Gene Transfection of Mammalian Cells Using ... - ACS Publications

Using pEGFP and pSEAP plasmids and NIH 3T3 fibroblasts as models, we demonstrate a new electroporation-based gene delivery method, called membrane ...
0 downloads 0 Views 368KB Size
Anal. Chem. 2007, 79, 5719-5722

Gene Transfection of Mammalian Cells Using Membrane Sandwich Electroporation Zhengzheng Fei,† Shengnian Wang,‡ Yubing Xie,‡,§ Brian E. Henslee,† Chee Guan Koh,† and L. James Lee*,‡

Department of Chemical and Biomolecular Engineering and NSF Nanoscale Science and Engineering Center for Affordable Nanoengineering of Polymer Biomedical Devices, The Ohio State University, 140 West 19th Avenue, Columbus, Ohio 43210

To avoid safety issues such as immune response and cytotoxicity associated with viruses and liposomes, physical methods have been widely used for either in vivo or ex vivo gene delivery. They are, however, very invasive and often provide limited efficiency. Using pEGFP and pSEAP plasmids and NIH 3T3 fibroblasts as models, we demonstrate a new electroporation-based gene delivery method, called membrane sandwich electroporation (MSE). The MSE method is able to provide better gene confinement near the cell surface to facilitate gene transport into the cells and thus shows significant improvement over transgene expression of mammalian cells compared to current electroporation techniques.

bulk electroporation requires the use of a high electric voltage, leading to low cell viability and limited transfection efficiency. In this paper, we present a much less invasive and more efficient gene delivery method, called membrane sandwich electroporation (MSE). The MSE platform is shown in Figure 1. We trapped the cells on a track-etched poly(ethylene terephthalate) (PET) membrane. Cell immobilization on a porous surface leads to localized cell electroporation, allowing the use of a low applied voltage to achieve temporarily dielectric breakdown of the cell membrane.8,9 When we placed another track-etched PET membrane on the top of the immobilized cells and sandwiched the cells between the two membranes, we observed a significant improvement of gene transfection with minimal cell damage.

A major challenge in gene therapy and drug delivery is to deliver genes and high molecular weight drugs into mammalian cells with a high transfection efficiency and minimal cell damage. Viruses and liposomes have been widely used as in vivo carriers, but safety issues such as immune response and cytotoxicity have limited their clinical applications.1,2 Physical and mechanical methods such as microinjection, particle bombardment (gene gun), and electroporation are more benign, because these methods can directly transfer naked DNA into cells and avoid the risks associated with introducing a secondary agent.3,4 Among the physical and mechanical methods, electroporationenhanced delivery of plasmid vectors is gaining acceptance for both in vitro and in vivo applications.5 The electroporation process induces transient openings in the plasma membrane by executing electric pulses on cells and driving genes or drugs into the cytoplasm.6,7 It is applicable to a wide variety of animal cells and tissues, simple to perform, and easy to use. However, conventional

MATERIALS AND METHODS Fabrication of Microfluidic Device. The microfluidic device consists of a pair of cross-channels connected by a center hole (shown in Figure 1a). One channel is present on the top of the device, and the other is on the bottom. Both channels are 500 µm in width and depth. The channel on the top of the device intersects with a 1-cm-diameter reservoir located at the center of the device where membranes can be fixed to the device. The microfluidic device was fabricated in a poly(methyl methacrylate) (PMMA) substrate using a high precision computer numerically controlled machine (AeroTech Inc, Pittsburgh, PA). A 45-µm-thick PMMA film was welded on the back side of the device using a thermal film laminator (Catena 35, GBC, Addison, IL), enclosing the bottom channel but allowing top access via reservoirs at the ends. Plasmids. Plasmid gWiz GFP (5757 bp) and secreted alkaline phosphatase (SEAP) (6569 bp) were purchased from Aldevron (Fargo, ND) and purified with an EndoFree Plasmid Maxi Kit from Qiagen (Valencia, CA) according to the manufacturer’s instructions. Cell Culture and Preparation. NIH 3T3 cells (mouse embryonic fibroblast cell line) were cultured in Dulbecco’s modified Eagle’s medium: Nutrient Mix F-12 (D-MEM/F-12) supplemented with L-glutamine (2 mM), sodium pyruvate (1 mM), and 10% (v/v) newborn calf serum. Cells were maintained in 25 cm2 T-flasks at 37 °C with 5% CO2 and subcultured using 0.25%

