**3. Discussion**

The increasing global demand for fish has caused an increase in fish byproducts, such as fish scales, bones, and skins. Studying these byproducts as biomaterials can reduce the number of fish byproducts and increase their value [33]. Although not conducted in this study, it is necessary to study the selection of materials suitable for tissue engineering by comparing the characteristics of the collagen derived from terrestrial organisms and the collagen derived from marine organisms. In the present study, we isolated and characterized MC and CHA from *P. olivaceus* using pepsin hydrolysis and alkaline hydrolysis, respectively [34]. Numerous studies have reported about HA and collagen extracted and isolated from different fish species, such as *Oncorhynchus keta*, *Thunnus obesus,* and *Oreochromis* [35–37]. Unlike thermally extracted HA, HA extracted through alkaline hydrolysis retains its carbonate groups, because carbonate cannot be separated without heat treatment [36]. Furthermore, CHA was reported to have a similar chemical composition to that of HA present in natural bone tissue [38]. According to previous studies, the HA derived from fish bones shows improved biocompatibility and osteogenic differentiation activity, and enhanced potential for bone formation on scaffolds fabricated with HA. Thus, CHA is a potential biomaterial that can be employed with various synthetic or natural polymers to fabricate tissue-engineered bone substitutes for bone tissue regeneration applications [39].

XRD, FTIR spectroscopy, and EDS analyses of isolated CHA and commercially available HA were performed to compare and determine the crystallinity, chemical composition, and atomic percentage, respectively. The XRD analysis of CHA showed the same peak international standard as HA (ICDD 01-086-0740) at 002, 211, 112, 202, 310, 222, 213, 321, and 004 of Bragg's reflection in the standard [35]. In addition, the results of both FTIR spectroscopy and EDS clearly indicated that the isolated CHA has a carbonate group attached to the HA unit, as shown in Figures 2A and 3. In Figure 2A, a larger band value of CHA was confirmed at 1400–1500 cm<sup>−</sup>1, which appears to be because the carbonate group increased [40]. The relatively large value of these two bands (1400–1500 cm<sup>−</sup>1) is because CHA has a relatively higher carbonate group than HA. EDS analysis was performed to evaluate the presence of trace elements, such as Ca and P, belonging to the isolated CHA and commercial HA, and the results showed a Ca/P weight ratio of 1.96 and1.578 for CHA and HA, respectively (Figure 3), indicating that the extraction process (alkaline hydrolysis) of CHA has affected the Ca/P ratio. However, according to previous studies, the Ca/P ratios of both CHA and HA are not significantly different, and they were within the accepted range for hydroxyapatite [39].

Collagen was extensively used to fabricate various tissue regenerative substitutes, because it is one of the main components of the ECM and is an excellent biomaterial that provides exceptional biological and functional properties, without an associated inflammatory response or cytotoxicity [21]. In particular, MC is an alternative and attractive type of collagen over land animal-derived collagen for tissue engineering applications, since it does not have religious restrictions and is not associated with a risk of disease transmission to humans. Hence, we extracted MC from *P. olivaceus* and characterized it using UV/Vis spectra, SDS-PAGE, FTIR spectroscopy, and amino acid composition analysis. The UV/Vis spectra recorded a relatively low absorption value at approximately 280 nm compared with general proteins, because collagen contains a relatively lesser amount of tyrosine and phenylalanine than general proteins (Figure 1A). Moreover, the SDS-PAGE results showed two distinguishable bands at approximately 130 kDa, corresponding to the α1 and α2 chains, and two bands at 250 kDa and 310 kDa, corresponding to the larger β and γ chains, respectively, suggesting that extracted *P. olivaceus* skin collagen is type I collagen, which is in agreemen<sup>t</sup> with the findings of previous studies [21]. The FTIR spectra of the extracted MC showed five characteristic peaks at 3300 cm<sup>−</sup>1, 2972 cm<sup>−</sup>1, 1635 cm<sup>−</sup>1, 1549 cm<sup>−</sup>1, and 1239 cm<sup>−</sup>1, which correspond to amide A, amide B, amide I, amide II, and amide III, respectively, similar to that of commercial collagen and the findings of previous publications [41]. Type I collagen forms a triple-helix structure with 20 different amino acids and is stabilized by its high content of glycine repeated every three residues, proline, and hydroxyproline. Moreover, proline and hydroxyproline showed lower values in MC, which is in agreemen<sup>t</sup> with the findings of previous reports [42]. Overall, the results suggested that MC and CHA were successfully extracted and characterized from *P. olivaceus*.

According to previous reports, various synthetic and natural biocompatible materials were employed to fabricate 3D-printed scaffolds and demonstrated good bone tissue regeneration effects. Among them, many researchers have focused on improving biological activities including cell adhesion, proliferation, migration, and differentiation through surface modifications including surface coating with natural or chemically modified biocompatible materials [43]. In the present study, we fabricated a 3D-printed porous PCL scaffold reinforced with CHA and surface-coated with MC to enhance osteogenic differentiation. To evaluate the osteogenic activity of the fabricated scaffolds, MC3T3-E1 cell-seeded scaffolds were analyzed, using ALP assay and Alizarin Red S staining. According to the results, the CHA/MC/PCL scaffolds significantly enhanced the mineral deposition through differentiation of MC3T3-E1 pre-osteoblasts to osteoblasts compared with the pure PCL scaffold, indicating that MC and CHA have excellent osteogenic activities and excellent synergetic effects on bone tissue regeneration. Moreover, the bone tissue regeneration effects of the fabricated scaffolds were evaluated using an in vivo calvarial defect mouse model. Micro-CT and histological analysis indicated that the 10% CHA/MC/PCL- and 10%

HA/MC/PCL-treated groups had prominent bone tissue reconstruction effects, compared with the non-treatment and PCL-treated groups (Figure 6). Overall, these results sugges<sup>t</sup> that materials obtained from marine byproducts have considerable potential for bone tissue regeneration and are attractive alternatives for terrestrial or synthetic materials. Based on this study, the possibility was evaluated of MC- and CHA-derived byproducts from *P. olivaceus* used as a substitute material for bone tissue. Furthermore, we will proceed with research on bone mimic scaffold containing MC and CHA. 4.

### **4. Materials and Methods**
