The Role of NF-κB Signaling in the Maintenance of Pluripotency of Human Induced Pluripotent Stem Cells

NF-κB signaling plays an essential role in maintaining the undifferentiated state of embryonic stem (ES) cells. However, opposing roles of NF-κB have been reported in mouse and human ES cells, and the role of NF-κB in human induced pluripotent stem (iPS) cells has not yet been clarified. Here, we report the role of NF-κB signaling in maintaining the undifferentiated state of human iPS cells. Compared with differentiated cells, undifferentiated human iPS cells showed an augmentation of NF-κB activity. During differentiation induced by the removal of feeder cells and FGF2, we observed a reduction in NF-κB activity, the expression of the undifferentiation markers Oct3/4 and Nanog, and the up-regulation of the differentiated markers WT-1 and Pax-2. The specific knockdown of NF-κB signaling using p65 siRNA also reduced the expression of Oct3/4 and Nanog and up-regulated WT-1 and Pax-2 but did not change the ES-like colony formation. Our results show that the augmentation of NF-κB signaling maintains the undifferentiated state of human iPS and suggest the importance of this signaling pathway in maintenance of human iPS cells.


Introduction
Mouse and human embryonic stem (ES) cells have differences in morphology, doubling times, and the expression of differentiation markers [1]. Self-renewal and the undifferentiated state of mouse ES cells depend on the activation of Stat3 by leukemia inhibitory factor (LIF) and Bmp4 signaling [2][3]. On the other hand, human ES cells require FGF2 signaling or ERK activation and co-operation with Activin/Nodal signaling [4], but not LIF/Stat3 and Bmp4 signaling. In 2006, Yamanaka et al. established induced pluripotent stem (iPS) cells [5,6]. Human iPS cells overcome two main problems of human ES cells: ethical issues related to derivation from human embryos, and the potential rejection of ES cell-derived cells by the immune system of the host. Consequently, human iPS cells are a promising resource in the field of regenerative medicine [5,6]. Although the function of human iPS cells and human ES cells is thought to be similar, an understanding of the molecular mechanism by which pluripotency and the undifferentiated state is maintained in human iPS cells is important for the clinical application of these cells.
NF-kB is a multi-functional transcription factor involved in various biological processes including inflammation, apoptosis, and immune regulation. NF-kB and Rel proteins are expressed in mammals: p65 (RelA), p50 (NF-IkB1), p52 (NF-kB2), c-Rel (Rel), and RelB [7]. The activation of NF-kB is triggered by the release of p50/RelA from the cytoplasm into the nucleus as a result of the degradation of IkappaBalpha (IkBa). The degradation of IkBa is regulated by various pathways, such as the phosphorylation of MAPK and IkB kinase (IKK) a/b. However, the functional role of NF-kB in maintaining the undifferentiated state of ES cells is controversial. Kim et al. reported the up-regulation of NF-kB upon the differentiation of mouse ES cells [8], and Torres et al. reported that Nanog maintains the undifferentiated state of mouse ES cells by inhibiting NF-kB [9]. On the other hand, Armstrong et al. reported that specific inhibition of NF-kB reduces expression of undifferentiation markers such as Oct4, Nanog, SSEA-4 and induces differentiation of human ES cells [4]. These reports suggest that NF-kB has different roles in mouse and human ES cells, but its role in human iPS cells remains to be clarified.
In the present study, we used monolayer differentiation and p65 siRNA, and compared the roles of NF-kB signaling and undifferentiated markers such as Nanog and Oct4, with undifferentiated human iPS cells. Our results indicate that NF-kB, especially p65, plays crucial role for maintenance of pluripotency of human iPS cells.

