| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 281, Issue 33, 23319-23325, August 18, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Institute of Pharmacology, Department of Biological Sciences & Biotechnology, State Key Laboratory of Biomembrane and Membrane Biotechnology, Institutes of Biomedicine, School of Medicine, Tsinghua University, Beijing 100084, China
Received for publication, February 24, 2006 , and in revised form, May 15, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
The promoter of Rex-1 has been shown to contain an octamer motif that is required for its activity in undifferentiated F9 cells and is involved in RA-mediated down-regulation (5). Subsequent analysis has demonstrated that Oct-3/4 and Oct-6 can regulate the activity of Rex-1 promoter through this octamer motif in a dose-dependent manner (6). Sox2, a member of the HMG (high mobility group) domain DNA-binding protein family, forms a ternary complex with Oct-3/4 protein on the enhancer element of Fgf4 and regulates its expression as a binary complex (7, 8).
Nanog is a newly identified transcription factor that maintain mouse ES cells in the pluripotent state independently of leukemia inhibitory factor (9, 10). Upon differentiation, ES cells lose the expression of Nanog (10). We have recently demonstrated that Nanog functions as a transcription activator through binding to a consensus recognition motif determined by SELEX, despite the fact that Nanog was originally proposed as a transcription repressor to inhibit the expression of genes important for cell differentiation (9, 11, 12). Interestingly, Nanog is regulated by an adjacent pair of highly conserved Octamer- and Sox-binding sites through an interaction between Oct-3/4 and Sox2 (13). Our recent work has demonstrated that transcription factors Oct-3/4, Nanog, Sox2, and FoxD3 anchor a negative feedback loop to maintain the expression of pluripotent factors at a steady state (14). In this report, we present evidence that Rex-1 is regulated by Nanog, Oct-3/4, and Sox2 at the transcription level.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
The expression plasmids Nanog, Oct-3/4, Sox2, Dmu-mSox2, and FoxD3 have been described or will be detailed elsewhere (12).3 Nanog-GFP was made by cloning the full-length Nanog cDNA into EcoRV site in-frame to a FLAG and GFP coding sequence in vector-pCR3.1. Nanog N1 was described previously (12). The Vp16 chimera was prepared by inserting a PCR fragment encoding amino acids of the Vp16 activation domain to the modified N1 mutant. Rex-1 promoter fragments were amplified by PCR from the mouse liver genome DNA and inserted into the SmaI site of a promoterless luciferase reporter vector, pGL3-Basic (Promega, Madison, WI). The forward primers were: 5'-ccttgccacagcctcaccctgatag-3' for the promoter of 1500-bp Rexp1500; 5'-cggctgccaatgcattttttaaaacgc-3' for the promoter of 1000-bp Rexp1000; 5'-tggaaaaagttcaggcaactagtgtac-3' for the promoter of 500-bp Rexp500; 5'-gaggtactgagatgtgactgagtctca-3' for the promoter of 286-bp Rexp286; and 5'-cgacccaagacaaggggacagctct-3' for the promoter of 187-bp Rexp187). The reverse primer was 5'-ctccttggacccctccctttttagatg-3'. The Nanog siRNA construct was prepared by inserting annealed oligonucleotides encoding a double-stranded RNA for Nanog to pBSU6 vector. The target sequence is ggg aac gcc tca tca atg cct for Nanog. A complete GFP expression cassette, including a cytomegalovirus promoter, GFP coding sequence, and poly(A), was inserted into the siRNA vector for the purpose of visualization.
