The authors have read the journal's policy and have the following conflicts: Both JD (non-executive Director and Founder) and JL (Employee) are part of nanoTherics Limited, which is involved in the commercialisation of the oscillating magnetic array and related magnetic nanoparticles. Jenson Lim (JL) is a consultant for and former employee of nanoTherics Limited. Jon Dobson (JD) is a consultant for, shareholder in, and founder and non-executive director of nanoTherics Limited. JD also holds key patents on this technology, which are owned by nanoTherics Limited. nanoTherics Limited markets the magnefect-nanoTM devices used in this study and is also a reseller of nTMagTM magnetic nanoparticle reagents. This does not alter the authors' adherence to the PLOS ONE policies on sharing data and materials.
Conceived and designed the experiments: JL MAC JD. Performed the experiments: JL MAC. Analyzed the data: JL JD. Contributed reagents/materials/analysis tools: JL JD. Wrote the paper: JL.
Current address: GE Healthcare, The Maynard Centre, Cardiff, United Kingdom
Gene delivery technologies to introduce foreign genes into highly differentiated mammalian cells have improved significantly over the last few decades. Relatively new techniques such as magnetic nanoparticle-based gene transfection technology are showing great promise in terms of its high transfection efficiency and wide-ranging research applications. We have developed a novel gene delivery technique, which uses magnetic nanoparticles moving under the influence of an oscillating magnetic array. Herein we successfully introduced short interfering RNA (siRNA) against green fluorescent protein (GFP) or actin into stably-transfected GFP-HeLa cells or wild-type HeLa and rat aortic smooth muscle cells, respectively. This gene silencing technique occurred in a dose- and cell density- dependent manner, as reflected using fluorescence intensity and adhesion assays. Furthermore, using endocytosis inhibitors, we established that these magnetic nanoparticle-nucleic acid complexes, moving across the cell surface under the influence of an oscillating magnet array, enters into the cells via the caveolae-mediated endocytic pathway.
Recent decades have seen the rise of gene delivery technologies to introduce foreign genes into highly differentiated cells like neurons or leukocytes, as such cells are known to be resistant to either accepting or expressing exogenous genes. Such technologies range from the relatively inexpensive lipid-based (e.g Lipofectamine) or non-lipid based (e.g. Fugene) reagents to more costly nucleofection (e.g. Amaxa) or gene gun (e.g. Helios) methods (reviewed in
Short interfering RNA (siRNA) or plasmid DNA is attached to magnetic nanoparticles and incubated with cells in culture (left). An oscillating magnet array below the surface of the cell culture plate pulls the particles into contact with the cell membrane (A) and drags the particles from side-to-side across the cells (B), mechanically stimulating endocytosis (C). Once the particle/nucleic acids complex is endocytosed, proton sponge effects rupture the endosome (D) releasing the nucleic acids (E) which either transcribes the target protein or silences the target genes (F)
Although we, and others, have shown successful transfection with this technology
Remarkable differences were observed using human lung epithelial cells NCI-H292 transfected with a plasmid containing the luciferase reporter gene. A 2 Hz/0.2 mm frequency and amplitude of displacement of the oscillating magnet array showed higher transfection efficiency with little negative impact on cell viability compared with a static magnet system and two commercially available lipid-based reagents
Here we show successful gene silencing of GFP and actin in stably-transfected GFP-HeLa cells and wild-type HeLa cells, respectively using this novel transfection system which outperformed a leading lipid reagent and a static magnet array system. Using endocytosis inhibitors, we also confirm that the route of entry for these nanoparticle-nucleic acid complexes is via the caveolae-mediated endocytic pathway, a process that appears to be enhanced by mechanical stimulation of the cells due to the oscillatory motion of the particle complexes across the cell surface.
Silencer GFP siRNA (siGFP) and the Negative Control (scrambled sequences, SCR) were purchased from Ambion/Invitrogen (Paisley, UK). Stealth siRNA against human Actin (siActin) was purchased from Invitrogen (Paisley, UK). Phosphate buffered saline, 24-well tissue cell culture plates and flasks (Costar) were purchased from Sigma (Dorset, UK). HeLa cells were purchased from ECACC/Sigma (Dorset, UK). Rat Aortic Smooth Muscle cells were a kind gift from Eva Pantazaka/Colin Taylor (University of Cambridge)
Eukaryotic expression vector pEGFP-N1 (CMV promoter driving gene encoding green fluorescence) was purchased from Clontech (Mountain View, USA). Plasmid DNA was prepared using the Qiagen EndoFree Plasmid Purification kit (Qiagen, Crawley, UK), and maintained in endonuclease-free water (Sigma, Dorset, UK) at −80°C.
