Sp1 transcriptionally regulates BRK1 expression in non-small cell lung cancer cells

Gene 542 (2014) 134–140

Contents lists available at ScienceDirect

Gene journal homepage: www.elsevier.com/locate/gene

Sp1 transcriptionally regulates BRK1 expression in non-small cell lung cancer cells Meng Li 1, Bing Ling 1, Ting Xiao, Jinjing Tan, Ning An, Naijun Han, Suping Guo, Shujun Cheng, Kaitai Zhang ⁎ State Key Laboratory of Molecular Oncology, Cancer Institute (Hospital), Peking Union Medical College & Chinese Academy of Medical Sciences, Beijing 100021, China

a r t i c l e

i n f o

Article history: Received 25 September 2013 Received in revised form 27 January 2014 Accepted 21 March 2014 Available online 25 March 2014 Keywords: BRK1 NSCLC Promoter Sp1 Transcriptional regulation

a b s t r a c t Following a previous study reporting that BRK1 is upregulated in non-small cell lung cancer (NSCLC), the present study sought to clarify the role of specificity protein 1 (Sp1) in the transcriptional regulation of the BRK1 gene. Therefore, a construct, named F8, consisting of the − 1341 to −1 nt sequence upstream of the start codon of the BRK1 gene inserted into pGL4.26 was made. A series of truncated fragments was then constructed based on F8. Segment S831, which contained the −84 to −1 nt region, displayed the highest transcriptional activity in the A549, H1299 and H520 NSCLC cell lines. Bioinformatic analysis showed a potential Sp1-binding element at −73 to −64 nt, and a mutation in this region suppressed the transcriptional activity of S831. Then the RNAi assays of Sp1 and its coworkers Sp3 and Sp4 were performed, and suppression of Sp1 by siRNA inhibited the mRNA expression of BRK1. Both an electrophoretic mobility shift assay (EMSA) and a chromatin immunoprecipitation (ChIP) assay demonstrated that Sp1 bound to the promoter area of the BRK1 gene. Our data identified a functional and positive Sp1 regulatory element from −73 to −64 nt in the BRK1 promoter, which may likely explain the overexpression of BRK1 in NSCLC. © 2014 Elsevier B.V. All rights reserved.

1. Introduction BRK1 protein, also known as BRICK1, C3orf10 and HSPC300, is a subunit of the WAVE/SCAR complex, which activates the Arp2/3 complex. BRK1 combines with WAVE/SCAR, PIR121/Sra-1, Nap125 and Abi subunits to form the WAVE/SCAR complex (Gautreau et al., 2004), a consensus structure in plants and animals. When activated by the Rac1 signaling pathway, the WAVE complex releases active WAVE-BRK1, leading to the assembly of actin filaments (Eden et al., 2002; Stradal and Scita, 2006). Although little is known about the mechanism of BRK1 function, it plays a key role in cellular processes that rely on actin filaments, such as cell attachment, stretching, endocytosis, division and mobility. Cell migration is an essential step in the metastasis of various tumor types and is mediated by the dynamic polymerization of actin filaments (Vasioukhin et al., 2000). In our previous study on NSCLC, gene expression profile data showed that BRK1 is more highly expressed in tumor tissues than in adjacent normal tissues (Liu et al., 2007; Sun et al., 2004). Immunohistochemistry (IHC) performed on lung squamous cell Abbreviations: NSCLC, non-small cell lung cancer; EMSA, electrophoretic mobility shift assay; ChIP, chromatin immunoprecipitation; IHC, immunohistochemistry; RLU, relative luciferase units; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; PVDF, polyvinylidene difluoride; ECL, enhanced chemiluminescence. ⁎ Corresponding author at: 17 Panjiayuan Nanli, Chaoyang District, Beijing 100021, China. E-mail address: [email protected] (K. Zhang). 1 The first two authors contributed equally to this paper.

http://dx.doi.org/10.1016/j.gene.2014.03.043 0378-1119/© 2014 Elsevier B.V. All rights reserved.

