The Oxidative Stressor Arsenite Activates Vascular Endothelial Growth Factor mRNA Transcription by an ATF4-dependent Mechanism
2005; Elsevier BV; Volume: 280; Issue: 21 Linguagem: Inglês
10.1074/jbc.m411275200
ISSN1083-351X
AutoresC. Nathaniel Roybal, Lucy A. Hunsaker, Olena Barbash, David L. Vander Jagt, Steven F. Abcouwer,
Tópico(s)Advanced Glycation End Products research
ResumoAberrant retinal expression of vascular endothelial growth factor (VEGF) leading to neovascularization is a central feature of age-related macular degeneration and diabetic retinopathy, two leading causes of vision loss. Oxidative stress is suggested to occur in retinal tissue during age-related macular degeneration and diabetic retinopathy and is suspected in the mechanism of VEGF expression in these diseases. Arsenite, a thiol-reactive oxidative stressor, induces VEGF expression by a HIF-1α-independent mechanism. Previously, we demonstrated that homocysteine, an endoplasmic reticulum stressor, increases VEGF transcription by a mechanism dependent upon activating transcription factor ATF4. Because ATF4 is expressed in response to oxidative stress, we hypothesized that ATF4 was also responsible for increased VEGF transcription in response to arsenite. We now show that arsenite increased steady state levels of VEGF mRNA and activated transcription from a VEGF promoter construct. Arsenite induced eIF2α phosphorylation, resulting in increased ATF4 protein levels. Inactivation or loss of ATF4 greatly diminished the VEGF response to arsenite treatment. Overexpression of ATF4 was sufficient to activate the VEGF promoter, and arsenite cooperated with exogenous ATF4 to further activate the promoter. A complex containing ATF4 binds a DNA element at +1767 bp relative to the VEGF transcription start site, and DNA binding activity is increased by arsenite treatment. In addition, the ability of a thiol antioxidant, N-acetylcysteine, to inhibit the effect of arsenite on VEGF expression coincided with its ability to inhibit phosphorylation of eIF2α and ATF4 protein expression. Thus, arsenite-induced up-regulation of VEGF gene transcription occurs by an ATF4-dependent mechanism. Aberrant retinal expression of vascular endothelial growth factor (VEGF) leading to neovascularization is a central feature of age-related macular degeneration and diabetic retinopathy, two leading causes of vision loss. Oxidative stress is suggested to occur in retinal tissue during age-related macular degeneration and diabetic retinopathy and is suspected in the mechanism of VEGF expression in these diseases. Arsenite, a thiol-reactive oxidative stressor, induces VEGF expression by a HIF-1α-independent mechanism. Previously, we demonstrated that homocysteine, an endoplasmic reticulum stressor, increases VEGF transcription by a mechanism dependent upon activating transcription factor ATF4. Because ATF4 is expressed in response to oxidative stress, we hypothesized that ATF4 was also responsible for increased VEGF transcription in response to arsenite. We now show that arsenite increased steady state levels of VEGF mRNA and activated transcription from a VEGF promoter construct. Arsenite induced eIF2α phosphorylation, resulting in increased ATF4 protein levels. Inactivation or loss of ATF4 greatly diminished the VEGF response to arsenite treatment. Overexpression of ATF4 was sufficient to activate the VEGF promoter, and arsenite cooperated with exogenous ATF4 to further activate the promoter. A complex containing ATF4 binds a DNA element at +1767 bp relative to the VEGF transcription start site, and DNA binding activity is increased by arsenite treatment. In addition, the ability of a thiol antioxidant, N-acetylcysteine, to inhibit the effect of arsenite on VEGF expression coincided with