| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1
* Institute of Health Sciences and Institute of Stem Cell Research, Shanghai Jiao Tong University School of Medicine and Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China;
Key Laboratory of Stem Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences and Shanghai JiaoTong University School of Medicine, Shanghai, China; and
Key Laboratory of Cell Differentiation and Apoptosis of Chinese Ministry of Education, Shanghai Jiao Tong University School of Medicine, Shanghai China
1Correspondence: Institute of Health Sciences, 225 South Chongqing Rd., Shanghai, China 200025. E-mail: yjin{at}sibs.ac.cn
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: embryonic stem cells • sumoylation • protein stability • cYes
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In mice, Oct-4 is normally expressed exclusively in pluripotent cells of the developing embryo and their in vitro counterparts, including ES and embryonic germ (EG) cells, respectively. It is absent from the differentiated somatic cell types in vitro and in vivo. Recently, it has been shown that ubiquitous expression of Oct-4 in transgenic mouse embryos was associated with the lethality of embryos (2)
. In addition, involvement of ectopic Oct-4 expression with tumorigenesis has been documented (3
, 4)
. In contrast, deletion of Oct-4 in mice causes complete loss of pluripotent cells in early embryonic life (5)
. In ES cells, Oct-4 knockdown induces expression of trophectoderm differentiation markers, whereas its overexpression induces differentiation into primitive endoderm and mesoderm lineages (6)
. Obviously, Oct-4 is a crucial and concentration-dependent cell fate determinant both in embryonic development and adult life. Therefore, its expression must be strictly controlled in a developmental stage- and cell type-specific manner. However, the precise mechanism by which Oct-4 levels are regulated in vivo is not well understood, and very little is known about its post-translational modification. Previous works in our laboratory have demonstrated that ubiquitin E3 ligase, Wwp2, interacts with Oct-4 and targets it for ubiquitination. Furthermore, we could show that the post-translational modification of Oct-4 by ubiquitin suppresses its transcriptional activity (7)
, suggesting post-translational modification is involved in controlling the function of Oct-4.
Post-translational modification of transcriptional factors is an important mechanism to achieve dynamic regulation of gene expression. In addition to ubiquitin, the small ubiquitin-like modifier (SUMO-1) has attracted a great deal of interest due to its involvement in a wide range of cellular processes, including gene transcription, cell cycle, protein stability, protein-protein interaction, subnuclear localization, and chromatin dynamics (8)
. Similar to ubiquitination, sumoylation is performed by an enzymatic cascade. SUMO-1 is first activated by E1 activating enzyme, a heterodimer of Aos1 and Uba2 subunits, in an ATP-dependent manner and transferred to E2 conjugating enzyme Ubc9 and then covalently attached to substrate proteins via an isopeptide bond between its C-terminal glycine and a lysine residue in the substrate proteins (9)
. In contrast to the ubiquitin system, it was proposed that E3 ligase would not be required for SUMO-1 conjugation, although three classes of E3 ligase, PIAS, RanBP2, and Pc2, have been identified for promoting transfer of SUMO from Ubc9 to specific substrates (10)
. Notably, Ubc9 is the only known SUMO E2 conjugating enzyme (11)
, and it can directly bind substrate proteins. The typical motif for sumoylation consists of the sequence 
KxE, where is a large hydrophobic residue, K is the site of SUMO modification, x is any residue, and E or D is an acidic residue (12)
. The essential role for Ubc9 in mammalian development is underscored by a recent Ubc9-knockout experiment, where Ubc9-deficient embryos died at the early post-implantation stage (13)
. Moreover, Taylor and LaBonne found that post-translational modification of the SoxE transcription factors by SUMO regulates specific developmental programs, highlighting the biological significance of sumoylation in gene expression (14)
. Identification of more SUMO substrates and investigation of functional consequences of their sumoylation will provide new insights about the functions of the SUMO pathway.
To explore the molecular mechanism controlling Oct-4 function in ES cells, we constructed a mouse ES cell cDNA library and performed yeast two-hybrid screening using the POU domain of transcription factor Oct-4 as bait. Intriguingly, Ubc9 was found to interact with Oct-4. Therefore, SUMO modification of Oct-4 was investigated and SUMO-1 acceptor site in Oct-4 was identified. We subsequently went further to compare the Oct-4 protein stability and self-renewal capacity between ES cells expressing wild-type Oct-4 and cells expressing sumoylation-deficient Oct-4. In addition, the study shows that cYes could be a new Oct-4 downstream gene. The findings provide novel clue for elucidating mechanisms by which Oct-4 function is precisely controlled in ES cells.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and electroporated with Flag-tagged SUMO or Ubc9 plasmids to generate the stable cell line. ZHBTc4 ES cells (a kind gift from Austin Smith) were cultured as described (6)
. To establish the stable ES cell line expressing the wild-type or mutant Oct-4 (K118R, E120A), ZHBTc4 ES cells were electroporated with 50 µg plasmid DNA and cultured in the presence of 1 µg tetracycline (Tc) per ml. Puromycin was used as the selection marker. The pools of stable ES cell lines were obtained and used in the experiments.