carcinomas showed that the high level of BRK1 in tumor tissue is associated with lymph node metastasis, pathological grade and poor differentiation (Cai et al., 2009). Silencing of the BRK1 gene in a NSCLC cell line causes the reorganization of actin filaments, inhibits the formation of pseudopodia and blocks the migration of cells (Cai et al., 2009). The association between BRK1 expression and NSCLC metastasis prompted us to investigate the exact mechanism of BRK1 upregulation. Specificity protein 1 (Sp1) is a transcription factor that is ubiquitously expressed in various cells and tissues. Sp1 recognizes GC-rich regions and binds to DNA through three C2H2-type zinc fingers in the C-terminal domain (Kadonaga et al., 1988; Suske, 1999; Philipsen and Suske, 1999). Each zinc finger of Sp1 recognizes three bases in one strand and a single base in the complementary strand, constituting a consensus binding sequence of 5′-(G/T)GGGCGG(G/A)(G/A)(C/T)-3′ (Narayan et al., 1997; Pavletich and Pabo, 1991). Sp protein family regulates the expression of genes required for cell growth, apoptosis and angiogenesis in cancer (Black et al., 2001; Bouwman and Philipsen, 2002; Safe and Abdelrahim, 2005). The relative expression of Sp1 in cancer cells has been shown to be higher than that of adjacent normal cells in several tumor models, including gastric tumors, breast cancers thyroid tumors and lung cancers (Chiefari et al., 2002; Colon et al., 2011; Wang et al., 2003; Zannetti et al., 2000). Sixty-five percent of lung cancer patients have been shown to have a higher level of Sp1 in tumor tissues (Lin et al., 2010). Additionally, Sp3 and Sp4, two other members of the Sp family, combine with Sp1 to regulate several genes, such as VEGF, VEGFR1, EGFR, PTTG1 and c-MET (Abdelrahim et al., 2004; Abdelrahim et al., 2007; Chintharlapalli et al., 2011; Colon et al., 2011; Pathi et al., 2011), in multiple cancer cell lines.

M. Li et al. / Gene 542 (2014) 134–140

In this study, we found an Sp1-binding site in the promoter region of the BRK1 gene. Both silencing the Sp1 gene and mutating the binding site downregulated BRK1 transcription. An EMSA and a ChIP assay directly confirmed the association between the Sp1 protein and sequence upstream of the BRK1 gene. To the best of our knowledge, this is the first report to describe an association between the Sp1 protein and the high level of BRK1 expression in NSCLC cell lines.

135

according to the manufacturer's protocol (Promega). The promoter activity was presented as relative luciferase units (RLU) as follows: RLU = value of firefly luciferase unit / value of renilla luciferase unit (Guo et al., 2010). All experiments were performed at least three times, and the average RLU of triplicate treatments is shown for each experiment. 2.4. Mutagenesis analysis of the Sp1-binding site

2. Materials and methods 2.1. Cell culture The A549, H1299 and H520 NSCLC cell lines (American Type Culture Collection, Manassas, VA, USA) were used in this study. All cells were maintained in RPMI 1640 medium with 10% fetal calf serum at 37 °C in a 5% CO2 atmosphere. Medium for the H1299 cells was supplemented with 1 mM HEPES, 1 mM glucose, and 1 mM sodium pyruvate.

Using the protocol of the QuikChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA), two mutations, Sp1-mut1 and Sp1-mut2, were generated on the predictive Sp1-binding site of the S831 construct. Double-stranded mutation PCR primers, TGGGTGTGGCTTTTTAGCGCAGG CG and TGTGGCCTGGCTTTTTAGGCGCA, were used to replace the CTGGC and AGCGC sequences, respectively, with TTTTT. The mutated constructs were confirmed by DNA sequencing. 2.5. RNA interference (siRNA)

2.2. Generation of BRK1 promoter-luciferase constructs The −1341 to −1 nt sequence upstream of the ATG start codon of the BRK1 gene was cloned and inserted into pGL4.26 to generate the luciferase-reporter construct named F8. A series of truncated fragments was then constructed from F8. Human male genomic DNA (Novagen, Schwalbach, Germany) was used as the template for PCR amplifications using Phusion DNA polymerase (Promega, Madison, WI, USA). The PCR primers used in this study are listed in Table 1. KpnI (GAGGTACC) and XhoI (GTCGACTCGAG) restriction sites were added to the 5′ end of the forward and reverse primers, respectively. The PCR was performed as follows: 94 °C for 5 min; 30 cycles of 94 °C for 30 s (denaturation), 60 °C for 30 s (annealing), and 72 °C for 90 s (extension); and an elongation step at 72 °C for 10 min. The amplified PCR products were purified with a gel extraction kit (Qiagen, Hilden, Germany) followed by digestion with KpnI and XhoI. The digested DNA fragments were inserted into pGL4.26, a firefly luciferase expression vector (Promega, USA), and constructs were confirmed by DNA sequencing. 2.3. Transient transfection and dual-luciferase assay On the day before transfection, cells were plated in 96-well plates at a density of 10,000 cells/well for A549, 5000 cells/well for H1299, and 7500 cells/well for H520. When the cell culture reached 70% confluence on the next day, the growth medium was replaced with fresh medium. The cells were then transfected using 15 μl Vigofect (Vigorous Biotechnology, Beijing, China), according to the manufacturer's instructions, with 50 ng of the pRL-TK vector and 500 ng of the pGL constructs. Twenty-four hours later, the cells were harvested with passive lysis buffer and frozen at −80 °C for 2 h. Dual-luciferase activities were measured using a GENios Pro Reader (Tecan, Männedorf, Switzerland) Table 1 Primers used to generate BRK1 promoter-luciferase constructs. Sequence

