Snail associates with EGR-1 and SP-1 to upregulate transcriptional activation of p15INK4b Chi-Tan Hu1, Tsu-Yao Chang2, Chuan-Chu Cheng2, Chun-Shan Liu2, Jia-Ru Wu2, Ming-Che Li1 and Wen-Sheng Wu2 Research Centre for Hepatology, Buddhist Tzu Chi General Hospital and Tzu Chi University, Hualien, Taiwan Institute of Medical Biotechnology, College of Medicine, Tzu Chi University, Hualein, Taiwan Keywords EGR-1; p15INK4b; Snail; SP-1; transcriptional regulation Correspondence Wen-Sheng Wu, Institute of Medical Biotechnology, College of Medicine, Tzu Chi University, No 701, Chung Yang Rd, Sec 3, Hualien 970, Taiwan Fax: +8867 03 8571917 Tel: +8867 03 8565301; ext 2327 E-mail: wuwstcu1234@yahoo.com.tw (Received 10 October 2009, revised 10 December 2009, accepted 18 December 2009) doi:10.1111/j.1742-4658.2009.07553.x Snail is a multifunctional transcriptional factor that has been described as a repressor in many different contexts It is also proposed as an activator in a few cases relevant to tumor progression and cell-cycle arrest This study investigated the detailed mechanisms by which Snail upregulates gene expression of the CDK inhibitor p15INK4b in HepG2 induced by the tumor promoter tetradecanoyl phorbol acetate (TPA) Using deletion mapping, the TPA-responsive element on the p15INK4b promoter was located between 77 and 228 bp upstream of the transcriptional initiation site, within which the putative binding regions of early growth response gene (EGR-1) and stimulatory protein (SP-1) were found Gene expression of EGR-1, Snail and SP-1 can be induced by TPA within 0.5–6 h In addition, basal levels of SP-1, but not of the other two transcriptional factors, were observed Blockade of TPA-induced gene expression of Snail, EGR-1 or SP-1 suppressed activation of the p15–pro228 reporter plasmid harboring the TPA-responsive element More detailed deletion mapping and site-directed mutagenesis further concluded that the overlapping EGR-1/SP-1-binding site was required for TPA-induced p15–pro228 activation In an EMSA, a DNA–protein complex was elevated by TPA, which can be blocked by antibodies against EGR-1, SP-1 or Snail at h Immunoprecipitation/ western blotting demonstrated that TPA could trigger the association of EGR-1 with Snail or SP-1 Furthermore, a double chromatin immunoprecipitation assay verified that EGR-1 could form a complex with Snail or SP-1 on the TPA-responsive element after treatment with TPA for 2–6 h Finally, we demonstrated a novel Snail-target region which could be bound by Snail and was also required for TPA-induced p15–pro228 activation In conclusion, Snail associates with EGR-1 and SP-1 to mediate TPA-induced transcriptional upregulation of p15INK4b in HepG2 Structured digital abstract  MINT-7384899: Snail (uniprotkb:O95863) physically interacts (MI:0915) with EGR-1 (uniprotkb:P18146) by anti bait coimmunoprecipitation (MI:0006)  MINT-7384908: SP-1 (uniprotkb:P08047) physically interacts (MI:0915) with EGR-1 (uniprotkb:P18146) by anti bait coimmunoprecipitation (MI:0006) Abbreviations ChIP, chromatin immunoprecipitaion; EGR-1, early growth response gene 1; MMPs, matrix metalloproteinases; shRNA, short hairpin RNA; SP-1, stimulatory protein-1; TPA, tetradecanoyl phorbol acetate 1202 FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS Transcriptional activation of p15INK4b C.-T Hu et al Introduction The Snail family of zinc-finger transcription factors was first described in Drosophila melanogaster [1], where they were shown to be essential for formation of the mesoderm [2] Snail may trigger a phenotypic change called epithelial mesenchymal transition required for embryonic development [3,4] Recent studies have linked Snail to tumor metastasis because epithelial mesenchymal transition is a prerequisite for cell migration and invasiveness [5–8] Snail genes can be induced by different growth factors and cytokines, such as hepatocyte growth factor [9], transforming growth factor b [10,11] and WNTs [12], that may trigger tumor progression However, Snail can also function as a negative regulator of cell growth [13] Interestingly, cell division is impaired in Snailexpressing epithelial cells that have undergone epithelial mesenchymal transition [13–15] and Snail may trigger invasion while suppressing tumor growth [16] Our recent report also demonstrated that Snail may simultaneously trigger both growth inhibition and cell migration of HepG2 [17] Conventionally, Snail was known to be a negative regulator of gene expression and responsible for diverse cellular effects Snail was known to repress epithelial markers such as E-cadherin [18,19] Also, the Crumbs polarity complex, a key apico-basal polarity factor, was also found to be suppressed by Snail for epithelial mesenchymal transition [20] With regard to the negative regulation of cell growth, Snail may repress Cyclin D2 to block the cell cycle in the MDCK cell line [13] Recently, the possible role of Snail as a transcriptional activator was highlighted For example, Slug, a Snail-related transcriptional factor, was found to be capable of activating its own promoter Moreover, Snail was implicated in the upregulation of migration- and invasion-related genes including matrix metalloproteinase (MMP-9) [21–23] and integrin b subunits [24] Our recent report also demonstrated that Snail was responsible for upregulation of the CDK inhibitor p15INK4b required for tetradecanoyl phorbol acetate (TPA)-induced cell-cycle arrest [17] Snail family proteins contain a C-terminal tandem C2H2 zinc finger as a sequence-specific DNA-binding motif and an N-terminal SNAG repression domain The detailed mechanisms by which Snail acts as a transcriptional repressor have been intensively studied Snail may bind to a consensus sequence such as E-box, which is also the binding site for basic helix– loop–helix transcriptional factors on the target promoter, thus interfering with gene expression More recent reports have further shown that Snail may asso- ciate with polycomb repressive complex or protein arginine methyltransferase to repress E-cadherin expression [25,26] However, how Snail upregulates gene expression is not yet clear In a recent report, the Snail-responsive element(s) on the proximal MMP-9 promoter was identified in MDCK cells This region contains the putative binding sites of stimulatory protein-1 (SP-1) and Ets-1 which are critical for the transactivation of MMP-9 [23] However, whether Snail binds directly to this region and whether it might cooperate with other transcriptional factors to activate MMP-9 promoter were not addressed Recently, we investigated the mechanisms by which Snail mediates TPA-induced upregulation of p15INK4b and a TPA-responsive element was identified on the p15INK4b promoter [17] In this study, we further pinpoint the critical regions by which Snail associates with other transcriptional factors such as early growth response gene (EGR-1) and SP-1 to upregulate transcription of p15INK4b Results Deletion mapping for the TPA-responsive element on the p15INK4b promoter Initially, detailed deletion mapping using p15INK4b promoter constructs of various lengths was performed to pinpoint the exact region responsible for promoter activation (Fig 1, left) Three constructs, p15–profull, p15–pro461 and p15–pro228, contain regions encompassing 1006, 461 and 228 bp, respectively, upstream of the translational start site, whereas p15–pro233 contains 233 bp of the distal part of the promoter within p15–pro461, and p15–pro77 contains 77 bp in the proximal part of the promoter within p15–pro228 As demonstrated in Fig (right), p15–profull, p15– pro461, p15–pro228 and p15–pro233 exhibited basal promoter activities which were 4.95-, 4.12-, 1.67- and 2.29-fold higher, respectively, than that of the pGL3 vector After treatment of HepG2 cells with 50 nm TPA for 24 h, the promoter activity of p15–profull, p15–pro461 and p15–pro228 increased by 3.8-, 3.7and 4.6-fold, respectively, in comparison with that of untreated HepG2 in each experimental group It is worth noting that the TPA-induced promoter activity of p15–pro228 was slightly higher than that of p15– profull and p15–pro461, although its basal promoter activity decreased significantly Also, the promoter activity of p15–pro233, which contains the distal part of p15–pro461, could be induced by TPA by only FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS 1203 Transcriptional activation of p15INK4b C.-T Hu et al Fig Deletion mapping for identification of tetradecanoyl phorbol acetate (TPA)-responsive element for promoter activation of p15INK4b The full-length p15INK4b promoter (p15–profull) and other shorter promoter constructs are shown in the left-hand panel HepG2 cells were transfected with pGL3 vector or various p15INK4b promoter plasmids coupled with pRL control plasmid, and then untreated (white bar) or treated with 50 nM TPA (black bar) for 24 h Dual luciferase assays were performed The relative promoter activity of each sample was calculated, taking the data for pGL3 vector in untreated cells as 1.0 The results of 5–7 experiments were averaged with a C.V of 5.0–8.0% The numbers beside the solid bar represent the-folds of induction by TPA for each promoter construct **Statistical significance (P < 0.005) between the indicated groups 1.5-fold, much less than the promoter activity of p15– pro461 and p15–pro228 (Fig 1, right) Thus, it seemed that the TPA-responsive element is mainly located at the promoter region on p15–pro228, which belongs to the proximal part of p15–pro461 Furthermore, p15–pro77, which contained the proximal part of the promoter region in p15–pro228, did not exhibit basal or TPA-induced promoter activity This further narrowed the TPA-responsive element to the region between 77 to 228 bp upstream of the translational initiation site strating that TPA-induced Snail protein at h was suppressed by 40–55% by the transfection of sh1, sh2 and sh3 (Fig S1A) Induction of gene expression of Snail, EGR-1 and SP-1 by TPA To investigate whether any other transcription factors cooperate with Snail for activation of the p15INK4b Snail was required for TPA-induced activation of p15–pro228 Our previous report showed that Snail was required for TPA-induced activation of p15–pro461 [17], and we further investigated whether it was also required for activation of p15–pro228 For this purpose, a short hairpin RNA (shRNA) technique was used to observe whether knockdown of Snail gene expression prevents TPAinduced activation of p15–pro228 Three combinations of effective Snail shRNA, namely sh1 (fragments 18 and 20), sh2 (fragments 18 and 19) or sh3 (fragments 19 and 20) prevented TPA-induced activation of p15–pro228 at 24 h by 35, 50 and 40%, respectively, compared with Lamin A shRNA (used as control shRNA) (Fig 2) It appeared that Lamin A shRNA prevented TPA-induced activation of p15–pro228 by 10–20% (Fig and data not shown), probably because of the involvement of Lamin A in transcriptional regulation The effects of Snail shRNAs were verified by western blotting, demon1204 Fig Suppression of tetradecanoyl phorbol acetate (TPA)-induced p15–pro228 activation by Snail shRNA HepG2 cells were co-transfected with pGL3 and pRL, or with p15–pro228 and pRL coupled with combinations of Snail shRNA (sh1, sh2 or sh3 as indicated in the text) or Lamin shRNA as control Transfected cells were untreated (white bar) or treated with 50 nM TPA (black bar) for 24 h Dual luciferase assays were performed and the relative promoter activity of each sample was calculated, taking the data for pGL3 vector in untreated cells as 1.0 Statistical significance at *P < 0.05 and **P < 0.005 between the indicated groups FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS Transcriptional activation of p15INK4b C.-T Hu et al promoter, genomatix software (v GmbH 1998–2008) was used to search the putative transcriptional factorbinding regions within )226 and )80 bp on the TPAA B responsive element (Table S1) Interestingly, we found binding regions (located between )202 and )169 bp) for two transcriptional factors, EGR-1 and SP-1, which according to previous studies could be induced by TPA [27–29] Thus we set out to investigate whether these candidate transcriptional factors and their corresponding cis-acting recognition sequences are involved in TPA-induced p15INK4b promoter activation Initially, the gene expression profiles of these candidate transcriptional factors were investigated Quantitative real-time PCR analysis clearly demonstrated that, compared with control levels, EGR-1 mRNA was dramatically induced by 50 nm TPA at 30 (47.0fold), followed by a gradual decrease from to h (to  10-fold), finally returning to the basal level at h (Fig 3A, upper) Also, Snail mRNA was significantly induced by TPA by  1.5- to 2.3-fold within 30 to h, maximally induced by 5.0-fold at h, decreased to  3.0-fold at h, and returned to the basal level at h (Fig 3A, middle) SP-1mRNA was maximally induced by 5.2-fold after treatment of TPA for 1h, followed by a decrease within 2)4 h (to  2.1- to 2.3fold) and returned to the basal level at h (Fig 3A, lower) Notably, highly constitutive SP-1 mRNA expression was observed, which was 5.2- and 5.0-fold that of EGR-1 and Snail, respectively (Fig 3B) On the other hand, using western blot analysis, EGR-1 protein was found to increase dramatically by  5.0fold following treatment with TPA for h, gradually decrease from to h and had disappeared totally at h (Fig 3C) As seen in the mRNA level (Fig 3A), SP-1 protein exhibited constitutive expression After TPA treatment, SP-1 protein increased significantly by 2–2.5-fold within 1–4 h and returned to the basal level at h (Fig 3C) Also, Snail protein was significantly induced by TPA within 1–2 h, maximally induced by  2.5-fold at h, and decreased to the basal level at h (Fig 3C) Collectively, these results indicated that C Fig Tetradecanoyl phorbol acetate (TPA)-induced gene expression of EGR-1, Snail and SP-1 in HepG2 HepG2 cells were untreated (con) or treated with 50 nM TPA for 0.5, 1, 2, and h (A and C) Real-time RT/PCR (A) and western blot (C) of EGR-1, SP-1 and Snail were performed In (A), the relative mRNA level for EGR-1, SP-1 and Snail at each time point of TPA treatment was caculated, taking the basal expression of each gene (con) as 1.0 (B) Real-time PCR for comparison of the basal levels of the three genes, taking the amount of EGR-1 as 1.0 In (A) and (B), the results are the average of five experiments with a C.V of 5.0–8.5% (A) Statistical significance at *P < 0.05 and **P < 0.005 between the results for TPA-treated and untreated HepG2 (con) (B) Statistical significance at **P < 0.005 between the results for SP-1 and the other two genes ERK was the internal control in (C) M, molecular mass marker FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS 1205 Transcriptional activation of p15INK4b C.-T Hu et al EGR-1 and SP-1 were required for TPA-induced promoter activation of p15INK4b Fig Prevention of tetradecanoyl phorbol acetate (TPA)-induced activation of p15–pro228 by blocking Snail, EGR-1 and SP-1 expression (A) Time-course analysis of TPA-induced activation of p15–pro228 HepG2 cells were transfected with pGL3 vector or p15–pro228 plus pRL control plasmid followed by treatment with TPA for the times indicated Dual luciferase assays were performed The relative promoter activity of each sample was calculated, taking the data for pGL3 vector in untreated cells as 1.0 The results are the average of three experiments with a C.V of 5.0–8.5% (B) Knockdown of Snail, EGR-1 or SP-1 prevented TPA-induced p15–pro228 activation HepG2 cells were co-transfected with pGL3 and pRL, or with p15–pro228 and pRL coupled with shRNA of SP-1 (SP46), EGR1 (E33), Snail (SN18) or shRNA of Lamin as mock, followed by no treatment (white bar) or treatment with 50 nM TPA (black bar) for and 12 h Dual luciferase assays were performed and the relative promoter activity of each sample was calculated, taking the data for pGL3 vector in untreated cells as 1.0 The results are the average of 5–7 experiments with a C.V of 5.0–8.5% Statistical significance at **P < 0.005 between HepG2 cells co-transfected with the indicated shRNA and with mock shRNA in addition to Snail, EGR-1 and SP-1 can also be induced by TPA, which may be required for promoter activation of p15INK4b 1206 Because gene expression of all these transcriptional factors can be rapidly induced by TPA, the p15INK4b promoter may be activated at an early phase of TPA treatment As demonstrated in the time-course experiment (Fig 4A), promoter activity of p15–pro228 can be significantly induced by TPA between and h, followed by a dramatic increase at 12 h (by  20–25fold) and sustained until 24 h We further examined whether blocking gene expression of the aforementioned transcriptional factors may prevent TPAinduced p15–pro228 activation at earlier time points As demonstrated in Fig 4B, TPA-induced promoter activation of p15–pro228 at and 12 h was greatly suppressed by shRNA of SP-1 (fragment 46) and EGR-1 (fragment 36) by 90–95 and 80–95%, respectively, compared with the mock (Lamin A) shRNA In comparison, Snail shRNA (fragments 18) prevented less ( 45–80%) TPA-induced activation of p15– pro228 The effects of the shRNAs for these transcriptional factors were verified by western blot analysis The TPA-induced increase in EGR-1 protein at h was attenuated by transfection of EGR-1 shRNA (fragments 33 and 36) by  75–80%, compared with that of mock (Lamin A) shRNA (Fig S1B) Similarly, the TPA-induced increase in SP-1 at h was suppressed by  60–75% by SP-1 shRNA (fragments 46 or 47) (data not shown) Also, the TPA-induced increase in Snail was attenuated by Snail shRNA (fragments 18) at h by  50% (data not shown) Taken together, these results indicated that, in addition to Snail, both EGR-1 and SP-1 were required for TPA-induced promoter activation of p15INK4b Identification of the critical regions in the TPA-responsive element We further investigated whether the putative binding motifs for EGR-1 and SP-1 are critical for activation of the p15INK4b promoter According to genomatix software, there is an overlapping EGR-1/SP-1 binding region ()202 to )186 bp) and an adjacent single SP-1 region ()183 to )169 bp) within the TPA-responsive element (Table S1) To investigate which is crucial for TPA-induced p15INK4b promoter activation, three deletion constructs of p15–pro228 were employed (Fig 5A, left) In one, namely p15–pro228DE/S-S, both the overlapping EGR-1/SP-1-binding site and the adjacent single SP-1 site (the region between )223 and )142 bp) were deleted In the other two, namely p15–pro228DE/ S and p15–pro228DS, the region containing the over- FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS Transcriptional activation of p15INK4b C.-T Hu et al A Fig Promoter assay of p15–pro228 with deletions or point mutations on various putative transcriptional factor binding sites HepG2 were co-transfected with pRL control plasmid and wild type p15–pro228 or pRL and p15–pro228 with deletion (A) or point mutations (B) on the EGR-1/SP-1 overlapping site, the single SP-1 site and the proposed Snail-binding motif within )202 to )184, )183 to )169 and )207 to )202 bp upstream of the translational initiation site, respectively The map for the sites of deletion and point mutation on each region are show in the left-hand panel in (A) and (B) Transfected cells were either not treated (white bar) or treated with 50 nM tetradecanoyl phorbol acetate (TPA; black bar) for 24 h Dual luciferase assays were performed and the relative promoter activity of each sample was calculated, taking the data for pGL3 vector in untreated cells as 1.0 The numbers beside the black bar indicate the-fold of TPA-induced promoter activity compared with each untreated group The results are the average of 5–7 experiments with a C.V of 5.0–7.0 Statistical significance at **P < 0.005 between the indicated groups B lapping EGR-1/SP-1 site ()223 to )179 bp) and the single SP-1 site ()176 to )143 bp), respectively, were deleted Interestingly, the TPA-induced promoter activation of p15–pro228DE/S-S (3.17-fold) and p15– pro228DE/S (3.29-fold) decreased by  80% compared with those of the parental p15–pro228 (16.28-fold) (Fig 5A, right) By contrast, the TPA-induced activation of p15–pro228DS (16.38-fold) was the same as that of the parental p15–pro228, although its basal activity was slightly reduced Furthermore, promoter assays using p15–pro228 with point mutations in the putative EGR-1- and SP-1-binding sites were performed As demonstrated in Fig 5B, TPA-induced activation of a p15–pro228 mutant (p15–pro228 E/S*) with three altered nucleotides on the EGR-1/SP-1 overlapping site (GGG fi TAT at )194 to )192) was reduced by  80% compared with that of wild-type p15–pro228 By contrast, TPA-induced activation of the p15–pro228 mutants, namely p15–pro228 SP-1*, with three altered nucleotides in the single SP-1 region (TGG fi GAC at )176 to )174), decreased by only 20% Taken together, these results strongly indicated that the overlapping EGR-1/SP-1, but not the single SP-1, binding region was essential for TPA-induced p15–pro228 activation EMSA for in vitro DNA-binding activity of the candidate transcriptional factors The critical role of the overlapping EGR-1/SP-1 region was further investigated by EMSA using nuclear extract obtained from HepG2 with or without TPA treatment The probe p15proE/S contains a subregion ()210 to )181 bp) of the TPA-response element harboring the overlapping EGR-1/SP-1-binding site (Fig 6A) As demonstrated in Fig 6B, three mobility-retarded DNA protein complexes, denoted as SI, SII and SIII, increased in the EMSA of HepG2 treated with TPA for 1, and h SII increased significantly by  2.0-fold after treatment of the cell with TPA for h and dramatically increased at and h by  9.5- and 10.0-fold, respectively, compared with that of untreated HepG2 SI, which migrated more slowly than SII, increased significantly at and h by  1.5- to 2.3-fold SIII, which migrated faster than SII, increased significantly within 1–6 h by 2.0 to 3.0-fold In the competition analysis for the sample from cells treated with TPA for h (Fig 6B, lanes 6–7), SII could be 90% suppressed by addition of the unlabeled wild-type EGR-1/SP-1 overlapping fragment (denoted as E/S competitor in Fig 6A) which contains FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS 1207 Transcriptional activation of p15INK4b C.-T Hu et al Fig EMSA for subregions on the tetradecanoyl phorbol acetate (TPA)-responsive element of the p15INK4b promoter (A) Schematic representation of subregions in the TPA-responsive element ()228 to )77 bp), including the p15proE/S probe ()210 to )181 bp), the E/S competitor ()203 to )184 bp) used for EMSA in (B) and (C) and the p15–proSN probe ()218 to )197 bp) used for EMSA in Fig (B) Time-course study for in vitro DNA binding activity Nuclear extracts of untreated HepG2 (control) or HepG2 treated with 50 nM TPA for 1, and h were incubated with p15proE/S probe for EMSA For competition, unlabeled wild-type or mutant E/ S competitor was included in the EMSA reaction using a sample of HepG2 treated with TPA for h (C) Detection of the proteins bound on p15proE/S Nuclear extracts of untreated HepG2 (lane 1) and HepG2 treated with 50 nM TPA for and h were preincubated with antibodies (2 lg each) against the indicated transcriptional factors or Raf (used as the negative control antibody) for 30 min, followed by EMSA reaction Lane in (B) and (C) are the samples of probe only The results are representative of two reproducible experiments A B C the region spanning )203 to )184 bp SII was suppressed by only  40% by a mutant of the E/S competitor with three altered nucleotides (GGG fi TAT at )194 to )192) However, TPA-induced elevation of SI was suppressed slightly by the addition of unlabeled wild-type or mutant E/S competitor Also, TPAinduced elevation of SIII was not significantly influenced by either wild-type or mutant E/S competitor at h Thus, it appeared that among the three TPAinduced mobility-retarded bands of p15proE/S, SII was not only the most abundant, but also the most 1208 specific for EGR-1/SP-1-overlapping region Furthermore, antibody-blocking experiments were performed to examine which complex contained the candidate transcriptional factors induced by TPA As demonstrated in Fig 6C, TPA-induced elevation of complex SII at h was greatly reduced by preincubation of the nuclear extracts with antibodies of SP-1 and EGR-1 (lanes and 8), but decreased only slightly if Snail antibody was used (lane 7) At the h time point, SII was greatly reduced by preincubation of the nuclear extract with antibodies against each of the three transcription factors (lanes 9–11) Raf antibody (as the control antibody) did not block TPA-induced elevation of complex SII at either time point (lanes 12 and 13) By contrast, TPA-induced elevations of both complex SI and SIII were not significantly blocked by any antibodies Thus, SII, but not SI and SIII, is the most important DNA–protein complex that contains the candidate transcriptional factors induced by TPA Taken together, by examining the pattern of SII we suggest that the in vitro DNA-binding activity of all three transcriptional factors toward p15proE/S could be elevated within 2–6 h following treatment with TPA Immunoprecipitation/western blotting for TPA-induced association of the candidate transcriptional factors Thus far, Snail, EGR-1 and SP-1 appeared to act in concert for TPA-induced p15INK4b promoter activation We further investigated whether they may interact with each other during this process By immunopreciptation of Snail coupled with EGR-1 FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS Transcriptional activation of p15INK4b C.-T Hu et al A Fig Immunoprecipitation (IP)/western blotting for the association of Snail, EGR and SP-1 HepG2 cells were untreated (con) or treated with 50 nM tetradecanoyl phorbol acetate (TPA) for 0.5, 2, and h Eight hundred micrograms of protein for each sample was used for immunopreciptation followed by western blotting The antibodies used for immunopreciptation and western blotting are indicated on the right and left of the gel, respectively Positions of EGR-1 and nonspecific bands are indicated by an arrow Western blots of ERK were performed to monitor equal amounts of protein in each sample used for immunopreciptation The results are representative of 2–3 reproducible experiments western blot analysis (Fig 7, upper), EGR-1 could be detected in the immunopreciptates of Snail from cells treated with TPA at 6.0 h By immunoprecipitation of SP-1 coupled with EGR-1 western blot (Fig 7, lower), EGR-1 (indicated by large arrow) could be abundantly detected in the immunopreciptates of SP-1 from cells treated with TPA at the 2.0 h time point A nonspecific band (indicated by small arrow) below the EGR-1-specific band can be detected in all samples analyzed In addition, SP-1 could not be detected in the immunopreciptate of Snail from TPA-treated HepG2 (data not shown) Taken together, it appeared that TPA could induce the association of EGR-1 with both Snail and SP-1, but not the association of Snail with SP-1 B Chromatin immunoprecipitaion assay for in vivo DNA-binding activity of the candidate transcriptional factors Fig Binding and interaction of the transcriptional factors on tetradecanoyl phorbol acetate (TPA)-responsive element in vivo (A) Time-course analysis for single chromatin immunoprecipitaion (ChIP) HepG2 cells were treated with 50 nM TPA for 0, 1, 2, 6, 12 and 24 h, ChIP assays for binding of Snail, EGR-1 and SP-1 on Fragment 228 were performed Ab, antibody; IP, immunoprecipitation (B) HepG2 cells were treated with 50 nM TPA for 0, and h, doubled ChIP assays were performed using the indicated antibodies for first immunoprecipitation (left) and second immunoprecipitation (right) Raf antibody was used as MOCK antibody for the first immunoprecipitation in both experimental groups In both (A) and (B), histone antibody was used to precipitate the promoter region of GAPDH as the positive control group In (A), PCR products of Fragment 228 from each sample are shown as the Input These results are representative of three reproducible experiments To further examine whether TPA may induce DNA binding of the candidate transcriptional factors toward the p15INK4b promoter in vivo, chromatin immunoprecipitaion (ChIP) assays were performed As shown in Fig 8A, the DNA-binding activity of all three transcriptional factors toward the promoter fragment encompassing TPA-responsive element ()228 to )1 bp, denoted as Fragment 228) could be induced by TPA The maximal TPA-induced binding activity ( 3.0fold) for Snail was observed at h, whereas induction of EGR-1 ( 4.0-fold) was earlier at h, sustained until h and thereafter decreased SP-1 exhibited significant basal activity, which may be elevated by 2.6-, 3.5- and 2.8-fold by treatment with TPA for 2, and 12 h, respectively The irrelevant Raf antibody, employed as mock, did not precipitate Fragment 228 at all It is worth noting that all three transcriptional factors exhibited the maximal in vivo DNA-binding activity at h, which was also the time of maximal in vitro DNA-binding activity observed in EMSA FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS 1209 Transcriptional activation of p15INK4b C.-T Hu et al (Fig 6C) This implied that h is the most critical time point for TPA-induced binding of the transcriptional factors on the p15INK4b promoter Double ChIP assay for TPA-induced interaction of the critical transcriptional factors on the promoter region Using both ChIP assay (Fig 8A) and immunoprecipitation/western analysis (Fig 7) we found that TPA may induce both interaction of EGR-1 with Snail or SP-1 with TPA-responsive element of the p15INK4b promoter, and also protein–protein association of EGR-1 with Snail or SP-1 Thus, it is intriguing to examine whether these proteins interact with each other on the TPA-responsive element To address the issue, a double ChIP assay using Fragment 228 was performed As shown in Fig 8B (left), after the first and second immunoprecipitation by antibodies of EGR-1 and Snail, respectively, significant levels of Fragment 228 can be detected in chromatin from cells treated with TPA for h, and this further increased by 2.5-fold at h In the reverse double ChIP using Snail and EGR-1 antibody for the first and second immunoprecipitations, respectively, a similar pattern of TPAinduced binding with Fragment 228 was observed Also, after the first and second immunoprecipitation by antibodies of EGR-1 and SP-1, respectively, significant levels of Fragment 228 could be detected at h and this further increased by 3.0-fold at h In the reverse double ChIP using SP-1 and EGR-1 antibody for the first and second immunoprecipitations, respectively, a basal level of Fragment 228 could be detected, which increased slightly at h and greatly (by 5.0-fold) at h No PCR product of Fragment 228 could be detected if mock antibody was used in the first immunoprecipitation in either experimental group This result confirmed that TPA may induce the association of EGR-1 with Snail or SP-1 on the TPA-responsive element of the p15INK4b promoter A proposed Snail target site involved in TPA-induced p15–pro228 Because there is no putative binding region of Snail such as the E-box on the TPA-responsive element, whether Snail binds on an unidentified region around the EGR-1/SP-1 overlapping site for activation of the p15INK4b promoter is an intriguing issue to be explored Using genomatix software, there is a 5-bp consensus sequence motif (TCACA) upstream of the EGR-1/SP-1 overlapping site on promoters of p15INK4b (at )207 to )203), which is also found on the 1210 MMP-9 promoter, another Snail-upregulated gene [21– 23] It is tempting to speculate that the sequence around this motif is the potential Snail target site for p15INK4b promoter activation (see Discussion) To investigate whether the proposed Snail target region was required for TPA-induced p15INK4b promoter activation, a p15–pro228 mutant denoted as p15–pro228SN* with three nucleotides altered in this region (CAC fi GTG at )206 to )204) (Fig 5B, left) was employed As shown in Fig 5B (right), TPAinduced promoter activity of p15–pro228SN* decreased by 45% compared with that of wild-type p15–pro228 Thus, the proposed Snail target region was involved in TPA-induced p15–pro228 activation Binding of Snail with the proposed Snail target site To examine whether the proposed Snail target region can be bound by Snail, we performed EMSA using a probe denoted as p15–proSN ()218 to )197 bp) which contains this region (Fig 6A) As shown in Fig 9A, two of the mobility-shifted bands (SNI and SNII) increased significantly by 1.5–2.0-fold in EMSA using nuclear extract from HepG2 treated with TPA for h compared with that from untreated HepG2 Both bands further increased by 6.0- to 8.0-fold at h and decreased at h Another band, SNIII, had a rather abundant basal level, increased by 2.5- and 5.0-fold at and h, respectively, followed by a decrease at h In the competition group, SNII and SNIII were totally abolished at and h by the addition of 200-fold unlabeled p15–proSN, whereas SNI was not suppressed at h We further examined whether alteration in the 5-bp consensus sequence motif (TCACA) of the proposed Snail target region may influence the pattern of EMSA As shown in Fig 9B, the TPAinduced elevation of SNII and SNI at both and h decreased by 55–65% in the EMSA using p15–proSN mutant as the probe (p15–proSN* with CAC fi GTG at )206 to )204), compared with that using wild-type probe (compare lanes and with lanes and 8) In addition, SNIII decreased dramatically in all the samples using p15–proSN* as the probe (compare lanes and with lanes and 8) We further examined whether Snail protein could be contained within the band shifts As shown in Fig 9C, SNII and SNIII were suppressed by 95 and 65% at h if the EMSA reaction mix was preincubated with Snail antibody, but not Raf antibody (the control antibody), for 30 (compare lanes and 8) At h, the blocking effect of the Snail antibody was less prominent because the amount of both complexes had already FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS Transcriptional activation of p15INK4b C.-T Hu et al A Fig EMSA of the proposed Snail target site Nuclear extracts of untreated HepG2 (lane in A, C and lanes and in B), HepG2 treated with 50 nM tetradecanoyl phorbol acetate (TPA) for 1, or h (lanes 3–7 in A and 3–9 in C, as indicated) or and h (lanes 3–4 and 7–8 in B), and HepG2 transfected with Snail overexpressing plasmid or pcDNA3 vector (C) were incubated with p15–proSN wild type probe (all lanes in A and C, lanes 1–4 in B and lane 1–2 in D) or p15–proSN* mutant probe (lanes 5–8 in B and lanes 3–4 in D) followed by EMSA For competition analysis (lanes 6–7 in A), unlabeled wild-type p15–proSN was included in the EMSA using a sample of HepG2 treated with TPA for and h For antibody blocking analysis (lanes 6–9 in C), nuclear extracts from HepG2 treated with TPA for and h were preincubated with antibodies (2 lg each) against Snail or Raf (as the negative control antibody) followed by EMSA ‘Probe only’ in (A), (B) and (C) represents the sample without nuclear extract The results are representative of 2–3 reproducible experiments B C D decreased at this time (compare lanes and 9) By contrast, the amount of SNI was not significantly influenced by Snail antibody at any time To further validate the specificity of band shifts with regard to Snail, HepG2 was transiently transfected with a Snailexpressing plasmid for 36 h, followed by EMSA The Snail mRNA in the Snail-transfected cell increased by  16.0-fold, as detected by real-time RT/PCR (Fig S2) Interestingly, SNI and SNII (but not SNIII) increased by 3.0–3.5-fold in the EMSA using nuclear extract from HepG2 transfected with Snail, compared with the cell transfected with pcDNA3 vector (Fig 9D, compare lanes and 2) Moreover, the amount of SNII (but not SNI) in EMSA for HepG2 overexpressing Snail decreased dramatically (by 90%) if p15–proSN* was used as the probe instead of wildtype p15–proSN (Fig 9D, compare lanes 1,2 with 3,4) Taken together, it appeared that in EMSA for either HepG2 treated with TPA or Snail overexpressing HepG2, SNII is the most specific DNA–protein complex which may contain the proposed Snail target fragment bound by Snail Further, a ChIP assay was performed to investigate whether Snail may bind to the proposed target region in vivo The target DNA was an 84-bp promoter fragment ()200 to )284 bp) denoted as p15–proSN-ChIP, which contains the proposed Snail target region upstream of the EGR-1/SP-1 overlapping site (Fig 10A) As shown in Fig 10B, slight basal binding activity of Snail toward p15–proSN-ChIP was observed in untreated HepG2, which was further increased in HepG2 treated with TPA for and h by  4.5-fold compared with the basal level As a negative control, the binding of Raf with p15–proSN-ChIP was not increased in TPA-treated HepG2 Snail may enhance basal and TPA-induced p15–pro228 activation Thus far, we have found that Snail is not only associated with EGR-1 and SP-1 on the EGR-1/SP-1-overlapping region (Figs and 8B), but is also capable of binding to the proposed Snail target site (Figs and 10) Both regions were required for TPA-induced p15INK4b promoter activation (Fig 5B) In addition, we have previously shown that in HepG2 stably overexpressing Snail, the promoter activity of p15INK4b was higher than in the parental cell [17] Thus FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS 1211 Transcriptional activation of p15INK4b C.-T Hu et al A B Fig 10 Chromatin immunoprecipitaion (ChIP) of the proposed Snail binding region (A) Scheme indicating the position of the promoter fragment (p15–proSN-ChIP) for ChIP, which is located between )200 and )284 bp containing the Snail target region upstream of EGR-1/SP-1 (B) HepG2 cells were untreated or treated with 50 nM tetradecanoyl phorbol acetate (TPA) for and h, ChIP assay for binding of Snail on promoter fragment p15–proSN-ChIP were performed Raf antibody was used as the negative control antibody Acetylated (Lys9/14) histone H3 antibody was used to precipitate the promoter region of GAPDH as internal control group The PCR product of the target promoter fragment from chromosome of each sample is shown as the Input The results are representative of three reproducible experiments Ab, antibody; IP, immunoprecipitation it appears that Snail may act as a transcriptional activator in TPA-induced p15INK4b gene expression Because Snail is conventionally known to be a major transcriptional repressor, this needs to be investigated further Therefore, we examined whether overexpression of Snail may influence p15–pro228 promoter activation To our surprise, after transient transfection of HepG2 with Snail-expressing plasmid for 36 h, the TPA-induced p15–pro228 promoter activity increased by 2.1-fold, and the basal p15–pro228 promoter was elevated by 3.2-fold compared with that transfected with pcDNA3 vector in each group (Fig 11) This result suggested that Snail may both enhance the TPAinduced p15INK4b promoter activity (which might be achieved via cooperation of EGR-1 and SP-1), and also activate the p15INK4b promoter by itself Discussion Previously, the mechanisms by which Snail upregulates gene expression were not explored in as much depth as those by which Snail downregulates gene 1212 Fig 11 Effects of Snail on activation of p15–pro228 in the presence or absence of tetradecanoyl phorbol acetate (TPA) HepG2 cells were transfected with Snail expressing plasmid or control pcDNA3 vector followed by treatment with TPA (solid bar) for 12 h or were left untreated (hollow bar) Dual luciferase assays were performed and the relative ratio of Firefly/Renilla luciferase activity of each sample was calculated, taking the data for HepG2 transfected with pcDNA3 and without TPA treatment as 1.0 The results are the average of 5–7 experiments with a C.V of 5.0–7.0 Statistical significance at **P < 0.005 between the results of HepG2 transfected with Snail and pcDNA3 vector in either the presence or absence of TPA expression Our study demonstrated the first time that Snail cooperates with other transcriptional factors such as EGR-1 and SP-1 to upregulate gene expression using TPA-induced p15INK4b as a model This molecular event occurred in a time-dependent manner, in which SP-1, EGR-1 and Snail may be recruited successively to their target regions on the TPA-responsive element and interact with each other The time course for TPA-induced gene expression of these transcriptional factors correlated well with the timing of protein–protein and protein–DNA interactions Interaction of Snail with EGR-1 Previously, EGR-1 has been found to be involved in TPA-induced gene expression of tyrosine hydroxylase in PC12 cells [30] and MDR1 in hematopoietic cell K562 [31] Our results further demonstrated that EGR-1 acted with Snail for TPA-induced promoter activation of p15INK4b It is worth noting that the TPA-induced gene expression of EGR-1 was earlier than that of Snail (Fig 3A,C), which was also observed during hepatocyte growth factor-induced cell migration of HepG2 [9] Consistent with this, ChIP assays demonstrated that TPA-induced binding of EGR-1 with the TPA-responsive element Fragment 228 was earlier at h and was sustained until h, whereas that of Snail was induced at FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS Transcriptional activation of p15INK4b C.-T Hu et al h This is in accord with the result of both immunoprecipitation/western analysis (Fig 7) and double ChIP assays (Fig 8B) demonstrating that TPA-induced maximal association of EGR-1with Snail occurred at h One more interesting issue to be addressed is the regulation of Snail and EGR-1 gene expression by each other Previously, Grotegut et al [9] demonstrated that EGR-1 was required for hepatocyte growth factor-induced transcription of Snail According to the time course experiment (Fig 3A,C), TPA-induced maximal expression of EGR-1 mRNA and protein occurred earlier than that of Snail, and it is possible that induction of Snail may be mediated by EGR-1 As expected, quantitative real-time RT/PCR analysis showed that TPA-induced Snail mRNA expression at h can be suppressed by EGR-1 shRNA (fragment E33) by 68% compared with TPAinduced Snail mRNA expression in cells transfected with mock (Lamin) shRNA (Fig S3A, upper left) To our surprise, Snail shRNA (fragment S18) can also suppress TPA-induced EGR mRNA expression at the h time point by 42% (Fig S3A, lower right) These results demonstrated that EGR-1 gene expression induced by TPA at the earlier stage was required to trigger gene expression of Snail at h, which was then used to sustain TPAinduced EGR-1 mRNA expression at later period (i.e h) This implied that both transcriptional factors may be upregulated by each other in a positive feedback manner However, a negative feedback effect of Snail on EGR-1 was also suggested by Grotegut et al., which contradicts our observation This may be because the experimental systems used in the studies were different In Grotegut’s [9] report, an inducible system was used to overexpress exogenous Snail, which may prevent hepatocyte growth factor-induced EGR-1 By contrast, we used a shRNA technique to block the TPA-induced endogenous Snail, resulting in the prevention of TPA-induced EGR-1 This issue was further addressed by transfection of the cells with exogenous Snail, followed by examining whether TPA-induced EGR-1 mRNA is influenced Overexpression of Snail was verified by real-time PCR in which Snail mRNA in cells transfected with Snail plasmid increased by 20.0- and 10.0-fold at and h, respectively, compared with that in cells transfected with pcDNA3 vector (Fig S3B, upper) Interestingly, overexpression of Snail in HepG2 did not significantly suppress TPA-induced EGR-1 at h, although it greatly suppressed TPA-induced EGR-1 at h (by 80%) (Fig S3B, lower) This indicated that at high concentrations Snail may prevent TPA-induced EGR-1 expression at a later stage, consistent with that observed in Grotegut’s study Taken together, it can be suggested that Snail has dual effects on EGR-1 expression, i.e stimulatory and suppressive at low and high concentrations, respectively The detailed mechanism for this phenomenon is worthy of further investigation Interaction of EGR-1 with SP-1 The involvement of SP-1 in the transcriptional upregulation of p15INK4b was investigated decades ago Early reports demonstrated that a SP-1 consensus site was required for transforming growth factor b-induced p15INK4b promoter activation [32] Furthermore, the cooperation of SP-1 with Smad2, Smad3 and Smad4 was required for this transcriptional event [33] In our results, the role of SP-1 in TPA-induced p15INK4b promoter activation was also essential, as evidenced by basal and TPA-induced SP-1 gene expression (Fig 3A), SP-1 binding on Fragment 228 (Fig 8A) and the association of SP-1 with EGR-1 (Fig 7) Furthermore, the cis-acting element of SP-1 for TPAinduced p15INK4b promoter activation was also clarified in this study There are two SP-1-binding motifs on the TPA-responsive element, one overlaps with that of EGR-1 at )202 to )186 bp, the other is the adjacent single SP-1 site at )183 to )169 bp (Table S1) Strikingly, deletion of or point mutation in the EGR1/SP-1 overlapping site, but not the single SP-1 site, abolished TPA-induced activation of p15–pro228 (Fig 5A,B) Therefore, the possibility that SP-1 (together with EGR-1) may bind on the overlapping EGR-1/SP-1 region rather than on the adjacent single SP-1 was much higher Interestingly, simultaneous binding of EGR-1 and SP-1 to adjacent DNA-binding sites was also shown to be required for maximal activity of the interleukin-2 receptor b-chain promoter suggesting synergistic functions of SP-1 and EGR-1 [34] The time course for the interaction of SP-1 with EGR1 is another issue to be addressed Notably, SP-1, but not EGR-1, exhibited a basal level of gene expression and DNA-binding activity in untreated HepG2, as shown in Figs 3B and 8A Furthermore, the double ChIP assay (Fig 8B) indicated that association of the SP-1/EGR-1 immunocomplex with the TPA-responsive element was observed after TPA treatment for 2–6 h It may be proposed that in untreated HepG2, SP-1 may constitutively bind on the overlapping EGR-1/ SP-1 region in p15INK4b promoter After TPA treatment for h, EGR-1 may be recruited to its target site on the same regions to interact with SP-1 Potential Snail binding region on TPA-responsive element In our previous report [17], the critical role of Snail in transcriptional regulation of the p15INK4b promoter was FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS 1213 Transcriptional activation of p15INK4b C.-T Hu et al studied using EMSA The promoter activity of p15– pro228 and in vitro DNA-binding activity toward the TPA-responsive element were enhanced in HepG2 overexpressing Snail (S7) compared with that of the parental HepG2 Moreover, in the EMSA result (Fig 6C), we found that Snail antibody can suppress TPA-induced mobility shift of p15–proES containing the EGR-1/SP-1 binding region at the h time point However, whether there are Snail-specific target regions in the p15INK4b promoter has not been clarified Investigation of this issue can be initiated by comparing the promoter sequence of p15INK4b with that of other genes known to be upregulated by Snail Recently, we found that gene expression of MMP-9 can also be induced by TPA in an EGR-1-, SP-1- and Snail-dependent manner (unpublished results), implying that both p15INK4b and MMP-9 can be upregulated by similar transcriptional apparatus Furthermore, there is a 5-bp consensus sequence motif (TCACA) upstream of the EGR-1/SP-1-binding site on both the p15INK4b and MMP-9 promoter Interestingly, alteration of three nucleotides in this region partially suppress (by  45%) TPA-induced activation of p15– pro228 (Fig 5B) Moreover, using EMSA with a probe containing this proposed Snail-binding region (p15– proSN), we demonstrated that Snail antibody may block the formation of DNA–protein complexes (especially SNII) in the TPA-treated sample (Fig 9C) On the other hand, overexpression of Snail also increased the formation of SNII (Fig 9D) Furthermore, using a ChIP assay we demonstrated that after treatment with TPA, the binding of Snail with its proposed target region greatly increased in HepG2 (Fig 10) Taken together, we suggest that Snail protein may bind on the region upstream of the EGR-1/SP-1-binding site for TPA-induced activation of the p15INK4b promoter Finally, we demonstrated that overexpression of Snail not only enhanced the TPA-induced p15–pro228 promoter activity, but also elevated the basal promoter activity (Fig 11) It is possible that the Snail-induced basal p15INK4b promoter activation might be achieved by Snail itself via binding to the proposed Snail target region Alternatively, some other transcriptional transcriptional factors in addition to Snail may be required and other unknown Snail-binding regions might be involved in this process This interesting issue is worthy of further investigation We propose the model for TPA-induced sequential binding of SP-1, EGR-1 and Snail on the TPA-responsive element of p15INK4b promoter in a time-dependent manner as follows (Fig 12) Before TPA treatment, SP-1 is constitutively bound on the EGR-1/SP-1-overlapping region After 1–2 h of TPA treatment, EGR-1 is recruited to the same region and associates with 1214 Fig 12 Proposed model for tetradecanoyl phorbol acetate (TPA)-induced sequential binding of SP-1, EGR-1 and Snail on the TPA-responsive element in p15INK4b promoter (Upper) At time zero, SP-1 may be constitutively bound on the EGR-1/SP-1-overlapping region (red) (Middle) After treatment with TPA for 1–2 h, EGR-1 is recruited to the constitutively bound SP-1 (Lower) At 4–6 h, Snail may associate simultaneously with the preformed EGR-1/SP-1 complex via EGR-1 and the sequence upstream of EGR-1/SP-1-overlapping region The adjacent signal SP-1 site and the proposed Snailbinding region are given in blue and black, respectively The proposed binding motif (TCACA)of Snail is marked as italic SP-1 Subsequently (within 4–6 h of TPA treatment), Snail is also recruited Finally, Snail may bind with both EGR-1 and the region (TCACA) upstream of the EGR-1/SP-1 overlapping site simultaneously Thus a tri-component transcriptional complex may be formed to trigger p15INK4b promoter activation Materials and methods Cell culture, chemicals and plasmid The cultured condition for HepG2 cells was as described in our previous report [17] Tetradecanoyl phorbol acetate was purchased from Sigma-Aldrich (Poole, UK) Plasmid with a full-length Snail coding sequence under the CMV promoter was a gift from C C Chang (Tzu Chi University, Hualien, Taiwan) Construction of deletion mutants of the p15INK4b promoter The reporter plasmids p15–profull, p15–pro461 and p15– pro228 were as used in the previous report [17] Plasmid p15–pro233 was derived from p15–pro461 by double digestion with SmaI/KpnI followed by ligation of the digested small fragment into pGL3 Plasmids p15–pro77, p15– pro228DE/S-S, p15–pro228DE/S and p15–pro228DS were derived from p15–pro228 by double digestion with MluI/ XhoI, KpnI/PvuII, SauI/NheI and SauI/PvuII, respectively, followed by filling in the restriction site overhangs by Klenow enzyme Subsequently, the digested DNA fragments FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS Transcriptional activation of p15INK4b C.-T Hu et al were self-ligated The restriction enzyme map of the TPAresponsive element within p15–pro228 is shown in Fig S4 Site-directed mutagenesis on TPA-responsive element of p15INK4b promoter The reporter plasmid p15–pro228 was used as a template for site-directed mutagenesis using a GeneEditorÔ in vitro site-directed mutagenesis system (Promega, Madison, WI, USA) The sequences of the oligonucleotides for mutagenesis were 5¢-ATCACAGCGGACAGGTATCGGAGCCTAA GGGGG-3¢ (on the EGR-1/SP-1-overlapping site); 5¢-GG GCGGAGCCTAAGGGGGGACGGAGACGCCGGCCCC-3¢ (on the adjacent SP-1-binding site); 5¢-TGGCCTC CCGGCGATGTGAGCGGACAGGGG (on the proposed Snail-binding site) Underlined letters indicate the changed nucleotides The mutagenesis reaction began with annealing of the selection oligonucleotides and the mutagenic oligonucleotides to the DNA template, followed by synthesis of the mutant strand with T4 DNA polymerase and T4 DNA ligase The heteroduplex DNA was then transformed into the repair minus Escherichia coli strain BMH71-18 mutS, and the cells were grown in selective media to obtain clones containing the mutant plasmid Plasmids resistant to the novel GeneEditorÔ Antibiotic Selection Mix were then isolated and transformed into the final host strain, TOP10, using the same selection conditions Promoter activity assay Various truncated formats of pGL3-based reporter plasmids of p15INK4b promoter were transfected with or without various shRNAs into HepG2 cells using the effectene transfection system (Invitrogen Ltd, Renfrew, UK) After the required treatments, cells were harvested and assayed for dual luciferase activity, as described in the manual protocol (Promega, Madison, WI, USA) The pRL vector expressing Renilla luciferase activity was co-transfected with experimental reporter plasmid as an internal control The normalized promoter activity was calculated as: Firefly luciferase activity divided by Renilla luciferase activity In general, the activity of the control vector pRL in each sample was stable and not significantly influenced by TPA The raw data of one dual luciferase assay are given in Table S2 The coefficient of variation of pRL activity of each sample was within 2.0–9.0% by statistical analysis of triplicate data (n = 3) Nonradioactive EMSA A DNA probe was biotinylated using a Biotin 3¢-end DNA labeling kit (Pierce, Rockford, IL, USA) The LightShift Chemiluminescent EMSA Kit (Pierce) was used for the DNA–protein binding reaction, as described in the manual Subsequently, the binding reaction mixture was separated by nondenaturing gel electrophoresis using 6.0–8.0% acrylamide The gel was transferred to a nitrocellulose membrane followed by UV cross-linking Membranes were then incubated with streptavidin–horseradish peroxidase coupled with Luminol substrate for chemiluminescent detection of the protein– DNA complex To verify the transcriptional factors bound on DNA, antibodies raised against zinc-finger motif (amino acids 602–610) of SP-1, the C-terminus of EGR-1 or the middle region (amino acids 21–150) of Snail (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was incubated separately with nuclear extracts prior to EMSA reaction ChIP assay A ChIP assay was performed using a modification of the standard protocol Briefly, after cross-linking chromosome with the bound proteins, immunoprecipitation of the transcriptional factors was performed using polyclonal antibodies against Snail, EGR-1 and SP-1 Raf-1 is used as a mock antibody As a positive control, antibody of acetylated (Lys9/14) histone H3 (Ac-histone H3) (Santa Cruz Biotechnology) was used to precipitate the histone-bound GAPDH promoter The recovered DNA was analyzed by PCR with primers flanking the putative transcriptional factor binding sites as indicated For quantitation, the gels were scanned and the intensity of each PCR band was estimated with geldigitizing software un-scan-it gel v 5.1 Primers used for the p15INK4b promoter (Fragment 228), p15–proSN-ChIP and GAPDH promoter fragment were designed using primer software as follows: Fragment 228, 230 bp: (F)5¢-TACTAGCGGTTTTACGGGCG-3¢, (R)5¢-TCGAA CAGGAGGAGCAGAGAGCGA-3¢; p15–proSN-ChIP, 84 bp: (F)5¢-ATGCGTCCTAGCATCTTTGG, (R)5¢-TGT CCGCTGTGTCGCCG; GAPDH promoter fragment, 150 bp: (F)5¢-TAGCCGGGCCTGGCCTCC-3¢, (R)5¢-GG ATTGCTTCTGGGAAA-3¢ Quantitative RT-PCR Real-time PCR of EGR-1 and Snail were performed by QuantiTect SYBR PCR kit (Qiagen, Crawley, UK) using ABI 7300 real-time PCR system (Applied Biosystems, Foster City, CA, USA) The primer sequences used were Snail: 5¢-TCCAGAGTTTACCTTCCAGCA-3¢ (forward) and 5¢-CTTTCCCACTGTCCTCATCTG-3¢ (reverse); EGR-1: 5¢-CTCTTAGGTCAGATGGAGGTTCTC-3¢ (forward) and 5¢-GGCATATGATGGAGATGATACTGA-3¢ (reverse) SP-1: 5¢-TAGGAGAGATTGGGAGAAATCATC-3¢ (forward) and 5¢-AAGATACCAGAAGGTCGAGAGA GA-3¢ (reverse) GAPDH was included as internal control to correct for equal RNA amounts HotStar Taq DNA polymerase was used for primer extension PCR mixtures were preincubated at 95 °C for 15 to activate the polymerase Each of the 40 PCR cycles consisted of 16 s of FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS 1215 Transcriptional activation of p15INK4b C.-T Hu et al denaturation at 94 °C, annealing of primers for 30 s at 55 °C and 15 s of extension at 72 °C The relative amounts of mRNA were calculated using 7300 system sds Software Western blot Western blot was performed as described in our previous report [17] Briefly, the cells were lyzed and equal amount of proteins in each sample were separated by 12% SDS/ PAGE and transferred to a poly(vinylidene difluoride) membrane Membranes were blocked in 5% dry milk and probed with antibodies against molecules of interest Following incubation with alkaline phosphatase-conjugated secondary antibody, proteins were visualized with NBT/ BCIP for color development For quantitation, the intensity of each specific band was estimated with gel digitizing software un-scan-it gel v 5.1 Immunoprecipitation/western blot Immunoprecipitaion was performed by standard method Briefly, 800 lg of protein in cell lysate was incubated with antibodies bound on protein A–Sepharose beads on roller at °C for h After extensive washing, proteins were eluted from the beads with loading buffer followed by western blot of the indicated proteins shRNA technology Lentiviral plasmids each encoding shRNA targeting different regions of the indicated mRNA were obtained from RNAi Core Laboratory (Academia Sinica, Taiwan) Cells at 60% confluence were transfected with 0.26 ngỈlL)1 shRNA either alone or in combination, using Effectene transfection reagent (Invitrogen) according to the manufacturer’s protocol Lentiviral plasmid encoding human Lamin A shRNA was used as mock shRNA The shRNA fragments targeting different regions of Snail, EGR-1 and SP-1 mRNA are given in Table S3 Statistical analysis Data were analyzed using Student’s t-test in excel All the quantitative studies were performed at least in triplicate, as appropriate Differences were considered significant at P < 0.05 Statistical significance between groups is indicated by *P < 0.05 and **P < 0.005 Acknowledgements We thank National Science Council in Taiwan and Research Centre of Hepatology in Buddhist Tzu Chi General Hospital for financial support, and 1216 Ms Sindy Wang and Mr Joseph Long for technical assistance References Boulay JL, Dennefeld C & Alberga A (1987) The Drosophila developmental gene snail encodes a protein with nucleic acid binding fingers Nature 330, 395–398 Alberga A, Boulay JL, Kempe E, Dennefeld C & Haenlin M (1991) The snail gene required for mesoderm formation in Drosophila is expressed dynamically in derivatives of all three germ layers Development 111, 983–992 Yamashita S, Miyagi C, Fukada T, Kagara N, Che YS & Hirano T (2004) Zinc transporter LIVI controls epithelial–mesenchymal transition in zebrafish gastrula organizer Nature 429, 298–302 Fritzenwanker JH, Saina M & Technau U (2004) Analysis of forkhead and snail expression reveals epithelial–mesenchymal transitions during embryonic and larval development of Nematostella vectensis Dev Biol 275, 389–402 Blanco MJ, Moreno-Bueno G, Sarrio D, Locascio A, Cano A, Palacios J & Nieto MA (2002) Correlation of Snail expression with histological grade and lymph node status in breast carcinomas Oncogene 21, 3241– 3246 Rosivatz E, Becker I, Specht K, Fricke E, Luber B, Busch R, Hofler H & Becker KF (2002) Differential expression of the epithelial–mesenchymal transition regulators snail, SIP1, and twist in gastric cancer Am J Pathol 161, 1881–1891 Olmeda D, Jorda M, Peinado H, Fabra A & Cano A (2007) Snail silencing effectively suppresses tumour growth and invasiveness Oncogene 26, 1862– 1874 Olmeda D, Montes A, Moreno-Bueno G, Flores JM, Portillo F & Cano A (2008) Snai1 and Snai2 collaborate on tumor growth and metastasis properties of mouse skin carcinoma cell lines Oncogene 27, 4690– 4701 Grotegut S, von Schweinitz D, Christofori G & Lehembre F (2006) Hepatocyte growth factor induces cell scattering through MAPK/Egr-1-mediated upregulation of Snail EMBO J 25, 3534–3545 10 Medici D, Hay ED & Olsen BR (2008) Snail and Slug promote epithelial–mesenchymal transition through beta-catenin–T-cell factor-4-dependent expression of transforming growth factor-beta3 Mol Biol Cell 19, 4875–4887 11 Kokudo T, Suzuki Y, Yoshimatsu Y, Yamazaki T, Watabe T & Miyazono K (2008) Snail is required for TGFbeta-induced endothelial–mesenchymal transition FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS Transcriptional activation of p15INK4b C.-T Hu et al 12 13 14 15 16 17 18 19 20 21 22 23 of embryonic stem cell-derived endothelial cells J Cell Sci 121, 3317–3324 Stemmer V, de Craene B, Berx G & Behrens J (2008) Snail promotes Wnt target gene expression and interacts with beta-catenin Oncogene 27, 5075–5080 Vega S, Morales AV, Ocana OH, Valdes F, Fabregat I & Nieto MA (2004) Snail blocks the cell cycle and confers resistance to cell death Genes Dev 18, 1131– 1143 Valdes F, Alvarez AM, Locascio A, Vega S, Herrera B, Fernandez M, Benito M, Nieto MA & Fabregat I (2002) The epithelial mesenchymal transition confers resistance to the apoptotic effects of transforming growth factor Beta in fetal rat hepatocytes Mol Cancer Res 1, 68–78 Peinado H, Quintanilla M & Cano A (2003) Transforming growth factor beta-1 induces snail transcription factor in epithelial cell lines: mechanisms for epithelial mesenchymal transitions J Biol Chem 278, 21113– 21123 Barrallo-Gimeno A & Nieto MA (2005) The Snail genes as inducers of cell movement and survival: implications in development and cancer Development 132, 3151– 3161 Hu CT, Wu JR, Chang TY, Cheng CC & Wu WS (2008) The transcriptional factor Snail simultaneously triggers cell cycle arrest and migration of human hepatoma HepG2 J Biomed Sci 15, 343–355 Cano A, Perez-Moreno MA, Rodrigo I, Locascio A, Blanco MJ, del Barrio MG, Portillo F & Nieto MA (2000) The transcription factor snail controls epithelial– mesenchymal transitions by repressing E-cadherin expression Nat Cell Biol 2, 76–83 Batlle E, Sancho E, Franci C, Dominguez D, Monfar M, Baulida J & Garcia De Herreros A (2000) The transcription factor snail is a repressor of E-cadherin gene expression in epithelial tumour cells Nat Cell Biol 2, 84–89 Whiteman EL, Liu CJ, Fearon ER & Margolis B (2008) The transcription factor snail represses Crumbs3 expression and disrupts apico-basal polarity complexes Oncogene 27, 3875–3889 Sun L, Diamond ME, Ottaviano AJ, Joseph MJ, Ananthanarayan V & Munshi HG (2008) Transforming growth factor-beta promotes matrix metalloproteinase-9-mediated oral cancer invasion through snail expression Mol Cancer Res 6, 10–20 Miyoshi A, Kitajima Y, Kido S, Shimonishi T, Matsuyama S, Kitahara K & Miyazaki K (2005) Snail accelerates cancer invasion by upregulating MMP expression and is associated with poor prognosis of hepatocellular carcinoma Br J Cancer 92, 252–258 Jorda M, Olmeda D, Vinyals A, Valero E, Cubillo E, Llorens A, Cano A & Fabra A (2005) Upregulation of MMP-9 in MDCK epithelial cell line in response to 24 25 26 27 28 29 30 31 32 33 34 expression of the Snail transcription factor J Cell Sci 118, 3371–3385 Haraguchi M, Okubo T, Miyashita Y, Miyamoto Y, Hayashi M, Crotti TN, McHugh KP & Ozawa M (2008) Snail regulates cell–matrix adhesion by regulation of the expression of integrins and basement membrane proteins J Biol Chem 283, 23514– 23523 Herranz N, Pasini D, Diaz VM, Franci C, Gutierrez A, Dave N, Escriva M, Hernandez-Munoz I, Di Croce L, Helin K et al (2008) Polycomb complex is required for E-cadherin repression by the Snail1 transcription factor Mol Cell Biol 28, 4772–4781 Hou Z, Peng H, Ayyanathan K, Yan KP, Langer EM, Longmore GD and Rauscher FJ III (2008) The LIM protein AJUBA recruits protein arginine methyltransferase to mediate SNAIL-dependent transcriptional repression Mol Cell Biol 28, 3198– 3207 Eyries M, Agrapart M, Alonso A & Soubrier F (2002) Phorbol ester induction of angiotensin-converting enzyme transcription is mediated by Egr-1 and AP-1 in human endothelial cells via ERK1/2 pathway Circ Res 91, 899–906 Ahn BH, Park MH, Lee YH & Min S (2007) Phorbol myristate acetate-induced Egr-1 expression is suppressed by phospholipase D isozymes in human glioma cells FEBS Lett 581, 5940–5944 Hewson CA, Edbrooke MR & Johnston SL (2004) PMA induces the MUC5AC respiratory mucin in human bronchial epithelial cells, via PKC, EGF/TGFalpha, Ras/Raf, MEK, ERK and Sp1-dependent mechanisms J Mol Biol 344, 683–695 Papanikolaou NA & Sabban EL (2000) Ability of Egr1 to activate tyrosine hydroxylase transcription in PC12 cells Cross-talk with AP-1 factors J Biol Chem 275, 26683–26689 McCoy C, Smith DE & Cornwell MM (1995) 12-O-tetradecanoylphorbol-13-acetate activation of the MDR1 promoter is mediated by EGR1 Mol Cell Biol 15, 6100–6108 Li JM, Nichols MA, Chandrasekharan S, Xiong Y & Wang XF (1995) Transforming growth factor beta activates the promoter of cyclin-dependent kinase inhibitor p15INK4B through an Sp1 consensus site J Biol Chem 270, 26750–26753 Feng XH, Lin X & Derynck R (2000) Smad2, Smad3 and Smad4 cooperate with Sp1 to induce p15(Ink4B) transcription in response to TGF-beta EMBO J 19, 5178–5193 Lin JX & Leonard WJ (1997) The immediate–early gene product Egr-1 regulates the human interleukin-2 receptor beta-chain promoter through noncanonical Egr and Sp1 binding sites Mol Cell Biol 17, 3714– 3722 FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS 1217 Transcriptional activation of p15INK4b C.-T Hu et al Supporting information The following supplementary material is available: Fig S1 Effective shRNA of EGR and Snail prevented TPA-induced EGR-1 and Snail Fig S2 Overexpression of Snail mRNA in HepG2 transfected with plasmid encoding Snail Fig S3 Quantitative RT/PCR for cross-regulation of EGR-1 and Snail Fig S4 Restriction map on the TPA responsive element Table S1 The putative transcriptional factor binding regions on the TPA-responsive element (between )226 and )80 bp) 1218 Table S2 The raw data of pRL activity Table S3 shRNA targeting sequences of Snail, EGR-1 and SP-1 used for knockdown approach This supplementary material can be found in the online version of this article Please note: As a service to our authors and readers, this journal provides supporting information supplied by the authors Such materials are peerreviewed and may be re-organized for online delivery, but are not copy-edited or typeset Technical support issues arising from supporting information (other than missing files) should be addressed to the authors FEBS Journal 277 (2010) 1202–1218 ª 2010 The Authors Journal compilation ª 2010 FEBS ... expression of EGR-1, Snail and SP-1 in HepG2 HepG2 cells were untreated (con) or treated with 50 nM TPA for 0.5, 1, 2, and h (A and C) Real-time RT/PCR (A) and western blot (C) of EGR-1, SP-1 and Snail. .. interaction of EGR-1 with Snail or SP-1 with TPA-responsive element of the p15INK4b promoter, and also protein–protein association of EGR-1 with Snail or SP-1 Thus, it is intriguing to examine whether... that SP-1 (together with EGR-1) may bind on the overlapping EGR-1/ SP-1 region rather than on the adjacent single SP-1 was much higher Interestingly, simultaneous binding of EGR-1 and SP-1 to adjacent