High-Efficient Transfection of Human Embryonic Stem Cells by Single-Cell Plating and Starvation

Abstract

Nowadays, the low efficiency of small interfering RNA (siRNA) or plasmid DNA (pDNA) transfection is a critical issue in genetic manipulation of human embryonic stem (hES) cells. Development of an efficient transfection method for delivery of siRNAs and plasmids into hES cells becomes more and more imperative. In this study, we tried to modify the traditional transfection protocol by introducing two crucial processes, single-cell plating and starvation, to increase the transfection efficiency in hES cells. Furthermore, we comparatively examined the transfection efficiency of some commercially available siRNA or pDNA transfection reagents in hES cells. Our results showed that the new developed method markedly enhanced the transfection efficiency without influencing the proliferation and pluripotency of hES cells. Lipofectamine RNAiMAX exhibited much higher siRNA transfection efficiency than the other reagents, and FuGENE HD was identified as the best suitable reagent for efficient pDNA transfection of hES cells among the tested reagents.

Introduction

Since human embryonic stem cells (hES cells) were successfully isolated from the inner cell mass of human blastocysts in 1998 [1], they have been widely used in fundamental research, mechanistic studies about various diseases, and transplantation medicine because of their properties of pluripotency and potentially unlimited capacity for self-renewal [2,3]. Genetic manipulation of hES cells such as transient or stable transgene expression and siRNA-based gene knockdown is an attractive strategy to treat genetic and degenerative diseases, providing both a powerful tool for genetic research and a key element to realize the full potential of hES cells [4 –7]. However, this strategy, which was most widely used to manipulate different mammalian cells, has worked poorly in hES cells partly because of the notoriously low efficiency of gene transfection.

Generally, the uptake of exogenous nucleic acids by the cell in the case of transfection reagent occurs through cell endocytosis or through forming complexes of nucleic acids/reagent and generating vesicles that are in some ways similar to the structure of a cell and can fuse with the cell membrane, followed by release of the exogenous nucleic acids into the cell [8]. A good strategy to increase transfection efficiency would be to expose single cells to transfection reagents as best as possible. hES cells can be passaged as single cells using different enzymatic solutions such as Trypsin-EDTA, TrypLE, Versene solution, and Accutase depending on the application [9 –11]. We have done extensive screening in our laboratory with Dispase, Versene solution, Trypsin-EDTA, EDTA/PBS, and Accutase and observed that Accutase performed exceptionally well in preparation of single-cell suspension and maintained high cell viability. Accutase was often recommended for single-cell passaging of hES cells in most studies [9,12]. Accutase-treated hES cells can be passaged as single cells and propagated as small monolayer clusters, which significantly improved the transfection efficiency [13]. Additionally, it was reported that Y-27632, a selective inhibitor of p160-Rho-associated coiled-coil kinase (ROCK) [14,15], could markedly diminish dissociation-induced apoptosis of hES cells, enhance colony formation of dissociated hES cells after passaging, and facilitate subcloning after gene transfer [10,11,16 –20]. Recent research shows that treatment of hES cells with Y-27632 before and during transfections can effectively enhance nonviral gene delivery to hES cells [21].

Several reports have demonstrated that performing transfection in medium without serum could improve transient transfection efficiency [22]. The cationic nanoparticle can be taken up by cells through endocytosis, which will be greatly reduced in the presence of serum instead [23,24]. In the earlier work, we found that the three-dimensional shapes of clones became more prominent, the space among cells increased, and the morphology of most single cells distinctly elongated after starvation of hES cells for several hours without influencing the proliferation and pluripotency of cells. Serum-reduced minimal essential medium (Opti-MEM) that allows for a reduction of serum supplementation by at least 50% with no change in growth rate is often recommended for use in gene transfection. Combination of the use of Y-27632 and starvation using Opti-MEM before transfection will not only increase the cell–cell spacing obviously but will also provide the beneficial conditions for serum-free transfection. Additionally, we found that the auxiliary Dulbecco's phosphate-buffered saline (DPBS) treatment before starvation could achieve desired results more readily.

In this study, to increase the transfection efficiency of small interfering RNA (siRNA) or plasmid DNA (pDNA) into hES cells with chemical material-based transfection, we introduced modifications to the traditional transfection protocols. The crucial steps in this modified protocol were single-cell plating and starvation before transfection in the presence of Y-27632. To study the effect of the single-cell plating and starvation on the efficiency of transfection, we used five different methods of siRNA or pDNA transfection of H9 hES cells (Fig. 1). Both siRNAs and pDNAs (encoding enhanced green fluorescent protein, EGFP) transfections were done in this study. Our results showed that the new developed method that combined single-cell plating with the treatment of starvation before transfection markedly enhanced the efficiency of siRNA and pDNA transfection without influencing the proliferation and pluripotency of hES cells.

FIG. 1.

Scheme of different transfection methods used in this study group I: single-cell plating and starvation before transfection (+/+) (as a new method); group II: single-cell plating was performed and cells were cultured normally in conditioned medium instead of starvation before transfection (+/−); group III: cells were cultured conventionally on a Matrigel-coated six-well plate, passaged, and plated as clones using Collagenase without the treatment of Accutase and ROCK inhibitors, while starvation was performed before transfection (−/+); group IV served as the traditional method control: cell clone plating was used and cells were cultured normally before transfection (−/−); group V: reverse transfection method: the siRNA/Reagent or pDNA/Reagent complexes were prepared inside the wells, after which single-cell suspensions were added (+/RT) (for more information, see the Materials and Methods section).

Recently, more and more transfection reagents have been extensively used in transfection experiments and have displayed distinct and specific affinities for a variety of different target cells. However, little is known about the effectiveness of these reagents in mediating transfection of siRNA or pDNA into hES cells. Therefore, we further comparatively examined the transfection efficiency of eight commercially available siRNA transfection reagents and eight DNA transfection reagents in both H9 and H1 hES cells. Results showed that the RNAi efficiency of Lipofectamine RNAiMAX was significantly higher than the other seven siRNA transfection reagents. Among the eight DNA transfection reagents tested, FuGENE HD showed the best efficiency of pDNA transfection under the same conditions.

Materials and Methods

Transfection reagents, siRNAs, and pDNAs

The used siRNA transfection reagents include Lipofectamine RNAiMAX (Invitrogen), Oligofectamine Reagent (Invitrogen), Lipofectamine 3000 Reagent (Invitrogen), X-tremeGENE siRNA Transfection Reagent (Roche), HiPerFect Transfection Reagent (Qiagen), Dharma FECT 1 Transfection Reagent (Thermo Fisher), Entranster-R RNA Transfection Reagent (Engreen), and N-TER Nanoparticle siRNA Transfection System (Sigma).

The used pDNA transfection reagents include Lipofectamine 2000 Reagent (Invitrogen), Lipofectamine 3000 Reagent (Invitrogen), Lipofectamine LTX and PLUS Reagent (Invitrogen), FuGENE HD Transfection Reagent (Promega), TurboFect Transfection Reagent (Thermo Fisher), Attractene Transfection Reagent (Qiagen), Neofect DNA Transfection Reagent (Neofect), and LipoFiter Liposomal Transfection Reagent (Hanbio).

