Bone Tissue Regeneration by Collagen Scaffolds with Different Calcium Phosphate Coatings: Amorphous Calcium Phosphate and Low-Crystalline Apatite

Surface-mineralized collagen sponges have attracted much attention as scaffolds for bone tissue engineering. Recently, we developed amorphous calcium phosphate (ACP) and low-crystalline apatite coating processes on collagen sponges. In the present study, we applied these coating processes to granular collagen sponges (referred to as Col) to compare the bone tissue regeneration capabilities of ACP-coated and apatite-coated Col (referred to as Col-ACP and Col-Ap, respectively) using a rat cranial bone defect model. According to micro-CT and histological analyses, Col-Ap enhanced bone tissue regeneration compared to Col, whereas Col-ACP did not. These results not only demonstrated the superior bone tissue regeneration capability of Col-Ap, but also indicated limitations of the in vitro simulated body fluid (SBF) test used in our previous study. Despite the apatite-forming ability of Col-ACP in SBF, it was ineffective in improving bone tissue regeneration in vivo, unlike Col-Ap, most likely due to the quick resorption of the ACP coating in the defect site. The present results clarified the importance of the coating stability in vivo and revealed that the low-crystalline apatite coating was more beneficial than the ACP coating in the fabrication of surface-mineralized collagen sponges for use as bone tissue engineering scaffolds.


Introduction
Biodegradable scaffolds with a three-dimensional (3D) porous structure play an important role in bone tissue engineering. They support cell adhesion, growth, and osteogenic differentiation; facilitate the formation of blood vessels; and finally, resorb while being replaced with the newly formed bone. In human bone, collagen fibrils mineralized with nonstoichiometric low-crystalline apatite constitute the basic building blocks [1]. Hence, biocompatible and bioresorbable collagen sponges coated with low-crystalline apatite have attracted attention as scaffolds for bone tissue engineering owing to their similarity to human bone. Several reports have shown that such apatite-coated collagen sponges enhance bone tissue regeneration compared to bare sponges without coating [2][3][4]. However, apatite is formed via the formation of thermodynamically unstable precursor phases under physiological conditions [5][6][7]. In bone biomineralization, a non-crystalline phase of calcium phosphate is concentrated within the intracellular vesicles of osteoblasts [8,9], and later, infiltrates the collagen fibrils where it crystallizes to form low-crystalline apatite [10,11]. Because of the critical role of amorphous calcium phosphate (ACP) in biomineralization, ACP-coated collagen sponges have also emerged as potential scaffolds for bone tissue engineering [12,13]. However, there have been few comparative analyses of apatite-and ACP-coated collagen sponges for use as bone tissue engineering scaffolds.
Recently, we fabricated two types of mineralized porous collagen scaffolds for bone tissue engineering: collagen rectangular blocks coated with low-crystalline apatite [14] and collagen cubic granules coated with ACP nanoparticles [15]. The former scaffold with an apatite coating enhances the adhesion of pre-osteoblastic MC3T3-E1 cells in vitro and enhances cell-ingrowth within its porous body in rat subcutaneous tissue compared to an uncoated scaffold [14]. Moreover, the apatite-coated scaffold significantly promotes cranial bone formation in a rat skull defect model and alveolar bone regeneration in a dog premolar furcation defect model compared to the uncoated scaffold [16]. The latter scaffold with an ACP coating forms an apatite layer on its surface in a simulated body fluid (SBF) test [17], suggesting its apatite-forming ability in vivo and subsequent osteoconductive abilities [15]; however, its bone tissue regeneration capability remains to be elucidated.
Here, we compared the effects of ACP and apatite coatings on a porous collagen scaffold for bone tissue regeneration. In the previous studies described above, the dimensions of the two mineralized collagen scaffolds were different: cubic granules (1 mm × 1 mm × 1 mm) were used for ACP coating [15], whereas larger rectangular blocks (6 mm × 6 mm × 3 mm) were used for apatite coating [14,16]. In the present study, to achieve our aim, both ACP and apatite coatings were prepared on collagen scaffolds of the same size, i.e., cubic granules of a porous collagen sponge (referred to as Col). The ACP-coated and apatite-coated Col (referred to as Col-ACP and Col-Ap, respectively) and uncoated Col as a control were implanted in rat cranial bone defects to compare their bone tissue regeneration capabilities and their potential as bone tissue engineering scaffolds.

