Lactoferrin Coating Improves the Antibacterial and Osteogenic Properties of Alkali-Treated Titanium with Nanonetwork Structures

State Key Laboratory of Oral Diseases and National Clinical Research Center for Oral Diseases, West China School of Stomatology, Sichuan University, Chengdu 610041, China Department of Removable Prosthodontics and Occlusion, Osaka Dental University, 8-1 Kuzuha-hanazono-cho, Hirakata, Osaka 573-1121, Japan Department of Oral and Maxillofacial Surgery, Shenzhen Hospital, Southern Medical University, 1333 Xinhu Road, Shenzhen, 518100 Guangdong, China Department of Stomatology Center, Shenzhen Hospital, Southern Medical University, 1333 Xinhu Road, Shenzhen, 518100 Guangdong, China Faculty of Health Sciences, Department of Oral Health Engineering, Osaka Dental University, 1-4-4 Makino-Honmachi, Hirakata, Osaka 573-1121, Japan Osaka Dental University, 8-1 Kuzuha-hanazono-cho, Hirakata, Osaka 573-1121, Japan


Introduction
With the increasing clinical use of dental implants and the expansion of indications for implant therapy, dental implants have become an important option for the treatment of patients with dentition defects and dentition loss [1][2][3]. This fact further highlights the need for scientific research towards promoting the continuous optimization of implant materials. Titanium and its alloy materials have become the main materials used for dental implantation surgery, as well as the focus of scientific researchers because of their suitable mechanical strength, biocompatibility, durability, nontoxicity, and other advantages. At present, the continuous scholarly indepth study of titanium and its alloy materials has revealed them as being increasingly multifunctional surfaces, that is, materials that can simultaneously respond to the colonization by different cells (osteoblasts, fibroblasts, vascular endothelial cells, etc.) and infectious factors (bacteria, etc.) in order to better adapt to their clinical applications, which have become the focus of research on titanium materials in recent years [4][5][6].
The different approaches used to produce multifunctional titanium surfaces include inorganic coating, functional organic coating treatment, and chemical surface treatment. However, researchers are progressively inclined to combine various surface functionalization methods to obtain better biological effects [4,7]. Kim et al. used alkali and heat treatment to improve the biological properties of these materials [8]. Because this treatment is relatively simple and inexpensive and might be able to increase the possibilities of clinical applications, many researchers have subsequently carried out in-depth research on it [7,[9][10][11][12][13][14][15][16][17]. Through our research, we defined suitable conditions for forming a nanonetwork structure on the surface of titanium (TNS) using alkali treatment of titanium and its alloys at room temperature [10]. We subsequently analyzed the surface of TNS in detail and proved the superiority of this material on osteogenesis in both in vitro and in vivo experiments [11,12]. On this basis, we coated the TNS surface with fluorinated hydroxyapatite and amelogenin and the superior properties of these materials were experimentally verified [13,14]. On the contrary, it is undeniable that TNS does not possess antimicrobial properties. Therefore, we did some research and found that UV irradiation can effectively improve the antimicrobial properties of TNS [15,16].
In healthy subjects, lactoferrin circulates at a physiological concentration of 2-7 μg/mL, and acts as a growth factor in inducing the growth and activity of osteoblasts, inhibiting the development of osteoclasts in vitro, and promoting bone growth in vivo [22,[28][29][30][31][32]. There have been numerous studies on the effects of different concentrations of LF on osteogenesis and vascularization, as well as on its antimicrobial activity. From these results, we could identify that for LF concentrations in the range of 1-100 μg/mL, the osteogenic induction ability increased with the increase of LF concentration [22,30,31,33,34]. On that note, 8 μg/mL was the maximum concentration of LF not resulting in a negative impact on endothelial cell adhesion, whereas a higher concentration was used to inhibit vascularization [35]. Additionally, these LF concentrations were shown to play a certain role in the antibacterial properties of these materials [20,21]. Therefore, in this experiment, in order to further optimize the osteogenic induction ability of TNS and increase its antimicrobial properties, we chose LF at a concentration of 10 μg/mL as the material for optimizing the biological properties of the TNS surface and formed a new TNS-LF implantation system. Subsequently, the surface properties of TNS and TNS-LF materials were analyzed in detail. The antimicrobial properties of both materials were analyzed using Staphylococcus aureus as an experimental model organism. In addition, the biological properties of these 2 materials were evaluated through cell and animal experiments. In this study, we hope to provide some experimental basis for future research and clinical applications of this simple and inexpensive surface treatment method used for the generation of pure titanium implants.

