A Morphological Analysis on the Osteocytic Lacunar Canalicular System in Bone Surrounding Dental Implants

Abstract

Osseointegration is the most preferable interface of dental implants and newly formed bone. However, the cavity preparation for dental implants often gives rise to empty lacunae or pyknotic osteocytes in bone surrounding the dental implant. This study aimed to examine the chronological alternation of osteocytes in the bone surrounding the titanium implants using a rat model. The distribution of the osteocytic lacunar canalicular system (OLCS) in bone around the titanium implants was examined by silver impregnation according to Bodian's staining. We also performed double staining for alkaline phosphatase (ALP) and tartrate-resistant acid phosphatase (TRAP), as well as immunohistochemistry for fibroblast growth factor (FGF) 23—a regulator for the serum concentration of phosphorus—and sclerostin, which has been shown to inhibit osteoblastic activities. Newly formed bone and the injured bone at the early stage exhibited an irregularly distributed OLCS and a few osteocytes positive for sclerostin or FGF23, therefore indicating immature bone. Osteocytes in the surrounding bone from Day 20 to Month 2 came to reveal an intense immunoreactivity for sclerostin. Later on, the physiological bone remodeling gradually replaced such immature bone into a compact profile bearing a regularly arranged OLCS. As the bone was remodeled, FGF23 immunoreactivity became more intense, but sclerostin became less so in the well-arranged OLCS. In summary, it seems likely that OLCS in the bone surrounding the dental implants is damaged by cavity formation, but later gradually recovers as bone remodeling takes place, ultimately inducing mature bone. Anat Rec,, 2011. © 2011 Wiley-Liss, Inc.

The cavity formation for dental implants can sometimes injure osteocytes, giving rise to pyknosis or cell death, that is, empty lacunae in the peripheral bone of the implant (Fujii et al.,1998; Futami et al.,2000; Haga et al.,2009). Damaged bone without osteocytes remains even after osseointegration—the optical interface of the implant and bone—takes place (Fujii et al.,1998; Futami et al.,2000; Haga et al.,2009). Thus, special attention should be paid to the bone around dental implants to avoid unforeseen clinical failures after dental implantation. Recently, using an implant model of rat maxilla, we have noted physiological bone remodeling—a gradual replacement of damaged bone to a compact one—after the achievement of osseointegration (Haga et al.,2009). As a consequence, bone mineralization—a parameter for bone quality—appeared to improve as well. It would seem that the bone regeneration associated with the dental implant depends on the biological activities of the recipient bone.

Osteocytes, the most abundant cells in mature bone, extend long cytoplasmic processes and enable cell-to-cell interaction: a cellular network, through which embedded osteocytes and osteoblasts can communicate with each other, referred to as the osteocytic lacunar canalicular system (OLCS) (Aarden et al.,1994; Burger and Klein-Nulend,1999). The osteocytic function is associated with bone remodeling and osteocytes are accepted as transducers of the mechanical strain that is translated into biochemical signals affecting this OLCS (Lanyon,1993; Aarden et al.,1994; Burger et al.,1995; Klein-Nulend et al.,1995; Burger and Klein-Nulend,1999; Knothe et al.,2004). Thus, the distribution of the OLCS was shown to reflect osteocytic function (Hirose et al.,2007). According to osteocyte viability in the maintenance of bone tissue (Noble and Reeve,2000; Knothe et al.,2004), the OLCS appears to be a crucial component for successful implantation.

Dentin matrix protein (DMP) 1, fibroblast growth factor (FGF) 23, and sclerostin/SOST are useful markers for osteocyte (Winkler et al.,2003; Atkins et al.,2009; Ubaidus et al.,2009; Kramer et al.,2010). For example, DMP1, a bone matrix protein specifically expressed in osteocytes—including their processes—plays a role in bone mineral homeostasis due to its high calcium ion-binding capacity (Toyosawa et al.,2001). Sclerostin, a product of SOST gene, suppresses the activity of osteoblasts and viability of osteoblasts and osteocytes; its expression is restricted to osteocytes in adult bone (van Bezooijen et al.,2004; Lin et al.,2009). Investigations into the biological functions of FGF23 have broadened our understanding of the systemic regulation of phosphate homeostasis as well as of the maintenance of proper mineralization in the bone matrix (Liu and Quarles,2007).