Plasmid construction and expression of recombinant proteins
Oct-4 cDNA containing POU domain (amino acid residues 127–282) was prepared by PCR from full-length mouse Oct-4 cDNA (7)
and introduced into pGBKT7 as bait of yeast two-hybrid. Oct-4 mutants were generated by PCR-based site-directed mutagenesis. Wild-type and mutant Oct-4 were cloned into either pET-30a (+) (Novagen, Madison, WI, USA) and pGEX-4T-1 (Amersham Biosciences, Uppsala, Sweden) vectors for expression in bacteria or pCMV-Not, pcDNA3 (Invitrogen, Carlsbad, CA, USA), pcDNA-Flag (a kind gift from Gang Pei), and pPyCAGIP (a kind gift from Ian Chambers) vectors for expression in mammalian cells. Ubc9 and SUMO-1 were obtained by RT-PCR and cloned as amino-terminal FLAG-tagged proteins in pcDNA-Flag or pPyCAGIP. For in vitro assay, Ubc9 cDNA was subcloned into pGEX-4T-1. SUMO-1 cDNA was prepared to generate the plasmid expressing GST or His tagged SUMO-1 (amino acid residues 1–97). Mouse Aos1 and Uba2 cloned in pET were kindly provided by Marry Dasso.
Yeast two-hybrid screening
The BD MatchmakerTM Library construction & Screening Kit (Clontech, Palo Alto, CA, USA) was utilized to construct the two-hybrid library of mouse ES cells and the screen was performed with mouse Oct-4 POU domain (amino acid residues 127–282) as bait according to the manufacturer’s instructions. Positive clones were selected based on the ability of the cells to grow on Trp, Leu, His, and Ade dropout media supplemented with 2.5 mM of 3-aminotriazole (3-AT, an inhibitor of HIS3), and blue colony color on plates of medium containing 5-bromo-3-indoyl-ß-D-galactopyranoside (X-gal). Plasmid DNA from positive colonies was isolated, propagated in E. coli, and sequenced to identify encoded genes.
GST pull-down and immunoprecipitation assays
For GST pull-down experiments, 0.5 µg of GST fusion proteins were incubated with 0.5 µg of His fusion proteins in 250 µl of TBS-N (20 mM Tris-HCl, pH 7.6, 200 mM NaCl, and 0.1% Nonidet P-40) at 4°C for 2 h followed by the addition of glutathione-sepharose 4B beads for another hour. For immunoprecipitation, cells were treated with 1% formaldehyde for 15 min at room temperature, and the crosslinking reaction was stopped by adding glycine to a final concentration of 0.125 M. Cell lysates were prepared in a IP buffer (50 mM Tis-HCl, pH 7.5; 150 mM NaCl; 0.5% Nonidet P-40; 2 mM EDTA; 1 mM NaF; 10% glycerol; and 1 mM phenylmethylsulfonyl fluoride). Following sonicate and centrifuge, the supernatant was incubated with anti-FLAG M2 affinity beads (sigma) for 2 h at 4°C. The samples from immunoprecipitation or GST pull-down assays were analyzed by Western blot.
To detect the endogenous interaction between Ubc9 and Oct-4, nuclear extract from F9 stable cell line expressing Flag tagged Ubc9 was incubated with anti-FLAG M2 affinity beads overnight at 4°C. After immunoprecipitation, Western blotting was performed using the antibody against Oct-4.
In vitro sumoylation assay
For in vitro sumoylation assay, the purified GST-Oct-4 protein was incubated with a reaction mixture containing 50 mM Tris-HCl (pH 7.4), 2 mM DTT, 5 mM ATP, 10 mM MgCl2, His-Aos-1 (150 ng), His-Uba2 (400 ng), GST-Ubc9 (500 ng), and GST- or His- tagged SUMO-1 (amino acid residues 1–97, 1 µg) for 2 h at 30°C, and the Oct-4 protein was detected by Western blotting with anti-Oct-4 antibody.
