Directing Assembly and Disassembly of 2D MoS2 Nanosheets with

Apr 28, 2017 - World Premier International (WPI) Research for Materials Nanoarchitectonics (MANA), National Institute for Materials Science (NIMS), 1-...
0 downloads 5 Views 8MB Size
Download PDF
Research Article www.acsami.org

Directing Assembly and Disassembly of 2D MoS2 Nanosheets with DNA for Drug Delivery Bang Lin Li,†,○ Magdiel I. Setyawati,†,○ Linye Chen,† Jianping Xie,† Katsuhiko Ariga,‡ Chwee-Teck Lim,§,∥,⊥,# Slaven Garaj,§,∥,∇ and David Tai Leong*,† †

Department of Chemical and Biomolecular Engineering, National University of Singapore, Singapore 117585, Singapore World Premier International (WPI) Research for Materials Nanoarchitectonics (MANA), National Institute for Materials Science (NIMS), 1-1 Namiki, Tsukuba 305-0044, Japan § Department of Biomedical Engineering, National University of Singapore, Singapore 117575, Singapore ∥ Centre for Advanced 2D Materials, Graphene Research Centre, National University of Singapore, Singapore 117546, Singapore ⊥ Mechanobiology Institute, National University of Singapore, Singapore 117411, Singapore # NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore 117456, Singapore ∇ Department of Physics, National University of Singapore, Singapore 117542, Singapore ‡

S Supporting Information *

ABSTRACT: Layer-by-layer (LbL) self-assembled stacked Testudo-like MoS2 superstructures carrying cancer drugs are formed from nanosheets controllably assembled with sequence-based DNA oligonucleotides. These superstructures can disassemble autonomously in response to cancer cells’ heightened ATP metabolism. First, we functionalize MoS2 nanosheets (MoS2-NS) nanostructures with DNA oligonucleotides having thiol-terminated groups (DNA/MoS2-NS) via strong binding to sulfur atom defect vacancies on MoS2 surfaces. The driving force to assemble into a higher-order DNA/MoS2-NS superstructure is guided by a linker aptamer that induced interlayer assembly. In the presence of target ATP molecules, these multilayer superstructures disassembled as a consequence of stronger binding of ATP molecules with the linking aptamers. This design plays a dual role of protection and delivery by LbL stacked MoS2-NS similar in concept to a Greek Testudo. These superstructures present a protective armor-like shell of MoS2-NS, which still remains responsive to small and infiltrating ATP molecules diffusing through the protective MoS2NS, contributing to an enhanced stimuli-responsive drug release system for targeted chemotherapy. KEYWORDS: molybdenum disulfide, DNA functionalization, layer-by-layer assembly, 3D cell culture, targeted chemotherapy

■

INTRODUCTION

defects of the basal plane without compromising on the 2Dness. Here, instead of tackling the reactivity issue head-on, we exploited the inertness of the MoS2 surface and the inherent phenomenon that comes with all bulk crystals, atomic defect sites.16−18 Taking MoS2 as our model TMD, S atom defect sites are always present on its exfoliated surface, and we utilize a Satom-terminated DNA molecule as the anchor to explore the role of S atoms in the attachment of MoS2 nanosheets (MoS2NS). We further used the inertness of the surface to our advantage in reducing the possibility of any spurious side reactions that may have occurred. This “plug and play” Sanchor strategy allows the synthesis to be carried out at room temperature with very fast kinetics in extremely mild buffer conditions as filling in a defect site is thermodynamically

A fascinating new class of two-dimensional (2D) materials, layered transition metal dichalcogenides (TMDs), has garnered much attention over the past few years. Although reports of layered TMDs’ exploitable properties and exciting applications cover several diverse research areas, such as biomedicine,1,2 electronic devices,3 catalysis,4 battery materials,5,6 and sensors,7−9 TMDs’ full potential is hamstrung by their inherent resistance toward functionalization.10 Despite sharing similar morphology with graphene that is renowned to be easily functionalized, 2D TMDs surfaces comprise transition metal sites fully embedded beneath two chemically inert chalcogen atomic layers. To tackle this problem, significant efforts have been dedicated to controlling the chemical functionalization at the surfaces of these 2D materials, which includes promoting the interaction between the ligands (e.g., organic halides,11 thiol groups,12,13 diazonium salts,14 and metal-acetate salts15) and the unsaturated metal or chalcogen atoms at the edges or © XXXX American Chemical Society

Received: February 21, 2017 Accepted: April 19, 2017

A

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 1. Thiol-terminated DNA increases MoS2-NS colloidal stability. (a) AFM image and (b) characteristic absorption spectrum of the as-obtained MoS2-NS. (c) Absorbance records (605 nm) of MoS2-NS in ultrapure water and 1× TAE buffer solution for different standing times. Data are average ± SD, n = 3, Student’s t-test, p < 0.05, *significant against the original MoS2-NS dispersion without treatments. (d) Absorbance changes (605 nm) of MoS2-NS dispersion before (A0h) and after 12 h (A12h) of standing time in the presence of various kinds of protective single-stranded DNA (ssDNA) including ssDNA without modification (unmodified P1), disulfide-terminated ssDNA dimer (P1-S−S-P1), and thiol-terminated ssDNA (P1). (e) Left panel: Schematic illustration of DNA functionalization on the surface of MoS2-NS, indicating that the thiol group is essential for functionalization of MoS2-NS with DNA. Right panel: The influence of DNA hybridization on the dispersity of MoS2-NS in a buffer solution and pictures of MoS2-NS aqueous dispersions in the presence of different kinds of DNA with standing time of 12 h. The thiol-terminated duplex (D1) was obtained via hybridization of the ATP aptamer with P1. (f) Effect of the thiol and disulfide group-terminated D1 sequence on the aggregation of MoS2-NS. (g) Fluorescence intensity of SYBR Green I binding to unbound duplex DNA as found in the supernatant of the treated mixture of the MoS2-NS and duplex. Data are average ± SD, n = 3, Student’s t-test, p < 0.05, *significant against the sample without treatment of MoS2-NS, #the minimum value of the SYBR Green signal. (h) Z-average sizes and ζ-potential values of MoS2-NS, P1-functionalized MoS2-NS (P1/MoS2-NS), and D1-functionalized MoS2-NS (D1/MoS2-NS). Data are average ± SD, n = 3, Student’s t-test, p < 0.05, *significant against control groups.

