Self-Assembled Cationic Micelles Based on PEG-PLL-PLLeu Hybrid

Article pubs.acs.org/Biomac

Self-Assembled Cationic Micelles Based on PEG-PLL-PLLeu Hybrid Polypeptides as Highly Effective Gene Vectors Jizhe Deng,† Ningning Gao,† Yanan Wang, Huqiang Yi, Shengtao Fang, Yifan Ma,* and Lintao Cai* Key Lab of Health Informatics of Chinese Academy of Sciences, Guangdong Key Laboratory of Nanomedicine, Shenzhen Key Laboratory of Cancer Nanotechnology, Institute of Biomedicine and Biotechnology, Shenzhen Institutes of Advance Technology, Chinese Academy of Sciences, P. R. China ABSTRACT: Developing safe and effective nonviral gene vector is highly crucial for successful gene therapy. In the present study, we designed a series of biodegradable micelles based on hybrid polypeptide copolymers of poly(ethylene glycol)-b-poly(L-lysine)-b-poly(L-leucine) (PEG-PLL-PLLeu) for efficient gene delivery. A group of amphiphilic PEG-PLLPLLeu hybrid polypeptide copolymers were synthesized by ring-opening polymerization of N-carboxyanhydride, and the chemical structure of each copolymer was characterized by 1H NMR and FT-IR spectroscopy measurement. The PEG-PLLPLLeu micelles were positively charged with tunable sizes ranging from 40 to 90 nm depending on the length of PLL and PLLeu segment. Compared with PEG-PLL copolymers, PEGPLL-PLLeu micelles demonstrated significantly higher transfection efficiency and less cytotoxicity. Furthermore, the transfection efficiency and biocompatibility of the micelles can be simultaneously improved by tuning the length of PLL and PLLeu segments. The transfection efficiency of PEG-PLL-PLLeu micelles in vivo was two to three times higher than that of PEI25k, which was attributable to their capability of promoting DNA condensation and cell internalization as well as successful lysosome escape. Hence well-defined PEG-PLL-PLLeu micelles would serve as highly effective nonviral vectors for in vivo gene delivery.

■

INTRODUCTION Gene therapy has been applied as an important therapeutic strategy to treat a variety of human diseases, such as cancer, monogenic disorders, cardiovascular diseases, infectious diseases, neurological diseases, and ocular diseases.1,2 Despite encouraging results from clinical trials, lacking safe and effective gene vectors remains the major barrier for successful gene therapy. Although viral gene vectors are currently the most effective vectors applied in 70% of gene therapy clinical trials, they are confronted with severe immunogenic effects and possible tumorigenesis as well as production problems. Over the past decade, nonviral vectors have drawn broad attention because of their potential of overcoming the defects of viral vectors.3 Among them, cationic polymers, such as polyethylenimine (PEI), polylysine (PLL), and polyamidoamine (PAMAM) dendrimer, have been intensively studied because of their well-defined chemo-physical characteristics, strong capability of DNA binding, and flexibility of chemical modification.4,5 However, their transfection efficiency in vivo is usually poor, which could be attributable to the poor stability and short half-life in vivo, insufficient uptake by target cells, incapability of avoiding lysosomes, and so on.6 Recently, cationic micelles self-assembled from amphiphilic cationic copolymers have been developed as novel gene vectors for DNA or siRNA delivery. In contrast with water-soluble cationic polymers, cationic micelles have a core−shell structure in aqueous medium with hydrophobic segments as the core and © XXXX American Chemical Society

