Current Medicinal Chemistry, 2008, 15, 305-313 305

Bone Regeneration by Synthetic Octacalcium Phosphate and its Role in Biological

Mineralization
O. Suzuki*,1, H. Imaizumi2, S. Kamakura3 and T. Katagiri4
1
Division of Craniofacial Function Engineering, Tohoku University Graduate School of Dentistry, Sendai, Japan
2
Department of Orthopedic Surgery, Osaki Citizen Hospital, Osaki, Japan
3
Division of Clinical Cell Therapy, Department of Translational Research, Center for Translational and Advanced Animal Research,
Tohoku University School of Medicine, Sendai, Japan
4
Division of Pathophysiology, Research Center for Genomic Medicine, Saitama Medical School, Hidaka, Japan
Abstract: Octacalcium phosphate (Ca8H2(PO4)6･5H2O; OCP) has been advocated to be a precursor of biological apatite crystals in bone
and tooth. Recent studies, using physical techniques, showed that OCP is present as a transient phase during biological apatite formation
in human dentin, porcine enamel and murine bone. However, there is still a controversy regarding the chemical nature of the first mineral
formed in the biominerals. A number of studies have demonstrated that synthetic OCP shows bone regenerative and biodegradable char-
acteristics, rather than other calcium phosphate bone substitute materials, such as hydroxyapatite (Ca10(PO 4)6(OH) 2; HA) ceramic. It
seems likely that synthetic OCP may be an alternative to autogenous bone graft. It is known that OCP contains alternative layers of water
molecules and an apatite structure, and that the transition of OCP to HA is likely to be spontaneous and irreversible. The conversion
process induces modification of local environment adjacent to OCP surface, including the changes in adsorption of serum proteins and
concentration of calcium and inorganic phosphate ions. This article reviews the possible application to bone regeneration by synthetic
OCP and the mechanism to enhance bone regeneration in relation to biological mineralization in bone and tooth.
Keywords: Octacalcium phosphate (OCP), hydroxyapatite (HA), conversion, biomineralization, biomaterials, bone regeneration.

1. INTRODUCTION phosphate (TCP) can be transformed to the most stable phase HA

under physiological condition [17]. Synthetic TCP, known as a
The prototype for the mineral in mature bone and tooth is usu- biodegradable bioceramic, is clinically used in a  form as an artifi-
ally considered to be the basic calcium phosphate hydroxyapatite cial material for bone defects [16,18]. Tetracalcium phosphate
(Ca10(PO4)6(OH)2; HA) [1-3]. Amorphous calcium phosphate (Ca3 (TTCP) is another artificial material, which is used as a component
(PO4)2 ･nH2O) and/or octacalcium phosphate (Ca8H2(PO4)6 ･5H2O; in calcium phosphate cement with
-TCP to form HA [19]. HA,
OCP) have been suggested as precursor phases to biological apatite having a stoichiometric Ca/P molar ratio 1.67, usually obtained by
crystals [4-6]. In fact, several studies have reported that OCP is synthesis, is widely recognized biocompatible and non-resorbable
present as an intermediate to HA in human dentine [7], porcine bioceramic for bone substitute [15,16]. The Ca/P molar ratio is one
enamel [8] and in sutures undergoing premature suture closure dur- of the indices used to identify specific calcium phosphate com-
ing intramembranous mineralization of mouse calvaria [9]. On the pounds. It is more likely that OCP precursors, if they exist, may
contrary, the direct precipitation of very small poorly crystalline exhibit a variety of compositions and structural environments.
HA has been reported to be present as the nucleating phase for ver- Mathew et al. reported the formula of OCP, Ca16H4+X
tebrate bone apatite mineral without the involvement of these pre- (PO4)12(OH)X･(10-X)H2O [20]. The authors stated that the maxi-
cursor phases [10-12]. Thus the mechanism of the initiation of mum acid phosphate assessed by pyrolysis corresponds to about
biomineralization remains controversial [13,14]. However, apart 40% of total phosphorus. The non-stoichiometric composition of
from the controversy as to the chemical nature of the first mineral OCP can be obtained in a hot solution synthesis condition [21,22].
phase in biomineralization of vertebrate, it is of interest to examine The Ca/P molar ratio of ACP has also been confirmed to vary de-
whether synthetic OCP is capable of initiating bone regeneration in pending on the synthesis conditions from 1.15 to 1.5 [23,24]. The
bone defects, thereby OCP can be used as a bone substitute material Ca/P molar ratios of Ca-deficient HA are in those under 1.67 for
in bone regeneration. It seems likely that bone regeneration by syn- stoichiometric composition of HA, but more than 1.5 [24]. It has
thetic OCP may simulate biomineralization process in bone and been well-documented that mature tooth and bone mineral are com-
tooth formation at least in the maturation process from OCP to HA posed of non-stoichiometric apatite crystals containing various
within their tissue formations, if OCP is involved in biomineraliza- impurities, such as carbonate and acid phosphate [1-3]. The exis-
tion. Studies of the mechanism of bone regeneration by synthetic tence of another type of OCP, containing carbonate has been re-
OCP may be relevant to the elucidation of the mechanism of ported [25]. This OCP has been referred to as OCP carbonate and
biomineralization during bone formation. We review the experi- suggested to contain carbonate replacing HPO42- in OCP structure
mental approach to bone regeneration by synthetic OCP in com- [25], which was later ascribed to be a state that carbonate resides on
parison with other calcium phosphate bone substitutes, such as inner surfaces of HA formed during the hydrolysis of OCP [26]. It
sintered HA ceramic, a widely used bioceramic clinically [15,16]. is also becoming clear that the major incorporation of impurities
(e.g., carbonate ions) into biominerals takes place during the hy-
2. CALCIUM PHOSPHATE COMPOUNDS drolysis process of OCP [27-29].

Table 1 shows the calcium phosphate compounds, having dif- 3. STRUCTURAL RELATION BETWEEN OCP AND HA
ferent Ca/P molar ratio, including OCP and HA. It has been sug-
gested that dicalcium phosphate dihydrate (DCPD), the anhydrous The crystal structure of OCP was determined by Brown in the
form of DCPD, DCP anhydrous (DCPA), OCP and tricalcium –
first place [5]. OCP is triclinic with space group P 1, with lattice
parameters a=19.692Å, b=9.523Å, c=6.835Å,
=90.15°, =92.54°,
=108.65° and Z=2. HA is hexagonal with P63/m with a=b=9.432Å
*Address correspondence to this author at the Division of Craniofacial Function Engi-
neering, Tohoku University Graduate School of Dentistry, 4-1 Seiryo-machi, Aoba-ku,
and c=6.881Å, Z=1 [30]. The epitaxial growth of apatite crystals on
Sendai 980-8575, Japan; Fax: +81-22-7171-7637; OCP was proposed to occur during biological apatite formation
E-mail: suzuki-o@[Link] based on the structural similarity between OCP and HA [6,31]. The
0929-8673/08 $55.00+.00 © 2008 Bentham Science Publishers Ltd.
306 Current Medicinal Chemistry, 2008 Vol. 15, No. 3 Suzuki et al.