Detection of sumoylation in vivo
HEK293 cells were cotransfected with wild-type or mutant Oct-4 with a Flag-tagged SUMO-1 plasmid. Cell were harvested and boiled in lysis buffer (100 mM Tris-HCl, pH7.5, 1% SDS). The lysate was then diluted 10-fold in IP buffer and sonicated briefly, followed by centrifugation at 14000 g for 20 min. The supernatant was immunoprecipitated with anti-FLAG M2 affinity beads. To detect SUMO-1 modification of endogenous Oct-4, the stable cell lines of CGR8 ES cells or F9 cells transfected with pPyCAGIP vector or SUMO-1-pPyCAGIP were established.
Reverse transcriptase-polymerase chain reaction (RT-PCR)
Total RNA was extracted from the cells using TRIzol (Invitrogen) and reverse-transcribed into cDNA using oligo (dT)15 and ReverTra Ace reverse transcriptase (Toyobo, Osaka, Japan). PCRs were carried out with 1 µl of cDNA template, 250 nM of each primer, 200 uM dNTP mix, and 1 U of TaqDNA polymerase (Hua Nuo, Shanghai, China) in a volume of 25 µl. Samples were amplified in a thermocycler under the different conditions. The primers used were: cYes, 5' TGGGATGTCTAATGTTGGTG 3' (forward) and 5' CAGGGATTTGAAACTTGGTG 3' (reverse); Oct-4, 5' ATGGCATACTGTGGACCTCA 3'(forward) and 5' AGCAGCTTGGCAAACTGTTC 3'(reverse); GAPDH, 5' GTCGTGGAGTCTACTGGTGTC 3'(forward) and 5' GAGCCCTTCCACAATGCCAAA 3' (reverse).
Quantitative real-time PCR
Raw data were obtained on an ABI PRISM 7900 using fluorogenic SYBR Green I double-stranded DNA-binding dye chemistry during real-time RT-PCR. Total RNA (6 µg) was reverse transcribed to cDNA after treatment with RNase-Free DNase. For analysis, cDNA was mixed with 250 nM forward and reverse primers, water, and 2xSYBR Green I master mix to a total volume of 10 µl and amplified by 40 PCR cycles. To adjust for variations in loading, the Ct values for each gene were normalized against that for the housekeeping gene GAPDH. Amplification data were analyzed by the Sequence Detection System 2.0 software (ABI, Foster City, CA, USA).
Colony forming assay
ZHBTc4 ES cells expressing the wild-type or mutant Oct-4 (K118R, E120A) were trypsinized to obtain single cell suspension and 1000 cells plated per 3.5 cm cell culture dish. The cell culture medium is same with that for the ZHBTc4 ES stable cell lines. After 7 days, culture plates were stained for alkaline phosphatase and colonies were scored in the categories: pure or mixed stem cell and fully differentiated.
Chromatin immunoprecipitation assay (CHIP)
Mouse ES cells were cross-linked/fixed in 1% formaldehyde for 10 min at room temperature. The crosslinking was halted with 125 mM glycine treatment for 5 min at room temperature. Cells were extracted in cell lysis buffer (5 mM PIPS pH 8.1, 85 mM KCl, 0.5% Nonidet P-40 with protease inhibitors) for 10 min on ice. Nuclei were recovered by centrifugation and extracted with nuclei lysis buffer (50 mM Tris-HCl pH 8.1, 10 mM EDTA, 1% SDS with protease inhibitors) for 10 min on ice, and sonicated chromatin to averaged length of 
0.5–1 kb while keeping samples on ice. Nuclear lysates were centrifuged, and the supernatant was precleared with protein A sepharose. Chromatin was captured with rabbit IgG or specific antibody for Oct-4. After overnight capture the chromatin was precipitated with protein A sepharose, and washed sequentially twice in dialysis buffer and four times in IP wash buffer (100 mM Tris-HCl pH 7.5, 25 mM EDTA, 1.25% SDS). Chromatin was eluted with IP elution buffer (50 mM NaHCO3, 1% SDS) for 30 min at room temperature, and RNase A and NaCl were added to the eluate. Chromatin was incubated at 67°C for 4–5 h to reverse formaldehyde cross-link. DNA was obtained through ethanol precipitation and used for PCR analysis. The following PCR primers were employed: cYes, 5' TTGTACCTCAGTTCCTCCTAAACC 3' (forward) and 5' AGAGTGGGACAGCATTGTATTTG3' (reverse); Rex1, 5' AAATGACCGGTACCTCCCTGATAAG3'(forward) and 5'ACCCAGAGCCACAGTGGAAATCTAG3' (reverse); GAPDH, 5' CCCATGTTTGTGATGGGTGTG3' (forward) and 5' TGGCATGGACTGTGGTCATGA 3'(reverse).
