Nanoparticles facilitate gene delivery to microorganisms via an electrospray process

Journal of Microbiological Methods 84 (2011) 228–233

Contents lists available at ScienceDirect

Journal of Microbiological Methods j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j m i c m e t h

Nanoparticles facilitate gene delivery to microorganisms via an electrospray process Yi-Hsuan Lee a, Bing Wu a, Wei-Qin Zhuang b, Da-Ren Chen a, Yinjie J. Tang a,⁎ a b

Department of Energy, Environmental, and Chemical Engineering, Washington University, St. Louis, MO 63130, USA Department of Civil and Environmental Engineering, University of California, Berkeley, CA 94720, USA

a r t i c l e

i n f o

Article history: Received 21 June 2010 Received in revised form 27 November 2010 Accepted 27 November 2010 Available online 7 December 2010 Keywords: CaCl2 Gene delivery Gold NPs Non-competent cells Plasmid

a b s t r a c t In this study, we developed a technique for delivering genes to microorganisms via electrospray of gold nanoparticles. During the electrospray process, charged monodisperse nano-droplets (a mixture of pET30aGFP plasmid and nano-sized gold particles) were accelerated and deposited on a thin layer of non-competent Escherichia coli cells. Via antibiotic selection, transformed cells containing green fluorescent protein appeared on the agar plates. PCR amplification and restriction enzyme analysis further confirmed that pET30a-GFP plasmid had successfully been delivered into the non-competent E. coli cells. The transformation efficiencies were optimized under different electrospray conditions. Among several electrospray buffer solutions, CaCl2 (0.01 M) was found to be the best for gene delivery. Furthermore, gold nanoparticles (NPs, 50 nm diameter) significantly improved plasmid transformation efficiency by 5– 7 fold (up to 2 × 106 CFU/μg plasmid) compared with that obtained using naked plasmid. Electronic microscopy images and gel electrophoresis showed that the morphology of plasmids remained unchanged during the electrospray process, but cellular membrane integrity was reduced after being electrosprayed with gold NPs and CaCl2 buffer solutions. This gene delivery method has the potential to work for many other microorganisms. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Gene delivery, the process of introducing foreign genes into living cells, is an important technique in fields such as metabolic engineering and gene therapy. Previously developed viral-based and non-virus gene delivery techniques, including chemical/biological approaches (either using lipids, cationic polymers, or virus as vectors, or using conjugation) and physical approaches (heat shock, electroporation, gene gun, microinjection, and sonoporation), have been applied in bacterial, mammalian, and plant cells. The aforementioned gene delivery methods have also been modified for transformation of difficult-to-transform bacteria. For example, by electrotransformation with special current oscillation programs, plasmid DNA can be delivered into thermophilic anaerobes such as Clostridium thermocellum, Clostridium acetobutylicum, and Thermoanaerobacterium saccharolyticum (Tyurin et al., 2004, 2005; Peng et al., 2006). However, these approaches are successful only in a few bacterial species because of the low efficiency of transformation, the complicated operation protocols, the severe damage to cells, and the high cost of complex devices (Henry, 2001; Okubo et al., 2008; Pathak et al., 2009).

⁎ Corresponding author. Tel.: + 1 314 935 3441. E-mail address: [email protected] (Y.J. Tang). 0167-7012/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2010.11.022

In this study, an electrospray technique, formally called “electrohydrodynamic atomization”, has been applied for gene delivery. Electrospray employs an electric field to disperse and accelerate liquid droplets or fine particles. An important advantage of electrospray is that the bioactive materials (i.e., protein or plasmid DNA) do not lose their activity during the process. Previous proof-of-concept studies had shown the electrospray technique could deliver DNA materials into cells in two steps: 1) the plasmid was charged and accelerated by electric force during the electrospray process; and 2) the plasmid penetrated the cell membrane, thus delivered targeted genes into cells. This novel transformation method does not require the preparation of competent cells, which is the crucial and timeconsuming step in traditional DNA transformation methods (Chen et al., 2000; Davies et al., 2005; Okubo et al., 2008). Previous studies on aerosol delivery of genetic materials were qualitative rather than quantitative, and the technique was mostly targeted to transfection of mammalian cells (which lack a cell wall) and required nonviral vectors (e.g., polyethylenimines) to carry DNA (Wu et al., 2009). This study focused on microbial transformation of an industrial microbial host, Escherichia coli, by examining the key factors for controlling gene delivery efficiency. It has been reported that non-toxic nanoparticles (e.g., gold NPs, silica NPs, and carbon nanotubes) can facilitate the macromolecules' entry into host cells (Jen et al., 2004; Rojas-Chapana et al., 2005; Galbraith, 2007). In particular, gold nanoparticles (Au NPs) are promising vehicles for gene delivery because these

