Portal de Periódicos da CAPES

Global analysis of induced transcription factors and cofactors identifies Tfdp2 as an essential coregulator during terminal erythropoiesis

2014; Elsevier BV; Volume: 42; Issue: 6 Linguagem: Inglês

10.1016/j.exphem.2014.03.001

1873-2399

Cynthia Chen, Harvey F. Lodish,

Epigenetics and DNA Methylation

Key transcriptional regulators of terminal erythropoiesis, such as GATA-binding factor 1 (GATA1) and T-cell acute lymphocytic leukemia protein 1 (TAL1), have been well characterized, but transcription factors and cofactors and their expression modulations have not yet been explored on a global scale. Here, we use global gene expression analysis to identify 28 transcription factors and 19 transcriptional cofactors induced during terminal erythroid differentiation whose promoters are enriched for binding by GATA1 and TAL1. Utilizing protein–protein interaction databases to identify cofactors for each transcription factor, we pinpoint several co-induced pairs, of which E2f2 and its cofactor transcription factor Dp-2 (Tfdp2) were the most highly induced. TFDP2 is a critical cofactor required for proper cell cycle control and gene expression. GATA1 and TAL1 are bound to the regulatory regions of Tfdp2 and upregulate its expression and knockdown of Tfdp2 results in significantly reduced rates of proliferation as well as reduced upregulation of many erythroid-important genes. Loss of Tfdp2 also globally inhibits the normal downregulation of many E2F2 target genes, including those that regulate the cell cycle, causing cells to accumulate in S phase and resulting in increased erythrocyte size. Our findings highlight the importance of TFDP2 in coupling the erythroid cell cycle with terminal differentiation and validate this study as a resource for future work on elucidating the role of diverse transcription factors and coregulators in erythropoiesis. Key transcriptional regulators of terminal erythropoiesis, such as GATA-binding factor 1 (GATA1) and T-cell acute lymphocytic leukemia protein 1 (TAL1), have been well characterized, but transcription factors and cofactors and their expression modulations have not yet been explored on a global scale. Here, we use global gene expression analysis to identify 28 transcription factors and 19 transcriptional cofactors induced during terminal erythroid differentiation whose promoters are enriched for binding by GATA1 and TAL1. Utilizing protein–protein interaction databases to identify cofactors for each transcription factor, we pinpoint several co-induced pairs, of which E2f2 and its cofactor transcription factor Dp-2 (Tfdp2) were the most highly induced. TFDP2 is a critical cofactor required for proper cell cycle control and gene expression. GATA1 and TAL1 are bound to the regulatory regions of Tfdp2 and upregulate its expression and knockdown of Tfdp2 results in significantly reduced rates of proliferation as well as reduced upregulation of many erythroid-important genes. Loss of Tfdp2 also globally inhibits the normal downregulation of many E2F2 target genes, including those that regulate the cell cycle, causing cells to accumulate in S phase and resulting in increased erythrocyte size. Our findings highlight the importance of TFDP2 in coupling the erythroid cell cycle with terminal differentiation and validate this study as a resource for future work on elucidating the role of diverse transcription factors and coregulators in erythropoiesis. The adult human generates roughly 2.4 million red blood cells every second, a process that requires the intricately regulated proliferation and differentiation of hematopoietic stem cells into mature erythrocytes. Much of this regulation occurs during terminal erythroid differentiation, the highly coordinated final step that begins with the committed erythroid progenitor colony-forming unit erythroid (CFU-E), and is characterized by several cellular phenomena, including heme biosynthesis, 3–5 cell divisions followed by permanent cell cycle exit, chromatin condensation, and enucleation [1Cantor A.B. Orkin S.H. Transcriptional regulation of erythropoiesis: an affair involving multiple partners.Oncogene. 