Material and Methods
2.1. Sample Preparation. Titanium disks (JIS 2 grade, diameter 15 mm, thickness 1 mm; Engineering Test Service, Osaka, Japan), sequentially polished with several grades of abrasive paper (Waterproof Paper® Nos. 800, 1000, and 1500; Riken Corundum Co. Ltd., Saitama, Japan), and titanium screw implants (external diameter 1.2 mm, length 12 mm; Nishimura Metal, Fukui, Japan) were used in this study. In order to reduce pollution on the surface of titanium disks and implants before alkali treatment, we ultrasonically cleaned samples for 10 min in a succession of acetone, ethanol, and deionized water. After the samples were completely dried, they were immersed in alkali solution (10 M NaOH) for 24 h at 30°C and then transferred into ultrapure water. Ultrapure water was replaced every 5 min, until rinsate conductivity was <5 μS/cm 3 , and subsequently, samples were removed and dried at room temperature (20-28°C). After dry heat sterilization, the dried TNS material was used as a control in the following experiments. The obtained TNS was sterilized by dry heat, and the experimental steps thereafter were carried out in a sterile environment. Bovine LF (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) concentration was adjusted by gradient dilution with deionized water to 10 μg/mL, and 60 μl of LF solutions was applied twice to the surface of TNS materials; the amount of protein coating on the surface of each sample is 0.6 μg. After completely drying, TNS-LF materials were used in subsequent experiments. The whole process guaranteed the stability of the operation and strictly abided by aseptic principles.

Characterization of Surfaces
2.2.1. Morphological Analysis. Surface morphology of TNS and TNS-LF samples were examined by scanning electron microscopy (SEM) (S-4800; Hitachi, Tokyo, Japan) at a magnification of 50.0k and 100.0k with an accelerating voltage of 5 kV, and 4 images were taken for each sample.
2.2.2. Surface Topography Analysis. Surface topography, mean average surface roughness (Ra), and mean peak-tovalley height (Rz) were examined using atomic force microscopy (AFM) (SPM-9600; Shimadzu, Tokyo, Japan). Scanning areas were approximately 2 × 2 μm, and 4 different observation points were randomly selected for each sample. (1) Cell Culture of Rat Bone Marrow Mesenchymal Stem Cells (rBMMSCs). Rat bone marrow mesenchymal stem cells (rBMMSCs) were extracted from the femur of 8-week-old Sprague-Dawley rats (Shimizu Laboratory Supplies Co., Kyoto, Japan) and incubated in 75 cm 2 flasks abiding by a previously described method [7]. Following incubation, well adherent-grown cells (third-and fourth cell generations) were digested by trypsin, and after digestion was stopped with added fresh medium, cell cultures were centrifuged, supernatant was discarded, cell pellets were resuspended in fresh medium, and cells were counted as4 × 10 4 cells/mL for subsequent in vitro experiments.
(2) Cell Culture of Rat Periodontal Ligament Cells (rPLCs). Rat periodontal ligament cells (rPLCs) were purchased from Lonza (Walkersville, MD, USA), and according to recommendations, medium (SCBM™ Stromal Cell Basal Medium, Lonza, Basel, Switzerland) was allocated at appropriate concentrations for cell culture. The methods of thawing, resuscitating, and subculturing of cells were reported in previous studies [14]. immediately before the PBS wash). After 1 h incubation, 100 μL of the reagent from each well was added to a 96-well tissue culture plate and fluorescence was measured at 560/590 nm using a microplate reader (SpectraMax M5; Molecular Devices LLC, Sunnyvale, CA, USA). Measurements of rBMMSC and rPLC cell morphology were obtained at 3 and 6 h. Confocal laser-scanning microscopy (LSM 700; Carl Zeiss, Jena, Germany) was used for cell examination, and the respective dyeing method was performed according to the previously described experimental method [7].

