SEARCH

SEARCH BY CITATION

Keywords:

  • Stem cells;
  • Telomerase;
  • TERT;
  • KLF4;
  • Transcriptional activation

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

The zinc finger Krüppel-like transcription factor 4 (KLF4) has been implicated in cancer formation and stem cell regulation. However, the function of KLF4 in tumorigenesis and stem cell regulation are poorly understood due to limited knowledge of its targets in these cells. In this study, we have revealed a surprising link between KLF4 and regulation of telomerase that offers important insight into how KLF4 contributes to cancer formation and stem cell regulation. KLF4 sufficiently activated expression of the human telomerase catalytic subunit, human telomerase reverse transcriptase (hTERT), in telomerase-low alternative lengthening of telomeres (ALT), and fibroblast cells, while downregulation of KLF4 reduced its expression in cancerous and stem cells, which normally exhibits high expression. Furthermore, KLF4-dependent induction of hTERT was mediated by a KLF4 binding site in the proximal promoter region of hTERT. In human embryonic stem cells, expression of hTERT replaced KLF4 function to maintain their self-renewal. Therefore, our findings demonstrate that hTERT is one of the major targets of KLF4 in cancer and stem cells to maintain long-term proliferation potential. STEM Cells 2010; 28:1510–1517.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Telomeres are the specialized DNA–proteins complexes at the ends of chromosomes that allow intact ends to be distinguished from broken chromosomes, prevent chromosomes from end-to-end fusion, degradation, genomic instability, and the associated risk of cancer [1, 2]. Telomere length is mainly influenced by changes in the activity of telomerase reverse transcriptase (TERT) that elongates telomeres [3]. Transcriptional regulation of telomerase is a major factor limiting telomerase activity in human cells. Human TERT (hTERT) expression is low or absent in most somatic cells, but it is activated during embryonic development, in stem cells, and in most tumors [4]. It has been shown that shortened telomeres are associated to increased age [5], which may contribute to organismal aging by limiting the proliferative capacity of adult stem cells [6]. Telomerase activity has been found to be upregulated in different kinds of cancers. Studies have shown that two oncogenes, c-MYC and SP1, play crucial roles in the regulation of hTERT transcription during carcinogenesis [7, 8]. Also, it is known that embryonic stem cells (ESCs) maintain high levels of telomerase activity, which are important for long-term stem cell self-renewal [9]. Deletion of telomerase in mouse adult stem cells can lead to their depletion and thus cause tissue degeneration [10, 11]. However, it is not completely understood how stem cells maintain high levels of TERT expression.

The Krüppel-like transcription factor (KLF) families are transcriptional factors that regulate a diverse array of cellular processes, including development, differentiation, proliferation, and apoptosis [12, 13]. For example, KLF1 is essential for expression of the ß-globin gene and for erythropoiesis [14], and KLF2 is involved in lung development [15]. KLF4, a member of this family, functions as a transcriptional activator or repressor depending on the interaction partner and the context of the binding sites. Recently, KLF4 has been shown to be one of four transcriptional factors (OCT3/4, SOX2, KLF4, and c-MYC) essential for reprogramming different kinds of differentiated cells into induced pluripotent stem (iPS) cells [16–22]. Although full characterization of iPS cells is still in progress, iPS cells seem to share some properties with ESCs.

Although telomerase activity has been previously shown to be increased during iPS cell generation using a combination of those four reprogramming factors [17, 21], c-MYC was later found dispensable for telomerase activation in iPS cells [23]. Here, we show that KLF4 binds directly to a KLF4-consensus site in the hTERT gene promoter and activates its expression in alternative lengthening of telomeres (ALT) cells and fibroblasts. Depletion of KLF4 in cancer cells and human ESCs (hESCs) decreases hTERT expression. Therefore, our findings have demonstrated that KLF4 is required for maintaining hTERT expression in hESCs and cancer cells by directly activating its transcription, and may also explain why KLF4 can promote induction of iPS cells from differentiated cells.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Yeast One-Hybrid Assay

