*3.6. Immunofluorescence Microscopy (IF) Analysis*

Among the cells attached on the SLA surfaces, osteogenic cells needed to be identified because they are the key cells in bone formation. Accordingly, the attachment of osteogenic cells to the implant surface was confirmed through the use of osteocalcin—antibody targeted for rabbits in vivo. Immunofluorescence microscopy enabled the development of images of osteogenic cells on the SLA implants and the surrounding bone after 10 days, and consequently, confirmed the existence of osteogenic cells on the implant surface. While osteogenic cells were detected on the implant surface, the surrounding bone showed few osteogenic cells (Figure 3A).

**Figure 3.** (**A**) Immunofluorescence microscopy (IF) of the SLA implants and surrounding bone on day 10, including nucleus, marker, and colocalization. Osteogenic cells (red arrow) are attached to the SLA implant surface rather than to the surrounding bone, which showed few osteogenic cells. The magnification of the photographs is ×200. (**B**) Transmission electron microscopy (TEM) analysis of retrieved SLA implant at day 10. (**C**) Macrophages (blue arrow) and osteogenic cell (red arrow) can be seen in the SLA implants. (**D**) TEM analysis of surrounding bone on day 10. (**E**) In the surrounding bone, osteocytes (white arrow) were detected.

#### *3.7. Transmission Electron Microscopy (TEM) Analysis*

The samples were also scrutinized by TEM. The TEM images of the SLA implants revealed macrophages and osteogenic cells (Figure 3B,C), while in the surrounding bone, osteocytes were detected (Figure 3D,E).

#### **4. Discussion**

In the present study, we aimed to determine the cell activity that occurs during the bone-forming process. We targeted the challenges concerning the lack of actual visualization of bone formation, and put much effort into presenting data on the active process in a bony environment with an actual 3D implant structure rather than the flat Ti discs used in in vitro studies. Our experimental data showed that a positive cell reaction occurred on the SLA surfaces, whereas the turned surfaces lacked cell adhesion. Meanwhile, the surrounding bone of the turned surface implants exhibited active cellular events. This may be confirmation of contact osteogenesis on the SLA surfaces. Whilst distance osteogenesis appears to occur around the smooth turned surfaces, it is well known that rougher turned surfaces (that were not investigated in this paper) also display contact osteogenesis [19,22–24].

According to the IF results seen, confirmation of osteogenic cells on the roughened implant surfaces might be further evidence of contact osteogenesis. However, further investigation is required to determine why only few osteogenic cells were detected on the bone surface—which is considered to be the place for distance osteogenesis. In addition, although limitations exist, in that cell classification is difficult in FIB specimens, the results reveal further evidence of contact osteogenesis on the SLA surface. Investigations are needed to better understand the link between such a phenomenon and the higher clinical long-term survival rate of implants with the SLA surfaces (over 95%), compared to that of turned implants (81–91%) [25,26].

The cells on the Ti implant surfaces seemed to be able to read the configuration of the structural parts of the implant. Considering the fact that implant geometry is a major factor in the initial stability of an implant inserted into bone and that osseointegration contributes to the subsequent stability, such SEM results imply that the initial, or primary, stability is associated with the shape of the crest area and that the secondary, or biological, stability is mainly connected to the cellular behavior at the root area [27,28].

Under the circumstances of immobility, exogenous foreign material such as Ti implants, exhibit bone demarcation instead of implant rejection; hence, the stability-enhancing structures of an implant may be of particular importance [29]. Cylindrical implants without threads have uniform but weak attachment to the bone, which is especially weak to shear stress. This weakness may have been one reason why the cylindrical implants displayed a lot of bone resorption in situ [30]. With regards to the electron microscopic images captured in the in vivo environment of this study, all the implant components shown, including the thread structure and microtopography, are important in the cellular response during the osseointegration process. Altering the surface roughness of a material may affect the biological processes regulating the behavioral mechanisms (e.g., cell activity, adhesion) of osteoblastic/immune cells, such extracellular protein deposition at the moment of implantation has an influence on the cellular behavior which later leads to differences in in vivo outcomes [31,32]. This study was qualitative. Quantitative investigations are necessary for various modified surfaces. Recently, implants made of other materials—including ceramic and polyetheretherketone (PEEK)—have been developed, the surfaces of which need to be further investigated with respect to this in vivo cellular response [33,34].

This study successfully presented direct evidence of the behavior of osteogenic cells on the implant surface in an in vivo environment at the electron microscopic level. According to the interpreted data herein, in the bone surrounding dental implants, cell behavior is determined by the treated surface of the implant, whereas cells attached to the SLA implants seem to be able to read the configuration of different implant structures and develop an attachment pattern that conforms to those structures.

**Author Contributions:** Conceptualization, J.-Y.C. and Y.-J.J.; Methodology, J.-Y.C., Y.-J.J., and T.A.; Software, J.-Y.C.; Validation, J.-Y.C., T.A., and I.-S.L.Y.; Formal Analysis, J.-Y.C. and Y.-J.J.; Investigation, J.-Y.C.; Resources, J.-Y.C., T.A., Y.-J.J., and I.-S.L.Y.; Data Curation, J.-Y.C., T.A., Y.-J.J., and I.-S.L.Y.; Writing—Original Draft Preparation, J.-Y.C.; Writing—Review & Editing, J.-Y.C., T.A., and I.-S.L.Y.; Visualization, J.-Y.C., T.A., Y.-J.J., and I.-S.L.Y.; Supervision, I.-S.L.Y.; Project Administration, J.-Y.C. and I.-S.L.Y.; Funding Acquisition, I.-S.L.Y.

**Funding:** This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Ministry of Science, ICT and Future Planning, the Korean Government [No. NRF-2016R1A2B4014330].

**Conflicts of Interest:** The authors declare that there are no conflicts of interest regarding the publication of this paper.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