design, this colloidal stability immediately reverses when we mix species of mutually complementary DNA oligonucleotides stably anchored on other MoS2-NS; they quickly self-assemble layer-by-layer (LbL) due to the complementary hybridization of the DNA oligonucleotide pairs. The stimulus strategy was achieved by introducing DNA-based aptamers as linkers, which then gives our design, stimuli-responsive capabilities, that works at body temperature and without requiring any external inputs and behaves as an autonomous cell-specific control system. The layered superstructures formed are analogous and functionally

favorable and there are likely many Boltzmann collisions between the larger-sized MoS2 sheets and the smaller-sized Sterminated DNA. DNA offers sequence designability and complementary binding specificity, which when applied on another MoS2-NS allowed for linking, such as for building higher-order structures.19−22 The intrinsic negative charge of the DNA segment adds to the negative charge of the MoS2 basal plane.23,24 As non-complementary DNA sequences do not hybridize, the anchored DNA oligonucleotides on MoS2-NS further repel one another, resulting in high colloidal stability. By B

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces similar to the defensive shield-like Testudo invented by the ancient Greeks. By exploiting chemically inert MoS2 as shields in this nano-Testudo superstructure, we can protect the precious cargo from in vivo and intracellular degradative enzymes. We further exploit the DNA sequence specificity recognition of the autonomous stimulus to disassemble our superstructure into nanosheets, achieving a high level of nanoscale control. Our above designability was achieved through stacked MoS2-NS superstructures with doxorubicin (Dox), which disassembled in the highly metabolically active environment of cells, and was proven in cancer models, showing its direct applicability to the nanomedicine field and, by extension possibly, to other areas of bionanotechnology.

anchored down through the terminal thiol group, after allowing D1 to bind to MoS2-NS, the resultant mixture was centrifuged and the supernatant was collected. The concentration of freefloating unbound DNA found in the supernatant was quantified using fluorescent SYBR Green I assay through binding to dsDNA. As expected, there was no change in the supernatant’s dsDNA content. Essentially, for the control non-S-modified D1 group, negligible amount of the DNA duplex remained bound to the MoS2-NS after centrifugation (Figure 1f). In contrast, the fluorescence intensities of SYBR Green I decreased in step with increasing concentrations of MoS2-NS (Figure 1f) in the case of the D1 and D1-S−S-D1 groups. Considering that the same initial intensities of SYBR Green I were detected prior to its introduction to these different groups (Figure S1), the intensity drop in the supernatant of the D1 and D1-S−S-D1 groups strongly indicated their persistent binding on the MoS2-NS. Several theories have been proposed to explain how the thiol ligand functionalization of the MoS2-NS occurs. Jung et al. proposed that the thiol-terminated small molecules (mercaptoethylamine and 1H,1H,2H,2H-perfluorodecanethiol) can bind with the sulfur vacancies through the chemisorption process, thus controlling the electrical properties of the MoS2NS.12 In a more recent study, McDonald and co-workers demonstrated that thiol-group-containing cysteine was physiosorbed on MoS2.31 However, our results suggest that the thiol group interacts with a fixed number of sites on the MoS2 surface, instead of untargeted adsorption interaction. The possible S-binding sites are the same in the thiolated duplex and disulfide duplex groups. If it were a specific type of S-binding on the MoS2 surface, then it would be expected that twice the amount of DNA would be actually bound in the disulfide duplex group as the chance of a successful capture is doubled. That is to say, the binding effect of the thiol and disulfide groups to the MoS2-NS is up to the number of hindrance-free sulfur atoms and the thiol-terminated duplex bound on the MoS2-NS was based on the S-atom-vacancies rather than unspecific physio- or chemi-sorption. Indeed, we observed that the amount of DNA captured on the MoS2-NS was indeed twice when D1-S−S-D1 was investigated (Figure 1g). This further reinforced our hypothesis that capturing of the DNA oligonucleotide proceeded via an interaction of its S with a Sdefect site on the MoS2-NS. This is in line with the previous studies which report that the thiol pendant of different backbones functionalize the MoS2-NS surface by filling the defect sites on the surface of MoS2-NS with the sulfur atoms.12,32,33 DNA functionalization through S-terminated anchoring on MoS2-NS was further shown through the dynamic light scattering and ζ-potential measurements in which DNA/MoS2-NS were collected via centrifugation and were subsequently re-dispersed into ultrapure water. We expected and observed a slight increase in size following MoS2-NS functionalization with thiolated ssDNA P1 (P1/ MoS2-NS) and thiolated dsDNA D1 (D1/MoS2-NS) (Figure 1h), supporting the successful formation of the DNA/MoS2-NS structure. Successful formation of the DNA/MoS2-NS structure was also reflected by the decrease in the ζ potential of the MoS2-NS from −26.3 to −35.4 and −41.3 eV after being functionalized with P1 and D1, respectively (Figure 1h), further corroborating the high colloidal stability observed with thiol DNA-functionalized D1/MoS2-NS groups (Figures 1e,f and S2). Drug Loading and ATP-Responsive Character. Even though the beneficial properties of TMDs in photothermal,

■

RESULTS AND DISCUSSION Thiol-Terminated DNA Increases MoS2-NS Colloidal Stability. Exfoliated MoS2-NS, consisting of three to six monolayers, were obtained by exposing commercial MoS2 powders to ultrasonic treatment in aqueous sodium cholate solution.8,25 Morphological and optical characterizations of MoS2-NS showed that the lateral size of MoS2-NS was around 100 nm, whereas their thickness ranged from 2 to 4 nm (Figure 1a). The absorption peaks corresponded to the four electronic transitions in the band structure of exfoliated MoS2-NS, termed as A, B, C, and D (Figure 1b).25 Peaks A and B at 666 nm (∼1.862 eV) and 605 nm (∼2.043 eV), respectively, are derived from the interband excitonic transitions at the K point of the Brillouin zone for 2D MoS2.26,27 Two broad peaks at 450 nm (C exciton) and 395 nm (D exciton) indicate a strong blue shift in optical absorption. This blue shift from the quantum size effect showed a plane size of less than 50 nm.8,28 To understand the stability mechanism of MoS2-NS in complex biological solutions, we investigated the absorbance of MoS2-NS with variable standing times that corresponded to the MoS2-NS colloidal stabilities.8 No absorbance change from 605 nm for MoS2-NS aqueous colloidal suspensions was observed within 12 h, indicating their good colloidal stability in water (Figure 1c). However, positive Mg2+ ion neutralization of the MoS2-NS surface leading to rapid aggregation and sedimentation significantly decreased the 605 nm absorbance of MoS2-NS (Figure 1c).8 We hypothesized that the thiol group can anchor down a ssDNA through binding at the intrinsic S defects on MoS2 surfaces. A high colloidal stability was observed when we treated layered MoS2-NS with protective thiol-terminated single-stranded DNA (ssDNA) oligonucleotides (P1) and disulfide-linked ssDNA oligonucleotides (P1-S−S-P1) as compared to the treatment group of the non-S-modified DNA oligonucleotides (unmodified P1) of the same DNA sequence, molar concentrations, and treatment times (Figure 1d and Table S1). To control for weak van der Waals forces between the double-stranded DNA (dsDNA) and nanosheet surfaces, we showed that nonthiolated duplex control (unmodified D1), hybridized from nonthiolated P1 and ATP aptamer, could not stably adsorb on MoS2-NS surfaces because the ring structure of nucleobases was effectively shielded through complementary base pairing.8,29,30 In stark contrast, thiolated duplex (D1) or disulfide-linked duplex (D1-S−S-D1) decisively reversed the trend (Figure 1e), again suggesting that the thiol group was very important in anchoring down the oligonucleotides and therefore forming a repulsive DNA “surface layer” to prevent aggregation between MoS2-NS nanostructures. To further demonstrate that DNA is firmly C