Double-stranded siRNAs corresponding to Oct4 (siOct4) were designed with the following sense and antisense sequence and were synthesized by RiboBio:

Sense (5′-3′) GGAUGUGGUCCGAGUGUGGU dTdT

Antisense (5′-3′) ACCACACUCGGACCACAUCC dTdT [25]

siR-Ribo eGFP (siEGFP) and siR-Ribo Negative Control (siControl) were purchased from RiboBio. The sequence of siControl was described as universal negative control designed for use in human, mouse, or rat cells.

Two different EGFP reporter plasmids were used for analysis of the transfection efficiency of pDNA: pEGFP-C1 expression vector (Clontech) and pSUPER.neo+GFP RNAi vector (Oligoengine).

hES cell culture

Undifferentiated H9 and H1 hES cells were routinely cultured and passaged as clumps, as described in our published article [26–27]. The presence of the feeder layer complicates mRNA or protein-level analysis, and also limits the transfection efficiency. So, hES cells were passaged to Matrigel-coated plates under feeder-free conditions in MEF-conditioned medium (CM) to maintain pluripotency as described previously before transfection [26 –28]. The hES cells cultured on Matrigel-coated plates were passaged as single cells using Accutase (Millipore).

Single-cell passaging and plating

The moderate-sized exponentially growing cell clones, which have been maintained for one to three passages on Matrigel-coated plates, were used for transfection. After 2 h of pretreatment with Y-27632 (Millipore) at a working concentration of 10 μM [29], the cells were carefully dissociated into single-cell suspension using Accutase at 37°C for 5 min, and then were collected by centrifugation (800 rpm for 5 min), resuspended in CM supplemented with 10 μM Y-27632, and used directly for reverse transfection (transfection complexes were prepared in multiwell plates, after which cells and medium were added) at a density of 1 × 10⁶ cells per well or replated into wells of a six-well plate (coated with Matrigel) for forward transfection (cells were first plated in multiwell plates, and transfection complexes were generally prepared and added the next day) at suggested cell density (eg, lipofectamine RNAiMAX: 2 × 10⁵ cells per well, FuGENE HD Transfection Reagent: 3 × 10⁵ cells per well).

Immunocytochemistry, alkaline phosphatase staining, and karyotype analysis

Immunocytochemistry was performed as previously described with minor modifications [30,31]. Fixed cells were permeabilized with 0.3% Triton X-100 and stained with antibodies against Nanog (1:500; Millipore), Oct4 (1:300; Abcam), Sox2 (1:300; Millipore), SSEA4 (1:50; Abcam), and TRA-1-60 (1:200; Abcam). The following secondary antibodies were used: Alexa Fluor 594-labeled donkey anti-mouse IgG (H+L) antibody (1:500; Invitrogen) or Alexa Fluor 488-labeled donkey anti-rabbit IgG (H+L) antibody (1:500; Invitrogen). Nuclei were stained with Hoechst 33342 (1:2,000; Invitrogen). Fluorescence signals were detected with a Nikon Eclipse TE2000-S Inverted Microscope (Nikon). For alkaline phosphatase (AP) staining, hES cells were stained using SIGMAFAST BCIP/NBT tablet (Sigma) according to the manufacturer's protocols. For karyotype analysis, cells were treated with 0.1 μg/mL of colcemid (Gibco) at 37°C for 2 h. The cells were then trypsinized, resuspended, incubated in 0.075 M potassium chloride for 15 min at 37°C, and fixed with 3:1 methanol:acetic acid, and then dropped onto slides to spread the chromosomes. The chromosomes were visualized by Giemsa staining.

Establishment of stable hES cell line expressing EGFP protein

Accutase-digested single hES cells (H9 and H1) produced by the above described method were replated into wells of a six-well plate (coated with Matrigel) at a density of 3 × 10⁵ cells per well. The cells were transfected with a complex consisting of FuGENE HD and pEGFP-C1 vector according to the manufacturer's protocol. Stable EGFP-positive transformants (named as H9-EGFP and H1-EGFP) were selected with CM containing G418 (Sigma) (300 μg/mL), and then further isolated and continuously cultured in wells of a six-well plate (coated with Matrigel) with CM in the presence of G418.

siRNA/pDNA transfection of hES cells

As shown in Fig. 1, a new method was applied for group I (+/+). First, 1 day before transfection, hES cell clones (H9 and H1, grown on Matrigel) should be digested with Accutase and replated into wells of a six-well plate (coated with Matrigel) as single cells using the protocol described above. About 3–6 h before transfection, the cells were incubated with 1 mL DPBS for 5 min to increase cell–cell spacing, and then these cells were cultured in reduced serum medium Opti-MEM I (GIBCO) supplemented with 10 μM Y-27632 instead of CM for starvation. At the end of starvation, transfection complexes (siRNA/Lipofectamine RNAiMAX or pDNA/FuGENE HD complexes) were prepared and added to each well containing cells according to the manufacturer's protocol. The medium may be changed after 6–8 h. For siRNA transfection, cells were often transfected again after 24 h using the same protocol.

Other methods used for group II (+/−) were as follows: single-cell plating was performed and cells were cultured normally in CM instead of starvation before transfection. As for group III (−/+), cells were cultured conventionally in wells of a six-well plate (coated with Matrigel), passaged, and plated as clones using Collagenase without the treatment of Accutase and Y-27632, while they were treated with starvation before transfection. Group IV (−/−) served as the control or cell clone plating. In this group, cells were cultured normally before transfection. These methods used most of the same protocols as the new method except for different treatments mentioned above (Fig. 1). The reverse transfection method (single-cell suspension transfection) (group V; +/RT) was applied as follows: transfection complexes were prepared inside the Matrigel-coated wells according to the manufacturer's protocol, after which single-cell suspension produced by the above described method was added, followed by incubation at 37°C in a CO₂ incubator overnight (Fig. 1).

RNA isolation, reverse transcription, and quantitative real-time PCR

Total RNA was isolated with TRIzol reagent (Invitrogen), followed by treatment with deoxyribonuclease I (DNase I; Promega) to remove any contaminating genomic DNA. The first-strand cDNA was synthesized using the RevertAid First-Strand cDNA Synthesis Kit (Fermentas). Quantitative real-time PCR (qPCR) analysis was carried out using FastStart Essential DNA Green Master (Roche) with MJ Mini Personal Thermal Cycler (BioRad). Primer sequences for qPCR are shown in Table 1.

Table 1.

