Journal of Molecular Neuroscience

, Volume 41, Issue 2, pp 252–262

Trichostatin A Stimulates Steroid 5α-Reductase Gene Expression in Rat C6 Glioma Cells via a Mechanism Involving Sp1 and Sp3 Transcription Factors

Authors

  • Song Her
    • Division of Bio-ImagingChuncheon Center, Korea Basic Science Institute
  • Mi-Sook Lee
    • Division of Bio-ImagingChuncheon Center, Korea Basic Science Institute
    • Department of Nursing, Laboratory of NeuropharmacologyShikoku University School of Health Sciences
Article

DOI: 10.1007/s12031-009-9284-6

Cite this article as:
Her, S., Lee, M. & Morita, K. J Mol Neurosci (2010) 41: 252. doi:10.1007/s12031-009-9284-6

Abstract

The adrenergic and serotonergic stimulations of rat C6 glioma cells have previously been shown to induce the activation of steroid 5α-reductase (5α-R) gene expression, resulting in their differentiation through the production of neuroactive 5α-reduced steroid metabolites. In addition, progesterone and histone deacetylase (HDAC) inhibitors have also been reported to promote the glial cell differentiation with the enhancement of serotonin-stimulated brain-derived neurotrophic factor gene transcription through the production of 5α-reduced neurosteroids, thus suggesting that glial cell differentiation is probably implicated in the protection and survival of neuronal cells in the brain. Therefore, the expression of 5α-R gene in glial cells seems physiologically important in maintaining the neural function in the brain, but little is known about the mechanism underlying the regulation of 5α-R gene transcription. In the present study, the effect of a HDAC inhibitor trichostatin A (TSA) on 5α-R gene transcription in the glioma cells was examined, and TSA was shown to induce the elevation of 5α-R mRNA levels through the activation of the 5α-R promoter via a mechanism involving Sp1 and Sp3 transcription factors in a time- and concentration-dependent manner. Thus, both Sp1 and Sp3 are considered to play a physiological role in the regulation of 5α-R gene expression, and hence the production of 5α-reduced neurosteroids in glial cells.

Keywords

Histone deacetylase inhibitorSteroid 5α-reductase mRNANeuroactive 5α-reduced steroidsSp transcription factorsGlial cell differentiation

Introduction

Steroid 5α-reductase (5α-R), the enzyme catalyzing the conversion of progesterone and other steroid hormones to their neuroactive 5α-reduced metabolites, has been shown to promote the differentiation/maturation of astrocytes (Melcangi et al. 1996). Furthermore, both adrenergic and serotonergic stimulations of rat C6 glioma cells have been shown to induce the elevation of 5α-R mRNA levels (Morita et al. 2004, 2005), thus resulting in the differentiation of the glioma cells through the production of 5α-reduced steroid metabolites (Morita et al. 2006). Moreover, progesterone and its 5α-reduced metabolite have been shown to considerably enhance the stimulatory action of serotonin on brain-derived neurotrophic factor (BDNF) gene expression by promoting the differentiation of these glioma cells (Morita and Her 2008). Therefore, it seems conceivable that differentiation of glial cells may be closely associated with survival and regeneration of neuronal cells in the brain, and hence leading to the improvement of mood disorders and other symptoms in depression.

Recently, histone deacetylase (HDAC) inhibitors, such as trichostatin A (TSA), sodium butyrate, and valproic acid are regarded as useful and effective agents modulating the gene transcription and then promoting the differentiation of various cells. Therefore, it seems possible to speculate that similar to both adrenergic and serotonergic stimulations of glial cells, these inhibitors may be able to induce the differentiation of glial cells through the production of 5α-reduced steroid metabolites, thus improving depressive behavior in the experimental animals and depressive mood disorders in the patients. In fact, these HDAC inhibitors have very recently been shown to induce the differentiation of the glioma cells probably through the production of neuroactive 5α-reduced steroid metabolites, resulting in the elevation of BDNF gene expression in these differentiated cells (Morita et al. 2009). Preclinical study has furthermore suggested that a HDAC inhibitor sodium butyrate may be able to exert its antidepressant-like action in mice in conjunction with a transient increase in BDNF gene expression under the experimental conditions in which the short-lasting hyperacetylation of histone in the brain is observed (Schroeder et al. 2007). However, the antidepressant-like actions of HDAC inhibitors still remain to be unestablished and should be further explored in different experimental systems.

