Multiple-Responsive Mesoporous Silica Nanoparticles for Highly

5 days ago - The in vitro data exhibited that our nanocarriers may provide an effective strategy for accurate cancer treatment. ... Fluorescence–Flu...
0 downloads 12 Views 6MB Size
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article Cite This: ACS Omega 2018, 3, 4306−4315

Multiple-Responsive Mesoporous Silica Nanoparticles for Highly Accurate Drugs Delivery to Tumor Cells Ronghua Jin,†,§ Zhongning Liu,‡,§ Yongkang Bai,† Yongsheng Zhou,‡ and Xin Chen*,† †

School of Chemical Engineering and Technology, Shaanxi Key Laboratory of Energy Chemical Process Intensification, Institute of Polymer Science in Chemical Engineering, Xi’an Jiaotong University, Xi’an 710049, P. R. China ‡ Department of Prosthodontics, National Engineering Laboratory for Digital and Material Technology of Stomatology, Beijing Key Laboratory of Digital Stomatology, Peking University School and Hospital of Stomatology, Beijing 100081, P. R. China S Supporting Information *

ABSTRACT: A core−shell nanocarrier with triple layers, where each layer is sensitive to one specific physiological stimulus, has been fabricated for highly accurate cancer therapy. The nanocarrier consists of mesoporous silica nanoparticles (core structure for drug loading), fluorescein isothiocyanate-labeled hyaluronan (FITC−HA, first shell for imaging with enzymatic response), disulfide bond-embedded silica (SiO2, second layer with glutathione response), and switchable zwitterionic surface (third layer with pH response). The nanocarrier decorated with zwitterionic surface is able to offer long blood circulation time due to the weak nonspecific protein absorption. After these nanocarriers were gradually gathered around tumor cells through enhanced permeability and retention effect, the zwitterionic surface could switch to positive charge in low-pH environment, which was in favor of cellular uptake due to the strengthened positive nanocarrier− negative cellular membrane interaction. Once internalized into tumor cells, the high concentration of glutathione in cytoplasm could cleave disulfide bonds to remove the SiO2 shell and the HA layer would be exposed, which would be further degraded by hyaluronidase to trigger payload release. The fluorescent spectrum and images reveal that both glutathione and hyaluronidase are required for the release of preloaded drugs from these nanocarriers. By employing the multiple response, our nanocarriers could achieve effective antibiofouling ability while maintaining enhanced cellular internalization and targeted drug delivery, resulting in preferred cancer cell cytotoxicity, which is much higher than that of free doxorubicin. The in vitro data exhibited that our nanocarriers may provide an effective strategy for accurate cancer treatment.



INTRODUCTION Nanocarrier for drug delivery has attracted increasing interest in localized chemotherapy, offering potential solutions to decrease adverse effects and improve therapeutic efficacy for tumor therapy.1−3 Typically, mesoporous silica nanoparticles (MSNs), which show tremendous advantages, such as large surface area, abundant surface functionalization sites, and ease of cell internalization, have been widely used for drug delivery.4−6 In addition, silica-based materials were considered to have good biocompatibility and biodegradability, which gradually degraded to orthosilicic acid, drained by the blood or lymphatic system, and finally excreted through the kidneys.7 This clear metabolism indicates no toxicity of the silica-based materials.8 However, an ideal MSN nanocarrier, which is able to (1) circulate in blood for long time, (2) selectively accumulate around tumor tissue, (3) effectively uptake by tumor cells, and (4) intracellularly release antitumor drug, has not been developed. Accordingly, “stealthy” poly(ethylene glycol)-modified MSNs with tumor-targeting moieties at terminal are rationally © 2018 American Chemical Society

designed to realize enhanced circulation of the nanocarriers and effective cancer cell internalization.9,10 However, the introduction of active targeting ligands, e.g., arginine-glycineaspartate peptide, only yields limited benefits to the two procedures mentioned above.11 As to further address this issue, our group developed an MSN with pH-responsive zwitterionic surface, which would adjust the antibiofouling ability of the nanocarriers to promote internalization to cancer cells after preferential aggregation around solid tumor by changing zwitterionic surface to positive charge.12 Nevertheless, the resultant positive charge on MSNs could inhibit the release of positively charged drugs, meanwhile absorb plenty of negatively charged drugs, resulting in an obvious limitation of these nanocarriers. Thus, MSNs covered by degradable shield with switchable zwitterionic surface, where the surface would “turn on” the antibiofouling property for prolonged blood circulation Received: March 7, 2018 Accepted: April 9, 2018 Published: April 18, 2018 4306

DOI: 10.1021/acsomega.8b00427 ACS Omega 2018, 3, 4306−4315

Article

ACS Omega

Scheme 1. Formation of FITC−HA, Disulfide Bond-Embedded Silica, and Zwitterionic Surface Multiple-Functionalized MSN (MSN−HA−SiO2−TSA/DMA), as Well as the Selective Cell Internalization, Targeted Drug Delivery, and Responsive Drug Release in Tumor Cell Based on This Nanocarrier

during circulation (alkaline environment) and provides efficient internalization into tumor cells (acidic environment)23,28 due to the pH-induced charge reversal. When the nanoparticles are around healthy cells (HaCaT cells, pH = 7.4), the zwitterionic surface terminal with −COO− and −N+(Me)3 would offer high resistance to cell endocytosis. The net charge of the nanoparticles would switch from zwitterionic to positive charge in the acidic environment surrounding tumor cells (e.g., HeLa cells) owing to the cleavage of the negatively charged −COO− group and the formation of positively charged NH3+. The positive nanoparticles are easy to bind with tumor cell membrane, which has negative charge, leading to preferential uptake of these nanoparticles by tumor cells. The MSN’s core serves as a container for efficient loading of antitumor drug, whereas HA and disulfide bond are in charge of intracellular drug release, due to the degradation triggered by high concentration of hyaluronidase (HAase) and glutathione (GSH) in tumor cells.29,30 Moreover, the introduction of FITC results in the ability of real-time drug tracing and selective tumor cell imaging.

