pH-Switchable Coordinative Micelles for Enhancing Cellular

May 22, 2019 - (19) The cellular uptake would be improved effectively while introducing ... 8 mmol) were orderly dissolved in 20 mL of dichloromethane...
3 downloads 0 Views 6MB Size
Download PDF
Research Article www.acsami.org

Cite This: ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

pH-Switchable Coordinative Micelles for Enhancing Cellular Transfection of Biocompatible Polycations Qingyan Zhang,† Zhanwei Zhou,† Chenzi Li, Pengkai Wu, and Minjie Sun* State Key Laboratory of Natural Medicines and Department of Pharmaceutics, China Pharmaceutical University, Nanjing 210009, China

Downloaded by BUFFALO STATE at 22:06:45:575 on May 30, 2019 from https://pubs.acs.org/doi/10.1021/acsami.9b04668.

S Supporting Information *

ABSTRACT: Inefficient transfection of biocompatible lowmolecular-weight (LMW) polycations, such as 1.8k polyethylenimine (PEI), is a major challenge for successful nucleic acid delivery. Current strategies to improve transfection efficiency are bottlenecked by maintaining the balance between efficient gene encapsulation and on-demand cargo release. Here, we developed a new class of Zn(II)-coordinated micelles, which showed tight small interfering RNA (siRNA) binding and pH-switchable release. The dipicolylaminemodified PEI 1.8k (PD) and dopamine-conjugated cholesterol (Chol−Dopa) assemble into coordinative micelles (Zn−PD/ Chol−Dopa) via the coordination of 2,2′-dipicolylamine (DPA) and Dopa through Zn(II) as a bridge. The high phosphatebinding affinity of Zn−DPA enhanced the siRNA packaging and the interaction between cholesterol and cell membranes enhanced cellular uptake. Moreover, the coordination effect weakened in the acidic environment of lyso/endosome, triggering the disassembly of micelles and siRNA release. These properties of the micelles resulted in strong siRNA transfection efficiencies in various cell lines. Our strategy of constructing coordinative micelles improves the transfection efficiency of LMW PEI and holds tremendous potential to develop the endogenous responsive gene delivery systems. KEYWORDS: zinc coordination, pH switchable, cholesterol modification, serum stability, siRNA delivery

■

INTRODUCTION Since the discovery of the RNA interference pathway,1 the delivery of small interfering RNA (siRNA) has fueled tremendous interest for both research and therapeutics.2 A critical step for siRNA application is to enter cells and achieve higher transfection efficiency.3 Hence, many different types of synthetic vectors, such as cationic polymers, have been developed to deliver siRNA into cells. The delivery efficiency of cationic polymers usually depends on the N/P ratio, which means higher transfection efficiency with higher net-positive surface charge as well as an excess of free cationic polymers within the mixture.4 However, the normal cellular condition could be inhibited while the polyplexes with highly positive charges interact with cellular membranes and other components, resulting in destabilization of the clathrin-mediated endocytosis, membrane receptors, and cause cytotoxicity, eventually.5,6 For tuning the balance of siRNA delivery efficiency and cytotoxicity, multiple strategies were tried to circumvent this contradiction, including a new elaborate design and construction of gene delivery systems, functionalization of polymers with fluorinated polymers,7 aminoacids,8 lipids,9 cyclodextrins,10 peptides,11 other polymers,12 and inorganic nanoparticles.13 Developing novel polymeric vectors that balance higher efficiency and lower toxicity is the challenge to current clinical application.14 For example, the Cheng group © XXXX American Chemical Society

designed the vector of the supramolecular strategy constructed by a natural polyphenol (−)-epigallocatechin-3-O-gallate to achieve high transfection efficiency while having minimal toxicity.15 Moreover, weak interactions with nucleic acid before entering cells or inefficient release of the entrapped nucleic acid after entering cells also hurdle the nucleic acid delivery. It is great of significance to avoid the intrinsic defects of former cationic polymer vectors and further explore the range of applications of polycationic carriers. A prominent gene delivery carrier must not only have a high binding affinity toward siRNA but also have a superior ability to release entrapped siRNA after it enters the cell. Here, we were devoted to develop smart vectors and strategies for achieving high affinity and controlled release of nucleic acid. Therefore, we explored a new class of pH-sensitive cationic coordinative micelles by using zinc(II) coordination to combine the cationic shell to the hydrocarbon core. Recently, zinc coordination has become a developed strategy to enhance transfection efficiency of nucleic acid because of its high affinity to phosphodiester moieties16 and its virus-mimicking nature with endosomal membrane breaking.17 Binding of zinc coordination with phosphate-containing components would Received: March 15, 2019 Accepted: May 21, 2019 Published: May 22, 2019 A

DOI: 10.1021/acsami.9b04668 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Scheme 1. (A) Schematic Illustration of the Formation of Zn−PD/Chol−Dopa@siRNA-Coordinative Micelles. (B) pHTriggered Coordination Disassembly and siRNA Release: (a) Cell Uptake of Coordinative Micelles; (b) Endosomal Escape and pH-Triggered Micelles Disassembly of Zn−PD/Chol−Dopa@siRNA; (c) siRNA Silencing

