Enhanced Biocompatibility and Antibacterial Activity of Selective Laser Melting Titanium with Zinc-Doped Micro-Nano Topography

Selective laser melting (SLM) titanium is a suitable material for use in customized implants. However, long-term implant survival requires both successful osseointegration and promising antibacterial characteristics. Native SLM titanium thus requires proper modifications to improve its bioactivity and antibacterial efficacy. Micro-arc oxidation (MAO) was conducted on sandblasted SLM substrate to form a microporous TiO2 coating. Subsequently, hydrothermal treatment was applied to fabricate micro-nano zinc-incorporated coatings with different Zn content (1mM-Zn and 100μM-Zn). Surface characterization was performed using scanning electron microscopy, X-ray photoelectron spectroscopy, X-ray diffraction, a three-dimensional profilometer, and a contact angle measuring device. The osteoblast-like cell line MC3T3-E1, Subclone 14, was used in cell viability assays to evaluate adhesion, proliferation, and ALP activity. An antibacterial assay was conducted using Streptococcus sanguinis and Fusobacterium nucleatum. Zn-incorporated samples exhibited higher adhesion, proliferation, and differentiation than did SLM and MAO samples. 100μM Zn samples exhibited the highest proliferation, and 1mM-Zn samples manifested the highest ALP activity. In addition, Zn-incorporated samples exerted inhibitory effects on both Streptococcus sanguinis and Fusobacterium nucleatum. Combining micro-arc oxidation and hydrothermal treatment, we successfully fabricated a novel Zn-incorporated coating on a microporous SLM surface which possesses both outstanding bioactivity and antibacterial efficacy.


Introduction
Dental implants are the prior choice for patients suffering tooth loss due to predictable clinical outcomes [1]. However, patients whose conditions are complicated by irregular alveolar bone structure or extraction sockets are unsuitable to apply with such implants as their lengths, diameters, and tread parameters are predetermined and unmodifiable [2]. With the rapid development of computer-aided design (CAD) and additive manufacturing (AM), personalized implants can significantly simplify clinical procedure and shorten treatment times. Furthermore, customized implants can minimize stress shielding and pressure-induced bone loss when compared to traditional implants [3,4]. As manufacturing processes continue to advance, an increasing variety of customized implants are entering clinical use.
Selective laser melting (SLM), one of the most modern types of additive manufacturing (AM), is particularly well suited to fabricate customized implants or bone substitutes with free-form geometry [5][6][7]. Since the bone-implant interface is crucial to osseointegration, numerous studies have focused on evaluating the biocompatible properties of the SLM surface [8][9][10][11]. Compared to conventional machineprocessed titanium, the biocompatibility of SLM substrate remains controversial. Shaoki et al. reported that compared to the surface of machine-processed titanium, rates of osteoblast adhesion and differentiation were higher on an SLM surface but residual melted titanium particles decreased removal torque values [8]. Studies have also revealed that bacteria easily adhere to SLM substrate due to its rough and irregular surface [12,13]. After SLM modifications both in vivo and in vitro (i.e., anodic oxidation, alkali-heat treatment, etc.), remarkable antibacterial properties, bone regeneration, and soft tissue adhesion were demonstrated with an SLM surface [12,14,15]. However, few of them possess both biocompatibility and antibacterial activity at the same time.
Micro-arc oxidation (MAO) has been widely used in the modification of biomaterial surfaces and confers a wide variety of biomechanical advantages [16,17]. MAO-produced porous TiO 2 coating, incorporated with calcium and phosphorous, has been found to favorably influence osteoblast adhesion, proliferation, and differentiation [18]. In our previous study, we fabricated a rougher TiO 2 coating by combining sandblasting and acid etching technology with MAO, thus significantly enhancing biocompatibility [19]. Layer-by-layer manufacturing of native SLM specimens also resulted in a rough surface. Few studies, however, have focused on modifying the microscale surface of SLM specimens by MAO.
Peri-implantitis caused by bacteria is one severe complication that can eventually lead to implant failure [20]. Bacteria can not only aggregate on the implant surface, forming a biofilm and resisting antibacterial treatment, but also impede osseointegration and form alveolar absorption [21]. In order to enhance the antibacterial properties of biomaterials, various modification strategies have been applied, such as the loading of antibiotics, cross-linking of drugs, and construction of micro-nano topography [12,22,23]. Although antibacterial properties have improved, few methods with a focus on improving both osteogenesis and antibacterial properties have been reported.
Zinc, an essential trace element involved in multiple skeletal signaling pathways, reduces osteoclast resorption and improves osteoblast differentiation by regulating the expression of related genes [24,25]. Biomaterials incorporating Zn have been found to exhibit remarkable osteogenic and antibacterial activity [26][27][28][29][30]. Among the different Zn compounds, ZnO exhibits powerful antibacterial properties due to its membrane-damaging effects, generation of reactive oxygen species (ROS), and release of Zn ions [31]. However, a high concentration of Zn not only is toxic to bacteria but also causes damage to human cells [32]. It is thus vital to appropriately adjust the concentration of Zn in any biomaterial, enough to inhibit bacterial adhesion and reproduction while conferring no cytotoxicity.
In this study, we modified the surface of an SLM substrate using MAO thus forming a microporous morphology. We further doped Zn ions onto the MAO surface by hydrothermal treatment, constructing a micro-nano topography on the SLM specimen. Surface characterization was performed by scanning electron microscopy (SEM), energy-dispersive spectroscopy (EDS), X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), a 3D profilometer, and a contact angle measuring device. Antibacterial activity was assessed by oral pathogenic bacterial species, including S. sanguinis and F. nucleatum. Cytocompatibility of the modified SLM surface was evaluated by characterizing the morphology, attachment, proliferation, and differentiation of osteoblast-like cells (MC3T3-E1). Despite that MAO has been widely used in implant surface modification, here we present a novel method to combine MAO and zinc element of the SLM substrate, which both improves biological performance and enhances antibacterial efficacy with the ultimate aim of improving customized dental implant success rates and reducing the risk of peri-implant inflammation.