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 2. Duplex functionalized MoS2-NS for Dox loading and ATP-responsive character. (a) Schematic illustration of drug loading and ATPinduced release on the duplex/MoS2-NS while the FAM-labeled ATP aptamer for P1 hybridization was used as an indicator. (b) Fluorescence spectra of Dox (4.5 μM) in the absence and presence of D1, MoS2-NS, and D1/MoS2-NS. (c) Dox loading amount of D1, MoS2-NS, and D1/MoS2NS (D1: 1 μM, MoS2-NS: 200 μg mL−1, fixed concentrations) as a measure of the loading yield. (d) Fluorescence intensities of the Dox-composite aqueous dispersion after incubation with different concentrations of ATP at 37 °C for 1 h. (e) Fluorescence imaging of MDA-MB-468 cells incubated with Dox/D1 and Dox/D1/MoS2-NS for 60 min. The nuclei were stained (blue) with Hoechst 33342, and the ATP aptamer was labeled with FAM fluorophore (green). Scale bar: 10 μm. (f) ATP content in MDA-MB-468 cells after different kinds of treatments. The treatment in a medium with high-concentration glucose (4.5 g L−1, H-glucose) at 37 °C was used as a control, whereas the medium with low-concentration glucose (1.0 g L−1, L-glucose) and temperature of 4 °C were used to provide intracellular ATP-poor conditions. For L-glucose experiments, the cells were pretreated with L-glucose for 8 h before measurements. Data are means ± SD, n = 3, Student’s t-test, p < 0.05, *significant compared to cells treated with H-glucose medium at 37 °C. (g) Monitoring of the fluorescence intensity of the FAM dye released from the surface of the MoS2-NS in MDAMB-468 cells. The cells were incubated with Dox/D1/MoS2-NS in the H-glucose medium at 37 °C for 60 min and then incubated with the fresh Hglucose medium at 4 °C or with the L-glucose medium at 37 °C for additional 30 or 60 min after removal of the excess Dox/D1/MoS2-NS. Data are means ± SD, n = 3, Student’s t-test, p < 0.05, *significant against samples without further fresh medium treatments, #significant against fresh medium treatment for 30 min.

First, an ATP-molecule-responsive aptamer was incorporated as a complementary strand to the thiolated DNA oligonucleotide anchored on the MoS2-NS. The ATP-aptamer-based duplex was a GC base-pair-rich structure, thus exhibiting a desired loading efficacy to Dox.37−40 At high concentrations of ATP, the ATP aptamer will preferentially bind with ATP and therefore dissociate from the thiolated DNA oligonucleotide anchors (Figures 2a and S3).39,40 The anchoring on the MoS2NS surface brings the intercalated Dox very close to the surface, which in turn strongly quenches Dox’s inherent fluorescence (Figures 2b and S4). The Dox loading yield on the D1/MoS2NS coincides well with the sum total of individual D1 and MoS 2 -NS (Figure 2b,c). This suggests both the Dox intercalation between the dsDNA and its adsorption on the

photodynamic, and drug delivery aspects have been explored for cancer therapy, their targeting therapies are still underdeveloped. As such, it becomes pivotal to overcome this underdeveloped trait to allow practical application of TMDsbased cancer therapy.34−36 Recently, DNA-based anticancer drug carriers were developed, as Dox preferentially intercalates in consecutive GC base pair duplexes.37,38 In our proof-ofconcept experiments, we broke up the study into two segments. The first is the study of formation and efficacy of single-layer DNA oligonucleotide-Dox-MoS2. In the second study, we made slight alterations in the DNA oligonucleotide sequence design to produce a customizable stacked complex of multilayer DNA oligonucleotide-Dox-MoS2, much like a Testudo higher-order structure. D

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 3. Intracellular Dox delivery and cancer killing effect. (a) Dox release in MDA-MB-468 cells. The cells were first precultured in either Hglucose (ATP-rich) or L-glucose (ATP-poor) media for 8 h and then incubated with various Dox-composite groups at 37 °C for 60 min under ATPrich and ATP-poor conditions, respectively. The treated samples were subsequently incubated with fresh ATP-rich and ATP-poor media for additional 30 min after removing the excess Dox/carriers. (b) Schematic illustration showing intracellular processes of drug delivery and ATP-induced release from DOX/D1/MoS2-NS. The inset indicated the role of endosomes in the delivery of Dox/D1/MoS2-NS to the nucleus. (c) Intracellular delivery of Dox/D1/MoS2-NS for different incubation times observed by fluorescence microscopy. The nuclei (gray) were stained by Hoechst 33342, and the late endosomes and lysosomes (green) were stained by LysoTracker Green. Scale bar: 10 μm. (d) In vitro cytotoxicity effect of Dox/D1/MoS2NS on MDA-MB-468 and MCF-7 under intracellular ATP-rich and ATP-poor conditions. Data are means ± SD, n = 3, Student’s t-test, p < 0.05, *significant when compared to the Dox-only group, #significant when compared to the corresponding ATP-poor group.