Table 1. Calcium Phosphate Compounds Relating to Hydroxyapatite

Calcium phosphate Abbreviation Chemical formula Ca/P molar ratio

dicalcium phosphate, anhydrous DCPA CaHPO4 1.00

dicalcium phosphate dihydrate DCPD CaHPO4
2H2O 1.00
octacalcium phosphate OCP Ca8H2(PO 4)6
5H2O 1.33
tricalcium phosphate TCP Ca3(PO4)2 1.50

amorphous calcium phosphate ACP Ca3(PO4)2
nH2O 1.50

Calcium-deficient hydroxyapatite Ca-deficient HA Ca10-XH X(PO4)6(OH)2-X 1.5-1.67

hydroxyapatite HA Ca10(PO 4)6(OH)2 1.67

tetracalcium phosphate TTCP Ca4(PO4)2O 2.00

mechanism of carbonated HA overgrowth on OCP was demon- cally compared to analyze the osteoconductive characteristics. HA
strated in vitro [32]. The planar defects in synthetic HA, usually used was commercially available, sintered at 1150°C and had 90μm
observed in biological apatite crystals as the central dark line, was average diameter pore, and 30 to 40% porosity. The histomor-
ascribed as evidence of the existence of OCP precursor [32]. Be- phometric analysis (Table 2) showed that the area of bone apposi-
cause OCP is composed of hydrated layer stacked with apatite layer tion was significantly higher on OCP than that of the sintered po-
alternatively, its structure is closely related to HA structure [6]. rous HA ceramic between 2 and 3 weeks, whereas it subsequently
became smaller on OCP than on the HA ceramic. The area attacked
by MNGCs, including tartrate-resistant acid phosphatase (TRAP)-
4. OCP BONE SUBSTITUTE positive cells, was significantly higher for OCP than for the HA
Osteoconductive nature of synthetic OCP has been found first ceramic at 8 weeks. The results disclosed some osteoconductive
by its subperiosteal implantation in mouse calvaria 10 years ago characteristics of synthetic OCP in the bone marrow space: 1) en-
[33,34]. In these studies, however, new bone was not induced by hancement of bone regeneration at the initial bone apposition stage,
subcutaneous implantations, suggesting that OCP may stimulate and 2) stimulation of resorption of the newly formed bone, coupled
periosteal cells, which is absent from the soft tissue sites, to en- with OCP biodegradation in relation to TRAP positive osteoclast-
hance bone formation in mouse [35]. Several studies have been like cells [48]. OCP did not induce bone in ectopic site, such as the
conducted to investigate possible use of synthetic OCP as a bone implantation in mouse subcutaneous tissue [34], whereas induced
regenerative scaffold in various forms, such as coatings on metallic bone ectopically in goat muscle [37], suggesting that osteoinductive
implants [36-42], microscaffold self-assembled [43,44] and gran- characteristics may be acquired in some animal species because of
ules [33,34,45-49]. These studies disclosed various biological char- possible supply of osteoblast-lineage cells in peripheral blood to the
acteristics of OCP, including proliferation and differentiation of implantation site [37,53,54].
osteoblastic cell lineages on OCP [43,44,49] and the ability to form
bone by implantation of porous metal coated with OCP in back
muscle of goats [37], implantation of bone marrow cells seeded
OCP in nude mouse subcutaneous tissue [40] or bone defects
[33,34,45-49].
Several methods to prepare OCP have been reported [33,50,51].
Fig. (1) shows a scanning electron micrograph of OCP particles
synthesized at pH 5 to 6 and 70°C, according to a method previ-
ously reported [33]. OCP crystals exhibited platy morphology, sev-
eral μm in length; usually those crystals showed smooth surfaces
and smooth terminals [22]. OCP granules obtained from the synthe-
sized cake (sieved from 16 to 32 mesh: 0.5-1.0 mm in diameter)
were implanted intramedullary in rabbit femurs for 12 weeks [48].
Fig. (2) is an example that OCP would be an osteoconductive bone
regenerative material. These figures show undecalcified and hema-
toxylin-eosin stained histological sections of OCP at 1, 3 and 8
weeks after implantation. Many osteoblasts with a cuboidal shape
were aligned on the OCP implant surfaces at one week (Fig. 2a).
Non-calcified osteoid tissue was visible between osteoblasts and Fig. (1). A scanning electron micrograph of synthetic OCP crystals. OCP
calcified bone matrix directly formed on the OCP implant. Calci- crystals exhibited platy morphology, several μm in length; usually those
fied bone matrix was formed and observed as trabecular bone at 3 crystals showed smooth surfaces and smooth terminals. Bar=2μm.
weeks (Fig. 2b). A multinucleated giant cell (MNGC) was attached
to the bare OCP surface (Fig. 2b). Layered trabecular bones were 5. CONVERSION OF OCP INTO HA IN VITRO
formed at 8 weeks (Fig. 2c). Thus, OCP provided a site to bond
elongated tissue from surrounding bone. It has been reported that A number of studies in vitro hydrolysis experiments have been
OCP initiates bone apposition from its surfaces apart from host conducted to elucidate the mechanism of OCP conversion assuming
bone tissue in rat calvaria defects, which was demonstrated by his- the physiological conditions in the presence of various molecules. It
tological examination using serial sections of the OCP implant [52]. has been reported that the rate of conversion into HA is considera-
These results indicate the osteoconductive characteristics of OCP in bly affected by the solution pH [55,56]. The carboxylate ions have
orthotopic site. Appositional bone formation and MNGCs on the been shown to be incorporated into hydrated layer of OCP [57].
granules of OCP and sintered HA ceramic were histomorphometri-
Bone Regeneration and OCP Current Medicinal Chemistry, 2008 Vol. 15, No. 3 307

Fig. (2). Photographs of undecalcified histological sections of OCP implantation, stained with hematoxylin and eosin. (a): 1 week; (b): 3 weeks; (c): 8 weeks.
Note that: 1) osteoid tissue (arrows) was formed by osteoblasts with a cuboidal shape on the OCP implant surface (a); 2) mature appositional bone (trabecular
bone) was formed around OCP and MNGC (arrows) was directly attached to the OCP surface (b); 3) layered trabecular bone was formed and MNGC was
observed at the site of the concave surface of the OCP implant (c), indicated by an arrow. Asterisk: implants. Bars = 50μm. Reproduced from Imaizumi et al.
[48] with permission from Springer Science+Business Media, Inc.

Table 2. Histomorphometric Analysis of Appositional Bone Formation and MNGCs on OCP and Sintered HA Ceramic

Appositional bone formation1 %±SE MNGCs2 %±SE

weeks
OCP HA OCP HA

1 8.8(3.8) 7.0(2.8) 13.8(4.2) 26.0(3.9)

2 45.4(4.2) 21.6(7.3) 17.6(4.2) 17.9(2.3)

3 44.5(3.2) 19.2(5.1) - -
4 33.9(6.2) 28.7(1.7) 16.5(2.5) 11.9(1.0)

6 32.9(0.7) 23.3(5.8) - -
8 36.8(4.4) 33.1(8.0) 27.8(4.4) 14.9(1.7)

12 9.3(3.1) 19.2(1.2) - -
1
Calculated from the length of mineralized surface per implant circumference.
2
Calculated from the length attacked by multinucleated giant cells (MNGCs) per implant circumference.
Reproduced from Imaizumi et al., [48] with permission from Springer Science+Business Media, Inc.