2002; 21: 3368-3376Crossref PubMed Scopus (483) Google Scholar, 2Hattangadi S.M. Wong P. Zhang L. Flygare J. Lodish H.F. From stem cell to red cell: regulation of erythropoiesis at multiple levels by multiple proteins, RNAs, and chromatin modifications.Blood. 2011; 118: 6258-6268Crossref PubMed Scopus (309) Google Scholar, 3Zhang L. Sankaran V.G. Lodish H.F. MicroRNAs in erythroid and megakaryocytic differentiation and megakaryocyte-erythroid progenitor lineage commitment.Leukemia. 2012; 26: 2310-2316Crossref PubMed Scopus (49) Google Scholar]. Close regulation at the mRNA level of erythroid genes is essential for proper erythroid maturation, and global studies of the changing transcriptional landscape have yielded insight into gene regulatory networks during terminal erythropoiesis [4Wong P. Hattangadi S.M. Cheng A.W. Frampton G.M. Young R.A. Lodish H.F. Gene induction and repression during terminal erythropoiesis are mediated by distinct epigenetic changes.Blood. 2011; 118: e128-e138Crossref PubMed Scopus (90) Google Scholar, 5Merryweather-Clarke A. Global gene expression analysis of human erythroid progenitors.Blood. 2011; 117: e96-e108Crossref PubMed Scopus (90) Google Scholar]. The role of key transcription factors and coregulators, notably GATA1 and its cofactor friend of GATA protein 1 (FOG1) (Zfpm1) [6Welch J. Watts J. Vakoc C. Global regulation of erythroid gene expression by transcription factor GATA-1.Blood. 2004; 104: 3136-3147Crossref PubMed Scopus (333) Google Scholar, 7Johnson K. Boyer M. Friend of GATA-1–independent transcriptional repression: a novel mode of GATA-1 function.Blood. 2007; 109: 5230-5233Crossref PubMed Scopus (53) Google Scholar], in controlling gene expression have also been well characterized [8Pilon A.M. Ajay S.S. Kumar S.A. et al.Genome-wide ChIP-Seq reveals a dramatic shift in the binding of the transcription factor erythroid Kruppel-like factor during erythrocyte differentiation.Blood. 2011; 118: e139-e148Crossref PubMed Scopus (78) Google Scholar, 9Kassouf M.T. Hughes J.R. Taylor S. et al.Genome-wide identification of TAL1's functional targets: insights into its mechanisms of action in primary erythroid cells.Genome Res. 2010; 20: 1064-1083Crossref PubMed Scopus (136) Google Scholar], although a comprehensive view of all transcriptional regulators is lacking. To this end, we analyzed expression levels of 276 transcription factors and 213 transcriptional coregulators during distinct stages of erythroid differentiation [4Wong P. Hattangadi S.M. Cheng A.W. Frampton G.M. Young R.A. Lodish H.F. Gene induction and repression during terminal erythropoiesis are mediated by distinct epigenetic changes.Blood. 2011; 118: e128-e138Crossref PubMed Scopus (90) Google Scholar] and identified 28 induced transcription factors and 19 induced cofactors, the promoters for the vast majority of which are themselves bound by factors GATA1 and TAL1. Protein interaction analysis [10Szklarczyk D. Franceschini A. Kuhn M. et al.The STRING database in 2011: functional interaction networks of proteins, globally integrated and scored.Nucleic Acids Res. 2011; 39: D561-D568Crossref PubMed Scopus (2678) Google Scholar] was then used to identify cofactors for each induced transcription factor. Several interacting partners were found to be jointly upregulated and bound by GATA1 and TAL1 during terminal erythropoiesis, including