MoS2-NS, the FAM fluorescence signal was quenched due to its close proximity to the MoS2-NS.7,8,29 In the presence of ATP molecules, ATP binds with the ATP aptamer, resulting in the escape of the FAM-ATP aptamer from the surface of the MoS2NS and in turn dose-dependent increase of the FAM fluorescence signal was detected (Figure S5). We investigated the FAM-labeled Dox/D1/MoS2 NS construct in MDA-MB468, a breast cancer cell line. After a brief 60 min, the Dox/D1/ MoS2-NS entered the cell and was found to localize near the nucleus. Dox (red) and the ATP aptamer (green) were effectively internalized by the cells, and their merged images

MoS2-NS play substantial role in overall Dox loading on the D1/MoS2-NS structure. Both Dox-loaded D1 (Dox/D1) and Dox-loaded D1/MoS2-NS (Dox/D1/MoS2-NS) showed a dose-dependent Dox release in the presence of ATP (Figure 2d), whereas Dox/MoS2 remained unresponsive toward any ATP stimuli at any tested concentration. Dox/D1/MoS2-NS responsiveness to ATP is confirmed through the use of the 5carboxyfluorescein (FAM)-labeled single-stranded ATP aptamer (Figure 2e). The FAM-ATP aptamer hybridized with the complementary ssDNA (P1) to form FAM-labeled duplex structures (FAM-D1). After FAM-D1 was incubated with the E

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 4. Formation of multilayer DNA/MoS2-NS via LbL assembly. (a) Schematic illustration of formation of multilayer Dox/D2/MoS2-NS and their intracellular Dox release process. (I) ATP-aptamer-induced LbL assembly; (II) Dox loading in multilayer nanostructures; (III) in vitro treatment of Dox/D2/MoS2-NS; (IV) cell uptake through endocytosis; (V) endosomal/lysosomal escape; (VI) ATP-induced Dox release in the cyotosol. (b, c) AFM images of P2/MoS2-NS and P3/MoS2-NS mixtures in the absence (b1) and presence (c1) of the ATP aptamer, and the height profile along the white line (b2, c2) overlaid on the AFM images. (d) Z-average sizes of P2/MoS2-NS and P3/MoS2-NS mixtures in the presence of non-complementary DNA (non-cDNA) and ATP aptamer. (e) Z-average sizes of multilayer D2/MoS2-NS incubated with different concentrations of ATP. The control experiments were conducted via incubating the mixture of P1/MoS2-NS and P2/MoS2-NS with different concentrations of ATP. (f) Fluorescence intensity of Dox, P2/MoS2-NS, and P3/MoS2-NS before and after incubation with non-cDNA and ATP aptamer. Data are average ± SD, n = 3, Student’s t-test, *p < 0.05.

NS construct was autonomously sensitive to intracellular ATP, triggering Dox release from the structure. Dox Release and in Vitro Toxicity. With the assistance of the ATP-aptamer-containing D1/MoS2-NS, Dox can be delivered into the breast cancer cells, MDA-MB-468, with high efficacy, especially in the presence of high concentration of ATP (Figure 3a). Only D1-containing nanostructures exhibited time-dependent Dox release efficacy under ATP-rich conditions but not the groups where D1 was absent (Dox-only group). As not only cellular internalization but also cellular trafficking plays a determinant role in the efficacy of nanomedicine formulation, we further examined the intracellular trafficking of Dox/D1/ MoS2-NS constructs. We hypothesized that the Dox/D1/ MoS2-NS were first internalized by the cancer cell and subsequently transported through the cells’ endosomes and lysosomes (Figure 3b). To validate our hypothesis, we stained the endolysosome vesicles with LysoTracker Green to track the localization of Dox/D1/MoS2-NS nanostructures within the

(yellow) showed distinct co-localization of the release of the ATP aptamer with the simultaneous release of Dox (Figure 2e). However, Dox/D1-based delivery (without MoS2-NS) was restricted only to the cell membrane, whereas the Dox/D1/ MoS2-NS group showed a significantly higher delivery rate into the cells. Traversing the cell membrane is only the first of many steps. We further validated and demonstrated the intracellular ATP concentration-dependent Dox release from the Dox/D1/ MoS2-NS. We inhibited the production of ATP either by decreasing the temperature (4 °C) or reducing the concentration of glucose in the medium. Cell incubation at a lower temperature and at a much lower concentration of glucose led to a drastic decrease in the intracellular ATP content (Figure 2f). With this reduced ATP content, the release of the ATP aptamer from the duplex was concomitantly reduced, as observed by the lower fluorescence-recovered FAM-ATP aptamer (Figure 2g). This showed that our Dox/D1/MoS2F

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 5. Inhibition of tumor spheroid growth of LbL-Dox/D2/MoS2-NS. (a) Changes of fluorescence intensities of Dox/D2/MoS2-NS incubated with different concentrations of ATP. (b) Changes of florescence intensities of (1) Dox, (2) Dox/D1, (3) single-layer Dox/D1/MoS2-NS, and (4) LbL-Dox/D2/MoS2-NS (4) with treatments of DNase I for different incubation times. (c) Left panel: Schematic illustration depicting the formation of 3D tumor spheroids and the treatment regime that they received. 3D tumor spheroids were formed by adding the MDA-MB-468 cells into the agarose micromold52 (I) and incubating them for 24 h prior the treatment (II). Subchronic study was conducted by treating the 3D tumor spheroids directly in the well for 5 days (III), whereas the acute study was conducted by flushing the 3D tumor spheroids, moving them to the flat agarose plate, and treating them for 2 days (IV). Right panel: Representative immunostaining image of the 3D tumor spheroids. Nucleus (blue) was stained with Hoechst 33342, whereas actin (red) was visualized following phalloidin immunostaining. Scale bar: 100 μm. (d) Increase in the apoptotic cell population following 3D tumor spheroids acute exposure with different Dox/carriers. Data are average ± SD, n = 3, Student’s t-test, p < 0.05, *significant when compared to vehicle control. (e) Phase contrast image depicts the reduction in the tumor size following subchronic exposure of different Dox-composites on 3D tumor spheroids over the course of 5 day treatment. Scale bar: 50 μm. (f) Quantification analysis recapitulates the reduction in the tumor size following subchronic exposure of different Dox-composites on 3D tumor spheroids. The concentration of Dox for all treatments is 5 μM. Data are means ± SD, n = 3, Student’s t-test, p < 0.05, *significant against vehicle control.

cells. In the first 30 min of incubation, the Dox/D1/MoS2-NS nanostructures were quickly localized in the endolysosomes (Figure 3c). After 60 min of incubation, Dox, of which red fluorescence was originally quenched when bound to MoS2-NS, had dissociated from the Dox/D1/MoS2-NS nanostructures as observed by decoupled delocalization of red fluorescence from the green lysosomal tracker fluorophore. This suggested lysosomal escape of Dox in the presence of ATP. Following treatment with the fresh medium for 30 min, it was found that the released Dox has specifically accumulated at the nuclei. Dox/D1/MoS2-NS exhibited a significantly enhanced cytotoxicity toward the MDA-MB-468 and MCF-7 cancer cells compared to that of Dox, Dox/D1, and Dox/MoS2-NS (Figure