cationic hydrophilic segments as the shell. This special structure can not only facilitates DNA condensation but also improve the stability of DNA complexes in vivo.7 Moreover, cationic micelles tend to generate compact nucleic acid complexes, which should be favorable to in vivo gene transfection.8 Previous studies have shown that cationic micelles selfassembled from copolymers, such as P(MDS-co-CES), PEGb-P[NDAPM-co-(HEMA-PCL)], PEG-b-PCL-b-PPEEA, PEGPLL-PLP, and PDMAEMA-PCL-PDMAEMA, exhibited considerable transfection efficiency in vitro, implying their potential for in vivo gene delivery.9−13 The cytotoxicity and immunogenecity of cationic micelles also raised safety concerns. Although the introduction of PEG molecules has been shown to reduce dramatically the cytotoxicity of cationic micelles both in vitro and in vivo, PEGylation also inhibited the uptake of nanoparticles and significantly diminished the downstream gene transfection.14,15 Therefore, the trade-off of improving both transfection efficiency and biocompatibility of cationic micelles remains to be further explored. Synthetic polypeptides are biocompatible and biodegradable polymers that have been widely used for drug and gene delivery, tissue engineering, surgical sealants, and so on.16 Compared with traditional polymers, the component and Received: August 8, 2012 Revised: September 19, 2012

A

dx.doi.org/10.1021/bm3012538 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

(1:20:40), PEG-PLL30-PLLeu20 (1:30:20), and PEG-PLL30-PLLe40 (1:30:40), were synthesized. PEG-PLL and PEG-PLL-PLLeu copolymers were obtained by the deprotection of PEG-PLLZ and PEG-PLLZ-PLLeu, respectively. In brief, PEG-PLLZ or PEG-PLLZ-PLLeu was dissolved in certain volume of trifluoroacetic acid (5 wt %) and 4 equiv of a 33 wt % solution of HBr in HAc with respect to the benzyloxycarbonyl (Z) groups was added and stirred at 0 °C for 2 h under nitrogen. Next, the reaction mixture was precipitated with excessive diethyl ether to generate crude product. The crude product was then dissolved in DMF and dialyzed using a dialysis membrane (MWCO 3500 Da) against 0.1 wt % ammonia solution and then distilled water for 48 h. 1 H NMR spectra of the polymers were recorded on a Bruker 400 MHz nuclear magnetic resonance instrument using CF3COOD as the solvents. Fluorescence spectra were recorded on a LS55 luminescence spectrometer (Perkin-Elmer). Micelle Formation. Micelles were prepared by directly dissolving PEG-PLL-PLLeu copolymers in aqueous media at a concentration of 1 mg/mL with magnetic stirring overnight, followed by sonication for 1 h. The secondary structure of the polypeptide chain within the micelles in water was characterized by circular dichroism (CD) spectra using JASCO-80 at 25 °C. The particle size and zeta potential of micelles produced by PEG-PLL-PLLeu copolymer in water were measured by Nano-ZS ZEN3600 (Malvern Instruments) at 25 °C. The morphologies of micelles (1:15:20) were observed by TEM using a JEM-100CXII microscope operating at an acceleration voltage of 100 kV. The TEM samples were prepared by dropping the sample solution onto a copper grid, which had been precoated with a layer of Formvar film and then stained by 0.2% (w/v) phosphotungstic acid solution. The CMC of micelles self-assembled from PEG-PLL-PLLeu copolymers was determined according to the literature using pyrene as a hydrophobic fluorescent probe. Preparation and Characterization of Micelle/DNA Complexes. In the present study, plasmids pEGFP-N1 DNA (4733 bp, Invitrogen) was extracted from E. coli DH5α from same experiment using QIAfilter plasmid purification Giga Kit and then stored at −20 °C. The plasmid DNA was aliquoted into a small volume to avoid repeated freezing and thawing to maintain the quality of DNA. To prepare micelle/DNA complexes, we suspended the plasmid DNA in water at 1 mg/mL. The micelle/DNA complexes at varied N/P ratios were prepared by adding different micelles into plasmids. The mixture was gently vortexed for 5 s and incubated at 37 °C for 30 min to allow the formation of micelle/DNA complexes. In some experiments, PEGPLL copolymers with varied molecular weight were gently mixed with plasmid DNA to form DNA nanoparticles. Particle size and zeta potential of DNA complexes in water were measured by Nano-ZS ZEN3600 (Malvern Instruments) at 25 °C. Agarose Gel Electrophoresis. To assess DNA binding ability of PEG-PLL-PLLeu micelles and PEG-PLL copolymers, we prepared DNA complexes with varied N/P ratios from 0.25 to 4 as previously described. The DNA complexes (containing 0.2 μg DNA per sample) were loaded into a 0.7% (w/v) agarose gel containing Gel Red (Biotium) with Tris-acetate-EDTA (TAE, pH 8.5) running buffer at 80 V for 80 min and then visualized using a Vilber Lourmat imaging system (WealTec, Sparks, NV). In Vitro Cytotoxicity of PEG-PLL-PLLeu Micelles and PEGPLL Copolymers. The cytotoxic effect of different PEG-PLL-PLLeu micelles and PEG-PLL peptides was evaluated using MTT (3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, purchased from Sigma) assay. In brief, 293T cells were seeded in a 96-well plate at 1 × 104 cells/well in 100 μL of DMEM medium with 10% FBS and cultured with 5−80 μg/mL of different micelles or PEG-PLL at 37 °C for 48 h. Four hours before the experiment was stopped, half of culture medium was replaced with MTT solution. At the end of the experiment, supernatants were gently removed, and 100 μL of DMSO was added to each well to dissolve formazan crystals. The absorbance was measured using a Synergy 4 microplate reader (BioTek, Winooski, VT) at a wavelength of 570 nm. The cell viability index was calculated by following formula: the viability (%) = (ODexp − ODblank)/(ODcontrol − ODblank) × 100%, and four replicates were analyzed for each sample.