its ability to inhibit phosphorylation of eIF2α and ATF4 protein expression. Thus, arsenite-induced up-regulation of VEGF gene transcription occurs by an ATF4-dependent mechanism. Oxidative stress is involved in the pathogenesis of two leading causes of vision loss, age-related macular degeneration (AMD) 1The abbreviations used are: AMD, age-related macular degeneration; DR, diabetic retinopathy; VEGF, vascular endothelial growth factor; HIF, hypoxia-inducible factor; ER, endoplasmic reticulum; ATF, activating transcription factor; HO, heme oxygenase; NAC, N-acetylcysteine; MEF, mouse embryo fibroblast; Wt, wild type; DN, dominant negative; AARE, amino acid response element; AsnSyn, asparagine synthetase. 1The abbreviations used are: AMD, age-related macular degeneration; DR, diabetic retinopathy; VEGF, vascular endothelial growth factor; HIF, hypoxia-inducible factor; ER, endoplasmic reticulum; ATF, activating transcription factor; HO, heme oxygenase; NAC, N-acetylcysteine; MEF, mouse embryo fibroblast; Wt, wild type; DN, dominant negative; AARE, amino acid response element; AsnSyn, asparagine synthetase. and diabetic retinopathy (DR). This association is indicated by the beneficial effects of antioxidant supplementation in AMD patients (1Age-Related Eye Disease Study Research Group Arch. Ophthalmol. 2001; 119: 1417-1436Crossref PubMed Scopus (2594) Google Scholar) and in streptozotocin diabetic rats (2Obrosova I.G. Minchenko A.G. Marinescu V. Fathallah L. Kennedy A. Stockert C.M. Frank R.N. Stevens M.J. Diabetologia. 2001; 44: 1102-1110Crossref PubMed Scopus (155) Google Scholar). Furthermore, the breakdown of retinal antioxidant defenses has been demonstrated in aged and diabetic retinas (3Li W. Yanoff M. Jian B. He Z. Cell Mol. Biol. 1999; 45: 59-66PubMed Google Scholar, 4Liles M.R. Newsome D.A. Oliver P.D. Arch. Ophthalmol. 1991; 109: 1285-1288Crossref PubMed Scopus (162) Google Scholar). However, the mechanism by which oxidative stress may contribute to the development of AMD and DR is still unknown. Intraocular neovascularization is associated with both AMD and DR and is often a causative factor in the loss of vision during these pathologies. The neovascularization of the retina is frequently associated with increased expression of vascular endothelial growth factor (VEGF), a positive regulator of angiogenesis (for review, see Ref. 5Ferrara N. Endocr. Rev. 2004; 25: 581-611Crossref PubMed Scopus (2965) Google Scholar). Retinal pigment epithelial cells are a major source of VEGF in diseased retinas (6Lopez P.F. Sippy B.D. Lambert H.M. Thach A.B. Hinton D.R. Invest. Ophthalmol. Vis. Sci. 1996; 37: 855-868PubMed Google Scholar, 7Matsuoka M. Ogata N. Otsuji T. Nishimura T. Takahashi K. Matsumura M. Br. J. Ophthalmol. 2004; 88: 809-815Crossref PubMed Scopus (188) Google Scholar). However, the signaling mechanisms resulting in aberrant VEGF expression in aged and diabetic retinas are yet to be determined. Oxidative stressors, hydrogen peroxide and superoxide, are positive regulators of VEGF in the retinal pigment epithelial cells (8Kuroki M. Voest E.E. Amano S. Beerepoot L.V. Takashima S. Tolentino M. Kim R.Y. Rohan R.M. Colby K.A. Yeo K.T. Adamis A.P. J. Clin. Invest. 1996; 98: 1667-1675Crossref PubMed Scopus (414) Google Scholar). Antioxidant supplementation lowered VEGF protein levels in the retinas of diabetic rats (2Obrosova I.G. Minchenko A.G. Marinescu V. Fathallah L. Kennedy A. Stockert C.M. Frank R.N. Stevens M.J. Diabetologia. 2001; 44: 1102-1110Crossref PubMed Scopus (155) Google Scholar). Arsenite, an oxidative stressor, was shown previously to up-regulate VEGF mRNA and hypoxia-inducible factor 1α (HIF-1α) protein (9Duyndam M.C. Hulscher T.M. Fontijn D. Pinedo H.M. Boven E. J. Biol. Chem. 2001; 276: 48066-48076Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar). This led to the suggestion that HIF-1 was responsible for VEGF induction by arsenite. However, further investigation demonstrated that arsenite failed to induce a functional HIF-1 complex and that the VEGF induction was HIF-1-independent (10Duyndam M.C. Hulscher S.T. van der Wall E. Pinedo H.M. Boven E. J. Biol. Chem. 2003; 278: 6885-6895Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar). Therefore, an alternative pathway must regulate the induction of VEGF by arsenite. Previously, we reported that homocysteine, a known ER stressor (11Ji C. Kaplowitz N. World J. Gastroenterol. 2004; 10: 1699-1708Crossref PubMed Scopus (170) Google Scholar), which is elevated in DR (12Goldstein M. Leibovitch I. Yeffimov I. Gavendo S. Sela B.A. Loewenstein A. Eye. 2004; 18: 460-465Crossref PubMed Scopus (37) Google Scholar), up-regulated VEGF expression through an endoplasmic reticulum stress response pathway that was dependent on activating transcription factor 4 (ATF4) (13Roybal C.N. Yang S. Sun C.W. Hurtado D. Vander Jagt D.L. Townes T.M. Abcouwer S.F. J. Biol. Chem. 2004; 279: 14844-14852Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar). ATF4 protein levels increase in response to cellular stresses that cause global protein synthesis to be diminished. The ATF4 mRNA contains small upstream open reading frames that allow the ORF encoding ATF4 to be selectively translated when the translation initiation factor eIF2α is phosphorylated and levels of ternary initiation complex are low (14Harding H.P. Novoa I. Zhang Y. Zeng H. Wek R. Schapira M. Ron D. Mol. Cell. 2000; 6: 1099-1108Abstract Full Text Full Text PDF PubMed Scopus (2404) Google Scholar, 15Vattem K.M. Wek R.C. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 11269-11274Crossref PubMed Scopus (1122) Google Scholar). Activation of kinases that phosphorylate eIF2α leads to inhibition of global translation under conditions that perturb nascent protein processing, such as ER stress, nutrient deprivation, or exposure to oxidants and reactive metals (14Harding H.P. Novoa I. Zhang Y. Zeng H. Wek R. Schapira M. Ron D. Mol. Cell. 2000; 6: 1099-1108Abstract Full Text Full Text PDF PubMed Scopus (2404) Google Scholar, 16Fawcett T.W. Martindale J.L. Guyton K.Z. Hai T. Holbrook N.J. Biochem. J. 1999; 339: 135-141Crossref PubMed Scopus (368) Google Scholar). ATF4 is a key mediator in the adaptive response of cells to oxidants. Import of the amino acids glycine and cysteine, precursors for glutathione biosynthesis, is impaired in ATF4-null cells (17Harding H.P. Zhang Y. Zeng H. Novoa I. Lu P.D. Calfon M. Sadri N. Yun C. Popko B. Paules R. Stojdl D.F. Bell J.C. Hettmann T. Leiden J.M. Ron D. Mol. Cell. 2003; 11: 619-633Abstract Full Text Full Text PDF PubMed Scopus (2363) Google Scholar). Other ATF4-inducible genes are mediators of the cytoprotective program activated by oxidative stressors, an example being heme oxygenase-1 (HO-1) (18He C.H. Gong P. Hu B. Stewart D. Choi M.E. Choi A.M. Alam J. J. Biol. Chem. 2001; 276: 20858-20865Abstract Full Text Full Text PDF PubMed Scopus (382) Google Scholar). Our previous findings implicating ATF4 in VEGF expression along with the current understanding of ATF4 as a regulator of redox homeostasis led us to question the role of ATF4 in the arsenite-mediated up-regulation of VEGF. We now demonstrate that arsenite induces eIF2α phosphorylation and that ATF4 protein up-regulation precedes VEGF mRNA up-regulation. ATF4 is necessary for VEGF mRNA up-regulation and for transcriptional activation of the VEGF promoter in response to arsenite. N-Acetylcysteine (NAC), which blocks the effect of arsenite on VEGF expression, also inhibited eIF2α phosphorylation and ATF4 protein expression. These