Shape Effect on Particle-Lipid Bilayer Membrane Association, Cellular

Oct 15, 2015 - Although computer simulation and cell culture experiments have shown that elongated spherical particles can be taken up into cells more...
0 downloads 10 Views 2MB Size
Download PDF
Research Article www.acsami.org

Shape Effect on Particle-Lipid Bilayer Membrane Association, Cellular Uptake, and Cytotoxicity Thapakorn Tree-Udom,† Jiraporn Seemork,‡ Kazuki Shigyou,§ Tsutomu Hamada,⊥,§ Naunpun Sangphech,|| Tanapat Palaga,|| Numpon Insin,# Porntip Pan-In,# and Supason Wanichwecharungruang*,⊥,#,△ †

Nanoscience and Technology Program, Graduate School, ‡Program in Petrochemistry, Faculty of Science, ||Department of Microbiology, Faculty of Science and Interdisciplinary Program in Medical Microbiology, #Department of Chemistry, Faculty of Science, and △Nanotec-CU Center of Excellence on Food and Agriculture, Chulalongkorn University, Bangkok 10330, Thailand § School of Materials Science, Japan Advanced Institute of Science and Technology, Nomi 923-1211, Japan S Supporting Information *

ABSTRACT: Although computer simulation and cell culture experiments have shown that elongated spherical particles can be taken up into cells more efficiently than spherical particles, experimental investigation on effects of these different shapes over the particle−membrane association has never been reported. Therefore, whether the higher cellular uptake of an elongated spherical particles is a result of a better particle− membrane association as suggested by some calculation works or a consequence of its influence on other cellular transmembrane components involved in particle translocation process, cannot be concluded. Here, we study the effect of particle shape on the particle−membrane interaction by monitoring the association between particles of various shapes and lipid bilayer membrane of artificial cell-sized liposomes. Among the three shaped lanthanide-doped NaYF4 particles, all with high shape purity and uniformity, similar crystal phase, and surface chemistry, the elongated spherical particle shows the highest level of membrane association, followed by the spherical particle with a similar radius, and the hexagonal prism-shaped particle, respectively. The free energy of membrane curvature calculated based on a membrane indentation induced by a particle association indicates that among the three particle shapes, the elongated spherical particle give the most stable membrane curvature. The elongated spherical particles show the highest cellular uptake into cytosol of human melanoma (A-375) and human liver carcinoma (HepG2) cells when observed through a confocal laser scanning fluorescence microscope. Quantitative study using flow cytometry also gives the same result. The elongated spherical particles also possess the highest cytotoxicity in A-375 and normal skin (WI-38) cell lines, comparing to the other two shaped particles. KEYWORDS: upconverting nanocrystals, elongated spherical particle, ellipsoid particle, endocytosis, membrane curving

■

INTRODUCTION Cellular transport is a fundamental process for all organisms. Particles can enter cells by specifically interacting with receptors on cell membrane, directly penetrating across the lipid bilayer, or going through nonreceptor mediated endocytosis (membrane wrapping and budding off). Although it has been shown that size, shape, charge, and surface chemistry of particles together with properties of lipid bilayer membrane itself play important roles in both energy dependent and independent cellular uptake of particles, the understanding is still unclear.1 Effective sizes for cellular penetration were reported differently in studies carried out using different cell lines and particle materials.2−6 Experiments using artificial liposomes7 showed that 50 nm particles preferred different type of membrane from 200 nm particles. Positively charged nanoparticles were often reported to possess better cellular uptake than the negatively charged ones,8−10 however, some negatively charged carbon particles showed excellent penetration ability.11 Most theoreti© 2015 American Chemical Society

cal calculation reports on particle shape effect over membrane translocation indicated that particle with spherocylinder shape (elongated spherical particles) could translocate across the lipid bilayer membrane more efficiently than the spherical particles.12−14 Limited cellular experiments also indicated a better uptake of elongated particles over spherical ones.15,16 Nevertheless, similar cellular uptake level for both shapes was also reported.17 Better anticancer activity observed in cell culture for 10-hydroxycamptothecin in elongated spherical carrier as compared to the same drug in spherical carrier, was also observed.18 Interpretation of results from experiments with real cells can be complicated because adsorption of various components presence in cell media onto the particles, or ligand exchange Received: July 25, 2015 Accepted: October 15, 2015 Published: October 15, 2015 23993