Materials and Methods
2.1. Specimen Preparation and Treatment. The SLM specimens were manufactured as described previously [12,15]. Briefly, round-disk samples (diameter, 10 mm; thickness, 1 mm) were designed by SolidWorks® 12.0 (SolidWorks Corp, USA). According to the data, specimens were fabricated by an SLM machine (SLM125HL, SLM Solutions GmbH, Germany) using commercial grade II titanium powders (Western BaoDe, China) as raw materials. The main characteristics of SLM machine were as follows: the maximum laser power is 145 W, the layer thickness is 30 μm, the laser scanning speed is 355 mm/s, the laser spot size is 50 μm, and the hatch space is 45 μm. All SLM titanium specimens were ultrasonically cleaned with acetone, ethanol, and deionized water for 15 min each time considered as SLM samples. To eliminate the residual titanium spheres and the native oxide layer of surface, some of the SLM samples were sandblasted by ZrO 2 particles (diameter, 250 μm) and then chemically polished in 5% hydrofluoric acid solution for 30 sec. Subsequently, they were ultrasonically cleaned and dried. MAO was conducted in an electrolyte solution with 0.02 M glycerophosphate disodium salt pentahydrate (C 3 H 7 Na 2 O 6 P·5(H 2 O) and 0.05 M calcium acetate monohydrate (Ca(CH 3 COO) 2 ·H 2 O)) using a pulsed DC power supply (MHWYDM750-20, Nenghua Electrical Co., China). Frequency, duty, voltage, and time were 100 HZ, 26%, 410 V, and 6 min, respectively. These specimens were considered as MAO samples. After MAO, parts of the MAO samples were hydrothermally treated in a Teflonlined autoclave containing 1 mM and 100 μM of zinc acetate (Zn(CH 3 COO) 2 ) at 200°C for 3 h. These further treated specimens we termed as 1 mM-Zn and 100 μM-Zn samples. Each sample was ultrasonicated in deionized water after every single treatment step. Samples were divided into four groups, namely SLM, MAO, 1 mM-Zn, and 100 μM-Zn.

