Design of Small Nanoparticles Decorated with Amphiphilic Ligands

Jun 18, 2019 - ... with the spring constant KS = 128kBT and the equilibrium bond length req = 0.7rc. The force constraining the variation of bond angl...
0 downloads 0 Views 6MB Size
Research Article www.acsami.org

Cite This: ACS Appl. Mater. Interfaces 2019, 11, 23822−23831

Design of Small Nanoparticles Decorated with Amphiphilic Ligands: Self-Preservation Effect and Translocation into a Plasma Membrane Yuchi Liu,†,⊥ Shixin Li,‡,⊥ Xuejuan Liu,† Hainan Sun,§ Tongtao Yue,*,‡ Xianren Zhang,† Bing Yan,*,∥,§ and Dapeng Cao*,† †

State Key Laboratory of Organic-Inorganic Composites, Beijing University of Chemical Technology, Beijing 100029, China State Key Laboratory of Heavy Oil Processing, Center for Bioengineering and Biotechnology, College of Chemical Engineering, China University of Petroleum (East China), Qingdao 266580, China ∥ Key Laboratory for Water Quality and Conservation of the Pearl River Delta, Ministry of Education, Institute of Environmental Research at Greater Bay, Guangzhou University, Guangzhou 510006, China § School of Environmental Science and Engineering, Shandong University, Jinan 250100, China

Downloaded via NOTTINGHAM TRENT UNIV on July 28, 2019 at 05:35:36 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.



S Supporting Information *

ABSTRACT: Design of nanoparticles (NPs) for biomedical applications requires a thorough understanding of cascades of nano-bio interactions at different interfaces. Here, we take into account the cascading effect of NP functionalization on interactions with target cell membranes by determining coatings of biomolecules in biological media. Cell culture experiments show that NPs with more hydrophobic surfaces are heavily ingested by cells in both the A549 and HEK293 cell lines. However, before reaching the target cell, both the identity and amount of recruited biomolecules can be influenced by the pristine NPs’ hydrophobicity. Dissipative particle dynamics (DPD) simulations show that hydrophobic NPs acquire coatings of more biomolecules, which may conceal the properties of the as-engineered NPs and impact the targeting specificity. Based on these results, we propose an amphiphilic ligand coating on NPs. DPD simulations reveal the design principle, following which the amphiphilic ligands first curl in solvent to reduce the surface hydrophobicity, thus suppressing the assemblage of biomolecules. Upon attaching to the membrane, the curled ligands extend and rearrange to gain contacts with lipid tails, thus dragging NPs into the membrane for translocation. Three NP−membrane interaction states are identified that are found to depend on the NP size and membrane surface tension. These results can provide useful guidelines to fabricate ligand-coated NPs for practical use in targeted drug delivery, and motivate further studies of nano-bio-interactions with more consideration of cascading effects. KEYWORDS: nano-bio interaction, cascading effect, biomolecular corona, plasma membrane, translocation



INTRODUCTION Nowadays, bionanotechnology relying on the use of nanoparticles (NPs) is being developed for clinical diagnostics and therapeutics.1,2 In particular, NPs of various physicochemical properties are designed and engineered for delivering drugs, proteins, genes, and other molecules into cells.3,4 The promising application of NPs in drug delivery raises an important question of how NPs interact with the cell membrane.5 Studies on cell uptake using multiple experimental techniques have been conducted to promote our understanding of this issue, whereas the underlying molecular mechanisms still remain unclear, partly due to the rapidly occurring process involving dynamic remodeling of both the NP and the membrane. Besides, upon introduction into biological fluids, NPs acquire coatings of biomolecules, commonly referred to as the “biomolecular corona”,6−8 to increase the effective size of NPs, alter the NP surface properties, and affect the NP aggregation state.9−12 The © 2019 American Chemical Society

amount, composition, orientation, and conformation of biomolecules adsorbed onto NPs are possible factors that may impact the outcome of interactions with the cell membrane.11,13−20 In general, NPs enter cells through different pathways that can be influenced by the NP properties and cell membrane environment.21 In particular, endocytosis encompasses multiple mechanisms by which cells ingest NPs into transport vesicles derived from the plasma membrane. After internalization, NPs are either associated with the intracellular vesicles or freely distributed inside the cytosol. In comparison, the endocytic pathway is thought to prevail for larger NPs or NP aggregates.22−24 For smaller NPs with no assemblage of biomolecules, it was predicted by simulations that they prefer Received: February 27, 2019 Accepted: June 18, 2019 Published: June 18, 2019 23822

DOI: 10.1021/acsami.9b03638 ACS Appl. Mater. Interfaces 2019, 11, 23822−23831

Research Article

ACS Applied Materials & Interfaces to directly penetrate through the membrane,24,25 and the translocation can be further enhanced by applying forces or fields that can be induced, for example, by inherent charge imbalance between the extracellular and intracellular regions.24,26−28 For the application of drug delivery, the internalization pathway of endocytosis may be disadvantageous because the contents from endocytosis can easily enter the endo-lysosomal system to be digested by acid hydrolase.29,30 Therefore, design of small NPs exhibiting the ability to penetrate through the cell membrane is vital for practical use in drug delivery. Previously reported experimental and computational studies on cellular uptake of NPs highlighted the properties of assynthesized NPs in modulation of interactions with cell membranes.21,31−34 Recently, it has become increasingly clear that acquisition of coronas can alter the subsequent fate of NPs interacting with the cell membrane. By preparing a library of gold NPs coated with different ligands, Melby et al. found that the initial NP surface coating can influence interactions with the cell membrane by determining the assemblage of proteins.35 Besides, the presence of biomolecules in relevant media can influence the behaviors of ultrasmall NPs,36,37 the cell uptake of which can occur via both energy-dependent (e.g., endocytosis) and energy-independent (e.g., penetration) manners, depending on the initial surface coverage.38 More recently, using coupled methodologies of flow cytometry and Raman microspectroscopic mapping, Alnasser et al. identified the correlation between the biomolecular corona and specific cell receptors, suggesting a possible mechanism of endogenous interaction.20 We stress the importance of considering cascading effects of engineered NP surface properties on interaction with cell membranes by determining recruitment of biomolecules upon introduction in biological media. It is suggested that a superior surface coating on small NPs for targeted drug delivery should possess two facets of advantages: to prevent the protein assemblage and to maintain the affinity of NPs to membranes for translocation. Motivated by these issues, in this combined experimental and computational study, we utilized a spherical NP model with its surface coated with ligands of different hydrophobicity to investigate nano-bio interactions at both protein and cell membrane interfaces. Experimental data on cell uptake show that hydrophobic NPs are more heavily internalized by cells, while further simulations reveal an increase in the amount of biomolecules adsorbed onto NPs with increasing hydrophobicity. Based on these results, we propose an amphiphilic ligand coating onto NPs. Coarsegrained membrane simulations reveal the design principle, following which the amphiphilic ligands first curl in solvent to decrease the NP hydrophobicity for suppression of protein assemblage. Upon attaching to the membrane, the hidden hydrophobic segments extend to gain contacts with lipid tails, thus dragging NPs into the membrane for translocation.