The histological alternation of osteocytes around implants has not been fully investigated despite several studies on orthopedic implants by DMP1 and SOST (Xie et al.,2004; Atkins et al.,2009). Few previous studies about the drawbacks of dental implants have focused on osteocytes around the implant, although some investigations noted injury to the surrounding bone and osteocytes from the heat of drilling (Listgarten,1996; Futami et al.,2000; Marco et al.,2005).

In this study, we aimed to verify the distribution of the OLCS and the biological function of osteocytes around the implant by investigating immunohistochemistry for DMP1, sclerostin, and FGF23. Evaluation of the chronological changes in the surrounding bone around an implant was carried out by Bodian's staining (Schoen,1991; Hirose et al.,2007; Ubaidus et al.,2009), Azan staining, and a double staining for the enzymatic histochemical and immunohistochemical detection of tartrate-resistant acid phosphatase (TRAP) activity and alkaline phosphatase (ALP) activity, respectively.

MATERIALS AND METHODS

Animals and Experiment Procedures

All animal experiments in this study were performed under the Niigata University Guidelines for Animal Experimentation (approval number 31) and the Guidelines of the Nippon Dental University for the Care and Use of Laboratory Animals (approval number 94).

Fifty 4-week-old male Wistar rats were used in this experimental study. The implantation procedure used in this study has been reported in our previous study (Fujii et al.,1998; Haga et al.,2009). Briefly, under anesthesia with an intraperitoneal injection of 8% chloral hydrate (4 mL/kg body weight), the upper first molars on both sides were extracted with forceps. At 1 month after tooth extraction (8 weeks of age), a full thickness flap was elevated at the recipient site under the same anesthesia. The bone cavities for implantation were formed by drilling with a slow speed (500 rpm) dental hand piece equipped with an engine reamer, followed by a Peeso reamer (1.15 mm in diameter). Profuse irrigation with sterilized physiological saline was maintained throughout the drilling. The custom-made titanium implants had a machined surface and a bullet shaped base (1.13 mm in diameter; KS-50, Kobelco, Kobe, Japan). The pure titanium implants were inserted bilaterally into the surgically prepared bone cavities so that their tops were situated 0.5 mm from the cortical bone surface. The flaps were repositioned and sutured with silk sutures to cover the implants (i.e., submerged method). After implantation, the animals were housed with free access to water and provided a powder diet. No antibiotics were given to the operated rats.

Histological Procedure

At Days 5, 10, and 20 after implantation, and Months 1, 1.5, 2, 2.5, 3, 3.5, 4–9, and 12 after implantation, the animals were anesthetized deeply in the same manner as described above and transcardially perfused with a fixative containing 4% paraformaldehyde (pH 7.4). Following fixation, the maxillae including implants were immersed in the same fixative for an additional 24 hr and then decalcified with 10% EDTA solution for 4 weeks at 4°C. After demineralization, the installed implants were removed with a pincette as it has been reported that this technique does not cause the destruction of the tissue specimens (Fujii et al.,1998). Specimens without the implants were dehydrated through an ascending ethanol series and embedded in paraffin. Serial paraffin sections were sagittaly cut at a 4.5 μm thickness and stained either with hematoxylin and eosin (H–E) or with Azan staining for histologic observation.

The Staining for Silver Pigmentation for Visualization of OLCS

We used a modified staining protocol based on Bodian's protargol-S procedure. The dewaxed sections were soaked in a 1% Protargol-S solution, diluted in borax-boric acid (pH 7.4; Merck KGaA, Darmstadt, Germany) for 30 hr at 37°C. After rinsing with distilled water, the reaction was enhanced by an aqueous solution containing 0.2% hydroquinone (Wako Pure Chemical Industries, Osaka, Japan), 0.2% citric acid (Wako), and 0.7% nitric silver (Wako). After an additional rinsing, the sections were reduced for 5 min with an aqueous solution of 2.5% anhydrous sodium sulfate (Junsei Chemical, Tokyo, Japan), 0.5% potassium bromide (Wako), and 0.5% amidol diaminophenol dihydrochloride (Wako). They were then treated with 1% gold chloride (Wako) and subsequently reduced by 2% oxalic acid amidol (Wako) until black-stained osteocytic canaliculi could be seen. After rinsing with distilled water, the sections were fixed in 5% sodium thiosulfate (Wako) and stained with H–E.