3d). The killing efficacy is expectedly further enhanced under high intracellular ATP concentrations due to the increased release of Dox through the ATP aptamer dissociation from its complementary strand on Dox/D1/MoS2-NS. Vertical Growth via LbL Assembly. Expanding our toolkit platform, we showed that we can design MoS2-NS to self-assemble into higher-order superstructures simply through smart design of DNA sequences. LbL-assembled higherordered superstructures have been achieved for energy, sensing, and biomedical applications. 41−44 Although most LbL assemblies for hard inorganic materials such as 2D nanosheets rely on nonspecific interactions, LbL construction is not always sheet-specific.45−47 Thus, in this case, we could introduce G

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

measured by the dynamic light scattering technique, increased from about 200 to around 400 nm (Figure 4d), which fits well into the extremes of the maximal x−y plane overlapping conformation and minimal overlapping conformation. ATP molecules could bind to the ATP aptamer, resulting in a conformational change and leading to the dissociation of LbL superstructures (Figure 4e). Dox loading was achieved based on as-obtained LbL superstructures. The LbL-D2/MoS2-NS exhibit a higher quenching effect on Dox fluorescence compared to that of individual P2/MoS2-NS and P3/MoS2NS layer (Figure 4f). The interlayer assembly of P2/MoS2-NS and P3/MoS2-NS occurs only in the presence of the ATP aptamer linker, and compared to individual layers, LbL superstructures had a higher loading efficacy due to the existence of the void space created between as-formed multilayers to carry even more Dox than what dsDNA duplex could carry (Figure 4f). LbL Superstructures for Tumor Inhibition. A comparison of the loading efficacy for different carriers was made (Table S2). As expected, increasing ATP concentration released Dox from the LbL-D2/MoS2-NS structures as seen from the fluorescence of freed Dox (Figure 5a). We then checked whether the LbL-D2/MoS2-NS was able to withstand the extracellular harsh conditions. The LbL-D2/MoS2-NS could be a good vehicle to target Dox delivery into cells, protecting the loaded duplex from enzymatic cleavage until the drugs reached the target cells and significantly enhancing the targeted effect of chemotherapy while simultaneously reducing the off-target effects of the nanomedicine. As the DNA is the critical and weakest link that holds the stacked superstructure together, we subjected our LbL-D2/MoS2-NS structures to the DNase I degradation reaction. The consequences of DNase cleavage for different drug carriers were investigated. From Figure 5b, it was found that the Dox-carrying composites (without MoS2-NS) rapidly degraded in the DNA degradative reaction by DNase I, leading to the release of Dox (Figure 5b). However, with the protection of the MoS2-NS (analogous to Testudo shield defense), DNA degradation in the LbL-Dox/D2/MoS2-NS was practically nonexistent with negligible release of Dox (Figure 5b). Significantly, for the LbL-Dox/D2/MoS2-NS nanostructures, the duplexes were fully covered by the MoS2-NS. This is in good agreement with the previous studies in which DNA strands were reported to withstand enzymatic digestion when the free-hanging strand end was folded within the well-formed DNA tetrahedral and nanoshuriken structures.48−51 To give us a better idea on the LbL-Dox/D2/MoS2-NS efficacy in the real in vivo setting, we treated 3D tumor spheroid models52,53 (Figure 5c) with the LbL-Dox/D2/MoS2NS and found it to be highly effective in killing these tumor cells. Compared to the vehicle control and free Dox groups, our acute study shows that the multilayer Dox/D2/MoS2-NS induced close to 5-fold increase in the number of apoptotic cell population following just 48 h treatment (Figure 5d). Comparatively, the single-layer Dox/D1/MoS2-NS resulted a “mere” 3.4-fold increase in apoptosis induction, giving LbLDox/D2/MoS2-NS a net 1.5-fold improvement in its efficacy in killing the tumor cells. To further validate our result, we conducted a subchronic study in which the tumor spheroids were treated with 10% of the dose they received in the acute study over an extended treatment time of 5 days. Although both single-layer Dox/D1/ MoS2-NS and LbL-Dox/D2/MoS2-NS induced significant tumor size reduction, the greatest size reduction was observed

individual sheet-specificity through complementary DNA oligonucleotides. In this protocol, the sequence of the original P1 (Figure 1e) was separated into two thiolated shorter species, namely, P2 and P3. A single species of short-sequence P2functionalized MoS2-NS (P2/MoS2-NS) in one pot showed good colloidal stability and did not self-assemble to form larger structures. When we mixed P3-functionalized MoS2-NS (P3/ MoS2-NS) in the second pot with non-complementary DNA (non-cDNA) strands, again no self-assembly into higherordered structures was observed. In contrast, when we introduced a linker DNA sequence, in which a segment of the linker sequence is complementary to P2 and another segment is complementary to P3, it immediately induced selfassembly into stacked structures (D2/MoS2-NS) in a controllable manner (Figure 4a). Replacing the linker with a negative control of non-complementary DNA oligonucleotides that do not have any complementarity with P2 or P3 (Figure S6) did not result in any significant aggregation or self-assembly to higher-order superstructures. As the linker in use contains the ATP aptamer sequence, the resulting larger stacked structure was able to disassemble in response to the ATP stimulus into the original discrete P2/MoS2-NS and P3/MoS2-NS (Figures 4a and S6). Now adding Dox then completes the new higherorder superstructures of our nanosheet design assembled LbL (LbL-Dox/D2/MoS2-NS). An important point to note is that the negatively charged phosphate groups in DNA are repelled by the negatively charged MoS2 surface, thus forcing the projection of the ssDNA oligo into space to facilitate easy hybridization with the complementary ssDNA oligo sequence. There is one critical advantage of stacked DNA/MoS2-NS, voids are created as the DNA oligonucleotides themselves are steric molecules between the nanosheets. For any drug carrier like the classical micellar systems and mesoporous and microporous systems, void space is critical as it is where the cargo is stored and protected until released at the right setting. However, void space in these classical carrier systems is extremely difficult to tune as it often involves the stochastic nature of chemical reactions that created those voids in the first place. In our case, the voids are created by the steric hindrance of the DNA oligonucleotides, and we simply vary the length of the oligonucleotides to tune this valuable void space according to the cargo sizes over a large range of sizes at angstrom precision (at 0.34 nm lengthwise resolution) with effectively insignificant stochastic variations. The LbL-Dox/D2/MoS2-NS were subsequently delivered into the cells through the endocytosis process and transportation into the endosomes and lysosomes (Figure 4a). Once in the cell, small ATP molecules are able to gain access into the stacked nanosheets and bind with the ATP aptamer linker, resulting in a conformational change of the ATP aptamer that dissociates from the P2 and P3 parts of the MoS2-NS. This dissociation of the linker also dissembled the higher-order structures and simultaneously released Dox. The vertical sizes of ssDNA-functionalized MoS2-NS (P2/ MoS2-NS and P3/MoS2-NS) are around 2 nm (Figure 4b). After incubation with the ATP aptamer, stacked LbL-D2/ MoS2-NS formed and eventually their vertical sizes increased to above 10 nm (Figure 4c). Nevertheless, a slight increase was expectedly observed in the x−y plane as the x−y plane alignment is uncontrollable as the precise anchoring of P2 and P3 on the MoS2-NS as well as which specific P2 binds to which P3 oligonucleotide could not be exactly controlled in our current design. As such, the average size of the MoS2-NS, H