Specimen
Characterization. Surface topographies were observed using field-emission scanning electron microscopy (FE-SEM, SU70, Hitachi, Japan). Chemistry compositions of surface were analyzed using an electronic differential system (EDS) equipped with FE-SEM and X-ray photoelectron spectroscopy (XPS, ESCALAB 250Xi, Thermo Scientific, USA). Crystallinity of the coatings was evaluated using an X-ray diffractometer (XRD, PANalytical B.V., X'Pert-PRO, Holland) fitted with a Cu Kα (λ = 0 154056 nm) at 40 kV 2 Journal of Nanomaterials and 40 mA in the range from 20°to 80°(2θ) with a step size of 0.02. Surface profiles and roughness were scanned with a 3D profilometer (BMT EXPERT, Germany). The static contact angles were determined by an optical contact angle measuring device (OCA40 Micro, Data Physics, Germany). To determine Zn-related characteristics of each sample, 1 mM-Zn and 100 μM-Zn specimens were immersed in 3 ml phosphate-buffered saline (PBS, Gibco, USA) at 37°C for 1, 4, 7, and 14 days successively. At predetermined time points, the collected PBS that contained Zn was analyzed by inductively coupled plasma mass spectrometry (ICP-MS, Agilent 7900, Agilent Technologies Inc., USA).

Cell
Culture. The preosteoblastic cell line MC3T3-E1 subclone 14 (Shanghai Cellular Institute of the Chinese Academy of Sciences, China) was cultured in α-MEM medium (Gibco, USA) containing 10% fetal bovine serum (FBS, Gibco, USA) and 1% antibiotic/antimycotic solution (Gibco, USA) at 37°C under a humidified atmosphere of 5% CO 2 . The culture medium was changed every 2 days to remove suspended cells.

Cell Proliferation Activity Assay.
Cell counting kit-8 assay (CCK-8, Dojindo, Japan) was used to evaluate the cell proliferation of MC3T3-E1 quantitatively [33]. MC3T3-E1 cells were seeded on sample surfaces at a density of 1 × 10 4 cells ml −1 in 48-well plates and cultured for 1, 4, and 7 days. At these predetermined time points, each sample was rinsed three times with PBS. Adherent cells were incubated with 500 μl 10% CCK-8 mixed with culture medium for 2 h at 37°C. The optical density (OD) was measured at 450 nm on an ELISA reader (Tecan, SUNRISE, Switzerland).