Detection of ALP and TRAP

Several paraffin sections were used for double staining with a rabbit antisera against tissue nonspecific alkaline phosphatase (ALP; Oda et al.,1999) immunohistochemistry and for tartrate-resistant acid phosphatase (TRAP) enzyme histochemistry as previously reported (Haga et al.,2009). After the inhibition of endogenous peroxidase, deparaffinized sections were pre-incubated with 1% bovine serum albumin in phosphate-buffered saline (BSA-PBS) for 30 min at room temperature. Antisera against ALP were applied to the sections at a dilution of 1:200 overnight at 4°C. The sections were then incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG (dilution 1:100; Chemicon International, Temecula, CA). Immunoreaction sites were made visible with 0.04% 3-3′-diaminobenzidine tetrahydrochloride (DAB) and 0.002% H₂O₂ in a 0.05 M Tris-HCl buffer (pH 7.6). Immunostained sections were further processed for the histochemical demonstration of TRAP by the Azo-dye method. The incubation medium comprised 30 mg fast red violet LB salt (Sigma, St. Louis, MO), 5 mg naphthol AS-BI phosphate (Sigma), and 50 mM L (+) sodium tartrate (0.37 g; Sigma) in a 0.1 M acetate buffer (pH 5.4), and the incubation was carried out for 45 min at 60°C. The sections were faintly counterstained with methyl green.

In addition, the number of the osteoblasts and osteoclasts was counted on ALP/TRAP sections under a light microscope (AxioImager M; Carl Zeiss, Oberkochen, Germany) at an original magnification of ×200 at Days 5, 10, and 20, and Months 1, 2, 2.5, 3, 3.5, and 6–9. The observation area was set as a boxed area 0.5 × 1 mm in the lateral bone around the implant. All values were presented as means ± standard deviation. Data were subjected to one-way ANOVA with PASW software (SPSS, Chicago, IL). A P value of <0.05 was considered a significant difference.

Immunohistochemistry for DMP1, Sclerostin, and FGF23

Paraffin sections were processed for immunocytochemistry for DMP1, sclerostin, and FGF23. After pre-incubation with 1% BSA-PBS for 30 min at room temperature, sections were incubated with a rabbit antibody against DMP1 (dilution 1:600; Takara Bio, Otsu, Japan) and goat anti-sclerostin (1:30; R&D systems, Minneapolis, MN) overnight at 4°C. Sections for DMP1 and sclerostin immunohistochemistry were then reacted with the incubation with horseradish peroxidase-conjugated goat anti-rabbit IgG (1:100; Chemicon) and horseradish peroxidase-conjugated rabbit anti-goat IgG (1:100; Chemicon), respectively, followed with the development of a DAB solution. For the detection of FGF23, dewaxed paraffin sections were treated with rat anti-FGF23 (1:25; R&D systems) for 2 hr. The sections were subsequently incubated in ALPase-conjugated goat anti-rat IgG (1:100; Chemicon). For the visualization of the ALPase-conjugated complex, the treated sections were immersed in an aqueous solution of 6 mg of naphthol AS-BI phosphate (Sigma) and 50 mg of red violet LB salt (Sigma) diluted in 60 mL of a 0.1 M Tris-HCl buffer (pH 9.0) for 10 min at 37°C. All sections were faintly counterstained with methyl green. The specificity of the immunoreactions was checked by the following two negative control experiments; the omission of the primary antibodies and the replacement of the primary antibodies by normal sera. These control experiments did not show any specific immunoreaction (data not shown).

Detection of Apoptosis Reaction using TdT-Mediated dUTP-Biothin Nick-End Labeling

Paraffin sections were labeled with terminal deoxynucleotidyl transferase (TdT) using a TACS 2 TdT-Blue Label In Situ Apoptosis Detection Kit (Trevigen, Gaithersburg, MD) according to the manufacturer's protocol.

RESULTS

Rats receiving the implantation gained body weight steadily during the postoperative periods up to the day of fixation without any remarkable changes in condition including feeding habits, as shown in our previous reports (Fujii et al.,1998; Haga et al.,2009).

Tissue Responses to Implantation—Distribution of the OLCS

At Day 5, collagen bundles in the connective tissue ran circumferentially around the implant (Fig. 1a). Abundant empty osteocytic lacunae either with degenerated osteocytes or without osteocytes were recognizable in the pre-existing bone near the implant cavity, with a comparatively irregular arrangement of the OLCS (Figs. 1a, 2a). Some osteoclasts existed in the bone around the implant with many blood capillaries, particularly in the bone-implant surface (Figs. 1a, 2a). Osteocytic canaliculi ran regularly in the pre-existing bone but not in the damaged bone (Fig. 2a). TdT-Mediated dUTP-Biothin Nick-End Labeling (TUNEL) positivity was found in osteocytes in and around the damaged bone, including the bone-implant surface (Fig. 3a).

