Introduction
Viral vectors are commonly used to generate induced pluripotent stem cells (iPSCs) because of their high efficiency to introduce foreign DNA into mammalian cells [1-3]. However, virus-free and/or integration-free plasmids in combination with non-viral transfection methods are often recommended due to the concerns of clinical application [4,5]. So far, many scientific research groups have introduced numerous devices or materials for non-viral transfection methods including diethylaminoethyl-dextran (DEAE-dextran), calcium phosphate, liposome (lipofection), non-liposomal based transfection (NLT) reagent, polymer, magnetic bead and electroporator.
DEAE-dextran and calcium phosphate are the chemicals that are traditionally used to increase transfection efficiency of viral RNA and DNA into mammalian cells [6,7]. Due to their positively charged property, they can easily bind to DNA, form co-precipitates and ultimately deliver DNA either into nucleus or cytoplasm of the cells [8,9]. Liposome is one of the widely used transfection methods, taking advantage of the ability to fuse with lipid bilayers of the cells, and this endocytosis-like action can easily uptake foreign DNA and induce the expression of introduced gene [10]. Other than liposomal based transfection, recently developed non-liposomal based transfection reagent also became a potent transfection tool [11]. Electroporation makes electric field which drives negatively charged DNA into targeting cells [12,13]. Among polymers developed, polyethyleneimine (PEI) and poly-beta-amino ester are effective “proton sponge” that can facilitate the transfection efficiently [14,15]. Using the magnetic bead is in the limelight among recent transfection methods because of its relatively higher transfection efficiency and less toxic effects when compared with lipofection and NLT methods [16]. For magnetofection, PEI which binds to DNA is coated over magnetic bead [17]. Then magnetic force exerted by magnetic plate attracts magnetic bead-PEI-DNA complex into cells [18].
In general, primary cell cultures are more difficult to be transfected than immortalized cell lines. Recent reports suggest that transfection of a specific factor could ameliorate some injured tissues when subsequently transplant the transfected primary cells [19]. Hence, improving transfection efficiency for cultured primary cells is at the forefront of gene-, cell-based therapy, tissue engineering and regenerative medicine [20]. Moreover, there is less information about optimal transfection protocol for bovine primary cultured cells [21,22]. To increase the transfection efficiency other than viral infection in primary cell cultures, various transfection methods has been developed [23-25].
Recent advancement of transfection technologies can achieve iPSC generation using a couple of non-viral transfection methods above mentioned [4,26]. Due to its high efficiency for non-viral plasmid transfection, electroporation has been applied to iPSC generation from the cells in various mammalian species [27-29]. Other than electroporation, polymer-based transfection and magnetofection were also successfully applied to generate iPSCs in mouse model [30,31]. In spite of the iPSC's successful application, its comparably lower efficiency of virus-free transfection than viral transfection is remained as the biggest hurdle to be overcome [32]. When generating iPSCs, poor efficiency is mainly caused by individual co-transfection of several vectors. To overcome this problem, a polycistronic vector is cloned and applied by many groups [29]. In this case, however, as several factors need to be cloned in one polycistronic vector, the final vector product is more than twofold longer than the monocistronic vector, often larger than 10 kb, and this reversely decreases transfection efficiency [33]. Therefore the optimal transfection protocol needs be investigated to elevate transfection efficiency when delivering longer plasmid DNA into cells. Despite a series of successful reports on establishment of iPSCs, low efficiency is still observed in human or large animal iPSCs when compared with a murine model. Therefore, an optimal transfection protocol is required to the cells of these species.
In the present study, the episome-derived enhanced green fluorescent protein (eGFP) vector was used to transfect bovine fetal fibroblasts (bFF) and human dental pulp cells (hDPC) to compare three different transfection protocols, NLT, magnetofection and electroporation by examining their transfection efficiency and cell viability. This may elucidate transfection efficacy of larger plasmid when applying to human and bovine somatic cells.
Materials and Methods
Chemicals and plastic wares
All inorganic and organic compounds were obtained from Life Technologies (Grand Island, NY, USA) unless indicated in the text. All plastic ware were purchased from Corning (One Riverfront Plaza, NY, USA).
Primary culture of human and bovine cells
To isolate human dental pulp tissue, mesiodens (maxillary central supernumerary teeth) were extracted from children at the Department of Pediatric Dentistry in Dental Hospital of Seoul National University according to the guidelines provided by ethics committee (S-D20100005). Dental pulp tissues were dissociated using 1% (w/v) collagenase type I, 2.4 mg/ml dispase and 0.25% (v/v) trypsin, and then proceeded to whole-tissue culture. After culturing for several days, hDPC can be observed propagating from center of the tissue to the peripheral area. The hDPC were isolated from two individual children were named to hDPC-1 and hDPC-2, respectively.
The bFF were isolated from femoral skins of 60 – 90 days old fetuses. Bovine femoral skins were dissociated with 0.25% trypsin for 30 min and dissected using blades. Cells are well propagated after culturing in dishes. To eliminate heterogeneous cell populations, the cells were cultured at least for five passages before transfection.