Specificity protein 1 (Sp1), a member of the Sp family, is well known as a transcription factor which can bind to the GC-rich elements of the promoter/regulatory region through three C(2)H(2)-type zinc fingers located in its C-terminal domain. Among the Sp family, which consist of four members, Sp1 and Sp3 are ubiquitously expressed and known as a modulator implicated in the expression of a wide variety of genes. On the other hand, the biological significance of histone acetylation and deacetylation in the regulation of gene transcription has been investigated, and considerable evidence for suggesting a potential role of histone acetylation has recently been accumulated. In particular, the effects of HDAC inhibitors on the promoter activities of various genes have been actively examined, and these inhibitors have been suggested to modulate the gene transcription through the transcription factor Sp1 and Sp3 (Choi et al. 2002; Ghosh et al. 2007; Jaffe et al. 2007; Taniura et al. 2002; Walker et al. 2001; Zeissig et al. 2007). Together with our previous findings (Morita et al. 2004, 2005, 2006, 2009; Morita and Her 2008), these observations enable us to hypothesize that HDAC inhibitors may induce the differentiation of glial cells probably through the production of 5α-reduced steroid metabolites, resulting in the enhancement of BDNF production, which may somewhat contribute to their antidepressant-like actions. Then, to investigate a possible influence of HDAC inhibition on 5α-R gene expression, we first examined the effect of TSA on 5α-R mRNA levels and 5α-R promoter activity in C6 glioma cells, and furthermore demonstrated that TSA can induce the activation of 5α-R gene transcription probably through the induction of Sp1 and Sp3 transcription factors.

Materials and Methods

Materials

DNeasy Blood and Tissue Kit and SuperFect™ Transfection Reagent were obtained from QIAGEN (Valencia, CA, USA). SYBR™ Green polymerase chain reaction (PCR) Master Mix and Dual 384-well GeneAmp™ PCR System 9700 were purchased from Applied Biosystems (Foster City, CA, USA). Moloney-murine leukemia virus (M-MLV) reverse transcriptase and pGL3 basic vector were from Promega (Madison, WI, USA). Terminal transferase (TdT) and NEB buffer 4 were from New England BioLabs (Ipswich, MA, USA). QuickChange mutagenesis method was obtained from Stratagene (La Jolla, CA, USA). Mutagenic primers were from OPERON (Alameda, CA, USA). SuperScript II and random hexamer were from Invitrogen (Carlsbad, CA, USA). TRI Reagent was from Sigma Chemical Co. (St. Louis, MO, USA). PROBER™ Probe DNA purifying system was from Intron Biotechnology (Seongnam, Korea). Antibodies against Sp1 (PEP2), Sp2 (K-20), Sp3 (D-20), and Sp4 (V-20) were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rat C6 glioma cell line (CCL-107) was obtained from the American Type Culture Collection (Rockville, MD, USA). Other chemicals were commercially available reagent grade or ultrapure grade.

Nucleic Acid Preparation

Genomic DNA of rat brain was prepared using the DNeasy Blood and Tissue Kit following the manufacturer’s instructions. Total RNA of C6 glioma cells was isolated using TRI Reagent following the manufacturer’s instruction, and the integrity of RNA was then verified by a gel electrophoresis (Karssen et al. 2006).

Determination of 5a-R mRNA Levels

Real-time quantitative reverse transcription-PCR was applied to the determination of 5α-R mRNA levels in C6 glioma cells. The first strand cDNA was synthesized from total RNA (500 ng) prepared from the glioma cells using SuperScript II and random hexamer as a primer following the manufacturer’s instruction, and the reaction mixture was diluted tenfold with sterilized ultrapure water and then stored at −20°C until use. The cDNA was amplified by a real-time PCR in 10 μl of the mixture containing 2 μl of the diluted reverse transcription mixture (equivalent to 5 ng of the starting RNA), 10 pmol of the specific primers for 5α-R gene (Table 1), and 5 μl of 2× SYBR™ Green PCR Master Mix using Dual 384-well GeneAmp™ PCR System 9700 for 40 cycle at 94°C for 15 s followed by fluorescence capture at 60°C for 1 min. The critical threshold, or the point at which fluorescence signals exceeded the background, for each sample and each target gene was determined and then compared with the standard, which was made by amplify the tenfold serial dilution of standard cDNA in parallel. The expression of 5α-R gene was normalized to the mean transcription levels of glyceraldehyde 3-phosphate dehydrogenase and β-actin.
Table 1

Primer sequences used in this study

Name

Sequence

Purpose

5α-R/F(-1630)

CCAGTTCTGATAAGACATCCCATAATCGCC

Genomic PCR for cloning promoter

5α-R/F(-1377)

TTCGTGCTGCTCAGATGAACCGATC

Genomic PCR for cloning promoter

5α-R/R(+64)

TGAAACCTTCCAAGTAGACCAGCATGTCG

Genomic PCR for cloning promoter

5α-R/R(+226)