synthesized by ourselves, with the detailed synthesis procedure given elsewhere.40 The amount of hydrophilic and hydrophobic ligands was 20.3/0, 18.1/0.8, 16.1/1.6, 9.7/2.6, 6.0/3.2, 4.0/4.9, and 0/8.3 mg, respectively, for HY1 to HY7 GNPs. After stirring for 30 min at room temperature, sodium tetrahydroborate (10 mg, 0.246 mmol) in water was added to the mixture dropwise. The solution turned red immediately and was then stirred for 12 h at room temperature. Then, 1 N HCl was added to the reaction mixture dropwise to neutralize the excess sodium tetrahydroborate until the pH reached 7.0. To remove the free ligands and solvent molecules from the NPs, the reaction mixture was centrifuged at 15 000 rpm for 20 min. The colorless supernatant was discarded and the solid was dissolved in deionized water and centrifuged again at 15 000 rpm for 20 min. The wash−centrifugation cycle was repeated five times. Transmission electron microscopy (TEM) analysis of the GNPs was performed using a JEM-1011 (JEOL, Tokyo, Japan). The diameters of GNPs were measured using Image Pro Plus 6.6. The ζ-potential was measured on a Malvern ZetaSizer Nano ZS (Malvern Instruments Ltd, Malvern, U.K.). Internalization of GNPs by A549 and HEK293 Cells. Both the A549 and HEK293 cells were obtained from American Type Culture Collection, and have been widely used for the study of the NP cell uptake.41 All cells were incubated in Dulbecco’s modified Eagle’s medium. Cell culture medium was supplemented with 10% fetal bovine serum (FBS), 10 U/mL penicillin, and 10 mg/mL streptomycin. Cells were seeded into a culture dish (10 cm) with a density of 7.5 × 104 cells/well. Cultures were maintained at 37 °C under humidified conditions with 5% CO2. After 24 h of plating, the medium in the culture dish was discarded, and the solution of GNPs was added at a concentration of 50 μg/mL. Both types of cells were incubated with GNPs for 24 h and washed three times with 500 μL of PBS for each well. Such a procedure was to remove GNPs not internalized by the cells or those just bound to the cell surface for quantification of cell internalization. This clean-up procedure was equally effective in the case of both hydrophobic and hydrophilic GNPs, according to our previous studies.39,42 For TEM experiments, the cells were fixed with 2.5% glutaraldehyde in 0.1 M of Nacacodylate buffer for 30 min at room temperature. The fixed cells were collected for TEM measurement. TEM images were taken using a JEOL 1200 EX transmission electron microscope (JEOL, Tokyo, Japan) at 80 kV. The images were acquired using an AMT 2k CCD camera. For quantification of the cell uptake, cells were incubated with 200 μL of trypsin−ethylenediaminetetraacetic acid solution for 3 min at 37 °C. Then 200 μL of medium with 10% FBS was added to each well to terminate the trypsin digestion. The detached cells in each well were counted and incubated with 400 μL aqua regia at 37 °C for 12 h. The cell uptake of GNPs was then quantified using inductively coupled plasma mass spectrometry (ICP-MS). Coarse-Grained Models and Simulation System Setup. There are three components considered in the simulation system: ligand-coated NPs, model biomolecules, and a lipid bilayer membrane (Figure S1). The NP model was constructed by arranging a number of hydrophilic beads in a spherical shape of a defined diameter. The ligand molecule coating on the NP surface was coarse-grained into a linear chain of beads. The hydrophobicity of each bead can be tuned by varying the interaction parameter (Tables S2 and S3). Note that biomolecules are diverse and variable in real biological fluids, making molecular modeling of such complex systems rather challenging. Although most proteins are charged, generally containing both positively and negatively charged residues on their surfaces, electrostatic interactions were not considered in this work as only the surface hydrophobicity of NPs was tuned to modulate the recruitment of biomolecules. To reduce the complexity, a simplified model of linear amphiphilic chain with a random distribution of hydrophobic and hydrophilic beads was used to represent the biomolecules interacting with ligand-coated NPs. Due to simplification of the coarse-grained models used in our simulations, any direct quantitative comparison between our simulation results and experimental measurements should be conducted with caution. The model cell plasma membrane was prepared by arranging defined

MATERIALS AND METHODS

Synthesis and Characterization of Gold Nanoparticles (GNPs). In our previous work, we have synthesized a library of GNPs with a gradual change in surface hydrophobicity (HY).39 Taking GNPs decorated with either purely hydrophilic or purely hydrophobic ligands for instance, 1.25 mL of hydrogen tetrachloroaurate(III) trihydrate solution (1.25 mL, 0.061 mmol) was added to a solution of hydrophilic or hydrophobic ligand in dimethylformamide (see Table S1 for the chemical structure of both ligands). Both the hydrophilic and hydrophobic ligands were 23823

DOI: 10.1021/acsami.9b03638 ACS Appl. Mater. Interfaces 2019, 11, 23822−23831

Research Article

ACS Applied Materials & Interfaces

Figure 1. Cellular uptake of GNPs decorated with ligands of varying hydrophobicity. (a−d) TEM images of interactions between A549 cells and GNPs decorated with purely hydrophilic ligands (HY1). (e−h) TEM images of interactions between A549 cells and GNPs decorated with purely hydrophobic ligands (HY7). (i) The amount of both purely hydrophilic (HY1) and purely hydrophobic (HY7) GNPs internalized by A549 cells. (j) The amount of both purely hydrophilic (HY1) and purely hydrophobic (HY7) GNPs internalized by HEK293 cells. (k) The cellular uptake of GNPs in both cell lines as a function of the surface hydrophobicity (HY1−HY7). Error bars in (i)−(k) indicate mean ± standard deviations corresponding to three replicates. numbers of lipid molecules in a bilayer of lateral size 40 nm × 40 nm. The number of lipids inside the membrane was adjusted to represent different values of the membrane surface tension.43 Each lipid was constructed by connecting three hydrophilic beads with two tails, each containing five hydrophobic beads. Such a coarse-grained model proposed by Groot and Rabone represents dimyristoylphosphatidylcholine and has been found to form stable bilayers and show typical phase behaviors of the bilayer.44 In recent years, such a coarse-grained lipid model has been widely used for studies of membrane-mediated protein aggregation and NP−membrane interactions.19,45−47 The solid NP was constrained to move as a rigid body during the simulation. Water molecules, which were modeled as single beads, and other components were prevented from penetrating into the NPs. As only the NP hydrophobicity was tuned to modulate the recruitment of biomolecules, electrostatic interactions were not considered, and no ions were included in the simulation box. Dissipative Particle Dynamics. Multiple computational techniques have been developed for the study of NP−membrane interactions.23,48,49 Simulations presented in this work are based on the dissipative particle dynamics (DPD),50 which is a mesoscopic coarse-grained simulation method taking into account hydrodynamic interactions. DPD was first introduced to simulate the hydrodynamic behavior of complex fluids51 and proven to be particularly suitable for studying nano-biointeractions at both protein and membrane interfaces.24,25,46,52,53 In DPD, a small group of atoms is coarsegrained into a single bead to decrease the molecular degree of freedom, thus allowing simulations to run on a larger scale for a longer time than using the atomistic model. During DPD simulations, the elementary units are soft beads whose dynamics are governed by Newton’s equations of motion, dri/dt = vi and dvi/dt = f i/mi, similar

to the molecular dynamics method. Typically, beads i and j interact with each other via a short-ranged repulsive force consisting of three types of forces: conservative force (FCij ), dissipative force (FDij ), and random force (FRij ). The conservative force between beads i and j is