Primers Used in Quantitative Real-Time Polymerase Chain Reaction

Genes	Sequences (5′-3′) forward	Sequences (3′-5′) reverse
GAPDH	CCAGCAAGAGCACAAGAGGAA	ATGGTACATGACAAGGTGCGG
Oct4	GGCTCTCCCATGCATTCAAAC	CCAAAAACCCTGGCACAAACT
EGFP	AAGCTGACCCTGAAGTTCATCTGC	CTTGTAGTTGCCGTCGTCCTTGAA

Protein extraction and western blot analysis

Protein extraction and western blot analysis were performed as previously described [30]. The following antibodies were used in this study: anti-Oct4 antibody (1:500; Millipore), anti-EGFP antibody (1:20,000; Proteintech), anti-β-actin antibody (1:5,000; Sigma), and horseradish peroxidase-conjugated anti-mouse IgG antibody (1:2,000; Millipore). Images were recorded using the Molecular Imager ChemiDoc XRS+System (BioRad) and analyzed with Quantity One 1-D analysis software (BioRad).

Fluorescence microscopy and flow cytometric analysis

The morphology and the fluorescence intensity of hES cells or hES-EGFP colonies were observed and analyzed by fluorescence microscopy. For flow cytometric analysis, hES cells were washed with PBS and carefully dissociated into single-cell suspension using Accutase, followed by several washes with PBS. The single cells were resuspended in media, and then analyzed for EGFP fluorescence by FACSCanto II (Becton Dickinson).

Cytotoxicity assay

The cytotoxicity assay was examined in H9 hES cells by methyl thiazolyl tetrazolium (MTT) assay. Briefly, H9 hES cells were seeded as single cells in wells of a 96-well plate (coated with Matrigel) and transfected with siRNA (siControl)/reagent or pDNA (pEGFP-C1)/reagent complexes at 24 h. After 48 h in culture, the transfected cells were incubated in medium containing 0.45 mg/mL MTT (Sigma) at 37°C for 4 h. Supernatant was then removed and dimethyl sulfoxide (DMSO; Sigma) was added to each well to dissolve the crystals. Absorbance was measured by PARADIGM Detection Platform (Beckman Coulter) at the wavelength of 570 nm.

Statistical analysis

All in vitro experiments were independently carried out at least thrice. The mean value ± standard deviation was calculated using Microsoft Excel and analyzed by one-way ANOVA along with Bonferroni's multiple comparison test (Software: IBM SPSS Statistics 19). P value < 0.05 was considered to be statistically significant.

Results

Pluripotency of hES cells after single-cell plating and starvation

Undifferentiated H9 and H1 hES cells with high quality were used in this study. They were routinely maintained on feeder layer (Fig. 2A) and passaged by mechanical dissociation using collagenase for optimal maintenance of genetic stability. To provide optimal conditions for highly efficient gene transfection in hES cells, we transferred cells to feeder-free conditions on Matrigel (Fig. 2A) and maintained for one to three passages to make the cells adapted to Matrigel in CM. Under these conditions, the feeder cells were completely removed from the sample preparation, and hES cells were maintained as loose clones that were easily dissociated into single cells with the use of Accutase (Fig. 2A). Performing starvation in Opti-MEM supplemented with Y-27632 before transfection in combination with the auxiliary treatment of DPBS did increase the cell–cell spacing obviously without influencing the proliferation of hES cells (Fig. 2B, C).

FIG. 2.

Observation of cell morphology and growth state of human embryonic stem (hES) cells grown on feeder layer or Matrigel-coated plates. (A) Undifferentiated H9 and H1 hES cells were conventionally cultured on feeder or Matrigel; (B) the colony size and cell morphology of normally cultured H9 hES cells at 0 h, 30 h, and 96 h after cell plating; (C) the colony size and cell morphology of H9 hES cells treated by the new method (single-cell plating and starvation) at 0 h, 30 h, and 96 h after cell plating. Bars = 100 μm.

To confirm that the hES cells remained in an undifferentiated state after single-cell plating and starvation, we examined five key markers (Nanog, Oct4, Sox2, SSEA4, and TRA-1-60) and the AP activity. Immunostaining for these key markers revealed that a high proportion of hES cells after single-cell plating and starvation were positive for the markers, Nanog, Oct4, Sox2, SSEA4, and TRA-1-60 (Fig. 3A). Similar to cells cultured routinely, AP staining showed that almost all hES cells cultured after single-cell plating and starvation were positive for AP (Fig. 3B). Moreover, to verify the genetic stability of hES cells after the treatment of single-cell plating and starvation, the stable EGFP-positive hES cell line (H9-EGFP cells) generated by our new developed transfection method was karyotyped at passage 10. These cells exhibited a normal female 46, XX karyotype (Fig. 3C).

FIG. 3.

New method-treated hES cells express key markers of pluripotentcy. (A) 96 h after single-cell plating, the H9 hES cells were treated by starvation (about 3–6 h before transfection), and then stained with antibodies to Nanog (first row), Oct4 (second row), Sox2 (third row), SSEA4 (fourth row), and TRA-1-60 (fifth row). A high proportion of hES cells were positive for Nanog, Oct4, Sox2, SSEA4, and TRA-1-60. Cell nuclei were stained by Hoechst 33342. Bars = 100 μm. (B) hES cell colonies treated by our new method were stained using AKP. Images were taken at 100× (left) and 40× (right) magnifications, respectively. Bars = 100 μm. (C) Karyogram of H9-EGFP cells, indicating a normal female 46, XX karyotype.

Efficient transfection of siRNA into hES cells

H9 hES cells were transiently transfected with siOct4 by Lipofectamine RNAiMAX using five different methods and were collected and analyzed 48 h (the RNA level) or 72 h (the protein level) after siRNA double transfection (Figs. 1 and 4). As shown in Fig. 4B and C, single-cell plating apparently enhanced transfection efficiency as the transfection efficiency of single cell-plated groups (+/+ and +/−) was higher than cell clone-plated groups (−/+ and −/−) with or without the treatment of starvation (P < 0.05) (Table 2). Moreover, starvation before transfection could also enhance the transfection efficiency of hES cells whether plated as single cells (P < 0.001, +/+ vs. +/−) or clones (−/+ vs. −/−). Nevertheless, our results clearly showed that single-cell plating combined with the treatment of starvation before transfection (+/+) presented an extra additive effect, this new method (+/+) enhanced the efficiency of transfection significantly and resulted in a 96% ± 2% inhibition of Oct4 gene expression at the RNA level and 95% ± 5% at the protein level, while using the traditional method (−/−), the inhibition was only 67% ± 3% at the RNA level and about 73% ± 4% at the protein level (P < 0.01) (Table 2). The inhibition of Oct4 gene expression at the RNA/protein level was 83% ± 4%/81% ± 3% (+/−) and 73% ± 2%/74% ± 6% (−/+), respectively, and the efficiency of these two methods was significantly lower than that of the new method (P < 0.05). Although the reverse transfection method (+/RT) obtained about 91% ± 1%/88% ± 6% efficiency in transient silencing of Oct4 gene at the RNA/protein level, it has an obvious disadvantage that the state of hES cells could be severely affected and more than half of the transfected cells would die, especially after double transfection (Fig. 4A). In addition, our results showed that the hES cells exhibited similar morphology changes to those reported in previous literature when the expression of Oct4 was inhibited [25,32], while the hES cells displayed no obvious morphology changes when treated with negative control siRNA (Fig. 4A).