CGGCTGCAGGACGAATGTACTCGTAC

Genomic PCR for cloning promoter

Sp1 AB/F

TCGGGAGGGGGCGGGGCTCTCTGTGTGGGCGGGGCTCAACTA

Gel mobility shift assay

Sp1 AB/R

TAGTTGAGCCCCGCCCACACAGAGAGCCCCGCCCCCTCCCGA

Gel mobility shift assay

Sp1A/F

ATCGGGAGGGGGCGGGGCTCTC

Gel mobility shift assay

Sp1A/R

GAGAGCCCCGCCCCCTCCCGAT

Gel mobility shift assay

Sp1mutA/F

TCTAGATCGGGAGGGTTCGGGGCTCTCTGTGT

Gel mobility shift assay

Sp1mutA/R

ACACAGAGAGCCCCGAACCCTCCCGATCTAGA

Gel mobility shift assay

Sp1B/F

TCTCTGTGTGGGCGGGGCTCAA

Gel mobility shift assay

Sp1B/R

TTGAGCCCCGCCCACACAGAGA

Gel Mobility Shift Assay

5α-R/-198F 1st Sp1 mut

AGATCGGGAGGGTTCGGGGCTCTCT

Point mutation

5α-R/-177F 2nd Sp1 mut

TCTCTGTGTGGGCGGGGCTCAA

Point mutation

5α-R 421F

TGAGCCAGTTTGCGGTTTATG

QPCR

5α-R 1521R

GGGCAAAGCCTGTCAGGAA

QPCR

5α-R 350F

CACCCTCCTGGTCACCTTTGT

QPCR

5α-R 450R

TCTGCTCTGTACATAGCCGTTGA

QPCR

RNA Ligase-mediated Rapid Amplification of cDNA Ends (RACE)

The oligonucleotide complementary to the sequence of 226 to 252 nucleotides from the initiation ATG of 5α-R gene, designated rSrd5a1/+226R (Table 1), was used as a specific primer for the reverse transcription. First strand cDNA was synthesized from 2 μg of total RNA of C6 glioma cells in 25 μl of the reaction mixture containing 4 pmol of the primer (rSrd5a1/+226R) and 200 units of M-MLV reverse transcriptase. Reaction product was purified using PROBER™ Probe DNA purifying system, and poly A-tails were attached to purified cDNA using TdT in 50 μl of 1× NEB buffer 4 containing 0.25 mM of CoCl2, 0.2 mM of dATP, and 5 units of TdT at 37°C for 15 min, and then terminated by heating the mixture at 87°C for 10 min. Rapid amplification of cDNA ends (RACE) products were generated by PCR using 17T-adaptor and rSrd5a1/+64R as primers, and separated on 1% agarose gel electrophoresis. The major band was eluted out, digested with Xho1 and Pst1, and then subcloned into pBluescript SK. Transcription start site was determined by sequencing 16 individual clones.

Serial Deletion and Point Mutation of 5α-R Promoter-luciferase Reporter Gene Constructs

The PCR amplification of 5α-R gene promoter/regulatory regions was carried out using 100 ng of rat genomic DNA as a template and specific primers summarized in Table 1, and obtained 1.6 kb PCR product ranging from 5′ upstream region (1.5 kb) to exon 1 (77 bp) was subcloned into the pGL3 basic vector, which was designated pGL3RA1532. To construct serial deletion mutants of the promoter-reporter construct, the wild-type plasmid was digested with various restriction enzymes, and the DNA fragments containing the promoter regions ranging from +77 to −952 (Mlu I), −581 (Hinf III), −250 (Sac I), and −50 bp (Mlu I, Nhe I) were then subcloned into the polylinker of pGL3-basic vector, designated pGL3RA952, pGL3RA581, pGL3RA250, and pGL3RA50, respectively. To make the construct pGL3RA150, the DNA fragment ranging from −77 to −150 bp was synthesized by PCR and subcloned to the basic vector.

To introduce the point mutations to the Sp1 binding elements of 5α-R promoter-luciferase construct pGL3RA250, the site-directed mutation was carried out by QuickChange mutagenesis method (Stratagene, La Jolla, CA, USA) using the mutagenic primers. All constructs were verified by the restriction maps on an agarose gels, and the point mutations of the constructs were confirmed by sequencing around the mutation sites at the DNA sequencing facility of Macrogen Inc. (Seoul, Korea).