Y.-H. Lee et al. / Journal of Microbiological Methods 84 (2011) 228–233

particles are readily conjugated with biomolecules at a high packing density (Mirkin et al., 1996). Meanwhile, NPs (with plasmid) can increase the momentum of plasmid during electrospray and improve their penetration through cell wall. Using a newly designed electrospray device, we quantitatively tested several parameters, including NP sizes, cell growth stages, and electrospray buffer solutions, to obtain the optimal gene delivery operation conditions for E. coli transformation. Since the voltage in the electrospray device can be adjusted to achieve optimal momentum for the complex of Au NPs and plasmid, such gene delivery approach could be potentially used to bombard hard-to-transform microbes for achieving efficient transformation.

2. Materials and methods 2.1. Electrospray system description Fig. 1 shows a schematic diagram of an electrospray system for transferring plasmid DNA into bacterial cells. The system was composed of an electrospray head and a deposition stage, arranged in a point-to-plate configuration. The spray head consisted of a single capillary (inside diameter: 0.38 mm, length: 10 cm) in a precision syringe, which was connected with a programmable syringe pump. Plasmid DNA or a mixture of plasmid DNA with NPs in the buffer solution was fed into the spray capillary. The membrane deposited with bacterial cells was placed on the deposition stage. A voltage (~7 kV) of positive polarity was applied to the spray head, and

229

electrically grounded at the deposition stage (2 cm below the spray needle tip). The charged plasmid DNA or plasmid DNA-NP complex, with a flow rate of 1.5 μl/min, was accelerated and uniformly electrosprayed onto the cells for transformation. The total current that ran through the membrane was limited to less than 0.3 μA. Before all experiments, the electrospray device was carefully cleaned and sterilized using 70% ethanol.

2.2. Microorganism cultivation and plasmid DNA preparation E. coli BL 21 (DE3) strain was stored at −80 °C prior to use. E. coli was grown in a Luria-Bertani (LB) medium at 37 °C at a shaking speed of 200 rpm overnight and then was inoculated into fresh LB medium with an inoculation rate of 5% as subcultures under the same culture conditions. The gene encoding green fluorescence protein (GFP) was cloned in the NcoI and NotI restriction sites in a kanamycin-resistant pET30a vector (Novagen, USA) to create an expression vector pET30aGFP. The plasmid DNA was extracted from its host using a PureYield™ Plasmid Miniprep Kit (Promega, USA), and the plasmid concentration was measured using a spectrophotometer (Nanodrop, Thermo Scientific, USA). The genomic DNA was obtained using a Wizard® Genomic DNA Purification Kit (Promega, USA). The T7 promoter primer (TAATACGACTCACTATAGGG) and T7 terminator primer (GCTAGTTATTGCTCAGCGG) were employed to amplify a piece of gene (containing ~1047 bp) in the pET30a-GFP plasmid, and the universal 16S rRNA forward primer (AGAGTTTGATCCTGGCTCAG)

Fig. 1. Schematic diagram of an electrospray system for gene transformation (Wu et al., 2010).

