Early Growth Response 1 Acts As a Tumor Suppressor in Vivo and in Vitro Via Regulation of P53
Total Page:16
File Type:pdf, Size:1020Kb
Research Article Early Growth Response 1 Acts as a Tumor Suppressor In vivo and In vitro via Regulation of p53 Anja Krones-Herzig,1 Shalu Mittal,1 Kelly Yule,1 Hongyan Liang,2 Chris English,1 Rafael Urcis,1 Tarun Soni,1 Eileen D. Adamson,2 and Dan Mercola1,3 1Sidney Kimmel Cancer Center, San Diego, California and 2The Burnham Institute; 3Cancer Center, University of California at San Diego, La Jolla, California Abstract human tumor cell lines express little or no Egr1 in contrast to their normal counterparts (9–12). Furthermore, Egr1 is decreased or The early growth response 1 (Egr1) gene is a transcription factor that acts as both a tumor suppressor and a tumor undetectable in small cell lung tumors, human breast tumors promoter. Egr1-null mouse embryo fibroblasts bypass repli- (11, 13), and human gliomas (12). Reexpression of Egr1 in human tumor cells inhibits transformation. The mechanism of suppression cative senescence and exhibit a loss of DNA damage response h and an apparent immortal growth, suggesting loss of p53 involves the direct induction of TGF- 1 leading to an autocrine- functions. Stringent expression analysis revealed 266 tran- mediated suppression of transformation (8), increased fibronectin, scripts with >2-fold differential expression in Egr1-null mouse and plasminogen activator inhibitor (9). Egr1 also has been embryo fibroblasts, including 143 known genes. Of the 143 implicated in the regulation of p53 in human melanoma cells genes, program-assisted searching revealed 66 informative leading to apoptosis (14–16), and the proapoptotic tumor genes linked to Egr1. All 66 genes could be placed on a single suppressor gene PTEN also is directly regulated by Egr1 (17). Recently, we have identified and established a new role of Egr1 as regulatory network consisting of three branch points of known a ‘‘gatekeeper’’ of p53-dependent growth-regulatory mechanisms in Egr1 target genes: TGFb1, IL6,andIGFI. Moreover, 19 replicative senescence and cell growth (18). This result was additional genes that are known targets of p53 were identified, revealed by examination of primary mouse embryo fibroblasts indicating that p53 is a fourth branch point. Electrophoretic (MEF) isolated from Egr1-null mice developed previously by mobility shift assay as well as chromatin immunoprecipitation Lee et al. (19) and Topilko et al. (20). Egr1-null cells from either confirmed that p53 is a direct target of Egr1. Because deficient strain express no Egr1 protein and much reduced p53 protein. p53 expression causes tumors in mice, we tested the role of These cells completely bypass replicative senescence in culture, Egr1 in a two-step skin carcinogenesis study (144 mice) that thereby appearing to be immortal without passing through a revealed a uniformly accelerated development of skin tumors ‘‘crisis’’ stage. Moreover, these cells fail to arrest following treatment in Egr1-null mice (P < 0.005). These studies reveal a new role with DNA-damaging agents, such as g-irradiation (18) and 12-O- for Egr1 as an in vivo tumor suppressor. (Cancer Res 2005; tetradecanoylphorbol-13-acetate (TPA; see below). Replicative 65(12): 5133-43) senescence and the DNA damage response of the Egr1-null cells may be restored by infection with an Egr1-expressing retrovirus in Introduction contrast to p53-null MEFs. In the Egr1-null cells, p53 is not The transcription factor early growth response 1 (Egr1)isa expressed and the p53 gene remains entirely free of mutation to member of the immediate-early gene family and encodes a 59-kDa high passage numbers. In contrast, the wild-type (WT) cells express phosphoprotein observed at f80 kDa by electrophoresis. Egr1 is normal levels of p53, arrest at low passage in normal tissue culture, involved in the regulation of cell growth and differentiation in and undergo crisis. WT cells that survive crisis inevitably exhibit response to signals, such as mitogens, growth factors, and stress mutations of the p53 gene. We speculated that Egr1 exerts a stimuli (1–4). Egr1 has a COOH-terminal DNA-binding domain gatekeeper function over p53, thereby allowing the null cells to consisting of three zinc fingers that regulates transcription through bypass replicative senescence and avoid the mutation association GC-rich elements. The 9-bp DNA consensus-binding sequence with the survival of crisis. However, the exact mechanism whereby GCG(G/T)GGGCG has been identified in the promoter of several Egr1 functions in the regulation of p53-dependent growth arrest is growth-regulatory genes (5). Besides the highly conserved DNA- unknown, and there is no evidence that Egr1 mediates these roles binding domain, Egr1 also contains a nuclear localization signal, in a whole organism. two activator domains, and one repressor domain (6). In addition, Here, we examined gene expression in the genetically defined Egr1 binds to regulatory proteins called NAB1 and NAB2 (nerve Egr1-deficient