Cell Culture
Human iPS cells (253G1, 201B6, 201B7) [6] [11][12] were generously provided by Prof. Yamanaka (University of Kyoto). Human ES cells (KhES-1, 2 and 3) [13][14] [5,6,15]. The iPS cells were cultured according to a previously reported protocol [16]. Briefly, feeder cells (SNL) were grown to sub-confluence in Dulbecco's modified Eagle's medium (DMEM) (GIBCO BRL, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS), 0.1 mM MEM non-essential amino acids (SIGMA), 2 mM L-glutamine, 100 U/ mL penicillin, and 100 mg/mL streptomycin (GIBCO BRL) at 37uC in a 5% CO 2 environment. The SNL feeder cells were mitotically inactivated prior to the iPS culture. The SNL cells were added to 10 mg/mL of mitomycin C (SIGMA), incubated for 2 hours, and then washed with PBS to remove the mitomycin C. For the passaging of the iPS cells, the cells were seeded on SNL feeder cells in gelatin-coated tissue culture dishes and grown in ES cell medium (ReproCELL, Yokohama, Japan) supplemented with 5 ng/mL recombinant FGF2 (ReproCELL) at 37uC in a 5% CO 2 environment [6,17,18]. The medium was changed every other day. For monolayer differentiation, the harvested iPS cell colonies were seeded onto a dish without feeder cells and grown in DMEM supplemented with 10% FBS without FGF2 for 7 days.

RNA Extraction and Oligonucleotide Microarray Analysis
The total RNA was extracted from iPS and ES cells using TRIZOL (Invitrogen, Carlsbad, CA, http://www.invitrogen.com) according to the manufacturer's protocol. The purity of the RNA was checked using an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). We used the Human Genome U133A Plus2.0 arrays (Affimetrix, Santa Clara, CA), which contain almost 54,000 probe sets representing more than 47,000 transcripts including 38,500 well-characterized human genes [19]. The preparation of the cRNA and the hybridization of the probe arrays were performed according to the manufacturer's protocols.

siRNA Treatment
Human iPS cells, which grew to form small colonies for about 24 hours after seeding, were treated with both p65 siRNA and control siRNA, as previously reported [20]. Briefly, each siRNA treatments were gently mixed with each concentration with siRNA Transfection medium and reagent solution at room temperature for 30 minutes. The complex solution was added into ES cell medium of each iPS cells in 6-well plate. In this study, the siRNA treatment was performed twice every 36 hours.

Electrophoretic Mobility Shift Assay (EMSA)
Nuclear extracts were isolated and examined for NF-kB and activator protein 1 (AP-1) DNA-binding activity as described previously [20,21]. The sequences of the consensus doublestranded oligonucleotides used to detect the DNA binding sites were 59-AGTTGAGGGGACTTTCCCAGGC-39 for NF-kB and 59-CGCTTGATGACTCAGCCGGAA-39 for AP-1 (Santa Cruz). Consensus oligonucleotides biotinylated with a 39-end DNA labeling kit (Pierce, IL, USA) were incubated with 7 mg of nuclear extracts and reaction solutions for 30 minutes on ice. In the competition assays, a 20-fold excess of unlabeled NF-kB consensus oligonucleotides was added to labeled NF-kB consensus oligonucleotides, confirming the specificity of DNA-protein binding in every experiment. The binding reactions were separated on a 6% polyacrylamide gel and were subsequently transferred electrophoretically to a nylon membrane. Biotinylated DNA was detected using a streptavidin-horseradish peroxidase conjugate and a chemiluminescent substrate (Lightshift Chemiluminescent EMSA Kit, Pierce).

Western Blot Analysis
Cell extracts from the iPS cells were collected and subjected to Western blotting using previously described methods [22]. Equal amounts of protein (30 mg) as determined using the BCA Protein Assay (Pierce) were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Each antibody purchased from Santa Cruz Biotechnology was diluted 1:2000-4000 for the Western blot analysis. Reactivity was detected using the ECL Advance Western Blotting Detection Kit (GE Healthcare, UK).

Alkaline Phosphatase Staining
The pluripotent status of stem cells was detected using the ALP Staining Kit (System Biosciences, CA) [5,6]. The human iPS cells cultured on feeder layer were washed and then fixed with fix solution for 5 minutes. The fixed cells were responded with the prepared ALP substrate solution for 15 minutes, protected from light. The reaction was stopped with PBS. The red stained cell colonies were observed using a light microscope.