* Corresponding author. Phone: 614-292-2408. Fax: 614-292-8685. E-mail: [email protected]. † Department of Chemical and Biomolecular Engineering. ‡ NSF Nanoscale Science and Engineering Center for Affordable Nanoengineering of Polymer Biomedical Devices. § Current address: College of Nanoscale Science and Engineering, University of Albany, Albany, NY 12222. (1) Thomas, C. E.; Ehrhardt, A.; Kay, M. A. Nat. Rev. Genet. 2003, 4, 346358. (2) Niidome, T.; Huang, L. Gene Ther. 2002, 9, 1647-1652. (3) Mehier-Humbert, S.; Guy, R. H. Adv. Drug Delivery Rev. 2005, 57, 733753. (4) Wells, D. J. Gene Ther. 2004, 11, 1363-1369. (5) Li, S. Curr. Gene Ther. 2004, 4, 309-316. (6) Andre´, F.; Mir, L. M. Gene Ther. 2004, 11, S33-S34. (7) Gehl, J. Acta Physiol. Scand. 2003, 177, 437-447. 10.1021/ac070482y CCC: $37.00 Published on Web 06/29/2007

© 2007 American Chemical Society

(8) Khine, M.; Lau, A.; Ionescu-Zanetti, C.; Seo, J.; Lee, L. P. Lab Chip 2005, 5, 38-43. (9) Kurosawa, O.; Oana, H.; Matsuoka, S.; Noma, A.; Kotera, H.; Washizu, M. Meas. Sci. Technol. 2006, 17, 3127-3133.

Analytical Chemistry, Vol. 79, No. 15, August 1, 2007 5719

Figure 1. Experimental setup of (a) MSE platform and (b) schematic of DNA migration path. Table 1. Comparison of Conventional Bulk Electroporation (Single Cuvette), Localized Cell Electroporation, and MSEa localized cell electroporation and MSE

method field strength (V/cm) pulse frequency (Hz) pulse duration (ms) no. of pulses electroporator

bulk electroporation (single cuvette) 9 electroporation 1,600 40 0.4 1 Bio-Rad Gene Pulser Xcell

DNA attraction

35 3.5 1 100 500 5 5 300 homemde microfluidic cell

a The pulse type was a bipolar square wave. Field strength is defined by the peak-to-peak amplitude. The lower limits of both electroporators equal to zero.

(w/v) trypsin with EDTA‚4Na. All cell culture reagents were purchased from Invitrogen (Carlsbad, CA). Before experimentation, NIH 3T3 cells were plated on a 35mm-i.d. plastic petri dish and allowed to grow for 1 day. Cells were rinsed with trypsin-EDTA solution, washed with Dulbecco’s phosphate-buffered saline without calcium or magnesium, and adjusted to a final cell density of 1 × 106 cells/mL in Opti-MEM I reduced-serum medium (w/o phenol red). Electroporation Procedure. The 0.1-cm gap electroporation cuvettes were used for conventional bulk electroporation. The 100µL drop of suspended cells (1 × 105 cells) and 5-µg DNA sample were loaded into the cuvette. The electroporation conditions are given in Table 1. For localized cell electroporation and MSE, a 3-mm-diameter track-etched PET membrane (BD Biosciences, San Jose, CA) was used as the support membrane with an average pore size of 400 nm and fixed at the center reservoir of the microfluidic device by sealing tape (shown in Figure 1). First, a 10-µL drop of suspended cells (1 × 104 cells) was loaded onto the support membrane, and a vacuum of 34 ( 3 kPa was used to trap the cells on the support membrane. Next, another 3-mm-diameter track-etched PET membrane with an average pore size of 3 µm was added on top of the immobilized cells with a spacer of ∼10 µm between the two membranes. Opti-MEM I reduced-serum medium was then loaded into the channels and the center reservoir. Two thin silver wire electrodes were placed in the inlet and outlet reservoirs, and a 0.5-µg DNA sample was loaded into the reservoir with the cathode. 5720