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces Author Contributions

when the LbL-Dox/D2/MoS2-NS was utilized to deliver the Dox to the tumor spheroids (Figures 5e,f and S7). Compared to that with the vehicle control group, approximately 30 and 45% size reduction were registered on tumor spheroids treated with LbL-Dox/D2/MoS2-NS for 3 and 5 consecutive days, respectively (Figures 5e,f and S7). Both the acute and subchronic studies agreed well on the overall increased killing by the LbL-Dox/D2/MoS2-NS structure. This overall improvement shown by the LbL-Dox/D2/MoS2-NS could be expected as the construct allowed more Dox to be loaded in the created void space and in turn more Dox was delivered to the tumor cells when compared to that from the single-layer Dox/D1/ MoS2-NS without voids. Overall, the results further supported our hypothesis that the stacked capacious LbL-Dox/D2/MoS2NS are effective autonomously controlled responsive cancer nanomedicine.

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the funding provided by the National Research Foundation, Prime Minister’s Office, Singapore, under Competitive Research Program (Award No. NRFCRP13-2014-03).

■

■

(1) Ji, D.-K.; Zhang, Y.; Zang, Y.; Li, J.; Chen, G.-R.; He, X.-P.; Tian, H. Targeted Intracellular Production of Reactive Oxygen Species by a 2D Molybdenum Disulfide Glycosheet. Adv. Mater. 2016, 28, 9356− 9363. (2) Wang, S.; Chen, Y.; Li, X.; Gao, W.; Zhang, L.; Liu, J.; Zheng, Y.; Chen, H.; Shi, J. Injectable 2D MoS2-Integrated Drug Delivering Implant for Highly Efficient NIR-Triggered Synergistic Tumor Hyperthermia. Adv. Mater. 2015, 27, 7117−7122. (3) Song, W. G.; Kwon, H.-J.; Park, J.; Yeo, J.; Kim, M.; Park, S.; Yun, S.; Kyung, K.-U.; Grigoropoulos, C. P.; Kim, S.; Hong, Y. K. HighPerformance Flexible Multilayer MoS2 Transistors on Solution-Based Polyimide Substrates. Adv. Funct. Mater. 2016, 26, 2426−2434. (4) Li, H.; Tsai, C.; Koh, A. L.; Cai, L.; Contryman, A. W.; Fragapane, A. H.; Zhao, J.; Han, H. S.; Manoharan, H. C.; AbildPedersen, F.; Norskov, J. K.; Zheng, X. Activating and Optimizing MoS2 Basal Planes for Hydrogen Evolution through the Formation of Strained Sulphur Vacancies. Nat. Mater. 2016, 15, 48−53. (5) Teng, Y.; Zhao, H.; Zhang, Z.; Li, Z.; Xia, Q.; Zhang, Y.; Zhao, L.; ́ Du, X.; Du, Z.; Lv, P.; Swierczek, K. MoS2 Nanosheets Vertically Grown on Graphene Sheets for Lithium-Ion Battery Anodes. ACS Nano 2016, 10, 8526−8535. (6) Xie, X.; Makaryan, T.; Zhao, M.; Van Aken, K. L.; Gogotsi, Y.; Wang, G. MoS2 Nanosheets Vertically Aligned on Carbon Paper: A Freestanding Electrode for Highly Reversible Sodium-Ion Batteries. Adv. Energy Mater. 2016, 6, No. 1502161. (7) Li, B. L.; Wang, J.; Zou, H. L.; Garaj, S.; Lim, C. T.; Xie, J.; Li, N. B.; Leong, D. T. Low-Dimensional Transition Metal Dichalcogenide Nanostructures Based Sensors. Adv. Funct. Mater. 2016, 26, 7034− 7056. (8) Li, B. L.; Zou, H. L.; Lu, L.; Yang, Y.; Lei, J. L.; Luo, H. Q.; Li, N. B. Size-Dependent Optical Absorption of Layered MoS2 and DNA Oligonucleotides Induced Dispersion Behavior for Label-Free Detection of Single-Nucleotide Polymorphism. Adv. Funct. Mater. 2015, 25, 3541−3550. (9) Li, B. L.; Setyawati, M. I.; Zou, H. L.; Dong, J. X.; Luo, H. Q.; Li, N. B.; Leong, D. T. Emerging 0D Transition-Metal Dichalcogenides for Sensors, Biomedicine, and Clean Energy. Small, 2017, DOI: 10.1002/smll.201700527. (10) Chen, X.; McDonald, A. R. Functionalization of TwoDimensional Transition-Metal Dichalcogenides. Adv. Mater. 2016, 28, 5738−5746. (11) Voiry, D.; Goswami, A.; Kappera, R.; e Silva Cde, C.; Kaplan, D.; Fujita, T.; Chen, M.; Asefa, T.; Chhowalla, M. Covalent Functionalization of Monolayered Transition Metal Dichalcogenides by Phase Engineering. Nat. Chem. 2015, 7, 45−49. (12) Sim, D. M.; Kim, M.; Yim, S.; Choi, M.-J.; Choi, J.; Yoo, S.; Jung, Y. S. Controlled Doping of Vacancy-Containing Few-Layer MoS2 via Highly Stable Thiol-Based Molecular Chemisorption. ACS Nano 2015, 9, 12115−12123. (13) Chou, S. S.; De, M.; Kim, J.; Byun, S.; Dykstra, C.; Yu, J.; Huang, J.; Dravid, V. P. Ligand Conjugation of Chemically Exfoliated MoS2. J. Am. Chem. Soc. 2013, 135, 4584−4587.