RT-PCR—Total RNAs were extracted form NIH3T3, P19, F9, and ES cells (and from the P19 and F9 cells transfected with Nanog-GFP and Nanog siRNA, respectively), using TRIzol reagent (Sangon, Shanghai, China; or an RNA kit from Qiagen). First strand cDNAs were synthesized and analyzed by PCR to detect the expression of Nanog, Rex-1, and 
-actin. The Nanog primers were: forward, 5'-gcggactgtgtgttctctcaggc-3'; reverse, 5'-ttccagatccgttcaccagatag-3'. The Rex-1 primers were: forward, 5'-tgacaaaggggacgaagcaagag-3'; reverse, 5'-gccatcaaaaggacacacaaag-3'. The -actin primers were: forward, 5'-cggctccggcatgtgcaaag-3'; reverse, 5'-aggggccacacgcagctcattg-3'. Amplification following hot start (94 °C for 5 min) was carried out for 25 cycles for -actin, 25 cycles for Nanog, and 30 cycles for Rex-1. For real-time PCR, the Nanog primers were: forward, 5'-ctcaagtcctgaggctgaca-3'; reverse, 5'-tgaaacctgtccttgagtgc-3'. The Rex-1 primers were: forward, 5'-aggccagtccagaataccag-3'; reverse, 5'-taggtatccgtcagggaagc-3'. The -actin primers were: forward, 5'-agtgtgacgttgacatccgt-3'; reverse, 5'-tgctaggagccagagcagta-3'.
Transfection and Reporter Assay—All cells were seeded in 24 wells and transiently transfected with Rex-1 promoter reporters (0.25 µg/well) and effector plasmids (control empty vector, Nanog expression construct, and Nanog siRNA vector) with increasing doses (0.25–0.75 µg) using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instruction. pCMV-Renilla plasmid (200 ng/transfection, Promega) was included in each transfection as an internal reference, and the amount of total DNA for each transfection was normalized to the same level using the control pCR3.1 empty vector. 24 h later, cells were lysed in 50 µl of 1x passive lysis buffer (Promega). Luciferase activity was measured using the dual luciferase reporter assay system (Promega) using a TD2020 luminometer (Turner Design). Each transfection was replicated at least twice and carried out in duplicate. We use Student's t test to assess whether the difference between two means was significant (p value at 0.05).
|
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
One may argue that the decline of Rex-1 expression in ES cells with Nanog knocked down (Fig. 1A) may be the consequence of ES differentiation triggered by the suppression of Nanog expression by Nanog siRNA. To establish a direct relationship between Nanog and Rex-1, we activated Rex-1 by Nanog in cells that normally express no or very little Rex-1. Indeed, P19 cells appear to express no detectable Rex-1 mRNA (1) and a much reduced level of Nanog as well (10). To this end, we transfected P19 cells with either pCR3.1GFP or pCR3.1-NanogGFP and extracted their RNAs 48 h post-transfection for RT-PCR analysis. As shown in Fig. 1B, we observed the induction of Rex-1 by Nanog in P19 cells (middle panel, lane 3 versus 2), whereas the mRNAs for actin are comparable in both cells and those for Nanog were elevated as expected (top and bottom panels).
|
Nanog Activates the Rex-1 Promoter in Both differentiated and Pluripotent Cells—We then screened several more cell lines for Rex-1 expression by RT-PCR. As shown in Fig. 2A, mouse ES cells and F9 EC cells express Rex-1 as detected by RT-PCR, whereas P19 and NIH3T3 do not (lanes 4 and 5 versus 2 and 3). This expression pattern suggests that Rex-1 is regulated differentially in these cells. Furthermore, these cell lines could serve as models to characterize the mechanism of Rex-1 regulation. To this end, we cloned a 1529-bp-long promoter of Rex-1 into pGL-Basic vector bearing the luciferase reporter (Fig. 2B), designating it as Rexp1500. As shown in Fig. 2C, the activity of this reporter functions as expected with relatively high activities in F9 and ES cells and no or lower activities in P19 and NIH3T3 cells, in agreement with the Rex-1 mRNA levels shown in Fig. 2A.