Cell Culture and Transient Transfection Assay

Rat C6 glioma cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% heat-inactivated bovine calf serum, 2 mM L-glutamine, 50 units/ml of penicillin, 50 μg/ml of streptomycin, and 50 μg/ml of gentamicin sulfate at 37°C in a humidified incubator containing 95% air and 5% CO2 atmosphere. The transient transfection assay was carried out as described previously (Her et al. 1999). Briefly, the cells on a 24-well cluster plate (2 × 105 cells/well) were transiently transfected with the 5α-R promoter-luciferase reporter gene constructs in combination with the β-galactosidase control construct (1 to 3 μg of double-strand DNA in 100 μl) using SuperFect™ Transfection Reagent or polyethylenimine solution (Boussif et al. 1995). The cells were harvested at 36 h after the transfection, and the activities of luciferase and β-galactosidase in an aliquot of the cell extract were then determined.

Nuclear Extract Preparation and Gel Mobility Shift Assay

Nuclear extracts were prepared from rat C6 glioma cells according to the procedure reported previously (Andrews and Faller 1991). Briefly, the glioma cells were grown to form a subconfluent monolayer on a 100-mm culture dish (approximately 1 × 107 cells) and collected into 1.5 ml of ice-cold phosphate-buffered saline (PBS). The cells were washed once with ice-cold PBS and then lysed by incubating in 400 μl of 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)-KOH (pH 7.9) containing 10 mM KCl, 1.5 mM MgCl2, 0.5 mM dithiothreitol, and 0.2 mM phenylmethylsulfonyl fluoride (PMSF) for 10 min on ice. The nuclei were collected by centrifuging the lysates at 17,000 × g for 10 s at 4°C and then resuspended in 20–100 μl of the hypotonic lysis buffer consisting of 20 mM HEPES-KOH (pH 7.9), 25% glycerol, 420 mM NaCl, 1.5 mM MgCl2, 0.2 mM ethylenediaminetetraacetic acid (EDTA), 0.5 mM dithiothreitol, and 0.2 mM PMSF. The suspension was incubated for 20 min on ice, followed by centrifugation at 17,000 × g for 2 min, and the extracts were then separated from the pelleted debris and stored at −80°C until use.

Gel mobility shift assay was carried out as described previously (Her et al. 1999) using the 32P-labeled probes containing the sequences of Sp1 binding elements. Briefly, the nuclear extracts (3 μg of total protein) were incubated with radio-labeled probes in 20 μl of the binding buffer consisting of 25 mM HEPES buffer (pH 7.9), 50 mM KCl, 0.05 mM EDTA, 10% glycerol, 0.5 nM dithiothreitol, 0.5 mM PMSF, and 0.5 μg of poly(dI-dC)-poly(dI-dC) for 30 min on ice. To confirm Sp1 as an element of the DNA-protein complex, anti-Sp transcription factors antibodies were included in the binding mixture. For the gel mobility shift competition assay, the different amounts of non-labeled competitor oligonucleotides were added to the binding mixture. After incubating the binding mixture, the DNA-protein complexes were separated from the free probe on 5% polyacrylamide gel, and the complexes were visualized by autoradiography.

Western Blot Analysis

Cells on a 6-well plate were exposed to TSA for different time periods, washed twice with PBS, and then lysed with radioimmunoprecipitation assay buffer (50 mM Tris–HCl, pH 7.5, 150 mM NaCl, 1% Triton X-100, 0.1% sodium dodecyl sulfate (SDS), and 1% sodium deoxycholate) containing 1 µM EDTA, 1 µM sodium orthovanadate, 2 µM ethyleneglycoltetraacetic acid, 1 µM PMSF, 10 µg/ml leupeptin, and 10 µg/ml aprotinin. The lysates were collected and centrifuged at 15,000 × g for 30 min at 4°C, and the protein concentrations of obtained cell extracts were then determined by a Bradford assay using bovine serum albumin as a standard. Aliquots of the cell extracts were boiled in SDS sample buffer, and proteins were separated by SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membrane. The membrane was washed with Tris-buffered saline and Tween 20 and blocked with 5% non-fat milk, and then incubated with a rabbit anti-Sp1, a rabbit anti-Sp3, and a mouse anti β-actin antibodies. The immunocomplexes of target proteins were visualized using a horseradish peroxidase-conjugated secondary antibodies (1:5,000) and an ECL Western blotting assay kit (Amersham Co., England).

Statistical Analysis

The data were expressed as the mean ± SE, and the statistical significance between two groups was assessed using Student’s t test. A value of P < 0.05 was accepted as a statistically significant difference.

