The iPSC technology opens up an unlimited source of variable human cell types for regenerative medicine. Administrating economical transfection methods such as cationic polymers provide a simplified platform for large-scale production of iPSCs. This method can be used in all iPSC institutes, especially in the labs facing financial issues. In the current investigation, we represented a cost-effective and optimized protocol for transfecting HDFs through a gold standard reagent (i.e PEI) and oriP/EBNA1-based vectors to produce hiPSCs. Previous evidences have confirmed the capability of PEI to transfect several types of cells such as HEK293[2], mesenchymal[39], neural[40], and embryonic[41] stem cells. However, with the aim of generating hiPSCs, our investigation is of the pioneers setting and standardizing several parameters relevant to HDF transfection process using PEI25K.
Our results revealed that using 2 µg pDNA implies a lower transfection rate and higher cell death in comparison to 1 µg pDNA, which may be related to the aspect ratio of PEI/pDNA. Higher concentrations of pDNA require a higher amount of PEI for modification and permeabilization of endosomal membrane, which causes a higher cytotoxic effect of polyplexes on cell viability[42, 43]. In addition, regardless of PEI dosage, an excess of pDNA itself can also contribute to cellular toxicity[44, 45]. There are pieces of evidence showing that regardless of the transfection method, the concentration of nucleic acid is also critical for enhancing efficiency and minimizing toxicity[44, 46]. For example, Juckem et al. suggested through empirical testing that an optimal starting point is one microgram per well of a 12-well plate resulting in maximum protein expression and minimal toxicity[44]. However, previous studies reported different efficient amounts of pDNA in transfection procedures, partly due to the different sensitivity of diverse cell types to the dosage and concentration of exogenous nucleic acid[44]. In addition, we found that a lower concentration of PEI (dissolved in a low ionic solvent such as distilled water) could prominently increase both transfection efficiency and cell viability compared to concentrated PEI. Besides the lesser toxic effect of PEI at lower concentrations[47], low-concentrated branched PEI have different formation and hydration properties, facilitating transport and diffusion of the polyplexes through cell membrane[48]. Moreover, we hypothesized that diluted PEI might provide more free branches than a concentrated state, advancing DNA binding affinity and improving cellular uptake[49]. In addition, diluted PEI might form a thicker layer around pDNA than concentrated PEI, which enhances the protonation state of PEI/pDNA polyplex structure and provides an electrostatic and physical barrier to DNases with cationic surfaces [24].
Our results also represented that adding PEI to pDNA can increase transfection efficiency relative to the converse order related in part to the electrostatic interactions between the positive charges of cationic polymers and negative charges of pDNA [50]. Therefore, even subtle changes in the sequence of adding reagents may influence the physicochemical properties and reactivity of reagents and subsequently change the transfection efficiency of the polyplexes[51–53].
Furthermore, we found that the presence of serum in polyplexes significantly decreased transfection efficiency due to the interfering role of serum in dissociating and disrupting the structure of PEI/pDNA complexes [18]. However, the MTT test showed that the existence of FBS prominently increased the cell viability, which may be attributed to the potential of serum proteins in absorbing free PEIs[18]. Previous studies also confirmed that the presence of FBS in polycations-based transfection reduces both cytotoxicity and transfection efficiency[54]. Therefore, in the current study, using FBS during polyplex formation was not favored by transfection efficiency and is not recommended in HDF cell type.
Apart from the impact of FBS in the structure of polyplexes, it is evident that FBS in cell culture media before transfection can affect the efficiency of PEI/pDNA delivery to the cells. Serum starvation induces cell cycle synchronization (G0/G1 proliferation delay), which eventually increases transfection efficiency[55, 56]. Moreover, the competitive interaction of serum proteins and polyplexes with cell surface reduce the efficiency of transfection. Herein, we examined the impact of 1h vs. 2h serum starvation on HDFs' transfection efficiency and cell viability. We discovered that both groups showed the same transfection efficiency. Yet 2h serum starvation decreased cell viability in comparison to 1h starvation. Nevertheless, the optimized serum starvation period differs in other studies depending on the studied cell types[57].