The fragment −151 to +79 of hTERT [24] was cloned into pHISi and pLacZi (Clontech, Mountain View, CA, http://www.clontech.com/), at the upstream of a CYC1 promoter driving the expression of HIS3 and LacZ (Fig. 1A), respectively. These reporter plasmids were integrated into Saccharomyces cerevisiae strain YM4271. The yeast strain harboring the integrated pHISi plasmid was then transformed with HeLa cDNA libraries (Clontech). Yeast transformants with a prey plasmid were selected for leucine and histidine prototrophy under 15 mM 3-amino-1,2,4-triazole. Plasmids from positive candidates from the histidine screen were extracted and transformed into YM4271 harboring the empty pHIS3-1 to rule out the false positives. Positive clones were further transformed into the yeast strain harboring the integrated pLacZi plasmid and the interaction was confirmed by β-galactosidase activity in an agar overlay assay as manufacturer described (Clontech). All yeast manipulations were performed following standard procedures [25].

thumbnail image

Figure 1. KLF4 binds to the hTERT promoter in the yeast one-hybrid assay. (A): A schematic diagram shows that the fragment −151 to +79 of hTERT was placed at the upstream of a CYC1 promoter driving the expression of HIS3 or LacZ for a one-hybrid screen. (B): Reporter assay for yeast strains carrying the plasmid pGAD or pGAD-KLF4. Tenfold serial dilutions of two individual KLF4 clones were spotted on SC medium lacking leucine or without leucine and histidine. Plates were photographed after 3 days of growth. Abbreviations: AD, activation domain; 3AT, 3-amino-1,2,4-triazole; hTERT, human telomerase reverse transcriptase; KLF4, Krüppel-like transcription factor 4; SC, synthetic complete medium.

Download figure to PowerPoint

Cell Culture

VA13 (immortalized human lung fibroblasts) was obtained from American Type Culture Collection (http://www.atcc.org/). BJ1 (fibroblasts from neonatal foreskin) was kindly provided by Dr. Jing-Jer Lin. FaDu (squamous cell carcinoma) and oral epidermoid carcinoma (OECM1) cell lines have been described previously [26] and were maintained in RPMI (HyClone, South Logan, UT, http://www.perbio.com/) containing 10% fetal bovine serum (FBS, GIBCO, Grand Island, NY, http://zt.invitrogen.com/). BJ1 cells were maintained in MEM containing 10% FBS, 2 mM L-glutamine (Invitrogen, Carlsbad, CA, http://zt.invitrogen.com/), 4 M nonessential amino acids (Invitrogen) and 1 mM sodium pyruvate (Sigma, Madison, WI, http://www.sigmaaldrich.com/). 293T and VA13 cells were maintained in DMEM supplemented with 10% FBS. NTU1 (hESC line) [27] was maintained as undifferentiated cells on inactivated mouse embryonic fibroblast (MEF) feeder in DMEM/F12 supplemented with 20% knockout serum replacement, 1 mM glutamine, 0.1 mM nonessential amino acid, 4 ng/ml bFGF (Invitrogen), and 0.1 mM β-mercaptoethanol (Sigma, Madison, WI). Cells with 25–35 passages were used for lipofectamine RNAiMAX transfection [28].

Transfection, RNA Purification, and Quantitative Reverse Transcription Polymerase Chain Reaction

Transient transfection was performed using either Lipofectamine Plus Reagent or Lipo2000 as manufacturer described (Invitrogen). RNA was extracted using the TRIzol reageant (Invitrogen). cDNA was synthesized using the SuperScript III CellsDirect cDNA Synthesis Kit (Invitrogen). Quantitative reverse transcription polymerase chain reaction (qRT-PCR) was performed in a PRISM7500 Sequence Detection System (ABI, Foster City, CA; http://www.appliedbiosystems.com.tw/) and ribosome protein L 30 was used as an internal control. pcDNA3-KLF4, pcDNA3-KLF4-AS, and pcDNA3-KLF1 were kindly provided by Dr. Shaw-Fa Yet [29]. pcDNA3-hTERT was obtained from Dr. Bob Weinberg [30]. All primer sequences for PCR are available upon request.