Results

The effect of an HDAC inhibitor TSA on 5α-R mRNA levels in rat C6 glioma cells was first examined to elucidate a possible association of HDAC activity with 5α-R gene expression in rat C6 glioma cells. As shown in Fig. 1, 5α-R mRNA levels was significantly elevated by exposing the glioma cells to TSA (25 ng/ml) for 1 h and reached the maximum levels (approximately 3.5-fold higher than the basal levels) at 5 h, then declining toward the basal levels after that time. Thus, TSA was considered to increase 5α-R mRNA levels in the glioma cells probably through the transcriptional activation of 5α-R gene expression.
https://static-content.springer.com/image/art%3A10.1007%2Fs12031-009-9284-6/MediaObjects/12031_2009_9284_Fig1_HTML.gif
Figure 1

Stimulatory action of trichostatin A (TSA) on 5α-R gene transcription in rat C6 glioma cells. Cells were exposed to 25 ng/ml of TSA for different time periods, and 5α-R mRNA levels were determined as described in the text. Results were expressed as the ratio to the basal levels. Values are the mean ± SE (*P < 0.05, n = 9)

To investigate the molecular mechanism underlying TSA-induced 5α-R gene expression, a 5′-flanking region of the 5α-R gene was cloned and sequenced. The obtained clone was shown to have three bases at −316, −571, and −650 bp, which were different from the sequence registered with the GenBank/EMBL database (Fig. 2a). In this study, the original clone was used without any back-mutation. To characterize the cloned 5′-flanking region, the transcription start site was first determined by RNA ligase-mediated-RACE using two sets of oligonucleotide primers, one was CDS primer 1 and 17T-adaptor primer and the other was CDS primers 2 and Nested primer (Table 1). The combination of CDS primer 1 and 17T-adaptor generated a smeared product, whereas the PCR using CDS primers 2 and Nested primer gave a 127-bp product as expected (data not shown). Then, this PCR product was cloned for the analysis of DNA sequence, and the attachment of RNA to guanine residue at the nucleotide position of 107 bp upstream from the ATG initiation codon was detected in 13 of the obtained 16 clones, and hence the transcription of 5α-R gene was considered to be initiated at this position. Furthermore, there were CAAT boxes and five GC-rich sequences including four Sp1-binding elements located at −186, −167, −133, and −115 bp in the immediate upstream region from the transcription start site (Fig. 2a).
https://static-content.springer.com/image/art%3A10.1007%2Fs12031-009-9284-6/MediaObjects/12031_2009_9284_Fig2_HTML.gif
Figure 2

Effects of different deletions of 5′-flanking region on 5α-R promoter activity. Deletion mutants of the 5α-R promoter-luciferase reporter gene construct were transfected into rat C6 glioma cells, and luciferase activity was then determined as described in the text. a Nucleotide sequence of rat 5′-flanking region of 5α-R gene. Potential transcription factor binding sites are boxed in the sequence. b Luciferase activity in the glioma cells transfected with the mutant 5α-R promoter-luciferase reporter gene constructs. Relative luciferase activity was obtained as the ratio to the activity in the cells transfected with the construct plasmid pGL3-basic. Values are the mean ± SE (*P < 0.05, n = 9)

To determine the sequence required for the promoter activity of 5′-flanking region of 5α-R gene, the cloned 1,532-bp fragment upstream of the transcription start site was subjected to the serial deletion, and the obtained fragments were then cloned into the firefly luciferase reporter vector pGL3 for a transient transfection assay of the promoter activity. As shown in Fig. 2b, the construct pGL3RA250 containing the 250-bp fragment could almost fully express the promoter activity, and the construct pGL3RA150 containing the 150-bp fragment almost completely lost the promoter activity in rat C6 glioma cells. As predicted by a computer program for searching a putative site of transcription factor binding, the Transcription Element Search Software, there were a putative Myc-associated zinc finger protein binding element (−192 bp) and two Sp1 binding sites (−186 and −167 bp) identified on the promoter sequence between −250 and −150 bp. Interestingly, pGL3RA50 containing the 50-bp fragment expressed the slightly but significantly higher promoter activity (approximately threefold) as compared with the activity of pGL3RA150. Thus, the binding of Sp1 transcription factor to its consensus elements was shown to be associated with the expression of 5α-R promoter activity.

To further investigate a possible involvement of these Sp1 binding sites in the activation of 5α-R promoter, the site-directed point mutation of these binding elements in pGL3RA250 was carried out to produce the mutant promoter-luciferase reporter gene constructs, in which each Sp1 binding element located at −186 (Sp1A) and −167 bp (Sp1B), or both of them (Sp1AB), were mutated as illustrated in Fig. 3a, and designated pGL3RA250mutA, pGL3RA250mutB, and pGL3RA250mutAB, respectively. Then, these mutant constructs were transfected into rat C6 glioma cells, and the basal promoter activities were determined to confirm the significance of these Sp1 binding sites in the expression of 5α-R promoter activity. As shown in Fig. 3b, the basal activity of 5α-R promoter was dramatically reduced by point mutation of either the Sp1 binding element at −186 or −167 bp, and the promoter activity of the construct with the mutation of both Sp1 binding elements (pGL3RA250mutAB) was almost identical to that of the deleted construct pGL3RA150. Furthermore, the effect of TSA treatment on the promoter activities of the wild-type and mutant constructs was also examined, and the stimulatory action of TSA on the 5α-R promoter activity was partially reduced by mutating each one of these Sp1 binding elements and almost completely abolished by the mutation of both binding elements. These results showed that there were two functional Sp1 binding elements located at −186 and −167 bp on the 5α-R promoter and suggested that each one of these binding elements might be definitely required for expressing the basal activity of 5α-R promoter activity, and both of them might be essential for the stimulatory action of TSA on the promoter activity.
https://static-content.springer.com/image/art%3A10.1007%2Fs12031-009-9284-6/MediaObjects/12031_2009_9284_Fig3_HTML.gif
Figure 3