CONCLUSIONS DNA was first anchored on layered MoS2 nanostructures via the binding of DNA’s thiol groups to sulfur atom vacancies on their surfaces. The duplex/MoS2-NS was considered a base platform, achieving a high efficacy and autonomous ATPresponsive drug delivery. Subsequently, the vertical growth of the DNA/MoS2-NS could be guided by the targeted aptamer, and thus a Testudo-like multilayer MoS2-NS was obtained via LbL assembly of the thiol-terminated DNA-functionalized MoS2-NS. The as-obtained Testudo higher-ordered structures formed with the MoS2-NS as “shields” were highly inert and resistant to the damaging intracellular DNA enzymes. Nevertheless, under ATP-rich conditions, the intracellular role of drug loading nano-Testudo rapidly transits from a defense to an aggression state, contributing to highly efficient apoptosis of cancer cells. Therefore, with the synergetic effects of DNA specificity and 2D plane of MoS2, a designable autonomously stimuli-responsive drug delivery system has been achieved based on multilayered DNA/MoS2-NS, showing its direct applicability to the nanomedicine field and, by extension possibly, to other areas of nanotechnology fields.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b02529. Experimental details; additional data for DNA-functionalized MoS2 stability; Dox loading on the ATP duplex structure and the MoS2-NS; ATP response dissociation; 3D tumor size reduction (PDF)

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Jianping Xie: 0000-0002-3254-5799 Katsuhiko Ariga: 0000-0002-2445-2955 Slaven Garaj: 0000-0001-5529-4040 David Tai Leong: 0000-0001-8539-9062 Author Contributions ○

B.L.L. and M.I.S. contribute equally to this work. I

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces (14) Knirsch, K. C.; Berner, N. C.; Nerl, H. C.; Cucinotta, C. S.; Gholamvand, Z.; McEvoy, N.; Wang, Z.; Abramovic, I.; Vecera, P.; Halik, M.; Sanvito, S.; Duesberg, G. S.; Nicolosi, V.; Hauke, F.; Hirsch, A.; Coleman, J. N.; Backes, C. Basal-Plane Functionalization of Chemically Exfoliated Molybdenum Disulfide by Diazonium Salts. ACS Nano 2015, 9, 6018−6030. (15) Backes, C.; Berner, N. C.; Chen, X.; Lafargue, P.; LaPlace, P.; Freeley, M.; Duesberg, G. S.; Coleman, J. N.; McDonald, A. R. Functionalization of Liquid-Exfoliated Two-Dimensional 2H-MoS2. Angew. Chem., Int. Ed. 2015, 54, 2638−2642. (16) Fabbri, F.; Rotunno, E.; Cinquanta, E.; Campi, D.; Bonnini, E.; Kaplan, D.; Lazzarini, L.; Bernasconi, M.; Ferrari, C.; Longo, M.; Nicotra, G.; Molle, A.; Swaminathan, V.; Salviati, G. Novel NearInfrared Emission from Crystal Defects in MoS2 Multilayer Flakes. Nat. Commun. 2016, 7, No. 13044. (17) Ai, K.; Ruan, C.; Shen, M.; Lu, L. MoS2 Nanosheets with Widened Interlayer Spacing for High-Efficiency Removal of Mercury in Aquatic Systems. Adv. Funct. Mater. 2016, 26, 5542−5549. (18) Li, W.; Yang, Y.; Weber, J. K.; Zhang, G.; Zhou, R. Tunable, Strain-Controlled Nanoporous MoS2 Filter for Water Desalination. ACS Nano 2016, 10, 1829−1835. (19) Jones, M. R.; Seeman, N. C.; Mirkin, C. A. Programmable Materials and the Nature of the DNA Bond. Science 2015, 347, No. 1260901. (20) Ohta, S.; Glancy, D.; Chan, W. C. W. DNA-Controlled Dynamic Colloidal Nanoparticle Systems for Mediating Cellular Interaction. Science 2016, 351, 841−845. (21) Raeesi, V.; Chou, L. Y. T.; Chan, W. C. W. Tuning the Drug Loading and Release of DNA-Assembled Gold-Nanorod Superstructures. Adv. Mater. 2016, 28, 8511−8518. (22) Pandian, G. N.; Sugiyama, H. Nature-Inspired Design of Smart Biomaterials Using the Chemical Biology of Nucleic Acids. Bull. Chem. Soc. Jpn. 2016, 89, 843−868. (23) Sun, H.; Ren, J.; Qu, X. Carbon Nanomaterials and DNA: from Molecular Recognition to Applications. Acc. Chem. Res. 2016, 49, 461− 470. (24) Patil, A. J.; Vickery, J. L.; Scott, T. B.; Mann, S. Aqueous Stabilization and Self-Assembly of Graphene Sheets into Layered BioNanocomposites using DNA. Adv. Mater. 2009, 21, 3159−3164. (25) Li, B. L.; Chen, L. X.; Zou, H. L.; Lei, J. L.; Luo, H. Q.; Li, N. B. Electrochemically Induced Fenton Reaction of Few-Layer MoS2 Nanosheets: Preparation of Luminescent Quantum Dots via a Transition of Nanoporous Morphology. Nanoscale 2014, 6, 9831− 9838. (26) Qiu, D. Y.; da Jornada, F. H.; Louie, S. G. Optical Spectrum of MoS2: Many-Body Effects and Diversity of Exciton States. Phys. Rev. Lett. 2013, 111, No. 216805. (27) Stier, A. V.; McCreary, K. M.; Jonker, B. T.; Kono, J.; Crooker, S. A. Exciton Diamagnetic Shifts and Valley Zeeman Effects in Monolayer WS2 and MoS2 to 65 Tesla. Nat. Commun. 2016, 7, No. 10643. (28) Wang, K.; Wang, J.; Fan, J.; Lotya, M.; O’Neill, A.; Fox, D.; Feng, Y.; Zhang, X.; Jiang, B.; Zhao, Q.; Zhang, H.; Coleman, J. N.; Zhang, L.; Blau, W. J. Ultrafast Saturable Absorption of TwoDimensional MoS2 Nanosheets. ACS Nano 2013, 7, 9260−9267. (29) Zhu, C.; Zeng, Z.; Li, H.; Li, F.; Fan, C.; Zhang, H. Single-Layer MoS2-Based Nanoprobes for Homogeneous Detection of Biomolecules. J. Am. Chem. Soc. 2013, 135, 5998−6001. (30) Zhang, Y.; Zheng, B.; Zhu, C.; Zhang, X.; Tan, C.; Li, H.; Chen, B.; Yang, J.; Chen, J.; Huang, Y.; Wang, L.; Zhang, H. Single-Layer Transition Metal Dichalcogenide Nanosheet-Based Nanosensors for Rapid, Sensitive, and Multiplexed Detection of DNA. Adv. Mater. 2015, 27, 935−939. (31) Chen, X.; Berner, N. C.; Backes, C.; Duesberg, G. S.; McDonald, A. R. Functionalization of Two-Dimensional MoS2: On the Reaction Between MoS2 and Organic Thiols. Angew. Chem., Int. Ed. 2016, 55, 5803−5808. (32) Yu, Z.; Pan, Y.; Shen, Y.; Wang, Z.; Ong, Z.-Y.; Xu, T.; Xin, R.; Pan, L.; Wang, B.; Sun, L.; Wang, J.; Zhang, G.; Zhang, Y. W.; Shi, Y.;