Immunohistochemical Analysis
The expressions of NF-kB p65 and Oct3/4 were detected using the ABC high-HRP Immunostaining Kit (TOYOBO, Osaka, Japan). Both antibodies were purchased from Santa Cruz Biotechnology and were diluted 1:200-500 for the immunohistochemical analysis.

Characterization of Human iPS Cells
Human iPS cells (253G1) were cultured on SNL feeder cells with ES cell medium containing FGF2. The undifferentiated state of the cell was confirmed using ES cell marker staining (SSEA-4, TRA1-60, TRA1-81, and Oct3/4) ( Figure 1A). To confirm the character of the human iPS cells, we compared the global geneexpression profiles of human iPS cell and human ES cells (KhES-1, 2 and 3) (Figure 1B and C). A Pearson correlation analysis revealed that 253G1 was clustered closely with the human ES cells ( Figure 1C). These results confirmed that the human iPS cells used in this study (253G1) were properly re-programmed and were similar to human ES cells.

Down-regulation of NF-kB Activity and Undifferentiated Markers Upon Monolayer Differentiation
For monolayer differentiation, human iPS cells were cultured without feeder cells using a standard culture medium (DMEM+10% FBS) without FGF2. Unlike the ES-like tightly packed flat colonies, the monolayer differentiated cells formed irregular colonies with large cells (Figure 2A). Compared with the differentiated cells (PTECs), NF-kB activity was augmented in the undifferentiated human iPS cells, but was almost abolished in the monolayer differentiated cells ( Figure 2B). In the undifferentiated state, human iPS cells highly expressed Oct3/4, and Nanog, but did not express WT-1 or Pax-2. However, the monolayer differentiated cells highly expressed WT-1 and Pax-2, but did not express Oct3/4 or Nanog ( Figure 2C).

Specific Knockdown of NF-kB Activity by Treatment with p65 siRNA
Human iPS cells cultured on SNL feeder cells were re-seeded on SNL feeder cells with ES cell medium in the absence or presence of p65 siRNA ( Figure 3A-L). On day 4, treatment with p65 siRNA abolished the NF-kB activity but showed no effect on AP-1 binding as a control ( Figure 3M). Unexpectedly, treatment with p65 siRNA showed no effect on cell growth and ES-like colony formation until day4 ( Figure 3I and J). Compared with the no treatment and control siRNA treatments, the shape and size of the colonies were almost the same, but only the colonies treated with p65 siRNA were ALP negative ( Figure 3N-P).

Downregulation of Nanog and Oct3/4 by Knockdown of NF-kB
To confirm the specific knockdown of p65 by siRNA, we performed a Western blot analysis. Treatment with p65 siRNA markedly reduced the protein expression of p65, but showed no effect on that of IkBa ( Figure 4A). As in the case of monolayer differentiation, treatment with p65 siRNA abolished the expression of the undifferentiated markers Oct3/4 and Nanog and upregulated those of the differentiation markers WT-1 and Pax-2 ( Figure 4A). To clarify the down-regulation of p65 and Oct3/4 in situ, we performed immunohistochemistry studies. The no treatment colonies and the colonies that were treated with the control siRNA showed the strong expression of p65 and Oct3/4, but treatment with p65 siRNA markedly reduced the expression ( Figure 4B-H).