structure of synthetic polypeptides are more similar to natural proteins. Moreover, the synthetic polypeptides have stable secondary structures in physiological solutions, which would contribute to the high stability of delivery systems. In the present study, we synthesized a series of cationic micelles based on hybrid polypeptide copolymers poly(ethylene glycol)-bpoly(L-lysine)-b-poly(L-leucine) (PEG-PLL-PLLeu) as effective gene vectors. The physiochemical properties and the stability of self-assembled PEG-PLL-PLLeu micelles were assessed, and the gene transfection efficiency of different micelles was investigated both in vitro and in vivo. It is anticipated that the transfection efficiency and biocompatibility of the micelles can be simultaneously improved by varying the length of PLL and PLLeu segments in polypeptides.

■

EXPERIMENTAL PROCEDURES

Materials. O-(2-Aminoethyl)-O′-(2-methyl) polyethylene glycol (PEG-NH2, Mw = 2000) and branched polyethyleneimine (MW = 25 000, PEI25k) was purchased from Sigma-Aldrich (Natick, MA). LLeucine (LLeu) and ε-benzyloxycarbonyl-L-lysine (LLZ) were purchased from GL Biochem (Shanghai, China) and recrystallized from ethyl acetate three times. Triphosgene was purchased from J&K Scientific and recrystallized from diethyl ether before use. NCarboxyanhydride (NCA) of ε-benzyloxycarbonyl-L-lysine (LLZNCA) and N-carboxyanhydride of L-leucine (LLeu-NCA) were prepared according to the method of Daly and Poché.17 Hydrogen bromide 33 wt % solution in glacial acetic acid was purchased from ACROS Organics. Tetrahydrofuran (THF) and n-hexane were provided by Shanghai Chemical Reagent China and dried with sodium before use. N,N′-Dimethylformamide (DMF) and dimethylsulfoxide were provided by J&K Scientific and distilled under reduced pressure before use. All other solvents were of analytical grade and used without further purification. QIAfilter plasmid purification Giga Kit (5) was purchased from Qiagen (Hilden, Germany). GelRedTM was purchased from Biotium (Hayward, CA). HEK293T cell line (human embryonic kidney cells) was from the American Type Culture Collection (ATCC, Rockville, MD) and cultured in Dulbecco’s modified Eagle medium (DMEM, Thermoscientific, Waltham, MA) supplemented with 10% (v/v) fetal bovine serum (FBS) at 37 °C, 5% CO2. All animals were purchased from Guangdong Province Laboratory Animal Center (Guangzhou, China) and maintained in the institutional animal care facility. All animal protocols were approved by Institutional Animal Care and Usage Committee of Shenzhen Institutes of Advanced Technology. Synthesis and Characterization of PEG-PLL and PEG-PLLPLLeu Copolymers. First, the PEG-PLLZ copolymers were synthesized by ring-opening polymerization of LLZ-NCA using PEG-NH2 as initiator. In brief, given amounts of PEG-NH2 and LLZ-NCA were dissolved in dried DMF (10 wt %) and stirred under a N2 atmosphere at 40 °C for 48 h. Then, the product mixture was precipitated with an excess diethyl ether under vigorous stirring to give a white solid (PEG-PLLZ). The product was purified by repeated precipitation in diethyl ether and dried in a vacuum. By varying the feed molar ratio of PEG-NH2 to LLZ-NCA (1:15, 1:20, 1:30), three PEG-PLLZ copolymers (PEG-PLLZ15, PEG-PLLZ20, PEG-PLLZ30) with different molecular weight were obtained. The PEG-PLLZ-PLLeu copolymers were synthesized by further ring-opening polymerization of LLeu-NCA initiated by PEG-PLLZ. In brief, given amounts of PEG-PLLZ and LLeu-NCA were dissolved in dried DMF (10 wt %) and stirred under a N2 atmosphere at 40 °C for 48 h. The product (PEG-PLLZ-PLLeu) was precipitated by diethyl ether and purified by repeated precipitation in diethyl ether and dried in a vacuum. Three different PEG-PLLZ copolymers (PEG-PLLZ15, PEG-PLLZ20, PEG-PLLZ30) were used as initiator for different reactions. By varying the feed molar ratio of PEG-PLLZ to LLeuNCA (1:20, 1:40), six copolymers with different molecular weight, including PEG-PLL 15 -PLLeu 20 (1:15:20), PEG-PLL 15 -PLLeu 40 (1:15:40), PEG-PLL 20 -PLLeu 20 (1:20:20), PEG-PLL 20 -PLLeu 40 B