■

tremendously benefit the siRNA condensation of lowmolecular weight cationic polymers with low cytotoxicity,18 and strengthen the combining affinity of cationic polymers with cell membranes.19 The cellular uptake would be improved effectively while introducing the zinc coordination to vectors.20 It inspires us to introduce zinc to equip siRNA delivery systems with suitable assembly forms to achieve desired gene vector with properties of tight cargo binding and controlled release. Consequently, coordinative micelles with core of hydrocarbon chains from cholesterol were first utilized for high transfection siRNA delivery in this study. It could play an important role in controlling stability and interaction of polyplexes with cell membranes during uptake and intracellular trafficking.21 Most importantly, the binding affinity of siRNA and the micelles packing the cationic polymers on the surface was enhanced significantly because of the high charge density. Zinc-coordination linkages can be disrupted in the endo/ lysosome where the environment is highly acidic, thereby causing the release of the entrapped siRNA inside the cytoplasm. Based on these design features, the pH-switchable zinc coordination of cationic micelles would be a gene delivery vector with effective transfection due to the tight siRNA binding, on-demand pH-switchable release, and low toxicity in vivo. In the assembly of core−shell micelles, zinc-linked Chol− Dopa, and polyethylenimine (PEI)−2,2′-dipicolylamine (DPA) (Zn−PD/Chol−Dopa) with coordination sites from zinc/hydroxy of Chol−Dopa and zinc/nitrogen, respectively (Scheme 1). The high phosphate-binding affinity of Zn−DPA enhanced the siRNA condensing. Moreover, the coordination effect much weakened in the low-pH environment of lyso/ endosome, triggering the disassembly of micelles and ondemand siRNA release. In this study, the transfection efficiency of Zn−PD/Chol−Dopa in vitro and in vivo had eventually verified the expectation of “superfast” gene transfection in various cell lines. Therefore, these zinc-coordination micelles provided flexible assembly strategy and enriched pH-switchable drug/gene delivery system. It holds great promise and further inspires researchers in the field of nonviral gene delivery.

EXPERIMENTAL SECTION

Materials. Branched polyethyleneimine (1.8k PEI and 25k PEI), DPA, α,α′-dichloro-p-xylene (Dx), dopamine, and cholesteryl

Figure 1. Synthesis procedures of (A) Zn−PD and (B) Chol−Dopa. chloroformate were purchased from commercial suppliers and used without further purification. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and 4′,6-diamidino-2-phenylindole (DAPI) were obtained from Dibo Chemical Technology. LysoTracker Red and GelRed were purchased from KeyGen Biotech. siEGFP (targeting EGFP) was obtained from Beyotime Biotechnology. siScr (scrambled siRNA), FAM-siScr, and siLuc (targeting luciferase) were obtained from GenePharma. Synthesis of Zn−PD Monomer and Chol−Dopa Monomer. Dx−DPA was synthesized according to the previously reported procedure.22 Briefly, Dx (2.80 g, 16 mmol) and K2CO3 (1.59 g, 8 mmol) were orderly dissolved in 20 mL of dichloromethane (DCM), afterward anhydrous DPA (0.79 g, 4 mmol) was added to the mixture. Then, the mixture was stirred at 40 °C and refluxed for 24 h. Then the product was purified using silica gel column with CH2Cl2/MeOH (30:1, v/v) as the eluent. The final Dx−DPA was obtained as a yellowish solid (0.18 g, 36% yield). For PD synthesis, PEI (0.344 g, 2 mmol) and DPA (0.338 g, 1 mmol) were respectively dissolved in 20 mL DCM, followed by B

DOI: 10.1021/acsami.9b04668 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces dropwise addition Dx−DPA into PEI. Then, the mixture was stirred at room temperature (RT) for 12 h. The solvent was removed by evaporation. The crude product was dissolved in DW and dialysis for 2 days to remove the unreacted Dx−DPA (Figure 1A). For Zn−PD synthesis, calculated PD, Zn(NO3)2·6H2O was mixed in water, then the reaction continued at RT for 4 h (Figure 1A). Dopamine (0.612 g, 2 mmol) dissolved into 20 mL mixture solvent (DCM/Methanol = 1:1 v/v) was added dropwise into the same solution which contained cholesterol chloroformate (1.8 g, 2 mmol) under nitrogen protection and then the reaction was conducted overnight. Afterward, the DCM layer was dried by using anhydrous Na2SO4 and the solvent was removed by rotary evaporation to give a white powder. Then the column chromatography was used to purify the product (Figure 1B). Preparation and Characterization of Zn−PD/Chol−Dopa Micelles. The Zn−PD/Chol−Dopa-coordinative micelles were prepared by the coordination reaction between Zn−DPA and dopamine. Briefly, Zn−PD (10 mg) was dissolved in water (5 mL) and Chol−Dopa (dissolved in methanol) was added dropwise to Zn− PD to obtain a series of Dopa/DPA with molar ratios of 0.1, 0.15, 0.2, 0.25, and 0.3, the mixtures were stirred at RT for 6 h. The methanol was finally removed by a rotary evaporator. The size distribution, zeta potential of Zn−PD, and Zn−PD/ Chol−Dopa micelles were detected by Malvern Zetasizer Nano ZS. The morphological observation of Zn−PD/Chol−Dopa and Zn− PD/Chol−Dopa@siRNA polyplexes at pH 7.4 and pH 5.0 were obtained using transmission electron microscope (TEM). Preparation and Characterization of the siRNA Loaded Micelles. To quantify gene packaging ability, polyplexes prepared at different w/w were mixed with siRNA at a 1:1 volume ratio in 4-(2hydroxyethyl)-1-piperazineethanesulfonic acid buffer (pH 7.4) and incubated for 30 min. After that, the amount of unpackaged siRNA was measured by agarose gel electrophoresis. Particle sizes were measured with dynamic light scattering, and zeta potentials were determined with electrophoretic light scattering. To measure the sizes and zeta potentials, polyplexes were prepared with w/w of 2, 4, 6, and 8. The w/w ratios for Zn−PD@siRNA or Zn− PD/Chol−Dopa@siRNA polyplexes in this study were expressed as equivalent PEI/siRNA ratios (i.e., excluding Zn−DPA or Chol−Dopa contents) and w/w of 2, 4, 6, and 8 was equal to N/P ratios of 14.8, 29.6, 44.4, and 59.2. To analyze the colloidal stability of polyplexes (PEI 1.8k@siRNA, PD@siRNA, Zn−PD@siRNA and Zn−PD/Chol−Dopa@siRNA), the micelles were prepared at w/w 4 and stored at 4 °C. The particle size was monitored at time intervals (0.5, 2, 4, 6, 12, 24, 36, and 48 h). Analysis of pH-Sensitivity of Zn−PD/Chol−Dopa and siRNA Release. To verify that the micelles would be separated into two different peaks under acidic conditions, the pH sensitivity of Zn−PD/ Chol−Dopa was analyzed by high-performance liquid chromatography (HPLC) using a C18 column with pH 7.0 phosphate buffer− isopropylalcohol (3:7, v/v) elution at 210 nm. For the siRNA release study, the Zn−PD/Chol−Dopa@siRNA micelles prepared at different w/w ratios were cultured with phosphate-buffered saline (PBS) of pH 5.0 and 7.4 at 37 °C for 2 h. Afterward, the pH-sensitivity-released siRNA from the micelles was measured using agarose gel electrophoresis. Cell Culture. The Chinese hamster ovary (CHO) cell lines, murine melanoma cell lines B16F10 and mouse breast cancer cell lines 4T1 were maintained in RPMI-1640 medium with 10% fetal bovine serum (FBS). The human breast cancer cell line MDA-MB231 was cultured in Dulbecco’s modified Eagle medium containing 10% FBS. Cytotoxicity. The cytotoxicity was analyzed by the MTT method. Briefly, CHO cells were seeded in 96-well plates at a density of 1−2 × 104 cells per well and cultured at 37 °C. siRNA polyplexes (PEI 1.8k, PD, Zn−PD and Zn−PD/Chol−Dopa@siRNA) with increasing w/w ratio were added, after incubation for 24 h, 20 μL MTT solution (5 mg/mL) was added to each well. After incubation for 4 h, 100 μL dimethyl sulfoxide was added to replace the media to dissolve