90000 HeLa cells per well were seeded onto a 24-well tissue culture plate and left overnight in a 37°C, 5% CO2 incubator. 0.6 µg of pEGFP-N1 (Clontech, UK) was complexed with 0.6 µl of nTMag (nanoTherics, Stoke-On-Trent, UK) in serum-free MEM for 15 min before transferring to the wells containing HeLa cells. nTMag is Fe3O4 dispersed in a polyethylenimine-HCl matrix; zeta potential: +23.4 mV; particle size distribution: 1.8 (polydisperse index). Cell were transfected using the magnefect-nano II system (nanoTherics, UK) before transferring the 24-well plate to the incubator for 48 hr. Fresh medium was replaced containing 0.5 mg/ml G418 (Biosera, UK). After 14 days, brightest GFP-expressing colonies of HeLa cells were selected visually and transferred to a 96-well plate. Another round of selection was conducted after a further 20 days, with cells then transferred to a 6-well plate. Finally GFP-HeLa cells were expanded to a 25 cm2 flask and cell sorted (Flow Cytometry Core Service, University of Sheffield) to enrich the population of GFP-expressing HeLa cells. Approximately 1 million GFP-HeLa cells were frozen down (liquid nitrogen) in a mixture containing 10% DMSO/FBS.
10,000/50,000 RASMC, 50,000 HeLa or 100,000 GFP-HeLa in antibiotic-free medium were seeded in 24-well plates, incubated at 37°C, 5% CO2, for 24 hr before addition of transfection complexes. To prepare transfection complexes, 10–300 nM siActin or 0.1–20 nM siGFP were diluted in 100 µl serum-free media, added to 1.2–2.0 µl nTMag (nanoTherics, UK) and mixed. After 20 min, complexes were added drop-wise to cells, with control wells comprising untransfected cells. Plates were incubated in the presence of an oscillating magnetic field (2 Hz at 0.2 mm amplitude of displacement) using the magnefect-nano II (nanoTherics, UK) system.
Cells were seeded into 24-well tissue culture plates as described before. Lipofectamine 2000 complexes were prepared in serum free medium using 33 nM of siRNA with 1 µl of Lipofectamine 2000 per well following the manufacturer’s recommended protocol. After 20 min incubation, complexes were added drop-wise to cells, and incubated at 37°C, 5% CO2, before analysis.
Cells were transfected by nucleofection (Amaxa/Lonza, DE) using program I-13, according to manufacturer’s instructions. Cells were left to express plasmids for 72 or 168 hr before flow cytometric analysis.
Transfected cells were washed in 0.5% bovine serum albumin and phosphate-buffered saline (PBS) and analysed for the relative fluorescence of gated cells, using a FACSort analyser (Becton Dickinson, USA). Median fluorescence intensity of gated cells was determined through the FL1 channel.
Cell viability was assessed using the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay, which is based on the ability of a mitochondrial dehydrogenase enzyme from viable cells to cleave the tetrazolium rings of the pale yellow MTT to form dark blue formazan crystals. 0.5 mg/ml of MTT dissolved in serum-free medium was added to cells in wells. Cells were allowed to accumulate crystals for 2–4 hr in a 37°C, 5% CO2 incubator, before they were solubilised with dimethyl sulfoxide and read using spectrophotometer at a wavelength of 595 nm. Number of viable cells was directly proportional to the level of the formazan product created
This protocol was based on Dormoy-Raclet et al
Cells were treated with 2–50 µg/ml chlorpromazine (CHL), 1.0–25 µg/ml filipin, 0.2–5 µM 5-(N-ethyl-N-isopropyl) amiloride (EIPA) (Sigma, Dorset, UK), 40–1000 µM genistein (GENI) or 8–200 µM dynasore (DYNA) (Calbiochem/Merck, Nottingham, UK) in serum-free culture medium for 30 min at 37°C. Subsequently, nTMag complexed with GFP in serum-free medium were added drop-wise to cells, transfected at 2Hz with 0.2 mm amplitude of displacement using the magnefect-nano II system (nanoTherics, UK) for 30 min, and incubated for a further 1 hr (2 hr total contact time with inhibitors). Fresh serum-containing medium was replaced and cells were analysed for transfection efficiency and viability using fluorescence microscopy and the MTT assay after 24 hr, respectively. Total transfection was obtained by multiplying the proportion of GFP transfected cells by its corresponding %viability.
We sought to transfect short interfering ribonucleic acid (siRNA)-associated magnetic nanoparticles into mammalian cells using the oscillating magnet array. For simplicity, and as proof of principle, HeLa cells were first stably transfected with a plasmid encoding green fluorescence protein (GFP), selected (G418) and sorted on the basis of GFP as described in the
Stably-expressing GFP-HeLa cells were transfected, either with a lipid reagent, nucleofection method or magnetofection using an oscillating magnet (2Hz frequency amplitude and 0.2 mm displacement), as indicated. Magnetofection using nTMag only was set as negative control (Mock). Fluorescent intensity (Median) was determined 72 hr post-transfection and knockdown levels were normalised to a scrambled control. Results are expressed as the mean±SD of at least three independent experiments. *, p<0.05; **, p≥0.05.