All experiments were conducted for at least three times and representative data were shown.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To confirm the interaction between Oct-4 and Ubc9, we carried out a GST pull-down assay with a bacterially expressed GST fusion protein of Ubc9 and His fusion protein of Oct-4. As shown in Fig. 1
A, the immobilized GST-Ubc9 could pull down His-Oct-4, whereas GST alone was ineffective. This indicates that Ubc9 associates with Oct-4 directly in vitro. We next examined whether the two proteins interact in vivo. Coimmunoprecipitation experiments were performed with the cell lysates of HEK 293 cells expressing exogenous Oct-4 and Flag-tagged Ubc9 or vector alone. As shown in Fig. 1B
, Oct-4 was coimmunoprecipitated with Flag-tagged Ubc9, when both Oct-4 and Flag-tagged Ubc9 were expressed. As a negative control, Oct-4 was not detected in the Flag antibody-immunoprecipitated complexes when cells were expressing either Oct-4 or Flag-tagged Ubc9 alone. This result suggests that a specific interaction exists between Ubc9 and Oct-4 in mammalian cells.
|
To demonstrate specific interaction between Oct-4 and Ubc9 with endogenous protein levels, we generated a F9 cell line stably expressing Flag-tagged Ubc9 and coimmunoprecipitation experiments were conducted with the nuclear extract of the cells. As shown in Fig. 1C
, endogenous Oct-4 was readily detected in the Flag-antibody-immunoprecipitated protein complex in F9 cells expressing Flag-tagged Ubc9. By contrast, Oct-4 was not present in Flag-antibody-immunoprecipitated protein complex when the nuclear extract was prepared from cells transfected with vector, although endogenous Oct-4 protein level in the nuclear extract from the vector-transfected cells was comparable with that from Flag-Ubc9-transfected cells (Fig. 1C
, bottom row). Therefore, we conclude that Ubc9 is a novel Oct-4-interacting protein.
Oct-4 is modified by SUMO-1 both in vitro and in vivo
Since Ubc9 is an E2 conjugation enzyme for protein sumoylation, we sought to determine whether Oct-4 could be post-translationally modified by SUMO-1. An in vitro sumoylation assay was performed using bacterially expressed GST-Oct-4 as a substrate. The GST-Oct-4 was incubated with various combinations of bacterially expressed E1 (Aos1/Uba2 heterodimer), E2 (GST-Ubc9), and a mature form of His-SUMO-1. Western blotting analysis of the reaction products with anti-Oct-4 antibody revealed a mobility shift to a slowly migrating Oct-4 band of around 100 kDa only when E1 and E2 as well as His-SUMO-1 were present (Fig. 2
A, lane 4), and the presence of this band was Oct-4-dependent (not in lane 5 of Fig. 2A
). To further confirm that the slowly migrating form of Oct-4 was indeed due to conjugation with SUMO-1, we conducted in vitro sumoylation assay using GST-SUMO-1 in place of His-SUMO-1. A more slowly migrating Oct-4 band was detected in the presence of E1, E2, and GST-SUMO-1 (Fig. 2A
, lane 6), confirming the identity of the slowly migrating band. These results clearly demonstrate that Oct-4 is a substrate of sumoylation and Ubc9 as its interacting protein is essential for the process.
|
We next examined whether Oct-4 could be sumoylated in vivo. To probe for a sumoylated form of cellular Oct-4, HEK 293 cells were transfected with expression vectors encoding Oct-4 and Flag-tagged SUMO-1. The sumoylated proteins in the cell lysate were immunoprecipitated by anti-Flag M2 affinity beads and then analyzed by immunoblotting with anti-Oct-4 antibody. Coexpression of Oct-4 with Flag-tagged SUMO-1 caused sumoylation of exogenously expressed Oct-4 in the higher molecular weight form of around 75 kDa (Fig. 2B
, top panel). As a negative control, the sumoylated Oct-4 was not detected in immunoprecipitated complexes from the cells transfected with either Oct-4 or Flag-tagged SUMO-1 alone, although the protein level of Oct-4 and Flag-tagged SUMO-1 in the lysate was comparable with that in the cells expressing both (Fig. 2B
, middle and bottom panel). Lastly, we tested whether endogenous Oct-4 is modified by SUMO-1. To facilitate the detection of endogenous sumoylated Oct-4, we generated a mouse ES cell line (CGR8) stably expressing Flag-tagged SUMO-1. Flag-sumoylated Oct-4 was readily detected with anti-Oct-4 antibody in the immunocomplexes precipitated by anti-Flag antibody (Fig. 2C
). However, we did not find any Flag-sumoylated Oct-4 in the cells transfected with vector alone. The same result was obtained when the experiment was conducted in F9 cells stably expressing Flag-tagged SUMO-1 (data not shown). Taken together, our data indicate that both exogenously and endogenously expressed Oct-4 is subject to modification by SUMO-1.