Products

F8

Forward ATAGCCAGGTGTGGTAG Reverse GGCCGCCGCCTGAGG Forward ATTAGCCAGGTGTGGTGACGTGAGC Reverse GGCCGCCGCCTGAGG Forward TAAAAATACAAAAAAATTAGCCGGG Reverse GGCCGCCGCCTGAGG Forward GCCTGGAGCAGTTGAGGGAGACGGC Reverse GGCCGCCGCCTGAGG Forward ATGGGTGTGGCCTGGCAGCGCAGGC Reverse GGCCGCCGCCTGAGG Forward ACGCCGGCGAGGACGTGACGTTGC Reverse GCCTGCGCTGCCAGGCCACACCCAT Forward GCCTGGAGCAGTTGAGGGAGACGGC Reverse GCAACGTCACGTCCTCGCCGGCGT

−1341 to −1 nt

F82 F83 S831 S832 S833

2.6. Reverse transcription quantitative PCR (qRT-PCR) Total RNA was extracted from cells using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and was then reverse-transcribed into cDNA by SuperScript II Reverse Transcriptase (Invitrogen). Real-time PCR was performed using SYBR Premix Ex Taq™ (Takara, Kusatsu, Japan) and the Mx3005 QPCR System (Stratagene). The primers used were as follows: Sp1 forward, ATCCCACAGTTCCAGACCGT; Sp1 reverse, ATGTTGCCTCCACTTCCTCG; Sp3 forward, TGAAGAGTGGCAGCTCAGTG; Sp3 reverse, TGGTACCTCTTCCACCACCT; Sp4 forward, GCGGGATGAG CGATCAGAAG; Sp4 reverse, CAGAGGAGAGGGCTGAGAGT; BRK1 forward, CTGGGCTAACCGGGAGTACA; BRK1 reverse, TTGTCACCCGAGCT TCAATGT; 18S forward, GAAACGGCTACCACATCC; and 18S reverse, ACCAGACTTGCCCTCCA. The real-time PCR assays were performed as follows: 95 °C for 10 s followed by 40 cycles of 95 °C for 15 s (denaturation) and 60 °C for 30 s (annealing and extension). Finally, the dissociation curve was measured for each sample. Relative expression levels of mRNA were calculated against the 18S internal control by the ΔΔCT method. 2.7. Western blotting

Name

F81

ON-TARGET plus SMARTpool Sp1, Sp3 and Sp4 siRNA (Thermo Scientific, Lafayette, CO, USA) were used for gene silencing, and each contained a mixture of 4 siRNAs: GCCAAUAGCUACUCAACUA, GAAG GGAGGCCCAGGUGUA, GGGCAGACCUUUACAACUC and CUACAGAGGC ACAAACGUA for Sp1; GGUAUUCACUCUAGCAGUA, GAAAUUUGUUUG UCCAGAA, GAUAGGAACUGUUAAUACU and GCGAGAUGAUACUUUG AUU for Sp3; GGUAUUCACUCUAGCAGUA, GAAAUUUGUUUGUCCA GAA, GAUAGGAACUGUUAAUACU and GCGAGAUGAUACUUUGAUU for Sp4. The ON-TARGET plus Non-targeting Control Pool (Thermo Scientific) was used as the non-specific control. The cells were transfected with 50 nM siRNA pool using the DharmaFECT2 reagent (Thermo Scientific) for 48 h in accordance with the manufacturer's instructions.

−911 to −1 nt −654 to −1 nt −277 to −1 nt −84 to −1 nt −152 to −60 nt −277 to −128 nt

Total cellular protein was extracted with RIPA lysis buffer (Appygen, Beijing, China) followed by concentration measurement with the BCA Protein Assay Kit (Pierce, Rockford, IL, USA). Protein lysates (30 μg) were run on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels. Proteins were then transferred onto polyvinylidene difluoride (PVDF) membranes using a Trans-Blot SD Semi-Dry Transfer Cell (Bio-Rad, Hercules, CA, USA) at 75 mA for 45 min. The membranes were then blocked in 5% skim milk in PBST containing 10 mM Tris–HCl buffer saline (pH 7.6) plus 0.05% Tween-20. The resulting proteins were detected with primary anti-human Sp1, Sp3, Sp4 antibodies (Abcam, Cambridge, UK) and an anti-human β-actin antibody (Sigma, St. Louis, MO, USA) followed by an HRP-conjugated secondary antibody. The

136

M. Li et al. / Gene 542 (2014) 134–140

blots were visualized by the enhanced chemiluminescence (ECL) system (Pierce). 2.8. Electrophoretic mobility shift assay (EMSA) EMSA was performed with the Non-Radioactive EMSA Kit (Viagene Biotech, Beijing, China) according to the user manual. The doublestranded DNA probe P-BRK1 (5′-TGTGGCCTGGCAGCGCAGGCGC-3′;

−58 to −79 nt) containing the putative Sp1-binding site was synthesized with and without a biotin label. Fifteen micrograms of cell nuclear extracts and various unlabeled competing probes were mixed and incubated at room temperature for 20 min followed by the addition of 15 μl of labeled probes for 20 min. For the supershift assay, samples were incubated with 4 μg of an anti-Sp1 polyclonal antibody (Millipore) for an additional 15 min at room temperature. Protein/DNA complexes were then separated by non-denaturing polyacrylamide gel electrophoresis in

Fig. 1. Identification of the BRK1 gene regulatory promoter. (A) The −1341 to −1 nt sequence and 3 truncations were inserted in a cis-orientation into the pGL4.26 vector to generate luciferase reporter constructs. (B) The four reporters, F8, F81, F82, and F83, were transfected into A549, H1299 and H520 cells, and dual-luciferase assays were performed to assess the promoter activity in each cell line. The results are shown as RLU, and the pGL4.26 vector was used as a negative control. *P b 0.05. (C) The F831 fragment was cut into 3 pieces to generate luciferase reporter constructs. (D) The S831, S832, and S833 reporters were transfected into A549, H1299 and H520 cells and analyzed by a dual-luciferase assay. The results are shown as RLU. *P b 0.05.

M. Li et al. / Gene 542 (2014) 134–140

a 6% gel and transferred onto a binding membrane. Chemiluminescence was detected with streptavidin–HRP using a FluorChem FC3 System (Alpha Innotech, San Leandro, CA, USA). 2.9. Chromatin immunoprecipitation (ChIP) The EZ-Magna ChIP A/G Kit (Millipore) was used for ChIP assays according to the manufacturer's protocol. Approximately 5 × 106 cells were cross-linked with 1% formaldehyde and collected in lysis buffer. DNA was then sheared to an approximate length of 200 to 1000 bp using a Sonics Sonication Instrument (Xinchen, Nanjing, China). Sheared DNA (50 μl) was incubated with 5 μg of anti-Sp1 ChIP-grade antibody or normal rabbit IgG followed by immunoprecipitation with 20 μl of protein A magnetic beads during an overnight incubation at 4 °C with rotation. Enriched DNA was extracted from the DNA/antibody/protein A bead complexes by proteinase K digestion and purified with spin columns. Finally, endpoint PCR was performed to analyze the expected BRK1 sequence using the following BRK1 primers: forward primer, ATTC GTGGGTGCTCAAGAGG (−103 to −84 nt); and reverse primer, GCCC AGTCCTGGTGAATCTC (+31 to +50 nt). Unsheared DNA was used as an input control to analyze the immunoprecipitation results. 2.10. Statistical analysis The results of triplicate luciferase assays were expressed as the mean ± SD. The RLU values of various treatments from each assay were compared by the paired sample t-test with SPSS13.0, and P b 0.05 was considered to be a significant difference. All of the histograms in this article were constructed using GraphPad Prism 5. 3. Results 3.1. Dual-luciferase assay of the BRK1 promoter region The human BRK1 gene is located on chromosome 3p25.3 and is 11,542 bp in length, including 3 exons and 2 introns. The transcriptional start site (TSS) begins with a guanine (G) − 54 nt upstream of the transcriptional ATG start codon. The − 1341 to − 1 nt sequence was cloned from human genomic DNA and inserted into the pGL4.26 luciferase reporter vector, named F8. Several constructs truncated in the 5′ end were generated, including F81 (− 911 to −1 nt), F82 (−654 to