Alkaline Phosphatase (ALP) Activity
Assay. MC3T3-E1 cells were seeded on sample surfaces at a density of 2 × 10 4 cells ml -1 in 48-well plates. After culturing with osteogenic medium (α-MEM medium supplemented with 10 mM β-glycerophosphate, 0.05 mM l-ascorbic acid, and 100 mM dexamethasone) for 14 days, cells were lysed by 0.1% Triton X-100 at 4°C overnight and an ALP assay kit (Jiancheng, China) was subsequently used to evaluate the ALP activity measured at an OD value of 520 nm. A protein assay kit (BCA, Beyotime, China) was used to normalize readings to total protein content measured at a wavelength of 562 nm [33].  Sample EDS spectra are shown in Figure 2, and surface elemental compositions are shown in Table 1    Journal of Nanomaterials due to their low content or amorphous form in these asprepared samples. XPS analysis (Figure 4) was performed to evaluate the surface chemical states of as-prepared samples. In MAO samples, two peaks (at 532.5 and 530.9 eV) were observed in O 1s XPS spectrum, assigned to the binding energy of O in hydroxyl and lattice oxygen. Peaks at 458.6 eV and 464.3 eV in the Ti 2p XPS spectrum were, respectively, assigned to Ti 4+ 2p 3/2 and Ti 4+ 2p 1/2 . These XPS findings were in agreement with XRD analysis, confirming that the TiO 2 formed during the synthetic procedure. Peaks at 347.6 and 351.2 eV in Ca 2p XPS spectrum, respectively, corresponded to Ca 2+ 2p 3/2 and Ca 2+ 2p 1/2 . The binding energy at 133.8 eV for P 2p was assigned to the p 5+ in Ca 10 (PO 4 ) 6 (OH) 2 [34]. Compared to MAO samples, no obvious changes in the binding energy of P 2p and Ca 2p were observed in the 100 μM sample. The Zn 2p 3/2 at 1021.5 eV and 2p 1/2 at 1044.5 eV suggested that Zn 2+ content was successfully coated on the surface of TiO 2 . The O 1s could additionally be split into three peaks: 532.5 eV for hydroxyl in Ca 10 (PO 4 ) 6 (OH) 2 , 529.9 eV for lattice oxygen in TiO 2 , and 530.9 eV for lattice oxygen in ZnO. The greater binding energy of Zn-O than that of Ti-O is attributed to the greater electronegativity of Zn (1.65) than Ti (1.54). Our findings imply that the layer coating TiO 2 was composed of ZnO and Ca 10 (PO 4 ) 6 (OH) 2 .
Surface profile images and sample hydrophilicity as determined by water contact angle analysis are shown in Figure 5; findings are shown in Table 2. SLM samples exhibited the highest Ra value of 6.96 μm. The Ra value of MAO, 1 mM-Zn, and 100 μM-Zn samples were 1.85, 1.74, and 1.85 μm, respectively, and were smoother than SLM samples. The water contact angles of SLM and MAO samples were 59.07°and 15.7°, respectively, while 1 mM-Zn and 100 μM-Zn samples exhibited total hydrophilicity and their water contact angles could not be measured.
The Zn ion concentration released in PBS was measured by ICP-MS ( Figure 6). During immersion in PBS for 14 days, both Zn-incorporated samples released Zn ions continuously at a low level. The total amount of Zn ions in 1 mM-Zn samples was higher than in 100 μM-Zn samples which were in accordance with the surface EDS analysis above. Figure 7 shows results of MC3T3-E1 proliferation on each sample. At days 1, 4, and 7, MAO, 1 mM-Zn, and 100 μM-Zn samples were found to have greater proliferation than were SLM samples. The 100 μM-Zn group showed higher proliferation than did other groups on the 1st and 7th days. MC3T3-E1 cells exhibited a spindle and elongated shape on SLM surfaces but cellular morphology on MAO, 1 mM-Zn, and 100 μM-Zn samples extensively varied (Figure 8).

Alkaline Phosphatase (ALP) Activity.
The osteogenic function of MC3T3-E1 cells was evaluated by analyzing ALP activity after 14 days of culture ( Figure 9). SLM samples performed poorly compared to the other three groups in the ALP activity test. ALP activities in 1 mM-Zn and 100 μM-Zn groups were statistically significantly higher than in the MAO group, and the 1 mM-Zn group was found to have the highest statistically significant level (1 mM-Zn > 100 μM-Zn > MAO > SLM).
3.4. Antibacterial Activity. Antibacterial findings are detailed in Figure 10. Both S. sanguinis and F. nucleatum adhered to SLM samples at rates significantly higher than in other groups. The antibacterial efficacy of MAO samples against adhered F. nucleatum was 18 32 ± 4 88%; however, no antibacterial effects of MAO samples against S. sanguinis were found despite SLM samples possessing the roughest surface. Samples incorporated with Zn significantly enhanced antibacterial activity with 1 mM-Zn group samples exhibiting the most effective inhibition of both bacterial species. Morphologies of S. sanguinis and F. nucleatum captured by SEM ( Figure 11) revealed how samples incorporated with Zn also reduced bacterial adhesion. Lower quantities of S. sanguinis and F. nucleatum were found on 1 mM-Zn and 100 μM-Zn sample surfaces, while organisms found adhered to SLM and MAO samples were densely aggregated.
Bacterial density in Zn-incorporated sample culture media was statistically significantly lower than in SLM and MAO groups (Figure 12). Compared with 100 μM-Zn samples, bacterial density in 1 mM-Zn samples was markedly lower, implying that bacterial density gradually decreased as the concentration of Zn ions in supernatant increased.

Discussion
Surface modification of titanium can significantly enhance the osteogenesis and antibacterial efficacy [35][36][37][38]. In this study, we fabricated a novel Zn-incorporated TiO 2 coating by MAO and subsequent hydrothermal treatment producing micro-nano topography on an SLM surface.
Surface roughness plays an important role in the adherence and proliferation of cells and bacteria. Albrektsson and Wennerberg demonstrated that dental implants with moderately rough surfaces (1 0 μm < Ra < 2 0 μm) evoke stronger bone responses than do smoother or rougher surfaces [39]. Rønold and Ellingsen [40] and Rønold et al. [41] proved that bone bonding forces increase with rougher surfaces to a threshold of 3.62-3.90 μm (range: 0.63 μm-11.03 μm). Increased roughness also promotes the formation and maturation of biofilms, especially when the Ra value exceeds 0.2 μm [42,43]. Due to the layer-by-layer manufacturing process and unmelted spheres, the original SLM specimens we fabricated had an Ra value of 6.96 μm. In order to decrease substrate roughness, we applied sandblasting and acid polishing to eliminate residual particles  Surface wettability is another important physicochemical property of biomaterials. Previous studies have shown that the hydrophilicity of surfaces influences both cellular and bacterial morphology, adherence, and proliferation. Compared to hydrophobic surfaces, hydrophilic surfaces significantly promote cellular attachment and proliferation [44]. In addition, implants with significant wettability decrease the postimplant healing period and accelerate early osseointegration [45][46][47]. MAO, 1 mM-Zn, and 100 μM-Zn samples were all evaluated with high surface hydrophilicity by water contact angle analysis. Most oral bacteria, including S. sanguinis and F. nucleatum, thrive better on surfaces with high wettability [48,49]. Further modifications should be applied to improve antibacterial effects.
Zn-incorporated coatings are well known to possess strong bacteriostatic properties, effectively inhibiting adhesion and proliferation of a wide variety of bacteria [26][27][28]. The ZnO in coatings also exerts antibacterial effects by ROS generation (i.e., H 2 O 2 , OH-, and O 2 2-) [50]. In addition, free Zn ions also penetrate bacterial membranes and cause DNA damage [51]. Research has proven that dental implant failure was commonly associated to dental plaque formation. As an early colonizer of dental plaques, S. sanguinis can attach directly to the surface of implants and facilitate the adhesion of later colonized bacteria, which eventually transfer the plaque to a "mature" community with high levels of Gramnegative anaerobic filamentous organisms [52,53]. F. nucleatum, one of the most common pathogens associated with peri-implantitis, is a later colonizer during biofilm formation [54]. We also noted that the antibacterial efficacy towards both S. sanguinis and F. nucleatum depended on the content of Zn within coatings. This is likely due to coatings with a higher Zn content releasing a greater quantity of Zn ions and more strongly inhibiting bacterial adhesion and proliferation by mechanisms such as ROS formation. In contrast to S. sanguinis, colony counts of F. nucleatum adhered to MAO samples were significantly lower than those of SLM samples. This difference is likely due to the catalytic effect of anatase in MAO coatings [55].
As a trace element, Zn plays a crucial role in the process of bone regeneration. Various studies have reported that biomaterials incorporating Zn exhibited promising biological characteristics both in vivo and in vitro [29,30]. Zn enhances osteoblastic adhesion and proliferation by upregulating the expression of FAK and integrin [56]. Zn exerts positive effects on bone formation by promoting the expression of genes involved in osteogenesis such as Runx2, ALP, OCN, and Col-I [30]. Moreover, Zn also decreases bone loss by suppressing osteoclast differentiation via inhibition of RANK expression [57]. Importantly, Zn can only confer favorable cytocompatibility properties and minimal cytotoxicity at an extremely low concentration. Yamamoto et al. demonstrated that the half maximal inhibitory concentration (IC) 50 of ZnCl 2 is 9 28 × 10 −5 mol/l (equals to 12.65 ppm of Zn ions) [58]. Ito et al. additionally reported that no cytotoxicity was observed at a Zn ion concentration of 3 ppm [59]. As shown in Figure 6, Zn ions released from 1 mM-Zn and 100 μM-Zn samples accumulated to levels lower than 3 ppm and thus remained in a relatively safe range.