siRNA, Western Blot Analysis, and Telomerase Activity Assay

KLF4 siRNA was synthesized by Silencer Select Predesigned siRNA system (Ambion, Foster City, CA, http://www.ambio.com) and targeted the coding region of the human KLF4 gene (NCBI accession number NM_004235.3). Silencer Select Negative Control #1 siRNA (NCsi) was used as the negative control, which was bioinformatically designed with the latest information about miRNA seed regions and sequence alignment algorithms to minimize interactions with any transcript in the transcriptomes of human, mouse, and rat. All oligo sequences for siRNA are available upon request. For transient transfections, FaDu and OECM1 cells were seeded at a density of 50%–70% in six-well plates. On the following day, transfections were performed by using KLF4-specific siRNA (10 nM) duplexes or control duplexes. Cells were harvested 48 hours after transfection. Western blot and telomerase activity (TRAP) assay were conducted and quantified as described previously [31–33]. The filter was immunoblotted with antibodies against KLF4 (ab34814, AbCam, Cambridge, United Kingdom, http://www.abcam.com/), P84 (NB100-174, Novus, Littleton, CO; http://www.novusbio.com/), or KLF1 (GTX104833, GenTex, Irvine, CA, http://www.gentex.com/). Images were captured and quantified by a bioluminescence imaging system (UVP BioSpectrum AC Imagine System, UVP, Upland, CA; http://www.uvp.com/). For protein quantification, KLF4 expression was normalized to P84 expression. For the ease of observation, 12.5, 12.5, 1,000, and 50 ng of FaDu, OECM1, BJ1, and VA13 lysates were subjected to the TRAP assay, respectively. The PCR reactions were conducted for 30 cycles. Telomerase activity was calculated by the ratio of the “sum intensity” of entire ladder to the signal of the amplified internal control. A 36-bp internal standard was used as an internal control. Telomerase activity was normalized between experiments using the “sum intensity” values of the cell line. Telomerase activity of the KLF4-transfected or NCsi cell line was set as 1.

Luciferase Reporter Assay

Various lengths of DNA fragments upstream of the initiating ATG codon of the hTERT gene were PCR amplified and inserted into the luciferase reporter vector, pXP2 [8]. This reporter construct (0.5 μg) was cotransfected with pcDNA3 and pcDNA3-KLF4 (1 μg) into 293T cells using the calcium phosphate transfection method. Renilla luciferase plasmids (0.1 μg) were cotransfected to serve as an internal control for transfection efficiency. Dual luciferase reporter assay (Promega, Madison, WI, http://www.promega.com/) was performed as manufacturer instruction. Each transfection was performed in triplicate. Data shown here are representative of three or more experiments from independent transfection, and standard deviation bars were shown. Point mutations were introduced into the hTERT promoter using QuikChange site-directed mutagenesis (Stratagene, La Jolla, CA, http://www.genomics.agilent.com/) as manufacturer described. All primer sequences for PCR and mutagenesis are available upon request.

Electrophoretic Mobility Shift Assay

The GST-KLF4 fusion protein was constructed and expressed. KLF4 was excised from pcDNA3-KLF4 using BamHI, subcloned into pGEX4T-1, and expressed as GST-KLF4 fusion proteins in E. coli. Electrophoretic mobility shift assay (EMSA) was performed as previously described [7]. Briefly, 2.4 μg recombinant proteins were used. 32P-end-labeled double-stranded synthetic oligonucleotides (5′-CCCCGGCCACCCCCGCGATGATCTG-3′) containing the KLF4 consensus in the core promoter were added and underlined sequences of the probes were altered to TT in the mutated oligonucleotides. Following electrophoresis on a 4% polyacrylamide nondenaturing gel, the gel was dried and subjected to autoradiography. Antibody against GST (27-4577, GE Healthcare, Uppsala, Sweden, http://www.gehealthcare.com/) was added to the binding reactions for the supershift assay.

