses therefore were essential techniques in complete interpretation of phases present at the interface. Amorphous silica in the reaction product was identified by selected area diffraction pattern analysis and X-ray maps indicated on the silica phase random appearance between HA-like layer and the pseudowollastonite substrate.
4. Discussion Overall results suggest that the mechanism of HA-like phase formation on pseudowollastonite in the human parotid saliva was similar to that of HA-like phase on pseudowollastonite in the simulated body fluid [12–14]. It is understood that the HA-like layer precipitated on the surface of the ceramic material from the human parotid saliva due to the high pH (10.32) conditions at the pseudowollastonite–human parotid saliva interface, resulting from the ionic exchange of Ca þþ from the pseudowollastonite network for 2H 3O þ from the saliva. In the pseudowollastonite–human parotid saliva system, the intermediate silica phase was found to be absent or present sporadically as very small regions along the interface. Usually, a continuous amorphous silica layer forms in silica-based bulk bioactive materials after immersion in simulated body fluid. The fact that the silica phase was discontinuous after immersion in the saliva implied that the diffusion of the protons from human parotid saliva towards the pseudowollastonite could not be as fast as in the pseudowollastonite/simulated body fluid.
112
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113
Fig. 7. (a) TEM image of the amorphous silica; (b) selected-area diffraction pattern of the phase; (c) EDS analysis of the area.
At the interface pH of 10.32, the amorphous silica phase would be expected to dissolve in saliva prior to the precipitation of HA-like phase. This agreed with the finding in this study, as in some regions the amorphous silica phase was found to be trapped inside the HA-like layer as it migrated towards the saliva. Selected area diffraction study performed in transmission electron microscope showed that the crystallography of the newly formed HA-like particles is similar to the mineral constituent of the natural cortical bone [31]. The same selected area analysis performed on thin section of a dentine sample in our TEM produced similar results. These findings agreed with comments reported for dentine in the ‘in vivo’ evaluation of diopside [16].
It was found that the pseudowollastonite material was reactive in human parotid saliva by forming compact HA-like layer at the pseudowollastonite–human parotid saliva interface. At this stage of the study pseudowollastonite could be of interest in some specific applications in dentistry for bone restorative purpose in periodontics.
5. Conclusions This study demonstrated the high reactivity of pseudowollastonite in a human parotid saliva for a period of 1 month. The HA-like layer formed at the interface was found to be compact, continuous and composed of many
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113
small crystallites which ultrastructure showed similarity to that of natural cortical bone and also dentine. It was important to confirm that the high pH conditions (10.32) that exist right at the pseudowollastonite/human parotid saliva interface are essential to promote HA-like precipitation. This finding suggested that the mechanism of HA-like phase formation was similar to that of HA-like phase on pseudowollastonite pellets in the simulated body fluid. No continuous silica rich interlayer was found in the pellets exposed to the saliva. In view of the results pseudowollastonite could be applicable in specific periodontal treatment for bone restorative purpose.
Acknowledgements Part of this work was supported by CICYT under project no. MAT97-1025. One of the authors acknowledges the Ministry of Education and Science of Spain for the postdoctoral fellowship given to her, no. PF95 00816891. Dr. M.P. Hector (Department of Pediatric Dentistry) helped to collect volumes of parotid saliva from one of the authors. The EPSRC support for the University of London Interdisciplinary Research Centre in Biomedical Materials is gratefully acknowledged.