Lysine residue at 118 is identified as major SUMO acceptor site for Oct-4
SUMO-1 is covalently attached to the lysine residues of substrate proteins, and it is known that the sumoylated lysine residues in many substrates locate within a consensus sequence of KxE (12)
. Studying the amino acid sequence of Oct-4, the sequence around lysine 118 (amino acid 117–120, VKLE) highly matches the consensus sequence for sumoylation. Therefore, we created an Oct-4 mutant (Oct-4 K118R), in which the lysine residue at 118 was changed into a similarly charged arginine residue. As shown in Fig. 3
A, the mutation completely abolished the slowly migrating SUMO-1 modified form of Oct-4 in sumoylation in vitro assay, indicating that lysine residue 118 is an Oct-4 SUMO-1 acceptor site. To test whether lysine residue 118 is the major SUMO-1 acceptor site of Oct-4 in vivo, a sumoylation assay was performed with lysate from HEK 293 cells expressing Flag-tagged SUMO-1 in combination with wild-type Oct-4 or mutant Oct-4 (K118R). The assay also included another mutant form of Oct-4, in which lysine residue at 244 was mutated into arginine, Oct-4 (K244R), since the sequence around lysine residue 244 also matches consensus sumoylation motif. As expected, a sumoylated form of Oct-4 can be detected with the wild-type of Oct-4 (Fig. 3B
, lane 3), but not Oct-4 (K118R) (Fig. 3B
, lane 4). In contrast, the mutant Oct-4 (K244R) behaved like wild-type Oct-4 (Fig. 3B
, lane 5). To further verify the finding described above, the same sumoylation assay in HEK 293 cells was performed with C-myc-tagged SUMO-1, the same results were obtained (data not shown), further confirming lysine residue at 118 is the major SUMO-1 modification site for Oct-4. Thus, we conclude that Oct-4 is sumoylated at lysine residue 118.
|
Disruption of Oct-4 sumoylation promotes its degradation through 26S proteasome
Sumoylation has been reported to affect stability of substrate proteins (15
, 16)
. We were prompted to examine whether sumoylation of Oct-4 affected its stability. To this end, stable ES cell lines constitutively expressing either wild-type Oct-4 or mutated Oct-4 (K118R) were generated with ZHBTc4 mouse ES cells, where Oct-4 is expressed only as a transgene and controlled by Tet-off system (6)
. The Tet-off system-regulated Oct-4 expression was suppressed by treating the cells with tetracycline (Tc). In addition, there is possibility that lysine residue at 118 of Oct-4 might be the target of a modification other than SUMO. Therefore, any functional consequences resulted from mutated Oct-4 (K118R) could be a result of inability of mutated Oct-4 to be modified by other mechanisms, but not by sumoylation. To test whether SUMO modification itself could affect Oct-4 stability, we mutated glutamic acid (E) residue at 120 in KxE motif into alanine (A, E120A), since the E residue at +2 of the motif has been shown to disrupt SUMO conjugation to K (12)
. To prove that the E residue mutation at 120 indeed disrupts Oct-4 SUMO conjugation, we conducted sumoylation assay using the cell lysate from HEK 293 cells expressing Flag-SUMO-1, together with either wild-type or mutant (K118R), or mutant (E120A) Oct-4. Similarly to the result shown in Fig. 3B
, SUMO-modified Oct-4 was readily detected when wild-type Oct-4 was expressed in the cells (Fig. 4
A, lane 3), but it was not detected when Oct-4 mutant (K118R) was expressed (lane 4). Importantly, SUMO conjugation to Oct-4 diminished sharply when Oct-4 mutant (E120A) was expressed (lane 5), confirming the prediction that mutation in the E120 of the consensus motif could disrupt SUMO modification at K118 residue of Oct-4. Afterward, the ES cell line stably expressing Oct-4 mutant (E120A) was also generated in ZHBTc4 ES cells. Next, ZHBTc4 ES cells expressing wild-type Oct-4 or mutated Oct-4 (K118R) and (E120A) were treated with Tc to suppress the Tet-off system-regulated Oct-4 expression and steady-state level of Oct-4 protein in the cell lysate of all three cell lines was measured by immunoblotting with anti-Oct-4 antibody. As shown in Fig. 4B