230

Y.-H. Lee et al. / Journal of Microbiological Methods 84 (2011) 228–233

and 16S rRNA reverse primer (ACGGCTACCTTGTTACGACTT) were employed to amplify the gene in the genomic DNA, by the polymerase chain reaction (PCR) using a Polymerase kit (Promega, USA). The PCR reactions were conducted with the following cycle conditions(~30 cycles): 1 min at 94 °C, 30 s at 58 °C and 2 min at 72 °C; finally, a 10 min extension process was performed at 72 °C. To examine the plasmid integrity, the pET30a-GFP plasmid DNA was digested with restriction enzymes (XhoI and Xbal, Promega, USA) in the buffer provided by the manufacturer. The final PCR products and digested products were observed directly on agarose gels after electrophoresis. 2.3. Gene transformation condition A thin layer of E. coli cells (~108) was deposited on a piece of sterile PVDF membrane (1 cm2, 0.2 μm, Millipore, US), which was then placed on a grounded stage, as illustrated in Fig. 1. After the electrospray process (~2 min), the membrane with the cells was quickly transferred into a 50 mL culture tube containing 5 mL of LB broth and vortexed. After incubation at 200 rpm (37 °C) for 3–5 h to recover cell growth, 100 μL of culture was plated onto kanamycincontaining LB agar plates for selection. The transformation efficiency was quantified based on the viable GFP-positive CFU (colony-forming unit) on the agar plates and the amount of plasmid sprayed (~5 ng plasmid in each test, n = 3). 2.4. Scanning electron microscope (SEM) and transmission electron microscope (TEM) observation After electrospraying the plasmid DNA, the cells on the membrane were fixed using 2% glutaraldehyde for 2 h. Then, the samples were washed in 0.10 M sodium cacodylate buffer for 20 min three times. Then the samples were dehydrated in a series of 10 min washes in 50%, 70%, 85% and 95% ethanol and further were dried with a freezedrier equipment (Labconco, USA). The samples were gold-coated using a sputter gun (SPI supplies, USA). Photographs of the sample structure were observed and taken with a SEM (FEI, UK). Additionally, a TEM was used to image the interaction of cells and NPs. After electrospray, cells were fixed in 4% paraformaldehyde/2.5% glutaraldehyde in a 0.1 M cacodylate buffer (pH 7.2) overnight. Samples were treated with osmium tetroxide and were dehydrated using ethanol before they were embedded into a Pelco Eponate 12 resin (Ted Pella Inc., Redding, USA). Sections of the samples were cut on an Ultramicrotome (Leica, Germany), inspected in the TEM (H7500, Hitachi, Japan) at 80 kV using the HR mode, and photographed by a digital camera (Hamamatsu, Japan). 2.5. Membrane integrity determination The integrity of the cytoplasmic membrane was determined using a BacLight Live/Dead Kit (Molecular Probes, Invitrogen, USA). In this kit, the SYTO 9 green-fluorescent nucleic acid stain generally labels all bacteria (those with intact membrane and damaged membrane) and propidium iodide penetrates only the bacteria with damaged membranes, causing a reduction in the fluorescence when both dyes are present. Cells were harvested by centrifugation and resuspended in 0.85% NaCl to adjust the cell density (OD600 ~ 0.04– 0.05). Then the sample (100 μl) was transferred onto a 96-well flatbottom microplate and a 100 μl pre-mixed dye solution (50 μl of 10 μM SYTO 9 and 50 μl of 60 μM propidium iodide) was added. The samples were incubated in dark for 15 min at ambient temperature. Fluorescence was measured with a fluorescence microplate reader (BioTek, USA); the excitation wavelength was set at 485 nm, and the emission wavelength was set at 490–700 nm. The ratio of fluorescence intensity at 530 nm (living cells can be detected) and that at 630 nm (membrane-damaged cells can be detected) was calculated.