while losing the antibiofouling ability in solid tumor, and then the degradable shield being easily removed in tumor cells to trigger the payload release and eliminate the influence of surface charge, are required. Various degradable shields with stimuli responsiveness to specific triggers, including pH,13−16 light,17 redox,18−20 magnetic field,21 light-enzyme,22 pH-enzyme,12,23−25 pHredox,26 and pH-adenosine triphosphate,27 have been successfully introduced to MSNs to ensure payload release only in given cell types for optimal cancer therapy. However, the drug release from these systems either only relied on single stimulus or required external triggers to assist. The former may cause certain amount of drug release at unexpected location, whereas the latter may complicate the chemotherapy. In addition, the combination of the stimuli-responsive cleavable shield with switchable zwitterionic surface to MSNs has not been reported yet. Inspired by our previous work, we intend to fabricate a drug-delivery capsule covered by shield with multiple functions, which simultaneously offer the prolonged blood circulation time, efficient tumor internalization, and stimuli-responsive drug release. In addition, these stimuli as mentioned above originate from the biochemical pathology so that the drug release could only happen in disease place and at designed time. In view of the above, we fabricated a nanocomposite by mesoporous silica nanoparticles, fluorescein isothiocyanatelabeled hyaluronan (FITC−HA), disulfide bond-embedded silica (SiO2), and pH-responsive zwitterionic surface terminal with −COO− and −N+(Me)3 for tumor therapy and imaging, which is denoted as MSN−HA−SiO2−N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride (TSA)/2,3-dimethylmaleic anhydride (DMA) (Scheme 1). The pH-responsive zwitterionic layer prevents nonspecific protein adsorption



RESULTS AND DISCUSSION MCM-41-type mesoporous silica nanoparticles (MSNs) were fabricated through a sol−gel approach catalyzed by base using cetyltrimethylammonium bromide (CTAB) as structure-directing agent. To fabricate the MSN−HA−SiO2−TSA/DMA nanocarriers, the resulting MSNs were derivatized with 3aminopropyltriethoxysilane (APTES) to introduce the amine groups (MSN-NH2). Then, the HA shell was formed on the exterior surfaces of MSN-NH2 through strong Coulombic interaction and hydrogen bonds between amino groups and HA (MSN−HA). The plenty of hydroxyls on HA were not 4307

DOI: 10.1021/acsomega.8b00427 ACS Omega 2018, 3, 4306−4315

Article

ACS Omega

Figure 1. Transmission electron microscopy images of (a) amino group-functionalized MSNs (MSN-NH2), (b) FITC−HA-fabricated MSN-NH2 (MSN−HA−FITC), (c) breakable organosilica matrix-covered MSN−HA−FITC (MSN−HA−SiO2), and (d) switchable zwitterionic surfacemodified MSN−HA−SiO2 (MSN−HA−SiO2−TSA/DMA).

only used to introduce fluorescein isothiocyanate (MSN−HA− FITC), but also serve as reaction points for further sol−gel process with tetraethyl orthosilicate (TEOS) and bis[3(triethoxysilyl)propyl]disulfide to generate the silica shell with disulfide bond embedding (MSN−HA−SiO2).31 The resulting MSN−HA−SiO2 was then modified by N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride (TSA) and APTES with a ratio of 1:1, following the conversion from amino groups to carboxyl group by reacting with 2,3-dimethylmaleic anhydride (DMA) to form the final zwitterionic surface terminal with −COO− and −N+(Me)3, which was denoted as MSN−HA−SiO 2−TSA/DMA. The resultant MSN-NH 2, MSNs−HA−FITC, MSN−HA−SiO2, and MSN−HA−SiO2− TSA/DMA were characterized by transmission electron microscopy (TEM). As shown in Figure 1a, the as-synthesized MSN-NH2 has a round shape with diameter of about 150 nm and an ordered arrangement of lattice array. Figure 1b obviously shows 5 nm thick organic layers on the MSNs, demonstrating the successful immobilization of HA. After silica coating, the shell increased to about 25 nm (Figure 1c), which has no obvious change after zwitterionic layer formation (Figure 1d). A similar contrast between the new shell and the MSN core verified the successful generation of silica layer on MSNs. The size distributions of MSN-NH2, MSNs−HA−FITC, and MSN−HA−SiO2−TSA/DMA were also investigated by dynamic light-scattering measurements. As shown in Figure S1, the size of MSN significantly increased from 100 to 160 nm after the formation of hydrophilic HA shell, which presents larger radius of hydration than its real size. The further

enhancement of the nanoparticle size to about 180 nm was observed after relative hydrophobic SiO2 modification, demonstrating the stepwise synthesis of MSN−HA−SiO2− TSA/DMA. The stepwise synthesis and functionalization of MSN−HA− SiO2−TSA/DMA were also investigated by IR spectra, ζpotential measurement, thermogravimetric analysis (TGA), and fluorescence spectrum (Figures S2−S5), which, respectively, present (1) characteristic covalent bond vibrations of HA, TSA, and DMA; (2) different charges of various MSNs in agreement with the electrical properties of their grafted groups (amine, HA, SiO2-OH, ammonium, and carboxyl); (3) exact weight percentage of each component; and (4) the characteristic emission peak of FITC at about 520 nm, revealing the successful formation of the MSN−HA−SiO2−TSA/DMA core−shell nanocarrier with triple shells. The pH-responsive zwitterionic property of MSN−HA− SiO2−TSA/DMA was investigated by ζ-potential measurements in phosphate-buffered saline (PBS) at different pH values. Figure 2a shows that the charge on the surface of the nanocarriers is neutral at pH 7.4 after co-functionalization of zwitterion (TSA and DMA), whereas the surface charge returns to the primary value without DMA modification (positive charge) because of the acid-triggered cleavage of linkages among amines and DMA.32 The zwitterionic surface and positive surface showed opposite performance for nonspecific protein resistance, where the protein absorption percentage of MSN−HA−SiO2−TSA/DMA at pH 6.8 (83%) is about 7 times larger than that of MSN−HA−SiO2−TSA/DMA at pH 7.4 (12%), indicating that our MSN−HA−SiO2−TSA/DMA 4308

DOI: 10.1021/acsomega.8b00427 ACS Omega 2018, 3, 4306−4315

Article

ACS Omega

Figure 2. (a) ζ-Potential of MSN−HA−SiO2−TSA/APTES and MSN−HA−SiO2−TSA/DMA with equal amounts of the two functional groups (COO− and N+(Me)3) at pH 7.4 or 6.8 (n = 3). (b) UV−vis spectrum of bovine serum albumin (BSA) solution before and after adsorption by MSN−HA−SiO2−TSA/DMA at different pH values. (c) Thermogravimetric analysis of MSN-NH2, MSN−HA, and MSN−HA−SiO2−TSA/DMA before and after the treatment of GSH and HAase. (d) Fluorescence spectrum of MSN−HA−SiO2−TSA/DMA before and after the treatment of GSH and HAase.