formazan, the absorbance was analyzed at a wavelength of 490 nm with a microplate reader. Gene Silencing in Vitro. CHO-EGFP cells were seeded in black 96-well plates at the density of 1 × 104 cells per well. Then polyplexes packing with siEGFP (100 nM) were added at different w/w ratios with FBS or FBS-free medium for 4 h, and the control group was treated with PBS. Then the polyplexes were replaced with complete medium containing 10% FBS. After 24 h, the expression of EGFP was measured by flow cytometry and fluorescence microscopy. Luciferase (Luc) silencing in vitro was performed on B16F10-Luc, 4T1-Luc, and MDA-MB-231-Luc cell lines. The cells (4 × 104) were typsinized and seeded into 48-well plates. After 24 h, the siLuc micelles at different w/w ratios diluted with FBS or FBS-free medium were added to the plates, and the control group was treated with PBS. After 4 h, the micelles were incubated with RPMI 1640 medium with 10% FBS for 24 h. Then, the cells were lysed with cell lysis buffer after being washed with PBS three times. The detection of luciferase activity was performed using a Luc assay kit according the specification. Cellular Uptake and Intracellular Trafficking. The cells (B16F10-Luc) were seeded into 24-well plates. After culturing to 60% confluence, the cells were incubated with the polyplexes prepared with FAM-siRNA (100 nM) for 2 and 4 h. The cells were then trypsinized and uptake-analyzed by BD FACSCalibur flow cytometer. Intracellular distribution was assessed by confocal microscopy following DAPI staining 4 h after incubation. Gene Silencing in Vivo. To assess Luc silencing in vivo, the left hind legs of C57BL/6 mice were subcutaneously injected of B16F10 cells (1 × 105) that stably express Luc. The D-luciferin potassium salt (200 μL, 15 mg·mL−1) was administered intraperitoneally to anesthetize the mice until tumor size reached 100 mm3. After measuring and recording the Luc bioluminescence on day 0 by using Tanon 5200 multi-imaging system (Tanon Science & Technology Inc), the mice were then divided into 6 groups (3 mice/group) and administered with the polyplexes by intratumoral injection (100 μL, 1.2 mg/kg siRNA, w/w 4). The treatments were repeated on day 1 and bioluminescence of the tumors was measured on day 2. In the end, the tumors were excised from the mice, homogenized in a lysis buffer followed by centrifugation at 12 000 rpm for 10 min for the detection of Luc expression. Statistical Analysis. Statistical assessment included two-sided Student’s t-test analysis for 2 groups. Differences were considered to be statistically significant at p < 0.05.

■

RESULTS AND DISCUSSION Synthesis and Characterization of Dx−DPA, PD, and Chol−Dopa. The Dx−DPA was synthesized as reported22 (Figure S1). As shown in Figure S2, the peak of 3.8 and 4.6 was the signal of methylene on benzyl chloride. The peaks among 7−9 ppm were assigned to the signal of pyridine of Dx and phenyl of benzyl chloride. The comparision of the peak area at 3.8 and 8.6 ppm proved the successful synthesis of Dx− DPA. The conjugation of Dx−DPA on PEI was obtained by substituting hydrogen atom with chlorine atom of Dx (Figure 1A). The peaks among 2−3 were the signal of methylene of PEI. The substation degree of Dx−DPA on PEI was 41.3% (Figure S2). Chol−Dopa was synthesized by amidation reaction (Figure 1B). It was characterized by 1H NMR shown in Figure S2. In the result, the peaks among 6−7 ppm were ascribed to the signals of the phenyl proton of dopamine and the peaks among 0−2 ppm were ascribed to methylene of cholesterol. Preparation of Zn−PD/Chol−Dopa-Coordinative Micelles. The amphiphilic-coordinative micelles (Zn−PD/ Chol−Dopa) were formed through coordination reaction between the zinc ion with positive charge of hydrophilic C

DOI: 10.1021/acsami.9b04668 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

condensed siRNA at w/w ratio 1.0 completely. As expected, Zn−PD/Chol−Dopa-coordinative micelles further improved the siRNA binding ability at w/w 0.75, which was probably because the siRNA binds to the surface of the coordinative micelles as opposed to bind with Zn−PD and assembling into the polyplexes. Hydrodynamic diameter and zeta potential of the Zn−PD/ Chol−Dopa@siRNA-coordinative micelles were then measured at various w/w ratios. As shown in Figure 3B,C, for PEI 1.8k@siRNA, PD@siRNA or Zn−PD@siRNA, the particle size gradually decreased with an increasing w/w ratio, indicating tighter binding with the higher w/w ratios. However, it is worth noting that Zn−PD/Chol−Dopa@ siRNA-coordinative micelles showed no significant change of the particle size at different w/w ratios. It proved that the micelles had more compacted structure than other complexes and siRNA loading could not condense the surface polymer. Meanwhile, coordinative micelles exhibited higher zeta potential at the same w/w ratio compared with other polyplexes, which further confirmed the stronger charge density and higher siRNA compacting ability of the micelles. These results manifested that siRNA was condensed more tightly with Zn−PD/Chol−Dopa-coordinative micelles at the same w/w ratio than other control polymers. Therefore, we assumed that larger charge density was achieved as the Chol− Dopa moieties were bound tightly with Zn−PD through coordination between Zn and phenolic hydroxyl group of dopamine. The Zn coordination could lightly undergo the acidresponsive structural separation, which likely effective releases siRNA at different pH conditions.23 To clarify this hypothesis, the acid-triggered coordination disassembly was first checked by the changed peak shape of Zn−PD/Chol−Dopa before and after coordination on the HPLC spectrum. As shown in Figure S3, the peak of free Chol−Dopa disappear with the formation of Zn−PD/Chol−Dopa micelles. Interestingly, after incubation at pH 5.0, the peak of free Chol−Dopa appears again, indicating the breakup of the coordination bond and disassembly of micelles. We next investigated whether the disassembly of micelles would result in release of the siRNA cargo. The gel electrophoresis was performed at pH 7.4 and 5.0 with an increasing w/w ratio. As shown in Figure 3D, Zn−PD/Chol− Dopa@siRNA micelles have little leakage at w/w range from 0.75 and 4.0 at pH 7.4, as reflected by the no band of siRNA signals. On the contrary, the initially entrapped siRNA was effused entirely from the gel even at w/w 4 while controlling pH at 5.0. Finally, we further demonstrated the acid-triggered degradation of the Zn−PD/Chol−Dopa@siRNA complexes by morphology changes observed by TEM. As shown in Figure 3E, the morphology of the Zn−PD/Chol−Dopa@siRNA complexes was visualized by TEM to be that of a sphere at pH 7.4. After incubation at pH 5.0, the complexes changed from a spherical shape to a loose, random morphology, indicating the degradation of the complexes. It confirmed that the Zn−PD/Chol−Dopa@siRNA complexes could easily release siRNA in the acid microenvironments such as lysosome compartments while holding on a steady state in the normal condition. Hence, the timely release of siRNA inside cells would be promoted in lysosome with Zn−PD/Chol−Dopa delivery. It was believed that the acid-responsive dissociation of the coordinative micelles results in the unpacking of the initially tightly entrapped siRNA under acid condition.

Figure 2. (A) Size distribution of Zn−PD/Chol−Dopa-coordinative micelles. (B) Zeta potential of Zn−PD/Chol−Dopa micelles. (C) Morphology observation of Zn−PD/Chol−Dopa micelles by TEM, scale bar: 100 nm. (D) Photograph and Tyndall effect of Zn−PD/ Chol−Dopa micelles.

cationic Zn−PD and phenolic hydroxyl group on hydrophobic Chol−Dopa. The molar ratio of Zn−PD/Chol−Dopa was optimized by the particle size. As can be seen in Figure S4A, with the increase of molar ratios from 0.1 to 0.3, the particle size of Zn−PD/Chol−Dopa presented a down and up trend, whereas the lowest size turns out at a point where the molar fraction of Chol−Dopa was 0.2. It meant that the introduction of Chol−Dopa condensed the coordinative micelles while the excess hydrophobic ligand unbalances amphiphilicity of the Zn−PD/Chol−Dopa polymer. Therefore, the ratio of 0.2:1 was selected for further research. The size and zeta potential of Zn−PD before and after Chol−Dopa complexation was shown in Figures 2A,B and S4B,C, the introduction of Chol−Dopa component condensed nanoparticles from 218.5 to 159.6 nm. With the decrease of the size, the zeta potential of the nanoparticles was increased from 21.4 to 56 mV, which proved that the formation of coordinative micelles could condense the charge density of polycations and compact the siRNA effectively. In addition, the Zn(II) contained in coordinative micelles was detected by inductively coupled plasma mass spectrometry as 9.51 mg/100 mg Zn−PD/Chol−Dopa. The particles imaged observed by TEM were spherical and the size was approximately 110 nm (Figure 2C). A beam of light was observed in solution (Figure 2D), which was the result of the Tyndall effect, indicating the micelles were formed. Preparation and Characterization of the Polyplexes. We hypothesized that the DPA/Zn ligand could improve the nucleic acid packaging due to the high binding affinity between Zn and the phosphate group. Moreover, DPA−Zn was utilized to improve the interaction between polymers and cell membranes because of the high affinity of zinc ions and phospholipids. In this study, siRNA encapsulation with Zn− PD/Chol−Dopa-coordinative micelles was prepared and evaluated by gel retardation assay. As can be seen in Figure 3A, as well as PEI 1.8k, PD without Zn coordination showed relative weak siRNA packaging ability while Zn−PD D

DOI: 10.1021/acsami.9b04668 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 3. (A) siRNA encapsulation assessed by agarose gel electrophoresis of polyplexes prepared at different w/w ratios. (B) Particle size and (C) zeta potential of PEI 1.8k@siRNA, PD@siRNA, Zn−PD@siRNA and Zn−PD/Chol−Dopa@siRNA polyplexes prepared at different w/w ratios. (D) siRNA release under acidic milieu. (E) Morphology observation of Zn−PD/Chol−Dopa@siRNA polyplexes at pH 7.4 and 5.0 by TEM. (F) Colloid stability of PEI 1.8k@siRNA, PD@siRNA, Zn−PD@siRNA, and Zn−PD/Chol−Dopa@siRNA polyplexes at w/w 4.