In our present experiment, MC3T3-E1 cells were used to test Zn-incorporated coating viability. SEM images revealed that both 1 mM-Zn and 100 μM-Zn samples were favorable    Figure 7: Cell proliferation of MC3T3-E1 cells on the surface of each sample after 1, 4, 7 days. # P < 0 05 and ## P < 0 01 compared with SLM; * P < 0 05 and * * P < 0 01 compared with MAO; $ P < 0 05 and $$ P < 0 01 compared with 1 mM-Zn. 8 Journal of Nanomaterials for cell adhesion. Cells were observed extensively spreading out on Zn-incorporated coating surfaces. In addition, firstday CCK-8 results revealed that cells proliferate better on Zn-incorporated samples when compared to SLM and MAO samples, likely due to the outstanding hydrophilicity and special micro-nano topography of the former groups. After incubation for 7 days, 100 μM-Zn samples were found to be the most ideal for cell proliferation. MAO and 1 mM-Zn samples exhibited similar biological characteristics and were both superior to SLM specimens. ALP activity, an early marker of osteogenesis, was markedly increased on Zn-incorporated specimens compared to SLM and MAO samples. Moreover, 1 mM-Zn samples expressed the highest ALP activity after 14 days, suggesting that higher loading of incorporated Zn exerts positive effects on cellular differentiation as long as Zn content does not exceed a concentration that is biologically safe. These results were not completely in accordance with the cell proliferation assays in which 100 μM-Zn samples showed the best proliferation ability. This might be because a high content of ZnO from Zn-incorporated samples could not only enhance ALP activity but also decrease the proliferation of cells in some degree at the same time despite the suitable surface topography, roughness, and hydrophilicity, which drew no conflicting conclusions to these two results.    Figure 10: Antibacterial ratio of S. sanguinis and F. nucleatum on the surfaces of each sample after 24 h of incubation. # P < 0 05 and ## P < 0 01 compared with the baseline (0% as SLM group). * P < 0 05 and * * P < 0 01 compared with MAO.  Figure 12: The bacteria density in supernatant of S. sanguinis and F. nucleatum after 24 hours of incubation. S. sanguinis (a) and F. nucleatum (b). # P < 0 05 and ## P < 0 01 compared with SLM; * P < 0 05 and * * P < 0 01 compared with MAO; and $ P < 0 05 and $$ P < 0 01 compared with 1 mM-Zn. 10 Journal of Nanomaterials With MAO and subsequent hydrothermal treatment, varying quantities of Zn can be incorporated into TiO 2 coatings to modify the native SLM surface. Compared with SLM and MAO specimens, Zn-incorporated samples exhibited better antibacterial activity and cytocompatibility.
Further studies should focus on animal experiments to evaluate whether this Zn-incorporated modification possesses similar biological capabilities in vivo as it does in vitro, thus establishing a solid foundation for future clinical application.

Conclusions
In this study, a novel micro-nano topographic coating incorporated with Zn was fabricated on the rough SLM surface by MAO and hydrothermal treatment. Due to a sustained release of Zn ions and superior hydrophilicity, Zn-incorporated samples highly promoted the adhesion, proliferation, and differentiation of MC3T3-E1 cells when compared to SLM and MAO samples. This modified surface also exhibited outstanding antibacterial effects against oral pathogenic flora (S. sanguinis and F. nucleatum), potentially significantly reducing the risk of peri-implant inflammation and ensuring long-term implant survival. Our study thus contributes to the advancement of customized biomedical implant fabrication with SLM technology.

Data Availability
All data included in this study are available upon request by contact with the corresponding author.

Conflicts of Interest
There is no conflict of interest regarding the publication of this paper.

Authors' Contributions
Fan Wu and Ruogu Xu contributed equally to this work.