Figure 1. Photomicrographs showing histological changes in the surrounding bone around an implant at Days 5 (a), 10 (b), and 20 (c), and Months 1 (d), 2 (e), and 7 (f). Paraffin sections with Azan staining. The bone with empty osteocytic lacunae (arrowheads) demonstrates the time-specific diminution (a–e); no bone with empty osteocytic lacunae is discernable at Month 7 (after Month 3) (f). The connective tissue is observed in the bone-implant surface at Days 5, 10, and 20 (a–c). At Days 10 and 20, and Month 1, azocalmine-stained structures (arrows) are recognizable in the matrix of the newly formed bone, indicating that this is woven bone (b–d). At Month 1(d), the woven bone appears to decrease in volume as compared with those at the previous stages and occupies in only a slight part of the surrounding bone at Month 2 (e). After Month 3, the newly formed bone appears to proceed with corticalization to become thick (f). The implant cavity is on the right side of all photomicrographs (a–f). Scale bars = 40 μm. In set in (a and b): higher magnification of the boxed area. Scale bars = 10 μm in inset.

Figure 2. Staining for silver pigmentation for visualization of the OLCS of the surrounding bone around the implant at Days 5 (a), 10 (b), and 20 (c), and Months 1 (d), 2 (e), and 7 (f). Note empty osteocytic lacunae (arrowheads) (a–e). Black-stained osteocytic canaliculi showing a nonarranged network in the damaged bone and some osteoclasts are found at Day 5 (a). Some osteocytes are distributed regularly in pre-existing bone but others irregularly in damaged pre-existing bone and woven bone at Days 10 and 20 (b and c). Though a random pattern of the OLCS is distributed in a major part of the bone around the implant at Month 1 (d), it appears to decrease in volume at Month 2 (e). Most of the bone around the implant at Month 7 (after Month 3) displays regular osteocytic canaliculi (f). The implant cavity is on the right side of all photomicrographs (a–f). Scale bars = 20 μm.

Figure 3. Apoptosis detection on the bone around implants at Days 5 (a) and 10 (b). Some apoptotic reactions (arrows) exist not only in the bone-implant surface but also in the inside bone at Day 5 (a); on the other hand, few apoptotic cells (arrow) react in the bone around the implant at Day 10 (b). The implant cavity is on the right side of both photomicrographs (a and b). Scale bars = 60 μm.

Numerous regenerated collagen bundles existed around the implant cavity at Day 10 (Figs. 1b, 2b) and remained at Day 20 (Figs. 1c, 2c). The newly formed bone showed a woven profile, and the formation of this bone proceeded towards the injured bone with empty osteocytic lacunae, resulting in a decrease in the volume of damaged bone (Figs. 1b,c, 2b,c). Round-shaped osteocytes were embedded in a disorderly fashion in new bone (Fig. 2b,c). Only a few osteocytes exhibited TUNEL positivity (Fig. 3b). The number of TUNEL-positive osteocytes did not change thereafter (data not shown).

At Month 1, our experimental model showed osseointegration in all the surface areas of the dental implants, as has been previously reported (Fujii et al.,1998; Haga et al.,2009) (Figs. 1d, 2d). In addition, the newly formed bone exhibited partial features of woven bone with an irregular arrangement of the OLCS (Figs. 1d, 2d). The bone around the implants was gradually replaced with the compact bone with a regular arrangement by Month 3 (Figs. 1e, 2e). Slender and flattened osteocytes were distributed in parallel after Month 3 (Figs. 1f, 2f). No apparent histological change was found from Months 3 to 12.