Wang, X. Towards Intrinsic Charge Transport in Monolayer Molybdenum Disulfide by Defect and Interface Engineering. Nat. Commun. 2014, 5, No. 5290. (33) Liu, T.; Wang, C.; Gu, X.; Gong, H.; Cheng, L.; Shi, X.; Feng, L.; Sun, B.; Liu, Z. Drug Delivery with PEGylated MoS2 Nano-sheets for Combined Photothermal and Chemotherapy of Cancer. Adv. Mater. 2014, 26, 3433−3440. (34) Liu, T.; Wang, C.; Cui, W.; Gong, H.; Liang, C.; Shi, X.; Li, Z.; Sun, B.; Liu, Z. Combined Photothermal and Photodynamic Therapy Delivered by PEGylated MoS2 Nanosheets. Nanoscale 2014, 6, 11219−11225. (35) Liu, T.; Shi, S.; Liang, C.; Shen, S.; Cheng, L.; Wang, C.; Song, X.; Goel, S.; Barnhart, T. E.; Cai, W.; Liu, Z. Iron Oxide Decorated MoS2 Nanosheets with Double PEGylation for Chelator-Free Radiolabeling and Multimodal Imaging Guided Photothermal Therapy. ACS Nano 2015, 9, 950−960. (36) Yin, W.; Yan, L.; Yu, J.; Tian, G.; Zhou, L.; Zheng, X.; Zhang, X.; Yong, Y.; Li, J.; Gu, Z.; Zhao, Y. High-Throughput Synthesis of SingleLayer MoS2 Nanosheets as a Near-Infrared Photothermal-Triggered Drug Delivery for Effective Cancer Therapy. ACS Nano 2014, 8, 6922−6933. (37) Kim, D.; Jeong, Y. Y.; Jon, S. A Drug-Loaded Aptamer−Gold Nanoparticle Bioconjugate for Combined CT Imaging and Therapy of Prostate Cancer. ACS Nano 2010, 4, 3689−3696. (38) Xiao, Z.; Ji, C.; Shi, J.; Pridgen, E. M.; Frieder, J.; Wu, J.; Farokhzad, O. C. DNA Self-Assembly of Targeted Near-InfraredResponsive Gold Nanoparticles for Cancer Thermo-Chemotherapy. Angew. Chem., Int. Ed. 2012, 51, 11853−11857. (39) Mo, R.; Jiang, T.; Gu, Z. Enhanced Anticancer Efficacy by ATPMediated Liposomal Drug Delivery. Angew. Chem. 2014, 126, 5925− 5930. (40) Mo, R.; Jiang, T.; DiSanto, R.; Tai, W.; Gu, Z. ATP-Triggered Anticancer Drug Delivery. Nat. Commun. 2014, 5, No. 3364. (41) Lee, T.; Min, S. H.; Gu, M.; Jung, Y. K.; Lee, W.; Lee, J. U.; Seong, D. G.; Kim, B.-S. Layer-by-Layer Assembly for Graphene-Based Multilayer Nanocomposites: Synthesis and Applications. Chem. Mater. 2015, 27, 3785−3796. (42) Lvov, Y.; Ariga, K.; Ichinose, I.; Kunitake, T. Layer-by-Layer Architectures of Concanavalin A by Means of Electrostatic and Biospecific Interactions. J. Chem. Soc., Chem. Commun. 1995, 22, 2313−2314. (43) Richardson, J. J.; Björnmalm, M.; Caruso, F. Technology-Driven Layer-by-Layer Assembly of Nanofilms. Science 2015, 348, No. aaa2491. (44) Ji, Q.; Honma, I.; Paek, S.-M.; Akada, M.; Hill, J. P.; Vinu, A.; Ariga, K. Layer-by-Layer Films of Graphene and Ionic Liquids for Highly Selective Gas Sensing. Angew. Chem., Int. Ed. 2010, 49, 9737− 9739. (45) Mashtalir, O.; Naguib, M.; Mochalin, V. N.; Dall’Agnese, Y.; Heon, M.; Barsoum, M. W.; Gogotsi, Y. Intercalation and Delamination of Layered Carbides and Carbonitrides. Nat. Commun. 2013, 4, No. 1716. (46) Correa, S.; Choi, K. Y.; Dreaden, E. C.; Renggli, K.; Shi, A.; Gu, L.; Shopsowitz, K. E.; Quadir, M. A.; Ben-Akiva, E.; Hammond, P. T. Highly Scalable, Closed-Loop Synthesis of Drug-Loaded, Layer-byLayer Nanoparticles. Adv. Funct. Mater. 2016, 26, 991−1003. (47) Zakaria, M. B.; Li, C.; Ji, Q.; Jiang, B.; Tominaka, S.; Ide, Y.; Hill, J. P.; Ariga, K.; Yamauchi, Y. Self-Construction from 2D to 3D: OnePot Layer-by-Layer Assembly of Graphene Oxide Sheets Held Together by Coordination Polymers. Angew. Chem., Int. Ed. 2016, 55, 8426−8430. (48) Setyawati, M. I.; Kutty, R. V.; Leong, D. T. DNA Nanostructures Carrying Stoichiometrically Definable Antibodies. Small 2016, 12, 5601−5611. (49) Tay, C. Y.; Yuan, L.; Leong, D. T. Nature-Inspired DNA Nanosensor for Real-Time in Situ Detection of mRNA in Living Cells. ACS Nano 2015, 9, 5609−5617. J

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces (50) Qian, H.; Tay, C. Y.; Setyawati, M. I.; Chia, S. L.; Lee, D. S.; Leong, D. T. Protecting microRNAs from RNase degradation with steric DNA nanostructures. Chem. Sci. 2017, 8, 1062−1067. (51) Lee, D. S.; Qian, H.; Tay, C. Y.; Leong, D. T. Cellular processing and destinies of artificial DNA nanostructures. Chem. Soc. Rev. 2016, 45, 4199−4225. (52) Chia, S. L.; Tay, C. Y.; Setyawati, M. I.; Leong, D. T. Nanotoxicity: Biomimicry 3D Gastrointestinal Spheroid Platform for the Assessment of Toxicity and Inflammatory Effects of Zinc Oxide Nanoparticles. Small 2015, 11, 760. (53) Tay, C. Y.; Muthu, M. S.; Chia, S. L.; Nguyen, K. T.; Feng, S.-S.; Leong, D. T. Reality Check for Nanomaterial-Mediated Therapy with 3D Biomimetic Culture Systems. Adv. Funct. Mater. 2016, 26, 4046− 4065.

K

DOI: 10.1021/acsami.7b02529 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX