Published online before print September 28, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
in human melanoma cells
* Departments of Molecular Health Sciences, Graduate School of Pharmaceutical Sciences, and
Molecular and Cellular Biology, Graduate School of Medical Sciences, Nagoya City University, Nagoya, Japan
1 Correspondence: Department of Molecular Health Sciences, Graduate School of Pharmaceutical Sciences, Nagoya City University, 3-1 Tanabe-dori, Mizuho-ku, Nagoya, Aichi 467-8603, Japan. E-mail: konozaki{at}phar.nagoya-cu.ac.jp
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In this study, we identified a sequence, CGCC, located at –48 to –45 upstream of the transcription start site, to be essential for the constitutive IL-1 gene activation. Specificity protein 1 (Sp1) and Sp3 bound to the nucleotide containing the sequence. Although the binding level to the nucleotide and expression level of Sp1 and Sp3 are comparable in A375-R8 and A375-6 cells, transactivation activity of Sp1 is higher in A375-R8 cells as compared with A375-6 cells. Sp3 could not transactivate the IL-1 promoter. These results suggest that Sp1 but not Sp3 is important for IL-1 gene activation. Trichostatin A (TSA), an inhibitor of histone deacetylase (HDAC), greatly augmented the IL-1 promoter activity in A375-6 cells to the level comparable with that in A375-R8 cells. TSA also induced IL-1 mRNA expression in A375-6 cells. Sp1 and Sp3 bound to HDAC1 in A375-R8 and A375-6 cells. The chromatin immunoprecipitation assay revealed the binding of Sp1 and HDAC1 to the promoter region of the IL-1 gene. The activities of HDAC bound to Sp1 and Sp3, and that of HDAC1 was lower in A375-R8 cells as compared with A375-6 cells. These results indicate that the reduction in the activity and interaction of HDAC1 with Sp1 are critical for the constitutive IL-1 gene expression.
Key Words: HDAC • trichostatin A • chromatin immunoprecipitation
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and β, and their genes share a similar intron/exon structure. Although they share the same biological activities by binding to the same receptor, type I receptor (IL-1RI), the precursor of IL-1β binds weakly to IL-1RI and cannot exert biological activity, and that of IL-1 can bind to IL-1RI and exerts biological activity. Therefore, IL-1β works only after proteolytic maturation followed by the release from its producing cells, and IL-1 can exert biological activity in precursor and mature forms. IL-1 is also active in a membrane-bound form.
The gene activation mechanism for human IL-1β has been studied extensively. Studies revealed that the C/EBPβ/NF-IL-6 site, NF-β1 site, cAMP response element-like site, AP-1 site, and NF-
B site are important for activation of the IL-1β gene [2
, 3
]. In contrast, the regulatory mechanism of the IL-1 gene expression is largely unknown [4
]. In macrophages and monocytes, IL-1 and -β are often induced by the same stimuli [1
]. However, in other cell types, including keratinocytes [5
], endothelial cells [6
], T cells [7
], melanoma cells [8
], and ovarian tumor cells [9
], gene expression of these two IL-1s appears to be regulated rather independently. In addition, sustained production of IL-1 is observed and is implicated in many chronic diseases, including juvenile rheumatoid arthritis [10
], inflammatory myopathies [11
], scleroderma [12
], cystic fibrosis [13
], HIV infection [14
], gastric carcinoma [15
], and ovarian cancer [16
]. Constitutively produced IL-1 is an autocrine growth factor for Kaposi sarcoma [17
] and induces senescence of endothelial cells [6
] and fibroblasts [18
].
In various experimental models, IL-1 increases tumor invasiveness and metastasis [1
, 19
] Melanoma cells in culture produce a variety of growth factors and cytokines, including IL-1, IL-6, and TGF- and -β [20
]. IL-1 is implicated in the pathogenesis of melanoma through effects on its own producing cells, adjacent cells, and infiltrating lymphoid cells. IL-1 also exhibits systemic effects, including fever, diarrhea, anorexia, somnolence, and hypercalcemia [1
]. The acquired resistance against antiproliferative cytokines IL-1, IL-6, and oncostatin M is implicated in malignancy of human melanoma [20
, 21
]. In addition, some melanoma cell lines established from a human melanoma legion produce IL-1 constitutively [11
].