DOI: 10.1021/acsami.5b06781 ACS Appl. Mater. Interfaces 2015, 7, 23993−24000

Research Article

ACS Applied Materials & Interfaces

or oleic acid-stabilized hexagonal prism (oleate-HP) was obtained with a lower oleic to octadecene ratio, spherical particles or oleic acid-stabilized nanosphere (oleate-Sph) was obtained with increasing ratio, and elongated spherical particles or oleic acid-stabilized elongated sphere (oleate-LSph) was obtained at the highest ratio (Figure 1). It is likely that the

between ligands on the particles and components in the cell media, may take place unknowingly. 15,19 In addition, discrimination between influence of particle shape on lipid bilayer membrane interaction and interaction with other transmembrane components cannot be made. Before any nanoparticles or micro-organisms can enter cells, they must associate with cell membrane. Unlike using the real cells, experiment using a model membrane such as a giant liposome can provide information on particle uptake based on passive particle−lipid bilayer membrane interaction without interferences from other cellular processes.20 Although investigation on particle− membrane association has been carried out on particles of different sizes,7 study of such association on different-shaped particles has never been reported. Using computer simulation via dissipative particle dynamics, Yang and Ma have suggested that the direct penetrating ability of a nanoparticle across a lipid bilayer membrane is governed by (1) a contact area between the particle and a lipid bilayer and (2) a local curvature of the particle at the contact point.12 With computer simulation, Murad team has shown that membrane curvature and elasticity play important roles in particle− membrane association, which then develops into the particle wrapping during endocytic internalization.21 The membrane curvature induced by particle−membrane interaction has been recently suggested to be an important step even in the receptormediated endocytosis. The Spakovitz team has used the computational model to demonstrate that the taking place of membrane curving during particle wrapping in clathrinmediated endocytosis is an important step that physically triggers the clathrin lattice reorganization necessary for the lipid bilayer vesicle formation.22 Here we compare the levels of particle−membrane association among spherical, elongated spherical and hexagonal prism-shaped particles, using lipid bilayer membrane of artificially made cell-sized liposomes. Explanation of the experimental result using free energy of the membrane curvature induced by particle−membrane association is demonstrated. This report also contains experiments on cellular internalization of different shaped particles, investigated by qualitative confocal fluorescence microscopy and quantitative flow cytometry. Here the study was carried out using three different shaped nanocrystals of NaYF4 doped with 30 mol % Yb3+ and 0.5 mol % Tm3+, all with similar crystal phase, similar surface chemistry, and negative surface charge, high shape purity and uniformity. The spherical, elongated spherical, and hexagonal prism-shaped particles used in the study all possess poly(ethylene oxide) carboxylate ligands on their surfaces. Information gained from this work can be used not only for the fundamental understanding of particle shape effect on membrane association and cellular uptake, but also for the selection of appropriate particle shapes for lanthanide-doped NaYF4 for imaging and photodynamic therapeutic applications.

Figure 1. Oleic acid to octadecene volume ratio used in the synthesis, TEM images, and dimensions of the obtained (30 mol % Yb3+, 0.5 mol % Tm3+)-doped NaYF4 particles (all with oleic acids as stabilizing ligands): oleate-HP (top), oleate-Sph (middle), and oleate-LSph (bottom).

concentration dependent self-organization of oleate ligands around the crystals affects the crystal growth rate in different directions. Because it has been speculated that the growth rates of the hexagonal (Yb3+, Er3+/Tm3+)-doped NaYF4 crystals in different directions govern directly their final shapes,23 the shape of the particles is influenced by the concentration of oleate ligands. The X-ray diffraction (XRD) patterns of the three shaped oleate-stabilized particles were similar to that of the standard βNaYF4 (JCPDS file No. 28−1192) (Figure S1), indicating that all of them possessed pristine hexagonal phase-NaYF4 structure. All three-shaped oleate-stabilized particles dispersed well in organic solvents such as hexane and chloroform. To make the three-shaped particles dispersible in hydrophilic medium such as water, the oleate ligands were replaced with poly(ethylene oxide) diacid of MW 600 or the so-called COOH-PEO-COOH ligands. Similar to oleic acid, here one COOH of the COOH-PEO-COOH probably formed stable interaction with the surface of the particle, leaving one COOH as a loose end. Therefore, the ligand-exchanged particles were denoted as COOH-PEO-particles, e.g., COOH-PEO-Sph, COOH-PEO-LSph, and COOH-PEO-HP. Successful exchange was confirmed through the FTIR spectrum and the good water dispersibility of the obtained COOH-PEO-particles. IR spectra of the three shaped COOH-PEO-particles showed characteristic peaks at 1610 and 1459 cm−1 (asym and sym str of carboxylate moieties on the surface of the particle), 1734 cm−1 (str of COOH group at the loose end) and 1104 cm−1 (str of C−O−C in the ethylene oxide) (Figure S2).24 All three shaped COOH-PEO-particles gave negative zeta potential values in the