Plasmid DNA preparation
pCXLE episomal DNA (plasmid 27082; Addgene, Cambridge, MA, USA) was obtained from Addgene. Midi prep was carried out either by using HiSpeed Midi Kit (Qiagen, Valencia, CA, USA) or Purelink Maxi Kit. Final plasmid DNA concentration was used between 1.0 μg/μl to 1.5 μg/μl.
Cell culture
The bFF were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Welgene, Daegu, Korea), supplemented with 10% fetal bovine serum (FBS; Atlas, Fort Collins, CO, USA), 100 U/ml penicillin – 100 μg/ml streptomycin (Sigma-Aldrich, Yong-in, Korea), 250 ng/ml fungizone. The hDPC were cultured in minimum essential medium – alpha (MEM-α), supplemented with 10% FBS, 100 U/ml penicillin–100 μg/ml streptomycin. Both bFF and hDPC were thawed at passage 4 and at least re-plated twice before transfection.
Cell transfection
For NLT and magnetofection, cells were seeded in 6-well plates at the density of 1.5×105 per well 24 h prior to transfection to match the cell density to 2.0×105 approximately as in electroporation. Electroporation was performed at the density of 2×105 cells per well. Before bFF and hDPC transfection, medium was suctioned out, washed with PBS for three times and freshly supplied with new medium for 1.8 ml in 6-well plates at least 30 min ago. Three transfection protocols are as follows.
For NLT, FuGENE-HD (Promega, Madison, WI, USA) and X-treme GENE 9 DNA transfection reagent (Roche, Mannheim, Germany) were used. The ratio of DNA and reagent were 2 to 8 (FuGENE-HD) and 2 to 6 (X-treme GENE 9) in hDPC, and 2 to 7 (FuGENE-HD) and 2 to 6 (X-treme GENE 9) in bFF. First, OptiMEM was aliquoted for 200 μl in 1.5 ml tube. Then transfection reagent and DNA mixture were added and mixed vigorously to make a homogeneous solution. After 15 min of incubation at 20-25°C , mixture solution was evenly scattered onto cell culture medium.
For magnetofection, plasmid DNA was dissolved in 200 μl OptiMEM and subsequently mixed with triple volume of Polymag (Chemicell GmbH, Berlin, Germany) for 20 min in room temperature. After replacing culture medium, Polymag-DNA complex dissolved in OptiMEM was added to the cells in culture, then incubated in a humidified 37°C incubator.
Electroporation was conducted using either NEPA™ or Neon™ transfection system. In NEPA™ electroporation system, for poring pulse, voltage, length, interval, number, decay rate, polarity were set respectively as 150 V, 5 ms, 50 ms, 2, 10%, ‘+’ for hDPC, and 275 V, 1.5 ms, 50 ms, 2, 10% and ‘+’ for bFF. For transfer pulse, voltage, length, interval, number, decay rate, polarity were set respectively as 20 V, 50 ms, 50 ms, 5, 40% and ‘+/-’ for both hDPC and bFF. In Neon™ electroporation system, voltage and length tested here were 1,400 V and 20 ms for hDPC, and 1,150 V and 30 ms for bFF, and the pulse was repeated twice in both cell groups.
Fluorescence detection
Fluorescence was observed using fluorescence microscope (Nikon Eclipse TE2000-U; Nikon, Tokyo, Japan) either on Day 1 or Day 2 to verify its GFP expression and cell condition.
Measurement of cell viability after transfection
Control and transfected cells were dissociated and harvested on Day 2 using trypsin. Cells were centrifuged and resuspended in 1 ml PBS. Then 10 μl samples are mixed with 10 μl trypan blue. Dead cells which stained by trypan blue were included to count potential survived cells.
Fluorescence activated cell sorting
Cells were sorted using fluorescence activated cell sorting Calibur (FACS Calibur; BD biosciences, San Jose, CA, USA) to measure transfection efficiency. Cells were all dissociated using 0.05% trypsin (v/v) and transferred into 15 ml tube. Then cells were centrifuged for 5 min to form pellets at the bottom of the tube. Supernatants are suctioned out and pellets are re-suspended with PBS. After another 5 min centrifugation and suction, 2% paraformaldehyde was added into tube to fix cells. Then centrifugation was performed twice and re-suspension using PBS was carried out. For each sample, 10,000 cells were sorted, and gated region was matched to control group. Within gated region, the transfection efficiency was measured according to FL-1 detected eGFP cell number.