dx.doi.org/10.1021/bm3012538 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

Scheme 1. Synthesis of PEG-PLL-PLLeu Copolymers and the Formation of Cationic Micelles for DNA Delivery

Uptake of Micelle/DNA Complexes by 293T Cells in Vitro. Plasmid DNA was labeled by a Label IT nucleic acid labeling kit (Mirus, Madison, WI) according to the manufacturer’s protocol. FITC-labeled plasmid DNA was then gently mixed with different PEG-PLL-PLLeu micelles or PEG-PLL peptides to generate DNA complexes (N/P = 20). 293T cells were seeded into a 24-well plate (2 × 105 cells/well) and cultured with different FITC-DNA complexes in DMEM medium containing 5% FBS at 37 °C for 4 h. Naked FITCDNA and PEI25k/FITC-DNA (N/P = 10) complexes were included as negative and positive controls, respectively. The uptake of FITC-DNA was measured using a Beckman Coulter Quanta SC Flow cytometer (Beckman Coulter, Brea, CA), and 5000−10 000 cells were counted for each sample. In Vitro Gene Transfection in 293T Cells. 293T cells (7 × 104 cells/well) were transfected with 0.8 μg of naked plasmid pEGFP-N1 DNA, micelle/DNA complexes (N/P = 20), PEG-PLL/DNA complexes (N/P = 20) or PEI25k/DNA complexes (N/P = 10) in DMEM containing 5% FBS. After 24 h of incubation, the transfection medium was replaced with DMEM containing 10% FBS, and cultured at 37 °C for another 24 h. The expression of EGFP in 293T cells was determined by Beckman Coulter Quanta SC cytometer, and the percentage of EGFP+ cells represented the transfection efficiencies. Intracellular Localization of DNA in 293T Cells. 293T cells were seeded into a chambered coverglass (Thermo Scientific) and cultured with 0.3 μg micelle/FITC-plasmid DNA complexes (N/P = 20) in DMEM containing 5% FBS. After 4 h of incubation, medium was removed and the cells were labeled with 50 nM Lyso Traker Red DND-99 (Invitrogen) for 20 min to visualize late endosomes and lysosomes. Cells were then rinsed twice with PBS, and the images were recorded using confocal microscope (Leica). The excitation/emission wavelengths were 490/520 nm for FITC and 577/590 nm for Lyso Tracker. Gene Transfection in Vivo. Balb/c female mice (4−6 weeks old) were randomly divided into nine groups (5 mice per group) and injected with 5 μg of naked plasmid EGFP DNA, DNA complexes (N/ P = 20), or PEI25000/pEGFP (N/P = 10) at posterior tibialis muscles. The muscles around injection sites were removed 5 days post injection and snap frozen in OCT medium. Frozen sections of 12 μm thick were