■

RESULTS AND DISCUSSION Particle Synthesis and Characterization. Different shapes of (30 mol % Yb3+, 0.5 mol % Tm3+)-doped NaYF4 nanocrystals were synthesized by thermal decomposition method using octadecene as solvent. Oleic acid was used as ligands to control particle shapes. Transmission electron microscopic (TEM) images showed that particles obtained from the reactions with different oleic acid to octadecene ratios were of different morphologies, i.e., hexagonal prism particles 23994

DOI: 10.1021/acsami.5b06781 ACS Appl. Mater. Interfaces 2015, 7, 23993−24000

Research Article

ACS Applied Materials & Interfaces range of −10 to −20 mV in water. Their TEM images confirmed that the ligand exchange process did not affect particle morphology (Figure S3). The dimensions of the three particles were similar to those shown in Figure 1. Therefore, the volume ratios among the sphere: elongated sphere: hexagonal prism particles were 1:1.3:30. In addition, upon 980 nm irradiation, the three shaped COOH-PEO-particles gave similar intense visible fluorescence emission (Figure S4), thus confirming their similar electronic structures. A presence of free carboxylic acid groups on COOH-PEOparticles allowed for further chemical conjugation of the particles. To quantitatively determine the effect of particle shapes on their association with lipid bilayer membrane by confocal fluorescence microscope, and their levels of being internalized into cells by flow cytometry, we covalently linked 6-aminofluorescein fluorophores to the carboxyl groups of the COOH-PEO-particles, via a carbodiimide assisted esterification reaction. The presence of a new absorption peak at ∼1640 cm−1 (amide I) in the FTIR spectra of all 6-aminofluoresceinlabeled particles (denoted here as f-particles: f-LSph, f-Sph and f-HP) indicated that the fluorescein moieties had been covalently linked to the surface of COOH-PEO-particles (Figure S5). The zeta potential values of f-LSph, f-Sph and fHP in water were −16.0 ± 0.35, −21.5 ± 1.10, and −7.3 ± 1.70 mV, respectively. Reactions were adjusted to obtain similar fluorescence intensity per mass among the three f-particles. Particle−Liposome Interaction. Here the levels of particle−membrane association among spherical, elongated spherical, and hexagonal prism-shaped particles, was studied using lipid bilayer membrane of artificially made cell-sized liposomes. Here the levels of particle−membrane association were measured through the relative amounts of particles bound to the lipid bilayer membrane of the liposomes. To investigate how particle morphology affected lipid bilayer membrane association, we employed cell-size liposomes with liquiddisordered phase constructed from dioleoyl L-α phosphatidylcholine (DOPC). Adhesion of the fluorescent particles on liposome surface was monitored by confocal laser microscopy (λex/λem 473/490−590 nm). At similar confocal microscope setting, obvious fluorescence was observed on liposomes that had been mixed with f-LSph, but not on those mixed with f-HP (Figure 2). The signal could be observed faintly for liposomes that had been mixed with f-Sph. It should be noted here that during the 20 min incubation of the f-particles with the liposomes, we did not observe any fluorescence signal at the inside of the liposomes. This indicated that these particles could not penetrate lipid bilayer membrane well under the nonenergy assisted condition. The result above clearly indicated that the degree of lipid bilayer membrane association was highest for the elongated spherical particle, followed with the sphere and prism particles. We explain these differences in membrane affinity of different shaped particles using the theory of membrane curvature.7 An association of particle on the membrane surface usually induces the membrane indentation, which is the curving of the membrane to wrap the particle. Dasgupta et al., reported that cube-like particles need higher adhesion energy to be wrapped by the membrane than spherical and rod particles. 25 Morphology of the particles strongly affects the curvation of the membrane. The hexagonal prism shaped particles possess sharp edges (90°) at the two ends and therefore require extreme curving of the membrane. This sharp turn leads to unstable packing of the phospholipid bilayer, thus it is a

Figure 2. Association of the three shaped f-particles on liposomes. Plots of fluorescence intensity (F.I.) of (a) f-LSph, (b) f-Sph, and (c) fHP along the white dashed line of the corresponding liposome shown on the right of the graph. The differential interference contrast pictures of the corresponding liposomes are shown next to the corresponding fluorescence images.

