Research Article www.acsami.org
Falling Leaves Inspired ZnO Nanorods−Nanoslices Hierarchical Structure for Implant Surface Modification with Two Stage Releasing Features Hang Liao,† Xinxin Miao,† Jing Ye,† Tianlong Wu,† Zhongbo Deng,† Chen Li,† Jingyu Jia,† Xigao Cheng,*,† and Xiaolei Wang*,†,‡ †
Department of Orthopedic Surgery, The Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi 330006, China Institute of Translational Medicine, NanChang University, NanChang, Jiangxi 330031, China
‡
S Supporting Information *
ABSTRACT: Inspired from falling leaves, ZnO nanorods−nanoslices hierarchical structure (NHS) was constructed to modify the surfaces of two widely used implant materials: titanium (Ti) and tantalum (Ta), respectively. By which means, two-stage release of antibacterial active substances were realized to address the clinical importance of long-term broad-spectrum antibacterial activity. At early stages (within 48 h), the NHS exhibited a rapid releasing to kill the bacteria around the implant immediately. At a second stage (over 2 weeks), the NHS exhibited a slow releasing to realize long-term inhibition. The excellent antibacterial activity of ZnO NHS was confirmed once again by animal test in vivo. According to the subsequent experiments, the ZnO NHS coating exhibited the great advantage of high efficiency, low toxicity, and long-term durability, which could be a feasible manner to prevent the abuse of antibiotics on implant-related surgery. KEYWORDS: antibacterial, zinc oxide, surface modification, nanoarrays, Ti, Ta, implant
■
INTRODUCTION Nowadays, titanium (Ti) and its alloys1,2 have been extensively used as implants,3 due to their excellent mechanical strength4,5 and moderate biocompatibility.6,7 However, because of the excessive elastic modules (over 100 GPa), implantation of Ti prosthesis has the potential risk of bone atrophy and reabsorption.8 In recent years, tantalum (Ta) with relative lower modulus (40−60 GPa),9 excellent corrosion resistance,10 good osseointegration,11,12 high surface friction,13 and impressive biocompatibility14 has become a promising alternative material in the implant field. In addition to mechanical requirements, the surface performance of the implant is also very important, especially for long-term antibacterial properties. At the current stage, implant-related infection is still a commonly disastrous problem in orthopedic surgery. It is a main cause for medical disputes and implant surgery failure. According to the previous reports, the incidence of implantrelated is approximately 0.9%.15 It is estimated to cost the U.S. healthcare system in excess of $1.8 billion per year.16 Unfortunately, in this aspect, neither the most commonly used Ti nor the emerging Ta possess significant antibacterial properties. Therefore, it is necessary to improve their © XXXX American Chemical Society
antibacterial properties through appropriate surface modification, so as to effectively reduce the risk of implant-related infection. The in vivo nanomaterials of zinc oxide is certified by the FDA, which can be used in the body.17 In the preliminary work of our group, we have studied the effects of different crystal planes,18 structures,19 and atomic arrangement20 of ZnO on the antibacterial properties. However, all of the above antibacterial studies were based on the single morphological structure of ZnO nanoarrays. In practical clinical applications, the releasing of antimicrobial components was preferred to be divided into two stages:21 at early stages (within 48 h), antimicrobial components expected a rapid releasing to kill the bacteria around the implant; at a second stage (over 2 weeks), it required a slow releasing to realize long-term inhibition. However, the single morphological structure of ZnO nanoarrays was insufficient to meet the above requirements. Received: January 14, 2017 Accepted: April 3, 2017 Published: April 3, 2017 A
DOI: 10.1021/acsami.7b00666 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
middle layer was the trunk-shaped ZnO nanorods, and the top layer was the leaves-like ZnO nanoslices. At the first stage (Figure 1C), the superficial ZnO nanoslices could be released quickly as the falling leaves, so as to effectively sterilize the surrounding environments during the initial implanting period. Afterward (Figure 1D), the middle layer of ZnO nanorods dissolved slowly to achieve low-dosage and long-term inhibition.