3. Results and discussion 3.1. The optimization of transformation conditions for E. coli In this study, the cell growth stage was considered as a key parameter influencing transformation efficiency (Fig. 2a). Gene transformation efficiency was low for cells in the early growth stage (OD600 b 0.4). The maximum GFP-positive colony count (a transformation efficiency of 2 × 106 CFU/μg plasmid) was achieved for exponential-growth cells (OD600 = 1–2) when both the cell division rate and total population were high. The transformation efficiency decreased for cells in the stationary phase. The electrospray buffer was another important factor in maintaining electrospray stability. Notably, the presence of saline may alter the ionic conditions surrounding the plasmid and further influence transformation efficiency (Bloomfield 1996). For plasmid in buffer solutions such as water, NaCl (0.01 M), Na2HPO4 (0.01 M) or MgCl2 (0.01 M), no GFPpositive colony was achieved using late-exponential-phase cells (Fig. 2b). Only a few GFP-positive colonies appeared after electrospray of naked plasmid in CaCl2 (0.01 M) buffer solution. However, after electrospray of gold NPs (50 nm) with the plasmid/CaCl2 solution, we found that the transformation efficiency was significantly enhanced, by 5–7 fold (Fig. 2a and b). Further increasing the CaCl2 concentration to 0.1 M did not increase the transformation efficiency and also decreased the NP stability in the spraying solution. The sprayed NP size and amount were also considered to be associated with the transformation efficiency. With an increase of particle size from 20 nm to 50 nm, the transformation efficiency improved twofold (P-value = 0.007) (Fig. 2c). Further increasing particle size from 50 nm to 100 nm did not significantly promote the transformation efficiency (P-value = 0.20). Also, increase of the sprayed NP (50 nm) amount enhanced the transformation efficiency. However, spraying too many NPs (N18 × 105 per test) might cause irreversible cellular damage and reduce gene transformation efficiency (Fig. 2d). 3.2. The proposed mechanism of gene delivery by electrospray PCR amplification and restriction enzyme digestion of plasmid extracted from E. coli transformants confirmed that the plasmid was successfully introduced into E. coli cells (Fig. 3). Gel electrophoresis analysis indicated that the treatment of plasmid with gold NPs or/and CaCl2 (0.01 M) buffer solution did not change the positions of the DNA bands in the agarose gel, which suggests that the presence of gold NPs or/and CaCl2 (0.01 M) buffer solution did not alter the conformation of plasmid DNAs (Fig. S1). A membrane integrity experiment revealed that the cellular membrane permeability increased without highly disruptive damage to the cell membrane structure (i.e., no significant DNA released from cells) after electrospray with NPs (Table 1). In a control experiment, direct mixing of NPs with cells (without electrospray) in a culture tube for ~ 5 h showed no change of cell membrane integrity. Via electrospraying NPs, the cell membrane permeability was enhanced, as suggested by lower F530/F630 ratios. F530/F630 values were 1.7 ± 0.10 and 1.6 ± 0.04 after electrospray with 50 nm and 100 nm NPs respectively, and further increases in NP size did not create additional membrane permeability and thus did not improve transformation. On the other hand, SEM images showed that the cell morphology was influenced by electrospraying plasmid in CaCl2 solution (0.01 M) (Fig. 4b). The CaCl2 solution could have reduced membrane stability and increased membrane permeability (i.e., leading to multiple transient pores) to enhance the transformation efficiency. Meanwhile, gold NPs help temporally create channels through the membrane surface to allow the plasmid into the cells (NPs may or may not enter the cells). Hence, the entrance mechanism of plasmid was thought to be the result of the electrospray buffer solution and gold NPs (Fig. 4c

Y.-H. Lee et al. / Journal of Microbiological Methods 84 (2011) 228–233

2.5

(a)

2.0 1.5

2

1.0 1 0.5

Efficiency (×106)

3

OD600

Efficiency (×106)

2.5

(b)

2.0 1.5 1.0 0.5

0.0

1

3

5

7

0

0.0

Water

Time (hr) 2.0

NaCl NaHPO4 MgCl2 CaCl2 CaCl2 Buffer solution (0.01M) with NPs

2.0

(c)

(d)

1.5

Efficiency (×106)

Efficiency (×106)

231

1.0

0.5

0.0 Control

20

50

100

Particle size (nm)

1.5

1.0

0.5

0.0

0

9

18

45

NP amount (×105 counts)

Fig. 2. Gene transformation efficiency (CFU/μg plasmid) of pET30a-GFP plasmid into E. coli under different conditions. (a) Different growth stages of E. coli; (White bar) Plasmid; (Dark bar) Plasmid with gold NPs (9 × 105); (Square symbol with dot line) cell growth (OD600) curve; (b) Different electrospray buffer solutions (0.01 M); (c) Different gold NP sizes; (d) Different gold NP amounts per test (For Panel a–c, 9 × 105 NPs were employed in each transformation experiment).

and d). However, such temporary cell damage could be repaired after incubation of cells in a rich medium for a few hours (Table 1). 3.3. Significance The electrospray gene delivery technique is similar to a gene gun, but uses electric force instead of high pressure gas or gunpowder. A

recent paper (Okubo et al. 2008) described how charged water droplets from an electrospray device generated a transient hole on cell membranes and facilitated DNA transport into E. coli (K12 strain). The operation required high current (100 μA) to create penetrations of the cell membrane (similar to electroporation). Their method achieved gene transformation efficiency 104–105 (CFU/μg plasmid). In this study, the current through the sample was limited to 0.2 μA. The