nonfavorable process (Figure 3a). Because the particles are very small comparing to the cell-sized liposome curvature, the

Figure 3. Association between lipid bilayer membrane and different shaped particles. (a) The packing of phospholipids around the sharp edge (top) and soft edge (bottom). (b) Free energy of the membrane that wraps spherical (Fsph) and elongated spherical (Flsph) particles as a function of aspect ratio (b/a).

membrane can be considered flat. Therefore, when the flat plane of the hexagonal prism is parallel to the membrane plane, favorable particle−membrane associate can take place without any need of membrane curving and this should lead to a favorable particle−membrane association. Nevertheless, among the unlimited random orientations of the particle toward the membrane plane, with only eight flat planes but with 18 sharp edges and 12 sharp corners of the hexagonal prism particle, there are only eight orientations that will lead to such parallel association. A little twist from the parallel orientation will introduce sharp edge or corner to the membrane and this will then lead to unfavorable association. This explains the minimal adhesion of the prism-shaped particles to the membrane observed in our experiment. When the soft edge particles, here including spherical and elongated spherical particles, adhere on the membranes, they are partially wrapped with angles θ1 and θ2 (Figure 3b). The 23995

DOI: 10.1021/acsami.5b06781 ACS Appl. Mater. Interfaces 2015, 7, 23993−24000

Research Article

ACS Applied Materials & Interfaces

Figure 4. Cellular uptake of various shaped f-particles. (a) Representative CLFM images of A-375 cells after being incubated with f-HP (column 1), f-Sph (column 2), and f-LSph (column 3) at 37 °C for 6 h, in differential interference contrast mode (top row) and fluorescence mode (bottom row). (b) Mean fluorescence intensity (F.I.) from flow cytometry measurement of the A-375 cells which had been incubated with the each of the three shaped f-particles for various times, and thoroughly washed. Data represent means ± SD (n = 6, from two independent experiments). F.I. of the control cells (cells incubated with no particles) was set at 1.0 and the F.I. shown were calculated in relative to the F.I. of the control cells. Statistical significant difference was calculated using one-way ANOVA and LSD (*p < 0.01).

their present elongated spherical shape.26 It should be mentioned here that, unlike the hexagonal prism particle in which only 8 particle−membrane orientations will lead to the stable association, favorable orientations are infinite for the sphere. This unlimited favorable orientations of the particle toward the membrane is also the case for the elongated spherical particle. Here the sphere and the elongated sphere possess similar radius, thus the orientations in which the elongated sphere is sit with its long axis perpendicular to the membrane plane will lead to similar membrane curving condition to the sphere. Because there are only two such possibilities (two ends of the particle) and other orientations will lead to a more favorable free energy of the membrane curving, the membrane association of the elongated spherical particles is more probable than that of the sphere. Cellular Association. The results from the cell-sized liposome experiment indicated a higher membrane association for f-LSph as compared to that of the f-Sph and f-HP. To see if the higher membrane association would correlate to the level of particle uptake into the real cells, we compared cellular uptake levels of the three shaped particles using human melanoma (A375) and human liver carcinoma (HepG2) cells. As expected, in the A-375 cells, the result clearly showed the highest cellular uptake for the elongated spherical particles. Confocal laser fluorescence microscopic (CLFM) images of cells incubated with the three f-particles for 6 h showed obvious fluorescence signals of the f-particles in the cells (Figure 4a) while the control cells (cells incubated with no particles) showed no fluorescence signal of the particles (not shown). Fluorescence signals from all three f-particles were observed only at cytoplasmic region, not in the nucleus, indicating that the three particles could not enter nucleus. Under the same fluorescence microscope setting, the highest fluorescence intensity was observed in the cells incubated with f-LSph. To confirm the relative cellular association levels among the three shaped particles, quantitative flow cytometry experiment was carried out. The mean fluorescence intensities of cells incubated with the f-particles for 1 to 6 h was calculated in relative to that of the untreated cells (F.I. = 1.0). The results clearly showed higher fluorescence intensity at longer incubation times, indicating that the cellular association of the particles was time dependent and longer incubation resulted in higher particle accumulation in the cells. More importantly, fluorescence intensities detected in the cells incubated with f-

adhesion surface areas of the sphere (Asph) and of the elongated sphere (Alsph) are given by A sph = 2πR2(1 − cos θ1)