Detection of ALP and TRAP

Many ALP-positive osteoblasts and TRAP-positive osteoclasts existed on the bone surface and in the bone marrow at Day 5 (Fig. 4a). The number of TRAP-positive cells appeared to decline after Day 10 (Fig. 4b–g), and the number of ALP-positive cells appeared to increase slightly at Days 10 and 20 (Fig. 4b,c,g). After Month 1, ALP-positive osteoblasts were also decreased in number (Fig. 4d–g). In addition, many ALP-immunopositive osteoblasts were localized close to the area occupied by the TRAP-reactive osteoclasts, suggesting the occurrence of a coupling phenomenon, which appeared to be intrinsic for active remodeling at Day 20 and Month 1 (Fig. 4c,d). Although osteoblasts with ALP immunoreactivity had earlier exhibited plump profiles (Fig. 4a–d), these cells became more spindle or flatter in shape in the newly formed bone at Month 2 than at previous stages (Fig. 4e). At (data not shown) and after (Fig. 4f) Month 3, ALP-positive osteoblasts and TRAP-reactive osteoclasts were rare in the lateral bone wall (Fig. 4f); this was found statistically significant by the quantitative analysis (Fig. 4g).

Figure 4. Photomicrographs of the localization of ALP/TRAP at Days 5 (a), 10 (b), and 20 (c), and Months 1 (d), 2 (e), and 7 (f), and a graph representing changes in the number of osteoblasts and osteoclasts (g). Double labeling of ALP by immunohistochemistry (arrowheads) and TRAP by enzymatic histochemistry (arrows). Numerous ALP-immunopositive cuboidal osteoblasts and TRAP-reactive osteoclasts localize along the implant cavity as well as in the bone marrow at Day 5 (a). TRAP-reactive osteoclasts appear to decline in cell volume after Day 5 (b–g). On the other hand, osteoblasts increase in number at Days 10 and 20 (b, c, and g) but thereafter decrease (d–g). Although osteoblasts with ALP immunoreactions exhibit cuboidal profiles at Day 5 to Month 1 (a–d), these cells become spindle shaped or flat in the newly formed bone, indicating that they are in an inactivate state after then (e and f). The graph shows that the number of osteoblasts and osteoclasts falls by Month 3, and levels off after Month 3 (g). The implant cavity is on the right side of all photomicrographs (a–f). Scale bars = 50 μm.

Immunohistochemical Detection of DMP1 and Sclerostin

The patterns of distribution for DMP1 positivity did not change but were similar throughout the entire experimental periods (Fig. 5a–c,g–i). Sclerostin immunopositivity was hardly seen in DMP1-positive osteocytes at Days 5 and 10 (Fig. 5d,e compared with Fig. 5a,b). The immunoreaction of sclerostin appeared in many osteocytes in the surrounding bone which showed a similar intensity of DMP1 immunoreactivity at Day 20 (Fig. 5f); sclerostin immunolocalization was discernible up to Month 2 (Fig. 5j). Sclerostin positivity seemed to be less intense than that of DMP1 at Month 2.5 (Fig. 5k compared with Fig. 5h). The intensity of sclerostin immunopositivity was attenuated but persisted at a plateau after Month 3 (Fig. 5l).

Figure 5. Photomicrographs of the localization of DMP1 (a–c and g–i) and sclerostin (d–f and j–l) immunoreactivity at Days 5 (a and d), 10 (b and e), and 20 (c and f), and Months 2 (g and j), 2.5 (h and k), and 7 (i and l). DMP1 immunopositivity is seen in the osteocytic lacunae and canaliculi in the whole region of the surrounding bone throughout all observation periods (a–c and g–i). Unlike the findings for DMP1 (a and b), osteocytes show little reaction for sclerostin in the pre-existing bone at Days 5 and 10 (d and e). Osteocytes in the surrounding bone from Day 20 to Month 2 reveal an intense immunoreactivity for sclerostin (f and j). Sclerostin-immunoreactive osteocytes thereafter decline in immunoreactivity (k and l). The implant cavity is on the right side of all photomicrographs (a–l). Scale bars = 35μm.

Immunohistochemical Detection of FGF23

At Day 5, FGF23-reactive osteocytes seemed to decrease in the region of the injured bone (Fig. 6a). FGF23-reactive cells were hardly seen in the new immature bone or in the damaged bone at Days 10 and 20, and at Month 1 (Fig. 6b–d). However, the newly formed bone with a comparatively regular arrangement of the OLCS revealed FGF23-immunoreactive osteocytes (Fig. 6e), and more intense FGF23 immunoreactivity was exhibited in the bone which tended to show a more regular OLCS at (data not shown) and after (Fig. 6f) Month 3.