[11]
[12] [13]
[14]
[15]
[16]
[17]
[18]
[19] [20]
References [1] Hench LL, Splinter RJ, Allen WC, Greenle TK Bonding mechanisms at the interface of ceramic prosthetic materials. Journal of Biomedical Material Research Symposium, 1971;2:117–141. [2] Kokubo T, Ito S, Sakka S, Yamamuro T Formation of a high-strength bioactive glass-ceramic in the system MgO-CaO-SiO 2-P 2O 5. Journal of Materials Science, 1986;21:536–540. [3] Ebisawa T, Kokubo T, Ohura K, Yamamuro T Bioactivity of CaO·SiO2¹ based glasses: in vitro evaluation. Journal of Materials Science: Materials in Medicine, 1990;1(4):239–244. [4] Ohura K, Nakamura T, Yamamuro T, Kokubo T, Ebisawa T, Kotoura T Bone-bonding ability of P2 O5¹ free CaO·SiO 2 glasses. Journal of Biomedical Materials Research, 1991;25:357–365. [5] Reck R, Storkel S, Meyer A Bioceramics: material characteristics versus in vivo behaviour. Annals of the New York Academy of Sciences, 1988;523:100–106. [6] Merwin GE, Review of bioactive materials for otologic and maxilofacial applications. In: Yamamuro T, Hench LL, Wilson J, editors. Handbook of bioactive ceramics. Bioactive glasses and glass-ceramics, vol. 1. Boca Raton, FL: CRC Press, 1990:323–328. [7] Beleites E, Neupert G, Augsten G, Vogel W, Schubert H SEM study of cell growth on mechanically produced biovitroceramic and carbon glass in vitro and in vivo. Laryngology, Rhinology and Otology, 1985;64(4):217–220. [8] Yamamuro T, Reconstruction of the iliac crest with bioactive glassceramic prostheses. In: Yamamuro T, Hench LL, Wilson J, editors. Handbook of bioactive ceramics. Bioactive glasses and glassceramics, vol. 1. Boca Raton, FL: CRC Press, 1990:335–342. [9] Zippel H, Hahnel H, Gummel J Surgical treatment possibilities in tumors of the spine. Indications, techniques, experiences. Zentralblatt Chirugie, 1987;112(1):63–75. [10] Yamamuro T, Replacement of the spine with bioactive glass-ceramic
[21]
[22]
[23] [24] [25]
[26]
[27] [28]
[29]
[30] [31]
113
prostheses. In: Yamamuro T, Hench LL, Wilson J, editors. Handbook of bioactive ceramics, vol. 1. Boca Raton, FL: CRC Press, 1990:343–351. Hench LL, Stanley HR, Clark AE, Hall M, Wilson J Dental applications of Bioglass威 implants. In: Bonfield W, Hastings GW, Tanner KE, editors. Bioceramics, vol. 4. Oxford: Butterworth-Heinemann, 1991. De Aza PN, Guitian F, De Aza S Bioactivity of wollastonite ceramics: in vitro evaluation. Scripta Materialia, 1994;31(8):1001–1005. De Aza PN, Guitian F, De Aza S, Polycrystalline wollastonite ceramics. Biomaterials free of P 2O 5. In: Vicenzini P, editor. Advances in science and technology, materials in clinical application, vol 12. Italy: Techna Srl, 1995. De Aza PN, Luklinska ZB, Anseau MR, Guitan F, De Aza S Morphological studies of pseudowollastonite for biomedical application. Journal of Microscopy, 1996;182:24–31. Nonami T In vivo and in vitro testing of diopside for biomaterials. Journal of the Society of Materials Engineering for Resources of Japan, 1995;8(2):12–18. Miake Y, Yanagisawa T, Yajima Y, Noma H, Yasui N, Nonami T High-resolution and analytical electron microscopic studies of new crystals induced by a bioactive ceramic (diopside). Journal of Dental Research, 1995;74(11):1756–1763. Muller-Maic C, Schmitz HJ, Strunz V, Furhmann G, Fritz TH, Gross UM Tissues at the surface of a new composite material titanium/glassceramic for replacement of bone and teeth. Journal Biomedical Materials Research, 1989;23(10):1149–1168. Kokubo T, Hayashi T, Sakka S, Kitsugi T, Yamamuro T Bonding between bioactive glasses, glass-ceramics or ceramics in simulated body fluid. Yogyo-Kyokai-hi, 1987;95(8):785–791. Hench LL Bioceramics: from concept to clinic. Journal of the American Ceramic Society, 1991;74(1):1487–1510. Van Blitterswijk CA, Grote JJ, Kuypers W, Blockvanhochk CJG, Daems WT Bioreactions at the tissue/hydroxyapatite interface. Biomaterials, 1985;6:243–251. Yoshii S, Kakutani T, Yamamuro T, Nakamura T, Kitsugi T, Oka M Strength of bonding between A-W glass-ceramic and the surface of bone cortex. Journal of Biomedical Materials Research, 1988;22(A):327–338. Ohtsuki C, Kokubo T, Yamamuro T Mechanism of apatite formation on CaO-SiO2¹ -P 2O 5 glasses in a simulated body fluid. Journal of Noncrystallline Solids, 1992;143:84–92. Gamble J, Chemical anatomy physiology and pathology of extracellular fluid. 