Preparation of Scaffolds
The three types of scaffold (Col, Col-ACP, and Col-Ap) were prepared. Cubic granules of Col with a dimension of 1 mm × 1 mm × 1 mm were prepared by cutting a porous sponge sheet of calf atelocollagen (Terudermis, Olympus Terumo Biomaterials Corp., Tokyo, Japan). Col-ACP was fabricated from Col by a temperature-controlled ACP coating process following our previously reported method [15]. Col-Ap was fabricated from Col by a plasma-and precursor-assisted biomimetic process using the same method as described earlier [14], except that 4 mg granules of Col were treated here, instead of three rectangular collagen blocks with dimensions of 6 mm × 6 mm × 3 mm.

Physicochemical Analyses
The scaffolds prepared in the preceding section were analyzed using a field emission scanning electron microscope (SEM; S-4800, Hitachi High-Tech Corporation, Tokyo, Japan) equipped with an energy dispersive X-ray (EDX) spectrometer (EMAX x-act, HORIBA, Ltd., Kyoto, Japan), an X-ray diffraction (XRD) instrument (M18X, MacScience, Tokyo, Japan) with CuKα radiation (λ = 0.154178 nm), and a Fourier transform infrared (FT-IR) spectrometer (FT/IR-4700, JASCO Corporation, Japan) equipped with an attenuated total reflection (ATR) accessory and a monolithic diamond crystal. Prior to the SEM and EDX analyses, the scaffolds were sputter-coated with gold. The XRD measurements were performed at 40 kV and 200 mA with the thin-film mode at an incidence angle of 1 • , a 2θ step width of 0.05 • , and a counting time of 6 s per step. Cross-sectional ultrathin specimens were prepared from Col-ACP by a conventional resin embedding method and analyzed using an analytical transmission electron microscope (TEM; Tecnai Osiris, FEI, Hillsboro, OR, USA) operated at 200 kV. The crystalline structure of the specimens was examined by selected area electron diffraction (SAED) analyses.
At two weeks post-surgery, rats were euthanized with an overdose of sodium pentobarbital (2.0 mL/kg, Somnopentyl; Kyoritsu Seiyaku Corp., Tokyo, Japan). The cranial bones were extracted and assessed by an X-ray micro-computed tomography (micro-CT) scanner (Latheta LCT-200, Hitachi, Ltd., Tokyo, Japan). From the captured micro-CT images, the radiolucent (non-calcified) area of the cranial bone and the volume of the newly formed bone were measured by image analysis software (ImageJ 1.41; National Institutes of Health, Bethesda, MD, USA). In histological analyses, the extracted cranial bones were fixed in 10% buffered formalin, decalcified with Plank-Rychlo's solution, and embedded in paraffin. After thin slicing, the tissue sections were stained with Masson's trichrome and observed using a light microscope (NanoZoomer S210 C13239-01, Hamamatsu Photonics K.K., Hamamatsu, Japan). The area and height of the newly formed bone were measured by image analysis software (ImageJ 1.41). Statistical differences were analyzed by one-way ANOVA with Tukey's HSD post hoc test (SPSS 11.0, IBM Corporation, Armonk, NY, USA), and differences with p < 0.05 were considered statistically significant.