We have reported that IL-1 is antiproliferative for a human melanoma cell line (A375-6) [22
, 23
]. During the study, we noticed that the IL-1-sensitive cells became resistant to IL-1, although they expressed functional IL-1RI [24
]. The IL-1-resistant clone, A375-R8, constitutively produced IL-1 but not IL-1β in protein and mRNA levels. As the resistant cells exhibited many features of advanced melanoma, including augmented expression of adhesion molecules and production of cytokines and matrix metalloproteinase [25
], it was suggested that the IL-1 resistance is associated with the malignant phenotype of melanoma. We reported previously that IL-1 stimulates its own gene expression and production through activation of NF-B in an autocrine manner [26
]. We also noticed that there is a region in the IL-1 promoter responsible for the constitutive gene activation. In this study, we analyzed the molecular mechanism of the constitutive activation of the IL-1 gene using IL-1-nonproducing A375-6 cells and IL-1-producing A375-R8 cells.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Plasmids
IL-1 genomic DNA fragments were cloned into pGL3-basic (Promega, Madison, WI, USA) to generate pGL3-IL1a(–421), pGL3-IL1a(–103), pGL3-IL1a(–70), pGL3-IL1a(–61), pGL3-IL1a(–53), pGL3-IL1a(–51), pGL3-IL1a(–31), and pGL3-(IL1a-GC)3. pGL3-IL1a(–421)mATAA, replacing CGCC to ATAA at –48 to –45, and pGL3-IL1a(–421)
CC, deleting CC at positions –46 to –45, were generated by PCR. All constructs were verified by sequencing. pCIneo-Sp1, pAC-NSp1, and pAC-CSp1 were generous gifts from Dr. Soichi Kojima (Riken, Japan) [27
] and pCMV4-Sp3 from Dr. Jonathan M. Horowitz (North Carolina State University, Raleigh, NC, USA) [28
]. Galactosidase 4 (GAL4)-Sp3 was generated by PCR. pFR-luciferase (Luc; 5xGAL4-Luc) was purchased from Stratagene (La Jolla, CA, USA).
Cell cultures
The human melanoma cell line A375 was given originally by Dr. Raymond Ruddon (National Cancer Institute, Bethesda, MD, USA). By limiting dilution, a subclone, IL-1-sensitive A375-6, was obtained [23
]. An IL-1-resistant subclone, A375-R8, was obtained by limiting dilution of A375-6 cells, which had acquired resistance to IL-1 after routine passage for 3 months [24
]. These cells were cultured in RPMI-1640 medium supplemented with 15 mM HEPES and 5% heat-inactivated FBS.
Immunoprecipitations and Western blot analysis
After being transfected with the indicated plasmid, 1 x 106 cells were lysed in radioimmunoprecipitation assay (RIPA)-A buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.1% SDS, 0.5% deoxycholate, and 1% Triton-X 100), supplemented with protease inhibitors and phosphatase inhibitors. The lysates were then subjected to immunoprecipitation, and coimmunoprecipitates or 1–2% of lysates were subjected to SDS-PAGE (12.5%), transferred onto polyvinylidene difluoride membranes (Millipore, Bedford, MA, USA), and probed with antibodies described in the figure legends. For immunoprecipitation, 4 µg/20 µl lysates of anti-Sp1, anti-Sp3, or anti-HDAC 1 antibodies were used. For immunoblotting, 0.5 µg/ml anti-Sp1, anti-Sp3, anti-HDAC1, or anti-β-actin antibodies were used. The immunoreactive proteins were visualized using ECL Western blotting detection reagents (Amersham Biosciences, Piscataway, NJ, USA), and the light emission was quantified by a LAS-1000 lumino image analyzer (Fuji Film, Tokyo, Japan).
Transfection and luciferase assay
For IL-1 promoter analysis and pGL3-(IL1a-GC)3 transcriptional assays, A375 cells were cotransfected with IL-1 promoter reporter plasmid or pGL3-(IL1a-GC)3 reporter plasmid together with pCMV-β-gal and pCMV5 in the presence or absence of Sp1 or Sp3 expression plasmids (pCIneo-Sp1, pAC-NSp1, pAC-CSp1, and pCMV4-Sp3) [27
, 28
] by using Effectene transfection reagent (Qiagen, Hilden, Germany), according to the manufacturer’s instructions. For GAL-4-Sp1and -Sp3 transcriptional assays, the cells were cotransfected with the reporter plasmid pFR-Luc and GAL-4-Sp1 or -Sp3 construct with pCMV-β-gal and pCMV5. To 5 x 10 5 cells, 0.03 µg reporter plasmid, 0.02 µg pCMV-β-gal, and 0.15 µg other plasmids were transfected. In some experiments, 24 h after the transfection, the medium was replaced with medium, with or without TSA. After 24 h incubation, cell lysates were collected, and luciferase assay was performed with the luciferase reporter gene assay kit (Roche, Germany), according to the manufacturer’s instructions. Luciferase activity was determined by using the β-gal value as a basis for normalization.