results suggest a model wherein oxidative stress induces VEGF in an ATF4-dependent mechanism thereby contributing to the neovascularization seen in retinal pathologies such as AMD and DR. Cell Culture—ARPE-19 cells (American Type Culture Collection, Manassas, VA) were maintained in Dulbecco's modified Eagle's medium (low glucose formulation) supplemented with 10% (v/v) fetal bovine serum, 100 units/ml penicillin, 100 μg/ml streptomycin, and 0.25 μg/ml amphotericin B. Immortalized mouse embryo fibroblasts (MEFs) were obtained from homozygous and heterozygous ATF4 knock-out MEF cultures (19Masuoka H.C. Townes T.M. Blood. 2002; 99: 736-745Crossref PubMed Scopus (176) Google Scholar) and were maintained in Dulbecco's modified Eagle's medium (high glucose formulation) supplemented with 10% (v/v) fetal bovine serum, 100 units/ml penicillin, 100 μg/ml streptomycin, 4 mm glutamine, and 50 μm thioglycerol. For Northern blotting experiments, cells were plated in 60-cm2 tissue culture dishes and grown to confluence. One day prior to being treated, ARPE-19 cells were fed with fresh media. MEF cultures were fed with fresh media containing 5 μm thioglycerol 1 day prior to treatment. Sodium arsenite (NaAsO2), N-acetylcysteine (NAC), ascorbic acid, trolox, and catalase (Sigma) were prepared fresh in Dulbecco's modified Eagle's medium and sterilized by filtration before being added to the cell cultures. For experiments utilizing adenoviral vectors, subconfluent cultures of ARPE-19 cells were treated with either ATF4 wild type (ATF4 Wt), ATF4 dominant negative mutant (ATF4 DN), or the empty AdEasy vector (Empty) (13Roybal C.N. Yang S. Sun C.W. Hurtado D. Vander Jagt D.L. Townes T.M. Abcouwer S.F. J. Biol. Chem. 2004; 279: 14844-14852Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar). The ATF4 DN mutant protein contains a 6-amino acid substitution in the DNA-binding domain (292RYRQKKR298 to 292GYLEAAA298) (18He C.H. Gong P. Hu B. Stewart D. Choi M.E. Choi A.M. Alam J. J. Biol. Chem. 2001; 276: 20858-20865Abstract Full Text Full Text PDF PubMed Scopus (382) Google Scholar). Twenty h post-infection, the percentage of cells infected was determined by examining the expression of enhanced green fluorescent protein by fluorescent microscopy. Northern Blot Analysis—RNA isolation and Northern blotting was performed as described previously (20Abcouwer S.F. Schwarz C. Meguid R.A. J. Biol. Chem. 1999; 274: 28645-28651Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar) using cDNAs corresponding to human VEGF (GenBank™ accession number dbEST189750), GRP78 (HAEAC89, ATCC), and heme oxygenase-1 (21Keyse S.M. Tyrrell R.M. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 99-103Crossref PubMed Scopus (1108) Google Scholar) and normalized to 18 S rRNA. Rat heme oxygenase-A (GenBank™ accession number AA874884) was used for Northern analysis of MEF RNA. 18 S rRNA template was generated using mouse RNA as described previously (13Roybal C.N. Yang S. Sun C.W. Hurtado D. Vander Jagt D.L. Townes T.M. Abcouwer S.F. J. Biol. Chem. 2004; 279: 14844-14852Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar). Membranes were hybridized with radiolabeled DNA probe for 6–8 h at 60 °C in a solution containing 7% (v/v) SDS, 0.25 m Na2HPO4, pH 7.2, as described (22Kevil C.G. Walsh L. Laroux F.S. Kalogeris T. Grisham M.B. Alexander J.S. Biochem. Biophys. Res. Commun. 1997; 238: 277-279Crossref PubMed Scopus (96) Google Scholar). Blotting results were quantified by overnight exposure to a phosphor screen followed by analysis using a STORM™ PhosphorImager and ImageQuant™ software (Amersham Biosciences). For each sample, hybridization to 18 S rRNA was used to normalize results for mRNAs. Fold inductions were determined by dividing normalized mRNA band intensity volumes for experimental samples to that of control (untreated or time 0) samples. VEGF Promoter Cloning—Initially, a HindIII restriction enzyme digest was performed on the human 137-kb PAC clone containing the VEGF-A gene (GenBank™ accession number AL136131). Following double gel purification (Qiagen, Valencia, CA) an 11.7-kb HindIII fragment was blunt-ended with the PCRTerminator™ end repair kit (Lucigen, Middleton, WI). The fragment was ligated into the pSMART-LC-Kan vector using the CLONESMART™ blunt end cloning kit (Lucigen, Middleton, WI). The resulting pSMART-LC-Kan-VEGF plasmid was digested with XhoI and the blunt end cutter EcoRV, excising an 8.2-kb fragment ranging from –6363 to +1886 kb relative to the VEGF transcription start site. The 8.2-kb fragment was then ligated into the XhoI- and SmaI-digested PGL-3-basic reporter vector (Promega, Madison, WI) creating the reporter vector pVEGF8.2-Luc. Sequencing performed by the University of New Mexico DNA Core Facility confirmed insert identity and orientation. Dual Luciferase Assays—ARPE-19 cells were plated at 50% confluence in 12-well plates and allowed to adhere overnight. Cells in each well were transfected with a CaCl2-DNA precipitate containing: 2.5 μg of pVEGF8.2-Luc, 20 ng of pRL-CMV vector (Promega) and incubated for 4 h. Cells were then treated with 1 ml of glycerol shock solution (4× Tris-buffered saline, 15% glycerol, and 600 μm Na2HPO4) for 90 s and washed twice with TBST (142 mm NaCl, 2.7 mm KCl, 25 mm Tris, 1% Tween 20). Cells were allowed to recover for 20 h post-transfection, fed with fresh media, and treated with arsenite for the indicated times and doses. For reporter experiments with exogenous ATF4 expression, cells were infected with either Empty, ATF4 Wt, or ATF4 DN adenovirus by adding viral vectors at the time of plating. Infected cultures were then transfected with the plasmids the following day as described above. Cell lysates were obtained with passive lysis buffer (Promega) and analyzed with the Dual-luciferase® assay (Promega) using a TD-20/20 luminometer (Turner Designs, Sunnyvale, CA). The firefly luciferase relative light units in each sample were normalized to the Renilla luciferase relative light units, and the sample mean ratio was divided by the control mean ratio to give fold inductions. Statistical significance of differences between triplicate samples was determined with the two-sample t test. Western Blot Analysis—ARPE-19 or MEF cells were grown to confluence in 60 cm2 plates and treated as described in the figure legends. Following treatment, cells were washed in cold phosphate-buffered saline and lysed for 20 min on ice in lysis buffer (0.1% SDS, 0.5% sodium deoxycholate, 1% Triton X-100, 50 mm Tris-HCl, pH 8.0) containing protease inhibitor mixture (Roche Applied Science). For samples to be analyzed with phospho-specific antibodies, phosphatase inhibitor mixture (50 mm NaF, 200 μm Na3VO4, 10 mm Na3PO4, and 100 μm EDTA) was included in the lysis buffer. Protein contents of cleared lysates were determined with a BCA protein assay kit (Pierce), and equal amounts of proteins (20 or 40 μg) were loaded into each lane and separated on a 10% SDS-PAGE gel. The protein bands were then transferred to nitrocellulose membrane (Bio-Rad) and probed with antibodies specific for ATF4 and eIF2α (Santa Cruz Biotechnology, Santa Cruz, CA), phosphorylated eIF2α (Cell Signaling Technology, Beverly, MA), or β-actin (Sigma). Proteins were detected with ECL™ Chemiluminescence Kit (Amersham Biosciences) according to manufacture's instructions. Membranes were scanned and viewed with a MultiGenius Bioimaging System® (Syngene, Cambridge, UK). MEF Immortalization—ATF4 heterozygous and homozygous null MEFs were obtained from Tim Townes (Department of Biochemistry and Molecular Genetics, University of Alabama at Birmingham). Cultures of 90% confluent ATF4+/– or ATF4–/– MEFs were transfected with a plasmid containing isolated SV40 early gene encoding the large T antigen alone (pD305 vector) (23Chang L.S. Pater M.M. Hutchinson N.I. di Mayorca G. Virology. 