Detection of Cell Osteogenic Markers.
Analysis of alkaline phosphatase (ALP) activity, quantification of calcium deposition in the extracellular matrix, osteocalcin (OCN) production, and expression levels of osteogenesis-related genes were used to detect the osteogenic differentiation induction ability of TNS and TNS-LF to rBMMSCs and rPLCs.
To evaluate induced ALP activity, rBMMSCs and rPLCs at a density of 4 × 10 4 cells/well were seeded on 24-well plates containing TNS and TNS-MAP disks, respectively. After the medium was cultured for 1 week, it was replaced with differentiation-inducing medium [7]; thereafter, cells were cultured for an additional 7 and 14 days, and then, the ALP activity of cells on the surfaces of TNS and TNS-LF was analyzed. Medium was changed every 3 days. Details on the respective method have been previously described [7,14]. As described above, the differentiation-inducing medium was replaced after 1 week of cell culture, cells were further cultured for 21 or 28 d, and consecutively, quantification of calcium deposition in the extracellular matrix and osteocalcin (OCN) production of cells on the surfaces of TNS and TNS-LF were evaluated using the methods previously described [7].
To evaluate the expression levels of osteogenesis-related genes, total RNA of rBMMSCs and rPLCs cultured for 3, 7, 14, and 21 d on TNS and TNS-LF sample disks was isolated using the RNeasy Mini Kit (Qiagen, Venlo, The Netherlands) and a TaqMan real-time RT-PCR assay (Life Technologies, Carlsbad, CA, USA), as previously described [7]. The expression levels of runt-related transcription factor 2 (Runx2), ALP, bone morphogenetic protein (BMP), and osteopontin (OPN) were measured. The StepOnePlus™ Real-Time PCR System (Life Technologies) was used to quantify these osteogenesis-related genes. Relative gene expression levels of each group were determined using the 2 −ΔΔCt method and normalized to that of the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) [7].

Osteointegration Characterization In Vivo
2.5.1. Animal Model and Surgical Procedures. As in our previous study, 10 male Sprague-Dawley rats aged 8 weeks (Shimizu Laboratory Supplies Co.) were randomly assigned to each of the TNS and TNS-LF groups. Previous reports have described in detail the experimental methods used, including the surgical procedure of sample implantation and postoperative care [7,17]. Animal studies were conducted in accordance with the National Institutes of Health guide and the ethical guidelines of the Animal Care and Use Committee of Osaka Dental University (Approval No. 18-03007).

Analysis of Animal Experimental
Results. Microcomputed tomography (micro-CT) and histological preparation and histomorphometric assessment and analysis performed were consistent with methods previously used [7,17]. The regions of interest (ROI; 500 μm around the implant and 2 mm below the epiphyseal line) in the CT images, bone volume fraction (BV/TV, %), trabecular number (Tb.N, mm -1 ), trabecular separation (Tb.Sp, μm), and trabecular thickness (Tb.Th, μm) were evaluated using the Morphometric software (TRI/3D-BON; Ratoc System Engineering, Tokyo, Japan). Bone area ratio (BA), boneimplant contact (BIC), and the area of new bone at 1, 4, and 8 weeks were also analyzed.

Characterization of Surfaces.
The surface characteristics of the 2 materials in both the control and experimental groups were analyzed as shown in Figure 1, Table 1, and  Table 2. Figure 1(a) presents SEM images of the surface morphology of the 2 kinds of materials at different magnifications. Under 50k and 100k magnification (especially at ×100k), it can be clearly seen that the surface of the TNS material has formed a uniform nanonetwork structure, and this nanonetwork is relatively sharp and slender. Although the nanonetwork was obviously thicker and rounder in the TNS-LF surface, after protein coating, the whole material still retained the morphological characteristics of the TNS nanoreticular structure. Figure 1(b) presents the AFM results obtained on the morphological characteristics of the surface of both materials. As shown in Figure 1, the surface of TNS exhibits a uniform nanoprotuberance structure. The nanoprotuberance on the surface of the TNS-LF material on the other hand is more obvious and rounded. Through the analysis of Ra and Rz on the surface of the two materials (Table 1), we found that both Ra and Rz of TNS-LF were slightly increased compared with TNS. Results of chemical element analysis on the surface of the 2 materials showed that the content of N, O, and C on the surface of TNS-LF was higher than that of TNS. Especially, the content of the N element increased from almost 0 to about 20%. Figure 1(c) illustrates the result of FTIR analysis. Both kinds of materials were tested and analyzed in the range of 350-4000 cm -1 . We found that there were different waveforms in the abscissa coordinate of around 2000 cm -1 . After further analysis in the range of 1350-1750 cm -1 , it was noted that there were obvious differences between the 2 materials. 4 Journal of Nanomaterials   Figure 2. Figure 2(a) shows that the 24 h antimicrobial rate of TNS-LF was 99%, indicating a remarkable antimicrobial effect of TNS after coating with LF. Similarly, results of the 24 h biofilm formation experiment (Figure 2(b)) also showed that biofilm formation on the surface of TNS-LF was less apparent than that on the surface of the TNS material. Results of Live/Dead and ROS staining for 1 and 6 h incubations are shown in Figure 2(c). Results following incubation for 1 h showed that the number of bacteria adhering to the surface of both materials was relatively low, and more specifically, the amount of bacteria adhering to the surface of the TNS-LF material was significantly (P < 0:001) less than that of TNS, and there were more dead bacteria present on TNS-LF than on the TNS material. After incubation for 6 h, bacteria that attached to the surface of the TNS material multiplied in large numbers, whereas the number of bacteria on the surface of the TNS-LF material was increased compared with that after 1 h incubation, but was still far lower than that of the TNS group. Regarding the results of ROS evaluation, no obvious ROS staining spots were observed in the TNS group at both 1 and 6 h, whereas obvious ROS staining spots could be observed at both time points, on the surface of the TNS-LF material. Figure 3(a) and Figure 4(a) illustrate the initial attachment of rBMMSCs and hPLCs on the surface of each material, respectively. This result indicated that TNS-LF could promote cell attachment on the surface of the material better than TNS at 1, 3, 6, and 24 h, thus providing a basis for cell proliferation and differentiation. Figure 4(b) presents the cell morphology of hPLCs on the surface of both materials at 3 and 6 h. We could clearly observe that hPLCs on the surface of TNS-LF displayed obvious advantages over TNS in both quantity and cell extension. Figure 3(b) and Figure 4(c), TNS-LF was able to promote alkaline phosphatase activity, Ca deposition in extracellular matrix, and osteocalcin (OCN) expression in both cells better than TNS materials.

Evaluation of Osteogenic Induction Ability In Vitro. As shown in
As shown in Figure 3(b), regarding the expression levels of osteogenic differentiation-related genes, the experimental group (TNS-LF) material demonstrated a significant promoting effect on the expression levels of these genes in rBMMSCs at various observation points compared with the control group (TNS) material.

Evaluation of Osteogenic Induction Ability In Vivo.
Results obtained from the in vivo experiments and their respective data analyses are shown in Figure 5. In the 3D image analysis of Figure 5(a), it can be clearly seen that the new bone that formed around the TNS-LF implant is greater than that around the TNS implant (implant in red, cortical bone in blue, and new bone in Kelly green). Concerning the quality of new bone, analysis of the BV/TV (P < 0:001), Tb.N (P < 0:05), and Tb.Th (P < 0:001) parameters of the new bone that formed around the implant revealed that they were significantly higher in the experimental group compared to the control group.
As shown in Figure 5(b), the area of new bone tissue that was generated around the implants in the experimental group was much larger than that in the control group. After further analysis and calculation of BIC and BA, it was found that the area of new bone around the TNS-LF implant was about 3 times greater than that of TNS, and BIC increased by nearly 20% compared with TNS.
Results of fluorescence staining visualized by fluorescence confocal microscopy are presented in Figure 5(c). New bone formation around TNS-LF implants increased significantly (P < 0:001; P < 0:01) compared with that around TNS. After calculating the new bone mass generated at 1, 4, and 8 weeks, it was revealed that the new bone that formed around the experimental group was slightly more than that of the control group at week 1, while the main time period of bone formation was between 4 and 8 weeks, during which period, new bone formation reached about 60% of the total. Additionally, the amount of new bone generated in the experimental group was greater than that in the control group, at each calculation time point.

Discussion
In this study, we analyzed the surface characteristics of TNS and TNS-LF materials in detail, evaluated the antimicrobial properties of the 2 materials using Staphylococcus aureus as a model organism, and analyzed the early attachment of rBMMSCs and hPLCs growing on the surface of these materials, as well as the expression levels of markers related to osteogenic differentiation. Concomitantly, animal experiments were also carried out to study the biological properties   Journal of Nanomaterials All these findings of characterization of the surfaces showed us that the surface of the TNS material following LF coating exhibits a new chemical form [36,37]. Results obtained by SEM and AFM investigation showed that alkali treatment at room temperature resulted in a uniform nano-scale network structure on the surface of pure titanium, as was previously described [7,38]. At the same time, we noticed that LF was uniformly adhered to the surface of the TNS material, but still retained the nanoreticular structure of the original substrate material. Chemical and physiochemical (XPS and FITR) characterization further validated the successful coating of the TNS surface with LF. The reason for choosing TNS as the base material in this study was that according to the results of Quartz Crystal Microbalance Sensor (QCM) analysis of TNS and Ti materials performed in previous studies, we knew that TNS adsorbs proteins more easily than Ti [14,39]. Concurrently, TNS has been shown to display increased hydrophilicity [7] relative to Ti Because there is no interface between the nanoreticular structure layer on the surface of TNS and Ti, the method used can efficiently coat proteins on the surface of TNS, offering a strong binding. Another reason was that our previous studies had concluded that this biocomposite implant was not only a protein-mechanical and TNS-physical-chemical structure, but also a functional combination of the 2 [7,14]. Moreover, no significant differences in the surface roughness of the TNS and TNS-LF surfaces were found.
A serious problem in the clinical setting stems from the fact that microorganisms adhere to abiotic surfaces and form biofilms. In implantation surgery, if a new implant material used could ensure reduced bacterial attachment, it would achieve certain clinical and commercial success, which is of great significance [4,40]. In this study, TNS-LF showed more effective antimicrobial properties relative to TNS, which is essential for long-term survival and implant success. Since the TNS material does not have antimicrobial properties [15,16], the antimicrobial properties exhibited by TNS-LF were due to LF coated on the surface of the TNS material. By performing experiments evaluating the in vitro antimicrobial properties of samples, we found that the LF coating provided obvious antibacterial attachment to the sample surface and also exhibited a certain bactericidal effect. After the initial discovery of the antimicrobial activity of LF [41,42], scholars have demonstrated that LF is a broad-spectrum antimicrobial activity protein and thus can inhibit the growth of many bacteria [43][44][45][46] through a direct action on bacteria themselves. It has been reported that the molecular mechanism of the bactericidal effect of LF on Staphylococcus aureus (Gram-positive bacteria) used in this study is likely to be similar to that of cationic and amphiphilic antimicrobial peptides [41]. Antimicrobial peptides bind to Gram-positive bacteria through electrostatic interaction between the negatively charged lipid matrix and cationic amino acid residues in the target membrane. After binding, due to hydrophobic interaction, the parent residues interfere with the interior of the nonpolar membrane [47,48]. This seems to correspond to the same N-terminal region of LF, where hydrophobic residues are located close to cationic residues, and amphiphilic cationic titanium obtained at the N-terminal of human or bovine LF has been shown to be many times more active to bacteria than the parent protein itself [41]. Likewise, it has been reported that LF can be used as an inhibitor of bacterial biofilm formation under certain conditions [49][50][51].
From the results obtained in this study, it could be clearly suggested that TNS-LF materials are able to promote the early attachment of cells and promote the osteogenic differentiation of rBMMSCs. The amount of new bone formed around TNS-LF implants and the implant bone binding rate in the observation area were higher than those in the control group. It has been reported that the mechanism of LF action in osteoblasts might be due to the mitogenic effect shown in osteoblasts, mainly mediated by LRP1 and related to the activation of IGF1, Ptgs2, and Nfatc1. LF-induced activation of p42/44 MAPK and PI3-kinase-dependent ATP signaling in osteoblasts is also another possibility [28]. Successful implant-bone bonding requires the concerted efforts of different types of cells, which should not be ignored in some cases of immediate implantation after tooth extraction. For example, the periodontal ligament plays a certain physiological function by attaching to the cementum and alveolar bone. It has been reported that there are still periodontal ligament cells attached to the side of the alveolar bone in the extraction socket after tooth extraction. The results of our study showed that the TNS-LF material could not only promote the adhesion and osteogenic differentiation of rBMMSCs around the material but also promote the adhesion of hPLCs and induce osteogenic differentiation, an additional advantageous property of this material.