EMSA
Preparation of nuclear extracts and EMSA were carried out as described previously [29
]. Samples of the nuclear extracts (15 µg) were analyzed by 6% nondenatured PAGE. For supershift assay, the nuclear extracts were incubated with 2 µg anti-Sp1 or anti-Sp3 antibodies before addition of the labeled probe. The sequences of the top strands of the individual oligonucleotide probe were as follows; 1, 5'-CGTAGCCACGCCTACTTAAG-3'; 2, 5'-TAGCCACGCCTACTTAAGAC-3'; 3, 5'-CCACGCCTACTTAAGACAAT-3'; 4, 5'-ACGCACTTGTAGCCACGTAG-3'; m1, 5'-CGTAGCCAATAATACTTAAG-3'; m2, 5'-CGTATAAACGCCTACTTAAG-3'. Competition experiments were performed with the corresponding unlabeled, double-stranded oligonucleotides.
RNA extraction and RT-PCR
Total RNA was extracted from A375 cells, and RT-PCR analysis was performed as described previously [24
]. For RT-PCR analysis, total RNA from A375 cells was reverse-transcribed into cDNA, and the amplification reaction was carried out using sense and antisense primers for IL-1 and GAPDH as previously reported [30
]. RT-PCR products were analyzed on a 1.5% agarose electrophoresis gel in the presence of ethidium bromide.
Chromatin immunoprecipitation (ChIP) assay
Formaldehyde was added directly to the cell culture medium to a final concentration of 1%. Fixation was carried out at room temperature for 10 min prior to quenching with 0.125 M glycine for 5 min. Cells containing 1 x 106 were lysed in Triton harvest buffer containing protease inhibitors, and nuclei were pelleted by centrifugation. The nuclear pellet was resuspended in RIPA-B buffer (10 mM Tris at pH 7.5, 100 mM NaCl, 1 mM EDTA, 0.5% sodium deoxycholate, 0.1% SDS, 1% Triton X-100, and protease inhibitors) and sonicated to an average fragment length 500–1000 bp. The chromatin suspension was precleared by centrifugation and then incubated overnight with 4 µg anti-Sp1, anti-Sp3, or anti-HDAC 1 antibodies or a no-antibody control. Chromatin-antibody complex was purified on protein A- or G-Sepharose beads, which have been preblocked with 100 µg/ml sheared salmon sperm DNA. The beads were washed six times with RIPA-B buffer. After the elution of immunoprecipitates and the reversal of the cross-links by heating to 65°C for 4 h, the DNA was recovered by phenol/chloroform extraction and precipitated by ethanol. Then, the association of Sp1 and HDAC1 with the IL-1 promoter was measured by PCR. The primers used for the amplification of the core IL-1 promoter region were forward: TCGGATTTCACGATTTCTCC and reverse: ACGGGGAATTTACAGGGAAG.
HDAC assay
HDAC assay was performed using the HDAC assay kit (Upstate Biotechnology), according to the manufacturer’s instructions. Briefly, 30 µl immunoprecipitated proteins were incubated with 20 µl [3H]acetate-labeled Histone H4 peptide in a total volume of 200 µl for 24 h at room temperature. The reaction was stopped by addition of 50 µl quenching solution. After centrifugation at 14,000 g for 2 min, a 100-µl aliquot of the organic phase was counted in 5 µl liquid scintillation cocktail.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
gene gene contained potential transcriptional control elements, including glucocorticoid responsive element (GRE)-like, NF-B, GC box, TATA-like sequence, and AP-1 binding sites. In previous studies, we found that A375-R8 cells, but not A375-6 cells, constitutively express IL-1 mRNA and secrete an active IL-1 molecule [24
]. IL-1β mRNA is not expressed in these cells. To analyze the promoter region involved in the constitutive activation of the IL-1 gene in A375-R8 cells, we constructed a series of 5'-end deletion mutants of the IL-1 gene with the luciferase expression plasmid. These reporter plasmids were transfected into A375-R8 and A375-6 cells, and the luciferase activity in the cell lysates was determined (Fig. 1A)
. As has been reported [26
], in A375-R8 cells, deletion of the upstream sequence to the nucleotide position from –421 to –103 resulted in an augmentation of the activity. By further deletion to –70, the transcriptional activity was decreased substantially as compared with –103. The activity was decreased markedly by further deletion to –31. These results indicate that the sequence between –103 and –70 contained a positively regulatory element(s) in A375-R8 cells but not in A375-6 cells. The positively regulatory element(s) was also found in the sequence between –70 and –31. Similar, but less effective, responses were observed in A375-6 cells.