6th ed. Cambridge: Harvard University Press, Cambridge, 1967. Hench LL, Wilson JW Surface-active biomaterials. Science, 1984;226:630–636. Hench LL, Bioactive glasses and glass-ceramics: a perspective. In: Yamamuro T, Hench LL, Wilson J, editors. Handbook of bioactive ceramics, vol. 1. Boca Raton, FL: CRC Press, 1990:7–23. Ohtsuki C, Aoki Y, Kokubo T, Bando Y, Neo M, Yamamuro T, Nakamura T Characterization of apatite layer formed on bioactive glass-ceramic A-W. Bioceramics, 1992;5:87–94. Lashley KS Reflex secretion of the human parotid gland. Journal of Experimental Psychology, 1916;1:461–493. Merlos A, Gracia I, Cane´ C, Esteve J, Bartroli J, Jiminez C CMOS flow-through pH-ISFET. Proc 5th Conference on Sensors and their Applications. Edinburgh, 1991. De Aza PN, Guitian F, Merlos A, Lora-Tamayo E, De Aza S Bioceramics-simulated body fluid interfaces: pH and its influence on hydroxyapatite formation. Journal of Materials Science: Materials in Medicine, 1996;7(7):399–402. Larelle CLB, Applied oral physiology. 2nd ed. London: Wright, 1988:158-172. Luklinska ZB, Bonfield W, Ultrastucture of bone-composite interface by high resolution transmission electron Microscopy. In: Ducheyne P, Kokubo T, Van Blitterswijk CA, editors. Bone-bonding materials. Reed Healthcare Communications, Reaal 2f, Leiderdorp, The Netherlands. 1992:73–78.
Bioactivity of pseudowollastonite in human saliva P.N. De Aza1a,*, Z.B. Luklinska b, M.R. Anseau c, F. Guitian a, S. De Aza d a Instituto de Ceramica, Universidad de Santiago de Compostela, Santiago de Compostela, Spain Materials Department/IRC in Biomedical Materials, Queen Mary and Westfield College, University of London, Mile End Road, London E1 4NS, UK c St Bartholomews and the Royal London School of Medicine and Dentistry, Department of Biomaterials in Relation to Dentistry, Medical Science Building, Queen Mary and Westfield College, University of London, Mile End Road, London E1 4NS, UK d Instituto de Ceramica y Vidrio (CSIC), Arganda del Rey, Madrid, Spain b
Accepted 30 April 1998
Abstract Objectives: Pseudowollastonite (CaO·SiO 2) was found to be bioactive in a simulated body fluid environment. In the present study, ‘in vitro’ bioactivity of pseudowollastonite was further assessed in human parotid saliva. The main objective was to compare behaviour of the material in a natural medium of high protein content (human parotid saliva) with its behaviour in an acellular protein-free solution (simulated body fluid). Methods: Samples of polycrystalline pseudowollastonite were immersed for one month in human parotid saliva at 37⬚C. Changes in ionic concentrations in the human parotid saliva and the pH right at the interface of pseudowollastonite/human parotid saliva were determined. The products of the interfacial reactions were studied by thin-film X-ray diffraction, scanning and transmission electron microscopy. Results: The results confirmed formation of a hydroxyapatite-like layer on the surface of the material, and also suggested that the mechanism of hydroxyapatite-like layer formation in saliva was similar to that showed in simulated body fluid. Conclusions: The hydroxyapatite-like layer formed at the interface was found to be compact, continuous and composed of many small crystallites with ultrastructure similar to that of natural cortical bone and dentine. The study also concluded that the high pH conditions (10.32) existing right at the pseudowollastonite/human parotid saliva interface promoted hydroxyapatite-like precipitation. At this stage of the study, similarities of the material behaviour in saliva and acellular simulated body fluid suggest that the pseudowollastonite could be of interest in specific periodontal applications for bone restorative purposes. 䉷 1999 Elsevier Science Ltd. All rights reserved. Keywords: Wollastonite; Bioactivity; Human parotid saliva; In vitro
1. Introduction Since the discovery of bioglasses by Hench et al. [1] in the early 1970s, various types of glasses and glass-ceramics have been used as bone replacement [2–4]. Some of these materials have been tested clinically as middle-ear implants [5–7] for the reconstruction of the iliac crest [8], and as vertebral prostheses [9,10] or dental implants [11]. Recently, the chain silicate minerals wollastonite (CaSiO 3) [12–14] and diopside (CaMgSi 2O 6) [15,16] * Corresponding author. Tel.: +34-981-563100 ext. 13544; fax: +34-981564242; e-mail: [email protected] 1 On leave at: St. Bartholomews and the Royal London School of Medicine and Dentistry, Department of Biomaterials in Relation to Dentistry, Medical Science Building, Queen Mary and Westfield College, University of London, Mile End Road, London E1 4NS, UK. This work was presented at the 3rd International Workshop on Interfaces held at Santiago de Compostela (Spain), September 16–19, 1996.