, the protein level in the mutated Oct-4 (K118R) and (E120A) cell lines was significantly lower than that in the wild-type Oct-4 cell line, although the expression of control protein, tubulin, was similar in all three cell lines. We consistently observed a faster mobility of mutated Oct-4 (K118R) and (E120A) than wild-type Oct-4 in Western blotting. It could be a change in Oct-4 conformation mediated by the mutation. We further performed RT-PCR to examine whether there was a decrease in Oct-4 mRNA level in the cells expressing mutated Oct-4. The result showed that no detectable difference in the Oct-4 mRNA level existed among these three cell lines (Fig. 4C
). We suspected that mutated Oct-4 has an enhanced degradation. To test this hypothesis, steady-state level of Oct-4 protein in the cells expressing the mutated Oct-4 (K118R) and (E120A) as well as wild-type Oct-4 was evaluated by Western blotting when the cells were treated with specific 26S proteasome inhibitors (lactacystin at 30 µM and epoxomicin at 10 µM), and compared to Oct-4 protein level when the cells were not treated with the inhibitors. Obviously, the steady-state Oct-4 protein level in the mutated Oct-4 cell lines (both K118R and E120A) was significantly increased in the presence of the inhibitors, although the treatment only enhanced Oct-4 protein level in wild-type Oct-4-expressing cells slightly (Fig. 4D
). In addition, similar Oct-4 protein level was observed in all three cell lines when the cells were treated with the inhibitors. The observation indicates that the lower Oct-4 protein level in the cells expressing mutated Oct-4 was caused by enhanced degradation through 26S proteasome.
|
Sumoylation of Oct-4 is important for maintaining self-renewal in ES cells
One of unique features of ES cells is its self-renewal ability and Oct-4 is known as one of key factors controlling self-renewal ability in ES cells. We wanted to investigate whether SUMO-1 modification of Oct-4 plays a role in its ability to control self-renewal. Therefore, colony forming assay, which has been widely used to determine the capacity of ES cells to maintain self-renewal (17)
, was conducted. The ZHBTc4 ES cells of three lines described above, expressing wild-type Oct-4, mutated Oct-4 (K118R) and (E120A), respectively, were cultured in the presence of Tc and seeded at low density. The colonies formed in 7 days were stained with alkaline phosphatase to highlight undifferentiated ES cells and scored in three categories: pure stem cell, mixed and fully differentiated, as described by Chambers et al. (18)
. We found that percentage of fully differentiated colony was significantly higher in the mutated Oct-4-expressing cells (both K118R and E120A) than that in wild-type Oct-4-expressing cells (Fig. 5
A). Furthermore, When Oct-4 protein level in the cells cultured in same condition was measured by Western blotting, it was found that the mutated Oct-4-expressing cells had significantly lower Oct-4 level than that in wild-type Oct-4-expressing cells (Fig. 5B
). The observation is consistent with the lower Oct-4 level in the mutated Oct-4-expressing cells when they were cultured under regular density (Fig. 4B
). This result establishes an important role of Oct-4 sumoylation in maintenance of self-renewal capacity in ES cells.
|
Mouse ES cell self-renewal is sustained by the cytokine leukemia inhibitory factor (LIF) (19)
, although human ES cells seem not dependent on it (20)
. Withdrawal of LIF from culture media will induce mouse ES cells to differentiate and reduce Oct-4 expression. To test whether the sumoylation play a role in LIF-induced decrease in Oct-4 expression, we analyzed the steady-state Oct-4 protein level in three ZHBTc4 ES cell lines cultured in the absence of LIF from 2 to 6 days. The data in Fig. 5C
show that the expression of Oct-4 declined gradually when cells were deprived of LIF in all three cell lines. This was anticipated from previous experiments. However, the reduction in Oct-4 protein level after withdraw of LIF was substantially faster in Oct-4 mutated cells than in cells expressing wild-type Oct-4. This observation further confirms our earlier finding that disruption of Oct-4 sumoylation promotes its degradation. The result also indicates that SUMO modification of Oct-4 is involved in regulation of Oct-4 expression level during LIF withdrawal-induced ES cell differentiation.