Effects of site-directed mutation of Sp1 binding elements on basal and trichostatin A (TSA)-stimulated 5α-R promoter activities. Rat C6 glioma cells were transfected with the Sp1-mutants of 5α-R promoter-luciferase reporter gene construct, and luciferase activity was determined as described in the text. a Schematic illustration of the Sp1-mutant 5α-R promoter-luciferase reporter gene constructs. b Basal and TSA-stimulated luciferase activity in the glioma cells transfected with the Sp1-mutant constructs. Relative luciferase activity was obtained as the ratio to the activity in the cells transfected with the wild-type construct pGL3RA250. Values are the mean ± SE (*P < 0.05, n = 9)

To examine a possible interaction of nuclear proteins with the Sp1 binding sites on the 5α-R promoter, the gel mobility shift assay was carried out using the nuclear extracts prepared from rat C6 glioma cells and the oligonucleotide containing two Sp1 consensus elements located on the promoter sequence at −186 and −167 bp (Sp1AB) or that containing one of each consensus element (Sp1A and Sp1B) as illustrated in Fig. 4a. The DNA-protein complex was formed by incubating the nuclear extracts with the radio-labeled Sp1AB probe, and the complex formation was significantly inhibited by the addition of non-labeled competitor oligonucleotides containing either Sp1AB, Sp1A, or Sp1B consensus sequences in a manner dependent on the concentrations of these oligonucleotides (Fig. 4b). Furthermore, to estimate the binding affinities of these Sp1 consensus elements to the nuclear proteins, the DNA-protein complexes were quantified using a scanning densitometry, and the 50% inhibitory concentration (IC50) of each competitor was then determined by the regression analysis of the correlation between the signal intensities and the competitor concentrations (Fig. 4c). Consequently, the IC50 values of these competitors were computed at 93.3, 67.6, and 186.2 ng/tube for the Sp1AB, Sp1A, and Sp1B consensus sequence, respectively. Therefore, the binding affinity of the Sp1A consensus element was estimated at approximately 2.8-fold higher than that of the Sp1B element.
https://static-content.springer.com/image/art%3A10.1007%2Fs12031-009-9284-6/MediaObjects/12031_2009_9284_Fig4_HTML.gif
Figure 4

Affinities of Sp1 binding elements to nuclear proteins prepared from rat C6 glioma cells. Gel mobility shift assay was carried out as described in the text using the radio-labeled wild-type oligonucleotide probe (Sp1AB) and the unlabeled competitor oligomucleotides (Sp1AB, Sp1A, Sp1B, and Sp1mutA) listed in Table 1. The DNA-protein complex formation was visualized and analyzed by a computer-assisted densitometry using the NIH Image 1.52 software. a Sequence of oligonucleotides used in the gel mobility shift assay. b Autoradiographic visualization of the DNA-protein complex. c Regression analysis of the correlation between the signal intensities and the competitor concentrations

The Sp family of zinc finger transcription factors is known to consist of four members, Sp1, Sp2, Sp3, and Sp4, and they have a highly conserved DNA-binding domain, which can bind to the consensus sequence of GGGGCGGGGC and the closely related sequences named the GC boxes. Then, to identify the Sp protein binding to the Sp1 consensus elements at −186 and −167 bp on the 5α-R promoter, the radio-labeled oligonucleotide probes containing the Sp1AB, Sp1A, or Sp1B consensus elements were incubated with the nuclear extracts of rat C6 glioma cells in the presence and absence of the specific antibodies against each Sp proteins, and the formation of DNA-protein complex was analyzed by a gel mobility shift assay. As shown in Fig. 5, several DNA-protein complexes were similarly formed from these probes and the nuclear extracts, and the major DNA-protein complex was supershifted by including anti-Sp1 and anti-Sp3 antibodies, but not anti-Sp2 and anti-Sp4 antibodies, in the reaction mixture. These results were considered to provide evidence for proposing both Sp1 and Sp3 as transcription factors expressed in rat C6 glioma cells and being responsible for the high basal activity of the 5α-R promoter observed in these cells.
https://static-content.springer.com/image/art%3A10.1007%2Fs12031-009-9284-6/MediaObjects/12031_2009_9284_Fig5_HTML.gif
Figure 5