In addition, to verify the stability in physiological conditions of Zn−PD/Chol−Dopa@siRNA micelles, its colloidal stability was evaluated by incubating them in PBS (pH 7.4) and monitoring the changes in hydrodynamic diameter over time. As shown in Figure 3F, while the control groups of PEI, PD/ siRNA polyplexes aggregated rapidly, Zn−PD polyplexes exhibited a certain resistance toward PBS, and Zn−PD/ Chol−Dopa@siRNA micelles could maintain a particle size below 200 nm for at least 24 h. As in the previous report, the beneficial effect of Zn coordination22 and cholesterol24 contributed to the steady colloidal stability of Zn or cholesterol-included polyplexes. Transfection Efficiency in CHO-EGFP Cell Lines. Because those Zn−PD/Chol−Dopa-coordinative micelles showed high siRNA-binding affinities and feasibility in releasing the entrapped siRNA into cells, superfast transfection efficiency was expected in the following study. Before assaying transfection efficiency in CHO-EGFP cells, we first examined the cytotoxicity of polyplexes formed by polymer and siScr because of the critical role of toxicity of the gene vector in clinical application. As shown in Figure 4A, Zn−PD/Chol− Dopa@siRNA micelles exhibited the toxicity with cell viability remaining about 60% up to w/w 8. On the contrary, the PEI

25k group decreases to 10% in cell viability at w/w 8, which is commonly used as positive control for gene transfection assessment. We determined the transfection efficiencies of polyplexes with w/w = 4 based on the abovementioned cytotoxicity data. We first excluded the nonspecific effects of the polymer on CHO cells by treating the cells with micelles formulated with control siScr (Figure S5A). Then, as shown in Figure 4B−C, Zn−PD/Chol−Dopa-coordinative micelles proved its advantage in helping the gene delivery based on up to 90% gene silencing. In contrast, Zn−PD just showed a moderate EGFP silencing at 70%. Certainly, the transfection efficacy of polyplexes formed by Zn−PD and Zn−PD/Chol−Dopacoordinative micelles was higher than the parent PEI 1.8k polyplexes and PEI 25k polyplexes. It was thought that the coordinative micelles would undergo dissociation induced by the destabilization of the complexes when they undergo the endosome-to-lysosome progression along with increasing acidity, leading to the efficient siRNA release. Moreover, the satisfied transfection efficiency also could attribute to cholesterol, which improved the interaction of polyplexes with cell membranes during uptake and intracellular trafficking. With these properties, Zn−PD/Chol−Dopa-coordinative E

DOI: 10.1021/acsami.9b04668 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 4. (A) Cytotoxicity of PEI 25k@siRNA, PEI 1.8k@siRNA, PD@siRNA, Zn−PD@siRNA, and Zn−PD/Chol−Dopa@siRNA on CHO cells at various w/w ratios. (B) EGFP silencing mediated by PEI 25k@siEGFP, PEI 1.8k@siEGFP, PD@siEGFP, Zn−PD@siEGFP, and Zn−PD/ Chol−Dopa@siEGFP at w/w 4. (C) Fluorescence images of EGFP protein expression after different treatments detected by fluorescence microscopy, and the control group was treated with PBS. Scale bar = 200 μm. Data are shown as mean ± SD (n = 3). **p < 0.01.

4T1, and MDA-MB-231 cell lines with Luc expression. As shown in Figure 5, the Luc silencing effect was improved in all polyplexes with increasing w/w ratio. Zn−PD/Chol−Dopacoordinative micelles@siLuc polyplexes achieved the best silencing effect when compared with other polyplex groups even in primary cells (4T1 and MDA-MB-231), which belong to hard-to-transfect cell types.22 One of the critical challenges for gene therapy is that gene vector would only introduce gene into immortalized cell lines while little transfection to the nonproliferating cell types.22,25 Like PEI 25k, it just exhibited a weak Luc silencing at ∼30% in 4T1 and MDA-MB-231 cell lines, while Zn−PD showed an improvement in these two cell lines compared with PEI 1.8k and PEI 25k at w/w 4, which was consistent with the previous report.24 Furthermore, Zn− PD/Chol−Dopa-coordinative micelles mediated a robust gene transfection in 4T1 and MDA-MB-231 cells and the Luc silencing could achieve 70% at w/w 4. Serum Resistance of Zn−PD/Chol−Dopa-Coordinative Micelles. For clinical applications of gene delivery systems, serum resistance is a significant parameter because cationic polymers were easily be caught by the inversely charged blood components during the circulation time.26 Many studies have reported that cholesterol introduction into the polycation was a beneficial factor in improving serum stability of the polyplexes for the unique nature of cholesterol.27,28 Herein, transfection efficacy was demonstrated with increasing serum content in the incubation media on the siEGFP and siLuc transfection activity. First, we excluded the nonspecific effects of the polymer on different cells with or without serum by treating the cells with micelles formulated

Figure 5. Luciferase silencing of PEI 25k@siLuc, PEI 1.8k@siLuc, Zn−PD@siLuc, and Zn−PD/Chol−Dopa@siLuc on B16F10-Luc, 4T1-Luc, and MDA-MB-231-Luc cells in the absence of FBS.

micelles were speculated to be a prominent gene vector, prospectively, achieving robust gene transfection in many cell lines. Transfection Efficiency in Cell Lines Expressing Luc. To verify the EGFP silencing results in other target gene and cell lines, we have also the tested silencing effect in B16F10, F