{

determined by FijC = aij riĵ max 1 −

rij rc

}

, 0 , where aij is the maximum

repulsive strength, rij = rj − ri (ri and rj are positions of beads i and j), rij = |rij| is the distance between the two beads, r̂ij = rij/|rij| is the unit vector, and rc is the cutoff radius. Note that in DPD all interactions are soft and repulsive. If the interaction parameter is larger than 25 (the water−water interaction parameter), the interaction can be effectively regarded as repulsive. Otherwise, the interaction is attractive if the parameter is smaller than 25. The values of interaction parameters aij are shown in Tables S2 and S3, respectively, for the biomolecular and membrane systems. Briefly, the dissipative force is determined by

(

FijD = − γ 1 −

rij 2 rc

) (r̂ ·v )r̂ , where γ is the friction coefficient, and ij ij

ij

vij = vi − vj, (vi and vj are the velocities of the two beads). The

(

random force is given by FijR = − σ 1 −

rij 2 rc

) θ r̂ , in which σ is the ij ij

noise amplitude, and θij is an uncorrelated random variable with zero mean and unit variance. In models of lipids and biomolecules, the interaction between neighboring beads in the same molecule is described by a harmonic spring force, given by FS = KS(rij − req)r̂ij, with the spring constant KS = 128kBT and the equilibrium bond length req = 0.7rc. The force constraining the variation of bond angle is given by Fφ = −∇Uφ, where Uφ = Kφ(1 − cos(ϕ − ϕ0)) is the bond angle potential with the constant Kφ = 10.0 and the equilibrium angle ϕ0 = π. 23824

DOI: 10.1021/acsami.9b03638 ACS Appl. Mater. Interfaces 2019, 11, 23822−23831

Research Article

ACS Applied Materials & Interfaces

Figure 2. Assemblage of coarse-grained biomolecules onto NPs decorated with ligands of different hydrophilic−hydrophobic ratios: 0:16, 8:8, and 12:4. (a) Time evolutions of the amount of biomolecules adsorbed onto NPs of different ligand hydrophobicity. (b−d) Final simulated snapshots for different cases: (b) 16:0; (c) 8:8, and (d) 4:12. Beads in the NP core are colored pink. Hydrophilic and hydrophobic beads in ligands are colored blue and red, respectively. Hydrophilic beads in biomolecules are colored cyan. Water beads are not shown for clarity. In this work, we mainly performed two series of simulations: one is the assemblage of model biomolecules complexing ligand-coated NPs, focusing on the effect of surface hydrophobicity of NPs on the adsorption of biomolecules. Next, we studied the interaction of NPs coated with ligands of different hydrophobicity with a lipid bilayer membrane. All simulations were performed in the canonical (NVT) ensemble, i.e., the total particle number (N), box volume (V), and temperature (T) were kept constant during the simulation. Periodic boundary conditions were adopted in all three directions. The time evolution of the systems was obtained from a modified version of the velocity-Verlet algorithm with a time step of Δt = 0.02.51

hydrophilic GNPs, possibly due to protein adsorption. By contrast, sizes of hydrophobic GNPs remained the same or slightly decreased likely due to the effects of both protein adsorption (size increase) and breakdown of aggregates (size decrease, see Figure S3). Furthermore, after protein adsorption, ζ-potential values of all GNPs were shifted from −10 to −40 mV, suggesting higher suspension stability due to stronger repulsion between individual GNPs. By observing GNP−cell interactions with TEM (Figure S4), we rarely observed GNP aggregation when they encountered cells. To quantitatively determine the effect of GNP hydrophobicity on the cell uptake, cells were washed three times with PBS to remove GNPs that were not internalized or bound to the cell surface. Then the number of GNPs internalized by cells was counted using ICP-MS analysis. As is seen in Figure 1i, the amount of hydrophobic GNPs internalized by A549 cells was over 40-fold higher than that of hydrophilic GNPs, suggesting that GNPs coated with more hydrophobic ligands are more heavily internalized by A549 cells. Our previous works have shown that the cell uptake of GNPs in different cell lines is quite similar.39,42,63 To examine whether the above experimental finding is general or specific only for the A549 cell line, we then repeated the cell uptake experiments using the HEK293 cell line. The selection of A549 and HEK293 is because the former is a popular cell model for lung exposure and the latter for kidney. When NPs are inhaled into lungs, they can be easily absorbed into blood and then translocated to the liver and kidneys. Therefore, both cell lines have been commonly used for the study of possible mechanisms of cell uptake. Our experiments showed similar results of cell uptake, i.e., the hydrophobic GNPs were more heavily internalized by the HEK293 cells (Figure 1j). To systematically understand the effect of NP hydrophobicity on the cell uptake, we further compared the amount of internalized GNPs with a gradual change in hydrophobicity



RESULTS AND DISCUSSION Cell Uptake of GNPs. Decades of experimental and theoretical studies have promoted our understanding of cell uptake of various NPs, with both the pathway and the mechanism being expected to depend on the physicochemical properties of NPs, such as size,54,55 shape,24,56−58 elasticity,33,49,59,60 and surface chemistry.34,61,62 In our previous work, we have synthesized a library of GNPs of an average diameter 6 nm with a gradual change in hydrophobicity by varying the ratio of hydrophobic/hydrophilic ligands coated on the GNP surface (Table S1, Figures S2 and S3).39 Here, we focus on the effect of GNP surface hydrophobicity on cell uptake. After adding GNPs (50 μg/mL) coated with purely hydrophilic ligands (HY1) to A549 cells and incubating GNPs with cells for 24 h, most hydrophilic GNPs were found to locate outside of the cells, and only a small number of GNPs were internalized by the cell (Figure 1a−d). For GNPs coated with purely hydrophobic ligands (HY7), the extent of cell uptake was strikingly increased (Figure 1e−h). Given that hydrophobic GNPs exhibited higher aggregation tendency than more hydrophilic GNPs in aqueous solution, we suspended all GNPs by sonicating them in cell culture medium containing serum proteins. In separate experiments, we found an increase in the hydrodynamic diameter of 23825