To test whether Nanog can transactivate Rex-1 promoter directly, we cotransfected the Rex-1 reporter together with the control or Nanog expression constructs into NIH3T3, P19 and F9 cells. Luciferase activities in these cells were determined, normalized (with Renilla), and are presented in Fig. 2D. Nanog up-regulated the Rex-1 reporter significantly in NIH3T3 (>3-fold), P19 (1.5-fold), and F9 (1.4-fold) cells compared with the control expression vector (Fig. 2D, lanes 2, 4, and 6 versus lanes 1, 3, and 5). Although Nanog can activate Rex-1 promoter in both differentiated cells (NIH3T3) and pluripotent cells (P19 and F9), the magnitude of activation appears to be greater in NIH3T3 cells than in P19 and F9 cells, perhaps because of the high endogenous levels of Nanog in both of the latter cells (data not shown). To test this possibility, we tested the dependence of Rex-1 expression in F9 cells by knocking down Nanog through siRNA in F9 cells. We considered F9 cells an ideal model because of high levels of expression for both endogenous Nanog and Rex-1 mRNAs (10). First, we tested the efficiency of Nanog knockdown by extracting the mRNA of the cells transfected with pBSU6-GFP and siNanog. As shown in Fig. 2F, Nanog siRNA inhibited most of the expression of endogenous Nanog after a 24-h transfection (top panel, lane 3 versus 2). Then we cotransfected the Rex-1 reporter construct with control and the Nanog siRNA constructs into F9 cells and measured luciferase activities 24 h later. As shown in Fig. 2E, the Nanog siRNA construct inhibited the Rex-1 reporter significantly (3.8-fold) compared with the control siRNA vector (lane 3 versus 2), demonstrating that Nanog plays a critical role in sustaining Rex-1 promoter activity.
Localization of Nanog-responsive Element between –187 and –286 in Rex-1 Promoter—Recent evidence from our group demonstrated that Nanog activates a reporter construct bearing multiple copies of its consensus binding motif (11), suggesting that Nanog activates its downstream target directly through DNA binding. To localize the DNA sequence that is responsive to Nanog-mediated activation, we generated a series of promoter deletion constructs (shown in Fig. 3A) bearing Rex-1 promoter sequences up to –1000, –500, –286, and –187, respectively, from the transcription start site at +1. To assess their responsiveness to Nanog, we transfected these deletion constructs with Nanog expression vector into P19 cells. A constant amount (0.25 µg) of each deletion Rex-1 promoter construct was cotransfected with control vector (Fig. 3B, lanes 1, 5, 9, 13, and 17) or increasing amounts of the Nanog expression construct (Fig. 3B, lanes 2–4, 6–8, 10–12, 14–16, and 18–20), respectively, to measure a dose-dependent activation of the reporter by Nanog in P19 cells. As shown in Fig. 3B, the luciferase activities driven by Rexp1500, Rexp1000, Rexp500, and Rexp286 Rex-1 promoter sequences were stimulated by Nanog in a dose-dependent fashion (lanes 5–20), whereas those by Rexp187 were not (lanes 1–4). Therefore, these results suggest that Nanog bind to the Rex-1 promoter between –187 and –286. Interestingly, the luciferase activities driven by Rexp1500 and Rexp1000 were less responsive than the Rexp500 and Rexp286 to Nanog, suggesting that additional effectors may modify the Nanog-mediated activation of Rex-1.
Nanog has been shown to bind an ATTA-containing motif (9, 15). We then screened the Rex-1 promoter between –187 to –286 and found a similar sequence at –244 (ATTC). To determine whether this motif is responsible for Nanog-mediated activation of Rex-1, we designed and generated four mutations within this motif (Fig. 3C, upper portion). Reporters bearing these mutations were analyzed by co-transfecting increasing amounts of the Nanog in P19 cells (Fig. 3C, lanes 2–4, 6–8, 10–12, 14–16, and 18–20). As shown in Fig. 3C, the luciferase activities driven by 286-mut-1, 286-mut-2, 286-mut-3, and 286-mut-4 mutant promoter sequences were no longer responsive to Nanog (lower portion, lanes 5–20), whereas Rexp286 was responsive to Nanog in a dose-dependent fashion (lanes 1–4). These results demonstrate that the ATTC motif is responsible for Nanog-mediated activation of Rex-1 promoter.