Figure 6. Photomicrographs of the localization of FGF23 immunoreactivity at Days 5 (a), 10 (b), and 20 (c), and Months 1 (d), 2 (e), and 7 (f). FGF23 shows hardly any positivity in osteocytes in the injured bone near the implant at Day 5 (a). When observed at Days 10, 20, and Month 1, few FGF23 positive osteocytes are seen in the new woven bone added to the damaged bone (b–d). In contrast, as the areas of these bone gradually decrease, more osteocytes exhibit intense FGF immunoreactivity (arrows) where the OLCS is regularly distributed in newly formed bone (e and f). The implant cavity is on the right side of all photomicrographs (a–f). Scale bars = 50 μm.

DISCUSSION

This study successfully demonstrated the histological alteration of osteocytes in the surrounding bone around implants. To our knowledge, this is the first report to reveal chronological alterations in the distribution of the OLCS as well as the immunolocalization of DMP1, sclerostin, and FGF23 in the bone around the implant. The present study confirmed our previous finding (Haga et al.,2009) that the newly formed bone changed from woven bone to compact bone by continuous bone remodeling, by employing a double staining with ALP immunohistochemistry and TRAP histochemistry, and by Azan staining. In addition, we concluded that a regular OLCS would reflect enhanced bone maturation. Woven bone was formed rapidly after implantation, where osteocytes were embedded irregularly. On the other hand, the bone was gradually remodeled, giving rise to a well-arranged OLCS. We agreed with a previous study (Ubaidus et al.,2009) which postulated that the speed of bone deposition was the most important factor influencing the regularity of the OLCS. Clinical failure in dental implantation often occurs at relatively early stages after the surgery (Friberg et al.,1991; Salonen et al.,1993; Tolstunov,2007; Manor et al.,2009). It is easy to imagine the relation between early failure and early bone conditions, that is, the woven bone of an irregular OLCS.

Another notable finding is that FGF23 positivity was abundant in osteocytes at the maturation stage of bone healing after the dental implantation, where the regularly distributed OLCS have been established after bone remodeling with reductions of ALP and TRAP reactivities. Assuming that FGF23 is produced by osteocytes in organized systems connected with the proper mineral metabolism (Ubaidus et al.,2009), this compact bone around the implant should contain a functional and appropriate microenvironment. In addition, the reduction of ALP immunoreaction after the peak of sclerostin reactivity suggests the suppressed activities of osteoblasts by osteocytes, controlling bone remodeling by sclerostin. Immunoreaction for DMP1, which is involved in homeostasis of calcification, was recognizable in osteocytes at all stages in this experiment. This study supports the idea that osteocytes serve as a local regulator of mineralization and bone remodeling (Lanyon,1993; Knothe et al.,2004).

In our TUNEL experiment, we noted one slight increase in the number of apoptotic osteocytes. Microdamage or osteocytic necrosis by drilling may make neighboring osteocytes apoptotic, accelerating resorption of the bone that contains injured osteocytes as shown by intense TRAP reactivity and subsequent bone remodeling both with intense ALP/TRAP reactivities. Molecular transport through osteocytic canaliculi could be efficiently performed from one osteocyte to other osteocytes and osteoblasts (Klein-Nulend et al.,1995; Knothe et al.,1998; Burger and Klein-Nulend,1999; Wang et al.,2005). With its high degree of complexity and organization, the OLCS seems to be disturbed by apoptosis (Hirose et al.,2007). Current findings are consistent with the function of osteocyte death as a trigger of bone resorption (Gu et al.,2005).

The present morphological data showed a close relationship between the arrangement and function of osteocytes in the bone around the dental implants, suggesting to us that a geometric constitution of the OLCS is an important factor for the functional maintenance of the bone around the implant. As described above, the bone remodeling negatively regulated by sclerostin-immunoreactive osteocytes may be involved in the spatial arrangement of the OLCS. On the other hand, the regularity of the OLCS appears to affect the synthesis of FGF23 in osteocytes. Qin et al. (2007) also reported the regulation of phosphate homeostasis by DMP1 through FGF23. Thus, osteocytes seem to play a key role in the intact healing process of the surrounding bone around implants. Taking this study together with our previous ones (Haga et al.,2009), it appears that the injured bone with empty osteocytic lacunae causes a delay in new mature bone formation, indicating a possible risk factor for initial fixation which is related to the prognosis and maintenance of the mineral homeostasis.

Acknowledgements

The authors cordially thank the staffs of the Division of Oral Anatomy, Niigata University Graduate School of Medical and Dental Sciences, and Department of Histology, The Nippon Dental University School of Life Dentistry at Niigata, for their encouragement throughout this study.