DOI: 10.1021/acsami.9b03638 ACS Appl. Mater. Interfaces 2019, 11, 23822−23831

Research Article

ACS Applied Materials & Interfaces

value, and ionic strength), and protein source, are thought to influence the formation and composition of coronas. In particular, there exist plenty of proteins and other biomolecules of different types that may interact with exotic NPs synergistically or competitively.10 In addition to the amount of biomolecules recruited to the corona, the identity of recruited biomolecules could also be influenced by the pristine NP hydrophobicity. By preparing a library of gold NPs coated with different ligands, it was found that each type of functionalized NPs formed complexes with a unique ensemble of serum proteins that depended on the initial surface coating of the NPs.35 Besides, natural plasma proteins are usually folded under normal conditions with defined secondary and tertiary structures. Although the hydrophobic core can be opened via binding to NPs, which may induce protein denaturation,67,68 the protein assemblage complexing NPs can also be driven by other forces, such as electrostatic interactions and H-bonding. Despite these disparities hindering direct experimental verification of the self-preservation effect revealed by the above simulations, earlier experiments have demonstrated the hydrophobic interaction as a main force governing interactions between NPs and biomolecules.69,70 Since nearly all biomolecules are amphiphilic, we speculate that NPs of different surface hydrophobicity exhibit different capability for recruiting surrounding biomolecules. By calculating the free energy change of protein adsorption onto NPs of different hydrophobicity, it was found that BSA proteins can spontaneously adsorb onto hydrophobic NP surfaces, whereas there exists a finite energy barrier hindering adsorption onto hydrophilic NP surfaces, suggesting the crucial role of NP surface hydrophobicity in acquisition of biomolecular corona.18 Membrane Response to Ligand-Coated NPs. Although the above simulations suggested that the amount of biomolecules recruited by NPs can be effectively reduced by decreasing the NP surface hydrophobicity, NPs coated with purely hydrophilic ligands may not be applicable for drug delivery due to the lower level of cell internalization revealed by our experiments (Figure 1). We, thus, proposed an amphiphilic ligand coating that can suppress biomolecular assemblage in the self-preservation mechanism revealed by our simulations (Figure 2). Next, we focus on the interaction between modeled plasma membranes and NPs decorated with amphiphilic ligands. To that end, each ligand-coated NP was positioned in close proximity above the top surface of a membrane of lateral dimensions 40 nm × 40 nm. Three NP diameters were considered, including 3 nm, 5 nm, and 10 nm, respectively. The hydrophobic ratio of ligands was fixed at 4:12, which has been demonstrated to effectively suppress the adsorption of biomolecules on the NP surface. Note that multiple active mechanisms can be used by living cells to regulate their membrane surface tension.71,72 For example, the contraction of the membrane-associated actomyosin cytoskeleton often leads to a contracted membrane with negative surface tension.73 Previous computations have confirmed the linear relation between membrane surface tension and lipid density in a defined range.43,74 Here, we prepared five lipid bilayers with the lipid density ranging from 1.2 to 1.5. The average surface tension was calculated as γ = ⟨Pzz − (Pxx + Pyy)/2⟩Lz with the diagonal components of the pressure tensor

(HY1−HY7). The statistical data showed that the amount of internalized GNPs increased with their hydrophobicity (Figure 1k), suggesting that the hydrophobicity dependence of NP cell uptake can be general for different types of cells. Protein Assemblage onto NPs of Different Hydrophobicity. The above experimental results have suggested that the cell uptake of NPs is strongly influenced by the NP surface hydrophobicity. As discussed earlier, NPs are thought to acquire coatings of biomolecules in biological fluids that may impact subsequent interactions with the cell membrane. For practical use in drug delivery, the protein assemblage on the NP surface may reduce the targeting specificity to induce potential side effects. Earlier experiments have shown that specific interactions between protein coatings on NPs and the cell surface influence the outcome of cell uptake.20 However, due to the high complexity of proteins in biological fluids (blood, lymph, cytoplasm, respiratory tract fluid, and cell culture media), precise regulation of the protein assemblage via change of the NP physicochemical properties should be rather difficult. Therefore, a good functionalization of NPs designed for targeted drug delivery should avoid or suppress acquisition of protein coatings before reaching the target cell. To understand how the assemblage of biomolecules is influenced by the NP surface hydrophobicity, we utilized a simplified model of biomolecules and a spherical NP model with its surface coated with ligands of different hydrophobicity to perform coarse-grained simulations. Considering the larger hydrodynamic diameter of GNPs measured in experiments, a larger NP of diameter 10 nm was used in our simulations. Each NP was uniformly coated with 114 ligand chains with the hydrophobicity of each ligand tuned by varying the number of hydrophobic beads at the terminus (Figure S1). For clarity, we defined the hydrophobic ratio as the ratio between hydrophobic and hydrophilic bead numbers in each ligand, and three ratios were considered, including 16:0, 8:8, and 4:12. Note that the ligand density was lower than that for GNPs used in our experiments (Table S1). Shown in Figure 2a are time evolutions of the number of biomolecules adsorbed onto NPs coated with ligands of three hydrophobic ratios. As is seen, after a finite simulation time, apparently more biomolecules were adsorbed onto NPs coated with ligands of higher hydrophobic ratios. Given the fact that NPs coated with amphiphilic ligands all have similar original hydrophobicity due to the presence of hydrophobic termini, we ascribe the difference in protein adsorption to the conformational change of such amphiphilic ligands. We monitored the change in the ligand conformation and the process of protein adsorption. Before encountering surrounding biomolecules, the amphiphilic ligands first curled to prevent exposure of the hydrophobic segments to solvent. Such a conformation, in which the hydrophilic segments pointed outward to form a “hydrophilic” shield, can decrease the effective surface hydrophobicity, thus generating a self-preservation effect to reduce contacts between hydrophobic segments with surrounding biomolecules. With an increase in the hydrophobic ratio, it became more difficult to sufficiently hide hydrophobic segments inside the shield. Besides, the thinner hydrophilic shield can be perturbed by the surrounding biomolecules more easily to gain contacts with the hidden hydrophobic segments. We note that our coarse-grained assembly simulations may diverge from the high complexity of real biological systems, in which a variety of factors, such as the protein concentration,12 NP size,64 incubation condition65,66 (e.g., temperature, pH

(

∂U

)

Pαα = NkBT − ∑i αi ∂α /V , where α ∈ {x, y, z}. As i

expected, the calculated surface tension decreased linearly 23826

DOI: 10.1021/acsami.9b03638 ACS Appl. Mater. Interfaces 2019, 11, 23822−23831

Research Article

ACS Applied Materials & Interfaces

Figure 3. Membrane interaction states of NPs coated with amphiphilic ligands of a hydrophilic/hydrophobic ratio 12:4. (a) Phase diagram of the interaction states as functions of the NP size and the target membrane surface tension. (b−d) Final simulated snapshots of the penetration (b), adhesion (b), and engulfment (d) states. Each state is displayed from both top and cross-sectional views. Lipid heads and tails are colored cyan and yellow, respectively. Lipid tails are not displayed in the snapshot from the cross-sectional view for clarity.