Figure 3. (a) Glutathione (GSH) and hyaluronidase (HAase) multiple-dependent release kinetics of doxorubicin (DOX)-loaded MSN−HA−SiO2− TSA/DMA. (b) The corresponding UV−vis absorption spectra of DOX, FITC, and DOX-loaded MSN−HA−SiO2−TSA/DMA after GSH/HAase treatment.

could not only prevent biosystem clearance (pH 7.4),33 but also able to be preferentially captured and internalized by tumor cells (pH 6.8).34 The GSH-responsive cleavage of silica shell and the enzyme (HAase)-induced degradation of HA shell were investigated by thermogravimetric analysis (TGA, Figure 2c). As can be seen from this figure, the percentage of organic matter in MSN−HA−SiO2−TSA/DMA dropped from about 25 to 20% after the GSH treatment, and a further 15% of organic matter was lost after addition of HAase into the GSH solution, which come from the loss of TSA/DMA, organosilica matrices, and HA. Moreover, the TGA curve and the residual ratio of MSN−HA−SiO2−TSA/DMA after GSH and GSH/

HAase treatment, respectively, match the data of MSNs−HA and MSNs, indicating that most of silica shell and HA shell were effectively removed from MSN−HA−SiO2−TSA/DMA under appropriate condition. The dual stimuli-responsive decomposition results in the release of fluorescent probe (FITC), which could be used to track the metabolism and release procedure of preloaded drugs (Figure 2d). As a nanocarrier for biomedical applications, the stability of MSN−HA−SiO2−TSA/DMA under physiological condition is also important, which has been investigated by TEM after incubation in PBS and serum for 7 days (Figure S6). As can be seen from this image, there is no obvious change of the size and 4309

DOI: 10.1021/acsomega.8b00427 ACS Omega 2018, 3, 4306−4315

Article

ACS Omega

Figure 4. Fluorescence microscopy images of Hela cells (tumor cell) and HaCaT cells (normal cell) after incubation with free DOX and DOXloaded MSN−HA−SiO2−TSA/DMA for 1 and 12 h. The scale bar is 50 μm.

morphology of MSN−HA−SiO2−TSA/DMA after the incubation, which means that MSN−HA−SiO2−TSA/DMA is quite stable under physiological condition. Doxorubicin (DOX), a model anticancer drug, was selected to evaluate the drug loading and stimuli-responsive release of MSN−HA−SiO2−TSA/DMA. The drug-loading capacity and encapsulation efficiency of MSN-NH2 and MSN−HA−SiO2− TSA/DMA at different mass ratios of DOX/MSNs are available in Table S1. The DOX-loading capacity gradually raised to 31.2 and 24.9 wt % for MSN-NH2 and MSN−HA−SiO2−TSA/ DMA, respectively, when the mass ratio of DOX/MSNs increased to 1:2. Further increase of the DOX would not change the number, which indicates that the saturated loading capacities of MSN-NH2 and MSN−HA−SiO2−TSA/DMA for DOX are 31.2 and 24.9 wt %, respectively. The encapsulation efficiencies of MSN-NH2 and MSN−HA−SiO2−TSA/DMA showed peak values of 98.5 and 91.8 wt %, respectively, and then decreased due to overloading. These data show that both MSN-NH2 and MSN−HA−SiO2−TSA/DMA exhibit high drug-loading capacity, and the surface modification would cause a small drop of the drug-loading capacity and efficiency, due to the DOX leak from the nanoparticles during the functionalization. To evaluate the performance of MSNs−HA−SiO2−TSA/ DMA for real antitumor therapy, the stimuli-responsive release of drug from this nanocarrier was investigated under different solutions (PBS buffer in the presence of 10 mM GSH and 0.1 mg/mL HAase or GSH/HAase) to mimic the biosystem. The amount of DOX released from MSN−HA−SiO2−TSA/DMA was analyzed by fluorescence spectrometry through monitoring the fluorescence intensity of the DOX in the solution. As shown in Figure 3a, there is no visible DOX release from the nanocarrier before triggering even after 20 h incubation (pH 7.4). After GSH addition (silica shell removal from the nanocarrier by cleavage of disulfide bond), a slight leakage of DOX about 5% appeared, which could be attributed to small amount of physical absorption. However, the DOX release immediately appeared after further addition of HAase, owing to the enzyme-induced degradation of the HA layer, which ends in 40 h with over 90% DOX release. The corresponding fluorescence spectrum is displayed in Figure 3b, which shows the obvious characteristic peak of both DOX and FITC after the stepwise treatment of GSH and HAase. The release profiles

of DOX from MSN−HA−SiO2−TSA/DMA in the solution containing only HAase, only GSH, and HAase/GSH were also compared (Figure S7). As can be seen from this figure, only 3 and 8% of DOX were released from MSN−HA−SiO2−TSA/ DMA after up to 48 h incubation in PBS with only HAase and only GSH, respectively. It can be attributed to the full sealing of MSN by the antibiofouling property of TSA/DMA, SiO2 shell, and HA layer, which would effectively encapsulate DOX and prevent the DOX release. However, the obvious DOX release appeared after treatment by both HAase and GSH, which reach about 92% within 48 h incubation. These results not only show that GSH and HAase are all needed to induce DOX release from MSN−HA−SiO2−TSA/DMA, but also further verify the drug-tracking ability of our nanocarrier, due to the concomitant DOX and the FITC probe. As to further understand the necessity of each component of our design, the release profile of DOX from MSN−HA−SiO2− TSA/DMA with no disulfide bonds in the SiO2 shell was also investigated (Figure S8). Briefly, we used only 100 μL of TEOS to fabricate the silica shell for the control nanoparticle instead of using 30 μL of tetraethyl orthosilicate (TEOS) and 70 μL of bis[3-(triethoxysilyl)propyl]disulfide for the nanoparticle containing disulfide bonds. Figure S8 indicates that only very tiny amount of DOX appeared in the solution even after up to 48 h incubation in PBS with glutathione (GSH) and hyaluronidase (HAase), due to the complete block of DOX fully covered by SiO2 shell. However, the MSN−HA−SiO2−TSA/DMA with disulfide bonds presented sustained DOX release in 48 h and ended with over 90% of final release. These results indicate that the introduction of disulfide bonds to form a cleavable SiO2 shell is required for effective drug delivery. To further investigate the mechanism of the responsive drug release, the hydrodynamic size change of MSN−HA−SiO2− TSA/DMA after treatment by GSH and HAase was measured by dynamic light scattering (DLS) (Figure S9). As can be seen from this figure, the size of MSN−HA−SiO2−TSA/DMA decreased to about 125 nm after 48 h treatment with GSH and HAase, which is close to the size of MSN-NH2, indicating the complete removal of the SiO2 shell and HA shell. These results demonstrate that the DOX release comes from the decomposition of the SiO2 shell and HA shell. To confirm the integration behavior of the MSN−HA− SiO2−TSA/DMA for selective imaging and targeted drug 4310