prepared using a CM1950 cryomicrotome (Leica), and the fluorescence images of tissue sections were recorded using confocal laser scanning microscope (CLSM, Leica). To semiquantify the expression of EGFP genes in muscular tissues, we measured the fluorescence intensity of each field were measured using Image Pro software (Media Cybernetics, Bethesda, MD), and a minimum of 30 pictures were measured for each group. Statistical Analysis. Data are reported as mean ± SE. The differences among groups were determined using one-way ANOVA analysis, followed by Tukey’s post test (Graphpad Prism, GraphPad Software, La Jolla, CA).

■

RESULTS AND DISCUSSION Synthesis of PEG-PLL-PLLeu Copolymers. As shown in Scheme 1, the PEG-PLL-PLLeu copolymer was synthesized through three steps. The first step was to prepare the diblock copolymer PEG-PLLZ by ring-opening polymerization of LLZNCA using mPEG-NH2 as initiator. Then, the triblock copolymer PEG-PLLZ-PLLeu was synthesized by further ring-opening polymerization of LLeu-NCA using aminoterminated PEG-PLLZ as a macromolecular initiator. The amphiphilic PEG-PLL-PLLeu triblock copolymers were obtained after the deprotection of PEG-PLLZ-PLLeu by HBr/HAc in TFA solution. Meanwhile, diblock copolymer PEG-PLL was obtained after the deprotection of the PEGPLLZ using the same method. By varying the feed molar ratio of LLZ-NCA and LLeu-NCA to mPEG-NH2, six copolymers with different molecular weight, including PEG-PLL15-PLLeu20 (1:15:20), PEG-PLL15-PLLeu40 (1:15:40), PEG-PLL20-PLLeu20 (1:20:20), PEG-PLL20-PLLeu40 (1:20:40), PEG-PLL30-PLLeu20 (1:30:20), and PEG-PLL30-PLLe40 (1:30:40) were synthesized. The structure of the polymers was characterized using 1H NMR and FT-IR spectroscopy. The 1H NMR spectra of PEGPLLZ-PLLeu and PEG-PLL-PLLeu in CF3COOD are shown in Figure 1A,B, respectively. As shown in Figure 1A, the peaks C

dx.doi.org/10.1021/bm3012538 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

Figure 1. Characterizations of the polymers. (A) 1H NMR spectrum of PEG-PLLZ-PLL copolymers. (B) 1H NMR spectrum of PEG-PLL-PLLeu copolymers. (C) FT-IR spectrum of PEG-PLLZ-PLLeu and PEG-PLL-PLLeu copolymers. (D) Circular dichroism spectrum of PEG-PLL-PLLeu micelles in aqueous solution.

around 3.8 ppm (b) are attributable to the protons (−CH2CH2O) in the PEG chain, the peaks around 7.3 ppm (p) are attributable to the protons (−C6H5) in PLLZ, and the peaks present between 0.8 and 0.9 ppm (j) are attributable to the protons (−CH3) in PLLeu. The polymerization degrees of PLLZ and PLLeu were determined by calculating the ratio of peak areas of p to b and j to b, respectively. Compared with Figure 1A, the peaks p and q belonging to the group of Z are disappeared in Figure 1B, indicating the complete deprotection of PEG-PLLZ-PLLeu. Other peaks are also illustrated in Figure 1A,B. The compositions of six PEG-PLL-PLLeu triblock polymers are listed in Table 1, which indicates that the molecular weight of synthetic PEG-PLL-PLLeu is almost consistent with our designation. The results of the FT-IR spectra in Figure 1C furthermore confirmed the successful