To realize the two-stage release mentioned above, dual layered ZnO nanorods−nanoslices hierarchical structures (NHS) were constructed, for the first time, to modify the surfaces of Ti and Ta, respectively. As show in Figure 1A, the
■
RESULTS AND DISCUSSION The morphology of the samples was observed by FE-SEM. Figure 2A,G show dense coverage of ZnO nanoslices arrays on the surface of Ti and Ta, respectively. The ZnO nanorods arrays are exhibited in Figure 2C,I. Figure 2E,K, exhibited the two layered structure of ZnO NHS. The bottom of the hierarchical structure was ZnO nanorods arrays. The density of the ZnO nanorods arrays could be controlled by adjusting the initial concentration of zinc acetate dihydrate, which acted as the precursors of ZnO. The sparse ZnO nanorod arrays could be obtained when we used half a dose of zinc acetate dihydrate solution (Figure S3). The top of the hierarchical structure was ZnO nanoslices arrays. To evaluate the stability of two different nanoarrays, ultrasound treatment was then applied to these samples. As a result, most of the ZnO nanoslices were removed (Figure 2B,H), suggesting the quickly dispersing characteristic of these materials. Under the same treatment, only a small part of the ZnO nanorods was removed (Figure 2D,J). These results indicated that the ZnO nanorod array was more firmly adhered to the substrate than ZnO nanoslices. One unexpected interesting discovery was that, after ultrasound treatment, the preserving rate of ZnO nanoslices on the ZnO NHS (Figure 2F,L) was significantly higher than its counterpart on the naked substrate (Figure 2B,H). The detailed mechanism of this phenomenon was still unknown, so we assumed that the middle layer of ZnO nanorod arrays could act as a “mini-lawn” which provided a certain cushion effect. These stability test results were further confirmed by the mass variation of the substrate surface (Figure S1). After ultrasound treatment, the ZnO
Figure 1. (A) Idea of the proposed dual layered ZnO NHS was inspired from falling leaves. The relative 3D schematic diagrams: (B) in the bottom layer, substrate plate represents Ti or Ta; in the middle layer, the trunk-shaped cylinder represents ZnO nanorods; and in the top layer, the leaves-like slices represents ZnO nanoslices. The twostage release: (C) early rapid releasing stage and (D) second slow releasing stage.
idea of the proposed dual layered ZnO NHS was inspired from the falling leaves. In the relative 3D schematic diagrams (Figure 1B): the bottom layer was the Ti or Ta substrate plate, the
Figure 2. SEM photos of ZnO nanoslices (A, B, G, and H), ZnO nanorods (C, D, I, and J), and ZnO NHS (E, F, K, and L) were observed before and after ultrasound. The images were processed by pseudo color, each color represented different composition: green (Ti or Ta), purple (ZnO nanoslices), and pink (ZnO nanorods). B
DOI: 10.1021/acsami.7b00666 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces nanoslices groups had highest mass alter from 7.3 ± 0.4 mg/L (Ti) and 7.33 ± 0.23 mg/L (Ta) to 0.43 ± 0.17 mg/L (Ti) and 0.33 ± 0.17 mg/L (Ta). The mass variation of ZnO nanoslices was obviously higher than ZnO nanorods. As the combination of these two types of nanoarrays, ZnO NHS thus possessed a two-stage release behavior, which was further demonstrated through the releasing curve by using inductive coupled plasma (Figure S2). The amounts of released Zn were 0.64 ± 0.1 mg/L (Ti) and 0.73 ± 0.11 mg/L (Ta) in the first 2 days and decrease quickly afterward. In the first 6 days, the Zn release on Ti declined slowly, whereas it declined more sharply on Ta. The detailed mechanism of this interesting phenomenon was still unclear. One possible explanation could be the differential adhesion between the two substrates (Ti and Ta) to the ZnO nanorod arrays. After 14 d, the releasing of zinc tended to be stable at 0.05 ± 0.02 mg/L. In orthopedic clinical applications, the early stage of the orthopedic implant-related infection rate was much higher than the late.22 ZnO NHS with two-stage release properties could meet two crucial clinical requirements of quick sterilization and long-lasting inhibition simultaneously. Figure 3A shows a typical transmission electron microscopy (TEM) image of the ZnO nanoslices, in which a large number of leaf-like structures can be observed. Figure 3B was the TEM image of a ZnO nanorod. Atomic-resolution high angle annular dark field (HAADF) images of two types of ZnO nanomaterials are provided in Figure 3C,D. Peaks from X-ray diffraction (XRD) demonstrated the formation of three different ZnO nanoarrays on the surface of Ti or Ta substrate, separately (Figure 3E and 3F). The antibacterial activities of ZnO nanorod arrays, ZnO nanoslices arrays, and ZnO NHS contact with the bacterial in different time were subsequently investigated. As show in Figure 4A,D, ZnO nanoslices (98 ± 1%) and ZnO NHS (99 ± 1.1%) had higher antibacterial rates in 8 h for Escherichia coli (E. coli) in Ti, compared to the ZnO nanorods (13.4 ± 6.6%). The results showed that ZnO nanoslices had a better performance for antibacterial treatment. However, after cocultured with bacteria for 1 day, the antibacterial ratio of ZnO nanorods increased to 81.3 ± 6.2%, which was very close to the level of ZnO nanoslices (82.2 ± 9.3%) and ZnO NHS (87.4 ± 7.2%) in Figure 4B,E. On the other hand, after ultrasound treatment, the ZnO nanorods (98.2 ± 3.6%) and ZnO NHS (90.5 ± 5.9%) groups still maintain a high antibacterial property. As a comparison, the performance of ZnO nanoslices groups declined sharply (Figure 4C,F). The same results were obtained from the groups of Ta. For Staphylococcus aureus (S. aureus), the results were similar. The ZnO NHS had better antibacterial activity to S. aureus. This is a valuable discovery, because in the actual clinical application, a majority of the implant-related infections were caused by S. aureus.23 Collectively, the above data further proved that the ZnO nanoslices could be quickly released to kill bacteria effectively. In contrast, the ZnO nanorods had better stability but needed more time to achieve antibacterial capacity. ZnO NHS, with the advantages of both, was more suitable for practical implant modification, such as Kirschner wire, fracture fixed plate, and screw. The incidence of implant-related infection may be greatly reduced if the implants with ZnO NHS have been widely used in orthopedics. Animal tests were then carried out to evaluate the practical performance of ZnO NHS modification. To increase the ratio of infection, eight different implantsnaked Ti implant (Ti), naked Ta implant (Ta), Ti implant with ZnO nanoslices
Figure 3. Typical TEM and atomic-resolution HAADF images of the ZnO with two different morphologies, (A and C) ZnO nanoslices and (B and D) ZnO nanorods. XRD spectra of the two substrate with three different ZnO coatings (E and F).
modification, Ta implant with ZnO nanoslices modification, Ti implant with ZnO nanorods modification, Ta implant with ZnO nanorods modification, Ti implant with ZnO NHS modification, and Ta implant with ZnO NHS modification were prepared and soaked in the solution of S. aureus before surgery. After 2 weeks, these implants were taken out from mice for further study. According to the plate counting method (Figure 5A), the antibacterial effects of implants modified with ZnO NHS were obviously better than the naked. Apart from the implant surfaces, the surrounding areas were also effectively sterilized by this two layered NHS (Figure 5B). In orthopedic surgery, the adhesion of bacteria on the surface of implants would have occurred if surgical operators did not have a strict C
DOI: 10.1021/acsami.7b00666 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 4. Antibacterial studies of different ZnO coatings against E. coli and S. aureus under different times and conditions in vitro, coculture for 8 h (A and D) and 24 h (B and E) (**p < 0.01 vs Ti-ZnO nanorods; ##p < 0.01 vs Ta-ZnO nanorods), (C and F) the relative data of cocolture for 48 h after ultrasound treatment (US) (**p < 0.01 vs Ti-ZnO nanoslices; ##p < 0.01 vs Ta-ZnO nanoslices).
Figure 5. (A) The optical diagram of four groups of mice after implant surgery. Two weeks after surgery, four corresponding plate counting photos were arranged in the bottom to show the bacterial number on the surfaces of each implant. (B) Antibacterial activity of the ZnO NHS modified Ti and Ta on the surface and surrounding, 2 weeks after surgery.
sterile concept or surgery time had been extended.24 The as prepared ZnO NHS coating can remarkable reduces this risk, and thus prevents the abuse of antibiotics. To obtain more detailed information, the tissues (skin and muscle) around the implant materials were stained by the hematoxylin and eosin method. As shown in Figure 6, the dark blue dots represented the nucleus in skin and muscle. From the blank control groups (Ti or Ta), we can observed the normal nucleus in skin and muscle (Figure 6A,B,K,L). The surroundings of ZnO nanoslices (Figure 6C,D,M,N), ZnO nanorods (Figure 6E,F,O,P), ZnO NHS (Figure 6G,H,Q,R) implant materials, after being soaked in a bacterial suspension of S. aureus, had a similar results. On the contrary, the tissues around the naked implant materials, which were treated with S. aureus solution, exhibited high-density nucleus of inflammatory cells in both skin and muscle samples (Figure 6I,J,S,T). More cleared evidence were provided with higher magnification (Figure 6i,j,s,t). Then, the morphology of the samples was again observed by the FE-SEM. The ZnO NHS group in Ti or Ta still had retained a lot of ZnO nanorods, compared to the ZnO nanorods group. Only a small part of ZnO had been left in the ZnO nanoslices group (Figure S4). Once again, this result proved that the ZnO NHS group was more stable. In summary, the ZnO NHS had excellent and durable antibacterial activity in
vivo. Most of the implant-related infection occurred within 1 week after surgery.25 The antibacterial activity of the ZnO NHS could last for more than 2 weeks in animal experiments, particularly suitable for application to modify the surface of the implant. Although the antibacterial activity of nanosized ZnO was excellent, as an implant coating for clinical application, it is equally important to test its biocompatibility. Cytotoxicity test is one of the most important indicators of the biocompatibility evaluation system for medical devices. In this study, the osteoblast (MC3T3-E1 cells) was used to test the cytotoxicity of Ti or Ta modified with different ZnO. No obvious cytotoxicity was observed from the ZnO nanoslice group, the ZnO nanorods group, and the ZnO NHS group, compared to the Ti or Ta control group (Figure S5). Moreover, according to the results of the control group, we also concluded that the biocompatibility of Ta was better than Ti. Therefore, Ta based implants, after being modified with ZnO NHS, could have broad application prospects in the future.
■
CONCLUSION Falling-leaves-inspired two layered ZnO NHS were constructed to modify the surface of Ti and Ta based implants. Compared with traditional single layered nanoarrays, ZnO NHS exhibited D
DOI: 10.1021/acsami.7b00666 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 6. Two weeks after coculture with S. aureus in vivo, the tissues (skin and muscle) around the implant materials were stained by the hematoxylin and eosin method. (A, B, K, and L) The blank control groups; (C, D, M, and N) implant materials modified with ZnO nanoslices; (E, F, O, and P) implant materials modified with ZnO nanorods; (G, H, Q, and R) implant materials modified with ZnO NHS; (I, J, S, and T) naked implant materials without ZnO, and high magnification of images (i, j, s, and t). of the specimens. The crystalline structure of the samples was determined by X-ray diffraction (XRD) and transmission electron microscope (TEM). Zn Release and Area Mass Alter. The amount of Zn released from ZnO NHS was investigated in phosphate buffered saline (PBS). The samples were immersed in 5 mL of PBS for 2 d and then retrieved to be immersed in fresh PBS of 5 mL for another 2 d. This process was repeated for a total 2 weeks. The amounts of Zn leached to the PBS solutions were measured by inductively coupled plasma atomic emission spectrometry (ICP-AES). For the 1 cm2 samples, they were divided into three groups: ZnO nanoslices, ZnO NHS, and ZnO nanorods. The variation of the samples surfaces was measured by a thousandth leveled electronic balance. After ultrasonication for 2 s, we measured the samples mass again. Antibacterial Properties Assay in Vitro. E. coli and S. aureus were chosen to assess the antibacterial ability of these ZnO samples. E. coli and S. aureus were cultured in Luria−Bertani broth at 37 °C for 24 h and adjusted to a concentration of 107 CFU/mL. Then, 100 μL of bacterial stock solution was taken into 5 mL of Luria−Bertani broth, with control group (Ti or Ta), ZnO nanoslices, ZnO nanorods, and ZnO NHS incubated for 8 h, 1 day, respectively. The same new samples were ultrasonicated (100W) for 2 s. These new samples were cultured in the bacterial culture solution for 2 days in a similar manner. After that, 100 μL of the coculture media was taken out for 105 degrees of dilution and 50 μL of the diluents was used to coat on the Petri dishes which were placed in the constant temperature incubator (37 °C) for 24 h. The antibiotic potency of the samples with four different groups was compared by plate counting method. The antibacterial rates (R) for E. coli and S. aureus in the medium were calculated based on the following formula: R = (A − B)/A × 100%, where A was the number of bacteria in the control group and B was the number of bacteria in the experimental group.
unique two-stage release features, which integrated rapid sterilization and long-lasting inhibition against both Grampositive bacteria and Gram-negative bacteria. Nanosized ZnO has been well recognized as a low toxic and low cost material for in vivo usage. The proposed ZnO NHS could be a feasible way to reduce the risk of implant surgery failure caused by bacteria. In addition, the abuse of antibiotics could be also minimized accordingly on implant-related surgery. Ongoing research will be focused on the long-term safety of the ZnO NHS.