137

−1 nt), and F83 (−277 to − 1 nt) (Fig. 1A). Dual-luciferase assays were performed to evaluate the transcriptional activities of the truncated constructs. The RLU values for F8, F81, F82, F83 and pGL4.26 vectors were as follows: 134.62 ± 3.86, 39.17 ± 8.47, 61.97 ± 17.65, 52.61 ± 4.17, and 0.63 ± 0.24, respectively, in A549 cells; 198.14 ± 40.99, 25.63 ± 6.03, 50.43 ± 5.87, 54.90 ± 6.89, and 0.79 ± 0.14, respectively, in H1299 cells; and 156.71 ± 25.60, 62.19 ± 7.56, 78.97 ± 21.53, 75.82 ± 12.15, and 0.44 ± 0.15, respectively, in H520 cells (Fig. 1B). There were statistically significant differences (P b 0.05) in the 3 cell lines between F8 and F81 as well as between F83 and the vector control, indicating that cis-enhanced elements may exist in the −1341 to −911 nt or − 277 to − 1 nt regions of the promoter. However, the dualluciferase assay of the −1341 to −911 nt sequence showed little transcriptional activity (data not shown). Three truncations of the F83 segment were made, S831 (− 84 to −1 nt), S832 (−152 to −60 nt) and S833 (− 277 to − 128 nt), and only S831 showed obvious luciferase transcriptional activity (Fig. 1C and D). The RLU values of the S831 and pGL4.26 vectors were as follows: 116.74 ± 37.51 and 0.95 ± 0.63, respectively, in A549 cells; 107.82 ± 32.79 and 0.95 ± 0.35, respectively, in H1299 cells; and 163.53 ± 41.22 and 0.44 ± 0.33, respectively, in H520 cells. These results suggested that there may be one or more major regulatory elements in the promoter region located between − 84 and − 1 nt. 3.2. Prediction and mutagenesis analysis of the Sp1-binding sequence To search for potential transcription factor binding sites, we used the TFsearch tool (http://www.cbrc.jp/research/db/TFSEARCH.html) to analyze the sequence from − 84 to − 1 nt inserted in S831. A candidate Sp1-binding site (CTGGCAGCGC) at − 73 to − 64 nt was found with an identity consensus score of 65.8 to the classic Sp1-binding sequence, which is a GC-rich element. We then generated two mutated constructs, namely S831-Sp1-mut1 and S831-Sp1-mut2, which replaced the CTGGC and AGCGC sequences, respectively, in S831 with AAAAA (Fig. 2A). While the RLU value of the S831-Sp1-mut1 showed little difference compared to that of S831, the S831-Sp1-mut2 displayed a much weaker transcriptional activity than S831 (Fig. 2B). The RLU values of the S831 and Sp1-mut2 vectors were as follows: 134.83 ± 38.89 and 4.01 ± 2.67, respectively, in A549 cells; 81.42 ± 22.53 and 5.18 ± 3.08, respectively, in H1299 cells; and 221.94 ± 61.45 and 2.44 ± 0.30, respectively, in H520 cells. Thus, AGCGC (− 69 to − 64 nt) may be a core

Fig. 2. Mutation of the putative Sp1-binding site in S831 leads to a decrease in RLU. (A) Analysis of S831 by TFsearch showed a candidate Sp1-binding site located at −73 to −64 nt. Two mutations were generated to change the −73 to −69 nt and −68 to −64 nt sequences into AAAAA. (B) The S831-Sp1-mut1 had a RLU value similar to that of S831, but the S831-Sp1-mut2 showed little transcriptional activity. *P b 0.05.

138

M. Li et al. / Gene 542 (2014) 134–140

cis-upregulated element of the BRK1 gene, and its corresponding transacting factor(s) may be Sp1, with or without participation of Sp3 and Sp4. 3.3. Suppression of Sp1 inhibits BRK1 transcription To define the role of the Sp protein family in regulating BRK1 transcription, we performed the RNA interference assays of Sp proteins using siRNA. The Sp1 siRNA pools effectively suppressed both Sp1 mRNA and protein expression levels (Fig. 3). As a consequence, the expression of BRK1 was significantly inhibited in the mRNA level, but not in the protein level, and the relative mRNA levels were 0.73 ± 0.03 in A549 cells, 0.59 ± 0.14 in H1299 cells, and 0.71 ± 0.03 in H520 cells (Fig. 3A; P b 0.05). Unfortunately, the Sp3 and Sp4 siRNA pools took effect only in the mRNA level rather than the protein level

for some reason, and BRK1 expression showed little difference in both the mRNA and protein levels. The suppression of Sp1 led to inhibition of BRK1 transcription, thereby indicating that Sp1 promoted the expression of BRK1 gene. 3.4. Identification of an Sp1-binding site in the BRK1 promoter through EMSA and ChIP The consensus binding sequence of Sp1 is 5′-(G/T)GGGCGG(G/A)(G/ A)(C/T)-3′ (Narayan et al., 1997). To confirm the binding activity of the putative Sp1-binding element, we performed an EMSA and a ChIP assay using Sp1 antibody. The P-BRK1 probe used in the EMSA involved the putative Sp1binding site (CTGGCAGCGC) plus 6 bp at both sides. The results of the

Fig. 3. Suppression of Sp1 inhibits BRK1 transcription. Cells were transfected with siRNA targeting Sp1, Sp3 and Sp4, and the expression of Sp proteins and BRK1 was analyzed by qRT-PCR (A) and Western blotting (B) with 18S and β-actin as the internal controls, respectively. The mRNA level of BRK1 decreased after Sp1 was knocked down, while the protein level showed little difference. The results were from three independent experiments. *P b 0.05 and **P b 0.01.

M. Li et al. / Gene 542 (2014) 134–140

EMSA are shown in Fig. 4A. The shift bands showed that one or more nuclear proteins extracted from A549, H1299 and H520 cells combined with P-BRK1 and that this binding activity could be blocked by 50× and 100× unlabeled probes (cold probe). The Sp1 antibody blocked the mobility of the bands (supershift bands), especially in A549 and H1299 cells, demonstrating that Sp1 was involved in the proteins binding to the P-BRK1 probe. The ChIP assay results showed that the Sp1 protein effectively enriched the − 103 to + 50 nt DNA fragment, which covered the putative Sp1-binding element, especially in A549 and H520 cells (Fig. 4B). The ChIP assay confirmed the binding activity between Sp1 and sequence upstream of the BRK1 gene. 4. Discussion Invasion and metastasis of malignant tumors are the main causes of death among cancer patients, and the biological structural basis of these processes is the actin filament, which can mediate formation of pseudopodia and ultimately drive tumor cell migration. Aberrant expression of actin or actin-related genes may contribute to the ability of a tumor cell to acquire metastatic potential (Escobar et al., 2010; Wang et al., 2005). BRK1 protein is a component of the WAVE/SCAR complex that helps to activate the Arp2/3 complex and promote actin nucleation and extension (Takenawa and Suetsugu, 2007). In our previous study, both gene expression profile data and immunohistochemistry of clinical specimens revealed the relationship between high BRK1 expression and NSCLC metastasis using mRNA and protein level measurements, respectively (Cai et al., 2009; Liu et al., 2007; Sun et al., 2004).

139

Additionally, cells that lack BRK1 expression have a defect in lamellipodia formation and show a blebbing phenotype on the membrane (Cai et al., 2009; Derivery et al., 2008). However, the exact mechanism of BRK1 upregulation in tumor cells has been seldom studied. In our study, we chose the A549, H1299 and H520 cell lines to represent lung adenocarcinoma, large cell lung carcinoma and squamous cell lung carcinoma, respectively. Here, we analyzed the sequence 5000 nt upstream of the ATG start codon using TFsearch and MATCH™, but no classic promoter was found. According to this result, the approximate putative promoters were mainly located in a fragment 1.4 kb upstream of the ATG codon. Therefore, we cloned the −1341 to −1 nt sequence in which the core promoter should exist to perform dual-luciferase assays. Among the 4 fragments we evaluated, F8 (− 1341 to − 1 nt) showed the highest RLU value, and its truncations, namely F81 (− 911 to − 1 nt), F82 (− 654 to − 1 nt), and F83 (− 277 to − 1 nt), displayed lower RLU values compared to F8. However, all of these vectors had higher RLU values than the pGL4.26 empty vector (Fig. 1B; P b 0.05). These data strongly indicated that cis-enhanced element(s) existed in the −1341 to − 911 nt and − 277 to − 1 nt regions, but further study excluded the former region (data not shown). When the latter region was cut into 3 pieces, only the S831 segment (− 81 to − 1 nt) had a higher RLU value than the empty vector control, which suggested that the promoter may be located in this region. Many types of tumors have a much higher level of the Sp1 transcription factor than normal cells, and this trend is associated with metastasis progress (Guo et al., 2010; Wang et al., 2003; Zannetti et al., 2000). A previous study has shown that A549 cells express increased Sp1 mRNA

Fig. 4. Identification of the binding activity of Sp1 and the BRK1 promoter. (A) DNA-binding activity was determined by EMSA. The −73 to −64 nt DNA probe bound to the nuclear extracts of the 3 cell lines, and this binding was blocked by an unlabeled probe (shift). An Sp1 antibody was added to detect the specificity of the binding activity (supershift). (B) ChIP was used to confirm the Sp1-binding activity. The Sp1 antibody effectively enriched the DNA sequence covering the putative binding element. Normal rabbit IgG was used as a negative control, and an anti-RNA polymerase antibody was used as a positive control.