Fig. 3. Confirmation of transformation in E. coli by restriction enzyme analysis and PCR amplification. Lane 1 and Lane 5: 100 bp DNA ladder; Lane 2: pET30a-GFP plasmid used for electrospray and digested with XhoI and Xbal (the expected sizes of the band are 904 bp and 5196 bp respectively); Lane 3 and Lane 4: pET30a-GFP plasmid recovered from E. coli transformants and digested with XhoI and Xbal; Lane 6: PCR amplification of pET30a-GFP plasmid used for electrospray using T7 promoter and T7 terminator primers (the expected size of the band is 1047 bp); Lane 7: PCR amplification of pET30a-GFP plasmid rescued from E. coli transformants using T7 promoter and T7 terminator primers; Lane 8: PCR amplification of genomic DNA from wild-type E. coli using T7 promoter and T7 terminator primers (negative control); Lane 9: PCR amplification of genomic DNA from wild-type E. coli using universal 16S rRNA primers (positive control). The similar DNA bands on the agarose gels (Lane 6 and 7 for PCR products; Lane 2, 3 and 4 for enzyme digestion products) confirmed successful transformation in E. coli by the electrospray approach.

232

Y.-H. Lee et al. / Journal of Microbiological Methods 84 (2011) 228–233

Table 1 The effect of electrospray on membrane integrity and intracellular release of E. coli DNA due to cellular structure damage (Pagan and Mackey 2000). E. coli cells

Membrane integrity (F530/F630)

Released DNA (ng/μl)

Before electrospray After electrospray

3.2 ± 0.1 3.2 ± 0.5 (0.99)* 1.4 ± 0.3 (0.001) 1.7 ± 0.1 (0.006) 2.8 ± 0.1 (0.10)

14.7 ± 2.7 18.0 ± 8.4 (0.31) 13.4 ± 1.9 (0.48) 16.7 ± 5.3 (0.36) Not measured

Recovery for ~ 5 h

Water NPs in water Plasmid in CaCl2 (0.01 M) with NPs Plasmid in CaCl2 (0.01 M) with NPs

* The data in brackets are the P-values, which were calculated based on the data before and after electrospray.

Fig. 4. SEM and TEM images of E. coli cells after electrospray. (a) SEM image of cells after electrospraying plasmid in water; (b) SEM image of cells after electrospraying plasmid in CaCl2 buffer solution (0.01 M). (c, d) TEM images of cells after electrospraying plasmid in CaCl2 buffer solution (0.01 M) in the presence of NPs: Cells with 50 nm NPs (Panel c); Cells with plasmid (Panel d). The images presented here were considered representative of the entire sample because the same observations were seen for multiple samples.

Fig. 5. Transformation of pET30a-GFP (a) and pET30a-Cherry (b) into E. coli. The transformed cells formed colorful colonies.

Y.-H. Lee et al. / Journal of Microbiological Methods 84 (2011) 228–233

creation of membrane permeability and consequent transformation resulted from the particle bombardment. The electrospray of NPs could achieve stable gene delivery efficiency (~ 106 CFU/μg plasmid) for two types of plasmids (pET30a-GFP and pET30a-Cherry) in E. coli (Fig. 5). This approach was also applicable to other microbial species (i.e., Saccharomyces cerevisiae, unpublished data). Although electrospray gene delivery still showed relatively low transformation efficiency compared to the heat shock and electroporation (up to 108 CFU/μg plasmid in E. coli) (Dower et al., 1988; Inoue et al., 1990), this method might provide an alternate approach for engineering these difficult-to-transform bacteria. 4. Conclusions In summary, we have demonstrated an electrospray gene delivery technique by spraying a mixture of plasmid and nanoparticles. This transformation method does not require the preparation of competent cells, which is a time-consuming step for the traditional DNA transformation methods. The momentum of nanoparticles (Au NPs) generated by the electronic field can facilitate gene delivery into microbial hosts and significantly improve plasmid transformation efficiency. The transformation efficiency can be optimized by adjusting the nanoparticle size, the cell growth condition, and the suitable buffer solution. Moreover, by controlling voltage in the electrospray process, the momentum of the nanoparticle delivery can be further optimized for transformation of microorganisms with different characteristics. Acknowledgements This study was supported by an NSF Career Grant (MCB0954016). The authors are also grateful to visiting students (Angela Horst, Zheng Zuo, and Peter Colletti) for their kind help with measurements. Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10.1016/j.mimet.2010.11.022.