from 5.8 to −1.48 with the increase of lipid density (Figure S5). All typical interaction states identified by our simulations are summarized in a phase diagram shown in Figure 3a. Depending on the NP size and the membrane tension, NPs were found to interact with the membrane in three distinct pathways, including penetration, adhesion, and engulfment. Scrutinizing typical snapshots in Figure 3b, we can find that in the final state of penetration, the amphiphilic-ligand-coated NP finally penetrated into the membrane. A hydrophilic pore was opened, in which the amphiphilic-ligand-coated NP was anchored. Under actual circumstances, successfully entering the lipid bilayer via opening hydrophilic pores means that NPs are likely to enter cells directly without the formation of a lipid shell and realize their biomedical functions. According to the phase diagram, the state of penetration occurred preferentially for NPs of sizes comparable to the membrane thickness when the membrane has a higher surface tension. In the second state of NP adhesion that occurred preferentially under lower surface tensions, NPs interact only with the outer leaflet of the membrane but failed to penetrate through the membrane to reach the inner leaflet (Figure 3c). In this case, complete NP translocation is hard to be achieved. In the third case, as shown in Figure 3d, larger NPs entered the inner core of the lipid bilayer and remained inside the bilayer, but were engulfed by the swollen membrane. Such a state was observed for membranes under lower surface tensions. Besides the amphiphilic-ligand-coated NPs, we also examined membrane interactions of NPs decorated with purely hydrophilic ligands, aiming to interpret why hydrophilic GNPs were less readily internalized by the cell (Figure 1). Our simulations revealed that most NPs preferred to adhere on the membrane surface, except for one NP of diameter 5 nm, which was found to penetrate into the membrane under a high membrane surface tension (Figure S6). Pathways of Membrane Interaction of AmphiphilicLigand-Coated NPs. Combining the above experimental and simulation results, we stress that NPs decorated with either purely hydrophilic or purely hydrophobic ligands may not be adequate for practical use in targeted drug delivery. It was suggested that purely hydrophilic-ligand-coated NPs are hard to be internalized by cells, whereas purely hydrophobic-ligandcoated NPs exhibit a strong ability to acquire coatings of biomolecules, thus reducing the targeting specificity. By contrast, we propose an amphiphilic ligand coating that can be used for both suppressing biomolecular assemblage and

achieving efficient membrane translocation. To reveal the mechanism following which the design strategy can be optimized for practical use in targeted drug delivery, we further analyzed, at the molecular level, the pathways of three membrane interaction states of amphiphilic-ligand-coated NPs under different simulation conditions. The hydrophobic ratio of ligands was fixed at 4:12, and similar membrane responses were observed when the ratio was set to 8:8. Time sequences of typical snapshots depicting the processes of membrane interactions of amphiphilic-ligand-coated NPs of three different sizes are given in Figure 4. In all three cases, the

Figure 4. Time sequences of typical snapshots depicting membrane interaction processes of amphiphilic-ligand-coated NPs under different conditions. (a) Penetration of a NP of diameter 5 nm through the membrane of lipid surface density 1.2. (b) Membrane adhesion of a smaller NP of diameter 3 nm. (c) Membrane engulfment on a larger NP of diameter 10 nm.

hydrophobic part of each ligand first interacted with the membrane and quickly entered the hydrophobic core to generate a pulling force that dragged NPs toward the membrane. Notably, the final membrane response to NPs is NP size dependent. For the NP of diameter 5 nm, which is similar to the membrane thickness, membrane insertion of all ligands caused embedding of NPs into the membrane. Due to hydrophilicity of both the proximal segments of ligands and the bare NP surface, surrounding lipids were rearranged to open a hydrophilic pore, in which the NP was anchored 23827

DOI: 10.1021/acsami.9b03638 ACS Appl. Mater. Interfaces 2019, 11, 23822−23831

Research Article

ACS Applied Materials & Interfaces

the membrane (Figure 4). Shown in Figure 6 are time evolutions of the average angle between vectors of ligands and the membrane normal direction. Combining typical snapshots (Figure 4), both the orientation and conformation of ligands evolved in different manners during membrane interactions in three representative pathways. Before membrane attachment, the average angle fluctuated around 90°, suggesting that all ligands orientated randomly albeit curling to prevent unfavorable contacts between hydrophobic segments and the solvent. Upon being captured by the membrane via vertical insertion of ligands into the membrane, we found a rapid decrease of the average angle (Figure 6a,b), except for the engulfment state where the steric hindrance due to the larger NP size hindered vertical insertion of more ligands (Figure 6c). For the state of membrane adhesion, due to the small NP size (3 nm), ligands inserted into the membrane remained tilted to maintain the average angle fluctuating around a lower value (Figure 6a). As we increased the NP diameter to 5 nm, which is comparable to the membrane thickness, we found a further rearrangement of ligands from being vertical to horizontal to the membrane surface, aiming to further increase contacts with the surrounding lipid tails. Such a ligand movement was clearly reflected in a gradual increase in the average angle (Figure 6b). Nearly no change in the average angle was observed for a larger NP of 10 nm (Figure 6c), suggesting that the final membrane engulfment was achieved with minor rearrangement of the ligands.

(Figure 4a). For the NP of diameter 3 nm, which is smaller than the membrane thickness, it kept floating on the upper leaflet of the membrane. Although all ligands were inserted into the membrane, the small NP size hindered opening of the hydrophilic pore, thus inducing a state of adhesion (Figure 4b). In contrast, for a larger NP of 10 nm, the steric hindrance was formed that prevented the top ligands from entering the membrane due to the longer distance. Instead, lipid molecules were found to climb along the NP surface to gain contacts with the ligands, finally leading to a state of engulfment (Figure 4c). The size-dependent membrane response to NP invasion is also influenced by the membrane surface tension, as revealed by evolutions of the NP position along the membrane normal direction (Figure 5). For smaller NPs of 3 nm, the NP



CONCLUDING REMARKS The biological interactions of NPs of various physicochemical properties have been studied extensively over the past decades,5 whereas relatively little is known about the cascading effect that should be taken into account when designing NPs for any biomedical application. In particular, NPs acquire coatings of biomolecules upon introduction into biological fluids, thus impacting both the targeting specificity and fate of subsequent interactions with cell membranes. In this work, we have combined experiments, coarse-grained assembly, and membrane simulations to provide an integrated picture of cascading interactions of NPs with both biomolecules and cell membranes. Cell culture experiments showed that NPs of higher hydrophobicity are heavily internalized by cells, but acquire coatings of more biomolecules that may reduce the targeting specificity, as indicated by our simulations. Based on these results, we proposed an amphiphilic ligand coating on

Figure 5. Time evolutions of the center of mass positions of NPs of different diameters along the membrane normal direction under different membrane surface tensions. The membrane regions are labeled with green color. Three NP diameters are considered, including 3 nm (a), 5 nm (b), and 10 nm (c).

insertion into the upper leaflet of the membrane was more time consuming under a lower surface tension, due to the closer packing of lipid molecules hindering penetration of NPs (Figure 5a). With an increase in the NP size, the effect of membrane surface tension became less evident as more ligands coating on larger NPs provided a higher driving force to facilitate penetration into the membrane (Figure 5b,c). At the molecular level, the NP invasion into membranes was accomplished via rearranging ligands as being immersed into

Figure 6. Time evolutions of the average angle between ligands and the Z-axis for three typical interaction states. The inset shows a schematic illustration of three distinct membrane interaction pathways of amphiphilic-ligand-coated NPs of different sizes. (a) NPs with size smaller than the membrane thickness adhere on the membrane surface via vertical insertion of ligands into the membrane. (b) NPs with size close to the membrane thickness first adhere on the membrane surface via vertical ligand insertion, then penetrate into the membrane via further rearrangement of ligands. (c) NPs larger than the membrane thickness are engulfed by the membrane with nearly no change in the ligand arrangement. 23828

DOI: 10.1021/acsami.9b03638 ACS Appl. Mater. Interfaces 2019, 11, 23822−23831

ACS Applied Materials & Interfaces



NPs for both preventing biomolecular adsorption and achieving membrane translocation. Our simulations revealed the mechanism. Before reaching the target cell membrane, the amphiphilic ligands first curled to hide their hydrophobic segments, thus decreasing the surface hydrophobicity to suppress adsorption of biomolecules. Upon attaching to the membrane, the curled ligands extend to gain contacts with lipid tails, thus dragging NPs into the membrane for translocation. Further simulations identified three states of NP−membrane interactions, including penetration, adhesion, and engulfment, depending on the NP size and the membrane surface tension. Although current experimental and simulation systems diverge from the real complexity of biological environments, these results are expected to facilitate molecular-level understanding on nano-bio interactions, and provide motivation for future studies on designing intelligent NPs for biomedical use with more consideration of the cascading effects.