Carbonate, citrate, fluoride, and cations have been found to influ- contamination with a lower Ca/P molar ratio, such as dicalcium
ence the conversion of OCP into HA [58-61]. Amino acid mole- phosphate dihydrate (DCPD). It seems likely that the OCP batch
cules have been demonstrated to affect the crystallization and mor- used included excess protons, as expected by Mathew et al. [20],
phology of OCP through interaction between the molecules and based on the previous work which analyzed acid phosphate in OCP
OCP surface [62,63]. [21]. The OCP hydrolyzate (referred to as HL hereafter) in 48 hours
Tung et al. indicated that OCP hydrolysis in a physiological of hydrolysis obtained at 70°C had a non-stoichiometric composi-
condition takes place in two stages: the fast initial process, which is tion, exhibiting a low Ca/P value of 1.46, compared to the
attributed to the surface topotactical conversion, followed by the stoichiometric Ca/P value of 1.67. It was found that acid phosphate
main, slower process, which involves the nucleation and crystal decreased consistently with the progression of OCP-apatite conver-
growth [64]. Tseng et al. observed central dark line in HA forma- sion; however, the resulting apatite crystals still contained high acid
tion through topotaxial OCP hydrolysis in vitro [65]. OCP can be phosphate, as observed for biological apatite crystals [2].
hydrolyzed into HA without changing its morphology through the Synthetic OCP, having a non-stoichiometric composition of
topotactical conversion [6]. OCP-apatite conversion brings about Ca/P molar ratio 1.26, was hydrolyzed by incubating the crystals in
progressive increase of Ca/P molar ratio in the crystal chemical a 2ppm fluoride-containing solution at 37°C in 150mM Tris buffer
composition toward that of HA [22]. Table 3 shows the chemical (pH 7.4) [22]. The hydrolysis was completed within after 10 days,
composition of OCP and the converted products obtained experi- which can be confirmed by XRD and Fourier transform infrared
mentally through hydrolysis at 70°C hot water. Stoichiometric HA (FTIR) analyses. The chemical composition of the hydrolysis prod-
with a Ca/P ratio of 1.67 is also shown as reference in the table. The ucts had a similar sequential change with the 70°C hydrolysis prod-
OCP batch was Ca-deficient, with a Ca/P molar ratio of 1.23. The ucts in individual element, such as Ca. Fluoride content of the crys-
acid phosphate (38.9% of the total P) was slightly higher than that tals was relatively constant regardless of the degree of hydrolysis
expected for OCP stoichiometry (33% of the total P), although (0.22-0.25 wt%). Table 4 shows solution composition and its satu-
analyses by x-ray diffraction (XRD) ruled out calcium phosphate ration levels during OCP hydrolysis. Remarkable changes with the
308 Current Medicinal Chemistry, 2008 Vol. 15, No. 3 Suzuki et al.

Table 3. Chemical Composition of Synthetic OCP and its Hydrolyzed Apatitic Products (OCP Hydrolyzates; HLs) at 70°C Hot Water

Calcium phosphates Ca wt% P wt% HPO4 % of total P Ca/P molar ratio SSA2 m2/g

Original OCP 31.1 19.6 38.9 1.23 16.0

HL, hydrolyzed for 6 h 31.4 19.5 32.1 1.24 26.0
HL, hydrolyzed for 48 h 35.5 18.8 16.4 1.46 45.0
1 1 1
Stoichiometric HA 38.3 18.9 -
1.67 -

1
Numerical values, calculated theoretically; 2SSA, specific surface area.
Reproduced from Suzuki et al., [49] with permission from Elsevier Ltd.

advance of OCP-apatite conversion were that [22,66]: (1) Ca con- bility of OCP follows this mechanism during bone formation. Our
centration in the supernatants decreased from 0.4mM to 0.16mM; previous studies showed that synthetic OCP in granule form en-
(2) phosphate concentration in the supernatants increased from hanced bone formation more than various synthetic HAs, including
0.62mM to 0.96mM; (3) degree of supersaturation (DS) with re- stoichiometric non-sintered HA [33,70] and sintered HA ceramic
spect to HA in the supernatants decreased from 6.67
107 to [48], having Ca/P molar ratio of 1.67, and also non-sintered Ca-
2.10
106, and DS with respect to OCP also decreased from 8.7
10- deficient HA obtained via direct precipitation [33], and further
1
to 7.4
10-2. When OCP was incubated in the absence of fluoride, more than commercially available -TCP ceramic [70]. However,
they were 1.21
108 with respect to HA and 1.2 with respect to careful examination should be required to conclude the osteogenic
OCP. DS values were calculated using the solubility product con- capability of calcium phosphate bone substitute materials. Recent
stants 2.63
10-60 for HA [67] and 1.05
10-47 for OCP [51]. The studies disclosed that the microstructure of calcium phosphate ce-
change in the DS value revealed that hydrolysis advances with un- ramics greatly influences their bone regenerative property and bio-
dersaturation with respect to OCP under supersaturation with re- degradation [71,72].
spect to HA. The results also suggest that OCP dissolves and then Ca-deficient HA obtained via OCP hydrolysis, rather than Ca-
re-precipitates as HA. deficient HA through direct precipitation [33, 73] or sintered HA
ceramic, would probably be the control material in essence to OCP
6. CONVERSION OF OCP INTO HA IN VIVO COUPLED for such the comparative purpose. Ca-deficient HA formed via OCP
WITH OSTEOGENIC RESPONSE contained a small amount of OCP as a residual inclusion within the
It has been found that synthetic OCP can be converted into HA formed apatite and further retained the original plate-like morphol-
when implanted in subperiosteal region of mouse calvaria [33,34], ogy of OCP [22,74]. The granules of OCP or HL obtained at 70°C
in thigh muscle pouches of mice [68] and in critical sized calvaria (sieved from 32 to 48 meshes: 300-500μm) were implanted into rat
bone defect [49]. Fig. (3) shows XRD patterns of the OCP granules calvaria critical sized 9mm defects in diameter to examine bone
before (Fig. 3a) and after implantation in rat calvarial defects for 10 regenerative property of OCP. The critical sized defect is known as
days (Fig. 3b) and 21 days (Fig. 3c). The patterns showed an in- a defect that does not heal spontaneously [75]. Fig. (4) shows his-
crease of amorphous material, suggesting that the crystallinity of tomorphometrical findings on the percentage of newly formed bone
OCP implanted decreases with implantation period. The remarkable (n-Bone%) in the defect. In week 4, the n-Bone% ± standard devia-
changes were the reduction of both intensity of OCP primary (100) tion (SD) in the untreated, and OCP and HL treated areas was 11.65
reflection around 4.8 degrees at 2 and (700) peak around 33.6 ± 4.66, 7.17 ± 6.57, and 5.14 ± 3.57, respectively. In week 12, the
degrees at 2. The (100) peak of HA around 11 degrees at 2 was n-Bone% ± SD in the untreated, OCP and HL treated areas was
observed at 10 days after implantation (indicated by arrow in Fig. 9.64 ± 4.34, 30.20 ± 10.60, and 4.36 ± 2.71, respectively. The bio-
3b) although it was unclear in 21 days implantation. These changes degradable nature of OCP may facilitate bone regeneration com-
could be an indication that OCP tends to convert to HA, based on pared to HL in this experimental model. However, previous in vitro
the XRD examinations of OCP hydrolysis into HA observed in study demonstrated that the differentiation of osteoblastic cells was
acellular culture medium [49] or in fluoride-containing solution facilitated only when incubated on OCP surface where OCP tends
[22] under physiological condition. The result suggests that syn- to convert to HA gradually, whereas such enhancement was not
thetic OCP tends to convert to an apatite structure by implantation substantiated on HL surface [49]. These results suggest that OCP-
into rat calvarial bone defect, identical to the structural changes HA conversion in itself may be involved in stimulating osteoblastic
observed for implantation into mouse calvaria and subcutaneous cell differentiation. Taken together, it is reasonable to assume that
tissue [33,34]. While it is well accepted that the acquisition of os- osteoconductive characteristics of OCP are induced not only by the
teoconductive characteristics requires the formation of bone-like formation of bone-like apatite on the OCP template but by the con-
apatite on the biomaterial surfaces before initiating direct apposi- version process from OCP to HA in itself.
tion of new bone [69], it is still uncertain whether osteogenic capa-