synthesis of the PEG-PLL-PLLeu copolymers. The peaks at 1650 and1550 cm−1 represent the amide I bond and amide II bond in the polypeptide, respectively. The peaks at 1105 cm−1 can be attributable to the C−O−C stretching vibration in the PEG segment. Additionally, the peak at 1710 cm−1 attributed to the CO stretching vibration of Z in the spectra of PEGPLLZ-PLLeu is completely disappeared in the spectra of PEGPLLZ-PLLeu, which confirmed the complete deprotection of PEG-PLLZ-PLLeu. Physiochemical Properties of Self-Assembled PEGPLL-PLLeu Micelles. PEG-PLL-PLLeu triblock copolymers were dissolved in water and simultaneously formed selfassembled micelles. The hydrodynamic particle sizes of these micelles were well-tunable ranging from 43.6 to 89.2 nm, which was strongly correlated with the length of PLL and PLLeu segments (Figure 2A). However, their surface charges did not significantly change. As shown in Figure 2B, the zeta potential of six PEG-PLL-PLLeu micelles was all above 55 mV, indicating their capability of DNA condensation. The TEM image further demonstrated that the PEG-PLL-PLLeu micelles could be welldispersed with a regular spherical shape (Figure 2C). The stability of micelles was evaluated by monitoring the particle size at room temperature for 6 weeks. The results showed that the sizes of all six micelles in water did not significantly change within 6 weeks, indicating their excellent stability (Figure 2D). The stability of PEG-PLL-PLLeu micelles can be attributable to the high charge density on their surface and the presence of PEG layer. Furthermore, the secondary molecular structures within proteins or peptides, including αhelices, β-sheets, and random coils, have been shown to contribute to the improved stability. 18 Therefore, we characterized the secondary structure of the PLLeu chain within the PEG-PLL-PLLeu micelles using the CD spectroscopy. The results in Figure 1D demonstrate a positive peak at

Table 1. Feed Composition and Final Composition of PEGPLL-PLLeu As Well As the CMC Value of the Micelles SelfAssembled from These Polypeptide triblock polypeptide

PEG/Lys-NCA/LeuNCA feed molar ratio

PEG/Lysa/Leub molar ratioc in copolymer

CMC (mg/L)c

PEG-PLL15PLLeu20 PEG-PLL15PLLeu40 PEG-PLL20PLLeu20 PEG-PLL20PLLeu40 PEG-PLL30PLLeu20 PEG-PLL30PLLeu40

1:15:20

1:16:21

113.5

1:15:40

1:16:44

44.6

1:20:20

1:21:21

177.8

1:20:40

1:21:44

63.1

1:30:20

1:30:22

398.5

1:30:40

1:30:44

79.4

a

Lysine unit. bLeucine unit. cCalculated from 1H NMR spectra D

dx.doi.org/10.1021/bm3012538 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

Figure 2. Characterizations of PEG-PLL-PLLeu micelles. (A) Particle size of PEG-PLL-PLLeu micelles in water. (B) Zeta potential of different PEGPLL-PLLeu micelles in water. (C) Transmission electron microscopy (TEM) image of PEG-PLL-PLLeu (1:15:20) micelles. Scale bar = 50 nm. (D) Stability of PEG-PLL-PLLeu micelles at 1−6 weeks post synthesis.