DOI: 10.1021/acsomega.8b00427 ACS Omega 2018, 3, 4306−4315

Article

ACS Omega

Figure 5. Viability of Hela cells and HaCaT cells incubated with (a) doxorubicin (DOX)-loaded MSN−HA−SiO2−TSA/DMA and (b) free DOX with increasing incubation time and the same amount of DOX (0.1 μg/mL).

observed even after up to 24 h incubation for the group of irresponsive zwitterionic nanoparticles. Moreover, the nucleus is pure blue and all nanoparticles are orange or yellow, which is a mixture of red DOX and green MSN−HA−SiO2−TSA/ DMA, indicating all DOX has been well encapsulated in the nanoparticles. It is obvious that the irresponsive zwitterionic nanoparticles could not enter the cytoplasm, where the high concentration of glutathione and hyaluronidase would quickly decompose these nanoparticles to release DOX.25 The tiny amount of fluorescent nanoparticles appeared in the image could be attributed to physical deposition during incubation. However, strong red fluorescence (DOX) appeared in the Hela nucleus only in 24 h incubation, accompanied by plenty of green dots (FITC-labeled nanoparticles) dispersed in the cytoplasm, demonstrating the effective uptake of nanoparticles and following DOX release. These results indicate that the zwitterionic nanoparticles would offer high resistance to cell endocytosis, whereas the positive nanoparticles are easy to uptake by cell. The data of cellular uptake of DOX-loaded MSNs before TSA and DMA modification (MSN−HA−SiO2) are provided in Figure S12. As shown in the figure, plenty of red DOX appeared in the Hela cell nucleus within 12 h incubation, accompanied by strong green fluorescence (FITC) dispersed in the cytoplasm, whereas the intensities of both the red and green fluorescences are all weaker than those of the group of MSN− HA−SiO2−TSA/DMA. These results indicated that MSN− HA−SiO2 could be taken by cancer cells and then release DOX, but the uptake efficiency is relatively lower than that of MSN− HA−SiO2−TSA/DMA. 3-(4,5-Dimethyl-thiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays were applied to evaluate the tumor therapy efficiency of MSN−HA−SiO2−TSA/DMA because the selective cellular internalization and stimuli-triggered drug release were expected to show high performance and specific cytotoxicity only to tumor cells. Thus, the in vitro cytotoxicities of Hela cells and Hacat cells were collected after co-culture with DOX or DOX-loaded MSN−HA−SiO2−TSA/DMA with the same amount of DOX (0.1 μg/mL) against different periods of time (Figure 5). As can be seen from the figure, pure DOX exhibited equally low cytotoxicity to either normal or tumor cells. Nevertheless, using MSN−HA−SiO2−TSA/DMA to deliver DOX can enhance the cytotoxicity of DOX to Hela cells with only 25% cell viability after 48 h incubation. In addition, the in vitro anticancer capacity of DOX-loaded MSN−HA−SiO2−TSA/DMA to Hela cells was 3-fold higher

delivery, DOX-encapsulated nanocarriers were incubated with Hela (human cervical carcinoma cell) and HaCat (human keratinocyte cell) and then a fluorescence microscope was used to investigate endocytosis of the nanocarriers and DOX release in both Hela and HaCat cells (Figures 4 and S10). As shown in these figures, yellow fluorescence (merging of the green FITC and red DOX) is strongly localized in the Hela cells after 1 h incubation and no obvious DOX fluorescence (red) was observed in both cytoplasm and nucleus, indicating the effective drug-tracking property, enhanced tumor-selective ability, and negligible premature release. Extending the incubation time to 12 h makes more DOX enter into the nucleus of Hela cells, accompanied by plenty of green dots (FITC) dispersed in the cytoplasm. The disassembly of MSN−HA−SiO2−TSA/DMA to green and red parts could be attributed to the stepwise decomposition of the disulfide bond-containing SiO2 shell and the hyaluronic acid layer by the high concentration of glutathione and hyaluronidase in cytoplasm, which would expose the MSN core and trigger the DOX release. The DOX would quickly enter the nucleus owing to the DOX’s natural property,25 whereas the FITC bound to hyaluronic acid fragment would be located in the cytoplasm, resulting in plenty of green fluorescence in the cytoplasm and red fluorescence only in the nucleus. The persistent green fluorescence in tumor cells would serve as fluorescent probe for specific cell imaging. As particle-associated DOX would show concurrent green and red fluorescence, which could not pass the nuclear pore owing to size mismatch, the separation of green and red fluorescence as well as the existence of DOX in nucleus lead us to believe that DOX has been released from the MSN−HA−SiO2−TSA/DMA.35 For HaCaT cells, only weak red fluorescence and few green dots appeared in HaCat cells even after 24 h incubation, demonstrating the tumor-targeted DOX delivery and selective tumor cell imaging by MSN−HA− SiO2−TSA/DMA. As a comparison, the internalization behavior of free DOX to Hela cells and HaCaT cells was also investigated, exhibiting similar low internalization percentages in these two types of cells. All of these results indicated the integration behavior of the MSN−HA−SiO2−TSA/DMA for selective imaging and targeted drug delivery. As to further verify our design, DOX-loaded irresponsive zwitterionic MSNs, DOX-loaded positive MSNs, and DOXloaded MSNs before TSA and DMA modification (MSN− HA−SiO2) were also used as controls to investigate the Hela cell uptake (Figures S11 and S12). As can be seen from Figure S11, only a tiny amount of fluorescent nanoparticles could be 4311