Alsph = 2πR2(1 − cos θ2) + LRθ2

The curvature free energies of the association for the sphere (Fsph) and the elongated sphere (Flsph) could then be estimated as follows Fsph = 4πKb(1 − cos θ1) Flsph = 4πKb(1 − cos θ2) +

π Kb(x − 1) 8

where x is the aspect ratio of elongated sphere (x = b/a = 1 + L/2R) and Kb is the bending modulus of the membrane. Notably, we here consider that the membrane is under zero tension and the bending energy of the free membrane part that is not interacting with the particle is neglected. When the spherical and the elongated spherical particles adhere on the membrane with the same surface area (Asph = Alsph), the difference of the free energies as a function of x (for simplicity, we assume that θ2 is a constant value of π/4) could be estimated (Figure 3b). When the aspect ratio increases, the adsorption of an elongated spherical particle is more stable than that of a sphere. The elongated sphere with x = 1.5 schematically shown in Figure 3b corresponds to our experimental condition. It should be noted that here previous investigation over effects of particle anisotropy on particle− membrane association and membrane wrapping via the calculation of lipid bilayer membrane curvature energy combined with a contact adhesion energy of the particle− membrane interaction, has also revealed that high aspect ratios with round tips particles associated with lipid bilayer membrane through their long side.25 It should be noted here that the volume of the elongated spherical particle was approximately 1.3 of the volume of the spherical particle. The similar range of volume of the two shaped particles ruled out the possibility that the significant difference membrane association level between the two is a result of their size difference. The fact that at similar volume range the elongated spherical particle can associate to lipid bilayer membrane better than the spherical one may help explain the evolution of microorganisms such as Salmonella, Shigella, Yersinia, and Listeria monocytogenes, into 23996

DOI: 10.1021/acsami.5b06781 ACS Appl. Mater. Interfaces 2015, 7, 23993−24000

Research Article

ACS Applied Materials & Interfaces

reported previousely.18,28,29 Interestingly, although our results indicated that the f-Sph particles associated better to lipid bilayer membrane and possessed higher cellular uptake level than the f-HP particles, the in vitro cytotoxicity of COOHPEO-HP particles was higher than that of COOH-PEO-Sph particles at the concentration of 1000 μg/mL after being incubated for 24 h in WI-38 and A-375 cell lines. We speculated that this small but significantly higher cytotoxicity of the COOH-PEO-HP particle over that of the COOH-PEO-Sph particles in these two cell lines might be a result of the different intracellular responses of the cells toward the two shaped particles. Future study on intracellular trafficking of the two shaped particles is needed to clarify this discrepancy.

LSph were the highest comparing to those incubated with f-Sph and f-HP (Figure 4b). The results obtained from HepG2 cells also showed similar trend (Figure S6). It was therefore concluded that among the three shaped particles, the f-LSph were taken up into cells most efficiently. Therefore, the cellular experimental results here agreed well with previous reports, which indicated that the elongated particles were taken up into cells more efficiently than spherical particles.15,16 The important point here is the correlation between the cellular uptake level of the three shaped particles and their ability to associate with lipid bilayer membrane. It is very likely that the higher cellular uptake of the elongated spherical particles is a result of their better membrane association. Although the cell-sized liposome experiment indicated only lipid bilayer membrane association with no significant penetration of the particles into the liposomes, here we observed the uptake of all three shaped particles into the cells. This suggested that the uptake probably occurred via energydriven process. In facts, previous reports have suggested that the membrane curvature could induce active cellular endocytosis.21,27 In Vitro Cytotoxicity Studies of COOH-PEO Particles. Cytotoxicity of the three shaped COOH-PEO-particles was tested on HepG2, A-375, and WI-38 by MTT assay. The viability of the untreated cells was set as 100% and the viability of the COOH-PEO-particle-treated cells was present relative to that of the untreated cells. All three cell lines showed IC50 for the three shaped particles of more than 1000 μg/mL. The test was also carried out at the extreme concentration of up to 10,000 μg/mL in the HepG2 cells, and all the three shaped COOH-PEO-particles showed similar viability of over 80%. In the A-375 and WI-38 cell lines, the COOH-PEO-LSph exhibited small but significantly higher toxicity than the other two shaped particles only when the concentration of the particles were at 1000 μg/mL, and 24 h incubation time was used (Figure 5a, b). Therefore, it could be concluded that all