Table 4. Solution Composition and its Saturation Levels During OCP Hydrolysis

Degree of supersaturation
Time days Calcium mM Phosphate mM
HA OCP
7
1 0.4 0.62 6.67
10 8.7
10-1
7
3 0.32 0.75 4.27
10 6.7
10-1
6
7 0.16 0.96 2.10
10 7.4
10-2
7* 0.48 0.53 1.21
108 1.2

Hydrolysis: pH=7.4 at 37, 150mM Tris-HCl buffer, initial [F]=2ppm, solid/solution=100mg/100mL.

*Initial [F]=0ppm.
Reproduced from Suzuki et al., [66] with permission from Wiley Periodicals, Inc.
Bone Regeneration and OCP Current Medicinal Chemistry, 2008 Vol. 15, No. 3 309

planted in subperiosteal region of 7-week-old BALB/c mice for 3

weeks. It was confirmed that the implantation over 10 days in
mouse calvaria is sufficient to initiate OCP-apatite conversion in
vivo under micro-beam XRD studies [33,34]. The calvarias contain-
ing OCP implant with cranial skins were dissected and fixed with a
mixture of 2% paraformaldehyde and 2% glutaraldehyde in 0.1M
cacodylate buffer at pH 7.3-7.4. After decalcification with 5%
EDTA in 0.1M cacodylate buffer, specimens were post-fixed with
1% OsO4, dehydrated with graded series of ethanol, passed through
propylene oxide and embedded in Epon. Ultra-thin sections
mounted on cupper grids were stained with uranyl acetate and lead
citrate and examined with a Hitachi H-700 transmission electron
microscope (TEM) at 100KV. Fig. (5) shows TEM micrographs of
the interface between an OCP granule implanted and newly formed
bone using the decalcified specimens in 13 days implantation. Os-
teoblasts aligned on the newly formed bone matrix (Fig. 5a). Many
fine filaments and small granular materials were observed in area
where OCP particles existed before decalcification (Fig. 5b), which
was identical to the biomolecules adsorbed onto OCP particles.
These biomolecules were previously characterized as circulating
serum proteins adsorbed on OCP implants [34]. The main compo-
nents of bone matrix were thick collagen fibers (Fig. 5c), whereas
the collagen invaded into the area where OCP particles existed
before the decalcification. It has been reported that OCP can be
converted to HA topotaxially in vitro [64,65,74]. The particle mor-
phology of original OCP remained unchanged during conversion.
On the contrary, it is possible that dissolution-re-precipitation proc-
ess is involved in the conversion to form de novo apatite [55,76]. If
Fig. (3). X-ray diffraction patterns of OCP implanted into rat calvarial de- OCP-apatite conversion of the implanted OCP advances topotaxi-
fects for 10 days and 21 days. Note the decrease in OCP intensity (100) ally even in vivo, then osteoblastic cells may utilize their surfaces as
reflection at 2= 4.8° and (700) reflection at 2= 33.6°, suggesting that the scaffold to deposit new bone directly during the conversion. In
crystallinity of OCP implanted decreases with implantation period. Also, order to confirm this assumption whether the implanted OCP can be
note the appearance of apatite reflection (arrow in b (10 days)), although it converted to HA with retaining its original morphology, TEM ob-
was not obvious in 21 days. It suggests that implanted OCP tends to be servation was carried out for the undecalcified specimens. The cal-
converted into apatite in vivo. Reproduced from Suzuki et al. [49] with varias containing OCP implant with cranial skins were dissected
permission from Elsevier Ltd. and fixed with 10% formalin, dehydrated by graded series of etha-
nol, and embedded in methylmethacrylate resin and polymerized at
36°C, without decalcification. Ultra-thin sections mounted on cup-
per grids were directly examined with a Hitachi H-700 TEM at 200
KV. Fig. (6) shows TEM micrographs of an interface between OCP
implanted and newly formed bone. It was apparent that plate-like
morphology of OCP particles (indicated by asterisk in Fig. 6a) was
retained even after 3 week implantation (after conversion from OCP
to HA). A number of nano-scale de novo crystals were also ob-
served on the OCP particles (Fig. 6b), which might be stemmed
from solution-mediated poorly crystalline HA formation in the
vicinity of the implanted OCP particles. These morphological char-
acteristics were identical to the de novo formation of HA on the
surface of OCP crystals during OCP-apatite conversion in vitro
[22]. Nevertheless, the platy morphology of the original OCP (Fig.
6a, 6b) was still retained in general. The results suggest that the
topotaxial conversion from OCP to HA is taking place in vivo. The
topotaxial conversion should be a distinctive feature of OCP
amongst calcium phosphate bone substitute materials.
Fig. (4). Experimentally stimulated bone formation by the implantation of
OCP and OCP hydrolyate (HL) granules. Histomorphometrical findings 8. BIODEGRADABLE NATURE OF SYNTHETIC OCP
regarding the area of newly formed bone (n-Bone%) in the defect. Data are
the means±SD of five specimens. p<0.05 (unpaired t test). n-Bone% was It has been shown that mature osteoclasts obtained from a rabbit
significantly higher in OCP at 12 weeks than in HL or the untreated group. long bone can resorb only synthetic carbonate apatite but not
Reproduced from Suzuki et al. [49] with permission from Elsevier Ltd. DCPD, OCP,
-TCP and HA [77,78]. However, OCP became to be
resorbed by the osteoclasts if pre-incubated in culture media, sug-
7. INTERFACE BETWEEN OCP AND NEWLY FORMED gesting occurrence of the surface conditioning of OCP, although
BONE AT ULTRASTRUCTURAL LEVEL crystalline phases other than OCP were not substantiated [77]. It
has been indicated that osteoclastic resorption was observed around
It is of interest to investigate how newly formed bone matrix is carbonate apatite and
-TCP in bone defects [18,79]. It has been
connected to implanted OCP and how OCP-apatite conversion ad- shown that implanted OCP in rat bone marrow was resorbed by
vances during bone formation at ultrastructural level. OCP granules MNGCs developing a ruffled border-like structure and a clear zone-
(sieved from 16 to 32 mesh: 0.5-1.0 mm in diameter) were im- like structure, showing ultrastructural characteristics with osteo-
310 Current Medicinal Chemistry, 2008 Vol. 15, No. 3 Suzuki et al.