REFERENCES

(1) Petros, R. A.; DeSimone, J. M. Strategies in the Design of Nanoparticles for Therapeutic Applications. Nat. Rev. Drug Discovery 2010, 9, 615. (2) Ferrari, M. Cancer Nanotechnology: Opportunities and Challenges. Nat. Rev. Cancer 2005, 5, 161. (3) Singh, R.; Lillard, J. W., Jr. Nanoparticle-Based Targeted Drug Delivery. Exp. Mol. Pathol. 2009, 86, 215−223. (4) Blanco, E.; Shen, H.; Ferrari, M. Principles of Nanoparticle Design for Overcoming Biological Barriers to Drug Delivery. Nat. Biotechnol. 2015, 33, 941. (5) Nel, A. E.; Mädler, L.; Velegol, D.; Xia, T.; Hoek, E. M.; Somasundaran, P.; Klaessig, F.; Castranova, V.; Thompson, M. Understanding Biophysicochemical Interactions at the Nano-Bio Interface. Nat. Mater. 2009, 8, 543. (6) Walkey, C. D.; Chan, W. C. Understanding and Controlling the Interaction of Nanomaterials with Proteins in a Physiological Environment. Chem. Soc. Rev. 2012, 41, 2780−2799. (7) Tenzer, S.; Docter, D.; Kuharev, J.; Musyanovych, A.; Fetz, V.; Hecht, R.; Schlenk, F.; Fischer, D.; Kiouptsi, K.; Reinhardt, C.; et al. Rapid Formation of Plasma Protein Corona Critically Affects Nanoparticle Pathophysiology. Nat. Nanotechnol. 2013, 8, 772. (8) Ke, P. C.; Lin, S.; Parak, W. J.; Davis, T. P.; Caruso, F. A Decade of the Protein Corona. ACS Nano 2017, 11, 11773−11776. (9) Monopoli, M. P.; Walczyk, D.; Campbell, A.; Elia, G.; Lynch, I.; Baldelli Bombelli, F.; Dawson, K. A. Physical-Chemical Aspects of Protein Corona: Relevance to in vitro and in vivo Biological Impacts of Nanoparticles. J. Am. Chem. Soc. 2011, 133, 2525−2534. (10) Casals, E.; Pfaller, T.; Duschl, A.; Oostingh, G. J.; Puntes, V. Time Evolution of the Nanoparticle Protein Corona. ACS Nano 2010, 4, 3623−3632. (11) Yang, J. A.; Lohse, S. E.; Murphy, C. J. Tuning Cellular Response to Nanoparticles via Surface Chemistry and Aggregation. Small 2014, 10, 1642−1651. (12) Monopoli, M. P.; Åberg, C.; Salvati, A.; Dawson, K. A. Biomolecular Coronas Provide the Biological Identity of Nanosized Materials. Nat. Nanotechnol. 2012, 7, 779. (13) Wang, L.; Li, J.; Pan, J.; Jiang, X.; Ji, Y.; Li, Y.; Qu, Y.; Zhao, Y.; Wu, X.; Chen, C. Revealing the Binding Structure of the Protein Corona on Gold Nanorods Using Synchrotron Radiation-based Techniques: Understanding the Reduced Damage in Cell Membranes. J. Am. Chem. Soc. 2013, 135, 17359−17368. (14) Ge, C.; Du, J.; Zhao, L.; Wang, L.; Liu, Y.; Li, D.; Yang, Y.; Zhou, R.; Zhao, Y.; Chai, Z.; Chen, C. Binding of Blood Proteins to Carbon Nanotubes Reduces Cytotoxicity. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 16968−16973. (15) Walkey, C. D.; Olsen, J. B.; Song, F.; Liu, R.; Guo, H.; Olsen, D. W. H.; Cohen, Y.; Emili, A.; Chan, W. C. Protein Corona Fingerprinting Predicts the Cellular Interaction of Gold and Silver Nanoparticles. ACS Nano 2014, 8, 2439−2455. (16) Wang, X.; Wang, M.; Lei, R.; Zhu, S. F.; Zhao, Y.; Chen, C. Chiral Surface of Nanoparticles Determines the Orientation of Adsorbed Transferrin and Its Interaction with Receptors. ACS Nano 2017, 11, 4606−4616. (17) Kokkinopoulou, M.; Simon, J.; Landfester, K.; Mailänder, V.; Lieberwirth, I. Visualization of the Protein Corona: Towards a Biomolecular Understanding of Nanoparticle-Cell-Interactions. Nanoscale 2017, 9, 8858−8870. (18) Ding, H.-m.; Ma, Y.-q. Computer Simulation of the Role of Protein Corona in Cellular Delivery of Nanoparticles. Biomaterials 2014, 35, 8703−8710. (19) Bai, X.; Xu, M.; Liu, S.; Hu, G. Computational Investigations of the Interaction between the Cell Membrane and Nanoparticles Coated with a Pulmonary Surfactant. ACS Appl. Mater. Interfaces 2018, 10, 20368−20376. (20) Alnasser, F.; Castagnola, V.; Boselli, L.; Esquivel-Gaon, M.; Efeoglu, E.; McIntyre, J.; Byrne, H. J.; Dawson, K. A. Graphene Nanoflakes Uptake Mediated by Scavenger Receptors. Nano Lett. 2019, 19, 1260−1268.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.9b03638. Structures and ratios of ligands coated on GNPs used in experiments, coarse-grained models and simulation system setup, interaction parameter during simulations, synthesis and characterization of gold nanoparticles, TEM images of GNPs−cell interactions showing no GNP aggregation, calculated membrane tension as a function of the lipid density, and additional simulation results of interactions between purely hydrophilicligand-coated nanoparticles and the plasma membrane under different conditions (PDF)



Research Article

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (T.Y.). *E-mail: [email protected] (B.Y.). *E-mail: [email protected] (D.C.). ORCID

Tongtao Yue: 0000-0002-8329-167X Xianren Zhang: 0000-0002-8026-9012 Bing Yan: 0000-0002-7970-6764 Dapeng Cao: 0000-0002-6981-7794 Author Contributions ⊥

Y.L. and S.L. contributed equally to this work.