Chromatin Immunoprecipitation

Chromatin immunoprecipitation (ChIP) assay was performed as previously described [34]. Approximately 1 × 107 cells were used for each immunoprecipitation. The following antibodies were used for the immunoprecipitation reactions: anti-KLF4 antibody (SC-20691, Santa Cruz Biotechnology, Santa Cruz, CA, http://www.scbt.com/) and V5 (46-0705, Invitrogen). DNA was amplified by PCR, using the primers in hTERT promoter and ribosome protein L. The reaction was resolved using 3% agarose gels following with ethidium bromide staining.

Immunocytochemistry

Alkaline phosphatase (AP) staining was performed according to the manufacture's instructions using the Alkaline Phosphatase Substrate Kit III (SK-5300, Vector, Burlingame, CA, http://www.vectorlabs.com/). Images were captured using a (Leica Leica, Wetzlar, Germany, http://www.leica-microsystems.com/) DMIRB microscope with a Nikon (Nikon, Japan, http://www.nikon.com/) digital sight DS-SMc photometric camera and processed by NIS-Element D 3.0 software.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Identification of KLF4 As a Regulator of hTERT Expression by the Yeast One-Hybrid Screen

Previous deletion analysis of hTERT promoter identified a core promoter region upstream of hTERT [24]. c-MYC and SP1 were further demonstrated to activate hTERT expression [7, 8] by directly binding to the core promoter region of hTERT. To identify additional factors that may regulate telomere expression, we utilized the yeast one-hybrid assay to screen for hTERT promoter binding proteins. The integrated bait sequence used for this assay was the defined hTERT promoter [24] located from −151 to +79 base pairs upstream of the transcription start site in the hTERT promoter that controls expression of two reporters, HIS3 and LacZ (Fig. 1A). The bait plasmids were integrated into the genome of the yeast strain BY4741. After screening 1 × 107 clones from a human HeLa cDNA library, only one positive yeast strain was recovered carrying the truncated sequence of a zinc finger transcription factor, KLF4. The full length KLF4 was then used to further confirm such interaction (Fig. 1B). These results showed that KLF4 is a potential hTERT promoter-binding protein.

Activation of hTERT by KLF4 in Fibroblasts and ALT Cells Having Low hTERT Expression

To determine whether KLF4 can indeed regulate hTERT expression, human fibroblasts (BJ1), or an ALT cell line (VA13), which lack or display reduced telomerase activity [35–37], were transfected with KLF4 expression plasmids, respectively, and semiquantitative RT-PCR analysis was used to determine hTERT expression. Interestingly, hTERT RNA levels increased significantly following KLF4 overexpression, which was confirmed by RT-PCR and western blotting (Fig. 2A). Furthermore, both cell types displayed telomerase activity (Fig. 2B and 2C; Supporting Information, Fig. S1). To further determine the specificity of KLF4 in the activation of hTERT, a KLF4 antisense plasmid, pcDNA3-KLF4-AS, or another closely related KLF4 family member, pcDNA3-KLF1 [38], was transfected into VA13, respectively. As shown in Figure 2D, these constructs failed to activate the expression of hTERT, suggesting KLF4 can specifically activate hTERT expression. Taken together, these results suggested that KLF4 is able to regulate hTERT transcription.

thumbnail image

Figure 2. KLF4 induces hTERT transcriptional expression and telomerase activity. (A): Analysis of telomerase activation in KLF4-overexpressed BJ1 and VA13 cell lines. RNA was harvested and RT-PCR and western blot analyses were performed. GAPDH and P84 were used as a loading control, respectively, for RT-PCR and western blot analysis. (B): Telomerase activity was measured by the telomerase activity (TRAP) assay using protein extracts from samples in (A). FaDu cell extract was used as a positive control. LB which lacks cell extract was used as a negative control. Heat refers to lyaste, which has been preincubated for 20 minutes at 85°C. (C): TRAP data were quantified. Note that heat-inactivated cells displayed the lowest telomerase activity was considered as zero and the level of KLF4-transfected cells was arbitrarily set as 1. (D): Quantitative RT-PCR was performed to detect activation of hTERT in cells expressing empty vector, sense KLF4, antisense KLF4, and KLF1 in VA13 cells. Black bars represent KLF4 mRNA level and the gray bars represent hTERT mRNA level. Bars represent the means and SD. KLF4 and KLF1 expression levels in these cells were analyzed by western blot analysis. P84 was used as a loading control. Abbreviations: GAPDH, Glyceraldehyde 3-phosphate dehydrogenase; hTERT, human telomerase reverse transcriptase; IC, internal control; KLF4, Krüppel-like transcription factor 4; LB, lysis buffer; RT-PCR, reverse transcription polymerase chain reaction; WB, western blot.