have been prepared synthetically for use as bioactive ceramic materials. Calcium phosphate ceramics, bioglasses and glassceramics have received much attention as potential bone implant materials, because they can form a bond with bone tissue [17–19]. It is well known that chemical reactions occurring on the ceramic surfaces play an important role in the bonding mechanism [20–22]. The ceramic surface initially reacts with the surrounding physiological fluid; therefore, the nature of the solids formed on the surface is determined by the chemistry of the ceramics and the constituents of the body fluid. De Aza et al. [12–14] showed that a hydroxyapatite-like (HA) layer formed on the surface of pseudowollastonite ceramic pellets exposed to acellular simulated body fluid [23]. The reaction starts with an ionic exchange of one Ca from the pseudowollastonite network for 2H 3O þ from the simulated body fluid, which progressively transforms the
0300-5712/99/$ - see front matter 䉷 1999 Elsevier Science Ltd. All rights reserved. PII: S0 30 0 -5 7 12 ( 98 ) 00 0 29 - 3
108
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113
compare its behaviour with the one in simulated body fluid. In view of the results some dental applications for pseudowollastonite in periodontics and restorative dentistry could be envisaged. In the study, changes in ionic concentration of the human parotid saliva and pH right at the pseudowollastonite/human parotid saliva (psW/HS) interface were measured. It was important to record these parameters as they are indicative of chemical changes taking place between the two media. The products of the interfacial reactions were characterised by thin-film X-ray diffraction, scanning and transmission electron microscopy (SEM and TEM).
2. Materials and methods Fig. 1. SEM image of the pseudowollastonite unexposed pellet.
pseudowollastonite crystals into an amorphous silica phase. As the reaction proceeds, the calcium concentration and the pH right at the pseudowollastonite/human saliva interface increase, creating the appropriate conditions for partial dissolution of the amorphous silica and subsequent HA-like phase precipitation. This type of HA-like layer has not been observed for non-bioactive materials, indicating that the HA-like layer plays an essential role in forming the tight chemical bond between the bioactive material and the bone [24–26]. Previous study found that bulk pseudowollastonite was bioactive in simulated body fluid environment [12–14] and therefore could be considered as hard tissue replacement material. The main purpose of the present investigation was to study the reactivity of polycrystalline pseudowollastonite ceramic pellets in human parotid saliva and
The starting material was polycrystalline pseudowollastonite ceramic (a-CaO·SiO 2) powder of 2–3 mm average particle size, synthesized at 1500⬚C for 2 h from an stoichiometric mixture of calcium carbonate (Merk R.G.) and silica (Aldrich R.G.). Pseudowollastonite bars were obtained by cold isostatic pressing at 200 MPa, followed by sintering at 1400⬚C for 2 h with a heating rate of 5⬚C/min. The bars were cut into pellets measuring 5 mm in diameter and 2 mm in thickness. Fifteen pellets of pseudowollastonite in total, prepared from the same batch of sintered bars, were divided into three groups and placed in three polyethylene bottles containing 40 ml of a stimulated human parotid saliva each, at the human body temperature (37⬚C). Several crystals of thymol (98% purity) were added to each bottle in order to avoid the formation of bacteria. The stimulated human parotid saliva was collected from one of the authors (healthy
Fig. 2. Thin film X-ray diffraction results of the pseudowollastonite pellet after one month soaking in human parotid saliva at different depth levels. (A ¼ inside the pseudowollastonite; B ¼ pseudowollastonite/human parotid saliva interface; X ¼ HA-like; o ¼ pseudowollastonite.)