cYes expression is regulated by Oct-4 sumoylation
As a transcriptional factor, Oct-4 functions mainly through activation or repression of target gene expression. We reasoned that gene expression profile in mutated Oct-4-expressing ES cells would be different from that in wild-type Oct-4-expressing ES cells, if SUMO modification plays an important role in transcriptional activity of Oct-4. Analysis of such expression profile could discover novel genes regulated by Oct-4, especially those genes whose expression might be affected by Oct-4 sumoylation. For this purpose, gene expression pattern in three established ZHBTc4 ES cell lines, including cell lines expressing wild-type Oct-4, mutated Oct-4 (K118R), and mutated Oct-4 (E120A) grown in the presence of Tc, was profiled using the Affymetrix Mouse Genome 430 2.0 Chip. Of the genes whose expression varied between wild-type Oct-4- and the mutated forms of Oct-4-expressing cells, our attention was drawn to cYes (NM009535). The expression of cYes in the mutated Oct-4 expressing ES cells (K118R mutation and E120A mutation) was significantly lower than that in the wild-type Oct-4-expressing ES cells, being 30% and 45% of the wild-type Oct-4- expressing cells for K118R mutation and E120A mutation, respectively. cYes is a member of the Src family of nonreceptor tyrosine kinases. It has been reported to highly express in mouse and human ES cells and is important for ES cell self-renewal. Therefore, we were interested in further investigation of relationship between cYes expression and Oct-4 sumoylation.
To verify the expression pattern of cYes gene detected by GeneChip, its expression in the three established ZHBTc4 ES cells was quantitatively determined by real-time PCR. Consistent with the result from GeneChip, mRNA level of cYes in the wild-type Oct-4- expressing ES cells was much higher than that in the mutated Oct-4 expressing ES cells (Fig. 6
A), suggesting that sumoylation status of Oct-4 might be involved in regulation of cYes expression. If cYes expression varies when SUMO modification of Oct-4 is disrupted, down-regulation of Oct-4 expression would also affect cYes expression. Thus, the parental ZHBTc4 ES cells were treated with Tc for 12 h, and then mRNA levels of both Oct-4 and cYes were examined by RT-PCR. As expected, Oct-4 mRNA levels in the cells were rapidly diminished (Fig. 6B
). Importantly, expression of cYes was reduced in parallel with Oct-4, indicating that Oct-4 might regulate cYes expression in ES cells. To obtain more evidence for the control of cYes expression by Oct-4, we searched for potential Oct-4 binding sites in genomic sequence region from –1000 to +300 base pairs of cYes transcription start site (http://www.cbrc.jp/htbin.nph-tfsearch). A putative Oct-4 binding site was found. Next, chromatin immunoprecipitation assay (CHIP) was performed with affinit-purified Oct-4 antibody and PCR primers specifically binding to DNA sequences flanking the putative Oct-4 binding site. As shown in Fig. 6C
, Oct-4 was specifically recruited to cYes promoter region, as it was at the promoter of Rex1, which is a well-known Oct-4 downstream gene. For negative control, equally weak PCR signal was found with PCR primers designed for GAPDH in IgG and Oct-4 antibody immunoprecipitated complexes, demonstrating the equal amounts of input and antibody used for IgG and Oct-4 precipitation. Taken together, our findings suggest that cYes could be a novel downstream gene of Oct-4 and its expression is regulated by Oct-4 sumoylation status.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
POU domain transcription factor Oct-4 plays a crucial role in maintaining self-renewal and pluripotent potential of ES cells. Recently, it has been proposed that Oct-4 is a concentration-dependent oncogenic fate determinant (4)
. However, how Oct-4 expression level is controlled in a developmental stage- and cell type-specific manner is still unknown. We have previously reported that the murine ubiquitin E3 ligase, Wwp2, targets Oct-4 for ubiquitination and promotes its degradation (7)
. It is not surprising to find both sumoylation and ubiquitination are involved in regulating Oct-4 function, since post-translational modification of cellular proteins has been proven to be a rapid and efficient means of controlling their functions, and the crosstalk between ubiquitination and sumoylation is found in a growing number of proteins (21)
. Generally, SUMO acts as an ubiquitin antagonist in a competitive manner. For example, SUMO blocks ubiquitination of I
B by direct competition for the same attachment site (22)
. The SUMO modification sites of Smad4 are also targeted by the ubiquitin pathway, such that mutation of these sites blocks sumoylation but at the same time reduces its ubiquitination, resulting in enhanced stability of Smad4 (15)
. On the other hand, cooperation between SUMO and ubiquitin has also been documented (21)