Fig. (5). Ultrastructure of newly formed bone matrix around OCP implanted in subperiosteal region of a 7-week-old BALB/c mouse calvaria for 13 days,
examined using decalcified sections. (a) new bone (NB) matrix formed by osteoblasts on OCP. Bar=1μm. (b) Interface between newly formed bone matrix and
OCP. Bar=0.4μm. (c) Interface between newly formed bone matrix and OCP of other area than (b). Bar=0.2μm. Newly formed bone matrix on OCP was com-
posed of many collagen fibers. At the interface between OCP and the collagen fibers, OCP (arrows) was invaded by the collagen fibers.

clasts, whereas MNGCs around synthetic non-sintered HA im- applications where rapid bone formation and concomitant implant
planted lacked such a feature [66, 80]. These results should be resorption are important considerations.
associated with the biodegradable characteristics of OCP
[38,48,70,80]. However, it is still unidentified whether OCP has a 9. OSTEOBLASTIC CELL RESPONSES TO SYNTHETIC
potential role to enhance osteoclast genesis from the precursor cells. OCP
There has been a report that an OCP synthesized via direct
precipitation [33] is more resorbable than
-TCP in rat cranial Human osteoblastic cell proliferation and differentiation have
defect [70], which is not inconsistent with the solubility of these been investigated with deposited OCP [39]. This study concluded
salts; OCP is thermodynamically less stable than
-TCP under that OCP coating favors osteoblast proliferation, activation of their
physiological condition [17]. A recent study, using
-TCP with metabolism and differentiation, compared to the original titanium
different pore sizes, showed that the biodegradability is markedly surfaces. Proliferation and differentiation of mouse bone marrow
affected by the pore size of such a ceramic scaffold [81]. It is stromal ST-2 cells and primary calvarial osteoblastic cells were
required to conclude as to which calcium phosphate is more resorb- investigated with OCP, HA obtained through hydrolysis of OCP
able in vivo milieu from the view point of the microstructure, (HL) [49] and non-sintered stoichiometric HA which were pre-
although both calcium phosphate scaffolds should be intrinsically coated on culture dishes [66]. The proliferation of ST-2 and primary
biodegradable in nature unlike thermodynamically stable HA under osteoblasts on OCP was initially inhibited, compared to HL and
physiological condition. Experimental evidence accumulated as to stoichiometric HA, whereas their differentiation to osteoblasts,
OCP, however, suggests that synthetic OCP would be a more useful examined by osteoblast-related genes, such as osteocalcin and Os-
bone substitute than HA in implant applications where rapid bone
Bone Regeneration and OCP Current Medicinal Chemistry, 2008 Vol. 15, No. 3 311

Fig. (6). Transmission electron micrographs of OCP implanted in subperiosteal region of a 7-week-old BALB/c mouse calvaria for 3 weeks, examined using
undecalcified ultra-thin section. The implantation period was ascertained to be sufficient to initiate conversion from OCP into HA in previous studies
([33,34]). (a) Interface between a couple of OCP crystals of the granule implanted (indicated by asterisk) and newly formed bone. Bone crystals in calvaria
were shown to be thin-blade in appearance. (b) Area of inside of an OCP granule, which was composed of aggregate of individual OCP crystals. Bars=0.5μm.
The platy morphology of original OCP crystals, shown in Figure 1, was retained even after the implantation (Figures 6a and 6b), suggesting that topotaxial
conversion of OCP into HA is taking place. Nano-scale de novo crystals were seen on the OCP template, which might be stemmed from solution-mediated
conversion from OCP to HA.

terix, was promoted on OCP. OCP tended to convert to apatite by (the central dark line), which can be observed in the center portion
the incubation, suggesting that OCP-apatite conversion may be of enamel crystals at the electron microscopic level, is believed to
involved in the capacity to facilitate osteoblastic cell differentiation be one or two unit cells of initially formed OCP crystals [32, 59, 91,
[49]. OCP did not inhibit the attachment of rat periodontal ligament 92]. It has been recognized that the apatite crystal deposition com-
cells, including potential osteogenic cell population, but enhanced mences in matrix vesicles which are secreted by osteoblasts at the
the proliferation [82]. Rat bone marrow stromal cells have been early stage of bone formation [93,94]. Careful examination by FTIR
cultured to three-dimensional OCP crystal assemblies [43,44] on demonstrated that the initial minerals formed within the isolated
the proliferation and differentiation, examined by osteoblast-related matrix vesicles were OCP [95,96]. Energy-dispersive X-ray micro-
genes, such as collagen type I and Cbfa1. The studies concluded analysis showed that Ca/P ratio of calcium phosphate within the
that OCP assemblies support positively the cell metabolism and matrix vesicles in calcifying bone increased with development [97].
could be used for ex vivo construction of bone defects [43,44]. It is A study by energy-filtering transmission electron microscope
uncertain how OCP acquires the properties of osteoconduction and (EFTEM) further determined the precise distribution of Ca and
in some case osteoinduction [37]. It was suggested that bone induc- phosphorus associated with bone matrix proteins around matrix
tive proteins such as bone morphogenetic proteins (BMPs) can be vesicles [98]. Kinetic study indicated that ultrafiltered human serum
adsorbed on OCP [83,84], and that OCP locally raises calcium ion had a composition suitable for the transformation of OCP to HA
concentrations around OCP [44]. A recent study showed that [99]. Weiner et al. proposed that initially depositing a less ordered
mononuclear macrophage-like cells can be observed adjacent to mineral transforms into a more crystalline mature phase and is
OCP implant where osteoblasts are actively forming bone [85]. The widespread in vertebrate, supporting a concept that the transient
elevated extracellular calcium stimulated secretion of BMP-2 by a precursor strategy occurs in vertebrate [100].
macrophage cell line in vitro, suggesting a possible role to stimulate
osteoblasts during bone formation and OCP biodegradation in vivo Biological function of OCP has not been proposed in biominer-
[85]. It may be required to investigate the local environment close alization of bone and tooth, except for the specific role to be the
to the OCP surface to elucidate the osteogenic cell functions. loci to nucleate HA through a process of OCP-apatite conversion. It
has been found out that serum constituents, such as
2HS glycopro-
10. POTENTIAL ROLE OF OCP IN BIOMINERALIZATION teins, are selectively adsorbed onto synthetic OCP shortly after its
implantation into the subperiosteal region of mouse calvaria
Mineralization events taking place in bone and tooth are sub- [33,34]. Moreover, it has been indicated that the hydrolysis of OCP
stantially organized under controls by the tissue-specific cells and affects the interaction with such serum constituents [22]. In vitro
the secreted matrix proteins. It is well known that the tooth enamel studies demonstrated that OCP is capable of stimulating osteoblas-
apatite crystals exhibited extremely elongated morphology and tic cell differentiation [43,44,49] compared to HA [49]. Although
oriented arrangement parallel to the crystal c-axis [86,87]. A previ- the mechanism of the initiation of biomineralization remains con-
ous study established the mechanism of c-axial-oriented growth of troversial [13,14], multiple lines of evidence by synthetic OCP
apatite along with microribbon of self-assembled amelogenin nano- obtained in vivo and in vitro support a concept that OCP, although
spheres [88]. Amelogenin is a protein which is the most abundant not sufficient in itself to induce new bone, when present as a pre-
matrix protein involved in enamel mineralization [89]. Iijima et al. cursor phase, play an important role in the bone-forming process
demonstrated that OCP crystals enlarged their sizes toward to c-axis [35].
of the crystal in the presence of porcine amelogenin, whereas the
elongated OCP crystals became to convert into HA with coexis- 11. CONCLUSION
tence of fluoride [90]. This elongation is explained by the inhibition
of OCP crystal growth toward the (010) face by amelogenin Transplantation of autologous bone grafts is still the gold stan-
through selective hydrophobic interaction with this face [90]. dard for orthopedic, oral and plastic surgical procedures because of
its excellent biocompatibility, biodegradation coupled with active
Indirect evidence to OCP inclusion in biomineralization has
bone regeneration [101]. However, it has disadvantages especially
been reported for tooth and bone tissues. The central planar defect
limited availability and the harvest procedure involves high donor
312 Current Medicinal Chemistry, 2008 Vol. 15, No. 3 Suzuki et al.