109
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113 Table 1 Changes in pH versus time at the interface of the pseudowollastonite/human saliva Time (min)
0
1
3
5
10
15
60
1 day
1 week
1 month
pH ( ⫾ 0.02)
8.00
9.10
9.73
10.28
10.30
10.32
10.32
10.32
10.32
10.32
and not under medication), using a Lashley cup [27] (purpose built at the Royal London Hospital, Department of Pediatric Dentistry, London, England based on Lashleys original paper and redesigned by Dr M.P. Hector) during five clinical sessions. The sessions were organized at the same time of the day after a very light lunch taken 1 h before the saliva collection. Each session lasted about 3 h, in order to collect the same amount of stimulated saliva. 1 vol.% solution of citric acid was used as the stimulus (1–2 ml/ 5 min). It was checked by chemical analysis that the collected saliva was a typical sample of human parotid saliva composition. The immersion period of the pellets in saliva was 1 month. This period was chosen based on the results from previous ‘in vitro’ experiments performed in the simulated body fluid [12,14]. One batch of pellets was used for pH measurements at certain time intervals. The second batch was used in thin film X-ray diffraction experiments and SEM examination, while the third was used in the TEM study. Changes in ionic concentrations of the human parotid saliva were determined using Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES, VARIAN Liberty 200, Varian Australia Pty Limited). An ionsensitive field-effect transistor (ISFET-Meter) of Si 3N 4 type [28,29] (purpose built at the Centro Nacional de Microelectronica CNN-CSIC, Universidad Autonoma de Barcelona. 08193 Bellaterra, Barcelona, Spain) was used to measure the pH exactly at the pseudowollastonite/ human parotid saliva interface. The structural changes of the pseudowollastonite ceramic were first analysed by thin-film X-ray diffraction, which enables analysis of up to a 1 mm thick layer, depending on the incident X-ray beam angle. The morphology of the cross-sections of the specimens was examined by SEM at
10 keV using a JEOL, JSM 6300 (Jeol Ltd, Tokyo, Japan) fitted with energy-dispersive and wavelength-dispersive X-ray spectrometers (EDS and WDS, Oxford Instruments UK, Ltd, High Wycombe, HP12 3S, England). The crosssections studied in the SEM were previously polished to a 1 mm finish using diamond paste, gently cleaned in an ultrasonic bath and carbon coated. X-ray elemental maps of Ca, Si and P of the cross-sections were also obtained. The product of the reaction was separately examined on a transmission electron microscope, JEOL JEM 2010 at 200 keV. (Jeol Ltd, Tokyo, Japan). These specimens were
Fig. 3. Hydroxyapatite-like layer formed on the surface of pseudowollastonite after 1 month immersion in human parotid saliva.
Fig. 4. X-ray maps of: Si, Ca and P, and SEM image of a cross-section of the pseudowollastonite pellet after soaking in human parotid saliva.
110
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113
Fig. 5. TEM micrograph of the morphology of the hydroxyapatite-like phase.
prepared by careful removal of the reaction layer from the surfaces of the pellets using a razor blade, and dispersing the powder on the surface of methanol in a Petri dish. The powder specimens were then collected on carbon coated TEM copper grids of 200 mesh and carbon coated. Electron beam transparent particles were chosen for TEM examination by selected area diffraction, low and high magnification imaging, and also energy dispersive X-ray spectroscopy (EDS).
3. Results SEM observation of pseudowollastonite control pellets (as sintered before immersion in saliva) showed that they were composed of nearly equiaxial grains, with an average Table 2 Concentration of Ca þþ, Si 4þ, HPO4¼ in human parotid saliva before and after immersion of pseudowollastonite pellets for 1 month mg/l
Ca þþ
Si 4þ
HPO4¼
Saliva control solution Saliva after 1 month of pseudowollastonite immersion
64.13 44.96
35.07 36.84
441.7 421.7
grain size of about 21.3 mm. The microstructure of a pseudowollastonite pellet is shown in Fig. 1. About 5% of pores were present in the structure and these were mainly found at the grain boundaries [13]. The average pore size was measured to be about 7.8 mm. A small amount of glassy phase was also observed, usually located at the triple points. SiO 2/CaO molar ratio of the glassy phase was 0.73, as determined by wavelength dispersive spectrometer technique (WDS) [13]. The pH, recorded exactly at the interface (not in the surrounding solution) of the specimens with the saliva, was found to increase sharply from 8.00 to 10.32 ⫾ 0.02 in the first 15 min of immersion. It then remained unchanged at 10.32 ⫾ 0.02 over 1 month (Table 1). Thin-film X-ray diffraction spectra collected at different depth levels below the surface of the samples exposed to saliva showed that the layer formed on the surface of the pellets is indeed composed of an HA-like phase (Fig. 2). Changes in the elemental ionic concentrations of human parotid saliva after 1 month of immersion are shown in Table 2. The composition of the original human parotid saliva solution is also enclosed for comparison. The chemical analysis of the human parotid saliva used in the study agreed with the composition of the one reported in the literature [30]. It was found that silicon ion concentration in human parotid saliva slightly increased over the exposure time indicating partial dissolution of the pseudowollastonite crystals. On the other hand, phosphorous and calcium ion concentrations decreased more significantly because of the precipitation of an HA-like phase on the surfaces of the pellets. The HA-like phase precipitated on the pseudowollastonite surface after 1 month exposure to saliva showed a characteristic globular morphology. This is clearly presented in Fig. 3. The HA-like globules formed a compact and continuous layer over the surface of the pellet. Fig. 4 shows the overall microstructure of the crosssection of pseudowollastonite pellet after 1 month in saliva. The elemental X-ray maps of silicon, calcium and phosphorus are also included in the figure. The outside layer indicated a well-textured calcium phosphate phase of 5–7 mm average thickness in direct contact with the pseudowollastonite substrate. This calcium phosphate phase was of HA-like type as previously identified by thin-film X-ray diffraction. In some regions between the pseudowollastonite and HA-like phase, a silica phase was detected. TEM was used to examine the ultrastructure of the surface product formed after the exposure of pseudowollastonite sample to saliva for one month. The characteristic feather-like morphology of the HA-like globular product is shown in Fig. 5. Two separate agglomerates seen in the figure are composed of crystallites radiating from the centre outwards. High-resolution lattice imaging was performed on the HA-like crystals formed at the surface of pseudowollastonite. The individual crystals were found
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113
111
Fig. 6. (a) High-resolution TEM image of hydroxyapatite-like crystals formed during exposure to human parotid saliva after one month; (b) EDS analysis of the area; (c) selected-area diffraction pattern of the region.
to grow in close contact forming a continuous phase. (002) lattice plane images with 0.344 nm spacing were well resolved in many areas (Fig. 6) and these appeared to be defect free. EDS analysis of the area confirmed the presence of calcium and phosphorus in the layer [Fig. 6(b)]. When the specimen was appropriately oriented, the selected area diffraction pattern often displayed (111) and (002) arcs corresponding to the 0.388 and 0.344 nm lattice spacings, respectively, indicating the preferential orientation of the HA-like crystals in the layer [Fig. 6(c)]. Sporadically, amorphous silica particles were also found. Fig. 7(a) and (b) shows, respectively, the low magnification image of the silica particles and their amorphous selected area diffraction pattern. EDS analysis of the particles confirmed their composition, as indicated in Fig. 7(c). Selected area diffraction and EDS analyses therefore were essential techniques in complete interpretation of phases present at the interface. Amorphous silica in the reaction product was identified by selected area diffraction pattern analysis and X-ray maps indicated on the silica phase random appearance between HA-like layer and the pseudowollastonite substrate.
4. Discussion Overall results suggest that the mechanism of HA-like phase formation on pseudowollastonite in the human parotid saliva was similar to that of HA-like phase on pseudowollastonite in the simulated body fluid [12–14]. It is understood that the HA-like layer precipitated on the surface of the ceramic material from the human parotid saliva due to the high pH (10.32) conditions at the pseudowollastonite–human parotid saliva interface, resulting from the ionic exchange of Ca þþ from the pseudowollastonite network for 2H 3O þ from the saliva. In the pseudowollastonite–human parotid saliva system, the intermediate silica phase was found to be absent or present sporadically as very small regions along the interface. Usually, a continuous amorphous silica layer forms in silica-based bulk bioactive materials after immersion in simulated body fluid. The fact that the silica phase was discontinuous after immersion in the saliva implied that the diffusion of the protons from human parotid saliva towards the pseudowollastonite could not be as fast as in the pseudowollastonite/simulated body fluid.
112
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113
Fig. 7. (a) TEM image of the amorphous silica; (b) selected-area diffraction pattern of the phase; (c) EDS analysis of the area.
At the interface pH of 10.32, the amorphous silica phase would be expected to dissolve in saliva prior to the precipitation of HA-like phase. This agreed with the finding in this study, as in some regions the amorphous silica phase was found to be trapped inside the HA-like layer as it migrated towards the saliva. Selected area diffraction study performed in transmission electron microscope showed that the crystallography of the newly formed HA-like particles is similar to the mineral constituent of the natural cortical bone [31]. The same selected area analysis performed on thin section of a dentine sample in our TEM produced similar results. These findings agreed with comments reported for dentine in the ‘in vivo’ evaluation of diopside [16].