1984; 133: 341-353Crossref PubMed Scopus (33) Google Scholar) by CaCl2 co-precipitation as described above. Forty-eight h later, transfected cells were subcultured (1:10 dilution) and maintained until fully confluent (1 week). Cells were then serially diluted in 60-cm2 plates and selected for clonal growth. Approximately 2 weeks after plating, cell colonies were either isolated with cloning cylinders or harvested together with all colonies on a plate to form pooled populations. Clonal and pooled populations were expanded, and expression of large T antigen was confirmed by Western blot analysis (data not shown). Pooled populations designated as ATF4–/–/D305/1:10 (ATF4–/–) and ATF4+/–/D305/1:10 (ATF4+/–) were utilized for the present study. The presence of ATF4 protein was confirmed in the ATF4+/– population and its absence proven in ATF4–/– population by Western blotting (Fig. 4B). DNA Binding Activity Analysis—Electrophoretic mobility shift assays were performed using Gel Shift Assay System (Promega) according to the manufacturer's protocol. The double-stranded DNA oligonucleotides containing putative ATF4-binding sites (underlined) known as amino acid response elements (AAREs) were as follows: CHOP1 AARE, 5′-TAGAGACAGGGTTTCACCATGTTGGCCAGG-3′; CHOP2 AARE in negative orientation, 5′-CCTGGGCAACATGGTGAAACACCATCTCTA-3′; DRAL AARE in negative orientation, 5′-TAGTGCACGAATGATGGAAAGGGAGGGTTG-3′; and asparagine synthetase (AsnSyn) AARE, 5′-GCGCGGAGCCGATTACATCAGCCCGGGCCT-3′ (Integrated DNA Technologies, Coralville, IA). In the binding reactions, α-32P-labeled DNA probes were incubated with 10 μg of nuclear extract. Binding reactions were performed at room temperature for 20 min, and the DNA-protein complexes were separated by electrophoresis on 5% Tris borate-EDTA-PAGE gel and visualized using a STORM™ PhosphorImager and ImageQuant™ software (Amersham Biosciences). For ATF4 supershifts, 2 μg of polyclonal anti-ATF4 (CREB-2, H-290; Santa Cruz Biotechnology) or monoclonal anti-myc (9B11; Cell Signaling Technology) were incubated with the binding mixture for 10 min before loading reactions on the gel. Dose Response and Time Course of VEGF, HO-1, and GRP78 mRNA Expression in ARPE-19 Cells Exposed to Arsenite—To characterize the induction of VEGF mRNA in response to arsenite, confluent ARPE-19 cells were treated with increasing doses of NaAsO2 (0, 5, 20, 100, and 300 μm) for 4 h. Total cellular RNA was isolated, and Northern blotting was performed to analyze VEGF mRNA content for each sample. (Fig. 1A). In addition, mRNA levels for HO-1 and GRP78 were examined. These genes served as indicators of oxidative and ER stress responses, respectively. In addition, ATF4 plays a role in regulation of both these genes (18He C.H. Gong P. Hu B. Stewart D. Choi M.E. Choi A.M. Alam J. J. Biol. Chem. 2001; 276: 20858-20865Abstract Full Text Full Text PDF PubMed Scopus (382) Google Scholar, 24Luo S. Baumeister P. Yang S. Abcouwer S.F. Lee A.S. J. Biol. Chem. 2003; 278: 37375-37385Abstract Full Text Full Text PDF PubMed Scopus (226) Google Scholar). The ARPE-19 cells contained detectable amounts of a single mRNA species corresponding to VEGF. The VEGF mRNA levels were normalized relative to 18 S rRNA and plotted. VEGF steady state mRNA increased maximally (5.1-fold) with 100 μm arsenite. The decrease observed with 300 μm arsenite coincided with apparent toxicity at this dosage. Cell death was not observed at lower concentrations of arsenite. HO-1 mRNA levels were also increased by arsenite as previously described (25Sardana M.K. Drummond G.S. Sassa S. Kappas A. Pharmacology. 