the known essential erythroid factors Tal1, Lmo2, and Gata1, each with Fog1, validating the use of this global study as a resource for finding potential critical transcriptional regulators. As further validation of this study, we functionally characterized the role of cofactor Tfdp2 whose interaction with E2f2 constituted the top candidate pair. The E2F family of transcription factors is essential for proper cell cycle control in many cell types and has also been shown to play critical roles in many differentiation pathways [11Korenjak M. Brehm A. E2F-Rb complexes regulating transcription of genes important for differentiation and development.Curr Opin Genet Dev. 2005; 15: 520-527Crossref PubMed Scopus (130) Google Scholar, 12Hernando E. Nahlé Z. Juan G. et al.Rb inactivation promotes genomic instability by uncoupling cell cycle progression from mitotic control.Nature. 2004; 430: 797-802Crossref PubMed Scopus (457) Google Scholar, 13Nahle Z. Polakoff J. Davuluri R.V. et al.Direct coupling of the cell cycle and cell death machinery by E2F.Nat Cell Biol. 2002; 4: 859-864Crossref PubMed Scopus (363) Google Scholar]. Specifically, loss of E2F family member E2F2 results in defective erythroid maturation, anemia, larger red blood cell (RBC) size, arrest in S phase, and other hematopoietic defects [14Li F.X. Zhu J.W. Hogan C.J. Degregori J. Defective gene expression, S phase progression, and maturation during hematopoiesis in E2F1/E2F2 mutant mice.Mol Cell Biol. 2003; 23: 3607-3622Crossref PubMed Scopus (73) Google Scholar, 15Dirlam A. Spike B.T. Macleod K.F. Deregulated E2f-2 underlies cell cycle and maturation defects in retinoblastoma null erythroblasts.Mol Cell Biol. 2007; 27: 8713-8728Crossref PubMed Scopus (49) Google Scholar]. Usually, E2Fs are bound to DNA in a heterodimer with one of three functional DNA-binding cofactors: TFDP1, 2, or 3 [16La Thangue N.B. DP and E2F proteins: components of a heterodimeric transcription factor implicated in cell cycle control.Curr Opin Cell Biol. 1994; 6: 443-450Crossref PubMed Scopus (141) Google Scholar, 17Rogers K.T. Higgins P.D. Milla M.M. Phillips R.S. Horowitz J.M. DP-2, a heterodimeric partner of E2F: identification and characterization of DP-2 proteins expressed in vivo.Proc Natl Acad Sci U S A. 1996; 93: 7594-7599Crossref PubMed Scopus (37) Google Scholar, 18Qiao H. Di Stefano L. Tian C. et al.Human TFDP3, a novel DP protein, inhibits DNA binding and transactivation by E2F.J Biol Chem. 2007; 282: 454-466Crossref PubMed Scopus (31) Google Scholar, 19Duronio R.J. Bonnette P.C. Farrell P.H.O. Mutations of the Drosophila dDP, dE2F, and cyclin E genes reveal distinct roles for the E2F-DP transcription factor and cyclin E during the G1-S transition.Mol Cell Biol. 1998; 18: 141-151Crossref PubMed Scopus (89) Google Scholar, 20Frolov M.V. Moon N. Dyson N.J. dDP is needed for normal cell proliferation.Mol Cell Biol. 2005; 25: 3027-3039Crossref PubMed Scopus (35) Google Scholar]; however, the specific role of these coregulators in E2F-mediated cell cycle control during terminal erythropoiesis has not been explored. Here, we use a primary mouse erythroid cell culture system to show that loss of TFDP2 results in ineffective erythropoiesis. GATA1 and TAL1 bind to regulatory regions of Tfdp2 and E2f2 genes to upregulate their expression levels. In contrast, expression levels of known E2F2 target genes decrease during terminal erythropoiesis, suggesting that E2F2 acts as a transcriptional repressor in terminally dividing erythroblasts. Consistent with this hypothesis, knockdown of Tfdp2 results in higher than normal levels of these cell cycle genes, causing cells to stall in S phase and fail to mature. Guided by our bioinformatics study, these findings suggest a novel model by which cells can coordinate their cell cycles with differentiation and serve as a roadmap for future functional and mechanistic studies of transcriptional regulators in erythropoiesis. Protein–protein interactions for a list of all expressed transcription factors and cofactors, defined by gene ontology [21The Gene Ontology ConsortiumGene ontology: tool for the unification of biology.Nat Genet. 