FIG. 4.

Specific Knockdown of Oct4 in hES cells using the new transfection method (+/+) or other methods (+/−, −/+, −/−, +/RT). (A) Morphology of H9 hES cells after siRNA transfection. The visible morphologic changes of cells following Oct4 RNAi using five different transfection methods were observed. Cell morphology was recorded at 48 h after double siRNA transfection. Control group indicates H9 hES cells treated with negative control siRNA. Bar = 100 μm. (B) Oct4 mRNA expression assessed by qPCR in hES cells transfected with siOct4 using five different transfection methods. Relative mRNA levels are shown after normalization against GAPDH mRNA. mRNA levels of Oct4 in cells transfected with siControl were arbitrarily set as 1 (100%). Error bars represent mean ± SD (n = 3). Asterisks represent significant differences between groups: ***P<0.001. (C) Oct4 protein expression assessed by western blot in hES cells transfected with siOct4 using five different transfection methods. Protein levels of Oct4 in cells transfected with siControl were set as 100%. The gray scale of protein bands was analyzed using Bio-Rad Quantity One software and calculated after normalization against β-actin loading control. Percentages at right indicate the knockdown efficiencies of Oct4 protein. The experiments were carried out in triplicate independently. ^# +/RT is the single-cell suspension reverse transfection method.

Table 2.

Analysis of RNAi Efficiency of hES Cells Transfected with siRNA by Different Transfection Methods

	RNAi efficiency of Oct4 gene (%)		RNAi efficiency of EGFP gene (%)
Transfection methods Single-cell plating/starvation	RNA level	Protein level	RNA level	Protein level
+/+	96 ± 2	95 ± 5	86 ± 2	83 ± 4
+/−	83 ± 4	81 ± 3	66 ± 8	64 ± 7
−/+	73 ± 2	74 ± 6	31 ± 9	32 ± 7
−/−	67 ± 3	73 ± 4	26 ± 6	21 ± 5
+/RT^#	91 ± 1	88 ± 6	71 ± 7	68 ± 6

Data are shown as mean ± SD.

+/RT is the single-cell suspension reverse transfection method.

hES, human embryonic stem; siRNA, small interfering RNA; EGFP, enhanced green fluorescent protein.

The H9-EGFP cell line stably expressing EGFP was first established to investigate whether the exogenous gene in hES cells could be equally and effectively downregulated by RNAi using the new method (Fig. 5A, B). Subsequently, H9-EGFP cells were transfected with siEGFP using five different transfection methods that were used to knock down Oct4 (Fig. 5C, D). The results showed that the effect of single-cell plating and starvation on the silencing efficiency of EGFP transgene was more obvious (Table 2). EGFP mRNA and protein levels were reduced by 86% ± 2% and 83% ± 4%, respectively, using the new method (+/+), while the inhibition rates of EGFP mRNA and protein were only 26% ± 6% and 21% ± 5%, respectively, using the traditional method (−/−) (P < 0.001). Meanwhile, for other methods, EGFP mRNA/protein levels were reduced by 66% ± 8%/64% ± 7% (+/−; P < 0.05, vs. +/+), 31% ± 9%/32% ± 7% (−/+; P < 0.001, vs. +/+), and 71% ± 7%/68% ± 6% (+/RT), respectively (Table 2). In conclusion, the new method enhanced the efficiency of transfection and achieved the highest RNA interference of endogenous and exogenous genes in hES cells.

FIG. 5.

Specific knockdown of enhanced green fluorescent protein (EGFP) in hES cells using the new transfection method (+/+) or other methods (+/−, −/+, −/−, +/RT). (A, B) H9 hES cells stably expressing EGFP were used to evaluate the interference effect of siEGFP. (A) Bright-field images (left) and fluorescence images (right) of H9-EGFP cell colonies. Bar = 100 μm. (B) The expression of EGFP was observed both at mRNA (left) and protein levels (right). (C) EGFP mRNA expression assessed by qPCR in H9-EGFP cells transfected with siEGFP using five different transfection methods. Relative mRNA levels are shown after normalization against GAPDH. mRNA levels in cells transfected with siControl were set as 1 (100%). Error bars represent mean ± SD (n = 3). Asterisks represent significant differences between groups: *P < 0.05 and ***P < 0.001. (D) EGFP protein expression assessed by western blot in H9-EGFP cells transfected with siEGFP using five different transfection methods. Protein levels of EGFP in cells transfected with siControl were set as 100%. The gray scale of protein bands was analyzed using Bio-Rad Quantity One software and calculated after normalization against β-actin loading control. Percentages at right indicate knockdown efficiencies of EGFP protein. The experiments were carried out in triplicate independently.^# +/RT is the single-cell suspension reverse transfection method.

Efficient transfection of pDNA into hES cells

H9 hES cells were transfected with two pDNA vectors (pEGFP-C1 and pSUPER.neo+GFP) encoding EGFP by FuGENE HD Transfection Reagent using the various methods to test whether the new method could also enhance the efficiency of pDNA transfection. At 36 h after pDNA transfection, cells were collected and analyzed by flow cytometry. As shown in Fig. 6B and C, when transfected with pEGFP-C1 and pSUPER.neo+GFP using the traditional method (−/−), 13.8% ± 6% and 6.7% ± 1.3% of cells were identified as EGFP positive, respectively (Table 3). In contrast, the new method (+/+) significantly increased the percent of EGFP-positive cells to 60.4% ± 3.2% for pEGFP-C1 transfection and 50.2% ± 3.7% for pSUPER.neo+GFP transfection (P < 0.001). Meanwhile, the percentage of GFP-positive cells obtained in other transfected conditions was significantly lower for both GFP reporter plasmids compared with the new method (P < 0.05) (Table 3). The expression of EGFP in H9 hES cells was also observed by an inverted fluorescence microscope (Fig. 6A). Clearly, single-cell plating combined with the treatment of starvation before transfection did also enhance the efficiency of pDNA transfection significantly. The new method demonstrated superior and highly efficient pDNA transfection in hES cells.

FIG. 6.

EGFP reporter expression in hES cells following transfection with the new method (+/+) or other methods (+/−, −/+, −/−, +/RT). (A) The bright-field (top panels) and fluorescence images (bottom panels) of H9 hES cells successfully transfected with EGFP reporter plasmid (pEGFP-C1) by various gene transfer methods. Bar = 100 μm. (B, C) Results of flow cytometric analysis of H9 hES cells transfected with the GFP reporter plasmid of pEGFP-C1 (B) and pSUPER.neo+GFP (C). Values above the horizontal bars in all graphs indicate the percentage of cells positive for EGFP. Three independent sets of replicate experiments were performed with the same trend observed. Data from one representative experiment are shown. ^# +/RT is the single-cell suspension reverse transfection method.

Table 3.

Measurement of EGFP Expression of hES Cells Transfected with the EGFP Reporter Plasmid by Different Transfection Methods

	GFP-positive cells (%)
Transfection methods Single-cell plating/starvation	pEGFP-C1	pSUPER.neo+GFP
+/+	60.4 ± 3.2	50.2 ± 3.7
+/−	41.6 ± 5.4	43.1 ± 5.2
−/+	20.4 ± 3.2	15.8 ± 1.8
−/−	13.8 ± 6	6.7 ± 1.3
+/RT^#	45.1 ± 7.2	39.5 ± 5.5

Average percentage of GFP-positive cells as measured by flow cytometric analysis (Fig. 6). Data are shown as mean ± SD.

+/RT is the single-cell suspension reverse transfection method.

Comparison of RNAi efficiency using various siRNA transfection reagents

In this part of the study, the eight commercially available siRNA transfection reagents were examined to identify the best suitable reagent for efficient transfection of hES cells using the new transfection method. First, the RNAi efficiency was tested using siRNAs against the Oct4 gene in two different hES cell lines (H9 and H1). As shown in Fig. 7, there was a significant difference between RNAi efficiency of eight different reagents in transfected cells. In particular, RNAiMAX mediated the highest efficiency in silencing of Oct4 genes (P < 0.05), followed by Dharma FECT in both H9 or H1 hES cells. Transfection with X-tremeGENE, HiPerFect, and Lipofectamine 3000 led to moderate efficiency in silencing of Oct4 genes. In contrast, transfection with N-TER Nanoparticle, Entranster-R, and Oligofectamine achieved lower efficiency in silencing of Oct4 genes in both H9 and H1 hES cells. Quantification of the qPCR and western blot results are summarized in Table 4. These results demonstrated that among the eight siRNA transfection reagents tested, RNAiMAX did transduce siOct4 into hES cells with highest efficiency and lead to a dramatic knockdown of Oct4 expression using the new transfection method.

FIG. 7.

Comparison of eight different siRNA transfection reagents used to introduce Oct4 siRNAs into hES cells. (A, C) Comparative analysis of RNAi efficiency of Oct4 gene by qPCR at the RNA level in H9 (A) and H1 (C) hES cells transfected with eight different transfection reagents at 48 h after double transfection. Relative mRNA levels are shown after normalization against GAPDH. mRNA levels in cells transfected with siControl were set as 1. Error bars represent mean ± SD (n = 3). Asterisks represent significant differences between groups: *P < 0.05, **P < 0.01, and ***P < 0.001. (B, D) Comparative analysis of RNAi efficiency of Oct4 gene by western blot at the protein level in H9 (B) and H1 (D) hES cells transfected with eight different transfection reagents at 72 h after double transfection. Protein levels of Oct4 in cells transfected with siControl were set as 100%. The gray scale of protein bands was analyzed using Bio-Rad Quantity One software and calculated after normalization against β-actin loading control. Percentages at right indicate knockdown efficiencies of Oct4 protein. Three independent sets of replicate experiments were performed with the same trend observed. Data from one representative experiment are shown.

Table 4.

Comparison of RNAi Efficiency of Oct4 Gene in H9 and H1 hES Cells Transfected with siOct4 by Different Transfection Reagents

	RNAi efficiency of Oct4 gene at the RNA level (%)		RNAi efficiency of Oct4 gene at the protein level (%)
Transfection reagents	H9 hES cells	H1 hES cells	H9 hES cells	H1 hES cells
RNAiMAX	95 ± 1	94 ± 1	96 ± 2	93 ± 4
Oligofectamine	57 ± 4	61 ± 2	55 ± 8	60 ± 8
Lipofectamine 3000	76 ± 3	65 ± 8	73 ± 6	73 ± 2
X-tremeGENE	79 ± 2	75 ± 4	75 ± 7	77 ± 4
HiPerFect	79 ± 4	75 ± 2	78 ± 6	76 ± 9
Dharma FECT	83 ± 6	79 ± 3	81 ± 7	85 ± 6
Entranster-R	61 ± 6	61 ± 1	69 ± 8	64 ± 8
N-TER nanoparticle	65 ± 7	72 ± 3	66 ± 6	67 ± 5

Data are shown as mean ± SD.

Subsequently, EGFP-specific siRNA knockdown experiments were also performed in two different hES cell lines (H9-EGFP and H1-EGFP) to further compare the eight siRNA transfection reagents more exactly and reliably. As shown in Fig. 8, very similar results as Oct4-specific siRNA knockdown experiments were obtained. RNAiMAX mediated maximal transfection efficiency, with dramatic reduction of EGFP expression at the RNA and protein levels in both siEGFP-transfected H9- and H1-EGFP cells, significantly higher than Oligofectamine, Lipofectamine 3000, X-tremeGENE, Entranster-R, and N-TER Nanoparticle (P < 0.01), which achieved lower efficiency in silencing of EGFP genes. Transfection with HiPerFect and Dharma FECT led to moderate efficiency in silencing of EGFP genes (Table 5). These results confirmed that RNAiMAX was an efficient transfection reagent for delivering siRNA into hES cells.

FIG. 8.

Comparison of eight different siRNA transfection reagents used to introduce EGFP siRNAs into hES cells expressing EGFP. (A, C) Comparative analysis of RNAi efficiency of EGFP gene by qPCR at the RNA level in H9-EGFP cells (A) and H1-EGFP cells (C) transfected with eight different transfection reagents at 48 h after double transfection. Relative mRNA levels are shown after normalization against GAPDH. mRNA levels in cells transfected with siControl were set as 1. Error bars represent mean ± SD (n = 3). Asterisks represent significant differences between groups: **P < 0.01 and ***P < 0.001. (B, D) Comparative analysis of RNAi efficiency of EGFP gene by western blot at the protein level in H9-EGFP cells (B) and H1-EGFP cells (D) transfected with eight different transfection reagents at 72 h after double transfection. Protein levels of EGFP in cells transfected with siControl were set as 100%. The gray scale of protein bands was analyzed using Bio-Rad Quantity One software and calculated after normalization against β-actin loading control. Percentages at right indicate knockdown efficiencies of EGFP protein. Three independent sets of replicate experiments were performed with the same trend observed. Data from one representative experiment are shown.

Table 5.

Comparison of RNAi Efficiency of EGFP Gene in H9-EGFP Cells and H1-EGFP Cells Transfected with siEGFP by Different Transfection Reagents

	RNAi efficiency of EGFP gene at the RNA level (%)		RNAi efficiency of EGFP gene at the protein level (%)
Transfection reagents	H9-EGFP	H1-EGFP	H9-EGFP	H1-EGFP
RNAiMAX	84 ± 3	83 ± 2	82 ± 3	79 ± 4
Oligofectamine	32 ± 4	38 ± 5	37 ± 7	27 ± 5
Lipofectamine 3000	46 ± 5	50 ± 5	53 ± 6	38 ± 3
X-tremeGENE	55 ± 6	55 ± 6	55 ± 7	60 ± 7
HiPerFect	75 ± 4	59 ± 7	73 ± 6	54 ± 9
Dharma FECT	72 ± 5	70 ± 4	68 ± 4	61 ± 6
Entranster-R	46 ± 4	50 ± 12	57 ± 8	39 ± 8
N-TER Nanoparticle	27 ± 14	12 ± 2	22 ± 3	26 ± 5

Data are shown as mean ± SD.

Comparison of exogenous gene expression using various pDNA transfection reagents

In the study, we also tested eight pDNA transfection reagents to identify which one was best suitable for high-efficiency gene transfection in hES cell lines (H9 and H1) using EGFP as an indicator. At 36 h after pDNA (pEGFP-C1 and pSUPER.neo+GFP) transfection, cells were processed for flow cytometric analysis. As shown in Fig. 9, under the same conditions, eight tested reagents differed widely in the efficiency of pDNA transfection of hES cells. The EGFP expression was highest in FuGENE HD-transfected cells, followed by Lipofectamine 3000 and Lipofectamine LTX-transfected cells. In contrast, low percentage of EGFP-positive cells was observed in other reagent-transfected cells, especially in Attractene-transfected cells (Table 6). Transfection with FuGENE HD resulted in high percentage of EGFP-positive cells for pEGFP-C1 and pSUPER.neo+GFP in both H9 and H1 hES cells, significantly higher than other seven transfection reagents (P < 0.05), demonstrating that this transfection reagent was best suitable for high-efficiency pDNA transfection in hES cells.

FIG. 9.

Flow cytometric analysis of EGFP expression in hES cells transfected with various pDNA transfection reagents. The expression vector, pEGFP-C1 (A), or the RNAi vector, pSUPER.neo+GFP (B), was transfected into H9 and H1 hES cells using eight different pDNA transfection reagents. Values above the horizontal bars in all graphs indicate the percentage of cells positive for EGFP. Three independent sets of replicate experiments were performed with the same trend observed. Data from one representative experiment are shown.

Table 6.

Flow Cytometric Analysis of EGFP Expression 36 h Post-Transfection Was Used to Quantitate Transfection Efficiency ( Fig. 9)

	GFP-positive cells (%) pEGFP-C1		GFP-positive cells (%) pSUPER.neo+GFP
Transfection reagents	H9 hES cells	H1 hES cells	H9 hES cells	H1 hES cells
Lipofectamine 2000	25.4 ± 2	32 ± 5.1	23.7 ± 3	24.8 ± 5.3
Lipofectamine 3000	40.8 ± 6.5	38.7 ± 8	32.7 ± 4.2	37.7 ± 7.6
Lipofectamine LTX	34 ± 3.7	36.9 ± 7.6	26.2 ± 4.9	31.1 ± 6.3
FuGENE HD	58.2 ± 5.1	48.3 ± 2.4	44.7 ± 3.8	43.2 ± 5.1
TurboFect	29.7 ± 2.5	33.8 ± 3.6	27.9 ± 2.5	24 ± 5.2
Attractene	10.4 ± 1.3	12.6 ± 4.3	10.6 ± 2.4	8.1 ± 2.1
Neofect	21.1 ± 3.3	27.9 ± 3.5	23.7 ± 5.8	21.7 ± 4.4
LipoFiter	22.8 ± 4.6	20.9 ± 6.3	21.1 ± 2.9	13.6 ± 3.5

Data are shown as mean ± SD.

Effects of siRNA/pDNA transfection reagents on the viability of H9 hES cells

Finally, we assessed the cell viability of H9 hES cells when exposed to transfection reagents. siRNA/pDNA transfection using transfection reagents under the same conditions used in the efficiency evaluation experiments resulted in some cytotoxicity compared with untransfected cells. As shown in Fig. 10A, the viability of cells exposed to N-TER Nanoparticle was significantly reduced (44.67% ± 11.53% survival) and lower than other reagents (P < 0.05). For the other siRNA transfection reagents, the cell viability was in a normal range (71%–89%). For pDNA transfection reagents, higher cell livability was found after transfection with Attractene (94% ± 13.33%). In contrast, Lipofectamine 2000 showed the lowest amount of cell viability (57.2% ± 12.8 survival), followed by Neofect (64.43% ± 12.97% survival). The cell viability of the other pDNA transfection reagents was in a normal range from 72.5% to 86.13% (Fig. 10B).

FIG. 10.

Effects of various transfection reagents on the viability of hES cells. Viability of H9 hES cells was examined by MTT assay at 48 h following transfection with various siRNA (A) and pDNA (B) transfection reagents in optimal conditions.

Discussion and Conclusion

Inefficient introduction of siRNA or pDNA into hES cells is a serious obstacle to biological and basic medical research that can restrict functional analysis of genes and proteins [33]. Over the past decade, several gene transfer techniques have been studied with the purpose of increasing the transfection efficiency of exogenous nucleic acids into hES cells, such as viral vector-based transfection (eg, lentiviral vectors, retrovirus vectors, adenovirus vectors) [34 –39], physical treatment-based transfection (eg, electroporation, optical transfection) [7,40 –43], and chemical material-based transfection (eg, liposomes or polycationic polymers, nanoparticles) [21,25,31,44]. Viral vector-based transfection is generally the most efficient method to transfer genetic materials into hES cells, but it has some major limitations related to cytotoxicity, immunogenicity, inflammation, and the risk of residual viral elements that can potentially cause insertional mutagenesis and oncogene activation [45 –47]. Recently, physical treatment-based transfection has also been developed and successfully improved gene transfer efficiency of hES cells. Unfortunately, this method often requires special equipment and often results in a high percentage of cell death. Chemical material-based transfection is the most common strategy in gene transfection. Although efficiency of this technique is lower than the other two techniques in gene transfer of hES cells sometimes, the low cytotoxicity, simplicity, less time-consuming, fewer safety concerns, and higher genetic material carrying capacity have made it more acceptable for gene delivery than other delivery systems [48,49]. The traditional protocol of chemical material-based transfection, termed forward transfection (FT), essentially consists of exposing cells to a nucleic acids/reagent complex 20–30h after cell plating. Unfortunately, efficiency of this standard protocol in hES cells is relatively low [50]. Alternatively, in the reverse transfection (RT), the complex is directly added to suspended cells while plated. The RT protocol dramatically increased gene transfer efficiency of hES cells especially with several modifications introduced, such as prolonging incubation time for suspending cells with transfection materials [13,50]. However, there are serious objections limiting its application, such as requiring a large number of cells before transfection, death of most of the cells during transfection, and poor state of successfully transfected cells. Therefore, there has been a push to develop the chemical material-based transfection technique to enhance gene transfer efficiency of hES cells.

In this study, we proposed an efficient, modified FT method to transfect hES cells using the chemical material-based transfection. Two key processing factors—single-cell plating and starvation—were introduced into this new method. Using this new developed method, our results showed that siRNAs (targeting endogenous or exogenous gene) or pDNAs (RNAi or expression vectors) could be delivered into H1 and H9 hES cell lines with a much higher efficiency than the traditional methods. Importantly, hES cells after single-cell plating and starvation, retained their pluripotency, comparative to normally cultured cells, through analyzing the presence of pluripotency markers. So far, we have found that the new method described here could also obviously enhance transfection efficiency using other transfection reagents examined in the second part of this study and was equally effective in delivering siRNAs and pDNA vectors into iPSCs (data not shown). Indeed, the new method does not require too much technical operation and can be quickly grasped by any person familiar with traditional transfection methods. Compared with other transfection methods, the new method is more simple, feasible, and far less expensive. Especially, good repeatability and reliability of the results can be achieved by using this new method. Highly efficient transfection obtained with the new method allows its use as a practical and valuable tool for studying gene function or other biomedical research in hES cells.