 View larger version (20K): [in a new window] |
Figure 1. Deletion analysis of the 5'-flanking region of the human IL-1 gene in A375 cells. (A–C) Putative consensus sequences in the 5'-upstream region of the human IL-1 gene are illustrated. Numbers indicate the distance in base pairs from the transcription start. Each reporter plasmid was transfected into A375 cell, which were harvested 48 h after transfection, and the luciferase activity was measured. After normalization with β-gal activity, the luciferase activity was indicated. Mean ± SD based on triplicate cultures is shown.
|
promoter activity. Similar, but less effective, responses were observed in A375-6 cells.
The region –51 and –31 contains a sequence CGCC (–48 to –45). To investigate the functional relevance of the sequence in the IL-1 promoter activity, we constructed mutant plasmids in the region by deletion [pGL3-IL-1a(421)CC] or substitution [pGL3-IL-1a(421)mATAA]. As shown in Figure 1C
, these mutants exhibited a marked decrease in the activity, indicating that the sequence CGCC is important for the IL-1 promoter activation.
Sp1 activates the IL-1 promoter activity in a manner dependent on the GC box in A375 cells
It is known that Sp1 and Sp3 bind to the GC box in the promoter region of many genes. To investigate the effect of Sp1 on the IL-1 promoter activity, we transfected A375 cells with an expression plasmid encoding Sp1, together with IL-1 reporter plasmids, and determined the luciferase activity. As shown in Figure 2A
, Sp1 markedly augmented the luciferase activity of pGL3-IL-1a(–51) and pGL3-IL-1a(–70) in A375-R8 cells. A small increase in the transcription activity was observed in pGL3-IL-1a(–31), suggesting that the GC box or GC box-like element present in pGL3-IL-1a(–31) or the vector may be operating. However, the augmenting effect was not observed in the GC box mutant, pGL3-IL-1a(421)CC or pGL3-IL-1a(421)mATAA. These results indicate that Sp1 activates IL-1 promoter activity in a manner dependent on the GC box. Next, we compared the effects of wild-type Sp1, mutant Sp1, and Sp3 on the IL-1 promoter activity in A375-6 and A375-R8 cells. As shown in Figure 2B
, wild-type Sp1 markedly augmented the IL-1 promoter activity in A375-R8 cells and less effectively in A375-6 cells. In contrast, Sp1N and Sp1C, which lack the transactivation domain and the DNA binding domain, respectively [27
], and wild-type Sp3 did not augment the promoter activity. These results suggest that Sp1 but not Sp3 activates the IL-1 promoter activity, and the transactivation domain and the DNA binding domain are necessary for Sp1 transactivation activity.
 View larger version (11K): [in a new window] |
Figure 2. Sp1 activates the IL-1 promoter activity in a manner dependent on the GC box in A375-R8 cells. (A) A375-R8 cells were cotransfected with the indicated reporter plasmids and pCMV-β-gal in the presence or absence of pCIneo-Sp1. (B) A375-6 and A375-R8 cells were cotransfected with pGL3-IL1a(–421) and pCMV-β-gal in the presence of indicated plasmids. Cells were harvested 48 h after transfection, and the luciferase activity was measured. After normalization with β-gal activity, luciferase activity was indicated as the ratio between each activity to that of pGL3-basic-transfected cells without Sp1 expression (A) and pGL3-IL-1a(–421)-transfected cells without Sp1 expression (B). Mean ± SD based on triplicate cultures is shown. WT, Wild-type.