■

CONCLUSION In conclusion, here we prepare (30 mol % Yb3+, 0.5 mol % Tm3+)-doped NaYF4 nanocrystals of sphere, elongated sphere, and hexagonal prism shapes. All the three shaped particles exhibit the same crystal structure, the same emission spectrum, and a similar range of negative surface charge in water. The elongated spherical particles possess an aspect ratio of 1.5 and similar radius to that of the spherical particles. Using these particles, we show here experimentally for the first time that the elongated spherical particles associate more efficiently with lipid bilayer membrane than the spherical particles. The higher membrane association of the elongated sphere over the sphere can be explained through the lower free energy of the membrane curving around the elongated spherical shaped particle as compared to that around the spherical shape particle. The degree of particle−membrane association directly influences the extent of cellular uptake/cytotoxicity of the particles. Since (30 mol % Yb3+, 0.5 mol % Tm3+)-doped NaYF4 nanocrystals can be used in various applications, this work provides essential information for selecting appropriate shaped particles for any particular applications, i.e., the prism particles are best for applications requiring minimal cellular association such as blood flow imaging, whereas the elongated spheres are suitable for drug delivery or photodynamic therapy applications where maximum cellular uptake is needed.

■

MATERIALS AND METHODS

All chemicals were used as received without further purification. YCl3· 6H2O (99.99%), YbCl3·6H2O (99.99%), TmCl3·6H2O (99.99%), oleic acid (90%), 1-octadecene (90%), ammonium fluoride, sodium hydroxide, poly(ethylene oxide)bis(carboxymethyl)ether (PEO-600diacid, or COOH-PEO-COOH, average Mn 600) and 6-aminofluorescein (95%, bioreagent) were purchased from Sigma-Aldrich (Steinheim, Germany). 1-(3-(Dimethylamino)propyl)-3-ethylcarbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) were purchased from Acros Organics (Geel, Belgium). Synthesis of Oleic Acid-Stabilized Sphere, Elongated Sphere, Hexagonal Prism-Shaped Particles. Oleic Acid-Stabilized Nanosphere (oleate-Sph). YCl3·6H2O (2.78 mmol, 843.3 mg), YbCl3·6H2O (1.2 mmol, 465 mg), and TmCl3·6H2O (0.02 mmol, 7.7 mg) were loaded into a three-neck round-bottom flask with the presence of oleic acid (24.0 mL) and octadecene (60.0 mL). The mixture was magnetically stirred and heated to130 °C under vacuum for 40 min to form a clear solution. After the solution was cooled to room temperature, a methanolic solution of NaOH (10 mmol, 400 mg) and NH4F (16 mmol, 592.64 mg) was added. The mixture was continuously stirred at 30 °C for 40 min then reaction temperature was increased to 75 °C in order to evaporate methanol from the mixture. Subsequently, the reaction temperature was rapidly increased to 300 °C and the mixture was continuously stirred under the nitrogen gas flow. After reaction time of 90 min, the mixture was cooled to

Figure 5. In vitro cytotoxicity of different shaped particles. Viability of (a) A-375 and (b) WI-38. after being incubated for 24 h with different shaped particles. Viability levels of the cells were expressed in relative to that of the untreated cells (viability =100%) and were normalized between two independent experiments. Data are shown as means ± SD (n = 6, from two independent experiments). Statistical significant differences were calculated using one-way ANOVA and LSD (p < 0.10) and significant difference between data sets is indicated with *.

the three shaped COOH-PEO-particles possessed low cytotoxicity to WI-38 (normal) and A-375 and HepG2 (cancer) cell lines. It is very likely that small but significantly higher cytotoxicity of the elongated spherical particle is a result of the better cellular uptake of the particle. Noted that higher toxicity observed for elongated spherical particles have also been 23997