1981; 23: 247-253Crossref PubMed Scopus (64) Google Scholar). HO-1 expression was much more sensitive to arsenite stress than that of VEGF. A dramatic increase in HO-1 mRNA was caused by 5 μm arsenite, representing roughly a 20-fold increase in sensitivity as compared with VEGF. Expression of GRP78/BiP mRNA was not induced by arsenite, suggesting that the unfolded protein response was not activated. Using the dose associated with the maximal VEGF induction, 100 μm arsenite, ARPE-19 cells were treated for various times (0, 0.5, 1, 2, 4, and 8 h). This time course study revealed a maximal (4.2-fold) increase of VEGF mRNA at 4 h and a decline by 8 h. Although the sensitivity to arsenite was different for VEGF and HO-1; they exhibited similar kinetics of induction. Again, no induction of GRP78 mRNA was observed. Activation of the VEGF Promoter by Arsenite—Further analysis of the effect of arsenite on VEGF transcription was carried out with a reporter vector, pVEGF8.2-Luc, containing 8.2 kb of the VEGF promoter region. This particular gene fragment was chosen because computer analysis of this 5′ region of genomic DNA revealed it to contain four putative ATF4-binding sites, termed AARE (13Roybal C.N. Yang S. Sun C.W. Hurtado D. Vander Jagt D.L. Townes T.M. Abcouwer S.F. J. Biol. Chem. 2004; 279: 14844-14852Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar). ARPE-19 cells were transfected with pVEGF8.2-Luc and pRL-CMV, allowed to recover for 20 h, treated with arsenite for the times, and doses indicated and analyzed using a dual luciferase assay. Dose-response samples were obtained after 8 h of exposure to 0, 20, 60, 100, and 140 μm arsenite. A maximal induction of 3.7-fold was achieved with 100 μm arsenite (Fig. 2A). Time course experiments were performed with exposure to 100 μm sodium arsenite for 0, 4, 6, 8, and 12 h (Fig. 2B). Arsenite induced reporter expression maximally at 8 h with an increase of 3.7-fold. eIF2α Phosphorylation and ATF4 Protein Expression in ARPE-19 Cells Exposed to Arsenite—eIF2α phosphorylation and the subsequent up-regulation of ATF4 protein expression following arsenite treatment were expected if increased VEGF transcription was mediated by ATF4. Previous studies have reported phosphorylation of eIF2α (26Duncan R.F. Hershey J.W. Arch. Biochem. Biophys. 1987; 256: 651-661Crossref PubMed Scopus (32) Google Scholar) and subsequent increased in ATF4 DNA binding activity (16Fawcett T.W. Martindale J.L. Guyton K.Z. Hai T. Holbrook N.J. Biochem. J. 1999; 339: 135-141Crossref PubMed Scopus (368) Google Scholar) upon treatment with arsenite. Therefore, the effects of 100 μm arsenite on phosphorylation of eIF2α and ATF4 protein levels in ARPE-19 cells were analyzed. Phosphorylation of eIF2α following arsenite treatment was rapid and transient (Fig. 3A). The level of phosphorylated eIF2α protein was greatly increased within 15 min and returned to basal levels within 120 min. Total eIF2α levels did not vary appreciably between the samples. To examine the effects on ATF4 protein levels, ARPE-19 cells were treated with 100 μm arsenite for 0, 1, 2, and 4 h and analyzed by Western blotting with an antibody to ATF4. The antibody detected a prominent band of ∼47 kDa, corresponding to ATF4 protein, only in samples obtained at 2 h and 4 h of arsenite treatment (Fig. 3B). The faster migrating band detected by this antibody was determined to be due to nonspecific binding, as demonstrated previously (13Roybal C.N. Yang S. Sun C.W. Hurtado D. Vander Jagt D.L. Townes T.M. Abcouwer S.F. J. Biol. Chem. 2004; 279: 14844-14852Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar) and by