Analytical Chemistry, Vol. 79, No. 15, August 1, 2007

Finally, a DNA attraction step was performed, followed by electroporation (the conditions are given in Table 1). After 1520 min, the support membrane with the cells was transferred to a 24-well plate, followed by a 24∼48-h cell culture before measuring the transfection efficiency. Detection of Green Fluorescence Protein (GFP) Expression. The transfection efficiency of pEGFP was qualified by the percentage of the cells with green fluorescence. An inverted fluorescence microscopy (TS100, Nikon) was used to detect GFP expression and cell viability 24 h after electroporation. Assay for Alkaline Phosphatase (AP) Activity. The transfection efficiency of pSEAP was quantified by the activity level of AP secreted by the transfected cells. Samples of culture media were collected 48 h after electroporation and determined by a colorimetric assay based on the hydrolysis of p-nitrophenyl phosphate (pNPP). A 100-µL sample of culture media and 30 µL of pNPP substrate solution (Sigma) were added into each well of a 96-well plate. The plate was incubated in the dark for ∼30 min at room temperature and read at 405 nm on a multiwell plate reader (GENios Pro). DNA Distribution Study by Spin-Disk Confocal Microscopy. A dilute λ-DNA (48.5 kbp, New England Biolabs, Ipswich, MA) solution (0.03 µg/mL) was used in the confocal microscopic experiments. It was prepared in Tris-EDTA buffer and labeled with a fluorescent dye (YOYO-1, Molecular Probes, Eugene, OR) at a dye-base pair ratio of 1:5, following the procedure given by Perkins et al.10 The DNA solution was loaded to the cathode side, while the anode side was loaded with the buffer solution only. A spin-disk confocal microscope (VisiTech Ltd., Alexandria, VA) with Z-stacking was used to trace the location of DNA molecules in the gap of the MSE setup. The Yokogawa spin-disk scanning unit (CSU-22) was connected to an inverted microscope (Olympus IX-81, Tokyo, Japan). A Jena piezocontroller was mounted underneath the 60× oil objective (1.42 NA) to carry out a z-direction scan with submicrometer accuracy (the minimum distance is 100 nm). RESULTS AND DISCUSSION Delivery of Plasmids pEGFP and pSEAP to NIH 3T3 Fibroblasts. Using plasmid GFP and NIH 3T3 fibroblasts as reporter gene and model cells, two different cases were tested for localized cell electroporation. Cells and genes were placed on (10) Tekle, E.; Astumian, R. D.; Chock, P. B. Proc. Natl. Acad. Sci. U. S. A. 1991, 88, 4230-4234.

Figure 2. Comparison of MSE with conventional bulk electroporation and localized cell electroporation using plasmids GFP (a-d) and SEAP (e). The green fluorescence indicated GFP expression 24 h after (a) bulk electroporation, localized cell electroporation with genes and cells on (b) opposite sides and (c) the same side of the support membrane, and (d) MSE. (e) The bars indicated the activity levels of SEAP expressed by NIH 3T3 cells 48 h after electroporation. Data were plotted with the standard deviation from the mean (n ) 3).

either opposite sides (Figure 2b) or the same side (Figure 2c) of the support membrane. In both cases, only a slight improvement was observed of GFP expression over the conventional bulk electroporation method (Figure 2a). When the MSE method was used, most cells survived after the treatment, and GFP expression (Figure 2d) was much higher than in bulk electroporation (Figure 2a) or in localized cell electroporation (Figure 2b and c). Using another plasmid SEAP, the levels of transgene expression mediated by localized cell electroporation and MSE were quantified. The amount of SEAP expression mediated by MSE was improved ∼40% over localized cell electroporation (Figure 2e). In the conventional bulk electroporation, a layer of foam was observed due to cell lysis in the high-intensity electric field (1600 V/cm). In comparison, a much low electric field with the amplitude of 35 V/cm was applied in our MSE method, and more than 90% of cell

viability was achieved. Cell viability was quantified right after electroporaion with the trypan blue method. Mechanism Analysis by a Spin-Disk Confocal Microscope. To explain why the MSE method promotes transgene delivery, Z-stacking was carried out using a spin-disk confocal microscope with a 60× oil objective to trace the local concentration of DNA molecules during the electroporation process. This system is similar to the confocal laser scanning microscopy micro-PIV system,11 but equipped with a Yokogawa CSU-22 spin-disk unit and Hamamatsu EM CCD camera. It is capable of scanning more than 120 full-frame images (1024 × 1024 pixels) per second, sufficient to directly measure the DNA distribution inside the sandwich gap in the MSE setup. To facilitate visualization, YOYO-1 (11) Perkins, T. T.; Smith, D. E.; Chu, S. Science 1997, 276, 2016-2021.