|
 View larger version (53K): [in a new window] |
Figure 3. Binding of nuclear proteins from A375 cells to the GC-rich region of the IL-1 promoter gene. (A) Sequences of probes are presented. (B) DNA binding activity of nuclear proteins from A375 cells to the GC-rich region of the IL-1 promoter gene. EMSA was performed using 32P-labeled probe 1 and the nuclear extracts of A375 cells. Competition analysis was performed to confirm the specific binding in the presence of 250-fold excess unlabeled probe 1. (C) Binding of Sp1 and Sp3 to the GC-rich region of the IL-1 promoter gene. EMSA was performed using 32P-probe 1 and the nuclear extracts of A375 cells. Supershift analysis was performed using anti-Sp1 or anti-Sp3 antibodies. Supershifted bands are indicated with arrows on the left. (D) Sequences required for the Sp1 and Sp3 binding in the GC-rich region of the IL-1 promoter gene. EMSA was performed using 32P-probe 1 and the nuclear extracts of A375-R8 cells in the presence of 250-fold excess of each unlabeled probe designated as probes 1, 2, 3, 4, m1, or m2 to determine the Sp1 and Sp3 binding sequence. The bands for specific binding are indicated with arrows on the left. Representative data of more than three experiments are shown.
|
promoter region. As shown in Figure 4D
, increasing amount of Sp1 led to activation of the luciferase activity in A375-R8 cells. However, Sp1 augmented the activity only slightly in A375-6 cells. In contrast, Sp3 did not induce the luciferase activity in any cells. These results indicate that Sp1 is activated in A375-R8 cells.
 View larger version (13K): [in a new window] |
Figure 4. Contribution of constitutive Sp1 activity to the activation of gene expression in A375-R8 cells. (A and B) A375 cells were cotransfected with pFR-Luc and pCMV-β-gal in the presence of GAL4-dbd, GAL4-Sp1, or GAL4-Sp3 plasmids. (C and D) A375 cells were cotransfected with pGL3-(IL1a-GC)3 and pCMV-β-gal in the presence of varying doses of pCIneo-Sp1 or pCIneo-Sp3. The ratio of the amount of DNA of pCIneo-Sp1 or pCIneo-Sp3 to pGL3-(IL1a-GC)3 was 0, 1, 2, and 4 from left to right. Cells were harvested 48 h after transfection, and the luciferase activity was measured. After normalization with β-gal activity, luciferase activity was indicated as the ratio between each activity to that of GAL4-dbd-transfected A375-6 cells (B) and as the ratio between each activity to that of A375-6 or A375-R8 cells without Sp1 or Sp3 expression (D). Mean ± SD based on triplicate cultures is shown.
|
 View larger version (10K): [in a new window] |
Figure 5. Expression levels of Sp1 and Sp3 and phosphorylation level of Sp1. A375 cells were lysed, and an equal amount of protein was subjected to SDS-PAGE. Immunoblotting (Blot) was performed with (A) anti-Sp1 or (B) anti-Sp3 antibodies. The expression level of each protein was assessed by immunoblotting of cell lysates with anti-β-actin (lower). (C) Cell lysates were immunoprecipitated (IP) with anti-Sp1 antibody and then treated with or without calf intestine phosphatase (CIP; 0.5 units/µg) and immunoblotted with anti-Sp1 or anti-phosphorylated (p)-Ser antibodies.
|
promoter activity and IL-1 mRNA expression in A375 cells. Recently, it was reported that interaction of Sp1 or Sp3 with HDAC is critical for suppressing induction of genes involved in cell cycle regulation, including p15INK4b, p19INK4d, and p21WAF1/CIP1 [33
]. To examine whether HDAC is involved in the differential activation of the IL-1 gene in A375 cells, TSA, an inhibitor of HDAC, was added to the reporter gene assay for the IL-1 promoter activity in A375-6 and A375-R8 cells. The luciferase activity without TSA was higher in A375-R8 cells as compared with A375-6 cells (Fig. 6A
). Increasing doses of TSA dramatically augmented the promoter activity in A375-6. TSA also augmented the promoter activity in A375-R8 cells, and the maximum level of the promoter activity in TSA-treated A375-6 cells was comparable with that of TSA-treated A375-R8 cells. TSA, however, did not augment the luciferase activity of the GC box mutants pGL3-IL-1a(421)CC or pGL3-IL-1a(421)mATAA (Fig. 6B)
and pGL3-basic but augmented that of pGL3-(IL1a-GC)3, which contained three tandem repeats of the GC-rich sequence in the IL-1 promoter region (Fig. 6C)
. These results indicate that the augmenting effect of TSA was specific to the GC box in the IL-1 promoter region. We then determined the effect of TSA on IL-1 mRNA expression in A375 cells. As shown in Figure 6D
, TSA induced IL-1 mRNA in A375-6 cells but not A375-R8 cells. These results suggest that HDAC is critical for the differential IL-1 mRNA expression at the transcriptional level in A375-R8 and A375-6 cells.