Given together, our optimized protocol consists of 1µg pDNA, diluted PEI (40 µg/ml), PEI to pDNA addition sequence, serum-free polyplexes, and 1h serum starvation of HDFs. Besides, since N/P ratio plays an essential role in influencing the degree of complexation, particle diameter, transfection efficiency, and cytotoxicity of carriers[34], in the next step, we investigated the effect of N/P ratio on transfection efficiency and cell viability using the optimized protocol. We observed that the PEI/pDNA ratio of 3:1 led to the highest transfection efficiency and cell viability in comparison to other higher N/P ratios representing the acceptable strength of low-dose PEI in successfully carrying a large-sized plasmid (~ 20kb) into HDFs.
We hypothesized two assumptions regarding this finding. Our first assumption is related to pDNA binding ability of PEI and particle aggregation. The pDNA dosage for all N/P ratios has been fixed (1µg) while the amount of PEI has been raised for increasing N/P ratio. We assumed that due to higher amounts of PEI and more positive charge in respect to the negative charge of pDNA, the pDNA binding ability of PEI has been increased at higher N/P ratios. Thus, despite the potential of higher buffering capacity at higher N/P ratios, extra condensing of pDNA has been occurred, preventing the easy release of pDNA into cytoplasm of cells[58]. Moreover, at higher N/P ratios, polyplexes might have more challenging cellular uptake due to greater sizes, and more particle aggregation [58, 59]. In line with our finding, Cheraghi et al. also assessed the effect of different PEI/ DNA (firefly luciferase) N/P ratios of 3, 6, 12, 18, 24 on transfection efficiency of MCF7 and BT 474 Cells. They observed a decrease (approximately more than 5 fold) in transfection efficiency from N/P ratio of 12 to 18 [58]. They also interpreted the occurrence of this difference to be as a result of higher particle aggregation, and consequently more challenging cellular uptake at higher N/P ratios. In addition, pDNA size may also have a role in either the mechanism of pDNA release from cationic reagent or intracellular migration of pDNA from the cytoplasm to nucleus[60]. In our study plasmid size was approximately high (~ 20kb), adding further challenges regarding particle aggregation at higher N/P ratios[61].
Our second assumption is related to the observed relative higher cytotoxic effect of PEI at higher N/P ratios in the current study. We assumed that even though the transfection might occur for a substantial number of cells at N/P ratios greater than 3 (e.g. 4, 6, 8, 10, and 12), but due to higher toxic effects of polyplexes at higher N/P ratios, the number of dead cells had been also relatively increased. Therefore, when we washed the transfected cells for flow cytometry analysis, dead cells (even transfected ones) were removed and washed from the plate and subsequently did not enter the flow cytometry analysis.
The maximum efficacy at a low-dose N/P ratio found in this study differs from previous studies, which used higher polyplex N/P ratios to improve the transfection efficiency even at the cost of cell viability[18, 34, 58, 62]. For example, Cheraghi et.al examined the effect of different N/P ratios (3, 6, 12, 18, 24) on transfection efficiency and cell viability of MCF7 and BT 474 Cells[58]. They observed significant cytotoxicity of polyplexes in N/P ratios greater than 3. They reported that according to Dynamic Light Scattering (DLS) results, by increasing N/P ratio, polyplexes showed more positive charges, which in turn produced more transfection efficiency and less cell viability. They postulated that among ratios of 3, 6, 12, 18, 24, N/P ratio of 12 could establish an optimized ratio between transfection efficiency and cytotoxicity of PEI/plasmid nanoparticles. In another investigation, Zhang et al. assessed the effect of extracellular nanovesicles (EVs) in PEI-mediated transfection of GFP-encoding plasmid into HEK293T cells, A549 cells, and in the zebrafish embryos[62]. They suggested a very low transfection efficiency for PEI25K and PEI60K at N/P ratios under 32. Due to the negative charge of both EV and DNA, the authors suggested that by increasing N/P ratio higher than 32, zeta potential (surface charge of polyplexes) increases, and subsequently, transfection efficiency will enhance. However, to achieve an acceptable transfection efficiency, they also observed an unavoidable decrease in cell viability by increasing N/P ratio.
We concluded that there could be two possible explanations regarding our finding of acceptable etransfection efficiency in a low N/P ratio. First, our optimization led to significant transfection even at a low N/P ratio of 3. In line with our results, evidence shows that the complex formation protocol can strongly influence transfection efficiency and cytotoxicity[33]. In other words, reagent/DNA ratio and complex formation conditions is of paramount importance for obtaining maximum efficacy with the least side effects in cell transfection process [63]. Second, in the current study, we used low ionic solvent to dissolve PEI. Evidence have shown that PEI/pDNA complexes in high ionic conditions (e.g. HEPES buffered 150 mm saline (HBS)) exhibited a strong tendency for aggregation at low N/P ratios, which can be avoided by increasing the N/P ratio to 6 or higher[33]. While polyplexes in low ionic solvents represented an opposite behavior[33]. Therefore, by administrating the optimized conditions and N/P ratio which we presented in the current study, we circumvented issues with either transfection success or high cytotoxicity.