193 nm and two negative peaks at 208 and 220 nm, all of which are characteristics of α-helical conformation. These data suggest that the structure of PLLeu segments within the micelles is predominantly a rigid α-helical conformation, which could also consequently enhance the stability of micelles.19 Furthermore, we measured the critical micelle concentration (CMC) of different micelles, which represents their thermodynamic stability. The CMC value of PEG-PLL-PLLeu micelles was primarily related to the length of hydrophobic blocks (PLLeu) and mass ratio of hydrophilic blocks (PEG-PLL) to hydrophobic blocks (PLLeu). With the length of PLLeu segments increased, the CMC value decreased (Table 1), suggesting that the induction of hydrophobic blocks is critical for micelle stability. These data are consistent with previous results that the induction of strong hydrophobic interactions into the core of micelles could prevent polymeric micelles from premature dissociation under physiological condition.20 The increased mass ratio of PEG-PLL to PLLeu also slightly increased the CMC value. Therefore, an appropriate mass ratio of hydrophilic blocks to hydrophobic blocks is critical for micelle stability. Cytotoxicity of PEG-PLL-PLLeu Micelles. The cytotoxicity of PEG-PLL-PLLeu micelles and PEG-PLL copolymers was evaluated in 293T cell using MTT assay. As shown in Figure 3, 40−80 μg/mL of PEG-PLL copolymers with different length of PLL segments (PLL = 15, 20, or 30) caused over 70− 100% of cell death after 48 h of incubation. In contrast, the presence of PEG-PLL-PLLeu micelles with short PLL blocks (PLL = 15) did not significantly affect the viability of 293T cells (Figure 3A). Similarly, PEG-PLL-PLLeu micelles with the length of PLL segments at 20 showed significantly less cytotoxicity than PEG-PLL (1:20) copolymers at the concentration of 40−80 μg/mL (P < 0.05, Figure 3B). Hence, the introduction of hydrophobic PLLeu blocks remarkably improved the biocompatibility of PEG-PLL cationic copolymer. When the length of PLL segments reached 30, the cytotoxicity of PEG-PLL-PLLeu micelles was dramatically increased, consistent with previous reports that the size of

Figure 3. Cell viability of 293T cells treated with PEG-PLL-PLLeu micelles or PEG-PLL in vitro. The 293T cells were cultured with 5 to 80 μg/mL of PEG-PLL-PLLeu micelles or PEG-PLL with different length of PLL segments for 48 h, and the cell viability was measured by MTT assay. (A) PLL = 15, (B) PLL = 20, and (C) PLL = 30. Data are shown as mean ± SE (n = 3−5).

PLL blocks directly contributed to the cytotoxicity of copolymers.21 PEG-PLL-PLLeu (40 μg/mL, 1:30:40) micelles caused less cell death than the same dose of PEG-PLL-PLLeu (1:30:20) micelles or PEG-PLL (1:30) did (p < 0.05, Figure 3C), which provided another evidence that hydrophobic PLLeu blocks could improve the biocompatibility of PEG-PLL cationic E

dx.doi.org/10.1021/bm3012538 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

copolymer. Overall, the toxicity of PEG-PLL-PLLeu micelle was strongly associated with the length of PLL segments. However, the introduction of a hydrophobic PLLeu block remarkably reduced the cytotoxicity of cationic copolymers. Moreover, the cytotoxic effect of polypeptide micelles could be further reduced by increasing the length of PLLeu segments, especially in those micelles with relatively long PLL segments. Hence, introduction of a hydrophobic PLLeu segment with an appropriate length is an effective strategy to improve the biocompatibility of cationic micelles. Characterization of DNA/Micelle Complexes. For efficient transfection, it is crucial that DNA is compacted into a stable particle to protect the DNA from degradation and allow for efficient uptake. In the present study, PEG-PLLPLLeu amphiphilic copolymers were designed to be able to self-assemble into core−shell structure micelles, which allowed the association of negatively charged DNA with positively charged outer surface. Therefore, their capability of DNA binding must be investigated. The DNA binding efficiency of different micelles was determined using the agarose gel electrophoresis performed at different N/P ratios from 0.25 to 4 (Figure 4). For comparison, DNA was also formulated

Figure 5. Characterizations of DNA complexes. The zeta potential (A) and the size (B) of different DNA complexes with PEG-PLL-PLLeu micelles or PEG-PLL copolymers at the N/P ratio of 20.

interaction and the hydrophobic interaction promote DNA condensation, which would consequently allow the formation of micelles with compact size and improved stability.13,22 The particle size of vector/DNA complexes is one of the key factors impacting DNA internalization and downstream gene transfection.6 Rejman et al.23 found that the nanoparticles with the particle size