. Recent investigation of NEMO, the regulatory subunit of IKK, suggests that SUMO and ubiquitin act sequentially on NEMO to mediate NF-B activation in genotoxic stress (23)
. Here, we report that disruption of the sumoylation in Oct-4 resulted in an enhanced degradation of Oct-4 through 26S proteasome. Although its underlying mechanism is unclear, it seems likely that sumoylation protects Oct-4 from degradation through antagonizing ubiquitin modification of Oct-4. In addition, lysine residue at 118 appears to be important for Oct-4 protein stability and function. The observation that the mutation of glutamic acid residue into alanine residue at 120 has a similar effect on Oct-4 protein stability and function as the mutation of lysine residue into arginine at 118 of Oct-4 provides direct experimental support for the important role of sumoylation. Sumoylation might alter configuration of Oct-4 or interfere with its association with other proteins to reduce its ability to interact with ubiquitin system, although the interaction between Oct-4 and Wwp2 was not affected by the mutation of lysine residue at 118 (data not shown). Ubiquitination of Oct-4 in the three established ZHBTc4 mouse ES cells, stably expressing wild-type Oct-4 and mutated forms of Oct-4 (K118R and E120A), is under investigation, which will facilitate our understanding how sumoylation and ubiquitination pathways interact to regulate Oct-4 protein level and functions. It is worthy to note that Oct-4 protein level was evidently lower in the mutated Oct-4- expressing ES cells than that in the wild-type Oct-4-expressing cells (Fig. 4B
and Fig. 5B
), suggesting that post-translational modification of Oct-4 contributes to control of Oct-4 expression in undifferentiated ES cells. Furthermore, we provide experimental evidence for the importance of SUMO modification in maintenance of Oct-4 expression in the differentiating ES cells (Fig. 5C
).
Although the effects of SUMO modification on transcriptional factors are diverse, there is growing evidence that post-translational modification by SUMO plays important roles in regulating transcription factor activity. Inhibition of transcription has been found in most cases. For example, sumoylation of transcription factors Elk1 (24)
, C-Jun (25)
, and Sp3 (26)
leads to a negative regulation of their transactivation function. However, SUMO-1 modification activated GATA4-dependent cardiogenic gene activity and increased DNA binding activity of p53 (27
, 28)
. Therefore, it is possible that sumoylation affects Oct-4 transcriptional activity in a gene-specific manner. In the present study, we found that disruption in Oct-4 SUMO modification reduced cYes gene expression significantly, demonstrating that SUMO modification is required for cYes gene expression in mouse ES cells. Although Oct-4 protein level was significantly lower in the mutated Oct-4 expressing ES cells, we did not observe a reduction in the expression of Rex1 (data not shown), which is known target gene of Oct-4, in these cells. Therefore, we believe that it is disruption in Oct-4 sumoylation itself, rather than a reduced Oct-4 protein level, that results in a decrease in cYes expression in the mutated Oct-4 expressing ES cells. We first propose that cYes might be a novel target gene of Oct-4, based on the following two observations: 1) Oct-4 specifically binds to cYes gene promoter; 2) downregulation of Oct-4 leads to decreased expression of cYes. Recently, several groups identified Oct-4 target genes with genome-wide CHIP assays (29
, 30)
and the possible recruitment of Oct-4 to genomic sequence of cYes gene was reported in the study (29)
. Furthermore, cYes has been found highly expressed in both mouse and human ES cells (31)
and is important for ES cell self-renewal (32)
. Anneren et al. proposed that cYes may represent a third, independent pathway, downstream of LIF in mouse ES cells. Our finding that Oct-4 regulates cYes expression further establishes the important role for cYes in maintaining undifferentatiated state for ES cells. In line with this, cYes may act as a common target of Oct-4 and LIF to maintain ES cells in an undifferentiated state.
During the revision and review process of this manuscript, Tsuruzoe et al. reported that Oct-4 is sumoylated on lysine residue 118 (33)
, further confirming our findings. Our study demonstrates that SUMO and ubiquitin could jointly exert tight control over intracellular protein level of Oct-4 and cYes may be a new downstream gene of Oct-4 in maintaining ES cells in an undifferentiated state. A more detailed understanding of the underlying mechanisms should shed more light on the function of sumoylation in development and ES cell proliferation.
|
ACKNOWLEDGMENTS |
|---|
Received for publication July 31, 2006. Accepted for publication April 19, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