confirming the size of endogenous ATF4 through comparison to exogenously expressed ATF4 (data not shown). β-Actin was used as a loading control and did not vary markedly between samples. ATF4 protein induction coincided with VEGF mRNA induction (Fig. 1). Thus, ATF4 was available to activate VEGF transcription in a temporal manner that agrees with the time course results of VEGF mRNA induction and VEGF promoter activation. The Role of ATF4 in VEGF Expression following Arsenite Treatment—To further examine the role of ATF4 in arsenite-induced up-regulation of endogenous VEGF mRNA, an adenoviral vector system was used to express either ATF4 wild type (ATF4 Wt) or ATF4 dominant negative (ATF4 DN) proteins, as described previously (13Roybal C.N. Yang S. Sun C.W. Hurtado D. Vander Jagt D.L. Townes T.M. Abcouwer S.F. J. Biol. Chem. 2004; 279: 14844-14852Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar). An empty vector (Empty) was used to control for the effects of infection and enhanced green fluorescent protein expression. Viral titers were assessed by determining the percentage of cells that exhibited enhanced green fluorescent protein fluorescence 20 h post-infection with these vectors. Viral infection of ARPE-19 cells with ATF4 Wt virus stock at a 1:320 dilution, ATF4 DN virus stock at a 1:160 dilution, and Empty virus stock at a 1:320 dilution resulted in infection rates between 90 and 100% (data not shown). ARPE-19 cells were thus infected with viruses at these dilutions of virus stocks. Twenty h after contact with the viruses, the cultures were fed with fresh media and subjected to the following: no treatment (incubated with control media), incubated with media containing 100 μm sodium arsenite, or incubated with 10 mm dl-homocysteine-containing media for 4 h. VEGF mRNA expression was then analyzed by Northern blotting (Fig. 4A). Relative VEGF mRNA levels in cells treated with empty vector were increased by arsenite (3.8-fold) and dl-homocysteine (5.5-fold). The VEGF mRNA inductions were further increased in all treatment groups when ATF4 Wt protein was overexpressed. In contrast, the expression of ATF4 DN greatly inhibited VEGF expression under all three conditions. To confirm the effect of the ATF4 DN on VEGF expression, replicate ARPE-19 cultures were infected with ATF4 DN virus stock at a 1:120 dilution or Empty virus stock at a 1:320 dilution and 48 h later were incubated with media containing 100 μm sodium arsenite for 4 h. ATF4 DN mutant significantly reduced VEGF mRNA expression by 71% (p < 0.001, n = 3). Heme oxygenase-1 was previously shown to be an ATF4-responsive gene via ATF4 binding on the HO-1 promoter in a complex containing the bZIP transcription factor Nrf2 (18He C.H. Gong P. Hu B. Stewart D. Choi M.E. Choi A.M. Alam J. J. Biol. Chem. 2001; 276: 20858-20865Abstract Full Text Full Text PDF PubMed Scopus (382) Google Scholar). In ARPE-19 cells infected with empty virus, HO-1 mRNA expression was induced 31-fold by arsenite treatment (Fig. 4A). In untreated control cells, exogenous overexpression of ATF4 Wt protein increased the expression of HO-1 mRNA by 3.8-fold, whereas ATF4 DN increased HO-1 expression by 54%. ATF4 Wt overexpression had no effect on HO-1 mRNA expression in arsenite treated cells, as the level was also 31-fold that of the untreated, empty virus-infected control. Expression of ATF4 DN reduced arsenite induction of HO-1 mRNA by 29%, to 22-fold compared with the untreated, empty virus-infected control. Thus, ATF4 Wt overexpression was sufficient to induce HO-1 expression and ATF4 DN had an inhibitory effect on HO-1 induction by arsenite, but these effects were not remarkable. In contrast to arsenite, homocysteine did not induce HO-1 mRNA expression. In fact, the induc
Referência(s)