DOI: 10.1021/acsami.9b04668 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 6. (A) Fluorescence images of EGFP protein expression on CHO-EGFP cells after different treatments in the presence of 10, 20, and 30% FBS detected by fluorescence microscopy, and the control group was treated with PBS. Scale bar = 200 μm. (B) EGFP silencing of PEI 1.8k@ siLuc, Zn−PD@siLuc, Zn−PD/Chol−Dopa@siLuc and PEI 25k@siLuc on CHO cell lines and luciferase silencing of PEI 1.8k@siLuc, Zn−PD@ siLuc, Zn−PD/Chol−Dopa@siLuc and PEI 25k@siLuc on B16F10-Luc, 4T1-Luc and MDA-MB-231-Luc cell lines in the presence of 10, 20, and 30% FBS.

cellular trafficking were evaluated by flow cytometry and confocal microscopy using FAM-siRNA-loaded Zn−PD/ Chol−Dopa. As shown in Figure 7A,B, Zn−PD/Chol− Dopa/siRNA complexes achieved the highest cell uptake reflected in the percentage of positive cells and the mean fluorescence intensity per cell. In contrast, Zn−PD and PEI 25k polyplexes just showed about half of transfection efficiency of Zn−PD/Chol−Dopa/siRNA complexes. We then confirmed these findings by confocal microscopy, as shown in Figure 7C, the green fluorescence intensity showed a decreased manner among groups of Zn−PD/Chol−Dopa, Zn−PD, and PEI 1.8k, respectively. The high uptake of Zn−PD/Chol−

with control siScr (Figure S5B). Then, as shown in Figure 6, Zn−PD/Chol−Dopa-coordinative micelles maintained a large proportion of transfection efficiency while the medium contains 10−30% serum. Contrarily, it was observed that much lower efficacy of PEI 1.8k, Zn−PD and PEI 25k in CHO-GFP cell lines. This much less influences of serum presence on the stability of Zn−PD/Chol−Dopa-coordinative micelles could also be observed in B16F10, 4T1, and MDAMB-231 cell lines. Cellular Uptake and Intracellular Trafficking. To visualize the intracellular behavior of the Zn−PD/Chol− Dopa@siRNA-coordinative micelles, cell uptake and intraG

DOI: 10.1021/acsami.9b04668 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 7. (A) FAM-siRNA positive cell analysis after treatment with PEI 25k/FAM-siRNA, PEI 1.8k@FAM-siRNA, Zn−PD@FAM-siRNA, and Zn−PD/Chol−Dopa@FAM-siRNA on B16F10 cells. (B) Cell uptake of PEI 25k@FAM-siRNA, PEI 1.8k@FAM-siRNA, Zn−PD@FAM-siRNA, and Zn−PD/Chol−Dopa@FAM-siRNA detected by FCM. Data are shown as mean ± SD (n = 3). **p < 0.01. (C) Confocal laser scanning microscopy observation of the cell uptake of PEI 1.8k@FAM-siRNA, Zn−PD@FAM-siRNA and Zn−PD/Chol−Dopa@FAM-siRNA. (D) Endosomal escape of PEI 1.8k@FAM-siRNA, Zn−PD@FAM-siRNA, and Zn−PD/Chol−Dopa@FAM-siRNA after 6 h incubation with B16F10 cells.

Dopa-coordinative micelles was attributed to the high affinity between Zn-coordinated polyplexes and phospholipid cell membranes and the beneficial effect of cholesterol as a good lipid anchor.29 It is important that the polyplexes are able to escape from the endosomes and release siRNA for subsequent gene silencing in addition to effective cell uptake.30 Accordingly, the endo/lysosomal escape of Zn−PD/Chol−Dopa complexes was evaluated by micelles loaded with FAM-siRNA. As shown in Figure 7D, lysosomes were stained with LysoTracker at 2 h and most of the endocytosed FAM-siRNA was located in lysosomes for PEI 1.8k polyplexes. Most of the FAM-siRNA of Zn−PD/Chol−Dopa@FAM-siRNA polyplexes separated from the LysoTracker signal compared with other polyplex groups, meaning efficient endo/lysosomal release. Overall, these results demonstrate that Zn−PD/Chol−Dopa can improve the cell uptake and achieve efficient release in cytoplasm, and thus contributing to the superior gene silencing activity. Luc Gene Silencing in Vivo. To validate whether Zn− PD/Chol−Dopa-coordinative micelles could achieve better silencing effect in vivo, the subcutaneous B16F10-Luc tumor model have been established. The tumor bioluminescent images were taken and analyzed before and after treatment with the siLuc-loading polyplexes by intratumoral injection. As shown in Figure 8, control animals were treated with PBS, free siLuc, and siScr and limited Luc silencing (or no Luc silencing) was observed, which could exclude the factor that the polymers had any nonspecific off-target effects. Treatment with PEI 1.8k@siLuc, Zn−PD@siLuc, and Zn−PD/Chol−Dopa@siScr polyplexes lead to the declining Luc expression in the tumors after two days injection (Figure 8A,B). Zn−PD/Chol−Dopa@ siLuc complexes exhibited the weakest fluorescence intensity

compared with other polyplex groups. To verify the bioluminescence findings, we excised and homogenized the tumors followed by measured the Luc activity using the previous way. Zn−PD/Chol−Dopa@siRNA polyplexes showed 80% Luc silencing, while the PEI 1.8k and Zn−PD polyplexes showed 22 and 46% Luc silencing (Figure 8C). Taken together, all these results indicated that there is a great potentiality of this acid-triggered responsive gene delivery system based on the zinc(II) coordination linker for hydrophobic core in the area of developing new gene vectors.

■

CONCLUSIONS

In summary, we have reported a novel siRNA delivery strategy that, for the first time, takes the combination of introducing hydrophobic modification and coordinative assembly into micelles. Both decorations enhanced the siRNA encapsulation, cellular uptake compared with PEI@siRNA polyplex. Besides, the cholesterol modification improved the colloidal stability of DPA/Zn-modified polyplexes. Interestingly, the coordinative assembly between Zn−PD and Chol−Dopa was pHresponsive which could remain stable in the physiological environment but undergo cleavage in the acidic tumor milieu, assisting with the selective release of siRNA in the tumor cell lines. Thus, the pH-switchable coordinative micelles achieved superior siRNA transfection efficiency on various cell lines in the absence or presence of FBS. This strategy was validated to give a new example of siRNA-delivery system design and it held tremendous potential to be developed as siRNA transfection reagent. H

DOI: 10.1021/acsami.9b04668 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 8. (A) Bioluminescence images of B16F10-Luc tumor bearing mice before and after treatment, siRNA dose: 1.2 mg/kg. (B) Quantification of luciferase expression from the bioluminescence images by ImageJ. (C) Relative luciferase expression of ex vivo tumor detected by the luciferase assay kit. Data are shown as mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.005.