|
Regulation of Rex-1 Promoter by Sox2 and Oct-3/4 but Not FoxD3—Recent data from ChIP-on-ChIP experiments revealed that Nanog, Sox2, and Oct-3/4 co-occupy common sites on the Rex-1 promoter of their target genes (16). Because Oct-3/4 can bind to the OCTA site on the Rex-1 promoter as described (6), we tested whether Oct-3/4 and its partner Sox2 could activate Rex-1 promoter. We cotransfected the Rexp500 constructs with increasing amounts of Sox2, FoxD3, or Oct-3/4 into 293T, NIH3T3, and P19 cells and measured their activities, as described in the legend for Fig. 4. As expected, Oct-3/4 activated the Rex-1 promoter significantly in 293T and P19 cells but not in NIH3T3 cells (Fig. 4, A–C, lanes 1–4). Interestingly, the Oct-3/4 partner, Sox2, activated the Rex-1 promoter in an almost identical pattern (Fig. 4, A–C, lanes 1 and 8–10), whereas FoxD3 did not have any effect on the Rex-1 promoter in any of these cells (Fig. 4, A–C, lanes 1 and 5–7). These results clearly demonstrated that whereas either Sox2 or Oct-3/4 could activate the Rex-1 promoter, Foxd-3 could not. Furthermore, neither Sox2 nor Oct-3/4 could activate the Rex-1 promoter in NIH3T3 cells (Fig. 4B), in contrast to Nanog (Fig. 2D).
We further evaluated whether Sox2 activated Rex-1 specifically. We cotransfected the Rexp500 reporter with a Sox2 mutant, Dmu-mSox2, which has much lower activity.4 As expected, Dmu-mSox2 activated the Rexp500 reporter poorly in 293T and P19 cells (Fig. 4, B and C, lanes 11–13).
The fact that Nanog, Oct-3/4, and Sox2 can all activate the Rex-1 promoter raises the possibility that these proteins may synergize with each other. To test this possibility, we cotransfected Nanog alone or in combination with Oct-3/4, Sox2, and Dmu-mSox2 with Rexp500 in 293T cells. As shown in Fig. 4D, Sox2 appear to be able to superactivate the Rex-1 promoter already activated by Nanog (lanes 3 and 4 versus 2), whereas Dmu-mSox2 failed to do so. On the other hand, Oct-3/4 plus Nanog and Oct-3/4 plus Sox2 failed to achieve any additional activation beyond the activity of Oct-3/4, Nanog, or Sox2 alone (data not shown). Unfortunately, the cooperativity among these three factors could not be demonstrated in P19 cells, suggesting that additional proteins may participate in the regulation of Rex-1. Nevertheless, our data demonstrate that Nanog is able to regulate Rex-1 with Sox2.