To date, more and more commercially available transfection reagents such as Lipofectamine RNAiMAX, HiPerFect Transfection Reagent, Lipofectamine2000, and FuGENE HD have been used to introduce siRNA or pDNA into hES cells [5,21,44,50,51]. Our previous studies have shown that the Lipofectamine RNAiMAX is a more efficient transfection reagent for delivering siRNA into hES cells compared with Oligofectamine and Lipofectamine 2000 [31]. However, as for many other commercially available transfection reagents, few studies have been performed for side-by-side comparison of their efficiency in mediating transfection of siRNA or pDNA into hES cells. In this study, we chose eight siRNA transfection reagents and eight pDNA transfection reagents commonly used for delivery of siRNA or pDNA into hES cells and performed a detailed parallel comparison of their efficiency in transfecting hES cells. Our results evidently indicated that the RNAi efficiency of Lipofectamine RNAiMAX was significantly higher than the other seven siRNA transfection reagents under the same conditions and FuGENE HD showed the best efficiency of pDNA transfection among the pDNA transfection reagents. Furthermore, according to our observations, Lipofectamine RNAiMAX and FuGENE HD showed no obvious cytoxicity on hES cells.

Footnotes

Acknowledgments

This work was supported by the National Natural Science Foundation of China (30871246, 81070993, 81272972, 81472355), Specialized Research Fund for the Doctoral Program of Higher Education of China (SRFDP) (20120162110059), National Basic Research Program of China (2010CB833605), Foundation of Hunan Provincial Science and Technology Department (2012FJ4040, 2013FJ4010, 2014FJ6003, 2014FJ6006), Program for New Century Excellent Talents in University (NCET-10-0790), Key Program of Central South University (2010QYZD006), and Open-End Fund for the Valuable and Precision Instruments of Central South University.

Author Disclosure Statement

The authors declare that no competing interests exist.

References

Thomson

, Itskovitz-Eldor

, Shapiro

, Waknitz

, Swiergiel

, Marshall

and Jones

. (1998). Embryonic stem cell lines derived from human blastocysts. Science, 282:1145–1147.

, Han

and Wu

. (2009). Transplantation of human embryonic stem cell-derived endothelial cells for vascular diseases. J Cell Biochem, 106:194–199.

Barad

, Schick

, Zeevi-Levin

, Itskovitz-Eldor

and Binah

. (2014). Human embryonic stem cells vs human induced pluripotent stem cells for cardiac repair. Can J Cardiol, 30:1279–1287.

Yates

and Daley

. (2006). Progress and prospects: gene transfer into embryonic stem cells. Gene Ther, 13:1431–1439.

Rassouli

and Matin

. (2009). Gene silencing in human embryonic stem cells by RNA interference. Biochem Biophys Res Commun, 390:1106–1110.

Kim

, Hwang

, Go

, Kim

and Choi

. (2015). Gene therapy strategies using engineered stem cells for treating gynecologic and breast cancer patients (Review). Oncol Rep, 33:2107–2112.

Huang

, Hu

, Braun

, Pallaoro

, Morales

, Zasadzinski

, Clegg

and Reich

. (2015). Light-activated RNA interference in human embryonic stem cells. Biomaterials, 63:70–79.

Al-Dosari

and Gao

. (2009). Nonviral gene delivery: principle, limitations, and recent progress. AAPS J, 11:671–681.

Bajpai

, Lesperance

, Kim

and Terskikh

. (2008). Efficient propagation of single cells Accutase-dissociated human embryonic stem cells. Mol Reprod Dev, 75:818–827.

10.

, Meng

, Krawetz

, Liu

and Rancourt

. (2008). The ROCK inhibitor Y-27632 enhances the survival rate of human embryonic stem cells following cryopreservation. Stem Cells Dev, 17:1079–1085.

11.

Gauthaman

, Fong

and Bongso

. (2010). Effect of ROCK inhibitor Y-27632 on normal and variant human embryonic stem cells (hESCs) in vitro: its benefits in hESC expansion. Stem Cell Rev, 6:86–95.

12.

Zhang

, Wu

, Hu

, Wang

and Li

. (2012). Rho kinase inhibitor Y-27632 and Accutase dramatically increase mouse embryonic stem cell derivation. In Vitro Cell Dev Biol Anim, 48:30–36.

13.

, Jin

, Dong

, Cheng

, Lin

, Huang

and Qiu

. (2010). High-efficiency siRNA-based gene knockdown in human embryonic stem cells. Rna-a Publication of the Rna Society, 16:2564–2569.

14.

Riento

and Ridley

. (2003). Rocks: multifunctional kinases in cell behaviour. Nat Rev Mol Cell Biol, 4:446–456.

15.

and Lee

. (2005). Rho kinase as potential therapeutic target for cardiovascular diseases: opportunities and challenges. Expert Opin Ther Targets, 9:715–736.

16.

Watanabe

, Ueno

, Kamiya

, Nishiyama

, Matsumura

, Wataya

, Takahashi

, Nishikawa

, Muguruma

and Sasai

. (2007). A ROCK inhibitor permits survival of dissociated human embryonic stem cells. Nat Biotechnol, 25:681–686.

17.

Pakzad

, Totonchi

, Taei

, Seifinejad

, Hassani

and Baharvand

. (2010). Presence of a ROCK inhibitor in extracellular matrix supports more undifferentiated growth of feeder-free human embryonic and induced pluripotent stem cells upon passaging. Stem Cell Rev, 6:96–107.

18.

Kurosawa

. (2012). Application of Rho-associated protein kinase (ROCK) inhibitor to human pluripotent stem cells. J Biosci Bioeng, 114:577–581.

19.

Holm

, Nikdin

, Kjartansdottir

, Gaudenzi

, Fried

, Aspenstrom

, Hermanson

and Bergstrom-Tengzelius

. (2013). Passaging techniques and ROCK inhibitor exert reversible effects on morphology and pluripotency marker gene expression of human embryonic stem cell lines. Stem Cells Dev, 22:1883–1892.

20.

Gharechahi

, Pakzad

, Mirshavaladi

, Sharifitabar

, Baharvand

and Salekdeh

. (2014). The effect of Rho-associated kinase inhibition on the proteome pattern of dissociated human embryonic stem cells. Mol Biosyst, 10:640–652.

21.

Yen

, Yin

and Cheng

. (2014). Enhanced non-viral gene delivery to human embryonic stem cells via small molecule-mediated transient alteration of cell structure. J Mater Chem B Mater Biol Med, 2:8098–8105.

22.