233

References Bloomfield, V.A., 1996. DNA condensation. Curr. Opin. Struct. Biol. 6, 334–341. Chen, D.R., Wendt, C.H., Pui, D.Y.H., 2000. A novel approach for introducing biomaterials into cells. J. Nanopart. Res. 2, 133–139. Davies, L.A., Hannavy, K., Davies, N., Pirrie, A., Coffee, R.A., Hyde, S.C., Gill, D.R., 2005. Electrohydrodynamic comminution: a novel technique for the aerosolisation of plasmid DNA. Pharm. Res. 22, 1294–1304. Dower, W.J., Miller, J.F., Ragsdale, C.W., 1988. High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res. 16, 6127–6145. Galbraith, D.W., 2007. Nanobiotechnology—silica breaks through in plants. Nat. Nanotechnol. 2, 272–273. Henry, C.M., 2001. Gene delivery—without viruses. Chem. Eng. News 79, 35–41. Inoue, H., Nojima, H., Okayama, H., 1990. High efficiency transformation of Escherichia coli with plasmids. Gene 96, 23–28. Jen, C.P., Chen, Y.H., Fan, C.S., Yeh, C.S., Lin, Y.C., Shieh, D.B., Wu, C.L., Chen, D.H., Chou, C.H., 2004. A nonviral transfection approach in vitro: the design of a gold nanoparticle vector joint with microelectromechanical systems. Langmuir 20, 1369–1374. Mirkin, C.A., Letsinger, R.L., Mucic, R.C., Storhoff, J.J., 1996. A DNA-based method for rationally assembling nanoparticles into macroscopic materials. Nature 382, 607–609. Okubo, Y., Ikemoto, K., Koike, K., Tsutsui, C., Sakata, I., Takei, O., Adachi, A., Sakai, T., 2008. DNA introduction into living cells by water droplet impact with an electrospray process. Angew. Chem. Int. Ed. 47, 1429–1431. Pagan, R., Mackey, B., 2000. Relationship between membrane damage and cell death in pressure-treated Escherichia coli cells: differences between exponential- and stationary-phase cells and variation among strains. Appl. Environ. Microbiol. 66, 2829–2834. Pathak, A., Patnaik, S., Gupta, K.C., 2009. Recent trends in non-viral vector-mediated gene delivery. Biotechnol. J. 4, 1559–1572. Peng, H., Fu, B., Mao, Z., Shao, W., 2006. Electrotransformation of Thermoanaerobacter ethanolicus JW200. Biotechnol. Lett. 28, 1913–1917. Rojas-Chapana, J., Troszczynska, J., Firkowska, I., Morsczeck, C., Giersig, M., 2005. Multiwalled carbon nanotubes for plasmid delivery into Escherichia coli cells. Lab Chip 5, 536–539. Tyurin, M.V., Desai, S.G., Lynd, L.R., 2004. Electrotransformation of Clostridium thermocellum. Appl. Environ. Microbiol. 70, 883–890. Tyurin, M.V., Sullivan, C.R., Lynd, L.R., 2005. Role of spontaneous current oscillations during high-efficiency electrotransformation of thermophilic anaerobes. Appl. Environ. Microbiol. 71, 8069–8076. Wu, B., Wang, Y., Lee, Y.-H., Horst, A., Wang, Z., Chen, D.R., Radhakrishna, S., Tang, Y.J., 2010. Comparative eco-toxicities of Nano-ZnO particles under aquatic and aerosol exposure modes. Environ. Sci. Technol. 44, 1484–1489. Wu, Y., Yu, B., Jackson, A., Zha, W., Lee, L.J., Wyslouzil, B.E., 2009. Coaxial electrohydrodynamic spraying: a novel one-step technique to prepare oligodeoxynucleotide encapsulated lipoplex nanoparticles. Mol. Pharm. 6, 1371–1379.