Finally, PEI polyplexes containing oriP/EBNA1 vector were used to generate iPSCs from HDFs through the optimized condition and N/P ratio. In this study, the efficiency of colony formation was about 2 colonies per 150000 input cells. Although efficiency was low, it is still sufficient to establish iPSC lines from HDFs [38]. In the same vein, it has been previously denoted that despite the development of numerous methods of introducing the reprogramming factors into the somatic cells, only a small percentage of cells may completely express the pluripotency factors and end up the road toward pluripotency state[64]. Our colonies exhibited positive AP staining indicating that iPSC colonies had been successfully developed.
Moreover, immunofluorescence staining and real-time PCR analysis demonstrated the presence of pluripotency-associated proteins including OCT4, cMYC, SOX2, KLF4 proteins and mRNAs including nanog, sox2, oct4. Successively, it was shown that cells could differentiate into different cell types from the three germ layers. The obtained hiPSCs showed a normal karyotype of 46 XY as demonstrated using chromosomal G-band analysis. These results suggest that hiPSC lines were successfully generated using our optimized PEI25K protocol.
It is noteworthy that other groups demonstrated other multiple transfection methods and reagents such as nucleofection[38], Lipotransfection[65], and FuGENE6/HD[25] for transfecting human fibroblasts to generate hiPSCs administrating oriP/EBNA1 or episomal vectors. Yu et al. described the derivation of hiPSCs using oriP/EBNA1 vectors via nucleofection[38]. Similar to our finding, they reported a low reprogramming efficiency (3–6 colonies/1000000 input cells). Wang et. al also generated hiPSCs via nucleofection and Binary Colloidal Crystals (BCC) as a feeder free system[66]. Although the authors reported a facilitating role of BCC on reprogramming process and emerging fully reprogrammed hiPSCs colonies, the nucleofection technique requires relatively expensive equipment and reagents[67]. Moreover, Skrzypczyk et al. successfully transferred oriP/EBNA1 vectors into human foreskin fibroblasts through lipotransfection and generated hiPSCs[65]. However, despite the common administration of lipotransfection in the generation of iPSCs, there is evidence showing that lipofection has its own adverse effects on transfected cells[68]. For example, evidence has been demonstrated that lipid-based methods have been linked with causing cell cycle arrest, proinflammatory reactions, and increased apoptosis[68]. In addition, the overall manufacturing cost of cell transfection is also a crucial factor for determining the transfection method[63]. Soe et al. compared three transfection reagents, including lipofectamine 2000, TransIT-PRO, and linear 25KD PEI, regarding cost-effectiveness and efficacy of transient expression of enhanced GFP in Chinese hamster ovary cells[63]. They observed a higher number of mRNA copies per viable cell and higher rProtein titre in PEI than lipofectamine group. They also stated that the cost of lipofectamine reagent would be the highest followed by TransIT-PRO and then PEI. Two grams of PEI supplies enough reagent for 2 Liter solution or 20.000 reactions (6 well plate)[41]. For the same price, a very lower number of reactions (~ 60) can be performed, if expensive reagents such as Lipofectamine are used. In fact, using Lipofectamine imposes much more cost on labs. So, PEI could serve as a cost-effective alternative for large-scale generation of iPSCs which can be a serious limiting factor for labs facing financial problems.
Despite different valuable investigations on PEI properties, to our knowledge, there was no study using PEI25K as a successful carrier for transfecting human fibroblast using oriP/EBNA1 vectors for obtaining hiPSCs. Herein, we introduce a cost-effective and straightforward method to use PEI/pDNA polyplexes in generating iPSCs. In this protocol, we overcome the cytotoxicity of PEI-mediated complexes via optimizing transfection conditions and minimizing PEI/pDNA ratio. Our method can be greatly useful in research projects, especially for labs with financial issues. Availability and Simplicity use of PEI reagents makes it a suitable candidate for large-scale transfection approaches.