Author Contributions

T.Y., X.Z., B.Y., and D.C. conceived and designed the project. Y.L. and S.L. performed the simulations. H.S. and B.Y. performed the experiments. The manuscript was written by Y.L. and T.Y. and revised by X.L., X.Z., B.Y., and D.C., with contributions from all authors. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the National Key Research and Development Program of China (2016YFA0203103), the National Natural Science Foundation of China (Nos. 21276007, 31871012, 91543204, 91643204), and the Natural Science Foundation of Shandong Province (No. ZR2018MC004). 23829

DOI: 10.1021/acsami.9b03638 ACS Appl. Mater. Interfaces 2019, 11, 23822−23831

Research Article

ACS Applied Materials & Interfaces (21) Gratton, S. E.; Ropp, P. A.; Pohlhaus, P. D.; Luft, J. C.; Madden, V. J.; Napier, M. E.; DeSimone, J. M. The Effect of Particle Design on Cellular Internalization Pathways. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 11613−11618. (22) Gao, H.; Shi, W.; Freund, L. B. Mechanics of ReceptorMediated Endocytosis. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 9469− 9474. (23) Reynwar, B. J.; Illya, G.; Harmandaris, V. A.; Müller, M. M.; Kremer, K.; Deserno, M. Aggregation and Vesiculation of Membrane Proteins by Curvature-Mediated Interactions. Nature 2007, 447, 461. (24) Yang, K.; Ma, Y.-Q. Computer Simulation of the Translocation of Nanoparticles with Different Shapes across a Lipid Bilayer. Nat. Nanotechnol. 2010, 5, 579. (25) Ding, H.-m.; Tian, W.-d.; Ma, Y.-q. Designing Nanoparticle Translocation Through Membranes by Computer Simulations. ACS Nano 2012, 6, 1230−1238. (26) Li, Y.; Li, X.; Li, Z.; Gao, H. Surface-Structure-Regulated Penetration of Nanoparticles across a Cell Membrane. Nanoscale 2012, 4, 3768−3775. (27) Wang, S.; Guo, H.; Li, Y.; Li, X. Penetration of Nanoparticles across a Lipid Bilayer: Effects of Particle Stiffness and Surface Hydrophobicity. Nanoscale 2019, 11, 4025−4034. (28) Lin, J.; Alexander-Katz, A. Cell Membranes Open “Doors” for Cationic Nanoparticles/Biomolecules: Insights into Uptake Kinetics. ACS Nano 2013, 7, 10799−10808. (29) Luzio, J. P.; Poupon, V.; Lindsay, M. R.; Mullock, B. M.; Piper, R. C.; Pryor, P. R. Membrane Dynamics and the Biogenesis of Lysosomes. Mol. Membr. Biol. 2003, 20, 141−154. (30) Liu, B. R.; Huang, Y.-w.; Winiarz, J. G.; Chiang, H.-J.; Lee, H.-J. Intracellular Delivery of Quantum Dots Mediated by a Histidine- and Arginine-rich HR9 Cell-penetrating Peptide through the Direct Membrane Translocation Mechanism. Biomaterials 2011, 32, 3520− 3537. (31) Tu, Y.; Lv, M.; Xiu, P.; Huynh, T.; Zhang, M.; Castelli, M.; Liu, Z.; Huang, Q.; Fan, C.; Fang, H.; Zhou, R. Destructive Extraction of Phospholipids from Escherichia coli Membranes by Graphene Nanosheets. Nat. Nanotechnol. 2013, 8, 594. (32) Shi, X.; von Dem Bussche, A.; Hurt, R. H.; Kane, A. B.; Gao, H. Cell Entry of One-Dimensional Nanomaterials Occurs by Tip Recognition and Rotation. Nat. Nanotechnol. 2011, 6, 714. (33) Yi, X.; Shi, X.; Gao, H. Cellular Uptake of Elastic Nanoparticles. Phys. Rev. Lett. 2011, 107, No. 098101. (34) Verma, A.; Uzun, O.; Hu, Y.; Hu, Y.; Han, H.-S.; Watson, N.; Chen, S.; Irvine, D. J.; Stellacci, F. Surface-Structure-Regulated CellMembrane Penetration by Monolayer-Protected Nanoparticles. Nat. Mater. 2008, 7, 588. (35) Melby, E. S.; Lohse, S. E.; Park, J. E.; Vartanian, A. M.; Putans, R. A.; Abbott, H. B.; Hamers, R. J.; Murphy, C. J.; Pedersen, J. A. Cascading Effects of Nanoparticle Coatings: Surface Functionalization Dictates the Assemblage of Complexed Proteins and Subsequent Interaction with Model Cell Membranes. ACS Nano 2017, 11, 5489− 5499. (36) Boselli, L.; Polo, E.; Castagnola, V.; Dawson, K. A. Regimes of Biomolecular Ultrasmall Nanoparticle Interactions. Angew. Chem., Int. Ed. 2017, 56, 4215−4218. (37) Bekdemir, A.; Stellacci, F. A Centrifugation-based Physicochemical Characterization Method for the Interaction Between Proteins and Nanoparticles. Nat. Commun. 2016, 7, No. 13121. (38) Gong, L.; Chen, Y.; He, K.; Liu, J. Surface Coverage-Regulated Cellular Interaction of Ultrasmall Luminescent Gold Nanoparticles. ACS Nano 2019, 13, 1893−1899. (39) Sun, H.; Liu, Y.; Bai, X.; Zhou, X.; Zhou, H.; Liu, S.; Yan, B. Induction of Oxidative Stress and Sensitization of Cancer Cells to Paclitaxel by Gold Nanoparticles with Different Charge Densities and Hydrophobicities. J. Mater. Chem. B 2018, 6, 1633−1639. (40) Li, S.; Zhai, S.; Liu, Y.; Zhou, H.; Wu, J.; Jiao, Q.; Zhang, B.; Zhu, H.; Yan, B. Experimental Modulation and Computational Model of Nano-Hydrophobicity. Biomaterials 2015, 52, 312−317.