DOI: 10.1021/acsomega.8b00427 ACS Omega 2018, 3, 4306−4315

Article

ACS Omega

without further purification. GSH was obtained from Shanghai Macklin Biochemical Co., Ltd., and HAase was purchased from Sigma-Aldrich. Preparation of Amino Group-Functionalized Mesoporous Silica Nanosphere through Disulfide Linkage (MSN-NH2). MCM-41-type MSNs were fabricated through a sol−gel approach catalyzed by base using CTAB as structuredirecting agent. First, 52.8 mL of NH3·H2O (29 wt % NH3 in water) was added to 1000 mL of deionized water containing 1.12 g of CTAB with magnetic stirring. After that, 5.8 mL of tetraethyl orthosilicate was added with rapid stirring, followed by heating up to 50 °C for 2 h. Subsequently, the mixture was placed overnight and then purified by repeating centrifuge and water/ethanol wash. To obtain the MCM-41-type MSNs, we used acidic methanol (9 mL of HCl/400 mL of methanol) to extract the surfactant templates at 70 °C for 36 h, which were then dried under vacuum for 20 h after completely washed with water and ethanol. The as-synthesized MSNs were suspended in 50 mL of anhydrous ethanol containing APTES (1 mL, 50% in ethanol), followed by stirring for 12 h. The resulting powder was dried under vacuum to get MSN-NH2 after completely washed with water and ethanol. Drug Loading and FITC−HA Shell Formation on MSNNH2. To load doxorubicin (DOX) into MSN-NH2 and introduce the hyaluronic acid (HA) shell on the surface as enzyme-degradable cap, 10 mL of absolute ethanol was used to disperse 200 mg of MSN-NH2 and 0.5 g of doxorubicin. After 24 h stirring at room temperature, 10 mL of HA solution (5% in water) was added into the suspension and agitated for another 8 h. The resulting powder was dried under vacuum after several times washing with water and ethanol to get DOXloaded MSN−HA. Furthermore, the DOX-loaded MSN−HA was then suspended in 50 mL of absolute ethanol containing 0.1 g of fluorescein isothiocyanate (FITC) under vigorous stirring. The mixture was agitated for 3 days, following by filtration and washing with copious ethanol by a fritted funnel. To achieve the complete removal of the unbound FITC, persistent washing was performed until there is no visible color in the eluant. The resulting MSN−HA−FITC was allowed to dry at room temperature overnight. Formation of Disulfide Bond-Embedded Silica Shell on MSN−HA−FITC. The formation of disulfide bondembedded silica shell on MSN−HA−FITC was exactly prepared according to our previous work,36 the only difference is using MSN−HA−FITC to replace protein. Surface Functionalization of MSN−HA−SiO2. The resulting MSN−HA−SiO2 powder was suspended in anhydrous ethanol. Then, N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride (TSA) and 3-aminopropyltriethoxysilane (APTES) were added with a mole ratio of 1:1. After 24 h stirring, the product (MSN−HA−SiO2−TSA/APTES) was dried overnight under vacuum and thoroughly washed with ethanol. The obtained MSN−HA−SiO2−TSA/APTES (200 mg) was suspended in dimethyl sulfoxide (6 mL) with 2,3-dimethylmaleic anhydride DMA (70.2 mg, 0.58 mmol). After 20 min deaeration by N2, triethylamine (0.2 mL) and pyridine (0.2 mL) were injected under N2 atmosphere. The suspension was then reacted for another 24 h and dried under vacuum after ethanol wash to get the zwitterionic mesoporous silica nanoparticle (MSN−HA−SiO2−TSA/DMA).

than that to HaCat cells after 48 h incubation. This is because of the different cellular microenvironments (acidity) around these two types of cells, which could either trigger or inhibit the switch of zwitterionic surface of MSN−HA−SiO2−TSA/DMA, resulting in disparate cellular uptake and toxicity. These data not only exhibit that MSN−HA−SiO2−TSA/DMA enhances the anticancer capability of encapsulated DOX, but also endow the encapsulated DOX with a highly selective cytotoxicity to cancer cells. The cytotoxicities of DOX-loaded MSN−HA−SiO2 to Hela cells and HaCat cells were explored as negative control. As shown in Figure S13, similar cytotoxicity to either Hela cells or HaCat cells was also observed due to the equal cellular uptake efficiency of MSN−HA−SiO2 to both cancer and normal cells. Moreover, the anticancer capacity of DOX-loaded MSN−HA− SiO2 to tumor cells is weaker than that of DOX-loaded MSN− HA−SiO2−TSA/DMA, which could be attributed to the enhanced endocytosis of the positive MSN−HA−SiO2−TSA/ DMA in the acid tumor environment. The cytotoxicity of MSN−HA−SiO2−TSA/DMA without drug loading was also evaluated by the MTT method (Figure S14). As shown in Figure S14, the pure MSN−HA−SiO2− TSA/DMA was nontoxic to either Hela or HaCaT cells even at 100 μg/mL and with 48 h incubation. These data further exhibit that (1) DOX is the source of the cytotoxicity of our system instead of nanocarriers and (2) the MSN−HA−SiO2− TSA/DMA exhibits safety and efficacy for fighting cancer even at high doses.



CONCLUSIONS In summary, we have fabricated an MSN−HA−SiO2−TSA/ DMA nanocarrier using MSNs (core for drug loading), FITC− HA (enzyme-degradable shell with fluorescent probe), disulfide bond-embedded silica (GSH disintegrable shell), and switchable zwitterionic surface (shell with pH-induced bioantifouling) for targeted tumor therapy and imaging. Owing to the multipleresponse-assisted drug-delivery strategy, the nonspecific absorption of these nanocarriers to proteins was greatly suppressed during blood circulation, whereas the cellular uptake would be enhanced in the acidic tumor microenvironment. After the selective internalization to tumor cells, the DOX inside the orifice of MSNs would be released in the cytoplasm due to the complete removal of the HA shell and SiO2 shell induced by the high concentration of GSH and HAase. Moreover, the whole procedure could be effectively tracked by the FITC probe with green fluorescence. These properties of high deactivation during blood circulation, preferential uptake by tumor cells, as well as intracellular drug release indicate that our MSN−HA−SiO2−TSA/DMA provided a promising prospect to achieve accurate therapy for tumor eradication.