Analytical Chemistry, Vol. 79, No. 15, August 1, 2007

5721

DNA molecules into the cells is driven by the electrophoretic force13 and is dependent on the duration and number of electric pulses.14 The nanoscale pores in the support membranes allow a focused electric field on the cell membrane and thus enhance cell permeabilization at a low electric voltage. However, negatively charged DNA molecules quickly migrate away from the negatively charged cell surface after the pulse duration because of electrical repulsion. This can seriously limit gene transfer into the cells. When a negatively charged track-etched PET membrane is placed on top of the cells, DNA molecules are prevented from diffusing away, as demonstrated in Figure 3. Thus, the sandwich membrane configuration is able to provide better gene confinement near the cell surface to facilitate genes transport into the cells.

Figure 3. DNA distribution in the gap between two membranes in the observed domain. The zero position is set at the surface of the top membrane, while the 10-µm position is near the surface of the bottom membrane.

conjugated large λ-DNA molecules were used instead of plasmids GFP and SEAP. Scanning was carried out every 0.4 µm across the 10-µm gap between two membranes. The scanning time interval between two adjacent planes was ∼0.04 s, about one-fifth of the time for DNA to diffuse across the two planes (estimated by t ≈ L2/2D ) 0.22 s, where the diffusion of λ-DNA molecules is calculated by D ∼ LDNA-ν ≈ 3.6 × 10-13 m2/s, ν ) 0.60, LDNA ) 17 µm for λ-DNA.12 Therefore, DNA molecules observed in adjacent planes must be different individual molecules. From Figure 3, a large number of DNA molecules were found in the gap after electroporation, with a decreasing number of DNA molecules near the membrane surface. Without the top membrane, DNA molecules were hardly seen this time within the same distance of 10 µm from the bottom membrane. During electroporation, the extent of cell permeabilization is dependent on the amplitude of electric pulses, while the transport of the polyanionic (12) Park, J. S.; Choi, C. K.; Kihm, K. D. Exp. Fluids 2004, 37, 105-119. (13) Smith, D. E.; Perkins, T. T.; Chu, S. Macromolecules 1996, 29, 1372-1373. (14) Mir, L. M.; Moller, P. H.; Andre´, F.; Gehl, J. Adv. Genet. 2005, 54, 83105. (15) Gabriel, B.; Teissie, J. Biophys. J. 1997, 73, 2630-2637.

5722

Analytical Chemistry, Vol. 79, No. 15, August 1, 2007

CONCLUSION A new membrane sandwich electroporation approach was demonstrated using plasmids GFP and SEAP as model materials. NIH 3T3 fibroblasts were tested and a significant improvement in transgene expression was observed compared to current electroporation techniques. In the MSE method, the focused electric field enhances cell permeabilization at a low electric voltage, leading to high cell viability; more important, the sandwich membrane configuration is able to provide better gene confinement near the cell surface, facilitating gene delivery into the cells. Successful examples of in vitro electroporation trials have been done on animal and human patients. Since typically cells or tissues from the patients are very limited and therapeutic materials such as plasmids and oligonucleotides are very expensive, our MSE method, with the ability to deal with small number of cells with high transfection efficiency and cell viability, offers a great impossibility for ex vivo gene therapy. The applicability of the MSE method to primary cells and hard-to-transfect cells (such as mouse embryonic stem cells and leukemia cells) is currently under investigation in our laboratory. ACKNOWLEDGMENT The authors thank the NSF Center for Affordable Nanoengineering of Polymeric Biomedical Devices for financial support (Grant EEC-0425626). Received for review March 8, 2007. Accepted May 25, 2007. AC070482Y