Results
For the transfection to the hDPC, cells were seeded 24 h prior to NLT and magnetofection. Electroporation was performed immediately after trypsinization. Fluorescence of eGFP was detected 48 h after transfection (Figure 1), and most hDPC were in normal condition and exhibited GFP except magnetofection group which showed large damages with no fluorescence activity (Figure 2). Fluorescence activated cell sorting by SSC and forward scatter (FSC) indicated that the cell condition is mostly defined by cell size and granularity in magnetofection group was also largely deviated from non-transfected control (Figure 3a) whereas most cells were in normal condition when using other transfection methods (Figure 3b, c, d and e). This made exclude magnetofection from the further experiments in hDPC. FuGENE-HD, X-tremeGENE 9, NEPA™ and Neon™ showed 35.6, 57.0, 7.9 and 25.6% of cell viabilities (Figure 4a), and 14.7, 6.7, 8.6 and 33.2% of transfection efficiencies based on viable cell population (Figure 4b), respectively.
On contrary to hDPC, magnetofection was applicable to bFF as the cells showed a well condition after 2 days of transfection as well as other transfection methods demonstrating theirapplicability either by its cell morphology or GFP fluorescence under microscope (Figure 5). Cell morphology analyzed by cell sorter also revealed that the cells in all experimental groups presented little damages as there was limited deviation from non-transfected control (Figure 6a, b, c, d and e).
After applying these conditions, magnetofection, FuGENEHD, X-tremeGENE 9, NEPA™ and Neon™ protocols exhibited 82.3, 67.3, 68.9, 24.2 and 29.4% for viabilities (Figure 7a) and 6.9, 4.2, 8.5, 19.0 and 49.5% for transfection efficiencies (Figure 7b) respectively.
The yield of final transfected cells from total cells were calculated and presented in Table 1 and 2. The cells which seeded on Day -1 for magnetofection and NLT are presumed to be proliferated to the density of 2.0×105 on Day 0, and followed by applying doubling time as 20 h for hDPC and 15 h for bFF in the following two days. Hence the number of cells on Day 2 for hDPC is 10.56×105 and 18.36×105 for bFF. The formula for calculating final transfected cells is multiplication of the number of cells on Day 2. The results demonstrates Neon™ transfection system is most suitable transfection method among magnetofection, NLT and electroporation, which could obtain 0.89×105 and 2.67×105 transfected cells after 2 days in culture from initial 2.0×105 cells in hDPC and bFF, respectively.
Discussion
Recently, iPSCs became one of the most important cell sources in studying differentiation and dedifferentiation processes, drug discovery and regenerative medicine by transplanting the cells [34-36]. However, the efficiency of iPSC generation is still low, mainly because of low simultaneous transfection efficacy of several key factors such as Oct4, Sox2, Klf4, C-myc. Viral transfection into mammalian cells is commonly used to generate iPSCs [1]. Despite its high transfection efficiency, it has several concerns in human medicine such as the possibility of unknown genetic mutation and tumor formation [37]. To avoid these concerns, virus-free transfection methods are required to apply to a human being [38,39]. In this study, NLT, magnetofection and electroporation methods were compared using hDPC and bFF as the cell sources. The results show that Neon™ electroporation is the best transfection method when compared with NLT, magnetofection and NEPA™ electroporation.
Magnetofection is a recently introduced gene delivery technique and is known to be effective in many cell lines [40]. Furthermore, it shortens the duration of transfection with high cell viability [41]. These advantages have been reported in many previous studies [42]. However, aggravated condition of transfected hDPC with deviated morphology indicates that magnetofection is an improper transfection method because it confers cell damage to hDPC. Although it exhibited comparably higher cell viability than other methods in bFF, magnetofection is so far behind electroporation in transfection efficiency.
Lipofection mediated transfection is one of the earliest transfection methods and its partially charged feature enables foreign genes to be delivered into cells [43]. Recently NLT also became more highlighted than conventional liposome-based transfection reagent [11,44]. In this study, transfection efficiency of NLT turns out to be low in transfecting DNA into hDPC and bFF than Neon™ electroporation method and little difference than NEPA™ electroporation. This contrasts the previous report demonstrating its higher transfection yield than electroporation [45]. There are also some reports suggesting that dental follicle cells or dental pulp stem cells are able to be transfected with other liposome-based transfection reagents, FuGENE-HD and Fugene 6. However, the efficiencies of gene delivery were lower than the electroporation method [46,47]. When compared with Neon™ electroporation method, these results are in consistence with our conclusion suggesting an inferior transfection method with NLT when transfecting hDPC.
Electroporation is a widely utilized transfection method in various types of cells [48,49]. Electric impulses could efficiently increase the delivery of DNA into desired cell targets [12,13]. In this study, Neon™ electroporation exhibited the highest transfection efficiency both in hDPC and bFF suggesting its effective application to human and large animals.
The results presented here may be applied to transfecting the specific growth factors for therapeutic uses [50]. In addition, these two cell types are all mesenchymal cell lineage which is known to generate iPSCs easier than other lineage types by the process of mesenchymal to epithelial transitions (MET) [51]. In conclusion, one of electroporation method tested here could improve transfection efficiency both in hDPCs and bFFs and this may improve gene delivery efficiency in human and bovine cell researches.