Download figure to PowerPoint

Requirement of Endogenous KLF4 for Maintaining hTERT Expression in Cancer Cells

To test whether hTERT is regulated by endogenous KLF4, we performed KLF4 siRNA knockdown experiments on FaDu and OECM1 telomerase-positive cancerous cells and then examined hTERT expression. The endogenous KLF4 levels were repressed in the KLF4 siRNA-transfected clones compared with the negative control following the transfection of KLF4 RNAi (Fig. 3A and 3B). Consistent with the idea that KLF4 is involved in the activation of hTERT, the hTERT level was decreased in the KLF4 knockdown cells (Fig. 3A). As expected, endogenous telomerase activity was slightly reduced in the KLF4 knockdown cells compared with the controls (Fig. 3C and 3D). These results demonstrate that endogenous KLF4 contributes to hTERT expression in human cancer cells.

thumbnail image

Figure 3. siRNA-mediated repression of endogenous KLF4 reduces the expression of hTERT in FaDu and OECM1 cells. (A): Quantitative RT-PCR shows KLF4 and hTERT expression in FaDu or OECM1 cells transiently transfected with oligo KLF4 siRNA or a NCsi. Black bars represent KLF4 mRNA level and the gray bars represent hTERT mRNA level. (B): Western blot analysis shows that introduction of KLF4si for 48 hours led to reduction in KLF4 protein levels. P84 was used as a loading control. The folds of the expression levels of KLF4 compared with those of the controls are shown below. (C): Telomerase activity was measured by the telomerase activity (TRAP) assay performed using protein extract from FaDu or OECM1 cells transiently transfected with KLF4si or a negative control oligo siRNA for 48 hours. (D): Independent TRAP assay results were quantitated. Black bars represent FaDu telomerase activity and the gray bars represent OECM1 telomerase activity. Bars represent the means and SD. Abbreviations: hTERT, human telomerase reverse transcriptase; IC, internal control; KLF4, Krüppel-like transcription factor 4; KLF4si, KLF4 siRNA; NCsi, negative control siRNA; OECM1, oral epidermoid carcinoma 1; RT-PCR, reverse transcription polymerase chain reaction.

Download figure to PowerPoint

Direct Activation of hTERT by KLF4 Through Binding to the Proximal Promoter Region

To further investigate whether KLF4 upregulates hTERT gene expression at the transcriptional level, we first search for KLF4 binding sites in the hTERT promoter region and then determine if they are responsible for hTERT expression using a luciferase (Luc) reporter gene fused to the hTERT promoter sequence. Various regions of the upstream of the TERT, which range from −210 to +77, were fused to the Luc reporter [8], and luciferase activities were measured in the transfected cells. Cotransfection of KLF4 led to a 2.3- to 3.7-fold increase for the reporter constructs carrying different promoter regions (Fig. 4A). Surprisingly, the shortest promoter region ranging from +18 to +77 could still respond to KLF4 activation, indicating that the KLF4 response element lies in the 60 bp DNA sequence. KLF4 was previously reported to bind to GC-rich regions [38]. A potential KLF4 binding site was readily found, which is similar to the mouse Klf4 consensus site [39], within the small promoter region (Fig. 4B). Supporting the idea that the putative KLF4 binding site is responsible for KLF4 response, a 2-bp substitution in the consensus site completely abolished the response of the smallest promoter to KLF4, while even a 3-bp substitution in the other position of the minimal promoter region still maintain the response (Fig. 4B). EMSA was performed to further determine whether KLF4 binds to these sites directly. Recombinant GST-KLF4, but not GST, bound to the end-labeled wild-type oligos, while the mutated version of this site, where the conserved CC sequence was changed to TT, significantly reduced its interaction with KLF4 (Fig. 4C). These results indicate that KLF4 binds to the hTERT promoter directly. We then examined whether KLF4 binds directly to the hTERT promoter in vivo using ChIP assay. As shown in Figure 4D, endogenous KLF4 showed specific binding activity toward the hTERT promoter in telomerase-positive FaDu cells, indicating that endogenous KLF4 binds to the hTERT promoter as well. These results have demonstrated that KLF4 activates hTERT through direct binding to the hTERT promoter.