Conclusions
We used simple alkali treatment to obtain a stable experimental TNS base material and then applied a simple method to obtain an appropriate concentration of LF coating on the surface of TNS, which could promote the osteogenic differentiation of bone marrow mesenchymal stem cells as well as the osteogenesis of periodontal ligament cells. Concurrently, the LF physiological concentration was not able to exert too much effect on vascular endothelial cells, but was still adequate in hindering formation of osteoclasts, thus providing its positive role in regulating a variety of cell behaviors in the process of bone binding. In addition, TNS-LF also exhibited antimicrobial properties, providing a guarantee for the success of implants. Importantly, the effect was obvious despite of the whole process of material treatment being relatively simple, and the cost of treatment being relatively low. The purpose of this study was to provide a case of such an implant, and to analyze the possibility and value of its clinical use from various perspectives, so as to provide a basis for future experiments. Experiments evaluating the effect of these materials on macrophages, osteoclasts, and vascular endothelial cells need to be further performed, and we will continue to attempt to identify the ideal concentration of this material around physiological concentrations to achieve the best balance between promoting osteogenesis and antimicrobial properties.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Additional Points
Statement of Significance. The purpose of this work is to use simple and inexpensive methods to treat implant materials and integrate biological proteins on the surface to improve the biological properties of materials. Especially, the TNS material coated with LF functions as an antibacterial and promotes osteogenesis. It is noteworthy that this material can also induce osteogenesis of periodontal ligament cells, which provides a new implant material for immediate implantation after tooth extraction. On the basis of other scholars' research and our previous work, we chose the appropriate LF concentration to prove the superiority of this material from the aspects of material surface analysis experiment, cell experiment, and animal experiment, which provides a theoretical possibility for the clinical application of this material.

Conflicts of Interest
The authors declare that they have no conflicts of interest.