It was found that the pseudowollastonite material was reactive in human parotid saliva by forming compact HA-like layer at the pseudowollastonite–human parotid saliva interface. At this stage of the study pseudowollastonite could be of interest in some specific applications in dentistry for bone restorative purpose in periodontics.
5. Conclusions This study demonstrated the high reactivity of pseudowollastonite in a human parotid saliva for a period of 1 month. The HA-like layer formed at the interface was found to be compact, continuous and composed of many
P.N. De Aza et al. / Journal of Dentistry 27 (1999) 107–113
small crystallites which ultrastructure showed similarity to that of natural cortical bone and also dentine. It was important to confirm that the high pH conditions (10.32) that exist right at the pseudowollastonite/human parotid saliva interface are essential to promote HA-like precipitation. This finding suggested that the mechanism of HA-like phase formation was similar to that of HA-like phase on pseudowollastonite pellets in the simulated body fluid. No continuous silica rich interlayer was found in the pellets exposed to the saliva. In view of the results pseudowollastonite could be applicable in specific periodontal treatment for bone restorative purpose.
Acknowledgements Part of this work was supported by CICYT under project no. MAT97-1025. One of the authors acknowledges the Ministry of Education and Science of Spain for the postdoctoral fellowship given to her, no. PF95 00816891. Dr. M.P. Hector (Department of Pediatric Dentistry) helped to collect volumes of parotid saliva from one of the authors. The EPSRC support for the University of London Interdisciplinary Research Centre in Biomedical Materials is gratefully acknowledged.
[11]
[12] [13]
[14]
[15]
[16]
[17]
[18]
[19] [20]
References [1] Hench LL, Splinter RJ, Allen WC, Greenle TK Bonding mechanisms at the interface of ceramic prosthetic materials. Journal of Biomedical Material Research Symposium, 1971;2:117–141. [2] Kokubo T, Ito S, Sakka S, Yamamuro T Formation of a high-strength bioactive glass-ceramic in the system MgO-CaO-SiO 2-P 2O 5. Journal of Materials Science, 1986;21:536–540. [3] Ebisawa T, Kokubo T, Ohura K, Yamamuro T Bioactivity of CaO·SiO2¹ based glasses: in vitro evaluation. Journal of Materials Science: Materials in Medicine, 1990;1(4):239–244. [4] Ohura K, Nakamura T, Yamamuro T, Kokubo T, Ebisawa T, Kotoura T Bone-bonding ability of P2 O5¹ free CaO·SiO 2 glasses. Journal of Biomedical Materials Research, 1991;25:357–365. [5] Reck R, Storkel S, Meyer A Bioceramics: material characteristics versus in vivo behaviour. Annals of the New York Academy of Sciences, 1988;523:100–106. [6] Merwin GE, Review of bioactive materials for otologic and maxilofacial applications. In: Yamamuro T, Hench LL, Wilson J, editors. Handbook of bioactive ceramics. Bioactive glasses and glass-ceramics, vol. 1. Boca Raton, FL: CRC Press, 1990:323–328. [7] Beleites E, Neupert G, Augsten G, Vogel W, Schubert H SEM study of cell growth on mechanically produced biovitroceramic and carbon glass in vitro and in vivo. Laryngology, Rhinology and Otology, 1985;64(4):217–220. [8] Yamamuro T, Reconstruction of the iliac crest with bioactive glassceramic prostheses. In: Yamamuro T, Hench LL, Wilson J, editors. Handbook of bioactive ceramics. Bioactive glasses and glassceramics, vol. 1. Boca Raton, FL: CRC Press, 1990:335–342. [9] Zippel H, Hahnel H, Gummel J Surgical treatment possibilities in tumors of the spine. Indications, techniques, experiences. Zentralblatt Chirugie, 1987;112(1):63–75. [10] Yamamuro T, Replacement of the spine with bioactive glass-ceramic
[21]
[22]
[23] [24] [25]
[26]
[27] [28]
[29]