thumbnail image

Figure 4. KLF4 directly promotes hTERT expression. (A): Transient transfection of KLF4 increases hTERT promoter activity. 293T cells were transiently cotransfected with luciferase reporter constructs and the pcDNA3 or pcDNA3-KLF4 expression plasmid. For each construct, the pRL-TK plasmid was cotransfected to correct for differences in transfection efficiency. Fold activation data represents the degree of activation obtained with pcDNA3-KLF4 plasmid relative to empty vector (pcDNA3). Error bars indicate standard derivations from three independent experiments. (B): Identification of a KLF4 binding site proximal to the hTERT core promoter. A schematic diagram depicts locations of the binding sites that were mt. Another GC-rich region was mutated to serve as a nc. Bars represent the mean and standard errors. (C): Electrophoretic mobility shift assay (EMSA) was performed with WT or mt-radiolabeled double-stranded oligonucleotides. Recombinant GST or GST-KLF4 proteins were subjected to EMSA (Materials and Methods) in the presence or absence of antibodies. (D): Schematic diagram shows the hTERT promoter region and location of PCR primers for chromatin immunoprecipitation (ChIP) assay. ChIP assay was conducted using an anti-KLF4 antibody, other antibody (V5) or no antibody in FaDu cells. The PCR primer pair is specific to the KLF4 binding site of the hTERT promoter (lower panel). Ribosome protein L 30 (upper panel) was used as an internal control. Abbreviations: GST, glutathione-S-transferase; hTERT, human telomerase reverse transcriptase; KLF4, Krüppel-like transcription factor 4; mt, mutated; nc, negative control; PCR, polymerase chain reaction; RPL, ribosome protein L 30; WT, wild-type.

Download figure to PowerPoint

Requirement of KLF4 for Maintaining hTERT Expression in hESCs

Given that transducing fibroblasts with four transcription factors, OCT3/4, SOX2, KLF4, and c-MYC, converts fibroblasts to iPS cells [21, 22], but as c-MYC is dispensable for telomerase activation in this process [23], we wonder whether KLF4 is involved in telomerase activation in stem cells. Toward this aim, KLF4 was knocked down by transient transfection of KLF4 siRNAs. Like in cancer cells, the reduction of KLF4 expression, which was confirmed by RT-PCR and western blotting, caused a decrease in hTERT mRNA expression in hESCs (Fig. 5A and 5B). AP is highly expressed in undifferentiated hESCs, but is rapidly downregulated as they differentiated [40]. In addition, as hESCs differentiate, they undergo extensive morphological changes [27, 41]. Following KLF4 knockdown, most of the hESCs exhibited distinct morphologies of differentiated cells, which is consistent with the previous finding [27, 41] (Fig. 5C). Consistently, AP-positive colonies were decreased following the knockdown (Fig. 5C and 5D). It is worth to note that KLF4 expression does not regulate AP activity (Supporting Information, Fig. S2). Furthermore, endogenous KLF4 could specifically bind to the hTERT promoter in hESC (Fig. 5E). As KLF4 knockdown leads to the reduction of hTERT expression and hESC differentiation, we then investigated whether overexpression of hTERT can rescue KLF4 knockdown-induced hESC differentiation. Based on AP expression and cell morphology, overexpression of hTERT rescued this differentiation phenotype induced by KLF4 knockdown (Fig. 6A and 6B). These results have demonstrated that KLF4 is directly involved in sustaining hTERT expression and thus maintains hESC self-renewal.