2000; 25: 25-29Crossref PubMed Scopus (26624) Google Scholar] as “sequence-specific DNA binding transcription factor activity” and “transcription cofactor activity,” were obtained from the STRING database [10Szklarczyk D. Franceschini A. Kuhn M. et al.The STRING database in 2011: functional interaction networks of proteins, globally integrated and scored.Nucleic Acids Res. 2011; 39: D561-D568Crossref PubMed Scopus (2678) Google Scholar] containing known and predicted physical and functional protein associations. Interactions were filtered to remove those between two transcription factors or two cofactors. Publicly available chromatin-immunoprecipitation sequencing (ChIP-seq) databases for histone 3 lysine 9 acetylation (H3K9ac) in mouse erythroleukemia cells (GEO accession no. GSM1000141), H3K27ac in mouse E14.5 fetal liver (GSM1000113), and histone 3 lysine 4 monomethylation (H3K4me1) (GSM946536), histone 3 lysine 4 trimethylation (H3K4me3) (GSM946524), GATA1 (GSM923575), and TAL1 (GSM923582) in E14.5 Ter119 + mouse erythroblasts were analyzed on the UCSC Mouse Genome Browser (http://genome.ucsc.edu/), mm9 assembly using the model-based analysis of ChIP-seq (MACS) [22Zhang Y. Liu T. Meyer C.A. et al.Model-based analysis of ChIP-Seq (MACS).Genome Biol. 2008; 9: R137Crossref PubMed Scopus (8348) Google Scholar]. We used 293T cells as the retrovirus-packaging cell line and maintained them in Dulbecco's Modified Eagle Medium with 10% fetal bovine serum (FBS), 2 mmol/L L-glutamine, and 1% penicillin/streptomycin (P/S). For MCF-7 cells, 293T culture medium was supplemented with 10 μg/mL human insulin (Sigma, USA). The shRNA sequences targeting mouse Tfdp2 were obtained from the Broad Institute RNAi consortium shRNA library (Cambridge, MA). We then cloned shRNA sequences into the BbsI sites of the murine stem cell virus (MSCV)-pkgGFP-U3-U6P vector, which co-expresses GFP from a phosphoglycerate kinase (PGK) promoter. The following are the shRNA sequences: shTFDP2a, AaaaCCCTGTTCATTCAACGATGAAgtcgacTTCATCGTTGAATGAACAGGG; shTFDP2b, AaaaCCACAGGACCTTCTTGGTTAAgtcgacTTAACCAAGAAGGTCCTGTGG. An shRNA against the firefly luciferase gene was also cloned into the same vector as a positive control. For the luciferase reporter assay, putative promoter and enhancer regions of Tfdp2 (chr9:96158706-96159362 and chr9:96167467-96168140) were amplified from mouse genomic DNA and cloned into the pGL3-Basic luciferase reporter vector (Promega, USA). The XZ-GATA1-IRES-GFP and XZ-TAL1-IRES-GFP constructs were made by cloning the ORF of Gata1 or Tal1 into the XZ vector. Twenty-four hours before transfection, MCF-7 cells were seeded into 96 well plates at a density of 50,000–100,000 cells per well. For transfection of MCF-7 cells in each well, Lipofectamine 2000 (Invitrogen, USA) was used to cotransfect 10 ng pGL3-Basic luciferase reporter containing either the Tfdp2 putative promoter or enhancer regions detailed above, or a control empty vector plasmid, together with 90 ng each of XZ-GATA1 and XZ-TAL1 into MCF-7 cells, followed by culture for 48 hours. Luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega, USA). To obtain erythropoietin-dependent erythroid progenitors of high purity, homogenized mouse E14.5 fetal liver cells were labeled with the mixture of biotin-conjugated antibodies in the Lin-negative Kit (BD Pharmingen, USA) and subsequently incubated with Streptavidin Particles (BD Pharmingen, USA). Magnetic-based negative selection was then used to deplete mature erythrocytes and nonerythroid cells [23Zhang L. Flygare J. Wong P. Lim B. Lodish H.F. miR-191 regulates mouse erythroblast enucleation by down-regulating Riok3 and Mxi1.Genes Dev. 2011; 25: 119-124Crossref PubMed Scopus (88) Google Scholar]. Standard retroviral infection of purified erythroid progenitors with shRNA constructs was then performed, and the cells were cultured in erythropoietin (EPO)-containing differentiation medium at 37°C until experimental analysis [24Sankaran V.G. Ludwig L.S. Sicinska E. et al.Cyclin D3 coordinates the cell cycle during differentiation to regulate erythrocyte size and number.Genes Dev. 2012; 26: 2075-2087Crossref PubMed Scopus (80) Google Scholar]. For analysis of enucleation, standard immunostaining followed by flow cytometry analysis was performed [23Zhang L. Flygare J. Wong P. Lim B. Lodish H.F. miR-191 regulates mouse erythroblast enucleation by down-regulating Riok3 and Mxi1.Genes Dev. 2011; 25: 119-124Crossref PubMed Scopus (88) Google Scholar]. Enucleated cells were identified as Ter119-positive, Hoechst-negative. For analysis of apoptosis, standard immunostaining using the apoptosis detection kit (BD Pharmingen, USA) was performed according to the manufacturer's protocol [25Hattangadi S.M. Burke K.A. Lodish H.F. Homeodomain-interacting protein kinase 2 plays an important role in normal terminal erythroid differentiation.Blood. 2010; 115: 4853-4861Crossref PubMed Scopus (40) Google Scholar]. Cells undergoing apoptosis were identified as PE annexin V-positive and 7-aminoactinomycin D (7-AAD)-negative. For flow cytometry sorting of the R1 to R5 stages of late erythroid cells, homogenized mouse E14.5 fetal liver cells were stained with 1 μg/mL propidium iodide (PI), 1:100 APC-conjugated Ter119, 1:300 PE-conjugated CD71 and were sorted into 5 fractions (R1–R5) [26Zhang J. Socolovsky M. Gross A.W. Lodish H.F. Role of Ras signaling in erythroid differentiation of mouse fetal liver cells: functional analysis by a flow cytometry-based novel culture system.Blood. 2003; 102: 3938-3946Crossref PubMed Scopus (321) Google Scholar]. For flow cytometry sorting of GFP + cells, in vitro cultured erythroid cells were stained with PI (1 μg/mL). We sorted GFP-positive/PI-negative cells. For flow cytometry sorting of enucleated cells, in vitro cultured erythroid cells were stained with PI (1 μg/mL), 1:100 APC-conjugated Ter119, and Hoechst (1 μg/mL). We sorted GFP-positive/PI-negative/Ter119-positive/Hoechst-negative cells. Approximately 200,000 erythroid cells were centrifuged at 300 rpm for 4 min onto polylysine-coated slides in a cytocentrifugation apparatus. Standard May-Grünwald Giemsa staining was then performed [26Zhang J. Socolovsky M. Gross A.W. Lodish H.F. Role of Ras signaling in erythroid differentiation of mouse fetal liver cells: functional analysis by a flow cytometry-based novel culture system.Blood. 2003; 102: 3938-3946Crossref PubMed Scopus (321) Google Scholar]. Images were taken under a fluorescence microscope (40X objective; Zeiss AxioPlan2 upright microscope). Quantification of cell areas was performed using Image J software. In vitro cultured erythroid cells at indicated time points were pulsed with 10 μM bromodeoxyuridine (BrdU) for 30 min using a BrdU flow kit (BD Pharmingen, USA) according to the manufacturer's protocol. Pulsed cells were fixed and permeabilized, treated with DNase to expose incorporated BrdU epitopes, and stained with APC-conjugated anti-BrdU antibodies and 7-AAD for total DNA content. Cells were analyzed by flow cytometry [24Sankaran V.G. Ludwig L.S. Sicinska E. et al.Cyclin D3 coordinates the cell cycle during differentiation to regulate erythrocyte size and number.Genes Dev. 2012; 26: 2075-2087Crossref PubMed Scopus (80) Google Scholar]. Lineage-negative mouse fetal erythroid progenitor cells were infected with Tfdp2 shRNA and Luc control shRNA viruses. Immediately after infection, the number of seeded mouse fetal liver cells was determined using a hematocytometer. After 12 hours of infection, the GFP + percentages were measured by flow cytometry analysis and multiplied by the number of cells seeded; this was taken to be the day 0 cell number. The number of cells after 24 and 48 hours of culture were counted using a hematocytometer. We measured GFP + percentages for those time points by flow cytometry analysis. Erythroid cells were collected at specified time points using QIAzol (Qiagen, USA) and total cellular RNA was isolated using the miRNeasy minikit (Qiagen, USA) according to the manufacturer's protocol. Reverse transcription was carried out using SuperScript II Reverse Transcriptase (Invitrogen, USA). Relative transcript levels of different genes were quantified by real-time PCR, using SYBR Green (Applied Biosystems, USA) and the ABI 7900 Machine Real-Time PCR system (Applied Biosystems, USA). Results were normalized to 18S (criteria for differences in 18S Ct values for control and experimental sample set at <1.5). Primer sequences are found in Supplementary Table E1 (online only, available at www.exphem.org). Microarray experiments were performed and processed using the Affymetrix Mouse Genome 430 2.0 array at the Whitehead Institute Genome Core. Pairwise comparisons of experimental results were performed. Unpaired one-tailed Student's t tests for experimental data with corresponding biological replicates were used to determine the statistical significance of data. We considered p values <0.05 to be significant. Transcription factors and their cofactors play important roles in regulating terminal erythroid differentiation. To obtain a global view of these transcriptional regulators in this process, we analyzed the expression pattern of 276 transcription factors and 213 cofactors defined by gene ontology [21The Gene Ontology ConsortiumGene ontology: tool for the unification of biology.Nat Genet. 2000; 25: 25-29Crossref PubMed Scopus (26624) Google Scholar] as “sequence-specific DNA binding transcription factor activity” and “transcription cofactor activity,” respectively. Based on RNA-sequencing data on R1–R5 populations, which represent consecutive stages of in vivo erythropoiesis from CFU-Es to mature reticulocytes, the majority of transcription factors and cofactors are downregulated during terminal differentiation, while only a small number are upregulated (Fig. 1, A; Supplementary Figure E1, online only, available at www.exphem.org). Several upregulated factors, such as Sox6 and Zfpm1, have previously been shown to be indispensable for erythropoiesis [27Tsang A.P. Fujiwara Y. Hom D.B. Orkin S.H. Failure of megakaryopoiesis and arrested erythropoiesis in mice lacking the GATA-1 transcriptional cofactor FOG.Genes Dev. 1998; 12: 1176-1188Crossref PubMed Scopus (295) Google Scholar, 28Cantù C. Ierardi R. Alborelli I. et al.Sox6 enhances erythroid differentiation in human erythroid progenitors.Blood. 2011; 117: 3669-3679Crossref PubMed Scopus (47) Google Scholar]; however, the majority of these induced factors remain to be characterized. Many of these induced factors may also be targeted by erythroid-important transcription factors, specifically GATA1 and TAL1 [29Gutiérrez L. Tsukamoto S. Suzuki M. et al.Ablation of Gata1 in adult mice results in aplastic crisis, revealing its essential role in steady-state and stress erythropoiesis.Blood. 2008; 111: 4375-4385Crossref PubMed Scopus (74) Google Scholar, 30Hall M.A. Curtis D.J. Metcalf D. et al.The critical regulator of embryonic hematopoiesis, SCL, is vital in the adult for megakaryopoiesis, erythropoiesis, and lineage choice in CFU-S12.Proc Natl Acad Sci U S A. 2003; 100: 992-997Crossref PubMed Scopus (178) Google Scholar]. By mining existing ChIP-seq data for the promoter, exon, and intron regions of all induced and repressed factors, we found that the regulatory regions of 18 out of 28 (64%) genes encoding induced transcription factors and 12 out of 19 (63%) induced cofactors are cobound by GATA1 and TAL1 in mouse erythroblasts, compared with only 37 out of 226 (16%) repressed transcription factors and 41 out of 180 (22%) repressed cofactors (Fig. 1, B and C). Co-occupancy of gene regulatory regions by GATA1 and TAL1 for transcription factors and cofactors is therefore significantly associated with gene induction, consistent with previous studies [31Tripic T. Deng W. Cheng Y. et al.SCL and associated proteins distinguish active from repressive GATA transcription factor complexes.Blood. 2009; 113: 2191-2201Crossref PubMed Scopus (144) Google Scholar]. These ChIP-seq findings, together with analysis of gene expression patterns, serve as a resource to provide promising candidates for further functional analysis of their roles in erythroid transcriptional regulation. Since transcription factors and their cofactors normally function together to regulate biological processes, we further probed the expression correlation of paired interacting transcription factors and cofactors using the STRING protein–protein interaction database [10Szklarczyk D. Franceschini A. Kuhn M. et al.The STRING database in 2011: functional interaction networks of proteins, globally integrated and scored.Nucleic Acids Res. 2011; 39: D561-D568Crossref PubMed Scopus (2678) Google Scholar] and pinpointed several pairs as jointly upregulated (Fig. 2, A). Our analysis yielded Zfpm1 (Fog1) and its previously identified interacting transcription factors Tal1, Lmo2, and Gata1 [1Cantor A.B. Orkin S.H. Transcriptional regulation of erythropoiesis: an affair involving multiple partners.Oncogene. 2002; 21: 3368-3376Crossref PubMed Scopus (483) Google Scholar, 32Ferreira R. Ohneda K. Yamamoto M. Philipsen S. GATA1 Function, a Paradigm for Transcription Factors in Hematopoiesis.Mol Cell Biol. 2005; 25: 1215-1227Crossref PubMed Scopus (308) Google Scholar, 33Wilkinson-White L. Gamsjaeger R. Dastmalchi S. et al.Structural basis of simultaneous recruitment of the transcriptional regulators LMO2 and FOG1/ZFPM1 by the transcription factor GATA1.Proc Natl Acad Sci U S A. 2011; 108: 14443-14448Crossref PubMed Scopus (34) Google Scholar] as coinduced pairs, thus validating our approach. Regulatory regions of genes encoding many of these coinduced factors are also bound by GATA1 and TAL1, and binding peaks are shown for both the known Tal1 and Zfpm1 pair (Fig. 2, B), and a novel candidate pair, Ddit3 and Trib3 (Fig. 2, C). Our global analysis revealed transcription factor E2F2 and cofactor TFDP2 as the most highly induced pair (Fig. 2, A). Since TFDP2 has never been linked with erythropoiesis, we aimed to functionally characterize the potential role of TFDP2 in regulating erythroid differentiation. Based on RNA-sequencing gene expression data [4Wong P. Hattangadi S.M. Cheng A.W. Frampton G.M. Young R.A. Lodish H.F. Gene induction and repression during terminal erythropoiesis are mediated by distinct epigenetic changes.Blood. 2011; 118: e128-e138Crossref PubMed Scopus (90) Google Scholar], expression of both Tfdp2 and E2f2 mRNA were highly upregulated in vivo and followed similar patterns, with the greatest induction occurring at the R2 to R3 transition (Fig. 3, A). The expression patterns for Tfdp2 and E2f2 were confirmed by quantitative RT-PCR (Fig. 3, B). In contrast, expression of Tfdp1, the first TFDP family member identified, decreased approximately twofold from R2 to R3 and over fivefold from R3 to R4 (Fig. 3, A); there was no detectable expression of Tfdp3 at any stage. The significant upregulation of Tfdp2 expression, as opposed to the downregulation of Tfdp1 expression, suggests that TFDP2 plays an important role in regulating terminal erythropoiesis. In parallel with our bioinformatics analysis, we utilized publicly available ChIP-seq databases for GATA1 and TAL1 and found two regions of co-occupancy in both E2f2 and Tfdp2 genes (Fig. 3, C and D). We also analyzed the pattern of histone modifications, using H3K9ac as a marker of active chromatin, H3K4me1 as a marker of enhancer elements, and H3K4me3 as a marker of active promoter regions. The two peaks within the E2f2 gene (peak 1 and peak 2) are likely in enhancer regions, as they are found in introns and in H3K4me1-positive regions (Fig. 3, C; Supplementary Figure E2A, online only, available at www.exphem.org). Peak 1 in the first intron of E2f2 has been shown in previous studies as a region that is occupied by GATA1 and that likely modulates E2f2 expression [34Tallack M.R. Keys J.R. Humbert P.O. Perkins A.C. EKLF/KLF1 controls cell cycle entry via direct regulation of E2f2.J Biol Chem. 2009; 284: 20966-20974Crossref PubMed Scopus (57) Google Scholar]. In the region surrounding the Tfdp2 gene, we discovered a single peak (peak 3) present in a possible promoter region that is 5′ of the transcriptional start site, H3K4me3-positive, and H3K4me1-negative. A second peak (peak 4) lies within the first intron of Tfdp2, possibly representing an enhancer element (Fig. 3, D; Supplementary Figure E2B, online only, available at www.exphem.org). To determine whether GATA1 and TAL1 binding is functionally significant for increased expression of Tfdp2, we performed luciferase reporter assays on the two peaks in the Tfdp2 gene. Both the regions in the putative promoter of Tfdp2, peak 3, as well as in the first intron, peak 4, were responsive to GATA1 and TAL1, with an approximately threefold increase in relative luciferase activity after addition of GATA1 and TAL1 (Fig. 3, E). To characterize the role of Tfdp2 in terminal erythroid differentiation, we used in vitro culture of primary mouse fetal liver (FL) erythroid cells and abrogated Tfdp2 expression through retroviral-expressed shRNA hairpins. Both shRNAs (shTFDP2a and shTFDP2b) significantly reduced expression of Tfdp2 mRNA in FL cells after both 1 and 2 days of culture (Fig. 4, A). TFDP2 is critical for proper induction of erythroid-important genes, as knockdown reduced the expression of the α and β hemoglobin chains (Hbb-b1 and Hba-a1), GATA1, the erythrocyte membrane protein Epb4.1, enzymes required for heme biosynthesis, and other markers of late-stage differentiation (Fig. 4, B). Consistent with impaired differentiation, TFDP2 knockdown cells also contained abnormally high mRNA levels of the transcription factor and proto-oncogene c-Myc (Fig. 4, B), whose normal downregulation is essential for terminal erythropoiesis [35Jayapal S.R. Lee K.L. Ji P. Kaldis P. Lim B. Lodish H.F. Down-regulation of Myc is essential for terminal erythroid maturation.J Biol Chem. 2010; 285: 40252-40265Crossref PubMed Scopus (54) Google Scholar]. Given the reduced expression of key erythroid-important genes, we expected that subsequent enucleation also would be inhibited. We therefore analyzed FL cells by flow cytometry after 24, 48, and 72 hours of culture. Both shRNA constructs reduced the percentage of enucleated cells (Ter119-positive, Hoechst-negative), and quantification of enucleation rates in three biological replicates for all time points revealed that both shRNA constructs reduced enucleation to le