Effects of antibodies against Sp family proteins on the DNA-protein complex formation. Gel mobility shift assay was carried out using the nuclear extracts prepared from rat C6 glioma cells and the radio-labeled oligonucleotide probes containing the Sp1A element, Sp1B element, and both of them in the presence of specific antibodies against Sp1, Sp2, Sp3, and Sp4 proteins as described in the text

To address the question whether the stimulatory action of TSA on the 5α-R promoter activity was observed within the cells, the glioma cells were transfected with the construct pGL3RA250 and exposed to various concentrations of TSA for different time periods. As shown in Fig. 6, the 5α-R promoter activity was elevated by low concentrations of TSA (up to 12.5 ng/ml) and then suppressed by increasing the drug concentration up to 200 ng/ml (Fig. 6a). Also, the stimulatory action of TSA on the promoter activity was observed in a manner dependent on the time of drug treatment (Fig. 6b). Furthermore, to elucidate the molecular mechanism of TSA-stimulated 5α-R gene expression, the transcription factor Sp1 and Sp3 levels in rat C6 glioma cells was examined by a gel mobility shift assay using the Sp1A and Sp1B oligonucleotide probes. As shown in Fig. 7a, the exposure of the glioma cells to TSA enhanced the nuclear protein binding to both Sp1A and Sp1B consensus elements, and the enhancement of the DNA-protein complex formation reached the maximum levels by exposing these cells to the drug for 1 h, then declining toward the basal levels by exposing to the drug for 3 h or longer periods. Moreover, the DNA-protein complexes consisting of the Sp1 consensus element and the nuclear extracts prepared from both non-treated and TSA-treated cells were supershifted by adding anti-Sp1 and anti-Sp3 antibodies, but not anti-Sp2 and anti-Sp4 antibodies, to the binding reaction mixture (Fig. 7b). In addition, TSA treatment significantly enhanced the expression of Sp1 and Sp3 proteins in a time-dependent manner (Fig. 7c). Consequently, these results were considered to provide evidence that TSA treatment might induce the expression of Sp1 and Sp3 transcription factors and/or enhance their binding affinities to the Sp1 consensus element on the 5α-R promoter, resulting in the activation of 5α-R gene transcription in the glioma cells.
https://static-content.springer.com/image/art%3A10.1007%2Fs12031-009-9284-6/MediaObjects/12031_2009_9284_Fig6_HTML.gif
Figure 6

Effect of trichostatin A (TSA) on 5α-R promoter-driven luciferase expression in rat C6 glioma cells. Cells were transfected with the 5α-R promoter-luciferase reporter gene construct pGL3RA250. a Transfected cells were exposed to various concentrations of TSA for 9 h. b Transfected cells were exposed to 6.3 ng/ml of TSA for different time periods. Luciferase activities in the cell extracts were then determined as described in the text. Results were expressed as the ratio to the basal levels. Values are the mean ± SE (*P < 0.05, n = 9)

https://static-content.springer.com/image/art%3A10.1007%2Fs12031-009-9284-6/MediaObjects/12031_2009_9284_Fig7_HTML.gif
Figure 7

Effect of trichostatin A (TSA) on the expression of Sp family transcription factors in rat C6 glioma cells. a Nuclear extracts were prepared from the glioma cells treated with 25 ng/ml of TSA for different time periods, and the gel mobility shift assay was carried out as described in the text using the radio-labeled oligonucleotide probes containing the Sp1A or Sp1B element. b Gel mobility shift assay with both Sp1A element and Sp1B element in the presence and absence of specific antibodies against Sp1, Sp2, Sp3, and Sp4. c Western blot analysis. Cell extracts were prepared from the glioma cells treated with 25 ng/ml of TSA for different time periods, and the Western blot analysis was carried out using antibodies against Sp1 or Sp3. The expression of β-actin protein was also determined as an internal standard, and the ratio of Sp1 and Sp3 protein levels to the internal standard protein was then calculated to normalize the expression of these target proteins. Data were obtained from three independent experiments. Values are the mean ± SE (*P < 0.05)