Results and Discussion
The physicochemical analyses revealed successful coating of ACP and low-crystalline apatite on Col-ACP and Col-Ap, respectively. As shown in the SEM images (Figure 1a), spherical nanoparticles and a nanostructured layer appeared on the Col-ACP and Col-Ap surfaces, respectively. Both the spherical nanoparticles and nanostructured layer consisted of calcium phosphate, as indicated by the Ca and P peaks in the EDX spectra of Col-ACP and Col-Ap (Figure 1b). In their XRD patterns, no diffraction peaks were detected for Col-ACP, whereas peaks ascribed to low-crystalline apatite were detected for Col-Ap (Figure 2a). The amorphous nature of the calcium phosphate nanoparticles on Col-ACP was evident from the SAED pattern (Figure 2b). In the FT-IR spectra, both Col-ACP and Col-Ap showed a peak at~1025 cm −1 , representing a phosphate group which was absent in Col (Figure 2c). These analytical results are consistent with our previous reports [14,15] and proved that the plasma-and precursor-assisted biomimetic process, which was developed for apatite coating on rectangular collagen blocks [14], was also applicable to the smaller granules of Col.  Col-Ap significantly promoted bone regeneration in rat cranial bone defects compared to Col, although Col-ACP did not. Figure 3a shows representative micro-CT images of the cranial bones implanted with Col, Col-ACP, and Col-Ap. In the Col group, a nearly circular radiolucent (black) area (recognized as a non-calcified area) was clearly observed, indicating that new bone formation hardly occurred in the defect. In the Col-ACP group, a slightly radiopaque (dark gray) area partly appeared in the bone defect, implying its partial repair by the newly formed bone tissue. In the Col-Ap group, a brighter (light gray) radiopaque area spread widely over the bone defect and the residual radiolucent (black) area was very small, suggesting that the bone defect was almost fully repaired by the newly formed bone tissue. The quantitative analysis showed that the radiolucent area in the Col-Ap group was significantly smaller than that in the Col and Col-ACP groups (Figure 3b). Similarly, the volume of the new bone in the Col-Ap group was significantly higher compared to that in the Col and Col-ACP groups (Figure 3c). There was no significant difference in the radiolucent area between the Col and Col-ACP groups (p > 0.05). These micro-CT results were further supported by the histological observations. In the histological images (Figure 4a,b), the newly formed bone tissue (denoted by NB) was apparent in all groups. Around the newly formed bone tissue, the residual scaffolds (denoted by RS) along with fibroblast-like and macrophage-like giant cells were observed. The area of the newly formed bone tissue in the Col-Ap group was apparently larger than that in the Col and Col-ACP groups, suggesting enhanced bone regeneration by Col-Ap. However, the area of the residual scaffold in the Col-Ap group was apparently smaller than that in the Col and Col-ACP groups, indicating the faster resorption of Col-Ap. Severe inflammatory reactions were not noticed in any group. The histomorphometric analysis clarified that the area and height of the newly formed bone tissue in the Col-Ap group were significantly greater compared to those in the Col and Col-ACP groups (Figure 4c,d). Again, there was no significant difference between the Col and Col-ACP groups (p > 0.05). These results reconfirmed the beneficial effect of a low-crystalline apatite coating on Col for bone tissue regeneration. An osteogenic effect by a low-crystalline apatite coating has been reported for similar collagen scaffolds [2][3][4], including for the larger-sized rectangular blocks [16]. Despite its apatite-forming ability in SBF [15], Col-ACP induced different biological responses from Col-Ap; it did not enhance bone regeneration compared to Col in the rat cranial bone defect model (Figures 3 and 4). That is, the present ACP coating on Col had no beneficial effect on bone tissue regeneration apart from the apatite coating, although there are several reports demonstrating the osteogenic effect of ACP [18,19]. Inferior osteogenic effects of ACP coatings compared to apatite coatings have been reported for other coating processes [20,21]. We consider that the ACP nanoparticles on Col-ACP were insufficient in their amount and stability, thereby resorbing in the bone defect before inducing apatite layer formation. A similar phenomenon has been reported for some calcium salts with relatively high solubility, such as calcium sulphate hemihydrate [22] and dicalcium phosphate dihydrate [23]; these materials show poor osteogenic properties, most likely due to fast resorption in vivo (bone defects are repaired only partially by immature bone), despite their apatite-forming ability in SBF [24]. The most probable cause for the discrepancy between the SBF test result [15] and the present in vivo result is the acidified environment resulting from postoperative inflammation reactions [25] and/or acid secretion by osteoclasts [26] in the vicinity of the scaffold. It is known that ACP transforms into apatite under a normal physiological environment, because body fluid (SBF as well) is supersaturated with respect to ACP, apatite, and several other calcium phosphate phases, among which apatite is the most stable and insoluble phase [27]. However, a decrease in environmental pH from the physiological pH (~7.4) significantly increases the solubility of these calcium phosphates [28], which may lead to the dissolution and resorption of the relatively unstable ACP phase. The effects of body fluid circulation might also be involved in the dissolution and resorption of the ACP nanoparticles by facilitating the diffusion of ions and clusters in the amorphous-to-crystalline transformation process through dissolution-reprecipitation reactions [29]. Note that the present results were obtained from rat cranial bone defects. The rate of bone regeneration varies with animal species and the region of bone defects. The flat cranial bone defects are regenerated through the intramembranous ossification process, which is different from longer bone regeneration through the endochondral ossification process. There are reports available on the faster regeneration of longer bone defects compared to that of flat cranial bone defects [30,31]. Hence, these scaffolds might induce different biological responses in different bone defects of different animal species, which are issues for future studies.

Conclusions
The low-crystalline apatite coating on Col (Col-Ap) was more effective in improving bone tissue regeneration in rat cranial bone defects than the ACP coating (Col-ACP). The stability of the mineralized surface coating is an important factor in determining the bone tissue regeneration capability of the collagen scaffold.

Informed Consent Statement: Not applicable.
Data Availability Statement: All data from this study are included within the article.