■

ORCID

ASSOCIATED CONTENT

S Supporting Information *

Zhanwei Zhou: 0000-0002-7149-3223 Minjie Sun: 0000-0003-0582-6189

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.9b04668.

■

Author Contributions †

Q.Z. and Z.Z. contributed equally to the manuscript.

Synthesis procedures of Dx−DPA; 1H NMR spectra of DPA and Dx−DPA, solvent: CDCl3, PEI and PEI−DPA, solvent: D2O, Chol−Dopa, solvent: CD3OD; HPLC spectrum of Chol−Dopa, Zn−PD, Zn−PD/Chol−Dopa (pH 7.4), and Zn−PD/Chol−Dopa (pH 5.0); particle size determination of Zn−PD/Chol−Dopa at various molar ratios of Dopa/DPA, size, and zeta potential of Zn−PD and Zn−PD/Chol−Dopa micelles; fluorescence images of EGFP protein expression after different treatments detected by fluorescence microscopy and the control group was treated with PBS; scale bar = 200 μm; EGFP silencing of PEI 1.8k@siScr, Zn−PD@siScr, Zn−PD/Chol−Dopa@siScr, and PEI 25k@siScr on CHO cell lines and luciferase silencing of PEI 1.8k@ siScr, Zn−PD@siScr, Zn−PD/Chol−Dopa@siScr, and PEI 25k@siScr on B16F10-Luc, 4T1-Luc, and MDAMB-231-Luc cell lines in the presence of 0, 10, 20, and 30% FBS (PDF)

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was financially supported by The National Key Research and Development Program of China (2017YFA0205400), the National Natural Science Foundation of China (no. 81573377), and The Jiangsu Fund for Distinguished Youth (BK20170028).

■

REFERENCES

(1) Fire, A.; Xu, S.; Montgomery, M. K.; Kostas, S. A.; Driver, S. E.; Mello, C. C. Potent and Specific Genetic Interference by DoubleStranded RNA in Caenorhabditis Elegans. Nature 1998, 391, 806− 811. (2) Davis, M. E.; Zuckerman, J. E.; Choi, C. H. J.; Seligson, D.; Tolcher, A.; Alabi, C. A.; Yen, Y.; Heidel, J. D.; Ribas, A. Evidence of RNAi in Humans from Systemically Administered siRNA via Targeted Nanoparticles. Nature 2010, 464, 1067−1070. (3) Mastrobattista, E.; Hennink, W. E. Polymers for Gene Delivery: Charged for Success. Nat. Mater. 2012, 11, 10−12.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone/Fax: +86 25 83271098. I