 View larger version (21K): [in a new window] |
Figure 6. TSA potentiates the IL-1 promoter activity and IL-1 mRNA expression in A375-6 cells. (A–C) A375 cells were transiently transfected with (A) pGL3-IL1a(–70), (B) pGL3-IL1a(–421), pGL3-IL1a(–421)CC, or pGL3-IL1a(–421)mATAA, and (C) pGL3-basic or pGL3-(IL1a-GC)3 and pCMV-β-gal. After 48 h culture, cells were treated without or with 50, 100, 200, or 500 nM TSA, from left to right (A) or 500 nM TSA (B and C) for 16 h, and luciferase activity was measured. Luciferase activity was indicated after normalization with β-gal (A) or as the ratio between each activity to that of A375-6 cells transfected with pGL3-IL-1a(–70) (B) or pGL3-basic (C) without TSA. Mean ± SD based on triplicate culture is shown. (D) A375 cells treated with or without 500 nM TSA for 16 h and total RNA were isolated. The expression levels of IL-1 and GAPDH mRNA were determined by RT-PCR. Representative data of three experiments are shown.
|
 View larger version (14K): [in a new window] |
Figure 7. TSA up-regulates Sp1 and Sp3 transactivating activity in A375 cells. (A and B) A375 cells were transiently cotransfected with pFR-Luc and pCMV-β-gal in the presence of GAL4-Sp1 (A) or GAL4-Sp3 plasmids (B). After 48 h culture, cells were treated without or with 100, 200, or 500 nM TSA, from left to right, for 16 h, and luciferase activity was measured. After normalization with β-gal activity, the luciferase activity was indicated. Mean ± SD based on triplicate culture is shown.
|
 View larger version (70K): [in a new window] |
Figure 8. Expression of HDAC1 and its interaction with Sp1 or Sp3 in A375 cells (A), which were lysed, and an equal amount of protein was subjected to SDS-PAGE. Immunoblotting was performed with anti-HDAC1 antibody. The expression level of each protein was assessed by immunoblotting cell lysates with anti-β-actin (lower). (B and C) A375 cells were lysed and immunoprecipated with (B) anti-Sp1 or (C) anti-Sp3 antibodies and immunoblotted with anti-HDAC1 antibody. The same Western blot filters were probed with (B) anti-Sp1 or (C) anti-Sp3 antibodies to estimate the immunoprecipitation efficiency (lower). The same experiment as representative data of more than three experiments are shown.
|
gene in A375 cells gene activation, ChIP assay was performed. As shown in Figure 9A
, Sp1 and HDAC1 appeared to bind to the promoter region of the IL-1 gene in A375-R8 and A375-6 cells. However, the binding level of Sp1 and HDAC1 was not significantly different between A375-R8 and A375-6 cells. To confirm whether HDAC is critical for regulating the transactivation activity of Sp1, HDAC activity was measured. The activity of HDAC associated with Sp3 was higher as compared with Sp1 (Fig. 9B)
. HDAC activity bound to Sp1 or Sp3 was lower in A375-R8 cells as compared with A375-6 cells. In addition, the total activity of HDAC1 was lower in A375-R8 cells as compared with A375-6 cells. These results suggest that the reduced activity of HDAC1 is critical for the constitutive gene activation of IL-1.