140

M. Li et al. / Gene 542 (2014) 134–140

and protein levels as compared to normal bronchial epithelial cells (Chen et al., 2011). Moreover, Sp3 and Sp4 usually combine with Sp1 to regulate gene transcription. Coincidently, we found a putative Sp1binding site at −64 to −73 nt by TFsearch software, and mutation of the proximal 5 bases led to a substantial decrease of RLU, which was representative of a decrease in BRK1 transcription (Fig. 2). Furthermore, in the RNAi assays of Sp proteins, mRNA level of BRK1 decreased by 30– 40% after Sp1 was knocked down (Fig. 3A), which indicated that Sp1 did promote the transcription of BRK1. Nevertheless, probably due to the negative feedback mechanism, the ubiquitination pathway of BRK1 protein was blocked, resulting in the constancy of BRK1 protein level (Fig. 3B). It was difficult to prove the participation of Sp3 and Sp4 here, for the siRNA failed to knock down Sp3 and Sp4 in the protein level. In addition, it was unexpected that silencing of Sp1 led to an obvious increase of Sp4 mRNA (Fig. 3A), suggesting the Sp4 might act as a compensative role of Sp1. The direct binding activity of Sp1 was further confirmed by an EMSA and a ChIP assay. Among the 3 cell lines, H1299 cells (which are p53−/−) displayed the weakest association between Sp1 and the BRK1 promoter (Fig. 4B), which may be due to the absence of p53 in H1299 cells as p53 protein can form a complex with Sp1 to regulate gene expression (Lin et al., 2010). 5. Conclusion In summary, we identified an Sp1 positive regulatory element in the promoter region of the BRK1 gene, which is located at −64 to −73 nt upstream of the ATG start codon of BRK1 gene. As both Sp1 and BRK1 are known to promote a metastasis phenotype in tumor cells, it is likely that BRK1 acts as a downstream effector of Sp1 in accelerating tumor progression by promoting the assembly of actin filaments. Conflict of interest The authors declare no conflict of interest. Acknowledgments This work was supported by grants from the National Natural Science Foundation of China (30872548) and the National High Technology Research and Development Program of China (2012AA020206). References Abdelrahim, M., Smith III, R., Burghardt, R., et al., 2004. Role of Sp proteins in regulation of vascular endothelial growth factor expression and proliferation of pancreatic cancer cells. Cancer Research 64 (18), 6740–6749. Abdelrahim, M., Baker, C.H., Abbruzzese, J.L., et al., 2007. Regulation of vascular endothelial growth factor receptor-1 expression by specificity proteins 1, 3, and 4 in pancreatic cancer cells. Cancer Research 67 (7), 3286–3294. Black, A.R., Black, J.D., Azizkhan-Clifford, J., 2001. Sp1 and Kruppel-like factor family of transcription factors in cell growth regulation and cancer. Journal of Cellular Physiology 188 (2), 143–160. Bouwman, P., Philipsen, S., 2002. Regulation of the activity of Sp1-related transcription factors. Molecular and Cellular Endocrinology 195 (1–2), 27–38.

Cai, X., Xiao, T., James, S.Y., et al., 2009. Metastatic potential of lung squamous cell carcinoma associated with HSPC300 through its interaction with WAVE2. Lung Cancer 65 (3), 299–305. Chen, Y., Wang, X., Li, W., et al., 2011. Sp1 upregulates survivin expression in adenocarcinoma of lung cell line A549. Anatomical Record (Hoboken, N.J.: 2007) 294 (5), 774–780. Chiefari, E., Brunetti, A., Arturi, F., et al., 2002. Increased expression of AP2 and Sp1 transcription factors in human thyroid tumors: a role in NIS expression regulation? BMC Cancer 2, 35. Chintharlapalli, S., Papineni, S., Lee, S.O., et al., 2011. Inhibition of pituitary tumortransforming gene-1 in thyroid cancer cells by drugs that decrease specificity proteins. Molecular Carcinogenesis 50 (9), 655–667. Colon, J., Basha, M.R., Madero-Visbal, R., et al., 2011. Tolfenamic acid decreases c-Met expression through Sp proteins degradation and inhibits lung cancer cells growth and tumor formation in orthotopic mice. Investigational New Drugs 29 (1), 41–51. Derivery, E., Fink, J., Martin, D., et al., 2008. Free Brick1 is a trimeric precursor in the assembly of a functional wave complex. PLoS One 3 (6), e2462. Eden, S., Rohatgi, R., Podtelejnikov, A.V., et al., 2002. Mechanism of regulation of WAVE1induced actin nucleation by Rac1 and Nck. Nature 418 (6899), 790–793. Escobar, B., de Carcer, G., Fernandez-Miranda, G., et al., 2010. Brick1 is an essential regulator of actin cytoskeleton required for embryonic development and cell transformation. Cancer Research 70 (22), 9349–9359. Gautreau, A., Ho, H.Y., Li, J., et al., 2004. Purification and architecture of the ubiquitous Wave complex. Proceedings of the National Academy of Sciences of the United States of America 101 (13), 4379–4383. Guo, Z., Zhang, W., Xia, G., et al., 2010. Sp1 upregulates the four and half lim 2 (FHL2) expression in gastrointestinal cancers through transcription regulation. Molecular Carcinogenesis 49 (9), 826–836. Kadonaga, J.T., Courey, A.J., Ladika, J., et al., 1988. Distinct regions of Sp1 modulate DNA binding and transcriptional activation. Science 242 (4885), 1566–1570. Lin, R.K., Wu, C.Y., Chang, J.W., et al., 2010. Dysregulation of p53/Sp1 control leads to DNA methyltransferase-1 overexpression in lung cancer. Cancer Research 70 (14), 5807–5817. Liu, Y., Sun, W., Zhang, K., et al., 2007. Identification of genes differentially expressed in human primary lung squamous cell carcinoma. Lung Cancer 56 (3), 307–317. Narayan, V.A., Kriwacki, R.W., Caradonna, J.P., 1997. Structures of zinc finger domains from transcription factor Sp1. Insights into sequence-specific protein–DNA recognition. Journal of Biological Chemistry 272 (12), 7801–7809. Pathi, S.S., Jutooru, I., Chadalapaka, G., et al., 2011. GT-094, a NO-NSAID, inhibits colon cancer cell growth by activation of a reactive oxygen species-microRNA-27a: ZBTB10-specificity protein pathway. Molecular Cancer Research 9 (2), 195–202. Pavletich, N.P., Pabo, C.O., 1991. Zinc finger–DNA recognition: crystal structure of a Zif268–DNA complex at 2.1 A. Science 252 (5007), 809–817. Philipsen, S., Suske, G., 1999. A tale of three fingers: the family of mammalian Sp/XKLF transcription factors. Nucleic Acids Research 27 (15), 2991–3000. Safe, S., Abdelrahim, M., 2005. Sp transcription factor family and its role in cancer. European Journal of Cancer 41 (16), 2438–2448. Stradal, T.E., Scita, G., 2006. Protein complexes regulating Arp2/3-mediated actin assembly. Current Opinion in Cell Biology 18 (1), 4–10. Sun, W., Zhang, K., Zhang, X., et al., 2004. Identification of differentially expressed genes in human lung squamous cell carcinoma using suppression subtractive hybridization. Cancer Letters 212 (1), 83–93. Suske, G., 1999. The Sp-family of transcription factors. Gene 238 (2), 291–300. Takenawa, T., Suetsugu, S., 2007. The WASP–WAVE protein network: connecting the membrane to the cytoskeleton. Nature Reviews Molecular Cell Biology 8 (1), 37–48. Vasioukhin, V., Bauer, C., Yin, M., et al., 2000. Directed actin polymerization is the driving force for epithelial cell–cell adhesion. Cell 100 (2), 209–219. Wang, L., Wei, D., Huang, S., et al., 2003. Transcription factor Sp1 expression is a significant predictor of survival in human gastric cancer. Clinical Cancer Research 9 (17), 6371–6380. Wang, W., Goswami, S., Sahai, E., et al., 2005. Tumor cells caught in the act of invading: their strategy for enhanced cell motility. Trends in Cell Biology 15 (3), 138–145. Zannetti, A., Del Vecchio, S., Carriero, M.V., et al., 2000. Coordinate up-regulation of Sp1 DNA-binding activity and urokinase receptor expression in breast carcinoma. Cancer Research 60 (6), 1546–1551.