Wallenstein

, Barminko

, Schloss

and Yarmush

. (2010). Serum starvation improves transient transfection efficiency in differentiating embryonic stem cells. Biotechnol Prog, 26:1714–1723.

23.

Merhi

, Dombu

, Brient

, Chang

, Platel

, Le Curieux

, Marzin

, Nesslany

and Betbeder

. (2012). Study of serum interaction with a cationic nanoparticle: Implications for in vitro endocytosis, cytotoxicity and genotoxicity. Int J Pharm, 423:37–44.

24.

Smith

, Giroud

, Wiggins

, Gower

, Thorley

, Stolpe

, Mazzolini

, Dyson

and Rappoport

. (2012). Cellular entry of nanoparticles via serum sensitive clathrin-mediated endocytosis, and plasma membrane permeabilization. Int J Nanomedicine, 7:2045–2055.

25.

Zafarana

, Avery

, Moore

and Andrews

. (2009). Specific knockdown of OCT4 in human embryonic stem cells by inducible short hairpin RNA interference. Stem Cells, 27:776–782.

26.

Ren

, Zhao

, Shan

, Yang

, Yin

, Jiang

, Zhang

and Yao

. (2005). Establishment of human embryonic stem cell line stably expressing Epstein-Barr virus-encoded nuclear antigen 1. Acta Biochim Biophys Sin (Shanghai), 37:68–73.

27.

Ren

, Zhao

, Yang

, Li

, Jiang

, Wang

, Shan

, Yang

, Zhou

and Zhang

. (2006). Establishment and applications of epstein-barr virus-based episomal vectors in human embryonic stem cells. Stem Cells, 24:1338–1347.

28.

Wanggou

, Jiang

, Li

, Zhang

, Liu

, Li

, Feng

, Liu

, Zhu

, et al. (2012). HESRG: a novel biomarker for intracranial germinoma and embryonal carcinoma. J Neurooncol, 106:251–259.

29.

Narumiya

, Ishizaki

and Uehata

. (2000). Use and properties of ROCK-specific inhibitor Y-27632. Methods Enzymol, 325:273–284.

30.

, Ren

, Shi

, Huang

, Liu

, Feng

, Liu

, Zhu

, Zhang

, et al. (2013). Identification, expression and subcellular localization of ESRG. Biochem Biophys Res Commun, 435:160–164.

31.

Zhao

, Yang

, Jiang

, Zhou

, Zhu

, Zeng

, Yao

and Ren

. (2008). Lipofectamine RNAiMAX: an efficient siRNA transfection reagent in human embryonic stem cells. Mol Biotechnol, 40:19–26.

32.

Lacoste

, Berenshteyn

and Brivanlou

. (2009). An efficient and reversible transposable system for gene delivery and lineage-specific differentiation in human embryonic stem cells. Cell Stem Cell, 5:332–342.

33.

Renz

and Beyer

. (2016). A concise protocol for siRNA-mediated gene suppression in human embryonic stem cells. Methods Mol Biol, 1341:369–376.

34.

Jang

, Shaw

, Yu

, Petersen

, Pepper

, Lutzko

and Kohn

. (2006). Specific and stable gene transfer to human embryonic stem cells using pseudotyped lentiviral vectors. Stem Cells Dev, 15:109–117.

35.

Cao

, Xie

, Gollan

, Zhao

, Narsinh

, Lee

and Wu

. (2010). Comparison of gene-transfer efficiency in human embryonic stem cells. Mol Imaging Biol, 12:15–24.

36.

Tashiro

, Kawabata

, Inamura

, Takayama

, Furukawa

, Sakurai

, Katayama

, Hayakawa

, Furue

and Mizuguchi

. (2010). Adenovirus vector-mediated efficient transduction into human embryonic and induced pluripotent stem cells. Cell Reprogram, 12:501–507.

37.

Wurm

, Gross

, Sgodda

, Standker

, Muller

, Forssmann

, Horn

, Blasczyk

and Cantz

. (2011). Improved lentiviral gene transfer into human embryonic stem cells grown in co-culture with murine feeder and stroma cells. Biol Chem, 392:887–895.

38.

Bajpai

. (2011). Lentivirus-mediated modification of pluripotent stem cells. Methods Mol Biol, 767:315–331.

39.

Rapti

, Stillitano

, Karakikes

, Nonnenmacher

, Weber

, Hulot

and Hajjar

. (2015). Effectiveness of gene delivery systems for pluripotent and differentiated cells. Mol Ther Methods Clin Dev, 2:14067.

40.

Zwaka

and Thomson

. (2003). Homologous recombination in human embryonic stem cells. Nat Biotechnol, 21:319–321.

41.

Lakshmipathy

, Pelacho

, Sudo

, Linehan

, Coucouvanis

, Kaufman

and Verfaillie

. (2004). Efficient transfection of embryonic and adult stem cells. Stem Cells, 22:531–543.

42.

Mohr

, de Pablo

and Palecek

. (2006). Electroporation of human embryonic stem cells: small and macromolecule loading and DNA transfection. Biotechnol Prog, 22:825–834.

43.

Moore

, Atze

, Yeung

, Toro-Ramos

, Camarillo

, Thompson

, Ricupero

, Brenneman

, Cohen

and Hart

. (2010). Efficient, high-throughput transfection of human embryonic stem cells. Stem Cell Res Ther, 1:23.

44.

Liu

, Jones

, Sumer

and Verma

. (2009). Stable transgene expression in human embryonic stem cells after simple chemical transfection. Mol Reprod Dev, 76:580–586.

45.

Dave

, Jenkins

and Copeland

. (2004). Gene therapy insertional mutagenesis insights. Science, 303:333.

46.

Nayak

and Herzog

. (2010). Progress and prospects: immune responses to viral vectors. Gene Ther, 17:295–304.

47.

Ren

, Yang

, Liang

, Liu

and Deng

. (2013). [Advances in cationic polymers used as nonviral vectors for gene delivery]. Sheng Wu Gong Cheng Xue Bao, 29:568–577.

48.

Kobayashi

, Rivas-Carrillo

, Soto-Gutierrez

, Fukazawa

, Chen

, Navarro-Alvarez

and Tanaka

. (2005). Gene delivery to embryonic stem cells. Birth Defects Res C Embryo Today, 75:10–18.

49.

Yen

, Zhang

, Gabrielson

, Yin

, Guan

, Chaudhury

, Lu

, Wang

and Cheng

. (2013). Cationic, helical polypeptide-based gene delivery for IMR-90 fibroblasts and human embryonic stem cells. Biomater Sci, 1:719–727.

50.

Villa-Diaz

, Garcia-Perez

and Krebsbach

. (2010). Enhanced transfection efficiency of by the incorporation of DNA liposomes in extracellular matrix. Stem Cells Dev, 19:1949–1957.

51.

Shi

, Ren

, Liu

, Wang

, Zhu

, Huang

, Liu

, et al. (2015). An ESRG-interacting protein, COXII, is involved in pro-apoptosis of human embryonic stem cells. Biochem Biophys Res Commun, 460:130–135.