MATERIALS AND METHODS Materials. Doxorubicin (DOX), glutathione (GSH), 3aminopropyltriethoxysilane (APTES), tetraethyl orthosilicate (TEOS), bis[3-(triethoxysilyl)propyl]disulfide, N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride (TSA), 2,3dimethylmaleic anhydride (DMA), cetyltrimethylammonium bromide (CTAB), fluorescein isothiocyanate (FITC), hyaluronic acid (HA, Mw = 10 kDa), and ammonium hydroxide were obtained from Sigma-Aldrich. All of the chemicals and reagents used in this study were of analytical grade and used as received 4312

DOI: 10.1021/acsomega.8b00427 ACS Omega 2018, 3, 4306−4315

Article

ACS Omega

DOX only and equivalent DOX-loaded MSN−HA−SiO2− TSA/DMA with a DOX concentration of 0.1 μg/mL for 6, 12, 24, and 48 h. Then, 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Sigma-Aldrich) was used to measure cell viability following the manufacture’s protocol. The absorbance of the wells at 570 nm was read through Varioskan Flash multimode reader (Thermo Fisher Scientific). Characterization. Transmission electron microscopy (TEM) images were collected on a Philips CM200 transmission electron microscope under the operation voltage of 200 kV. The TEM samples were obtained by dropping 5 μL of solution onto carbon-coated copper grids. All of the TEM images were visualized without staining. The IR spectra were measured by an AVATAR 320 FT-IR spectrometer. The ultraviolet−visible (UV−vis) spectra were measured by a Hitachi U-2910 spectrophotometer. All pH values were measured by a Sartorius BECKMAN F 34 pH meter. The ζpotentials were measured by a Delsa Nano C particle analyzer (Beckman Coulter) running Delsa Nano software and using 4 mW He−Ne laser operating at a wavelength of 633 nm and an avalanche photodiode detector. In vitro and intracellular releases were, respectively, monitored by fluorescence spectroscopy and confocal fluorescence microscopy using an Olympus BX51 microscope equipped with a fluorescent lamp; Ex = 488 nm, Em = 590 nm for DOX and Ex = 488 nm, Em = 520 nm for FITC.

Drug-Loading Capacity and Encapsulation Efficiency Measurements. Quantitative analyses of loaded DOX for MSN-NH2 and MSN−HA−SiO2−TSA/DMA were conducted by fluorescence spectroscopy at 590 nm (Em). The DOXloading capacity and encapsulation efficiency were obtained by the following equations loading capacity =

weight of drug in MSNs weight of drug loaded MSNs

encapsulation efficiency =

weight of drug in MSN initial weight of drug

pH-Induced ζ-Potential Change of MSN−HA−SiO2− TSA/DMA. MSN−HA−SiO2−TSA/DMA was suspended in 0.1 mg/mL PBS solution with pH 6.8 or 7.4, followed by different incubation times at 37 °C. Periodically, part of the solution was withdrawn and measured by a particle analyzer (Delsa Nano C System, Beckman Coulter). Measurements were performed 30 times for each sample, and the results were treated by Delsa Nano software version 2.31. Antibiofouling Property of MSN−HA−SiO2−TSA/ DMA. Bovine serum albumin (BSA, Sigma-Aldrich) was used as a model protein to investigate the antibiofouling property of MSN−HA−SiO2−TSA/DMA by protein adsorption measurements. MSN−HA−SiO2−TSA/DMA was first mixed with PBS solution (pH 6.8 or 7.4) containing BSA for 1 day incubation at 37 °C. Then, 200 μL of the solution was withdrawn after vortex to ensure homogeneity and centrifuged at 10 000g for 5 min to collect the aggregate of the protein-adsorbed MSNs. The residual protein was analyzed using UV−vis spectroscopy by measuring the peak signal at 280 nm. In Vitro Drug Release. DOX-encapsulated MSN−HA− SiO2−TSA/DMA nanocarriers were placed in simulated biosystem (i) PBS buffer with pH 7.4; (ii) PBS buffer with pH 7.4, GSH; (iii) PBS buffer with pH 7.4, GSH and HAase at room temperature. After that, 2 mL of supernatant was withdrawn at designed time point and analyzed by fluorescence spectroscopy after centrifugation to measure the content of DOX released from the nanocarriers (Em at 590 nm). Confocal Microscopy Analysis of Tumor Targeting and Selective Drug Delivery. To observe the selective tumor-targeting drug delivery of the DOX-loaded MSN−HA− SiO2−TSA/DMA, Hela cells (cancer cells) and HaCaT cells (normal cells) were employed. The cells were first seeded at 24well plates containing glass coverslips with a concentration of 2.5 × 104 per well. Then, these cells were cultured in Dulbecco’s modified Eagle’s medium (Invitrogen) containing 10% fetal bovine serum (Gibico) and penicillin−streptomycin (100 U/mL and 100 μg/mL, Gibico), followed by 1 day incubation at 37 °C under 5% CO2 atmosphere. On the following day, DOX-loaded MSN−HA−SiO2−TSA/DMA containing 0.1 μg/mL DOX and/or 0.1 μg/mL free DOX was added to the cells. The cells were stained with 4′,6diamidino-2-phenylindole (Life Technologies). Then, the following procedures were conducted: 1 and 12 h incubation, PBS rinsing, 4% paraformaldehyde fixing, and permeabilization in 0.1% Triton X-100. Afterward, the cells were rinsed and mounted and the fluorescence was recorded by a fluorescence microscope (Olympus BX51, Olympus, Japan). In Vitro Cytotoxicity Analysis. Hela cells and HaCaT cells were seeded in a 96-well plate with a concentration of 3 × 103 per well for 24 h before treatment. These cells were exposed to