MEFs. These cells exhibit at least 266 significant growth factor-I A-binding protein) that repress its transcriptional differences in transcript expression compared with WT cells, many activity (7). of which could be recognized as Egr1 target genes or downstream Analysis of certain human tumor cells and tissues indicate that signal transduction mediators. Indeed, an extended regulatory Egr1 exhibits prominent tumor suppressor function (5, 8–11). Many network based on four ‘‘nodes’’ could be constructed. Three nodes are upstream of numerous recognized Egr1-regulated genes. However, a fourth node, and potential Egr1 target gene, p53,is Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/). upstream of at least 19 genes not previously recognized as Egr1- A. Krones-Herzig and S. Mittal contributed equally to this work. regulated genes but are well-recognized mediators of p53 function. Requests for reprints: Dan Mercola, Sidney Kimmel Cancer Center, 10835 Altman Electrophoretic mobility shift assay (EMSA) and chromatin Row, San Diego, CA 92121. Phone: 858-450-5990, ext. 370; Fax: 858-450-3251; E-mail: [email protected]. immunoprecipitation (ChIP) experiments directly confirmed the I2005 American Association for Cancer Research. binding of the p53 promoter by Egr1 in cells both in vitro and www.aacrjournals.org 5133 Cancer Res 2005; 65: (12). June 15, 2005 Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Cancer Research in vivo. Moreover, TPA-induced skin tumors appeared rapidly in TGFb1, TGFb1I, GRO1, IGFII, THSP2, and procollagen XI. Primers were Egr1-null mice compared with WT and heterozygous littermates. selected for these genes with the aid of Primer Express software (Applied These studies indicate that Egr1 regulates an extensive network of Biosystems) and are available on request. The results were normalized to genes that include direct regulation of p53 at the level of RNA and the relative amounts of glyceraldehyde-3-phosphate dehydrogenase (GAPDH). protein expression. Our observations indicate that this mechanism Chromatin immunoprecipitation assay. WT and null MEFs and may function in vitro and in vivo to control DNA damage–induced human prostate cancer DU145 cells were treated with 1% formaldehyde for growth arrest. Loss of this mechanism in vivo may be related to 20 minutes to cross-link protein to binding sites on DNA. Egr1-containing increased skin tumor formation in Egr1-deficient mice. fragments were recovered by immunoprecipitation as described earlier (24). The ChIP-captured DNA was then screened for p53 promoter fragments by PCR analysis using the following p53 specific primers: (mouse) 5V- Materials and Methods GTAGAGTAAGCCCCCGGAAG-3V and 5V-GGTTACCGGGATTCAAAACA-3V Cells and cell culture. MEFs were prepared as described earlier (21) (the amplified fragments correspond to À925 and À424 region of the mouse from 19-day-old embryos from Egr1 WT, Egr1-null, and Egr1 heterozygous p53 promoter from AF287146); (human) 5V-TGGGAGTTGTAGTCTGAA- mice kindly provided by Dr. J. Milbrandt (Washington University Medical CGCTTC-3V and 5V-GAGAAGCTCAAAACTTTTAGCGCC-3V (the amplified School, St. Louis, MO) (19). Cultures of WT and Egr1-null MEFs were fragments correspond to À693 to +89 region of the human p53 promoter established and maintained in parallel for >60 passages. The predicted from X54156). Genomic DNA was used as a control for the amplification genotype and expression properties of the MEFs derived from Egr1-null and efficiency of each primer pair. heterozygous mice were confirmed by PCR-based analysis of DNA and RNA Electrophoretic mobility shift assay. WT and Egr1-null MEFs were and by Western analysis for protein expression as described previously (18). grown in tissue culture to confluence, incubated in serum-free medium for Oligonucleotide microarray analysis. Total cellular RNA was isolated 1 hour, and then supplemented with 20% fetal bovine serum for 1 hour. by using RNeasy kits (Qiagen Inc., Valencia, CA) and quantified. The total Cells were then lysed in hypotonic buffer (10 mmol/L HEPES, 25 mmol/L RNA quality was determined using the standard Affymetrix protocol. For KCl, 1 mmol/L EDTA, 1 mmol/L EGTA) containing 1:100 protease hybridization, total RNA (10 Ag) from WT or Egr1-null MEFs was reverse inhibitor cocktail (Sigma, St. Louis, MO) on ice for 10 minutes. Samples transcribed using an oligo(dT) primer harboring a T7 RNA polymerase were centrifuged at 1,500 rpm for 7 minutes and the pellets were promoter at the 5V end (Genset, San Diego, CA). Following second-strand incubated on ice for 30 minutes in nuclear extract buffer (50 mmol/L Tris- synthesis, biotinylated cRNA probes were produced. The probes were then HCl, 420 mmol/L KCl, 5 mmol/L MgCl2, 0.1 mmol/L EDTA) with 1:100 fragmented and hybridized to Affymetrix MGU75Av2 array representing protease inhibitor cocktail. The samples were centrifuged at 5,000 rpm for 12,566 mouse transcripts using the standard Affymetrix protocol (http:// 30 minutes at 4jC and the supernatants were removed for protein www.affymetrix.com).