DOI: 10.1021/acsami.5b06781 ACS Appl. Mater. Interfaces 2015, 7, 23993−24000

Research Article

ACS Applied Materials & Interfaces

Interaction of f-Sph, f-LSph, and f-HP with Cell-Sized Liposomes. The obtained cell-sized liposome suspension were mixed with f-Sph or f-LSph or f-HP (the weight ratio of f-particles to liposomes was kept to approximately 6.4 to 1 for all three shaped particles) and subjected to confocal laser microscope (Olympus FV1000-D, Tokyo, Japan) using excitation wavelength of 473 nm and detection at 490−590 nm. The fluorescence intensity on liposome surface was acquired by the Image-J software. Cell Culture. The A-375 human melanoma cell line (ATCC CRL1619) was obtained from the American Type Culture Collection (ATCC) (Manassas, VA, USA). Cells were cultured in Dulbecco’s modified Eagle’s medium (HyClone, UT, USA) supplemented with 10% v/v fetal bovine serum (HyClone, UT, USA), 100 mM sodium pyruvate (HyClone, UT, USA), 10 mM HEPES (HyClone, UT, USA), 100 U/mL penicillin (General Drugs House Co., Ltd., Bangkok, Thailand) and 0.4 mg/mL streptomycin (M & H Manufacturing Co., Ltd., Samut Prakan, Thailand) in a humidified incubator at 37 °C and 5% CO2. Cells were harvested using 0.25% w/v trypsin-1 mM EDTA and were resuspended in fresh complete medium before plating. The HepG2 human liver carcinoma cell line (ATCC HB-8065) was obtained from the American Type Culture Collection (ATCC) (Manassas, VA, USA). Cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (HyClone, UT, USA) supplemented with 10% v/v fetal bovine serum (HyClone, UT, USA), 100 mM sodium pyruvate (HyClone, UT, USA), 10 mM HEPES (HyClone, UT, USA), 100 U/mL penicillin (General Drugs House Co., Ltd., Bangkok, Thailand) and 0.4 mg/mL streptomycin (M & H Manufacturing Co., Ltd., Samut Prakan, Thailand) in a humidified incubator at 37 °C and 5% CO2. Cells were harvested using 0.25% w/v trypsin-1 mM EDTA and were resuspended in fresh complete medium before plating. The WI-38 human diploid cell line (ATCC CCL-75) was supplied by the ATCC. Cells were cultured in Advanced Minimum Essential medium (Gibco, NY, USA) supplemented with 5% v/v fetal bovine serum (HyClone, UT, USA), 200 mM L-glutamine (HyClone, UT, USA), 100 U/ml penicillin (General Drugs House Co., Ltd., Bangkok, Thailand) and 0.4 mg/mL streptomycin (M & H Manufacturing Co., Ltd., Samut Prakan, Thailand) in a humidified incubator at 37 °C and 5% CO2. Cells were harvested using 0.25% w/v trypsin-1 mM EDTA and were resuspended in fresh complete medium before plating. Cellular Association Determined by Confocal Laser Scanning Fluorescence Microscopy. A-375 cells were seeded on an eight well cell culture black chamber slide at 6 × 103 cells per well overnight. After that, the culture medium was then replaced with complete medium containing the tested particles at 500 μg/mL. The cells were incubated in this medium for 6 h at 37 °C, 5% CO2 then washed three times with PBS. Cell imaging was performed under an Eclipse-C1 Ti series confocal laser scanning fluorescence microscope (Nikon, Tokyo, Japan) using excitation wavelength of 488 nm and detection at 525 nm. HepG2 cells were seeded on an eight-well cell culture black chamber slide at 1× 104 cells per well overnight. After that, the culture medium was then replaced with complete medium containing the tested particles at 500 μg/mL. The cells were incubated in this medium for 6 h at 37 °C, 5% CO2, washed three times with PBS, fixed by 4%w/v paraformaldehyde, washed three times with PBS, and then stained with 4′,6-diamidino-2-phenylindole (DAPI) followed by washing with PBS three times. Cell imaging was performed by CLFM. Cellular Uptake Determined by Flow Cytometry Analysis. The A-375 and HepG2 cell lines were used to investigate the uptake of sphere, elongated sphere and hexagonal prism shaped particles. A-375 and HepG2 cells were seeded into a 24-well plates at 1.5 × 105 cells per well and then incubated at 37 °C under 5% CO2 overnight. The culture medium was then replaced by complete medium containing the tested particles at 500 μg/mL. The cells were incubated with particles for 1−6 h at 37 °C under 5% CO2. The cells in complete medium with no samples were used as controls. After cell/particle incubation, the cells were washed with cold phosphate buffer saline pH 7.4 (PBS) three times to remove unbound particles, collected by cell scraper and then centrifuged at 3000 rcf for 5 min at 4 °C. The cell