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.8b00427. (1) DLS results of different functionalized MSNs (Figures S1 and S9); (2) FT-IR spectra (Figure S2); (3) ζ-potential (Figure S3); (4) TG analysis (Figure S4); (5) fluorescence spectrum (Figure S5); (6) microscopy images (Figure S6); (7) stimuli-responsive release (Figures S7 and S8); (8) cellular uptake experiment (Figures S10−S12); (9) cell viability (Figures S13 and S14); (10) DOX-loading capacity and encapsulation efficiency (Table S1) (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Yongsheng Zhou: 0000-0002-4332-0878 Xin Chen: 0000-0002-1224-0285 Author Contributions §

R.J. and Z.L. contributed equally to this work.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (81601606 to X.C. and 81400498 to Z.L.), the “Young Talent Support Plan” of Xi’an Jiaotong University (X.C.), the Technology Foundation for Selected Overseas Chinese Scholar of Shaanxi Province (X.C.), the Fundamental Research Funds for the Central Universities (2016qngz02 to X.C.), the One Hundred Talents Program of Shaanxi Province (X.C.), the National Natural Science 4313

DOI: 10.1021/acsomega.8b00427 ACS Omega 2018, 3, 4306−4315

Article

ACS Omega

Organosilica for Redox-Responsive Gene Delivery and Synergistic Cancer Chemotherapy. Adv. Mater. 2016, 28, 1963−1969. (20) Huang, P.; Chen, Y.; Lin, H.; Yu, L.; Zhang, L.; Wang, L.; Zhu, Y.; Shi, J. Molecularly organic/inorganic hybrid hollow mesoporous organosilica nanocapsules with tumor-specific biodegradability and enhanced chemotherapeutic functionality. Biomaterials 2017, 125, 23− 37. (21) Giri, S.; Trewyn, B. G.; Stellmaker, M. P.; Lin, V. S. Y. Stimuliresponsive controlled-release delivery system based on mesoporous silica nanorods capped with magnetic nanoparticles. Angew. Chem., Int. Ed. 2005, 44, 5038−5044. (22) Chen, X.; Liu, Z.; Parker, S. G.; Zhang, X.; Gooding, J. J.; Ru, Y.; Liu, Y.; Zhou, Y. Light-Induced Hydrogel Based on Tumor-Targeting Mesoporous Silica Nanoparticles as a Theranostic Platform for Sustained Cancer Treatment. ACS Appl. Mater. Interfaces 2016, 8, 15857−15863. (23) Chen, X.; Soeriyadi, A. H.; Lu, X.; Sagnella, S. M.; Kavallaris, M.; Gooding, J. J. Dual Bioresponsive Mesoporous Silica Nanocarrier as an “AND” Logic Gate for Targeted Drug Delivery Cancer Cells. Adv. Funct. Mater. 2014, 24, 6999−7006. (24) Chen, X.; Liu, Z. A pH-Responsive Hydrogel Based on a Tumor-Targeting Mesoporous Silica Nanocomposite for Sustained Cancer Labeling and Therapy. Macromol. Rapid Commun. 2016, 37, 1533−9. (25) Chen, X.; Liu, Z. N. Dual responsive mesoporous silica nanoparticles for targeted co-delivery of hydrophobic and hydrophilic anticancer drugs to tumor cells. J. Mater. Chem. B 2016, 4, 4382−4388. (26) Hu, L.-L.; Meng, J.; Zhang, D.-D.; Chen, M.-L.; Shu, Y.; Wang, J.-H. Functionalization of mesoporous organosilica nanocarrier for pH/glutathione dual-responsive drug delivery and imaging of cancer therapy process. Talanta 2018, 177, 203−211. (27) Chen, X.; Cheng, X. Y.; Soeriyadi, A. H.; Sagnella, S. M.; Lu, X.; Scott, J. A.; Lowe, S. B.; Kavallaris, M.; Gooding, J. J. Stimuliresponsive functionalized mesoporous silica nanoparticles for drug release in response to various biological stimuli. Biomater. Sci. 2014, 2, 121−130. (28) Du, J. Z.; Du, X. J.; Mao, C. Q.; Wang, J. Tailor-made dual pHsensitive polymer-doxorubicin nanoparticles for efficient anticancer drug delivery. J. Am. Chem. Soc. 2011, 133, 17560−17563. (29) Zheng, Z. B.; Zhu, G. Z.; Tak, H.; Joseph, E.; Eiseman, J. L.; Creighton, D. J. N-(2-hydroxypropyl)methacrylamide copolymers of a glutathione (GSH)-activated glyoxalase I inhibitor and DNA alkylating agent: Synthesis, reaction kinetics with GSH, and in vitro antitumor activities. Bioconjugate Chem. 2005, 16, 598−607. (30) Liu, D.; Pearlman, E.; Diaconu, E.; Guo, K.; Mori, H.; Haqqi, T.; Markowitz, S.; Willson, J.; Sy, M. S. Expression of hyaluronidase by tumor cells induces angiogenesis in vivo. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 7832−7837. (31) Prasetyanto, E. A.; Bertucci, A.; Septiadi, D.; Corradini, R.; Castro-Hartmann, P.; De Cola, L. Breakable Hybrid Organosilica Nanocapsules for Protein Delivery. Angew. Chem., Int. Ed. 2016, 55, 3323−3327. (32) Zhou, Z. X.; Shen, Y. Q.; Tang, J. B.; Fan, M. H.; Van Kirk, E. A.; Murdoch, W. J.; Radosz, M. Charge-Reversal Drug Conjugate for Targeted Cancer Cell Nuclear Drug Delivery. Adv. Funct. Mater. 2009, 19, 3580−3589. (33) Cao, Z.; Yu, Q.; Xue, H.; Cheng, G.; Jiang, S. Nanoparticles for drug delivery prepared from amphiphilic PLGA zwitterionic block copolymers with sharp contrast in polarity between two blocks. Angew. Chem., Int. Ed. 2010, 49, 3771−3776. (34) Arvizo, R. R.; Miranda, O. R.; Moyano, D. F.; Walden, C. A.; Giri, K.; Bhattacharya, R.; Robertson, J. D.; Rotello, V. M.; Reid, J. M.; Mukherjee, P. Modulating pharmacokinetics, tumor uptake and biodistribution by engineered nanoparticles. PLoS One 2011, 6, No. e24374. (35) Pan, L.; He, Q. J.; Liu, J. N.; Chen, Y.; Ma, M.; Zhang, L. L.; Shi, J. L. Nuclear-Targeted Drug Delivery of TAT Peptide-Conjugated Monodisperse Mesoporous Silica Nanoparticles. J. Am. Chem. Soc. 2012, 134, 5722−5725.