DOI: 10.1021/acsami.9b04668 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces (4) Ding, Y.; Jiang, Z.; Saha, K.; Kim, C. S.; Kim, S. T.; Landis, R. F.; Rotello, V. M. Gold Nanoparticles for Nucleic Acid Delivery. Mol. Ther. 2014, 22, 1075−1083. (5) Liu, H.; Wang, H.; Yang, W.; Cheng, Y. Disulfide Cross-Linked Low Generation Dendrimers with High Gene Transfection Efficacy, Low Cytotoxicity, and Low Cost. J. Am. Chem. Soc. 2012, 134, 17680−17687. (6) Zhou, J.; Liu, J.; Cheng, C. J.; Patel, T. R.; Weller, C. E.; Piepmeier, J. M.; Jiang, Z.; Saltzman, W. M. Biodegradable Poly(Amine-co-Ester) Terpolymers for Targeted Gene Delivery. Nat. Mater. 2011, 11, 82−90. (7) Cheng, Y. Fluorinated Polymers in Gene Delivery. Acta Polym. Sin. 2017, 8, 1234−1245. (8) Wang, L.-H.; Wu, D.-C.; Xu, H.-X.; You, Y.-Z. High DNABinding Affinity and Gene-Transfection Efficacy of Bioreducible Cationic Nanomicelles with a Fluorinated Core. Angew. Chem., Int. Ed. 2016, 55, 755−759. (9) Liu, J.; Jiang, Z.; Yang, Z.; Zhang, X.; Luo, X.; Jiang, Q.; Saltzman, W. M. Enzymatic Synthesis of Poly(Amine-co-Esters) for Gene Delivery. J. Controlled Release 2013, 172, E106−E107. (10) de Ilarduya, C. T.; Sun, Y.; Düzgüneş, N. Gene Delivery by Lipoplexes and Polyplexes. Eur. J. Pharm. Sci. 2010, 40, 159−170. (11) Lv, H.; Zhang, S.; Wang, B.; Cui, S.; Yan, J. Toxicity of Cationic Lipids and Cationic Polymers in Gene Delivery. J. Controlled Release 2006, 114, 100−109. (12) Zeng, H.; Little, H. C.; Tiambeng, T. N.; Williams, G. A.; Guan, Z. Multifunctional Dendronized Peptide Polymer Platform for Safe and Effective siRNA Delivery. J. Am. Chem. Soc. 2013, 135, 4962−4965. (13) Yuba, E.; Nakajima, Y.; Tsukamoto, K.; Iwashita, S.; Kojima, C.; Harada, A.; Kono, K. Effect of Unsaturated Alkyl Chains on Transfection Activity of Poly(Amidoamine) Dendron-Bearing Lipids. J. Controlled Release 2012, 160, 552−560. (14) Babu, A.; Muralidharan, R.; Amreddy, N.; Mehta, M.; Munshi, A.; Ramesh, R. Nanoparticles for siRNA-Based Gene Silencing in Tumor Therapy. IEEE Trans. Nanobiosci. 2016, 15, 849−863. (15) Shen, W.; Wang, Q.; Shen, Y.; Gao, X.; Li, L.; Yan, Y.; Wang, H.; Cheng, Y. Green Tea Catechin Dramatically Promotes RNAi Mediated by Low-Molecular-Weight Polymers. ACS Cent. Sci. 2018, 4, 1326−1333. (16) Chu, C.; Ren, E.; Zhang, Y.; Yu, J.; Lin, H.; Pang, X.; Zhang, Y.; Liu, H.; Qin, Z.; Cheng, Y.; Wang, X.; Li, W.; Kong, X.; Chen, X.; Liu, G. Zinc(II)-Dipicolylamine Coordination Nanotheranostics: Toward Synergistic Nanomedicine by Combined Photo/Gene Therapy. Angew. Chem., Int. Ed. 2019, 58, 269−272. (17) Liu, S.; Yang, J.; Jia, H.; Zhou, H.; Chen, J.; Guo, T. Virus Spike and Membrane-Lytic Mimicking Nanoparticles for High Cell Binding and Superior Endosomal Escape. ACS Appl. Mater. Interfaces 2018, 10, 23630−23637. (18) Liu, S.; Jia, H.; Yang, J.; Pan, J.; Liang, H.; Zeng, L.; Zhou, H.; Chen, J.; Guo, T. Zinc Coordinated Cationic Polymers Break Up the Paradox Between Low Molecular Weight and High Transfection Efficacy. Biomacromolecules 2018, 19, 4270−4276. (19) Liu, S.; Jia, H.; Yang, J.; Pan, J.; Liang, H.; Zeng, L.; Zhou, H.; Chen, J.; Guo, T. Zinc Coordination Substitute Amine: A Noncationic Platform for Efficient and Safe Gene Delivery. ACS Macro Lett. 2018, 7, 868−874. (20) Huang, S.; Li, J.; Han, L.; Liu, S.; Ma, H.; Huang, R.; Jiang, C. Dual Targeting Effect of Angiopep-2-Modified, DNA-Loaded Nanoparticles for Glioma. Biomaterials 2011, 32, 6832−6838. (21) Wu, P.; Luo, X.; Wu, H.; Yu, F.; Wang, K.; Sun, M.; Oupicky, D. Cholesterol Modification Enhances Antimetastatic Activity and siRNA Delivery Efficacy of Poly(ethylenimine)-Based CXCR4 Antagonists. Macromol. Biosci. 2018, 18, 1800234. (22) Liu, S.; Zhou, D.; Yang, J.; Zhou, H.; Chen, J.; Guo, T. Bioreducible Zinc(II)-Coordinative Polyethylenimine with Low Molecular Weight for Robust Gene Delivery of Primary and Stem Cells. J. Am. Chem. Soc. 2017, 139, 5102−5109.

(23) Wang, M.; Liu, H.; Li, L.; Cheng, Y. A Fluorinated Dendrimer Achieves Excellent Gene Transfection Efficacy at Extremely Low Nitrogen to Phosphorus Ratios. Nat. Commun. 2014, 5, 3053−3060. (24) Zhu, J.-Y.; Zeng, X.; Qin, S.-Y.; Wan, S.-S.; Jia, H.-Z.; Zhuo, R.X.; Feng, J.; Zhang, X.-Z. Acidity-responsive gene delivery for “superfast” nuclear translocation and transfection with high efficiency. Biomaterials 2016, 83, 79−92. (25) Wang, Y.; Li, J.; Chen, Y.; Oupický, D. Balancing Polymer Hydrophobicity for Ligand Presentation and siRNA Delivery in Dual Function CXCR4 Inhibiting Polyplexes. Biomater. Sci. 2015, 3, 1114− 1123. (26) Nishimura, T.; Yamada, A.; Umezaki, K.; Sawada, S.-i.; Mukai, S.-a.; Sasaki, Y.; Akiyoshi, K. Self-Assembled Polypeptide Nanogels with Enzymatically Transformable Surface as a Small Interfering RNA Delivery Platform. Biomacromolecules 2017, 18, 3913−3923. (27) Dolivet, G.; Merlin, J.-L.; Barberi-Heyob, M.; Ramacci, C.; Erbacher, P.; Parache, R.-M.; Behr, J.-P.; Guillemin, F. In vivo growth inhibitory effect of iterative wild-type p53 gene transfer in human head and neck carcinoma xenografts using glucosylated polyethylenimine nonviral vector. Cancer Gene Ther. 2002, 9, 708−714. (28) Yang, F.; Cho, S.-W.; Son, S. M.; Bogatyrev, S. R.; Singh, D.; Green, J. J.; Mei, Y.; Park, S.; Bhang, S. H.; Kim, B.-S.; Langer, R.; Anderson, D. G. Genetic Engineering of Human Stem Cells for Enhanced Angiogenesis Using Biodegradable Polymeric Nanoparticles. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 3317−3322. (29) Qiao, H.; Sun, M.; Su, Z.; Xie, Y.; Chen, M.; Zong, L.; Gao, Y.; Li, H.; Qi, J.; Zhao, Q.; Gu, X.; Ping, Q. Kidney-Specific Drug Delivery System for Renal Fibrosis Based on Coordination-Driven Assembly of Catechol-Derived Chitosan. Biomaterials 2014, 35, 7157−7171. (30) Wu, D.; Ensinas, A.; Verrier, B.; Primard, C.; Cuvillier, A.; Champier, G.; Paul, S.; Delair, T. Zinc-Stabilized ChitosanChondroitin Sulfate Nanocomplexes for HIV-1 Infection Inhibition Application. Mol. Pharmaceutics 2016, 13, 3279−3291.

J

DOI: 10.1021/acsami.9b04668 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX