- Email: [email protected]
GATA-1 binding sites in exon 1 direct erythroid-specific transcription of PPOX
Available online at www.sciencedirect.com
Gene 409 (2008) 83 – 91 www.elsevier.com/locate/gene
GATA-1 binding sites in exon 1 direct erythroid-specific transcription of PPOX Karen M.K. de Vooght, Richard van Wijk, Wouter W. van Solinge ⁎ Department of Clinical Chemistry and Haematology, Laboratory for Red Blood Cell Research, University Medical Center Utrecht, Utrecht, The Netherlands Received 28 May 2007; received in revised form 5 October 2007; accepted 24 November 2007 Available online 4 December 2007 Received by R. Di Lauro
Abstract We investigated erythroid-specific expression of the human PPOX gene. This gene encodes protoporphyrinogen oxidase, which is involved in synthesizing heme for red blood cells and heme as a cofactor for the respiratory cytochromes. In vitro luciferase transfection assays in human uninduced and hemin induced erythroleukemic K562 cells showed that the presence of exon 1 increased promoter activity fourfold as compared to reporter constructs lacking this exon. This transcriptional regulation was mediated by two GATA-1 sites in exon 1. Electrophoretic mobility shift and chromatin immunoprecipitation assays demonstrated that both GATA sites were able to bind GATA-1 in vitro and in vivo. Exon 1 did not affect promoter activity in human hepatoma HepG2 cells and U937 monocytic cells but its presence decreased promoter activity in HeLa human cervical carcinoma cells. We conclude that the GATA-1 binding sites in exon 1 constitute key regulatory elements in differential expression of PPOX in erythroid and non-erythroid cells. © 2007 Elsevier B.V. All rights reserved. Keywords: Red cells; Erythroid-specific expression; Heme biosynthetic enzyme; Transcriptional regulation; Promoter; 5′UTR
1. Introduction Protoporphyrinogen oxidase (PPOX, EC 1.3.3.4), the penultimate enzyme in the heme biosynthetic pathway, catalyzes the six-electron oxidation of protoporphyrinogen IX to protopor-
Abbreviations: PPOX, protoporphyrinogen oxidase gene; 5′UTR, 5′ untranslated region; bp, base pair(s); nts, nucleotides; PCR, Polymerase Chain Reaction; cDNA, DNA complementary to RNA; EMSA, Electrophoretic mobility shift assay; RT-PCR, Reverse Transcription-Polymerase Chain Reaction; ChIP, chromatin immunoprecipitation; ALAD, δ-aminolevulinate dehydratase gene; PBGD, porphobilinogen deaminase gene; UROS, uroporphyrinogen III synthase gene; UROD, uroporphyrinogen decarboxylase gene; CPO, coproporphyrinogen oxidase gene; FECH, ferrochelatase gene; HBG1, γA-globin gene; CCR3, CC Chemokine Receptor 3 gene; SPTA1, α-spectrin gene; PAX5, Pax5 gene. ⁎ Corresponding author. Department of Clinical Chemistry and Haematology, Laboratory for Red Blood Cell Research, University Medical Center Utrecht, Postbus 85500, 3508 GA, Utrecht, The Netherlands. Tel.: +31 30 2507604; fax: +31 30 2505418. E-mail address: [email protected] (W.W. van Solinge). 0378-1119/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2007.11.010
phyrin IX. Like other heme biosynthetic enzymes, PPOX is involved in synthesizing heme for red blood cells (erythroidspecific expression) and as a cofactor for the respiratory cytochromes (housekeeping expression). During erythroid differentiation, the expression of each enzyme of the heme synthetic pathway increases sequentially. PPOX activity is therefore higher in erythroid than in non-erythroid cells (Sassa, 1976; Conder et al., 1991; Fujita et al., 1991a; Lake-Bullock and Dailey, 1993; Taketani et al., 1995b; Takahashi et al., 1998). The human PPOX gene spans about 8 kb and contains 13 exons (Taketani et al., 1995a). Northern blot analysis of eight different human tissues (heart, brain, placenta, lung, liver, skeletal muscle, kidney and pancreas) showed evidence for a single ~ 1.8 kb transcript in all tissue types (Dailey and Dailey, 1996). Exon 1 is part of the 5′ untranslated region (5′UTR) and transcription of PPOX is initiated at two major sites in both erythroid and non-erythroid cells (Fig. 1) (Taketani et al., 1995a). Expression of PPOX is under control of a single promoter that lacks a canonical TATA box, a property common to housekeeping promoters (Taketani et al., 1995a).
84
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
Fig. 1. PPOX exon 1 and 60 nts of its 5′-flanking region. Putative regulatory elements are indicated by arrows (Transcription Element Search Software, assembly date: January 2006) (Schug and Overton, 1997). Exons are depicted as boxes (open boxes: coding regions; filled boxes: 5′UTR). An arrow indicates the translational ATG site. The two major transcriptional start sites (Taketani et al., 1995a) are underlined.
Although tissue-specific regulation of other heme biosynthetic enzymes is extensively studied (Romana et al., 1987; Chretien et al., 1988; Fujita et al., 1991b; Tugores et al., 1994; Taketani et al., 1995a; Bishop et al., 1996; Takahashi et al., 1998; Aizencang et al., 2000), there is little knowledge concerning transcriptional regulation of PPOX. Functional studies have only been performed on the murine PPOX gene in erythroid and non-erythroid cell lines, showing that erythroid-specific expression required elements in the − 1160 to − 746 bp promoter region, while transcriptional elements involved in housekeeping expression were located in the region − 746 to + 50 bp. The − 198 to + 50 bp region flanking the transcriptional start site exhibited no regulatory elements required for either housekeeping or erythroid-specific expression (Dailey et al., 2002). With regard to the human PPOX gene, no functional studies have been performed. Only Taketani et al. (1995a) suggested, based on the sequence, that certain regulatory elements in the promoter might be involved in housekeeping expression (CCAAT box and Sp1 binding sites) and other regulatory elements in erythroid-specific expression (GATA-1 binding sites). In this study we report on the role of exon 1 in erythroidspecific transcriptional regulation of the human PPOX gene. 2. Materials and methods 2.1. Promoter constructs Promoter reporter constructs were prepared according to the method described by de Vooght et al. (2005) The human wild-type PPOX promoter construct lacking exon 1 (pGL3-PPOX),
spanning nts −635 to +23 (GenBank, accession no. X99450) of PPOX, was amplified from the DNA of a healthy individual by use of forward primer PPOX-F2: 5′-AATTGGAGTCTTCTTGGGACC-3′ (nts −635 to −615), and reverse primer PPOXR3: 5′-CCGTCCACTCTGTTCTCG-3′ (nts +6 to +23). Forward primer PPOX-F2 and reverse primer PPOX-R-EXON1: 5′CCACAATAGGTAGGGATGAGAG-3′ (nts +243 to +265) were used to generate a PPOX promoter construct that included exon 1 (pGL3-PPOX + 1, spanning nts −635 to +265). Constructs with mutated GATA-1 binding motifs in exon 1 were obtained by site-directed mutagenesis using double-stranded plasmid DNA templates, as described (Braman et al., 1996). The most upstream GATA binding motif in exon 1, designated GATA-1A (Fig. 1, nts +155 to +160), was mutated (GATA → GTTA) by use of sense primer GATA-AmutS: 5′-GCCCTGCGAGGGCCGTTAGCGAGGGTGTGGC-3′ (nts +142 to +172) and its complementary primer GATA-AmutAS (pGL3-GATA_Amut). The other GATA binding motif in exon 1, designated GATA-1B (Fig. 1, nts +175 to +180), was mutated in a similar manner by use of sense primer GATA-BmutS: 5′-GGGTGTGGCCCTTAACTGCACCCAGCAGAG-3′ (nts +164 to +193) and its complementary primer GATA-BmutAS (pGL3-GATA_Bmut). The GATA-1B site of pGL3-GATA_Amut was mutated, as described above, generating the double GATA mutant pGL3-GATA_A/Bmut. These constructs rendered low levels of expression in HepG2, HeLa and U937 cells. Therefore, constructs were subcloned into the MluI and XhoI sites of the pGL3 enhancer vector (Promega Corporation, Madison, WI, USA), generating: pGL3e-PPOX, pGL3e-PPOX + 1, pGL3eGATA_Amut, pGL3e-GATA_Bmut and pGL3e-GATA_A/ Bmut. The consensus GATA binding motifs at position −304
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
and −300 were mutated using the wild-type construct, sense primer GATA-304/-300mutS: 5′-GGCCTTAAGTGTCCCAAACAATTCTCCCTACG-3′ (nts −318 to −287) and its complementary primer GATA-304/-300mutAS (pGL3-GATA_Dmut+ 1 and pGL3-GATA_Dmut − 1). The non-consensus GATA binding motif at position +256 was mutated by use of primer GATA256mutS: 5′-CTCTCATCCCTACCAATTGTGGCCTGAATTC3′ (nts +243 to +273) and complementary primer GATA256mutAS (pGL3-GATA_Cmut). All plasmids were verified by DNA sequence analysis. 2.2. Cell culture Human erythroleukemic K562 cells were cultured in RPMI 1640 medium (Invitrogen, Paisley, UK) supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 U penicillin, and 100 μg streptomycin (Invitrogen) per mL. To investigate the
85
erythroid-specific nature of the GATA-1 binding sites in exon 1, cells were induced with hemin three days prior to transfection or enzymatic shearing of chromatin (for ChIP assay), according to Fibach et al. (1995) and Aizencang et al. (2000). Human monocytic U937 cells were cultured in RPMI 1640 medium supplemented with 10% fetal calf serum, 0.05 mM β-mercaptoethanol, 2 mM L-glutamine, 100 U penicillin, and 100 μg streptomycin per mL. Human hepatoma HepG2 and human cervical carcinoma HeLa cells were cultured in MEM + Earle's + L-glutamine (Invitrogen) with the same supplements as described above. All cells were grown at 37 °C in a humidified atmosphere containing 5% CO2. 2.3. DNA transfections K562 cells and U937 cells were transiently transfected using Superfect (Qiagen, Valencia, CA), as previously described (de
Fig. 2. Exon 1 directs high-level expression of PPOX in K562 cells. PPOX reporter gene constructs without (pGL3-PPOX) or with exon 1 (pGL3-PPOX + 1) were transiently transfected in K562 cells (panel A). Similar promoter constructs were transfected in HepG2 and in HeLa cells (panel B). Luciferase activities were calculated relative to the pGL3-SV40 (control) vector (n = 4, samples in duplicate).
86
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
Vooght et al., 2005). An adapted protocol was used for transfection of HepG2 and HeLa cells. In brief, 24 h prior to transfection cells were seeded in 24-well plates, to give about
50% confluency at the time of transfection. Cells were transfected with 2 μg of the reporter plasmid DNA and 50 ng of Renilla Luciferase pRL-SV40 vector (Promega), which was
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
used as an internal control. Transfection complexes were removed after 150 min and fresh medium was added. Forty-eight hours after transfection, luciferase activity was measured with the Dual-Luciferase Reporter Assay System (Promega), using the Veritas™ Microplate Luminometer (Promega). Firefly luciferase activities were corrected for transfection efficiency by using Renilla luciferase activity measurements. The pGL3control vector and the promoterless pGL3-Basic Luciferase Reporter Vector (Promega) were used as positive and negative controls, respectively. 2.4. Quantitative RT-PCR for determination of relative mRNA levels Total RNA was isolated from K562 cells 48 h after transfection using the storage reagent RNAlater® (Ambion, Austin, Texas, USA) and the isolation kit RNAqueous®-4PCR (Ambion). Isolated RNA was treated with Turbo DNase (Ambion) to remove contaminating plasmid and genomic DNA. Total RNA was reverse transcribed using AMV reverse transcriptase (Roche, Diagnostics, Mannheim, Germany) and oligo d(T) primers. After first-strand synthesis, the cDNA was quantified by TaqMan realtime PCR using Luciferase and Renilla gene specific primers and dual-labeled probes (Applied Biosystems, Warrington, Cheshire, UK) (Chen et al., 2003). Fluorescence was detected with an ABI Prism 7900 sequence detection system (Applied Biosystems, Foster City, CA, USA). The sequence of the TaqMan probe for Firefly luciferase was as follows: 5′-VIC-CGCAGGTGTCGCAGG-MGB-3′. The sequence of the TaqMan probe for Renilla luciferase was: 5′-6-FAM-CCTCTTCTTATTTATGGCGACAT-MGB-3′. Firefly luciferase mRNA was normalized to Renilla luciferase content. Relative quantification of Firefly luciferase mRNA was performed by the comparative CT method (Livak method (Livak and Schmittgen, 2001)) and expressed as the percentage of Firefly luciferase mRNA measured in cells transfected with pGL3-PPOX + 1. 2.5. Electrophoretic mobility shift assay Electrophoretic mobility shift assay (EMSA) was performed as described previously (de Vooght et al., 2005) using nuclear extracts from K562, HeLa, and HepG2 cells. Extracts were prepared according to the method of Dignam et al. (1983). The following wild-type and mutant single-stranded primers (sense, mutations in bold), with their respective complementary antisense primers, were used to investigate DNA-protein interactions at both GATA sites in exon 1 of PPOX: GATA_AWTs, 5′-GCGAGGGCCGATAGCGAGGGT-3′; GATA_Amuts, 5′-GCGAGGGCCGTTAGCGAGGGT-3′; GATA_BWTs, 5′-GTGGCCCTTATCTGCACCCAG-3′; GATA_Bmuts 5′-GTGGCCCTTAACTGCACCCAG-3′.
87
For supershift assay, 1 μg of anti-GATA-1 (m-20 X) or antiNF-E2 (sc-16276 X) (Santa Cruz Biotechnology, Inc, Santa Cruz, CA, USA) was incubated with nuclear extract for 30 min at room temperature prior to the addition of the probe. 2.6. Chromatin immunoprecipitation assay To investigate in vivo binding of GATA-1 to the GATA-1 binding sites in exon 1 of PPOX we performed chromatin immunoprecipitation (ChIP) assay, by use of the ChIP-IT™ Express Enzymatic kit (Active Motif, Carlsbad, CA, USA) and the ChIP-validated antibody GATA-1 pAB (39025, Active Motif), according to the manufacturer's instructions. Briefly, K562 cells were fixed using formaldehyde and chromatin was sheared into small, uniform fragments by enzymatic digestion. Specific GATA-1/DNA complexes were immunoprecipitated using a ChIP-validated antibody directed against human GATA1. Normal rabbit IgG (sc-2027, Santa Cruz) was used as a control to confirm absence of non-specific IgG/DNA binding. Following immunoprecipitation, crosslinking was reversed, proteins removed by treatment with proteinase K and DNA recovered. 7.5 μL of immunoprecipitated DNA was PCR amplified using primers covering both GATA-1 binding sites in exon 1 of PPOX: primers PPOX ChIP forward 5′-GGAGTAGCGGATTTGAAGCA-3′ and PPOX ChIP reverse 5′TCACCCACAATAGGTAGGGAT-3′. The length of the amplified PPOX PCR product was 214 nts. 3. Results 3.1. Exon 1 directs erythroid-specific expression of PPOX To establish the functional relevance of exon 1 with regard to erythroid-specific expression of PPOX, we transfected PPOX promoter constructs without exon 1 (pGL3-PPOX) and with exon 1 (pGL3-PPOX + 1) into K562 cells. Luciferase activities were calculated relative to the pGL3 control vector. These results showed that the presence of exon 1 led to a fourfold increase in promoter activity (Fig. 2, panel A). To study the relevance of exon 1 with regard to housekeeping expression, transfections were performed in HepG2 and HeLa cells. Transfection of the promoter enhancer vectors in HepG2 and HeLa cells demonstrated that the presence of exon 1 did not affect promoter activity in HepG2 cells but reduced promoter activity by 50% in HeLa cells (Fig. 2, panel B). 3.2. Two GATA-1 binding sites in exon 1 mediate expression of PPOX on the transcriptional level Exon 1 of PPOX contains two putative GATA-1 binding sites. Since GATA-1 is a well-known modulator of erythroid-
Fig. 3. Two GATA-1 binding motifs in exon 1 mediate expression of PPOX in K562 cells on the transcriptional level. PPOX reporter plasmids containing wild-type and mutated GATA-1 binding motifs were transfected into K562 cells (panel A), HepG2 and HeLa cells (panel B). Luciferase activities were calculated relative to the PPOX promoter reporter vector with exon 1. TaqMan real-time RT-PCR was performed on Firefly and Renilla luciferase mRNA in K562 cells transfected with wildtype and mutant PPOX reporter plasmids (panel C). Quantification of Firefly luciferase mRNA was performed using the comparative CT method (Livak method (Livak and Schmittgen, 2001)) and expressed as percentage of wild-type + exon 1 (n = 3, samples in eightfold).
88
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
specific transcriptional regulation (Ferreira et al., 2005), we postulated these two binding sites to be of importance with respect to the observed increased promoter activity. The first GATA-1 motif (GATA-1A, CGATAG) constituted a nonconsensus sequence whereas the second motif (GATA-1B, AGATAA) constituted a full consensus sequence (WGATAR) (Merika and Orkin, 1993) for binding of GATA-1 (Fig. 1). To determine the functional importance of both GATA-1 sites, single and double mutant GATA-1 luciferase promoter reporter plasmids were generated and transfected into K562 cells. Mutation of the non-consensus GATA-1A site (CGATAG → CGTTAG, a mutation that has been shown to disrupt GATA-1 binding (Zon et al., 1991)) did not significantly alter promoter activity (Fig. 3, panel A). An identical mutation of the consensus GATA-1B site caused a 36% reduction in promoter activity. When both GATA-1 binding sites were mutated, promoter activity was reduced by 66%. The activity of this double mutant GATA-1 reporter construct (34% ± 4%) was comparable with the wild-type PPOX construct lacking exon 1 (23% ± 2%) (Fig. 3, panel A). Other putative GATA-1 binding sites in the PPOX promoter construct (at position − 304, − 300 and + 256) were found to be non-functional in K562 cells (Fig. 3, panel A). To confirm that the reduced promoter activity of the GATA-1 mutant PPOX promoter constructs was due to inhibition of transcription, we quantified luciferase mRNA levels. The amount of Firefly luciferase mRNA in cells transfected with wild-type and mutant PPOX promoter constructs (Fig. 3, panel
C) corresponded with the measured Firefly luciferase activity (Fig. 3, panel A) in these cells (mean ratio of luciferase activity versus luciferase transcripts was 0.98 (range: 0.88–1.10). Hence, the reduction in luciferase activity of the GATA-1 mutant PPOX promoter constructs is due to reduced transcription of Firefly luciferase. Subsequently we studied the effect of the two GATA-1 mutants on housekeeping expression of PPOX. Enhancer promoter reporter plasmids containing the mutant GATA-1 binding sites were transfected into the non-erythroid cell lines HepG2 and HeLa. Mutating either of the two GATA-1 binding motifs in exon 1, as well as mutating them simultaneously, did not affect promoter activity in both HepG2 and HeLa cells (Fig. 3, panel B). 3.3. GATA-1 binding sites in exon 1 direct erythroid-specific transcription of PPOX To verify the erythroid-specific character of both GATA-1 binding sites in exon 1 of PPOX, we repeated transfection experiments in K562 cells induced with hemin to differentiate into the erythroid direction. Monocytic U937 cells were used as (non-erythroid) control. As shown in Fig. 4, relative luciferase activities of the different promoter reporter vectors in induced K562 cells were similar to activities in uninduced K562 cells (Fig. 3). Exon 1 did not increase promoter activity in monocytic U937 cells (Fig. 4). These results confirm that the GATA-1 binding sites in exon 1 are essential for erythroid-specific
Fig. 4. GATA-1 binding sites in exon 1 direct erythroid-specific transcription of PPOX. PPOX reporter gene constructs were transiently transfected in K562 cells induced with hemin three days prior to transfection. Similar promoter constructs were transfected in U937 cells. Luciferase activities were calculated relative to the pGL3-SV40 (control) vector (n = 4, samples in duplicate).
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
expression of PPOX, and not involved in housekeeping expression of PPOX. 3.4. GATA-1 binds both GATA-1 binding sites in exon 1 of PPOX Further evidence for the role of GATA-1 in PPOX transcription came from EMSA studies. Both wild-type PPOX probes showed a specific complex with K562 nuclear extracts (Fig. 5A and B, lane 1). Formation of these complexes was effectively abolished by addition of excess unlabeled homologous oligonucleotide (competitor) (Fig. 5A and B, lane 2), but not by addition of excess mutant competitor (Fig. 5A and B,
89
lane 3). Addition of anti-human GATA-1 antibody caused a supershift of the specific oligonucleotide/GATA-1 complex (Fig. 5A and B, lane 4). This was a specific reaction, since no supershift occurred after addition of anti-human NF-E2 antibody (results not shown). In addition, EMSAs were carried out using labeled oligonucleotides harboring the respective GATA-1 mutations. These did not lead to formation of the specific oligonucleotide/GATA-1 complexes (Fig. 5A and B, lane 5). To evaluate in vivo binding of GATA-1 to exon 1 of PPOX we performed ChIP assays in K562 cells. Amplification of recovered sheared chromatin with primers annealing to exon 1 of PPOX revealed a 214 nts PCR product specific for PPOX
Fig. 5. GATA-1 binds to both GATA binding motifs in exon 1 of PPOX in vitro. Electrophoretic mobility shift assays were performed with K562 nuclear extract and labeled wild-type and mutant oligonucleotide probes. The absence (−) or presence (+) of 500 fmol of unlabeled competitors is indicated, as well as the addition of 1 μg of anti-GATA-1. The lowest arrow denotes the specific GATA-1-DNA complex, the upper arrow the GATA-1-DNA supershifted complex.
90
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
Fig. 6. GATA-1 binds to both GATA binding motifs in exon 1 of PPOX in vivo. Chromatin from K562 cells was immunoprecipitated with a ChIP-validated GATA-1 antibody (anti-GATA-1) and PCR amplified using primers flanking the two GATA-1 binding sites in exon 1 of PPOX (PCR product of 412 nts). Amplification of input chromatin (input) prior to immunoprecipitation served as positive control for chromatin extraction and PCR amplification. Chromatin immunoprecipitation using non-specific antibody (rabbit IgG) served as negative control. In addition, PCR in absence of DNA (DNA negative) was performed to exclude DNA contamination of the reaction mixture.
(Fig. 6). This product was absent when using normal rabbit IgG. From these results we concluded that GATA-1 directly interacts with exon 1 of PPOX in K562 cells. 4. Discussion In this study we describe a novel mechanism involved in the regulation of human protoporphyrinogen oxidase gene expression. We show that erythroid-specific regulation of PPOX depends on two GATA-1 sites in exon 1. The enzymes of the heme biosynthetic pathway serve as a model for differential regulation of housekeeping proteins. Heme is synthesized in red blood cells and in most other tissues where it serves as a cofactor for respiratory cytochromes. The enzymes involved in its biosynthesis are ubiquitously expressed. The genes encoding the second, third and fourth enzyme of the heme biosynthesis, δ-aminolevulinate dehydratase (ALAD), porphobilinogen deaminase (PBGD) and uroporphyrinogen III synthase (UROS), are each transcribed from two distinct promoters (Chretien et al., 1988; Bishop et al., 1996; Aizencang et al., 2000). In all of these genes the housekeeping promoter is located upstream of exon 1. The erythroid-specific promoters are located in intron 1, immediately upstream of exon 2, which is the first coding exon in all of these genes (Aizencang et al., 2000). Transcription of the genes that code for the fifth to eighth enzyme of the heme biosynthetic pathway (uroporphyrinogen decarboxylase (UROD), coproporphyrinogen oxidase (CPO), PPOX and ferrochelatase (FECH)) is regulated by a single promoter (Romana et al., 1987; Tugores et al., 1994; Taketani et al., 1995a; Takahashi et al., 1998). Until this report, no studies were performed with regard to (tissue-specific) regulation of the PPOX gene in erythroid and non-erythroid cells. Transfection experiments of wild-type and mutant reporter plasmids in uninduced and hemin induced K562 cells demonstrated that erythroid-specific transcriptional regulation of
PPOX was mediated by two GATA-1 binding sites in exon 1 (Figs. 3 and 4). In addition, both GATA-1 sites were able to bind GATA-1 in vitro and in vivo (Figs. 5 and 6). These results suggest therefore that both GATA-1 binding sites in exon 1 are involved in the high level of PPOX expression in erythroid cells. The importance of this regulation is supported by the observation that both GATA-1 binding sites are conserved in rat and mouse (data from Ensembl). The consensus GATAB binding site is fully conserved in rat and mouse, while the non-consensus GATAA binding site in exon 1 of the human PPOX gene is conserved except for the first nucleotide. Taketani et al. (1995a) suggested that putative GATA-1 binding sites in the promoter region might regulate PPOX expression during erythroid differentiation. In contrast, we show that these other putative GATA-1 binding sites (the consensus GATA binding motifs at position −304 and − 300, and the non-consensus GATA binding motif at position + 256), are non-functional (Fig. 3, panel A). 5′ Untranslated exons can regulate the expression of genes in two different ways. First, they may act as a tissue-specific translational regulator (Pickering and Willis, 2005). Second, the 5′UTR may regulate gene expression by mediating transcriptional regulation (Zimmermann et al., 2005). A pivotal role of the 5′UTR, and the transcription factors interacting with it, is described for several genes such as the γ-globin gene (HBG1) and the CC Chemokine Receptor 3 gene (CCR3) (Amrolia et al., 1995; Zimmermann et al., 2005). Our experiments show that the 5′UTR of human PPOX regulates gene expression on the level of transcription rather than translation (Fig. 3). More specifically, transcriptional regulation is mediated by binding of GATA-1 to two GATA-1 binding sites in the 5′UTR. There are only few examples of genes in which GATA-1 is known to interact with the 5′UTR (Amrolia et al., 1995; Wong et al., 2004; Zimmermann et al., 2005). In the human HBG1 gene GATA-1 is suggested to interact with the 5′UTR, causing transcriptional repression of γA-globin gene expression (Amrolia et al., 1995). GATA-1 also interacts with exon 1 of CCR3 and Zimmermann et al. suggest that GATA-1 has a regulatory role in CCR3 transcription (Zimmermann et al., 2005). GATA-1 binding in exon 1 of the α-spectrin gene (SPTA1) is also required for high-level expression of α-spectrin, as described by Wong et al. (2004). Our findings with regard to the importance of exon 1, and both GATA-1 sites in particular, for erythroid-specific expression of PPOX are in contrast with a recent report by Kotze and colleagues. They reported that the 5′UTR of the PPOX gene is particularly mutation prone, leading them to conclude that this region is probably not an important part of the gene (Kotze et al., 1998). Also, Dailey et al. (2002) showed that erythroidspecific transcription of the mouse PPOX gene required more upstream regulatory elements. However, the constructs used in this latter study lacked the part of exon 1 that contains four putative GATA-1 binding sites (+ 80 to + 288) (Schug and Overton, 1997). Hence, this may explain the observed lack of erythroid-specific promoter activity. In contrast to the key role of exon 1 in erythroid-specific expression of human PPOX, no effect of this exon was seen on
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
promoter activity in HepG2 or U937 cells (Figs. 2 and 4). Exon 1 even decreased promoter activity in HeLa cells by 50% (Fig. 2). The latter could be due to an effect on translation or due to the presence of repressor element(s) in exon 1. The occurrence of repressor elements in the 5′UTR of human genes has been reported in a number of human genes, for example in Pax5 (PAX5) (Rahman et al., 2001). Our results regarding expression in HeLa, HepG2 and U937 cells suggest that the two GATA-1 binding sites in exon 1 do not play a regulatory role other than in erythroid-specific expression (Figs. 3 and 4). Sp1 and/or GRα are candidate transcription factors that may repress PPOX promoter activity (Akerblom et al., 1988; Zaid et al., 2001). In conclusion, we demonstrate that erythroid-specific transcriptional regulation of the human PPOX gene is mediated by two GATA-1 binding sites in exon 1. These GATA-1 binding sites do not play a regulatory role in housekeeping expression. Our results contribute to a better understanding of the molecular mechanisms involved in transcriptional regulation of PPOX, and the transcriptional role of 5′UTR's in general. Acknowledgement The authors sincerely wish to thank Adri Thomas for helpful discussions. References Aizencang, G., Solis, C., Bishop, D.F., Warner, C., Desnick, R.J., 2000. Human uroporphyrinogen-III synthase: genomic organization, alternative promoters, and erythroid-specific expression. Genomics 70, 223–231. Akerblom, I.E., Slater, E.P., Beato, M., Baxter, J.D., Mellon, P.L., 1988. Negative regulation by glucocorticoids through interference with a cAMP responsive enhancer. Science 241, 350–353. Amrolia, P.J., Cunningham, J.M., Ney, P., Nienhuis, A.W., Jane, S.M., 1995. Identification of two novel regulatory elements within the 5′-untranslated region of the human Aγ -globin gene. J. Biol. Chem. 270, 12892–12898. Bishop, T.R., Miller, M.W., Beall, J., Zon, L.I., Dierks, P., 1996. Genetic regulation of δ-aminolevulinate dehydratase during erythropoiesis. Nucleic Acids Res. 24, 2511–2518. Braman, J., Papworth, C., Greener, A., 1996. Site-directed mutagenesis using double-stranded plasmid DNA templates. Methods Mol. Biol. 57, 31–44. Chen, L.S., Tassone, F., Sahota, P., Hagerman, P.J., 2003. The (CGG)n repeat element within the 5′ untranslated region of the FMR1 message provides both positive and negative cis effects on in vivo translation of a downstream reporter. Hum. Mol. Genet. 12, 3067–3074. Chretien, S., et al., 1988. Alternative transcription and splicing of the human porphobilinogen deaminase gene result either in tissue-specific or in housekeeping expression. Proc. Natl. Acad. Sci. U. S. A. 85, 6–10. Conder, L.H., Woodard, S.I., Dailey, H.A., 1991. Multiple mechanisms for the regulation of haem synthesis during erythroid cell differentiation. Possible role for coproporphyrinogen oxidase. Biochem. J. 275, 321–326. Dailey, T.A., Dailey, H.A., 1996. Human protoporphyrinogen oxidase: expression, purification, and characterization of the cloned enzyme. Protein Sci. 5, 98–105. Dailey, T.A., McManus, J.F., Dailey, H.A., 2002. Characterization of the mouse protoporphyrinogen oxidase gene. Cell. Mol. Biol. 48, 61–69. de Vooght, K.M.K., van Wijk, R., van Oirschot, B.A., Rijksen, G., van Solinge, W.W., 2005. Pyruvate kinase regulatory element 1 (PKR-RE1) mediates hexokinase gene expression in K562 cells. Blood Cells Mol. Diseases 34, 186–190.
91
Dignam, J.D., Lebovitz, R.M., Roeder, R.G., 1983. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 11, 1475–1489. Ferreira, R., Ohneda, K., Yamamoto, M., Philipsen, S., 2005. GATA1 function, a paradigm for transcription factors in hematopoiesis. Mol. Cell. Biol. 25, 1215–1227. Fibach, E., Kollia, P., Schechter, A.N., Noguchi, C.T., Rodgers, G.P., 1995. Hemin-induced acceleration of hemoglobin production in immature cultured erythroid cells: preferential enhancement of fetal hemoglobin. Blood 85, 2967–2974. Fujita, H., et al., 1991a. Sequential activation of genes for heme pathway enzymes during erythroid differentiation of mouse Friend virus-transformed erythroleukemia cells. Biochim. Biophys. Acta 1090, 311–316. Fujita, H., Yamamoto, M., Yamagami, T., Hayashi, N., Sassa, S., 1991b. Erythroleukemia differentiation. Distinctive responses of the erythroidspecific and the nonspecific δ-aminolevulinate synthase mRNA. J. Biol. Chem. 266, 17494–17502. Kotze, M.J., et al., 1998. Molecular analysis reveals a high mutation frequency in the first untranslated exon of the PPOX gene and largely excludes variegate porphyria in a subset of clinically affected Afrikaner families. Mol. Cell. Probes 12, 293–300. Lake-Bullock, H., Dailey, H.A., 1993. Biphasic ordered induction of heme synthesis in differentiating murine erythroleukemia cells: role of erythroid 5aminolevulinate synthase. Mol. Cell. Biol. 13, 7122–7132. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2− ΔΔCT method. Methods 25, 402–408. Merika, M., Orkin, S.H., 1993. DNA-binding specificity of GATA family transcription factors. Mol. Cell. Biol. 13, 3999–4010. Pickering, B.M., Willis, A.E., 2005. The implications of structured 5′ untranslated regions on translation and disease. Semin. Cell Dev. Biol. 16, 39–47. Rahman, M., et al., 2001. A repressor element in the 5′-untranslated region of human Pax5 exon 1A. Gene 263, 59–66. Romana, M., Dubart, A., Beaupain, D., Chabret, C., Goossens, M., Romeo, P.H., 1987. Structure of the gene for human uroporphyrinogen decarboxylase. Nucleic Acids Res. 15, 7343–7356. Sassa, S., 1976. Sequential induction of heme pathway enzymes during erythroid differentiation of mouse Friend leukemia virus-infected cells. J. Exp. Med. 143, 305–315. Schug, J., Overton, G.C., 1997. TESS: Transcription Element Search Software on the WWW. Computational Biology and Informatics Laboratory, School of Medicine. University of Pennsylvania, Pennsylvania. Takahashi, S., et al., 1998. Differential regulation of coproporphyrinogen oxidase gene between erythroid and nonerythroid cells. Blood 92, 3436–3444. Taketani, S., et al., 1995a. The human protoporphyrinogen oxidase gene (PPOX): organization and location to chromosome 1. Genomics 29, 698–703. Taketani, S., et al., 1995b. Induction of terminal enzymes for heme biosynthesis during differentiation of mouse erythroleukemia cells. Eur. J. Biochem. 230, 760–765. Tugores, A., Magness, S.T., Brenner, D.A., 1994. A single promoter directs both housekeeping and erythroid preferential expression of the human ferrochelatase gene. J. Biol. Chem. 269, 30789–30797. Wong, E.Y., Lin, J., Forget, B.G., Bodine, D.M., Gallagher, P.G., 2004. Sequences downstream of the erythroid promoter are required for high level expression of the human α-spectrin gene. J. Biol. Chem. 279, 55024–55033. Zaid, A., Hodny, Z., Li, R., Nelson, B.D., 2001. Sp1 acts as a repressor of the human adenine nucleotide translocase-2 (ANT2) promoter. Eur. J. Biochem. 268, 5497–5503. Zimmermann, N., Colyer, J.L., Koch, L.E., Rothenberg, M.E., 2005. Analysis of the CCR3 promoter reveals a regulatory region in exon 1 that binds GATA-1. BMC Immunol. 6, 7. Zon, L.I., Youssoufian, H., Mather, C., Lodish, H.F., Orkin, S.H., 1991. Activation of the erythropoietin receptor promoter by transcription factor GATA-1. Proc. Natl. Acad. Sci. U. S. A. 88, 10638–10641.
Gene 409 (2008) 83 – 91 www.elsevier.com/locate/gene
GATA-1 binding sites in exon 1 direct erythroid-specific transcription of PPOX Karen M.K. de Vooght, Richard van Wijk, Wouter W. van Solinge ⁎ Department of Clinical Chemistry and Haematology, Laboratory for Red Blood Cell Research, University Medical Center Utrecht, Utrecht, The Netherlands Received 28 May 2007; received in revised form 5 October 2007; accepted 24 November 2007 Available online 4 December 2007 Received by R. Di Lauro
Abstract We investigated erythroid-specific expression of the human PPOX gene. This gene encodes protoporphyrinogen oxidase, which is involved in synthesizing heme for red blood cells and heme as a cofactor for the respiratory cytochromes. In vitro luciferase transfection assays in human uninduced and hemin induced erythroleukemic K562 cells showed that the presence of exon 1 increased promoter activity fourfold as compared to reporter constructs lacking this exon. This transcriptional regulation was mediated by two GATA-1 sites in exon 1. Electrophoretic mobility shift and chromatin immunoprecipitation assays demonstrated that both GATA sites were able to bind GATA-1 in vitro and in vivo. Exon 1 did not affect promoter activity in human hepatoma HepG2 cells and U937 monocytic cells but its presence decreased promoter activity in HeLa human cervical carcinoma cells. We conclude that the GATA-1 binding sites in exon 1 constitute key regulatory elements in differential expression of PPOX in erythroid and non-erythroid cells. © 2007 Elsevier B.V. All rights reserved. Keywords: Red cells; Erythroid-specific expression; Heme biosynthetic enzyme; Transcriptional regulation; Promoter; 5′UTR
1. Introduction Protoporphyrinogen oxidase (PPOX, EC 1.3.3.4), the penultimate enzyme in the heme biosynthetic pathway, catalyzes the six-electron oxidation of protoporphyrinogen IX to protopor-
Abbreviations: PPOX, protoporphyrinogen oxidase gene; 5′UTR, 5′ untranslated region; bp, base pair(s); nts, nucleotides; PCR, Polymerase Chain Reaction; cDNA, DNA complementary to RNA; EMSA, Electrophoretic mobility shift assay; RT-PCR, Reverse Transcription-Polymerase Chain Reaction; ChIP, chromatin immunoprecipitation; ALAD, δ-aminolevulinate dehydratase gene; PBGD, porphobilinogen deaminase gene; UROS, uroporphyrinogen III synthase gene; UROD, uroporphyrinogen decarboxylase gene; CPO, coproporphyrinogen oxidase gene; FECH, ferrochelatase gene; HBG1, γA-globin gene; CCR3, CC Chemokine Receptor 3 gene; SPTA1, α-spectrin gene; PAX5, Pax5 gene. ⁎ Corresponding author. Department of Clinical Chemistry and Haematology, Laboratory for Red Blood Cell Research, University Medical Center Utrecht, Postbus 85500, 3508 GA, Utrecht, The Netherlands. Tel.: +31 30 2507604; fax: +31 30 2505418. E-mail address: [email protected] (W.W. van Solinge). 0378-1119/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2007.11.010
phyrin IX. Like other heme biosynthetic enzymes, PPOX is involved in synthesizing heme for red blood cells (erythroidspecific expression) and as a cofactor for the respiratory cytochromes (housekeeping expression). During erythroid differentiation, the expression of each enzyme of the heme synthetic pathway increases sequentially. PPOX activity is therefore higher in erythroid than in non-erythroid cells (Sassa, 1976; Conder et al., 1991; Fujita et al., 1991a; Lake-Bullock and Dailey, 1993; Taketani et al., 1995b; Takahashi et al., 1998). The human PPOX gene spans about 8 kb and contains 13 exons (Taketani et al., 1995a). Northern blot analysis of eight different human tissues (heart, brain, placenta, lung, liver, skeletal muscle, kidney and pancreas) showed evidence for a single ~ 1.8 kb transcript in all tissue types (Dailey and Dailey, 1996). Exon 1 is part of the 5′ untranslated region (5′UTR) and transcription of PPOX is initiated at two major sites in both erythroid and non-erythroid cells (Fig. 1) (Taketani et al., 1995a). Expression of PPOX is under control of a single promoter that lacks a canonical TATA box, a property common to housekeeping promoters (Taketani et al., 1995a).
84
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
Fig. 1. PPOX exon 1 and 60 nts of its 5′-flanking region. Putative regulatory elements are indicated by arrows (Transcription Element Search Software, assembly date: January 2006) (Schug and Overton, 1997). Exons are depicted as boxes (open boxes: coding regions; filled boxes: 5′UTR). An arrow indicates the translational ATG site. The two major transcriptional start sites (Taketani et al., 1995a) are underlined.
Although tissue-specific regulation of other heme biosynthetic enzymes is extensively studied (Romana et al., 1987; Chretien et al., 1988; Fujita et al., 1991b; Tugores et al., 1994; Taketani et al., 1995a; Bishop et al., 1996; Takahashi et al., 1998; Aizencang et al., 2000), there is little knowledge concerning transcriptional regulation of PPOX. Functional studies have only been performed on the murine PPOX gene in erythroid and non-erythroid cell lines, showing that erythroid-specific expression required elements in the − 1160 to − 746 bp promoter region, while transcriptional elements involved in housekeeping expression were located in the region − 746 to + 50 bp. The − 198 to + 50 bp region flanking the transcriptional start site exhibited no regulatory elements required for either housekeeping or erythroid-specific expression (Dailey et al., 2002). With regard to the human PPOX gene, no functional studies have been performed. Only Taketani et al. (1995a) suggested, based on the sequence, that certain regulatory elements in the promoter might be involved in housekeeping expression (CCAAT box and Sp1 binding sites) and other regulatory elements in erythroid-specific expression (GATA-1 binding sites). In this study we report on the role of exon 1 in erythroidspecific transcriptional regulation of the human PPOX gene. 2. Materials and methods 2.1. Promoter constructs Promoter reporter constructs were prepared according to the method described by de Vooght et al. (2005) The human wild-type PPOX promoter construct lacking exon 1 (pGL3-PPOX),
spanning nts −635 to +23 (GenBank, accession no. X99450) of PPOX, was amplified from the DNA of a healthy individual by use of forward primer PPOX-F2: 5′-AATTGGAGTCTTCTTGGGACC-3′ (nts −635 to −615), and reverse primer PPOXR3: 5′-CCGTCCACTCTGTTCTCG-3′ (nts +6 to +23). Forward primer PPOX-F2 and reverse primer PPOX-R-EXON1: 5′CCACAATAGGTAGGGATGAGAG-3′ (nts +243 to +265) were used to generate a PPOX promoter construct that included exon 1 (pGL3-PPOX + 1, spanning nts −635 to +265). Constructs with mutated GATA-1 binding motifs in exon 1 were obtained by site-directed mutagenesis using double-stranded plasmid DNA templates, as described (Braman et al., 1996). The most upstream GATA binding motif in exon 1, designated GATA-1A (Fig. 1, nts +155 to +160), was mutated (GATA → GTTA) by use of sense primer GATA-AmutS: 5′-GCCCTGCGAGGGCCGTTAGCGAGGGTGTGGC-3′ (nts +142 to +172) and its complementary primer GATA-AmutAS (pGL3-GATA_Amut). The other GATA binding motif in exon 1, designated GATA-1B (Fig. 1, nts +175 to +180), was mutated in a similar manner by use of sense primer GATA-BmutS: 5′-GGGTGTGGCCCTTAACTGCACCCAGCAGAG-3′ (nts +164 to +193) and its complementary primer GATA-BmutAS (pGL3-GATA_Bmut). The GATA-1B site of pGL3-GATA_Amut was mutated, as described above, generating the double GATA mutant pGL3-GATA_A/Bmut. These constructs rendered low levels of expression in HepG2, HeLa and U937 cells. Therefore, constructs were subcloned into the MluI and XhoI sites of the pGL3 enhancer vector (Promega Corporation, Madison, WI, USA), generating: pGL3e-PPOX, pGL3e-PPOX + 1, pGL3eGATA_Amut, pGL3e-GATA_Bmut and pGL3e-GATA_A/ Bmut. The consensus GATA binding motifs at position −304
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
and −300 were mutated using the wild-type construct, sense primer GATA-304/-300mutS: 5′-GGCCTTAAGTGTCCCAAACAATTCTCCCTACG-3′ (nts −318 to −287) and its complementary primer GATA-304/-300mutAS (pGL3-GATA_Dmut+ 1 and pGL3-GATA_Dmut − 1). The non-consensus GATA binding motif at position +256 was mutated by use of primer GATA256mutS: 5′-CTCTCATCCCTACCAATTGTGGCCTGAATTC3′ (nts +243 to +273) and complementary primer GATA256mutAS (pGL3-GATA_Cmut). All plasmids were verified by DNA sequence analysis. 2.2. Cell culture Human erythroleukemic K562 cells were cultured in RPMI 1640 medium (Invitrogen, Paisley, UK) supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 U penicillin, and 100 μg streptomycin (Invitrogen) per mL. To investigate the
85
erythroid-specific nature of the GATA-1 binding sites in exon 1, cells were induced with hemin three days prior to transfection or enzymatic shearing of chromatin (for ChIP assay), according to Fibach et al. (1995) and Aizencang et al. (2000). Human monocytic U937 cells were cultured in RPMI 1640 medium supplemented with 10% fetal calf serum, 0.05 mM β-mercaptoethanol, 2 mM L-glutamine, 100 U penicillin, and 100 μg streptomycin per mL. Human hepatoma HepG2 and human cervical carcinoma HeLa cells were cultured in MEM + Earle's + L-glutamine (Invitrogen) with the same supplements as described above. All cells were grown at 37 °C in a humidified atmosphere containing 5% CO2. 2.3. DNA transfections K562 cells and U937 cells were transiently transfected using Superfect (Qiagen, Valencia, CA), as previously described (de
Fig. 2. Exon 1 directs high-level expression of PPOX in K562 cells. PPOX reporter gene constructs without (pGL3-PPOX) or with exon 1 (pGL3-PPOX + 1) were transiently transfected in K562 cells (panel A). Similar promoter constructs were transfected in HepG2 and in HeLa cells (panel B). Luciferase activities were calculated relative to the pGL3-SV40 (control) vector (n = 4, samples in duplicate).
86
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
Vooght et al., 2005). An adapted protocol was used for transfection of HepG2 and HeLa cells. In brief, 24 h prior to transfection cells were seeded in 24-well plates, to give about
50% confluency at the time of transfection. Cells were transfected with 2 μg of the reporter plasmid DNA and 50 ng of Renilla Luciferase pRL-SV40 vector (Promega), which was
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
used as an internal control. Transfection complexes were removed after 150 min and fresh medium was added. Forty-eight hours after transfection, luciferase activity was measured with the Dual-Luciferase Reporter Assay System (Promega), using the Veritas™ Microplate Luminometer (Promega). Firefly luciferase activities were corrected for transfection efficiency by using Renilla luciferase activity measurements. The pGL3control vector and the promoterless pGL3-Basic Luciferase Reporter Vector (Promega) were used as positive and negative controls, respectively. 2.4. Quantitative RT-PCR for determination of relative mRNA levels Total RNA was isolated from K562 cells 48 h after transfection using the storage reagent RNAlater® (Ambion, Austin, Texas, USA) and the isolation kit RNAqueous®-4PCR (Ambion). Isolated RNA was treated with Turbo DNase (Ambion) to remove contaminating plasmid and genomic DNA. Total RNA was reverse transcribed using AMV reverse transcriptase (Roche, Diagnostics, Mannheim, Germany) and oligo d(T) primers. After first-strand synthesis, the cDNA was quantified by TaqMan realtime PCR using Luciferase and Renilla gene specific primers and dual-labeled probes (Applied Biosystems, Warrington, Cheshire, UK) (Chen et al., 2003). Fluorescence was detected with an ABI Prism 7900 sequence detection system (Applied Biosystems, Foster City, CA, USA). The sequence of the TaqMan probe for Firefly luciferase was as follows: 5′-VIC-CGCAGGTGTCGCAGG-MGB-3′. The sequence of the TaqMan probe for Renilla luciferase was: 5′-6-FAM-CCTCTTCTTATTTATGGCGACAT-MGB-3′. Firefly luciferase mRNA was normalized to Renilla luciferase content. Relative quantification of Firefly luciferase mRNA was performed by the comparative CT method (Livak method (Livak and Schmittgen, 2001)) and expressed as the percentage of Firefly luciferase mRNA measured in cells transfected with pGL3-PPOX + 1. 2.5. Electrophoretic mobility shift assay Electrophoretic mobility shift assay (EMSA) was performed as described previously (de Vooght et al., 2005) using nuclear extracts from K562, HeLa, and HepG2 cells. Extracts were prepared according to the method of Dignam et al. (1983). The following wild-type and mutant single-stranded primers (sense, mutations in bold), with their respective complementary antisense primers, were used to investigate DNA-protein interactions at both GATA sites in exon 1 of PPOX: GATA_AWTs, 5′-GCGAGGGCCGATAGCGAGGGT-3′; GATA_Amuts, 5′-GCGAGGGCCGTTAGCGAGGGT-3′; GATA_BWTs, 5′-GTGGCCCTTATCTGCACCCAG-3′; GATA_Bmuts 5′-GTGGCCCTTAACTGCACCCAG-3′.
87
For supershift assay, 1 μg of anti-GATA-1 (m-20 X) or antiNF-E2 (sc-16276 X) (Santa Cruz Biotechnology, Inc, Santa Cruz, CA, USA) was incubated with nuclear extract for 30 min at room temperature prior to the addition of the probe. 2.6. Chromatin immunoprecipitation assay To investigate in vivo binding of GATA-1 to the GATA-1 binding sites in exon 1 of PPOX we performed chromatin immunoprecipitation (ChIP) assay, by use of the ChIP-IT™ Express Enzymatic kit (Active Motif, Carlsbad, CA, USA) and the ChIP-validated antibody GATA-1 pAB (39025, Active Motif), according to the manufacturer's instructions. Briefly, K562 cells were fixed using formaldehyde and chromatin was sheared into small, uniform fragments by enzymatic digestion. Specific GATA-1/DNA complexes were immunoprecipitated using a ChIP-validated antibody directed against human GATA1. Normal rabbit IgG (sc-2027, Santa Cruz) was used as a control to confirm absence of non-specific IgG/DNA binding. Following immunoprecipitation, crosslinking was reversed, proteins removed by treatment with proteinase K and DNA recovered. 7.5 μL of immunoprecipitated DNA was PCR amplified using primers covering both GATA-1 binding sites in exon 1 of PPOX: primers PPOX ChIP forward 5′-GGAGTAGCGGATTTGAAGCA-3′ and PPOX ChIP reverse 5′TCACCCACAATAGGTAGGGAT-3′. The length of the amplified PPOX PCR product was 214 nts. 3. Results 3.1. Exon 1 directs erythroid-specific expression of PPOX To establish the functional relevance of exon 1 with regard to erythroid-specific expression of PPOX, we transfected PPOX promoter constructs without exon 1 (pGL3-PPOX) and with exon 1 (pGL3-PPOX + 1) into K562 cells. Luciferase activities were calculated relative to the pGL3 control vector. These results showed that the presence of exon 1 led to a fourfold increase in promoter activity (Fig. 2, panel A). To study the relevance of exon 1 with regard to housekeeping expression, transfections were performed in HepG2 and HeLa cells. Transfection of the promoter enhancer vectors in HepG2 and HeLa cells demonstrated that the presence of exon 1 did not affect promoter activity in HepG2 cells but reduced promoter activity by 50% in HeLa cells (Fig. 2, panel B). 3.2. Two GATA-1 binding sites in exon 1 mediate expression of PPOX on the transcriptional level Exon 1 of PPOX contains two putative GATA-1 binding sites. Since GATA-1 is a well-known modulator of erythroid-
Fig. 3. Two GATA-1 binding motifs in exon 1 mediate expression of PPOX in K562 cells on the transcriptional level. PPOX reporter plasmids containing wild-type and mutated GATA-1 binding motifs were transfected into K562 cells (panel A), HepG2 and HeLa cells (panel B). Luciferase activities were calculated relative to the PPOX promoter reporter vector with exon 1. TaqMan real-time RT-PCR was performed on Firefly and Renilla luciferase mRNA in K562 cells transfected with wildtype and mutant PPOX reporter plasmids (panel C). Quantification of Firefly luciferase mRNA was performed using the comparative CT method (Livak method (Livak and Schmittgen, 2001)) and expressed as percentage of wild-type + exon 1 (n = 3, samples in eightfold).
88
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
specific transcriptional regulation (Ferreira et al., 2005), we postulated these two binding sites to be of importance with respect to the observed increased promoter activity. The first GATA-1 motif (GATA-1A, CGATAG) constituted a nonconsensus sequence whereas the second motif (GATA-1B, AGATAA) constituted a full consensus sequence (WGATAR) (Merika and Orkin, 1993) for binding of GATA-1 (Fig. 1). To determine the functional importance of both GATA-1 sites, single and double mutant GATA-1 luciferase promoter reporter plasmids were generated and transfected into K562 cells. Mutation of the non-consensus GATA-1A site (CGATAG → CGTTAG, a mutation that has been shown to disrupt GATA-1 binding (Zon et al., 1991)) did not significantly alter promoter activity (Fig. 3, panel A). An identical mutation of the consensus GATA-1B site caused a 36% reduction in promoter activity. When both GATA-1 binding sites were mutated, promoter activity was reduced by 66%. The activity of this double mutant GATA-1 reporter construct (34% ± 4%) was comparable with the wild-type PPOX construct lacking exon 1 (23% ± 2%) (Fig. 3, panel A). Other putative GATA-1 binding sites in the PPOX promoter construct (at position − 304, − 300 and + 256) were found to be non-functional in K562 cells (Fig. 3, panel A). To confirm that the reduced promoter activity of the GATA-1 mutant PPOX promoter constructs was due to inhibition of transcription, we quantified luciferase mRNA levels. The amount of Firefly luciferase mRNA in cells transfected with wild-type and mutant PPOX promoter constructs (Fig. 3, panel
C) corresponded with the measured Firefly luciferase activity (Fig. 3, panel A) in these cells (mean ratio of luciferase activity versus luciferase transcripts was 0.98 (range: 0.88–1.10). Hence, the reduction in luciferase activity of the GATA-1 mutant PPOX promoter constructs is due to reduced transcription of Firefly luciferase. Subsequently we studied the effect of the two GATA-1 mutants on housekeeping expression of PPOX. Enhancer promoter reporter plasmids containing the mutant GATA-1 binding sites were transfected into the non-erythroid cell lines HepG2 and HeLa. Mutating either of the two GATA-1 binding motifs in exon 1, as well as mutating them simultaneously, did not affect promoter activity in both HepG2 and HeLa cells (Fig. 3, panel B). 3.3. GATA-1 binding sites in exon 1 direct erythroid-specific transcription of PPOX To verify the erythroid-specific character of both GATA-1 binding sites in exon 1 of PPOX, we repeated transfection experiments in K562 cells induced with hemin to differentiate into the erythroid direction. Monocytic U937 cells were used as (non-erythroid) control. As shown in Fig. 4, relative luciferase activities of the different promoter reporter vectors in induced K562 cells were similar to activities in uninduced K562 cells (Fig. 3). Exon 1 did not increase promoter activity in monocytic U937 cells (Fig. 4). These results confirm that the GATA-1 binding sites in exon 1 are essential for erythroid-specific
Fig. 4. GATA-1 binding sites in exon 1 direct erythroid-specific transcription of PPOX. PPOX reporter gene constructs were transiently transfected in K562 cells induced with hemin three days prior to transfection. Similar promoter constructs were transfected in U937 cells. Luciferase activities were calculated relative to the pGL3-SV40 (control) vector (n = 4, samples in duplicate).
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
expression of PPOX, and not involved in housekeeping expression of PPOX. 3.4. GATA-1 binds both GATA-1 binding sites in exon 1 of PPOX Further evidence for the role of GATA-1 in PPOX transcription came from EMSA studies. Both wild-type PPOX probes showed a specific complex with K562 nuclear extracts (Fig. 5A and B, lane 1). Formation of these complexes was effectively abolished by addition of excess unlabeled homologous oligonucleotide (competitor) (Fig. 5A and B, lane 2), but not by addition of excess mutant competitor (Fig. 5A and B,
89
lane 3). Addition of anti-human GATA-1 antibody caused a supershift of the specific oligonucleotide/GATA-1 complex (Fig. 5A and B, lane 4). This was a specific reaction, since no supershift occurred after addition of anti-human NF-E2 antibody (results not shown). In addition, EMSAs were carried out using labeled oligonucleotides harboring the respective GATA-1 mutations. These did not lead to formation of the specific oligonucleotide/GATA-1 complexes (Fig. 5A and B, lane 5). To evaluate in vivo binding of GATA-1 to exon 1 of PPOX we performed ChIP assays in K562 cells. Amplification of recovered sheared chromatin with primers annealing to exon 1 of PPOX revealed a 214 nts PCR product specific for PPOX
Fig. 5. GATA-1 binds to both GATA binding motifs in exon 1 of PPOX in vitro. Electrophoretic mobility shift assays were performed with K562 nuclear extract and labeled wild-type and mutant oligonucleotide probes. The absence (−) or presence (+) of 500 fmol of unlabeled competitors is indicated, as well as the addition of 1 μg of anti-GATA-1. The lowest arrow denotes the specific GATA-1-DNA complex, the upper arrow the GATA-1-DNA supershifted complex.
90
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
Fig. 6. GATA-1 binds to both GATA binding motifs in exon 1 of PPOX in vivo. Chromatin from K562 cells was immunoprecipitated with a ChIP-validated GATA-1 antibody (anti-GATA-1) and PCR amplified using primers flanking the two GATA-1 binding sites in exon 1 of PPOX (PCR product of 412 nts). Amplification of input chromatin (input) prior to immunoprecipitation served as positive control for chromatin extraction and PCR amplification. Chromatin immunoprecipitation using non-specific antibody (rabbit IgG) served as negative control. In addition, PCR in absence of DNA (DNA negative) was performed to exclude DNA contamination of the reaction mixture.
(Fig. 6). This product was absent when using normal rabbit IgG. From these results we concluded that GATA-1 directly interacts with exon 1 of PPOX in K562 cells. 4. Discussion In this study we describe a novel mechanism involved in the regulation of human protoporphyrinogen oxidase gene expression. We show that erythroid-specific regulation of PPOX depends on two GATA-1 sites in exon 1. The enzymes of the heme biosynthetic pathway serve as a model for differential regulation of housekeeping proteins. Heme is synthesized in red blood cells and in most other tissues where it serves as a cofactor for respiratory cytochromes. The enzymes involved in its biosynthesis are ubiquitously expressed. The genes encoding the second, third and fourth enzyme of the heme biosynthesis, δ-aminolevulinate dehydratase (ALAD), porphobilinogen deaminase (PBGD) and uroporphyrinogen III synthase (UROS), are each transcribed from two distinct promoters (Chretien et al., 1988; Bishop et al., 1996; Aizencang et al., 2000). In all of these genes the housekeeping promoter is located upstream of exon 1. The erythroid-specific promoters are located in intron 1, immediately upstream of exon 2, which is the first coding exon in all of these genes (Aizencang et al., 2000). Transcription of the genes that code for the fifth to eighth enzyme of the heme biosynthetic pathway (uroporphyrinogen decarboxylase (UROD), coproporphyrinogen oxidase (CPO), PPOX and ferrochelatase (FECH)) is regulated by a single promoter (Romana et al., 1987; Tugores et al., 1994; Taketani et al., 1995a; Takahashi et al., 1998). Until this report, no studies were performed with regard to (tissue-specific) regulation of the PPOX gene in erythroid and non-erythroid cells. Transfection experiments of wild-type and mutant reporter plasmids in uninduced and hemin induced K562 cells demonstrated that erythroid-specific transcriptional regulation of
PPOX was mediated by two GATA-1 binding sites in exon 1 (Figs. 3 and 4). In addition, both GATA-1 sites were able to bind GATA-1 in vitro and in vivo (Figs. 5 and 6). These results suggest therefore that both GATA-1 binding sites in exon 1 are involved in the high level of PPOX expression in erythroid cells. The importance of this regulation is supported by the observation that both GATA-1 binding sites are conserved in rat and mouse (data from Ensembl). The consensus GATAB binding site is fully conserved in rat and mouse, while the non-consensus GATAA binding site in exon 1 of the human PPOX gene is conserved except for the first nucleotide. Taketani et al. (1995a) suggested that putative GATA-1 binding sites in the promoter region might regulate PPOX expression during erythroid differentiation. In contrast, we show that these other putative GATA-1 binding sites (the consensus GATA binding motifs at position −304 and − 300, and the non-consensus GATA binding motif at position + 256), are non-functional (Fig. 3, panel A). 5′ Untranslated exons can regulate the expression of genes in two different ways. First, they may act as a tissue-specific translational regulator (Pickering and Willis, 2005). Second, the 5′UTR may regulate gene expression by mediating transcriptional regulation (Zimmermann et al., 2005). A pivotal role of the 5′UTR, and the transcription factors interacting with it, is described for several genes such as the γ-globin gene (HBG1) and the CC Chemokine Receptor 3 gene (CCR3) (Amrolia et al., 1995; Zimmermann et al., 2005). Our experiments show that the 5′UTR of human PPOX regulates gene expression on the level of transcription rather than translation (Fig. 3). More specifically, transcriptional regulation is mediated by binding of GATA-1 to two GATA-1 binding sites in the 5′UTR. There are only few examples of genes in which GATA-1 is known to interact with the 5′UTR (Amrolia et al., 1995; Wong et al., 2004; Zimmermann et al., 2005). In the human HBG1 gene GATA-1 is suggested to interact with the 5′UTR, causing transcriptional repression of γA-globin gene expression (Amrolia et al., 1995). GATA-1 also interacts with exon 1 of CCR3 and Zimmermann et al. suggest that GATA-1 has a regulatory role in CCR3 transcription (Zimmermann et al., 2005). GATA-1 binding in exon 1 of the α-spectrin gene (SPTA1) is also required for high-level expression of α-spectrin, as described by Wong et al. (2004). Our findings with regard to the importance of exon 1, and both GATA-1 sites in particular, for erythroid-specific expression of PPOX are in contrast with a recent report by Kotze and colleagues. They reported that the 5′UTR of the PPOX gene is particularly mutation prone, leading them to conclude that this region is probably not an important part of the gene (Kotze et al., 1998). Also, Dailey et al. (2002) showed that erythroidspecific transcription of the mouse PPOX gene required more upstream regulatory elements. However, the constructs used in this latter study lacked the part of exon 1 that contains four putative GATA-1 binding sites (+ 80 to + 288) (Schug and Overton, 1997). Hence, this may explain the observed lack of erythroid-specific promoter activity. In contrast to the key role of exon 1 in erythroid-specific expression of human PPOX, no effect of this exon was seen on
K.M.K. de Vooght et al. / Gene 409 (2008) 83–91
promoter activity in HepG2 or U937 cells (Figs. 2 and 4). Exon 1 even decreased promoter activity in HeLa cells by 50% (Fig. 2). The latter could be due to an effect on translation or due to the presence of repressor element(s) in exon 1. The occurrence of repressor elements in the 5′UTR of human genes has been reported in a number of human genes, for example in Pax5 (PAX5) (Rahman et al., 2001). Our results regarding expression in HeLa, HepG2 and U937 cells suggest that the two GATA-1 binding sites in exon 1 do not play a regulatory role other than in erythroid-specific expression (Figs. 3 and 4). Sp1 and/or GRα are candidate transcription factors that may repress PPOX promoter activity (Akerblom et al., 1988; Zaid et al., 2001). In conclusion, we demonstrate that erythroid-specific transcriptional regulation of the human PPOX gene is mediated by two GATA-1 binding sites in exon 1. These GATA-1 binding sites do not play a regulatory role in housekeeping expression. Our results contribute to a better understanding of the molecular mechanisms involved in transcriptional regulation of PPOX, and the transcriptional role of 5′UTR's in general. Acknowledgement The authors sincerely wish to thank Adri Thomas for helpful discussions. References Aizencang, G., Solis, C., Bishop, D.F., Warner, C., Desnick, R.J., 2000. Human uroporphyrinogen-III synthase: genomic organization, alternative promoters, and erythroid-specific expression. Genomics 70, 223–231. Akerblom, I.E., Slater, E.P., Beato, M., Baxter, J.D., Mellon, P.L., 1988. Negative regulation by glucocorticoids through interference with a cAMP responsive enhancer. Science 241, 350–353. Amrolia, P.J., Cunningham, J.M., Ney, P., Nienhuis, A.W., Jane, S.M., 1995. Identification of two novel regulatory elements within the 5′-untranslated region of the human Aγ -globin gene. J. Biol. Chem. 270, 12892–12898. Bishop, T.R., Miller, M.W., Beall, J., Zon, L.I., Dierks, P., 1996. Genetic regulation of δ-aminolevulinate dehydratase during erythropoiesis. Nucleic Acids Res. 24, 2511–2518. Braman, J., Papworth, C., Greener, A., 1996. Site-directed mutagenesis using double-stranded plasmid DNA templates. Methods Mol. Biol. 57, 31–44. Chen, L.S., Tassone, F., Sahota, P., Hagerman, P.J., 2003. The (CGG)n repeat element within the 5′ untranslated region of the FMR1 message provides both positive and negative cis effects on in vivo translation of a downstream reporter. Hum. Mol. Genet. 12, 3067–3074. Chretien, S., et al., 1988. Alternative transcription and splicing of the human porphobilinogen deaminase gene result either in tissue-specific or in housekeeping expression. Proc. Natl. Acad. Sci. U. S. A. 85, 6–10. Conder, L.H., Woodard, S.I., Dailey, H.A., 1991. Multiple mechanisms for the regulation of haem synthesis during erythroid cell differentiation. Possible role for coproporphyrinogen oxidase. Biochem. J. 275, 321–326. Dailey, T.A., Dailey, H.A., 1996. Human protoporphyrinogen oxidase: expression, purification, and characterization of the cloned enzyme. Protein Sci. 5, 98–105. Dailey, T.A., McManus, J.F., Dailey, H.A., 2002. Characterization of the mouse protoporphyrinogen oxidase gene. Cell. Mol. Biol. 48, 61–69. de Vooght, K.M.K., van Wijk, R., van Oirschot, B.A., Rijksen, G., van Solinge, W.W., 2005. Pyruvate kinase regulatory element 1 (PKR-RE1) mediates hexokinase gene expression in K562 cells. Blood Cells Mol. Diseases 34, 186–190.
91
Dignam, J.D., Lebovitz, R.M., Roeder, R.G., 1983. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 11, 1475–1489. Ferreira, R., Ohneda, K., Yamamoto, M., Philipsen, S., 2005. GATA1 function, a paradigm for transcription factors in hematopoiesis. Mol. Cell. Biol. 25, 1215–1227. Fibach, E., Kollia, P., Schechter, A.N., Noguchi, C.T., Rodgers, G.P., 1995. Hemin-induced acceleration of hemoglobin production in immature cultured erythroid cells: preferential enhancement of fetal hemoglobin. Blood 85, 2967–2974. Fujita, H., et al., 1991a. Sequential activation of genes for heme pathway enzymes during erythroid differentiation of mouse Friend virus-transformed erythroleukemia cells. Biochim. Biophys. Acta 1090, 311–316. Fujita, H., Yamamoto, M., Yamagami, T., Hayashi, N., Sassa, S., 1991b. Erythroleukemia differentiation. Distinctive responses of the erythroidspecific and the nonspecific δ-aminolevulinate synthase mRNA. J. Biol. Chem. 266, 17494–17502. Kotze, M.J., et al., 1998. Molecular analysis reveals a high mutation frequency in the first untranslated exon of the PPOX gene and largely excludes variegate porphyria in a subset of clinically affected Afrikaner families. Mol. Cell. Probes 12, 293–300. Lake-Bullock, H., Dailey, H.A., 1993. Biphasic ordered induction of heme synthesis in differentiating murine erythroleukemia cells: role of erythroid 5aminolevulinate synthase. Mol. Cell. Biol. 13, 7122–7132. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2− ΔΔCT method. Methods 25, 402–408. Merika, M., Orkin, S.H., 1993. DNA-binding specificity of GATA family transcription factors. Mol. Cell. Biol. 13, 3999–4010. Pickering, B.M., Willis, A.E., 2005. The implications of structured 5′ untranslated regions on translation and disease. Semin. Cell Dev. Biol. 16, 39–47. Rahman, M., et al., 2001. A repressor element in the 5′-untranslated region of human Pax5 exon 1A. Gene 263, 59–66. Romana, M., Dubart, A., Beaupain, D., Chabret, C., Goossens, M., Romeo, P.H., 1987. Structure of the gene for human uroporphyrinogen decarboxylase. Nucleic Acids Res. 15, 7343–7356. Sassa, S., 1976. Sequential induction of heme pathway enzymes during erythroid differentiation of mouse Friend leukemia virus-infected cells. J. Exp. Med. 143, 305–315. Schug, J., Overton, G.C., 1997. TESS: Transcription Element Search Software on the WWW. Computational Biology and Informatics Laboratory, School of Medicine. University of Pennsylvania, Pennsylvania. Takahashi, S., et al., 1998. Differential regulation of coproporphyrinogen oxidase gene between erythroid and nonerythroid cells. Blood 92, 3436–3444. Taketani, S., et al., 1995a. The human protoporphyrinogen oxidase gene (PPOX): organization and location to chromosome 1. Genomics 29, 698–703. Taketani, S., et al., 1995b. Induction of terminal enzymes for heme biosynthesis during differentiation of mouse erythroleukemia cells. Eur. J. Biochem. 230, 760–765. Tugores, A., Magness, S.T., Brenner, D.A., 1994. A single promoter directs both housekeeping and erythroid preferential expression of the human ferrochelatase gene. J. Biol. Chem. 269, 30789–30797. Wong, E.Y., Lin, J., Forget, B.G., Bodine, D.M., Gallagher, P.G., 2004. Sequences downstream of the erythroid promoter are required for high level expression of the human α-spectrin gene. J. Biol. Chem. 279, 55024–55033. Zaid, A., Hodny, Z., Li, R., Nelson, B.D., 2001. Sp1 acts as a repressor of the human adenine nucleotide translocase-2 (ANT2) promoter. Eur. J. Biochem. 268, 5497–5503. Zimmermann, N., Colyer, J.L., Koch, L.E., Rothenberg, M.E., 2005. Analysis of the CCR3 promoter reveals a regulatory region in exon 1 that binds GATA-1. BMC Immunol. 6, 7. Zon, L.I., Youssoufian, H., Mather, C., Lodish, H.F., Orkin, S.H., 1991. Activation of the erythropoietin receptor promoter by transcription factor GATA-1. Proc. Natl. Acad. Sci. U. S. A. 88, 10638–10641.