room temperature. The nanoparticles were precipitated by the addition of ethanol. Then the obtained particles were isolated via centrifugation at 5000 rpm (2688 rcf). Finally, the precipitation was washed several times with ethanol. Oleic Acid-Stabilized Elongated Sphere (oleate-LSph). The synthesis procedure was performed following a similar procedure as described above except that 42.0 mL of oleic acid and 42.0 mL of octadecene were used. Oleic Acid-Stabilized Hexagonal Prism (oleate-HP). The synthesis of the oleate-HP was similar to the oleate-Sph except that 12.0 mL of oleic acid and 72.0 mL of octadecene were used. To surround the entire three shaped particles (oleate-Sph, oleateLSph and oleate-HP) with COOH-PEO-COOH, all three materials were subjected to the ligand exchange as described below. Ligand Exchange. Fifty milligrams of oleate-stabilized particles was loaded into a glass vial containing 5.0 mL of ethanolic solution of PEO-600-diacid ligand (25 mg/mL). The vial was capped and sonicated for 15 min. The resulting solution was then maintained at 75 °C for 10 h. After the solution was cooled to 30 °C, hexane was then added dropwise in order to precipitate out the particles. The PEOcoated sphere, elongated sphere and hexagonal prism shaped particles (COOH-PEO-Sph, COOH-PEO-LSph, and COOH-PEO-HP) were isolated via centrifugation at 8000 rpm (6880 rcf) and repeatedly washed with ethanol. Fluorescence Conjugation. Twenty milligrams of EDC was added to a 20 mL aqueous solution of the COOH-PEO-particles (1 mg/mL) and stirred at 0 °C under nitrogen gas for approximately 15 min. Then 25 μL of acetone solution containing 1.2 mg of 6aminofluorescein together with 12 mg of NHS was added. The mixture was continuously stirred at room temperature overnight. The fluorescein-labeled particles (denoted as f-Sph, f-LSph, and f-HP for the products obtained from COOH-PEO-Sph, COOH-PEO-LSph and COOH-PEO-HP, respectively) were isolated via centrifugation at 8000 rpm (6880 rcf) and repeatedly washed with water/acetone until no 6-aminofluorescein was detected in the supernatant. Finally, the contents of fluorescein moieties labeled on f-Sph, f-LSph, and f-HP were quantitatively determined by fluorescence spectrophotometer with the aid of a calibration curve. Material Analyses. X-ray diffraction (XRD): XRD patterns of samples were obtained on a DMAX 2200/Ultima+ diffractometer (Rigaku, Tokyo, Japan) using Cu Kα radiation source and operating at 40 kV and 30 mA; the XRD spectra were collected with a scan range of 10−80° and scan speed of 1°/min. Transmission electron microscopy (TEM): TEM photographs of particles were obtained using a TECNAI 20 TWIN transmission electron microscope (FEI Company, OR, USA) operating at 120 kV; the diluted dispersion of oleatestabilized particles in hexanes or COOH-PEO particles in water was dropped on carbon film on 200 mesh copper grids and then dried in desiccators at room temperature; the average size was then estimated from several obtained TEM images using SemAfore program. Attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR): ATR-FTIR spectra were acquired using a Nicolet 6700 (Thermo Scientific, MA, USA) coupled with diamond ATR crystal at the resolution of 2 cm−1 and 64 numbers of scans. Dynamic light scattering (DLS): zeta potential of COOH-PEO-particles in water were measured on a Malvern Zetasizer nanoseries model S4700 (Malvern Instruments, Worcestershire, UK) using a He−Ne laser beam at 632.8 nm and scattering angle of 173°. Fluorescence spectrophotometry: Emission spectra were measured in 1 cm quart cuvettes using a Cary Eclipse Fluorescence Spectrophotometer (Agilent Technologies, CA, USA). Preparation of Mimic Lipid Bilayer Membranes. Dioleoyl L-α phosphatidylcholine (DOPC, Avanti Polar Lipids, Alabaster, AL, USA.) was used to prepare cell-size liposomes. Twenty microliters of the lipid in chloroform at the concentration of 2 mM was mixed with 12 μL of 10 mM glucose in methanol. The mixed solution was dried up in a glass vial by gentle nitrogen gas flow to make a thin film. Then, the film was kept under vacuum for 3 h before being hydrated with 200 μL of Milli-Q water at 37 °C for 2−3 h. The obtained cell-sized liposomes were used immediately. 23998

DOI: 10.1021/acsami.5b06781 ACS Appl. Mater. Interfaces 2015, 7, 23993−24000

Research Article

ACS Applied Materials & Interfaces pellet was resuspended in cold PBS and finally subjected to flow cytometry. Intracellular fluorescence detection was performed on a Cytomics FC500MPL (Becman Coulter Inc., NY, USA) flow cytometer using laser with the excitation wavelength of 488 nm and detected at 525 nm. At least 10 000 cells were measured for each sample. The acquired data were analyzed by FlowJo software (Tree Star, Inc.) using the mean fluorescence tool. All data were analyzed by one-way ANOVA and LSD using SPSS software. A p value of