Role of Nanog C Terminus in Mediating Rex-1 Activation—We have shown previously that the C-terminal domain of Nanog contains the main transactivation activity toward synthetic reporter constructs bearing Nanog-binding sites (11, 12). It is not clear whether the C terminus of Nanog is responsible for the observed activation of Rex-1. To this end, we tested the N1 Nanog mutant against the Rex-1 reporter. As shown in Fig. 5B, N1 is about one-third as active as the wild type Nanog in activating Rexp500 in NIH3T3 cells (lane 3 versus 1). However, N1 appeared to be inactive against Rex-1 reporter in P19 cells (Fig. 5C, lane 3 versus 1). Together, these data demonstrate that the C terminus of Nanog is required for Nanog-mediated Rex-1 activation. Given the fact that CD2 of Nanog is a potent transactivator, even comparable with the most active transactivator Vp16 from human herpesvirus (12), we constructed a chimera between Nanog and the transactivation domain from the viral protein Vp16 of human herpesvirus (Fig. 5A) and tested its activity on Rex-1 reporter. As shown in Fig. 5B, the chimera is a very robust activator of Rex-1 reporter in NIH3T3, outperforming the wild type Nanog significantly (Fig. 5B, lane 4 versus 2). Curiously, the same chimera is less active than the wild type Nanog in P19 cells, suggesting that the C terminus of Nanog interacts with factors that are specifically expressed in P19 or ES cells in mediating Rex-1 activation.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Rex-1 is regulated by multiple transcription factors. As a well known and much analyzed marker for stem cell pluripotency, Rex-1 has attracted considerable attention recently. However, little is known about why Rex-1 is expressed only in ES and EC cells (2). Previously, Oct-3/4 has been identified as a regulator of Rex-1 expression (6). Our findings in this study suggest that two additional factors known to regulate stem cell pluripotency, Nanog and Sox2, also regulate Rex-1 expression. These results reveal for the first time a validated target for Nanog. Furthermore, it appears that Nanog cooperates with Sox2 in regulating Rex-1 expression. Interestingly, we detected little cooperativity between Oct-3/4 and Nanog, suggesting that these two factors may be functioning in the same pathway. Surprisingly, Oct-3/4 and Sox2, known to regulate fgf4 cooperatively, exhibited no cooperativity toward Rex-1 promoter. Thus, Nanog, Sox2, and Oct4 may regulate common targets but through distinct mechanisms.
We reported earlier that Nanog regulates gene transcription through two strong transactivation domains at the C terminus, the CD2 and W-repeat domains, employing synthetic reporters (11, 12). In this study, we show for the first time that these two C-terminal domains are required for Nanog to activate a natural promoter from an important ES cell marker, Rex-1. Although the role of the C-terminal domain of Nanog in maintaining stem cell pluripotency remains unclear, the fact that it is involved in Rex-1 expression suggests that it plays an important role in ES cell self-renewal. The interactive partner for the C terminus of Nanog has not been identified. The Rex-1 promoter-based reporter of this study may serve as an important tool to evaluate Nanog function at the biochemical level and may help to unravel the role of Nanog in mediating stem cell self-renewal and pluripotency.
|
FOOTNOTES |
|---|
1 A Cheung Kong (Changjiang) scholar of the Li Ke-Shing Foundation and Ministry of Education, China. To whom correspondence should be addressed: Guangzhou Inst. of Biomedicine and Health, Guangzhou, China 510663. Tel.: 862032290209; Fax: 862032290606; E-mail: pei_duanqing{at}gibh.ac.cn.
2 The abbreviations used are: ES, embryonic stem; EC, embryonic carcinoma; RA, retinoic acid; RT, reverse transcriptase; ChIP, chromatin immunoprecipitation; GFP, green fluorescent protein; siRNA, small interfering RNA; PBS, phosphate-buffered saline.
3 W. Shi, H. Wang, G. Pan, Y. Geng, Y. Guo, and D. Pei, submitted for publication.
4 W. Shi, H. Wang, G. Pan, Y. Geng, Y. Guo, and D. Pei, manuscript in preparation.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. Toyooka, D. Shimosato, K. Murakami, K. Takahashi, and H. Niwa Identification and characterization of subpopulations in undifferentiated ES cell culture Development, March 1, 2008; 135(5): 909 - 918. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Aksoy, C. Sakabedoyan, P.-Y. Bourillot, A. B. Malashicheva, J. Mancip, K. Knoblauch, M. Afanassieff, and P. Savatier Self-Renewal of Murine Embryonic Stem Cells Is Supported by the Serine/Threonine Kinases Pim-1 and Pim-3 Stem Cells, December 1, 2007; 25(12): 2996 - 3004. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |
