The basic helix-loop-helix transcription factor OLIG2 is specifically expressed in cells of the oligodendrocyte lineage. It is also expressed in various tumors originating from glial cells; however, the expression of OLIG2 is rare or weak in glioblastomas, the most malignant gliomas. The role of OLIG2 in glioma remains unclear. To investigate the function of OLIG2 in glial tumor cells, we have established a glioblastoma cell line, U12-1, in which the expression of OLIG2 is induced by the Tet-off system. Induction of OLIG2 resulted in suppression of both the proliferation and anchorage-independent growth of U12-1. It also resulted in an increase in the expression of p27Kip1. A luciferase assay revealed that the CTF site of the p27Kip1 gene promoter was essential for OLIG2-dependent activation of p27Kip1 gene transcription. Electrophoretic mobility shift assays confirmed that a nuclear extract of OLIG2-expressing U12-1 cells contained a protein complex that binds to the CTF site of the p27Kip1 gene promoter. Furthermore, siRNA against p27Kip1 rescued the OLIG2-mediated growth and DNA synthesis inhibition of U12-1 cells. These results indicate that OLIG2 suppresses the proliferation of U12-1 and that this effect is mediated by transactivation of the p27Kip1 gene, and low expression of OLIG2 may be related to the malignant behavior of human glioblastoma.
Oligodendrocyte lineage transcription factor 2 (OLIG2) (Lu et al., 2000) is a member of the OLIG family of basic helix-loop-helix (bHLH) transcription factors and plays a key role in determining the ultimate location of motor neuron and oligodendrocytes in the spinal cord during development (Lu et al., 2000; Takebayashi et al., 2000; Zhou et al., 2001; Zhou et al., 2000). bHLH proteins are classified into six groups, designated A-F, on the basis of their structural and biochemical properties (Atchley and Fitch, 1997; Ledent and Vervoort, 2001). OLIG2 belongs to group A, members of which form heterodimers with other bHLH proteins, commonly referred to as E proteins, and bind to a DNA consensus sequence (CANNTG) known as the E box (Ledent et al., 2002).
Expression of the OLIG2 gene and protein has recently been found to be increased in human gliomas, primary tumors of the central nervous system, and OLIG2 has been proposed as a potential diagnostic marker for oligodendrogliomas (Lu et al., 2001; Marie et al., 2001; Yokoo et al., 2004). OLIG2 is highly expressed in the tumors of oligodendroglial lineage, which follow a relatively favorable clinical course. However, OLIG2 expression is low in the grade IV glioma, glioblastoma (Aguirre-Cruz et al., 2004; Ligon et al., 2004; Mokhtari et al., 2005). We have also shown that the expression level of OLIG2 in glioblastomas is significantly lower than that in other types of gliomas (Ohnishi et al., 2003). The cause of the difference in OLIG2 expression among gliomas remains unclear. We therefore hypothesized that the alteration of OLIG2 expression contributes to the malignant behavior of gliomas.
The p27Kip1 protein is a member of the Cip-Kip family of cyclin-dependent kinase (CDK) inhibitors (CKIs) and plays an important role in the regulation of the G1-S transition of the cell cycle by inhibiting the activities of the cyclin D-CDK4, cyclin E-CDK2 and cyclin-A-CDK2 complexes (Soos et al., 1996). It therefore acts as a tumor suppressor. The expression of p27Kip1 in human gliomas is inversely correlated with the cell proliferation activity (Zagzag et al., 2003). Immunoreactivity for p27Kip1 has been identified as a prognostic marker in oligodendrogliomas (Korshunov and Golanov, 2001; Korshunov et al., 2002) and is virtually absent in glioblastomas (Piva et al., 1997; Schiffer et al., 2002).
We have investigated the function of OLIG2 in cell proliferation using the Tet-off gene expression system in a glioblastoma cell line. Our results indicate that OLIG2 suppresses cell proliferation by increasing the expression of p27Kip1, and that it achieves this latter effect through the CCAAT-box-binding transcription factor (CTF) site in the promoter of the p27Kip1 gene.
Inhibition by OLIG2 of human glioblastoma cell proliferation
To investigate the possible effect of OLIG2 on the proliferation of glioblastoma cells, we established a Tet-off-based human glioblastoma cell line, designated U12-1, in which the expression of OLIG2 is inducible. In these cells, the abundance of OLIG2 was increased markedly 24 hours after removal of doxycycline from the culture medium (Fig. 1). Induction of OLIG2 expression resulted in a significant reduction in the growth rate of U12-1 cells during culture for 5 days, this effect being especially prominent after removal of doxycycline for 3 days (Fig. 2A). OLIG2 expression also inhibited DNA synthesis in U12-1 cells by 40% as revealed by measurement of BrdU incorporation (Fig. 2B). Furthermore, OLIG2 inhibited the anchorage-independent growth of U12-1 cells demonstrated by determination of the size and number of colonies formed by the cells on soft agar plates (Fig. 2C).
Upregulation of p27Kip1 mRNA and protein by OLIG2
To identify genes whose transcription is regulated by OLIG2 and might contribute to the effect of OLIG2 on cell proliferation, we compared the transcriptomes of U12-1 cells that had been cultured in Tet-on or Tet-off medium by cDNA microarray analysis (Table 1). Among genes associated with the cell cycle, the expression of that for CKI p27Kip1 was significantly increased in cells expressing OLIG2 (Table 1). This result thus suggested that OLIG2 might increase the transcriptional activity of the p27Kip1 gene and thereby inhibit cell proliferation.
To confirm this result of the cDNA microarray analysis, we examined the effect of OLIG2 induction on the amount of p27Kip1 mRNA in U12-1 cells by RT-PCR analysis. Removal of doxycycline resulted in a time-dependent increase in the amount of p27Kip1 mRNA in U12-1 cells (Fig. 3A). Immunoblot analysis revealed that induction of OLIG2 expression was also accompanied by a time-dependent increase in the amount of p27Kip1 protein (Fig. 3B). Transient expression of OLIG2 in U12-1 cells also led to an increase of p27Kip1 protein expression (data not shown). These results thus demonstrated that OLIG2 increased the expression of p27Kip1 at both the mRNA and protein levels.
siRNA against p27Kip1 attenuates OLIG2-mediated suppression of proliferation in U12-1 cells
To confirm the role of p27Kip1 in the inhibitory effect of OLIG2 on glioblastoma cell proliferation, we downregulated p27Kip1 expression with siRNA (Fig. 4A) and examined the growth rate and DNA synthesis in OLIG2-expressing U12-1 cells. Downregulation of p27Kip1 significantly attenuated the effect of OLIG2-mediated inhibition of both growth rate (Fig. 4B) and DNA synthesis (Fig. 4C) in U12-1 cells, suggesting that p27Kip1 plays a role in the suppression of glioblastoma cell proliferation in the presence of OLIG2. It has been reported that ectopic expression of p27kip1 blocked the proliferation of p27-47 cells at high density but had little effect on the growth of cells at low density (Zhang et al., 2000). To examine the effect of cell density on the growth of U12-1 cells expressing OLIG2, we performed cell growth experiments using U12-1 cells at high or low density as previously described (Zhang et al., 2000). The inhibition of growth rate of U12-1 cells by OLIG2 was more significant when cells were grown at low density (700 cells/cm2), compared with high density (50,000 cells/cm2) (Fig. 4D).
OLIG2-dependent transactivation of the p27Kip1 gene promoter in U12-1 cells
To investigate the molecular mechanism by which OLIG2 increases p27Kip1 expression, we measured the transcriptional activity of the human p27Kip1 gene promoter in U12-1 cells with the use of a luciferase reporter construct. The activity of the p27PF plasmid, which contains the wild-type p27Kip1 gene promoter (Fig. 5A), was increased about threefold in Tet-off cells expressing OLIG2 compared with that in Tet-on cells not expressing OLIG2 (Fig. 5B). To examine the region that is essential for OLIG2 to enhance p27Kip1 transcriptional activity, we next compared the transcriptional activity in the presence (Tet-off) or absence (Tet-on) of OLIG2 by using a series of 5′ deletion mutants (Fig. 5A). No substantial loss of OLIG2-dependent promoter activity was apparent after deletion of nucleotides -3568 to -550 (Fig. 5B). The transcriptional activity of p27Kip1 promoter in cells transfected with p27No.12 was not different in the presence or absence of OLIG2; however, transcriptional activity in cells transfected with p27No.2 was increased in the presence of OLIG2 (Fig. 5B). Since the difference between the two constructs is that the p27No.12 construct lacks the (-549/-5site, these results suggested that the region of the p27Kip1 gene promoter between nucleotides -549 and -512 is required for OLIG2-dependent transcription.
Given that two Sp1-binding sites (Sp1-1 and Sp1-2) and one CTF site are present within the region of the p27Kip1 gene promoter from nucleotides -549 to -512 (Minami et al., 1997; Zhang and Lin, 1997), we examined the effect of point mutations in these sites on OLIG2-responsive transcriptional activity (Fig. 6A). The OLIG2-dependent activity of p27PF-based reporter plasmids containing point mutations in the Sp1-1 or Sp1-2 sites did not differ markedly from that of the wild-type plasmid (Fig. 6B), suggesting that these sites do not contribute substantially to OLIG2-dependent promoter activity. By contrast, mutation of the CTF site abolished OLIG2-dependent luciferase activity (Fig. 6B). The CTF site thus appears to be essential for OLIG2-induced transactivation of the p27Kip1 gene promoter in U12-1 cells.
OLIG2 is required for the binding of protein(s) to the CTF site of the p27Kip1 gene promoter
To examine the relationship between OLIG2 and the CTF site of the p27Kip1 gene promoter, we performed electrophoretic mobility shift assay (EMSA) analysis with a 21-bp oligonucleotide (5′-CGCCGCAACCAATGGATCTCC-3′) containing the CTF site as a probe and nuclear extracts prepared from U12-1 cells that had been incubated for 48 hours in Tet-off or Tet-on medium (Fig. 7A). A single major band corresponding to a protein-DNA complex was obtained in the nuclear extract expressing OLIG2 (Fig. 7B). The specificity of complex formation was confirmed by competition analysis with the unlabeled probe. These results thus indicated that OLIG2 is required for the formation of a protein complex that binds to CTF site of the p27Kip1 gene promoter. To examine whether this complex contains OLIG2, the nuclear extracts were incubated with anti-OLIG2 antibody and the same amount of rabbit IgG was thereafter applied to the EMSA. However, there was no difference in the shifted bands in the presence of anti-OLIG2 or rabbit IgG (data not shown). Although we additionally performed the EMSA using a recombinant OLIG2 protein instead of the nuclear extracts from OLIG2-expressing U12-1 cells, we did not obtain a positive signal from recombinant OLIG2 protein (data not shown). Thus, these studies were not able to resolve whether OLIG2 directly binds to the CTF site of the p27Kip1 gene promoter.
Although the expression pattern of OLIG2 is known to differ among gliomas (Lu et al., 2001; Marie et al., 2001; Yokoo et al., 2004), the biological function of this bHLH transcription factor in these tumors has been unclear. We have now established a human glioblastoma cell line, U12-1, in which the expression of OLIG2 is inducible. With this cell line, we have shown that OLIG2 suppresses cell proliferation as a result of its induction of the expression of p27Kip1, a CKI that functions as a negative regulator of progression through G1 and S phases of the cell cycle. A colony-formation assay also revealed that OLIG2 suppresses the ability of U12-1 cells to grow in an anchorage-independent manner.
In our study, siRNA against p27Kip1 abrogated OLIG2-mediated suppression of growth and DNA synthesis of glioblastoma cells. In astrocytoma biopsies, the levels of p27Kip1 protein are correlated inversely with tumor grade and are a strong predictor of survival (Ding et al., 2005; Piva et al., 1997; Tamiya et al., 2001). Overexpression of p27Kip1 in various human glioblastoma cell lines, including U251 cells used in our study, suppresses growth of the cells, but not other CDKIs (p16INK4A, p18INK4C, p19INK4D or p21WAF1/CIP1) (Komata et al., 2003). Thus, it is suggested that p27Kip1 seemed to be one of the candidates for the treatment of human malignant glioma cells.
The ectopic expression of p27Kip1 has been reported to block the proliferation of fibroblasts at high density, but had little effect on the growth of cells at low density (Zhang et al., 2000). In this study, OLIG2 suppressed cell growth more significantly at low density compared with high density. The inhibitory effect of the OLIG2-p27Kip1 pathway on cell growth might be differently regulated in each cell type.
PTEN, which is referred to as the major tumor suppressor gene having a crucial role in glioma tumorigenesis and frequently mutated in advanced glioblastomas, has also been known to affect p27Kip1 levels (Mamillapalli et al., 2001). Mutation of PTEN in glioblastoma results in deregulation of maintaining proliferation and survival. Ectopic expression of PTEN in glioblastoma cell lines harboring PTEN mutations results in the inhibition of cellular proliferation and suppression of both soft agar colony formation and tumorigenicity in nude mice (Cheney et al., 1998; Furnari et al., 1997; Morimoto et al., 1999). PTEN has been shown to regulate the ubiquitin-dependent degradation of p27Kip1 through the ubiquitin E3 ligase SCFSKP2 (Mamillapalli et al., 2001) and reintroduction of the PTEN gene in glioblastoma cells inhibits S-phase entry by recruitment of p27Kip1 (Cheney et al., 1999). Given that OLIG2 regulates the transcriptional activity of p27Kip1 in glioblastoma cells, alteration of OLIG2 expression in glial cells might involve in tumorigenesis of glioblastoma, as well as PTEN-mutation.
p27Kip1 was identified as an inhibitor of cyclin-dependent kinase (CDK) complexes, and is a member of the Cip/Kip family of CKIs. CKIs associate with cyclin-CDK complexes to negatively regulate progression through the G1 phase of the cell cycle. Interestingly, cytoplasmic p27Kip1 has recently been shown to play a role in the regulation of cell migration independently of cyclin-CDKs complexes. The C-terminal domain of p27 induces actin cytoskeletal rearrangement in a Rac-dependent manner to mediate cell migration in hepatocellular carcinoma cells (McAllister et al., 2003) and p27kip1 regulates actin dynamics to promote cell migration through inhibition of RhoA activation in mouse embryonic fibroblasts (Besson et al., 2004). By contrast, Baldassarre et al. showed that p27Kip1 binds with stathmin, a well-known microtubule-regulatory protein, to inhibit the migration of fibrosarcoma cells (Baldassarre et al., 2005). In astrocytic tumors, RhoA, RhoB, and Rac1 expression decreases and inversely related to the grade, particularly lowest in glioblastoma (Forget et al., 2002). Thus, OLIG2 might regulate not only proliferation, but also migration and/or invasion of glioblastoma cells via p27Kip1 transcriptional mechanism.
OLIG2 plays an important role in the differentiation of oligodendrocyte precursor cells (OPCs) into oligodendrocytes (Lu et al., 2000; Takebayashi et al., 2000; Zhou et al., 2001; Zhou et al., 2000). OPCs possess a built-in mechanism that stops cell division and initiates differentiation at the appropriate time (Nakatsuji and Miller, 2001; Raff et al., 2001; Tokumoto et al., 2002), and accumulation of p27Kip1 in OPCs is necessary for this withdrawal from the cell cycle (Durand et al., 1997; Durand and Raff, 2000; Levine et al., 2000; Zezula et al., 2001). Recently, it has been reported that the treatment of IFN-γ leads to a decrease in the number of OLIG2-positive cells and to downregulation of p27Kip1 protein expression, resulting in inhibition of OPCs differentiation and enhancement of proliferation (Chew et al., 2005). Suppression of the proliferative activity of U12-1 cells by OLIG2 might be related to differentiation of the tumor cells.
We have also shown that OLIG2 increased the transcriptional activity of the p27Kip1 gene promoter independently of the CTF site. OLIG2 was previously shown to bind to a DNA consensus sequence known as the E box (CANNTG) (Ledent et al., 2002). The nature of the interaction of OLIG2 with this site remains to be determined, however. Several transcriptional factors, including NF-Y, C/EBP and NF-I, are capable of binding to the CCAAT sequence (Chodosh et al., 1988; Dorn et al., 1987; Hooft van Huijsduijnen et al., 1987). OLIG2 might increase the expression levels of these transcriptional factors and thereby indirectly enhance their binding to the CTF site.
In conclusion, we have shown that OLIG2 suppresses the proliferation of glioblastoma cells. Furthermore, our results indicate that OLIG2 achieves this effect by increasing the transcriptional activity of the p27Kip1 gene through the CTF site present in the gene promoter.
Materials and Methods
Human OLIG2 cDNA (GenBank accession number NM_005806) was generated from a brain cDNA library by the polymerase chain reaction (PCR) and was subcloned into the pGEM-T vector (Promega, Madison, WI) to generate pGEM-OLIG2. The OLIG2-coding sequence (nucleotides 105-1076 relative to the transcription start site) was amplified and subcloned into the EcoRI-XbaI sites of pTRE2hyg (BD Clontech, Palo Alto, CA) to yield pTRE2hyg-OLIG2, into the EcoRI-XhoI sites of pCMV-Myc (BD Clontech) to yield pCMV-Myc-OLIG2.
Establishment of a Tet-off cell line
The Tet-off regulator plasmid pTet-Off (BD Clontech) was introduced into the human glioblastoma cell line U251 (RIKEN Cell Bank, Tsukuba, Japan) by transfection. Selection of the transfected cells with G418 (Sigma, St Louis, MO) yielded a stable transformant, designated U12, that was shown by immunoblot analysis to express constitutively the tetracycline (Tet)-sensitive transactivator tTA. The U12 cells were then transfected with pTRE2hyg-OLIG2, which contains the hygromycin-resistance gene, and the transfected cells were subjected to selection with G418 and hygromycin B (Invitrogen, Carlsbad, CA). A Tet-off cell line, designated U12-1, that expresses OLIG2 under the control of tTA was thus established.
U12-1 cells were maintained under an atmosphere of 5% CO2 at 37°C in Tet-on medium, comprising Dulbecco's minimum essential medium (Nissui, Tokyo, Japan) supplemented with 10% fetal bovine serum (Invitrogen), 2 mM L-glutamine (Sigma), penicillin-streptomycin (Sigma), G418 (100 μg/ml), hygromycin B (50 μg/ml), and the tetracycline analog doxycycline (50 ng/ml) (Sigma). Expression of OLIG2 was induced by transfer of the cells to Tet-off medium, which is the same as Tet-on medium but lacks doxycycline.
Analysis of cell-growth capacity
U12-1 cells (1×105) were plated in a 10-cm tissue culture dish, and, after serum deprivation for 24 hours, they were cultured for 5 days in either Tet-on or Tet-off medium as previously described (Fuse et al., 2000). At intervals of 24 hours, cells in three dishes were separately harvested by exposure to trypsin and their number and viability were determined with a hemocytometer after staining with Trypan Blue. To examine the effect of cell density on cell growth in the presence of OLIG2, U12-1 cells were seeded at a density of 700 cells/cm2 in 6-cm dishes and 50,000 cells/cm2 in 12-well plates, and cultured for 3 days in either Tet-on or Tet-off medium. At intervals of 24 hours, cells in three dishes were separately harvested and their number and viability were determined.
Assay of BrdU incorporation
U12-1 cells were seeded at a density of 1×105 per well in a six-well plate and cultured for 72 hours in Tet-on or Tet-off medium. They were then incubated for 2 hours with 10 μM 5-bromo-2′-deoxy-uridine (BrdU) (BrdU labeling and detection Kit, Roche Molecular Biochemicals/Boehringer Mannheim), and fixed with cold 70% ethanol plus glycine (50 mM, pH 2.0) for 30 minutes at -20°C for immunocytochemistry. BrdU-incorporating cells were detected by immunostaining with a monoclonal anti-BrdU antibody (1:50) (Roche) and Alexa Fluor 488-conjugated goat anti-mouse IgG (Molecular Probes, Eugene, OR). Nuclei were counterstained with propidium iodide (10 μg/ml), and the immunofluorescence signals were then acquired with a confocal laser-scanning microscope (Olympus, Tokyo, Japan). Cell proliferation was determined by counting BrdU-positive nuclei per total nuclei from three random fields in three independent experiments.
Colony formation assay
For evaluation of anchorage-independent cell growth, Bacto agar solution (5%) at a temperature of 37°C was added to either Tet-on or Tet-off medium prewarmed to 42°C to yield a 0.55% agar solution. After rapid mixing, the solution was added to 6-cm dishes (5 ml per dish) and maintained at room temperature. U12-1 cells were harvested by exposure to trypsin and diluted with 4% noble agar in either Tet-on or Tet-off medium at 37°C to yield a solution containing 0.44% agar. The suspended cells (1×105 in 3 ml per dish) were then transferred to the dishes containing the 0.55% agar medium. After culture of the cells for 3 weeks, colonies with a diameter greater than 125 μm were counted in four 1-cm2 areas in each of three dishes.
cDNA microarray analysis
Total RNAs of U12-1 cells that had been cultured in either Tet-on or Tet-off medium for 48 hours were isolated with the use of an RNeasy column (Qiagen, Valencia, CA). cDNA microarray analysis was performed by the Biomedical Department of Kurabo Industries Ltd. (Osaka, Japan) according to the protocol recommended by Agilent Technologies (http://www.chem.agilent.com/scripts/phome.asp). In brief, total RNA from Tet-on or Tet-off cells was labeled with Cy5 and Cy3, respectively, and allowed to hybridize with cDNAs corresponding to 12,814 human genes. Hybridization data acquired with a fluorescence scanner was analyzed with Avadis software.
Total RNA was extracted from U12-1 cells with the use of an RNeasy Midi kit (Qiagen). Portions (2 μg) of DNase-treated RNA were subjected to reverse transcription (RT) with oligo(dT)12-18 as primer and a Superscript II RNase H- Reverse Transcriptase Kit (Invitrogen). The resulting cDNA was subjected to real-time PCR with a Gene Amp 5700 system and data were analyzed with software version 1.3 (Applied Biosystems, Foster City, CA). The specific primers were 5′-TGCCCTCCCCAGTCTCTCTTA-3′ (sense) and 5′-CCCAAGCACCTCGGATTTTT-3′ (antisense) for p27Kip1 cDNA and 5′-GAAGGTGAAGGTCGGAGTC-3′ (sense) and 5′-GAAGATGGTGATGGGATTTC-3′ (antisense) for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA. The amount of p27Kip1 mRNA was normalized by detecting GAPDH mRNA.
U12-1 cells in 6-cm dishes were lysed for 30 minutes in an ice-cold solution containing 10 mM Tris-HCl (pH 7.4), 5 mM EDTA, 150 mM NaCl, 10% glycerol, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, and 50 mM phenylmethylsulfonyl fluoride. The protein concentration of the lysate was measured with a protein assay kit (Bio-Rad), after which portions (20 μg of protein) were fractionated by SDS-polyacrylamide gel electrophoresis on a 10% gel. The separated proteins were transferred electrophoretically to a polyvinylidene difluoride membrane, which was then incubated for 1 hour at room temperature with 5% dried nonfat milk in Tris-buffered saline containing 0.1% Tween-20. The membrane was then incubated at room temperature first for 2 hours with antibodies to OLIG2 (1:2000 dilution) (Ohnishi et al., 2003), to p27Kip1 (1:2500) (BD Clontech), or to actin (1:500) (Chemicon, Temecula, CA), and then for 1 hour with horseradish-peroxidase-conjugated secondary antibodies. Immune complexes were detected with enhanced chemiluminescence reagents (Amersham Pharmacia Biotech, Freiburg, Germany) and quantified with an LAS1000 image analyzer (Fuji Film, Tokyo, Japan).
Stealth siRNA-mediated knockdown of p27Kip1
To confirm whether OLIG2-mediated suppression of proliferation is dependent on p27Kip1, the 25-nucleotide modified synthetic small interfering RNAs (stealth siRNA) corresponding to nucleotides 843-867 of p27Kip1 mRNA sequence were synthesized by Invitrogen. The siRNA sequence was as follows: 5′-GCAACCGACGAUUCUUCUACUCAAA-3′ (sense); 5′-UUUGAGUAGAAGAAUCGUCGGUUGC-3′ (antisense). Negative control stealth siRNA was also obtained from Invitrogen (catalog number: 12935-300). U12-1 cells were plated in a 3.5-cm tissue culture dish. Eighteen hours after serum deprivation, the cells were transfected with siRNA by Lipofectamine 2000 transfection reagent (Invitrogen). Six hours later, cells were detached with trypsin, counted, and reseeded for growth experiments and monitoring of DNA synthesis (1×105 cells per dish) in Tet-on or Tet-off medium. For growth experiments, cells in three dishes were separately harvested by exposure to trypsin and their number and viability were determined with a hemocytometer after staining with Trypan Blue. p27Kip1 protein expression level was examined by immunoblotting at 2 and 4 days after transfection of siRNA against p27Kip1. DNA synthesis was examined by immunostaining with anti-BrdU antibody at 3 days after transfection of siRNA against p27Kip1 in the presence or absence of OLIG2.
Luciferase reporter assay
Luciferase reporter plasmids based on pGVB2 and containing portions of the promoter of the human p27Kip1 gene (p27PF, p27ApaI, p27AflII, p27No.2, p27No.12, p27No.1, p27MB-435, p27SacII), or derivatives of p27PF containing point mutations (p27mSp1-1, p27mSp1-2, p27mCTF), were introduced into U12-1 cells by transfection (Inoue et al., 1999). The cells were plated in 12-well plates (1×105 cells per well) with Tet-on medium and were transferred to Tet-off medium 48 hours before transfection. The cells were transfected with 1 μg of each luciferase plasmid, as well as with 20 ng of a Renilla luciferase vector (pRL-TK) (Promega) as a control, with the use of the Lipofectamine 2000. Cells were assayed in triplicate for luciferase activity 48 hours after transfection as previously described (Okada et al., 2000). The experimental reporter luciferase activity was calculated by subtracting the intrinsic activity as measured by samples corresponding to the PicaGene Promoter Vector 2 (pGVB2) and then normalized to transfection efficiency as measured by the activity deriving from pRL-TK.
Electrophoretic mobility shift assay (EMSA)
For an EMSA, nuclear extracts were prepared from U12-1 cells as described (Ishida et al., 2002). The extracts were subjected to immunoblot analysis with antibodies to OLIG2, p27Kip1 and lamin A/C (Santa Cruz Biotec). A synthetic double-stranded oligonucleotide (5′-CGCCGCAACCAATGGATCTCC-3′) corresponding to a 21-bp DNA sequence including the CTF site of the human p27Kip1 gene promoter (nucleotides -530 to -510) was labeled with [γ-32P]ATP (Amersham Pharmacia Biotech) with the use of a Gel-Shift Assay Kit (Promega). Nuclear extract (10 μg of protein) was incubated for 30 minutes at room temperature with ∼30,000 cpm of the labeled probe in a final volume of 15 μl containing 10 mM Tris-HCl (pH 7.5), 50 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, 5% glycerol, and 1 μg poly(dI-dC):poly(dI-dC). For competition analysis, unlabeled competitor oligonucleotide was added to the nuclear extract before addition of the labeled probe. For `supershift' analysis, antibodies to OLIG2 were incubated with the nuclear extract for 30 minutes at room temperature before addition of the probe. Reaction products were subjected to electrophoresis (150 V for 150 minutes at 4°C) on a 4% polyacrylamide gel containing 0.25× Tris-borate-EDTA buffer.
Purification of recombinant proteins
Fusion protein, Glutathione-S-transferase (GST)-OLIG2 and control GST proteins were purified from Escherichia coli, BL21 (DE3) transformed with pGEX-OLIG2 cultured in LB medium containing 100 μg/ml ampicillin at 37°C. The E. coli was grown to an optical density (A600) of approximately 0.6, and isopropyl-thio-β-D-galactoside (IPTG) was added to a final concentration of 0.1 mM. The cultures were incubated for an additional 4 hours, and cells were harvested by centrifugation at 1250 g at 4°C. The pellet was suspended in lysis buffer [50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1 mM EDTA and 1 mM PMSF] and lysed with 0.5% Triton X-100 and 10 mg/ml lysozyme. The lysates were centrifuged at 9000 g for 30 minutes to remove unbroken cells and debris, and supernatants were loaded on 50% slurry of the glutathione-Sepharose 4B beads (Amersham Biosciences Corp, Piscataway, NJ). The beads were washed with 60 bed volumes of PBS and then the fusion proteins were eluted with 10 bed volumes of elution buffer [10 mM glutathione in 50 mM Tris-HCl (pH 8.0)]. Eluted samples were dialyzed against PBS at 4°C overnight. The dialysis product was analyzed by SDS-PAGE, followed by Coomassie Blue staining; and its concentrations were determined by comparison to a well-defined BSA control. Approximate 200 ng recombinant GST-OLIG2 was applied to EMSA.
Comparisons between experimental groups were made by Student's t-test. P<0.05 was considered significant.
We thank M. Watanabe for technical assistance as well as S. Kitajima, M. Tamamori-Adachi, J. Hamada and Y. Takahashi for helpful suggestions. T. Suzuki was supported by the Hokkaido University Clark Memorial Foundation. This work was supported in part by grants from the Ministry of Education, Science, Technology, Sports and Culture of Japan, the Ministry of Health, Labor, and Welfare of Japan, the Japan Human Science Foundation.
↵* These authors contributed equally to this work
- Accepted December 21, 2005.
- © The Company of Biologists Limited 2006