 View larger version (30K): [in a new window] |
Figure 9. Sp1 and HDAC1 bind to the promoter region of the IL-1 gene in A375 cells. (A) ChIP of the gene with anti-Sp1 or anti-HDAC1 antibodies from A375 cells detected by PCR targeting the promoter region. Input corresponds to PCR mixtures containing 4% of the total amount of proteins used in immunoprecipitation reactions. (B) A375 cells were lysed, and an equal amount of protein was immunoprecipitated with anti-Sp1, anti-Sp3, or anti-HDAC1 antibodies. HDAC activity was measured as described in Materials and Methods. Representative data of three experiments are shown.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
gene, interaction of HDAC1 with Sp1 or Sp3, and reduction of HDAC1 activity are critical for the constitutive activation of the IL-1 gene. Consistent with our previous report [26
], the sequence between –103 and –70 of the IL-1 gene contained a positive regulatory element(s), and the sequence between –421 and –103 contained a negative regulatory element(s). The NF-B site between –103 and –70 has been shown to be important for an autocrine effect of IL-1. The positive regulatory element(s) were also found in the sequence between –70 and –31, which contained the GC box. In our previous study, we found that this region, –70 to –31, is not involved in the autocrine effect of IL-1. Similar but less-ffective responses were observed in A375-6 cells. There are no differences in the sequence of the promoter of the IL-1 genomic gene, –70 to –1, between A375-6 and A375-R8 cells (data not shown). Recently, it is reported that methylation of the adjacent Sp1 binding site inhibits Sp1 binding to DNA [35
]. However, treatment of A375-6 cells with 5-aza-CdR, an inhibitor of methylation, for 6 days did not induce IL-1 mRNA, suggesting that methylation cannot account for the differential IL-1 mRNA expression either (data not shown).
We analyzed the region critical for the constitutive activation of the IL-1 promoter. By deletion and substitution mutation analyses, the sequence CGCC at –48 to –45 appeared to be important for the promoter activity. EMSA analysis revealed that Sp1 and Sp3 are bound to the nucleotide, –56 to –37, containing the GC box. In addition to the sequence CGCC, the sequence GCC at –52 to –50 contributed to the maximum binding of Sp1 and Sp3 to the sequence. In agreement with the findings, the transactivation activity of plasmid IL51, which lacks G at –52, was lower as compared with the wild-type plasmid IL71.
We next determined the effect of Sp1 or Sp3 on the transactivation activity of the IL-1 promoter. Overexpression of wild-type Sp1, but not mutant Sp1, up-regulated the promoter activity in A375-R8 cells and less effectively in A375-6 cells. In contrast, overexpression of wild-type Sp3 did not up-regulate the promoter activity. Therefore, Sp1 but not Sp3 appeared to contribute to the constitutive transactivation of the IL-1 promoter. As the augmenting effect of Sp1 is higher in A375-R8 cells as compared with A375-6 cells, it was suggested that there are differences in the mode of activation of Sp1 and Sp3 between A375-6 and A375-R8 cells.
Sp1 and Sp3 are expressed in a variety of tissues and play an important role in regulation of the activation of genes involved in the development, proliferation, and differentiation of cells and inflammation [36
]. Therefore, the expression of Sp1 and Sp3 is strictly regulated at multiple levels, including transcription, translation, and the proteasome-dependent degradation pathway. As it is reported that Sp1 and Sp3 are degraded through a ubiquitin-proteasome pathway [37
], the protein levels of Sp1 and Sp3 were compared between A375-R8 and A375-6 cells. However, there were no differences between these cells, indicating that the differential transactivation activity for IL-1 promoter activity was not a result of differential expression levels of Sp1 and Sp3.
Constitutive activation as well as a high level expression of Sp1 is reported to be essential for constitutive production of an endothelial growth factor in human pancreatic adenocarcinoma [38 ]. Therefore, we sought to determine whether Sp1 or Sp3 is constitutively activated in A375-R8 cells. As expected, Sp1 was markedly activated in A375-R8 cells. Although Sp3 was also activated in A375-R8 cells more than A375-6 cells, the activation level was lower as compared with Sp1 in any cells. The functions of Sp1 and Sp3 are regulated by post-translational modifications and an interaction with other regulatory molecules. The former type regulation includes phosphorylation [32 , 39 ], acetylation [40 ], and glycosylation [41 ]. The DNA binding ability of Sp1 is regulated by phosphorylation of Ser at the N-terminal Glu-rich region, and Sp1 is phosphorylated by ERK1/2 or JNK upon stimulation with fibroblast growth factor, hepatocyte growth factor, peroxide, or a variety of extracellular stimuli [32 , 42 ]. As hyperphosphorylation is found in melanoma, leukemia cells, and lung carcinoma, the DNA binding ability of Sp1 or Sp3 was determined in A375 cells. However, there were no differences in the amount of DNA-bound Sp1 and Sp3 between A375-6 and A375-R8 cells. In addition, Sp1 was phosphorylated constitutively and comparably in these two cell clones. Therefore, the differential transactivation activity of Sp1 was not a result of phosphorylation.