(41) Kostarelos, K.; Lacerda, L.; Pastorin, G.; Wu, W.; Wieckowski, S.; Luangsivilay, J.; Godefroy, S.; Pantarotto, D.; Briand, J.-P.; Muller, S.; et al. Cellular Uptake of Functionalized Carbon Nanotubes is Independent of Functional Group and Cell Type. Nat. Nanotechnol. 2007, 2, 108. (42) Zhou, H.; Jiao, P.; Yang, L.; Li, X.; Yan, B. Enhancing Cell Recognition by Scrutinizing Cell Surfaces with a Nanoparticle Array. J. Am. Chem. Soc. 2011, 133, 680−682. (43) Shillcock, J. C.; Lipowsky, R. Tension-Induced Fusion of Bilayer Membranes and Vesicles. Nat. Mater. 2005, 4, 225. (44) Groot, R. D.; Rabone, K. Mesoscopic Simulation of Cell Membrane Damage, Morphology Change and Rupture by Nonionic Surfactants. Biophys. J. 2001, 81, 725−736. (45) De Meyer, F. J.-M.; Venturoli, M.; Smit, B. Molecular Simulations of Lipid-mediated Protein-Protein Interactions. Biophys. J. 2008, 95, 1851−1865. (46) Yue, T.; Zhang, X. Cooperative Effect in Receptor-Mediated Endocytosis of Multiple Nanoparticles. ACS Nano 2012, 6, 3196− 3205. (47) Yue, T.; Wang, X.; Huang, F.; Zhang, X. An Unusual Pathway for the Membrane Wrapping of Rodlike Nanoparticles and the Orientation- and Membrane Wrapping-Dependent Nanoparticle Interaction. Nanoscale 2013, 5, 9888−9896. (48) Dasgupta, S.; Auth, T.; Gompper, G. Shape and Orientation Matter for the Cellular Uptake of Nonspherical Particles. Nano Lett. 2014, 14, 687−693. (49) Shen, Z.; Ye, H.; Yi, X.; Li, Y. Membrane Wrapping Efficiency of Elastic Nanoparticles during Endocytosis: Size and Shape Matter. ACS Nano 2019, 13, 215−228. (50) Español, P.; Warren, P. Statistical Mechanics of Dissipative Particle Dynamics. Europhys. Lett. 1995, 30, 191. (51) Hoogerbrugge, P.; Koelman, J. Simulating Microscopic Hydrodynamic Phenomena with Dissipative Particle Dynamics. Europhys. Lett. 1992, 19, 155. (52) Mao, J.; Chen, P.; Liang, J.; Guo, R.; Yan, L.-T. ReceptorMediated Endocytosis of Two-Dimensional Nanomaterials Undergoes Flat Vesiculation and Occurs by Revolution and Self-Rotation. ACS Nano 2016, 10, 1493−1502. (53) Chen, P.; Huang, Z.; Liang, J.; Cui, T.; Zhang, X.; Miao, B.; Yan, L.-T. Diffusion and Directionality of Charged Nanoparticles on Lipid Bilayer Membrane. ACS Nano 2016, 10, 11541−11547. (54) Jiang, W.; Kim, B. Y.; Rutka, J. T.; Chan, W. C. NanoparticleMediated Cellular Response is Size-Dependent. Nat. Nanotechnol. 2008, 3, 145−150. (55) Zhang, S.; Li, J.; Lykotrafitis, G.; Bao, G.; Suresh, S. SizeDependent Endocytosis of Nanoparticles. Adv. Mater. 2009, 21, 419− 424. (56) Vácha, R.; Martinez-Veracoechea, F. J.; Frenkel, D. ReceptorMediated Endocytosis of Nanoparticles of Various Shapes. Nano Lett. 2011, 11, 5391−5395. (57) Li, Y.; Yue, T.; Yang, K.; Zhang, X. Molecular Modeling of the Relationship Between Nanoparticle Shape Anisotropy and Endocytosis Kinetics. Biomaterials 2012, 33, 4965−4973. (58) Yue, T.; Xu, Y.; Sun, M.; Zhang, X.; Huang, F. How Tubular Aggregates Interact with Biomembranes: Wrapping, Fusion and Pearling. Phys. Chem. Chem. Phys. 2016, 18, 1082−1091. (59) Sun, J.; Zhang, L.; Wang, J.; Feng, Q.; Liu, D.; Yin, Q.; Xu, D.; Wei, Y.; Ding, B.; Shi, X. Tunable Rigidity of (Polymeric Core)-(Lipid Shell) Nanoparticles for Regulated Cellular Uptake. Adv. Mater. 2015, 27, 1402−1407. (60) Yue, T.; Zhang, X. Molecular Modeling of the Pathways of Vesicle-Membrane Interaction. Soft Matter 2013, 9, 559−569. (61) Verma, A.; Stellacci, F. Effect of Surface Properties on Nanoparticle-Cell Interactions. Small 2010, 6, 12−21. (62) Van Lehn, R. C.; Atukorale, P. U.; Carney, R. P.; Yang, Y.-S.; Stellacci, F.; Irvine, D. J.; Alexander-Katz, A. Effect of Particle Diameter and Surface Composition on the Spontaneous Fusion of Monolayer-Protected Gold Nanoparticles with Lipid Bilayers. Nano Lett. 2013, 13, 4060−4067. 23830

DOI: 10.1021/acsami.9b03638 ACS Appl. Mater. Interfaces 2019, 11, 23822−23831

Research Article

ACS Applied Materials & Interfaces (63) Wang, W.; Sedykh, A.; Sun, H.; Zhao, L.; Russo, D. P.; Zhou, H.; Yan, B.; Zhu, H. Predicting Nano-Bio Interactions by Integrating Nanoparticle Libraries and Quantitative Nanostructure Activity Relationship Modeling. ACS Nano 2017, 11, 12641−12649. (64) Tenzer, S.; Docter, D.; Rosfa, S.; Wlodarski, A.; Kuharev, J.; Rekik, A.; Knauer, S. K.; Bantz, C.; Nawroth, T.; Bier, C.; et al. Nanoparticle Size is a Critical Physicochemical Determinant of the Human Blood Plasma Corona: a Comprehensive Quantitative Proteomic Analysis. ACS Nano 2011, 5, 7155−7167. (65) Lundqvist, M.; Stigler, J.; Cedervall, T.; Berggård, T.; Flanagan, M. B.; Lynch, I.; Elia, G.; Dawson, K. The Evolution of the Protein Corona around Nanoparticles: A Test Study. ACS Nano 2011, 5, 7503−7509. (66) Mahmoudi, M.; Abdelmonem, A. M.; Behzadi, S.; Clement, J. H.; Dutz, S.; Ejtehadi, M. R.; Hartmann, R.; Kantner, K.; Linne, U.; Maffre, P.; et al. Temperature: the “Ignored” Factor at the Nanobio Interface. ACS Nano 2013, 7, 6555−6562. (67) Zuo, G.; Kang, S.; Xiu, P.; Zhao, Y.; Zhou, R. Interactions Between Proteins and Carbon-Based Nanoparticles: Exploring the Origin of Nanotoxicity at the Molecular Level. Small 2013, 9, 1546− 1556. (68) Zuo, G.; Huang, Q.; Wei, G.; Zhou, R.; Fang, H. Plugging into Proteins: Poisoning Protein Function by a Hydrophobic Nanoparticle. ACS Nano 2010, 4, 7508−7514. (69) Nie, S. Understanding and Overcoming Major Barriers in Cancer Nanomedicine. Nanomedicine 2010, 5, 523−528. (70) Tagliazucchi, M.; Peleg, O.; Kröger, M.; Rabin, Y.; Szleifer, I. Effect of Charge, Hydrophobicity, and Sequence of Nucleoporins on the Translocation of Model Particles through the Nuclear Pore Complex. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 3363−3368. (71) Charras, G. T.; Yarrow, J. C.; Horton, M. A.; Mahadevan, L.; Mitchison, T. Non-Equilibration of Hydrostatic Pressure in Blebbing Cells. Nature 2005, 435, 365. (72) Solon, J.; Pécréaux, J.; Girard, P.; Fauré, M.-C.; Prost, J.; Bassereau, P. Negative Tension Induced by Lipid Uptake. Phys. Rev. Lett. 2006, 97, No. 098103. (73) Charras, G.; Paluch, E. Blebs Lead the Way: How to Migrate without Lamellipodia. Nat. Rev. Mol. Cell Biol. 2008, 9, 730. (74) Yue, T.; Li, S.; Zhang, X.; Wang, W. The Relationship between Membrane Curvature Generation and Clustering of Anchored Proteins: A Computer Simulation Study. Soft Matter 2010, 6, 6109−6118.

23831

DOI: 10.1021/acsami.9b03638 ACS Appl. Mater. Interfaces 2019, 11, 23822−23831