site morbidity. Synthetic OCP is a possible alternative bone substi- [30] Kay, M. I.; Young, R. A.; Posner, A. S. Nature, 1964, 204, 1050.
[31] Brown, W. In Tooth Enamel II; Stack, M. V., Fearnhead, R. W., Eds.; John
tute since OCP shows a unique osteoconductive characteristic com- Wright: Bristol, 1965, p. 11.
pared to the calcium phosphates clinically used, such as HA ce- [32] Nelson, D. G. A.; Wood, G. J.; Barry, J. C.; Featherstone, J. D. B. Ultrami-
ramic. OCP is however, brittle in nature therefore can not be easily croscopy, 1986, 19, 253.
molded as scaffold to large bone defects. Compounding with poly- [33] Suzuki, O.; Nakamura, M.; Miyasaka, Y.; Kagayama, M.; Sakurai, M. To-
hoku J. Exp. Med., 1991, 164, 37.
mers such as collagen is one of a measure to overcome this poor [34] Suzuki, O.; Nakamura, M.; Miyasaka, Y.; Kagayama, M.; Sakurai, M. Bone
mouldability [102]. Collagen composites with HA [103-106] or
- Miner., 1993, 20, 151.
TCP [107-110] have been developed and recently studied for such [35] Eidelman, N.; Eanes, E. In Octacalcium Phosphate; Chow, L., Eanes, E. D.,
large defects. OCP combined with collagen has also been fabricated Eds.; Karger: Switzerland, 2001; Vol. 18, pp. 50-76.
[36] Barrere, F.; Layrolle, P.; van Blitterswijk, C. A.; de Groot, K. Bone, 1999,
and shown to have not only moldability but also to retain intrinsic 25, 107S.
osteoconductivity of OCP [102]. Synthetic OCP combined with [37] Barrere, F.; van der Valk, C. M.; Dalmeijer, R. A.; Meijer, G.; van Blitter-
such polymer is one of candidate to autologous bone grafts. The swijk, C. A.; de Groot, K.; Layrolle, P. J. Biomed. Mater. Res. A, 2003, 66,
elucidation of the molecular mechanism to enhance bone regenera- 779.
[38] Barrere, F.; van der Valk, C. M.; Dalmeijer, R. A.; van Blitterswijk, C. A.; de
tion by synthetic OCP may lead to the recognition of OCP as scaf- Groot, K.; Layrolle, P. J. Biomed. Mater. Res. A, 2003, 64, 378.
fold to bone regeneration because it is a potential precursor of bio- [39] Bigi, A.; Bracci, B.; Cuisinier, F.; Elkaim, R.; Fini, M.; Mayer, I.; Mi-
logical mineralization in bone and tooth. hailescu, I. N.; Socol, G.; Sturba, L.; Torricelli, P. Biomaterials, 2005, 26,
2381.
[40] Dekker, R. J.; de Bruijn, J. D.; Stigter, M.; Barrere, F.; Layrolle, P.; van
ACKNOWLEDGEMENTS Blitterswijk, C. A. Biomaterials, 2005, 26, 5231.
[41] Habibovic, P.; Van der Valk, C. M.; Van Blitterswijk, C. A.; De Groot, K.;
This study was supported in part by grants-in aid (17076001, Meijer, G. J. Mater. Sci. Mater. Med., 2004, 15, 373.
18659567,19390490) from the Ministry of Education, Science, [42] Lin, S.; LeGeros, R. Z.; LeGeros, J. P. J. Biomed. Mater. Res. A, 2003, 66,
819.
Sports and Culture of Japan, and grants from UEHARA MEMO- [43] Liu, Y.; Cooper, P. R.; Barralet, J. E.; Shelton, R. M. Biomaterials, 2007, 28,
RIAL FOUNDATION. The authors thank Mr. Masatoshi Yamada 1393.
of JGC Corporation for his help in the study of transmission elec- [44] Shelton, R. M.; Liu, Y.; Cooper, P. R.; Gbureck, U.; German, M. J.; Barralet,
tron microscope. The authors also thank Professors Ryutaro Ka- J. E. Biomaterials, 2006, 27, 2874.
[45] Sugihara, F.; Mandai, Y.; Yoshikawa, M.; Toda, T. J. Dent. Res., 1995, 74,
mijo, Masanori Nakamura of Showa University School of Den- 973.
tistry, Professors Hidetoshi Shimauchi, Keiichi Sasaki, Seishi [46] Sasano, Y.; Kamakura, S.; Nakamura, M.; Suzuki, O.; Mizoguchi, I.; Akita,
Echigo, and Yasuyuki Sasano of Tohoku University Graduate H.; Kagayama, M. Anat. Rec., 1995, 242, 40.
School of Dentistry for their joint research and helpful suggestions [47] Kamakura, S.; Sasano, Y.; Homma, H.; Suzuki, O.; Kagayama, M.; Motegi,
K. J. Dent. Res., 1999, 78, 1682.
and valuable criticism during the preparation of the manuscript. [48] Imaizumi, H.; Sakurai, M.; Kashimoto, O.; Kikawa, T.; Suzuki, O. Calcif.
Tissue Int., 2006, 78, 45.
REFERENCES [49] Suzuki, O.; Kamakura, S.; Katagiri, T.; Nakamura, M.; Zhao, B.; Honda, Y.;
Kamijo, R. Biomaterials, 2006, 27, 2671.
[1] Elliot, J. C. In Tooth Enamel; Fearnhead, R. W., Stack, M. V., Eds.; John [50] LeGeros, R. Z. Calcif. Tissue Int., 1985, 37, 194.
Wright: Bristol, 1965, pp. 20-22. [51] Moreno, E.; Brown, W.; Osborn, G. Soil Sci. Soc. Am. Proc., 1960, 24, 99.
[2] Aoba, T. Anat. Rec., 1996, 245, 208. [52] Kamakura, S.; Sasano, Y.; Homma, H.; Suzuki, O.; Kagayama, M.; Motegi,
[3] Kim, H. M.; Rey, C.; Glimcher, M. J. J. Bone Miner. Res., 1995, 10, 1589. K. Oral Dis., 2001, 7, 259.
[4] Betts, F.; Blumenthal, N. C.; Posner, A. S. J. Cryst. Growth, 1981, 53, 63. [53] Eghbali-Fatourechi, G. Z.; Lamsam, J.; Fraser, D.; Nagel, D.; Riggs, B. L.;