[30] [31]
113
prostheses. In: Yamamuro T, Hench LL, Wilson J, editors. Handbook of bioactive ceramics, vol. 1. Boca Raton, FL: CRC Press, 1990:343–351. Hench LL, Stanley HR, Clark AE, Hall M, Wilson J Dental applications of Bioglass威 implants. In: Bonfield W, Hastings GW, Tanner KE, editors. Bioceramics, vol. 4. Oxford: Butterworth-Heinemann, 1991. De Aza PN, Guitian F, De Aza S Bioactivity of wollastonite ceramics: in vitro evaluation. Scripta Materialia, 1994;31(8):1001–1005. De Aza PN, Guitian F, De Aza S, Polycrystalline wollastonite ceramics. Biomaterials free of P 2O 5. In: Vicenzini P, editor. Advances in science and technology, materials in clinical application, vol 12. Italy: Techna Srl, 1995. De Aza PN, Luklinska ZB, Anseau MR, Guitan F, De Aza S Morphological studies of pseudowollastonite for biomedical application. Journal of Microscopy, 1996;182:24–31. Nonami T In vivo and in vitro testing of diopside for biomaterials. Journal of the Society of Materials Engineering for Resources of Japan, 1995;8(2):12–18. Miake Y, Yanagisawa T, Yajima Y, Noma H, Yasui N, Nonami T High-resolution and analytical electron microscopic studies of new crystals induced by a bioactive ceramic (diopside). Journal of Dental Research, 1995;74(11):1756–1763. Muller-Maic C, Schmitz HJ, Strunz V, Furhmann G, Fritz TH, Gross UM Tissues at the surface of a new composite material titanium/glassceramic for replacement of bone and teeth. Journal Biomedical Materials Research, 1989;23(10):1149–1168. Kokubo T, Hayashi T, Sakka S, Kitsugi T, Yamamuro T Bonding between bioactive glasses, glass-ceramics or ceramics in simulated body fluid. Yogyo-Kyokai-hi, 1987;95(8):785–791. Hench LL Bioceramics: from concept to clinic. Journal of the American Ceramic Society, 1991;74(1):1487–1510. Van Blitterswijk CA, Grote JJ, Kuypers W, Blockvanhochk CJG, Daems WT Bioreactions at the tissue/hydroxyapatite interface. Biomaterials, 1985;6:243–251. Yoshii S, Kakutani T, Yamamuro T, Nakamura T, Kitsugi T, Oka M Strength of bonding between A-W glass-ceramic and the surface of bone cortex. Journal of Biomedical Materials Research, 1988;22(A):327–338. Ohtsuki C, Kokubo T, Yamamuro T Mechanism of apatite formation on CaO-SiO2¹ -P 2O 5 glasses in a simulated body fluid. Journal of Noncrystallline Solids, 1992;143:84–92. Gamble J, Chemical anatomy physiology and pathology of extracellular fluid. 6th ed. Cambridge: Harvard University Press, Cambridge, 1967. Hench LL, Wilson JW Surface-active biomaterials. Science, 1984;226:630–636. Hench LL, Bioactive glasses and glass-ceramics: a perspective. In: Yamamuro T, Hench LL, Wilson J, editors. Handbook of bioactive ceramics, vol. 1. Boca Raton, FL: CRC Press, 1990:7–23. Ohtsuki C, Aoki Y, Kokubo T, Bando Y, Neo M, Yamamuro T, Nakamura T Characterization of apatite layer formed on bioactive glass-ceramic A-W. Bioceramics, 1992;5:87–94. Lashley KS Reflex secretion of the human parotid gland. Journal of Experimental Psychology, 1916;1:461–493. Merlos A, Gracia I, Cane´ C, Esteve J, Bartroli J, Jiminez C CMOS flow-through pH-ISFET. Proc 5th Conference on Sensors and their Applications. Edinburgh, 1991. De Aza PN, Guitian F, Merlos A, Lora-Tamayo E, De Aza S Bioceramics-simulated body fluid interfaces: pH and its influence on hydroxyapatite formation. Journal of Materials Science: Materials in Medicine, 1996;7(7):399–402. Larelle CLB, Applied oral physiology. 2nd ed. London: Wright, 1988:158-172. Luklinska ZB, Bonfield W, Ultrastucture of bone-composite interface by high resolution transmission electron Microscopy. In: Ducheyne P, Kokubo T, Van Blitterswijk CA, editors. Bone-bonding materials. Reed Healthcare Communications, Reaal 2f, Leiderdorp, The Netherlands. 1992:73–78.