Discussion

The present study indicated for the first time that TSA elevated 5α-R mRNA levels in rat C6 glioma cells probably through the stimulation of 5α-R gene transcription (Fig. 1), and hence the cloning and characterization of the promoter/regulatory region of rat 5α-R gene were considered to be essential for understanding the molecular mechanism of TSA action on the transcription of 5α-R gene in the glioma cells. Then, the 1,532-bp fragment of the promoter was obtained, and the promoter-luciferase reporter gene constructs containing the different length of deleted promoter fragments were generated to determine the promoter sequence required for the expression of its basal activity. The transient transfection studies clearly indicated that the fragment containing a minimum of 250 bp upstream of the transcription start site was sufficient to express the basal promoter activity and suggested that two Sp1 consensus-binding elements identified at the positions of −186 and −167 bp might be required for maintaining the basal activity of the 5α-R promoter (Fig. 2). However, it seems still necessary to verify whether these Sp1 consensus elements are practically functional. Then, the site-directed point mutations of Sp1 consensus elements located at −186 (Sp1A) and −167 bp (Sp1B) were carried out, and the basal activities of these mutant promoter constructs were then determined. Consequently, the basal promoter activity was shown to be markedly reduced by mutating each one of the Sp1 consensus-binding elements. Moreover, the stimulatory action of TSA on the promoter activity was shown to be completely blocked by mutating both Sp1 binding elements (Fig. 3). Therefore, it seems possible to consider that both Sp1 consensus elements located at −186 and −167 bp may be practically functional and essential for the stimulatory action of TSA on the 5α-R promoter as well as for the expression of its basal activity.

The properties of these functional Sp1 consensus elements were furthermore studied, and the binding affinity of the Sp1 consensus element at −186 bp to the nuclear proteins was shown to be approximately 2.8-fold higher than that of the element at −167 bp (Fig. 4). On the other hand, the binding of nuclear proteins to these elements were also characterized, and two proteins of the Sp transcription factors, Sp1 and Sp3, were clearly shown to be capable of interacting with the consensus elements (Fig. 5). Moreover, TSA treatment was shown to increase the amounts of Sp1 and Sp3 proteins in the nucleus (Fig. 7) prior to the elevation of 5α-R mRNA levels in the glioma cells (Fig. 6). Thus, Sp1 and Sp3 are considered to play an essential role as an intrinsic factor regulating the basal activity of the 5α-R promoter preferentially through interacting with the distal Sp1 binding element and also suggested to be involved in the stimulatory action of TSA on the transcription of 5α-R gene in the glioma cells.

Since histone acetylation is well established as one of the intrinsic events regulating the gene transcription, HDAC is therefore suggested to play an important role in the regulation of various gens expression in a variety of the cells and tissues. Recent studies using various HDAC inhibitors, such as sodium butyrate and TSA, have shown that these inhibitors can activate the human insulin-like growth factor binding protein-3 promoter in breast cancer cells (Walker et al. 2001) and hepatoma cells (Choi et al. 2002). Furthermore, sodium butyrate has been shown to induce the activation of epithelial sodium channel gene expression in human colorectal tumor cells (Zeissig et al. 2007). Moreover, TSA has also been shown to activate cyclooxygenese-1 gene transcription in human astrocytes (Taniura et al. 2002) and baboon glycodelin gene transcription in COS cells (Jaffe et al. 2007). In contrast, TSA has been reported to induce the inhibition of TGF-b-induced collagen expression in skin fibroblasts (Ghosh et al. 2007). These observations have consistently suggested that HDAC inhibitors can specifically modify the transcriptional regulation of various genes through the expression of Sp1 and/or Sp3, thus proposing a possible implication of histone acetylation in the regulation of gene expression.

The present study indicated for the first time that an HDAC inhibitor TSA can activate the 5α-R promoter through a mechanism involving Sp family transcription factors Sp1 and Sp3, resulting in the stimulation of 5α-R gene transcription in rat C6 glioma cells. Therefore, it seems possible that the inhibition of HDAC activity, and hence the enhancement of histone acetylation may be closely associated with the production of neuroactive 5α-reduced steroid metabolites in the cells. On the hypothesis supported by our previous studies (Morita et al. 2004, 2005, 2006, 2009; Morita and Her 2008), it seems conceivable that TSA may be able to induce the differentiation of glial cells probably through the enhancement of 5α-reduced neurosteroid production, thus resulting in the augmentation of serotonin-stimulated BDNF gene expression in differentiated glial cells, and hence a chain of these events that occurred in the brain is considered as a hypothetical mechanism underlying the antidepressant-like actions of HDAC inhibitors. However, there is no practical evidence for answering the critical question of whether TSA can induce the differentiation of glial cells, resulting in the enhancement of serotonin-stimulated BDNF production, which may be related to the survival, regeneration, and plasticity of neuronal cells in brain. Further studies on the effects of HDAC inhibitors on the differentiation of glial cells and the production of BDNF in these differentiated cells are now in progress.

Acknowledgment

This work was supported by a grant of the Korean Basic Science Institute (T29740).

Copyright information

© Humana Press 2009