[5] Brown, W. E. Nature, 1962, 196, 1048. Khosla, S. N. Engl. J. Med., 2005, 352, 1959.
[6] Brown, W. E.; Smith, J. P.; Lehr, J. R.; Frazier, A. W. Nature, 1962, 196, [54] Sata M; Tanaka K; Nagai R. N. Engl. J. Med., 2005, 353, 737.
1050. [55] Tomazic, B. B.; Tung, M. S.; Gregory, T. M.; Brown, W. E. Scan. Electron
[7] Bodier-Houlle, P.; Steuer, P.; Voegel, J. C.; Cuisinier, F. J. Acta Crystallogr. Microsc.,1989, 3, 119.
D Biol. Crystallogr., 1998, 54, 1377. [56] Zhang, J. W.; Nancollas, G. H. J. Phys. Chem., 1992, 96, 5478.
[8] Tohda, H.; Yamada, M.; Yamaguchi, Y.; Yanagisawa, T. J. Electron Mi- [57] Markovic, M.; Fowler, B. O.; Brown, W. E. In Hydroxyapatite and Related
crosc., 1997, 46, 97. Materials; Brown, P. W., Constantz, B., Eds.; CRC Press: Boca Raton, 1994;
[9] Crane, N. J.; Popescu, V.; Morris, M. D.; Steenhuis, P.; Ignelzi, M. A., Jr. pp. 139-144.
Bone, 2006, 39, 434. [58] Bigi A, G. M.; Ripamonti, A.; Roveri, N. J. Inorg. Biochem., 1988, 32, 251.
[10] Wu, Y.; Glimcher, M. J.; Rey, C.; Ackerman, J. L. J. Mol. Biol., 1994, 244, [59] Iijima, M.; Tohda, H.; Suzuki, H.; Yanagisawa, T.; Moriwaki, Y. Calcif.
423. Tissue Int. 1992, 50, 357.
[11] Kuhn, L. T.; Wu, Y.; Rey, C.; Gerstenfeld, L. C.; Grynpas, M. D.; Acker- [60] Sharma, V. K.; Johnsson, M.; Sallis, J. D.; Nancollas, G. H. Langmuir, 1992,
man, J. L.; Kim, H. M.; Glimcher, M. J. J. Bone Miner. Res., 2000, 15, 1301. 8, 676.
[12] Wu, Y.; Ackerman, J. L.; Strawich, E. S.; Rey, C.; Kim, H. M.; Glimcher, M. [61] Tomazic, B. B.; Mayer, I.; Brown, W. E. J. Cryst. Growth, 1991, 108, 670.
J. Calcif. Tissue Int., 2003, 72, 610. [62] Bigi, A.; Boanini, E.; Bracci, B.; Falini, G.; Rubini, K. J. Inorg. Biochem.,
[13] Boskey, A. L. J. Cell Biochem. Suppl., 1998, 30-31, 83. 2003, 95, 291.
[14] Grynpas, M.D.; Omelon, S. Bone, 2007, 41, 162. [63] Bigi, A.; Boanini, E.; Walsh, D.; Mann, S. Angew. Chem. Int. Ed. Ingl.,
[15] Bucholz, R. W. Clin. Orthop. Relat. Res., 2002, 395, 44. 2002, 41, 2163.
[16] LeGeros, R. Z. Clin. Orthop. Relat. Res., 2002, 72, 81. [64] Tung, M. S.; Tomazic, B.; Brown, W. E. Arch. Oral Biol., 1992, 37, 585.
[17] Brown, W. E.; Mathew, M.; Tung, M. S. Prog. Crystal Growth Charact., [65] Tseng, Y. H., Mou. J. C., Chan, J. C. J. Am. Chem. Soc., 2006, 128, 6909.
1981, 4, 59. [66] Suzuki, O.; Kamakura, S.; Katagiri, T. J. Biomed. Mater. Res. B Appl. Bio-
[18] Ogose, A.; Hotta, T.; Kawashima, H.; Kondo, N.; Gu, W.; Kamura, T.; Endo, mater., 2006, 77, 201.
N. J. Biomed. Mater. Res. B Appl. Biomater., 2005, 72, 94. [67] Moreno, E. C.; Kresak, M.; Zahradnik, R. T. Caries Res., 1977, 11(Suppl 1),
[19] Chow, L. In Octacalcium Phosphate; Chow, L., Eanes, E. D., Eds.; Karger: 142.
Switzerland, 2001; Vol. 18, pp. 148-164. [68] Ban, S.; Jinde, T.; Hasegawa, J. Dent. Mater. J., 1992, 11, 130.
[20] Mathew, M.; Brown, W.; Schroeder, L.; Dickens, B. J. Cryst. Spectrosc., [69] Kokubo, T. Biomaterials, 1991, 12, 155.
1988, 18, 235-250. [70] Kamakura, S.; Sasano, Y.; Shimizu, T.; Hatori, K.; Suzuki, O.; Kagayama,
[21] Suzuki, O.; Yagishita, H.; Amano, T.; Aoba, T. J. Dent. Res., 1995, 74, 1764. M.; Motegi, K. J. Biomed. Mater. Res., 2002, 59, 29.
[22] Suzuki, O.; Yagishita, H.; Yamazaki, M.; Aoba, T. Cells Mater., 1995, 5, 45. [71] Okuda, T.; Ioku, K.; Yonezawa, I.; Minagi, H.; Kawachi, G.; Gonda, Y.;
[23] Eanes, E. D.; Gillessen, I. H.; Posner, A. S. Nature, 1965, 208, 365. Murayama, H.; Shibata, Y.; Minami, S.; Kamihira, S.; Kurosawa, H.; Ikeda,
[24] Winand, L. In Tooth Enamel; Stack, M., Fearnhead, R., Eds.; John Wright T. Biomaterials, 2007, 28, 2612.
and Sons: Bristol, 1965, pp. 15-19. [72] Yokozeki, H.; Hayashi, T.; Nakagawa, T.; Kurosawa, H.; Shibuya, K.; Ioku,
[25] Hayek, E. Klin. Wochenschr., 1967, 45, 857. K. J. Mater. Sci. Mater. Med., 1998, 9, 381.
[26] Brown, W. E. Clin. Orthop. Relat. Res., 1966, 44, 205. [73] Eanes, E. D. Calcif. Tissue Res., 1970, 5, 133.
[27] Brown, W. E.; Eidelman, N.; Tomazic, B. Adv. Dent. Res., 1987, 1, 306. [74] Iijima, M.; Nelson, D. G.; Pan, Y.; Kreinbrink, A. T.; Adachi, M.; Goto, T.;
[28] Chickerur, N. S.; Tung, M. S.; Brown, W. E. Calcif. Tissue Int., 1980, 32, 55. Moriwaki, Y. Calcif. Tissue Int., 1996, 59, 377.
[29] Siew, C.; Gruninger, S. E.; Chow, L. C.; Brown, W. E. Calcif. Tissue Int., [75] Hollinger, J. O.; Kleinschmidt, J. C. J. Craniofac. Surg., 1990, 1, 60.
1992, 50, 144. [76] LeGeros, R. Z.; Daculsi, G.; Orly, I.; Abergas, T.; Torres, W. Scan. Electron
Microsc., 1989, 3, 129; discussion 137.
Bone Regeneration and OCP Current Medicinal Chemistry, 2008 Vol. 15, No. 3 313