HDAC negatively regulates transcription of genes by inducing the conformational changes through removing the acetyl group from the histones comprising the nucleosome. It also regulates the activity of transcription factors, including p53, GATA-1, and TFIIE, through deacetylation and does so by recruitment of corepressors, such as m-Sin3A, nuclear receptor corepressor, and silencing mediator of retinoid and thyroid receptors, to Sp1 or NF-B [34
, 43
, 44
]. Aberrant regulation of HDAC recruitment is implicated in the pathogenesis of tumors, as inhibitors of HDAC cause growth arrest, cell differentiation, and/or apoptosis of tumor cells [45
]. For instance, HDAC1 interacts with Sp1 bound to the promoter region of p21WAF1/CIP1; consequently, it inhibits the transactivation activity of Sp1, which leads to down-regulation of p21 expression [33
]. In this study, TSA up-regulated the promoter activity of IL-1 in A375-6 and A375-R8 cells, which appeared to be specific to the GC box of the promoter. It is interesting that the activation rate was dramatically high (40-fold) in A375-6 cells, and the maximum level of the promoter activity in TSA-treated A375-6 cells was comparable with that of TSA-treated A375-R8 cells. Furthermore, treatment with TSA induced IL-1 mRNA expression in A375-6 cells. Therefore, HDAC appeared to be critical for the regulation of IL-1 expression in A375 cells. As TSA also activated the transactivation activity of Sp1 and Sp3, it was suggested that recruitment of HDAC to Sp1 or Sp3 or activity of HDAC recruited to Sp1 or Sp3 was inhibited by TSA.
Among HDAC family members, HDAC1 and HDAC2 are shown to interact with Sp1 or Sp3 [34
]. The expression levels of Sp1 and Sp3 were comparable between A375-R8 and A375-6 cells, and there were no differences in the levels of HDAC1 interaction with Sp1 or Sp3 in these cells. HDAC1 indirectly binds to DNA through the complex or transcription factors including Sp1 and Sp3 [34
]. ChIP assay confirmed the binding of Sp1 and HDAC1 to the promoter region of the IL-1 gene in A375-R8 and A375-6 cells. It is interesting that our study revealed that the activity of HDAC, associated with Sp1 or Sp3, and that of HDAC1 are lower in A375-R8 cells as compared with A375-6 cells. In addition, the activity of HDAC associated with Sp3 was higher than that with Sp1. Therefore, not only the interaction between HDAC1 and Sp1 or Sp3 but also the activity of HDAC1 are critical for the differential activation of the IL-1 gene. The mechanism of HDAC activity reduction in A375-R8 cells is not known. It is hypothesized that HDAC activity is reduced by oxidative stress, including NO and its derivatives [46
]. Therefore, it is tempting to speculate that oxidative stress is induced in A375-R8 cells, which may down-regulate HDAC activity.
The GC box in the promoter of the IL-1 gene has been reported to be important for the constitutive gene expression of IL-1 in many cells, including monkey kidney cell line COS-7, human osteosarcoma cell line MG-63, and human esophageal carcinoma cell line EC-GI [4
]. Alveolar macrophages and blood monocytes from patients with asthma exhibit enhanced cytokine production and reduced activity of HDAC [47
]. In patients of chronic obstructive pulmonary disease, alveolar macrophages are critical in orchestrating the chronic inflammation through the release of proteases and inflammatory cytokines [46
]. Alveolar macrophages of cigarette smokers show a reduction in HDAC activity and expression of HDAC2 compared with cells from healthy individuals [48
]. Therefore, it is possible that the GC box, Sp1, Sp3, and HDAC are similarly important in the aberrant inflammatory cytokine production in these cells as well.
Received January 6, 2006; revised August 2, 2007; accepted August 21, 2007.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B regulates IL-1 β transcription through a consensus NF- B binding site and a nonconsensus CRE-like site J. Immunol. 153,712-723[Abstract] Eur. Cytokine Netw. 5,533-538[Medline] antisense oligomer Science 249,1570-1574 in the growth of adult human T-cell leukemia cells Cancer Res. 49,1143-1147 in systemic sclerosis J. Clin. Invest. 103,1253-1260[Medline] and liver metastasis in gastric carcinoma Cancer 91,1272-1276[CrossRef][Medline] and β by normal and cancerous human ovarian tissues Eur. Cytokine Netw. 8,179-187[Medline] and β in early passage fibroblasts from aging individuals Exp. Gerontol. 28,505-513[CrossRef][Medline] gene expression in human melanoma cells: autocrine stimulation through NF-B activation by endogenous IL-1 Cytokine 10,872-879[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