Gene silencing was determined using fluorescence median within a gated population of cells and normalising to a scrambled siRNA control. Although the nucleofection method was able to silence GFP after 72 hr (77.7±0.4%) using a high dose of siRNA (500 nM), this was not sustained as observed after 168 hr (43.6±3.8%, data not shown). The oscillating magnet array system, with a lower dose of siRNA (20 nM), showed approximately 61.1±6.1% knockdown of GFP, both after 72 hr and 168 hr (
Next we wanted to silence an endogenous protein using the oscillating magnet array technology. Actin, a major cytoskeletal protein, found ubiquitously in eukaryotes, is known to be involved in a range of cellular activities including adhesion
In HeLa cells, we were successful in introducing siRNA against β-actin using the oscillating magnet array, as evidenced by a decrease in reattached HeLa cells onto gelatine-coated plates 72 hr after transfection
HeLa (A–C) or rat aortic smooth muscle cells (D, E) transfected with the either scrambled siRNA or siRNA against actin, as indicated, were seeded on six-well gelatine-coated plates 72 hrs post-transfection. Adherent, Hoechst–stained (nuclear) HeLa cells after transfection with scrambled siRNA (A) or actin siRNA (B) were counted by fluorescence microscopy, as described in the
A similar effect can be seen in a primary cell type, the rat aortic smooth muscle cells (
This contrasted with wells plated with 50,000 cells, which had more consistent levels of knock-down at all concentrations of siRNAs and viabilities (between 39.8±3.0% for 1 nM to 52.6±7.1% for 300 nM;
Finally, in order to understand the mechanism by which these nucleic acid-nanoparticle complexes enter into the target cells, we investigated the impact of endocytosis blockers on nanomagnetic transfection efficiency. It has been reported that non-viral gene carriers including magnetic nanoparticles, are energy dependent and exploit the endocytic pathway to gain entry into their target cells
On analysis 24 hrs post transfection, we found transfection of HeLa cells with nTMag associated with pEGFP-N1 plasmid was affected by low doses (
HeLa cells were transfected with pEGFP-N1 in the presence of varying doses of inhibitors, as indicated and described in the
Previously we have demonstrated the use of a horizontal/lateral motion to a magnet array improves the transfection of both easy- and hard-to-transfect cell lines (as described earlier). In this study we explored the potential of this technique for gene silencing using short-interfering RNA (siRNA) and to understand the molecular mechanism behind magnetofection using an oscillating magnet array. As proof of principle, we created stably expressed green fluorescence protein (GFP) in HeLa and HEK293 cells by transfecting using our oscillating magnet array system and selecting using G418 antibiotics. Following sorting based on GFP fluorescence, we succeeded in silencing the GFP gene using siRNA complexed with nTMag and transfected with our oscillating magnet array. Our oscillating magnet array system had the benefit of using less siRNA in order to achieve high levels of gene silencing, compared with the nucleofection method. Also our system appears to prolong siRNA knock-down function without the overgrowth of untransfected cells or cell death over time, something we observed with the nucleofection and lipid reagent methods, respectively.
Next we examined knock-down of an endogenous gene. Actin, a major cytoskeletal protein, found ubiquitously in eukaryotes, is known to be involved in cell division, phago-/endo-/exo-cytosis, motility/adhesion, etc
Another aim of this work was to determine route of entry of the magnetic nanoparticle complexes into the cell. Magnetofection of HeLa cells involves a combination of non-specific, clathrin- and caveolae- dependent endocytosis, as determined using antimycin A, a non-specific endocytic and other energy-dependent processes
Previously we have shown that using our oscillating magnet array (set to oscillate at 2Hz with a 0.2 mm displacement), NCI-H292 cells transfected with magnetic nanoparticles associated with a plasmid containing a luciferase reporter had 50% decrease in luciferase expression, in the presence of either antimycin A (1 µg/ml) or nystatin (25 µg/ml), in general agreement with
Using a combination of endocytosis inhibitors, we found that magnetic nanoparticles complexed with plasmid encoding GFP, entered HeLa cells via caveolae-mediated endocytosis. While our results are largely in agreement with other groups
While we are aware of the pitfalls of using pharmacological inhibitors to study the endocytic route of entry for the magnetic nanoparticles – e.g. lack of specificity on the part of the inhibitor itself, efficacy of inhibition and viability of cells after exposure to inhibitors and transfection is highly cell dependent
JL would like to thank Joshua Rappoport (University of Birmingham) for critically reading this manuscript and his colleagues at nanoTherics Ltd. and the Guy Hilton Research Centre, for their help and support.