thumbnail image

Figure 5. Effects of KLF4 gene knockdown on hESCs. (A): Quantitative reverse transcription polymerase chain reaction analyses of KLF4 and hTERT mRNA levels in hESCs transfected with oligo KLF4 siRNA at 48 hours post-transfection. (B): Western blot analysis demonstrates the reduction of KLF4 expression in KLF4 siRNA-treated hESCs. The folds of the expression levels of KLF4 compared with that of the loading control are shown below. (C): Repression of KLF4 expression induces hESC differentiation. Phase contrast images of hESCs were photographed at 48 hours post-transfection. Embryonic stem cells were stained for AP to identify undifferentiated cells. Scale bar = 100 μm. Black arrowheads represent undifferentiated hESCs, white arrowheads represent mouse embryonic fibroblast, and arrows indicate differentiated hESCs. (D): Four replicate experiments were evaluated by microscopic imaging to assess the percentage of undifferentiated cells, AP positive, within a colony (horizontal axis). The data are expressed as the average ratios ± SD of colonies for NCsi-, GAPDHsi- and KLF4si-treated hESCs. (**, p < .001, and *, p < .005; one-way ANOVA. Error bars denote SD.). (E): Chromatin immunoprecipitation assay was conducted as described in Figure 4D in hESCs. Abbreviations: AP, alkaline phosphatase; hESCs, human embryonic stem cells; hTERT, human telomerase reverse transcriptase; KLF4, Krüppel-like transcription factor 4; KLF4si, KLF4 siRNA; NCsi, negative control siRNA; RPL, ribosome protein L 30.

Download figure to PowerPoint

thumbnail image

Figure 6. hTERT expression rescues KLF4 siRNA-induced hESC differentiation. (A): hESCs were transiently cotransfected with KLF4 siRNA and the pcDNA3 or pcDNA3-hTERT expression plasmid. Phase contrast (upper panel) and AP images were photographed at 72 hours post-transfection. Scale bar = 100 μm. (B): Experiments were evaluated by microscopic imaging to assess the percentage of undifferentiated cells as described in Figure 5. (**, p < .001, and *, p < .05; two-way ANOVA. Error bars denote SD.). Abbreviations: AP, alkaline phosphatase; hESCs, human embryonic stem cell; hTERT, human telomerase reverse transcriptase; KLF4si, KLF4 siRNA; NCsi, negative control siRNA.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Transduction of four transcription factors, OCT3/4, SOX2, KLF4, and c-MYC, converts fibroblasts to iPS cells [21, 22]. Telomerase activity was observed to be upregulated in these iPSCs in comparison with the differentiated cells used for reprogramming [21]. In addition, the maintenance of iPSCs also requires continuous elongation of telomeres [23]. Although c-MYC has been shown to be required for activating telomerase in cancer cells [8], it is dispensable for telomerase activation in the stem cell reprogramming process. Therefore, one of the OCT3/4, SOX2, or KLF4 factor should be required to directly or indirectly activate telomerase during the reprogramming process, and it remains a mystery which factor is responsible for activation of telomerase activity. In this study, we showed that KLF4 plays a direct role in induction of the activity of telomerase, the ribonucleoprotein complex expressed in human cancer cells and ESCs. In addition, KLF4 sufficiently activated hTERT expression in the cells lacking or with low telomerase activity. Contrarily, another KLF family member, KLF1, cannot turn on hTERT expression. Finally, we demonstrated that KLF4 directly binds to the hTERT promoter and thus activates its transcription in cancer and stem cells. KLF4 knockdown led to differentiation of hESCs, and expression of hTERT rescued hESC differentiation caused by reduction of KLF4 function. Therefore, KLF4 directly activates hTERT expression in human cancer and ESCs, suggesting that KLF4 may be responsible for activation of telomerase activity during the iPS reprogramming.