[77] Doi, Y.; Iwanaga, H.; Shibutani, T.; Moriwaki, Y.; Iwayama, Y. J. Biomed. [94] Bonucci, E. J. Ultrastruct. Res., 1967, 20, 33.
Mater. Res., 1999, 47, 424. [95] Sauer, G. R.; Wuthier, R. E. J. Biol. Chem., 1988, 263, 13718.
[78] Doi, Y.; Shibutani, T.; Moriwaki, Y.; Kajimoto, T.; Iwayama, Y. J. Biomed. [96] Sauer, G. R.; Wu, L. N. Y.; Iijima, M.; Wuthier, R. E. J. Inorg. Biochem.,
Mater. Res., .1998, 39, 603. 1997, 65, 57.
[79] Nagayama, M.; Takeuchi, H.; Doi, Y. Dent. Mater. J., 2006, 25, 219. [97] Ozawa, H.; Yamamoto, T. J. Histochem. Cytochem., 1983, 31, 210.
[80] Kamakura, S.; Sasano, Y.; Homma-Ohki, H.; Nakamura, M.; Suzuki, O.; [98] Hoshi, K.; Ejiri, S.; Ozawa, H. J. Bone Miner. Res., 2001, 16, 289.
Kagayama, M.; Motegi, K. J. Electron Microsc. (Tokyo), 1997, 46, 397. [99] Eidelman, N.; Chow, L. C.; Brown, W. E. Calcif. Tissue Int., 1987, 41, 18.
[81] von Doernberg, M. C.; von Rechenberg, B.; Bohner, M.; Grunenfelder, S.; [100] Weiner, S.; Sagi, I.; Addadi, L. Science, 2005, 309, 1027.
van Lenthe, G. H.; Muller, R.; Gasser, B.; Mathys, R.; Baroud, G.; Auer, J. [101] Goulet, J. A.; Senunas, L. E.; DeSilva, G. L.; Greenfield, M. L. Clin. Orthop.
Biomaterials, 2006, 27, 5186. Relat. Res., 1997, 76.
[82] Hatakeyama, J.; Hatakeyama, Y.; Takahashi, I.; Suzuki, O.; Sasano, Y. Oral [102] Kamakura, S.; Sasaki, K.; Honda, Y.; Anada, T.; Suzuki, O. J. Biomed.
Dis., 2007, 13, 500. Mater. Res. B Appl. Biomater., 2006, 79B, 210.
[83] Liu, Y.; de Groot, K.; Hunziker, E. B. Bone, 2005, 36, 745. [103] Okazaki, M.; Ohmae, H.; Takahashi, J.; Kimura, H.; Sakuda, M. Biomateri-
[84] Liu, Y.; Huse, R.O.; de Groot, K.; Buser, D.; Hunziker, E. B. J. Dent. Res., als, 1990, 11, 568.
2007, 86, 84. [104] Du, C.; Cui, F. Z.; Feng, Q. L.; Zhu, X. D.; de Groot, K. J. Biomed. Mater.
[85] Honda, Y.; Anada, T.; Kamakura, S.; Nakamura, M.; Sugawara, S.; Suzuki, Res., 1998, 42, 540.
O. Biochem. Biophys. Res. Commun., 2006, 345, 1155. [105] Kikuchi, M.; Itoh, S.; Ichinose, S.; Shinomiya, K.; Tanaka, J. Biomaterials,
[86] Nylen, M. U.; Omnell, K. A.; Eanes, E. D. J. Cell Biol., 1963, 18, 109. 2001, 22, 1705.
[87] Kerebel, B.; Daculsi, G.; Kerebel, L. M. J. Dent. Res., 1979, 58, 844. [106] Itoh, S.; Kikuchi, M.; Koyama, Y.; Takakuda, K.; Shinomiya, K.; Tanaka, J.
[88] Du, C.; Falini, G.; Fermani, S.; Abbott, C.; Moradian-Oldak, J. Science, Biomaterials, 2002, 23, 3919.
2005, 309, 2166. [107] Sugaya, A.; Minabe, M.; Hori, T.; Tatsumi, J.; Watanabe, Y.; Ikeda, K.;
[89] Moradian-Oldak, J.; Iijima, M.; Bouropoulos, N.; Wen, H. B. Connect. Numabe, Y.; Hayashi, H.; Kamoi, K. J. Periodontal Res., 1990, 25, 60.
Tissue Res., 2003, 44(Suppl 1), 58. [108] Vicente, V.; Meseguer, L.; Martinez, F.; Galian, A.; Rodriguez, J.; Alcaraz,
[90] Iijima, M.; Du, C.; Abbott, C.; Doi, Y.; Moradian-Oldak, J. Eur. J. Oral. Sci., M.; Clavel, M. Ultrastruct. Pathol., 1996, 20, 179.
2006, 114(Suppl 1), 304; discussion 327, 382. [109] Shibata, Y.; Yamamoto, H.; Miyazaki, T. J. Dent. Res., 2005, 84, 827.
[91] Miake, Y.; Shimoda, S.; Fukae, M.; Aoba, T. Calcif. Tissue Int., 1993, 53, [110] Matsuno, T.; Nakamura, T.; Kuremoto, K.; Notazawa, S.; Nakahara, T.;
249. Hashimoto, Y.; Satoh, T.; Shimizu, Y. Dent. Mater. J., 2006, 25, 138.
[92] Nelson, D. G.; Barry, J. C. Anat. Rec., 1989, 224, 265.
[93] Anderson, H. C. J. Cell Biol., 1967, 35, 81.

Received: September 15, 2007 Revised: November 21, 2007 Accepted: November 25, 2007