Discussion
Human iPS cells have great therapeutic potential, but the role of NF-kB signaling in the maintenance of their pluripotency has not been fully understood. NF-kB activity is reportedly inhibited in undifferentiated mouse ES cells, and NF-kB activity is repressed by Nanog [9]. On the other hand, in human ES cells, treatment with the specific NF-kB inhibitor, PDTC, reduces expression of undifferentiated markers such as Oct4, Nanog, SSEA-4 and promotes prominent differentiation [4]. Compared with human ES cells, we used properly reprogrammed human iPS cells (253G1) and demonstrated the augmentation of NF-kB activity in undifferentiated human iPS cells. The augmented NF-kB activity was also confirmed using another 3 independent human iPS cells lines ( Figure S1). Furthermore, the knockdown of NF-kB downregulated Nanog and Oct3/4 expression, suggesting that NF-kB activity in human iPS cells is not suppressed by Nanog. These results suggest that NF-kB signaling may play some different roles in the maintenance of an undifferentiated state in mouse ES cells and human ES cells, but we should think about epiblasts stem cell. As in the case with human ES and iPS cells, mouse epiblasts stem cells [23][24] rely on FGF signaling, form flat disc-shaped colonies in vitro and show poor clonal survival in the absence of the ROCK inhibitor Y-27632. Taken together, the likely explanation behind the different role of NF-kB is not explained by species differences but that mouse ES cells reside in a different pluripotent state relative to human ES and iPS cells.
Armstrong et al. reported specific inhibition of NF-kB induces prominent differentiation of human ES cells [4], but the same group also reported the activation of NF-kB signaling pathway during differentiation as a result of embryoid body (EB) formation [25]. As Armstrong et al. reported, we confirmed augmented NF-kB activity in undifferentiated human ES cells and downregulation of NF-kB activity upon monolayer differentiation ( Figure S1). We also confirmed augmented NF-kB activity in EB formation ( Figure S1). This discrepancy may be explained by difference in culture condition. Inhibition of NF-kB using PDTC or siRNA was performed using 2-dimensional culture condition, but activation of NF-kB was confirmed using 3-dimensional culture condition, EB formation. As EB is a 3-dimensional mass, the culture condition of the cells inside of EB is not uniform. For example, oxygen concentration of center of EB is assumed to be much lower than outside of the mass. The cells outside of EB may utilize culture medium like monolayer 2-dimensional culture, but the cells located in the center of EB cannot directly contact or utilize culture medium. In other words, EB contains lots of cells cultured under different conditions and 2-dimensional monolayer culture condition may be more suitable to evaluate the role of NF-kB in the maintenance of pluripotency of human iPS cells.
Colony formation is one of the typical characteristics of undifferentiated ES and iPS cells, and an ES-like colony shape is the most potent indicators to select undifferentiated colonies during long-term ES cultures or the establishment of iPS cells. Interestingly, the knockdown of NF-kB by p65 siRNA abolished the expression of Oct3/4 and Nanog but never affected the colony growth and shape of human iPS cells until day 4. Since knockdown treatment with siRNA was transient ( Figure S2) and the precise mechanism remains to be determined, but our results show a minimal role of Oct3/4 and Nanog in ES-like colony formation of human iPS cells.
WT-1 exhibits a spatially and temporally defined expression in the developing genitourinary system, and the WT-1 promotor contains a conserved NF-kB site [26]. The Pax-2 gene is a kidneyspecific master gene that is required in both UB and mesenchymal cell lineages, normally optimizing UB branching and the mesenchymal-to-epithelial transformation in kidney development; several NF-kB binding motifs are present in the full length of the    59-promotor region of Pax-2. However, our results showed that the specific knockdown of NF-kB up-regulated WT-1 and Pax-2 expression. Recently, Chen et al. reported that NF-kB is not required for regulation of expression of the WT-1 gene [26], and Pax-2 gene expression requires both ROS generation and NF-kB activation. Taken together, the expression of WT-1 and Pax-2 in human iPS cells may not be directly regulated by NF-kB.
In conclusion, our results demonstrate that NF-kB signaling plays key roles in maintaining the undifferentiated state of human iPS cells suggesting that this signaling pathway can be another important indicator of the maintenance of human iPS cells. Figure S1 The relation of NF-kB activity to differentiation and undifferentiation in various iPS and ES cells. RPTECs and culture medium (REGM BulletKit) containing 0.5% FBS, hydrocortisone (0.5 mg/mL), epidermal growth factor (10 ng/mL), epinephrine (0.5 mg/mL), triiodothyronine (6.5 ng/ mL), transferrin (10 mg/mL), insulin (5 mg/mL), gentamicin (50 mg/mL), and amphotericin (50 ng/mL) were purchased from Cambrex Corporation (East Rutherford, NJ). R-iPS cells were established by the retroviral transduction of four transcription factors: Oct3/4, Sox2, Klf4, and c-Myc. We confirmed pluripotency of R-iPS by expression of undifferentiated markers and teratoma formation.