Foundation of Shaanxi Province (2017JM5023 to X.C.), and open fund of the State Key Laboratory of Military Stomatology (2017KA02 to X.C.)



REFERENCES

(1) Park, J.-H.; von Maltzahn, G.; Xu, M. J.; Fogal, V.; Kotamraju, V. R.; Ruoslahti, E.; Bhatia, S. N.; Sailor, M. J. Cooperative nanomaterial system to sensitize, target, and treat tumors. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 981−986. (2) Sailor, M. J.; Park, J. H. Hybrid Nanoparticles for Detection and Treatment of Cancer. Adv. Mater. 2012, 24, 3779−3802. (3) von Maltzahn, G.; Park, J. H.; Lin, K. Y.; Singh, N.; Schwoppe, C.; Mesters, R.; Berdel, W. E.; Ruoslahti, E.; Sailor, M. J.; Bhatia, S. N. Nanoparticles that communicate in vivo to amplify tumour targeting. Nat. Mater. 2011, 10, 545−552. (4) Ambrogio, M. W.; Thomas, C. R.; Zhao, Y. L.; Zink, J. I.; Stoddart, J. F. Mechanized silica nanoparticles: a new frontier in theranostic nanomedicine. Acc. Chem. Res. 2011, 44, 903−913. (5) Fang, Y.; Zheng, G.; Yang, J.; Tang, H.; Zhang, Y.; Kong, B.; Lv, Y.; Xu, C.; Asiri, A. M.; Zi, J.; Zhang, F.; Zhao, D. Dual-pore mesoporous carbon@silica composite core-shell nanospheres for multidrug delivery. Angew. Chem., Int. Ed. 2014, 53, 5366−5370. (6) Li, Z.; Barnes, J. C.; Bosoy, A.; Stoddart, J. F.; Zink, J. I. Mesoporous silica nanoparticles in biomedical applications. Chem. Soc. Rev. 2012, 41, 2590−2605. (7) Kortesuo, P.; Ahola, M.; Karlsson, S.; Kangasniemi, I.; Kiesvaara, J.; Yli-Urpo, A. Sol-gel-processed sintered silica xerogel as a carrier in controlled drug delivery. J. Biomed. Mater. Res. 1999, 44, 162−167. (8) Lai, W.; Garino, J.; Ducheyne, P. Silicon excretion from bioactive glass implanted in rabbit bone. Biomaterials 2002, 23, 213−217. (9) Aznar, E.; Oroval, M.; Pascual, L.; Murguia, J. R.; MartinezManez, R.; Sancenon, F. Gated Materials for On-Command Release of Guest Molecules. Chem. Rev. 2016, 116, 561−718. (10) Peer, D.; Karp, J. M.; Hong, S.; FaroKHzad, O. C.; Margalit, R.; Langer, R. Nanocarriers as an emerging platform for cancer therapy. Nat. Nanotechnol. 2007, 2, 751−760. (11) Khalil, I. A.; Kogure, K.; Akita, H.; Harashima, H. Uptake pathways and subsequent intracellular trafficking in nonviral gene delivery. Pharmacol. Rev. 2006, 58, 32−45. (12) Liu, Z.; Chen, X.; Zhang, X.; Gooding, J. J.; Zhou, Y. CarbonQuantum-Dots-Loaded Mesoporous Silica Nanocarriers with pHSwitchable Zwitterionic Surface and Enzyme-Responsive Pore-Cap for Targeted Imaging and Drug Delivery to Tumor. Adv. Healthcare Mater. 2016, 5, 1401−1407. (13) Angelos, S.; Khashab, N. M.; Yang, Y. W.; Trabolsi, A.; Khatib, H. A.; Stoddart, J. F.; Zink, J. I. pH clock-operated mechanized nanoparticles. J. Am. Chem. Soc. 2009, 131, 12912−12914. (14) Angelos, S.; Yang, Y. W.; Patel, K.; Stoddart, J. F.; Zink, J. I. pHresponsive supramolecular nanovalves based on cucurbit[6]uril pseudorotaxanes. Angew. Chem., Int. Ed. 2008, 47, 2222−2226. (15) Martínez-Carmona, M.; Lozano, D.; Colilla, M.; Vallet-Regí, M. Lectin-conjugated pH-responsive mesoporous silica nanoparticles for targeted bone cancer treatment. Acta Biomater. 2018, 65, 393−404. (16) Chen, X.; Yuan, P.; Liu, Z.; Bai, Y.; Zhou, Y. Dual responsive hydrogels based on functionalized mesoporous silica nanoparticles as an injectable platform for tumor therapy and tissue regeneration. J. Mater. Chem. B 2017, 5, 5968−5973. (17) Aznar, E.; Casasus, R.; Garcia-Acosta, B.; Marcos, M. D.; Martinez-Manez, R.; et al. Photochemical and chemical two-channel control of functional nanogated hybrid architectures. Adv. Mater. 2007, 19, 2228−2231. (18) Sun, L.; Wang, D.; Chen, Y.; Wang, L.; Huang, P.; Li, Y.; Liu, Z.; Yao, H.; Shi, J. Core-shell hierarchical mesostructured silica nanoparticles for gene/chemo-synergetic stepwise therapy of multidrugresistant cancer. Biomaterials 2017, 133, 219−228. (19) Wu, M.; Meng, Q.; Chen, Y.; Zhang, L.; Li, M.; Cai, X.; Li, Y.; Yu, P.; Zhang, L.; Shi, J. Large Pore-Sized Hollow Mesoporous 4314

DOI: 10.1021/acsomega.8b00427 ACS Omega 2018, 3, 4306−4315

Article

ACS Omega (36) Jin, R. H.; Liu, Z. N.; Bai, Y. K.; Zhou, Y. S.; Chen, X. Effective Control of Enzyme Activity Based on a Subtle Nanoreactor: A Promising Strategy for Biomedical Applications in the Future. ACS Appl. Nano Mater. 2018, 1, 302−309.

4315

DOI: 10.1021/acsomega.8b00427 ACS Omega 2018, 3, 4306−4315