■
EXPERIMENTAL SECTION
Sample Fabrication. Ti and Ta foils (99.9% pure) 10 × 10 × 0.1 mm3 in size were ultrasonically cleaned in deionized water. 50 mM Zn(NO3)2·6H2O, 80 mM NH3·H2O, and 25 mM HMT were added to 400 mL of deionized water, 85 °C water bath for 24 h, and a ZnO solution was synthesized. A 0.2 mL drop of the solution was dropped on the Ti and Ta surfaces, 55 °C drying, designated as ZnO nanoslices. A NaOH solution (0.03 m) was added slowly to a solution of zinc acetate dihydrate (0.01 m) in methanol at 60 °C and stirred for 2 h, as a normal concentration of the Zn2+ group. Then a NaOH solution (0.03 m) was added slowly to a solution of zinc acetate dihydrate (0.005 m) in methanol at 60 °C and stirred for 2 h, as half a dose of zinc acetate dihydrate. The resulting solution (0.3 mL) was dropped on a Ti and Ta surface at 150 °C, until the solution was evaporated to dryness. Then, the sample was placed in 500 mL of deionized water with 25 mM Zn(NO3)2·6H2O and 25 mM HMT, 85 °C water bath for 24 h, designated as ZnO nanorods.26 The zinc oxide solution was dropped on the surface of ZnO nanorods, 55 °C drying to form as ZnO NHS. Surface Characterization. Field-emission scanning electron microscopy (FE-SEM) was utilized to examine the surface topography E
DOI: 10.1021/acsami.7b00666 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
ACS Applied Materials & Interfaces
■
In Vivo Infected Studies. All experiments were performed in compliance with the relevant laws and approved by the Institutional Animal Care and Use Committee at Institute of Translational Medicine, Nanchang University. Thirty male KM mice between 35 and 40 g were employed for subcutaneous implantation of biomaterials. The studies were performed in accordance with the surgical procedures which were performed under standard aseptic conditions. Preoperatively, mice were anesthetized by intraperitoneal injection of 10% chloral hydrate (0.005 mL/g). Then, the dorsal area of each mouse was shaved and disinfected and a 2−4 cm midline incision was created. The sterilized 1 cm2 samples (Ti, Ta, Ti-ZnO nanoslices, Ta-ZnO nanoslices, Ti-ZnO nanorods, Ta-ZnO nanorods, Ti- ZnO NHS, and Ta- ZnO NHS) were soaked in a bacterial suspension of S. aureus at 107CFU/mL into the subcutaneous. Two groups of samples (free Ti and Ta) were soaked with physiologic saline solution (0.9% NaCl) into the subcutaneous as the blank control groups. Postoperatively, the mice were subcutaneously administered physiologic saline solution (1−2 mL/g) for fluid replacement. After recovery, the mice were housed in pairs and allowed to move in their cages without restriction. They were fed with commercial mice chow and water ad libitum. Two weeks after surgery, the mice were euthanized with carbon dioxide asphyxiation. The implanted material and surrounding secretions were removed and placed in 5 mL of physiologic saline solution. After that, 100 μL of the coculture media was taken out for 104 degrees of dilution, and 50 μL of the diluents was used to coat on the Petri dishes which were placed in the constant temperature incubator (37 °C) for 24 h. The antibiotic potency of the samples was compared by plate counting method. Then the morphology of the samples was again observed by the FE-SEM. The tissues (skin and muscle) around the implant materials were removed immediately and fixed in 4% neutral buffered formalin (Chempur, Poland), for paraffin embedding. The paraffin embedded tissues were cut into 5 mm serial sections, stained by the conventional hematoxylin and eosin method. Histological sections were examined under an Olympus BX41 microscope equipped with digital camera for capturing images. Cytotoxicity. MC3T3-E1 cells were seeded on 1 cm2 samples (Ti, Ta, Ti-ZnO nanoslices, Ta-ZnO nanoslices, Ti-ZnO nanorods, TaZnO nanorods, Ti- ZnO NHS, and Ta- ZnO NHS) placed in 24-well plates (Corning, USA) at a density of 104 per well, and cultured at 37 °C under a 5% CO2 atmosphere in Dulbecco’s Modified Eagle’s Medium (DMEM, Gibco, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, USA) and 1% penicillin/streptomycin. After 7 days of cell culture, CCK-8 (Dojindo, Japan) solution was added to each well for a further 1 h incubation to detect the cell viability. The medium was transferred to a new 96-well cell culture plate, and its optical density (OD) was measured by a spectrophotometer (Molecular Devices, USA) at a wavelength of 450 nm. The cell viability (%) was calculated as follows: Cell viability (%) = OD450 nm in test cells/OD450 nm controlcells × 100%. Statistical Analysis. The experiments were conducted in triplicate, and the data were expressed as means standard deviations. The oneway ANOVA combined with the Student−Newman−Keuls (SNK) post hoc test was applied to determine the level of significance. Significance was considered for p values