KLF4 has been shown as both an activator and a repressor [42]. The presence of both activation and repression domains may allow KLF4 to alter their positive or negative transcriptional effect as the situation dictates. In this study, KLF4 serves as an activator to promote hTERT expression, which may further sustain cell proliferation. Previous studies reveal that KLF4 is highly expressed in epithelial tissues including the gut and skin [43, 44]. These findings imply that KLF4 is usually expressed in adult somatic tissues with a higher rate of cell proliferation. In agreement with these data, while most normal somatic cells are telomerase negative, low levels of this enzyme have been detected in skin and gut [45, 46]. In the early stage of laryngeal squamous cell carcinoma, KLF4 is highly expressed and this is correlated with cancer progression [47]. In the skin, overexpression of KLF4 results in hyperplasia and dysplasia [48] and it eventually leads to squamous cell carcinoma [49]. A previous study [50] has indicated that activation of telomerase activity happens frequently in head and neck squamous cell carcinoma (HNSCC) and may occur in the early tumorigenesis process. These studies suggest that KLF4 may in fact present an oncogenic activity in certain types of cancers. Here, we showed in telomerase positive HNSCC cell lines, FaDu and OECM1, KLF4 contributes to their telomerase expression. It is possible that c-MYC and/or SP1 also contribute to telomerase expression in these cells. Less telomerase reduction observed in FaDu cells might be caused by high expression of c-MYC in this cell line [47]. Although KLF4 knockdown was able to reduce telomerase expression in human cancer cells and stem cells, overexpression of KLF4 could only partially elevate telomerase activity. KLF4 and SP1 work together on human pregnancy-specific glycoprotein PSG5 promoter [51]. Accordingly, KLF4 and SP1 may need to work together to turn on hTERT expression in hESCs. However, the fact that overexpression of KLF4 upregulated hTERT transcription 228 folds in VA13 cells has ruled out this possibility. VA13 cells express very low level of the RNA component of human telomerase (hTR) [52]. While KLF4 promoted hTERT expression, hTR was not equally upregulated (data not shown). This yielded only modest upregulation of telomerase activity, even though hTERT has been drastically elevated. This is consistent with the previous findings that ALT cells do not express or express extremely low hTERT and hTR, and only coexpression of both subunits can reconstitute robust telomerase activity [31].

Recent advances in the reprogramming of somatic cells to iPS cells using defined factors have generated tremendous enthusiasm in the field of regenerative medicine. However, the mechanism of the reprogramming process induced by the four reprogramming factors has only been partially unraveled. By using MEF cells which overexpress these four reprogramming factors, it has been shown that iPS cell colonies formed faster when MEFs were pretreated with KLF4 and c-MYC and then with OCT3/4 and SOX2. This result indicates that KLF4 and c-MYC act earlier during reprogramming, conceivably by inducing epigenetic alterations that assist the binding of OCT4 and SOX2 to their target genes [53]. In our studies, repression of KLF4 expression-induced hESC differentiation within two days. Overexpression of hTERT-rescued KLF4 knockdown-induced hESC differentiation, suggesting that hTERT may be required at the early stage to maintain genome stability. A transcriptional hierarchy for the four reprogramming factors was previously defined [54]. The study also indicated that KLF4 is an upstream regulator of a large feed-forward loop that contains OCT4, SOX2, and c-MYC, as well as other common downstream factors including NANOG [55].Combining the results of our studies, it appears that KLF4 may exert a crucial role in somatic cell reprogramming and maintenance of embryonic stem (ES) cell self-renewal.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Altogether, our results demonstrate that KLF4 directly binds to the promoter of hTERT and activates telomerase. Depletion of KLF4 in hESCs decreases hTERT expression. Expression of hTERT replaces KLF4 function to maintain self-renewal of hESCs. Therefore, KLF4 is required for maintaining hTERT expression in hESCs by directly activating its transcription. This also explains why KLF4 can promote induction of iPS cells from differentiated cells.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

We thank Dr. Ting Xie for his discussion and critical comments. We also thank Drs. Jing-Jer Lin, Muh-Hwa Yang, Shaw-Fang Yet, and Bob Weinberg for plasmids and strains. This work was supported by grants from the National Science Council (NRPGM-98-3112-B-002-039) and National Health Research Institute of Taiwan (NHRI-EX98-9727BI) to S.C.T.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
STEM_477_sm_suppfig.doc618KSupporting Figures

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.