scholarly journals Erythroid Maturation and Globin Gene Expression in Mice With Combined Deficiency of NF-E2 and Nrf-2

Blood ◽  
1998 ◽  
Vol 91 (9) ◽  
pp. 3459-3466 ◽  
Author(s):  
Florence Martin ◽  
Jan M. van Deursen ◽  
Ramesh A. Shivdasani ◽  
Carl W. Jackson ◽  
Amber G. Troutman ◽  
...  

NF-E2 binding sites, located in distant regulatory sequences, may be important for high level α- and β-globin gene expression. Surprisingly, targeted disruption of each subunit of NF-E2 has either little or no effect on erythroid maturation in mice. For p18 NF-E2, this lack of effect is due, at least in part, to the presence of redundant proteins. For p45 NF-E2, one possibility is that NF-E2–related factors, Nrf-1 or Nrf-2, activate globin gene expression in the absence of NF-E2. To test this hypothesis for Nrf-2, we disrupted the Nrf-2 gene by homologous recombination. Nrf-2–deficient mice had no detectable hematopoietic defect. In addition, no evidence was found for reciprocal upregulation of NF-E2 or Nrf-2 protein in fetal liver cells deficient for either factor. Fetal liver cells deficient for both NF-E2 and Nrf-2 expressed normal levels of α- and β-globin. Mature mice with combined deficiency of NF-E2 and Nrf-2 did not exhibit a defect in erythroid maturation beyond that seen with loss of NF-E2 alone. Thus, the presence of a mild erythroid defect in NF-E2–deficient mice is not the result of compensation by Nrf-2.

Blood ◽  
1998 ◽  
Vol 91 (9) ◽  
pp. 3459-3466 ◽  
Author(s):  
Florence Martin ◽  
Jan M. van Deursen ◽  
Ramesh A. Shivdasani ◽  
Carl W. Jackson ◽  
Amber G. Troutman ◽  
...  

Abstract NF-E2 binding sites, located in distant regulatory sequences, may be important for high level α- and β-globin gene expression. Surprisingly, targeted disruption of each subunit of NF-E2 has either little or no effect on erythroid maturation in mice. For p18 NF-E2, this lack of effect is due, at least in part, to the presence of redundant proteins. For p45 NF-E2, one possibility is that NF-E2–related factors, Nrf-1 or Nrf-2, activate globin gene expression in the absence of NF-E2. To test this hypothesis for Nrf-2, we disrupted the Nrf-2 gene by homologous recombination. Nrf-2–deficient mice had no detectable hematopoietic defect. In addition, no evidence was found for reciprocal upregulation of NF-E2 or Nrf-2 protein in fetal liver cells deficient for either factor. Fetal liver cells deficient for both NF-E2 and Nrf-2 expressed normal levels of α- and β-globin. Mature mice with combined deficiency of NF-E2 and Nrf-2 did not exhibit a defect in erythroid maturation beyond that seen with loss of NF-E2 alone. Thus, the presence of a mild erythroid defect in NF-E2–deficient mice is not the result of compensation by Nrf-2.


1989 ◽  
Vol 83 (3) ◽  
pp. 1032-1038 ◽  
Author(s):  
T J Ley ◽  
K A Maloney ◽  
J I Gordon ◽  
A L Schwartz

Blood ◽  
2004 ◽  
Vol 104 (11) ◽  
pp. 1602-1602
Author(s):  
Patrick G. Gallagher ◽  
Andre M. Pilon ◽  
Murat O. Arcasoy ◽  
David M. Bodine

Abstract EKLF is the founding member of the KLF family of transcription factors. EKLF-deficient (−/−) mice die at d14-15 of gestation of severe anemia, attributed to decreased expression of the β-globin gene. Morphology of fetal-liver derived circulating erythroid cells in EKLF −/− mice does not mimic that seen in thalassemia, and mice expressing high levels of γ-globin bred onto an EKLF −/− background do not correct the anemia or rescue the −/− mice. These data suggest additional red cell defects are present in EKLF −/− mice contributing to the anemia. To address this hypothesis, we performed subtractive hybridization and microarray analyses with RNA isolated from d13.5 fetal livers of wild type (WT,+/+) and EKLF-deficient mice. In subtractive hybridization, WT (1X) was the tester population and EKLF −/− (30X) was the driver. >200 differentially expressed clones were sequenced. 122 clones were β-globin; 21 alpha hemoglobin stabilizing protein (AHSP); and the remainder were RBC membrane proteins including band 3 and β-spectrin. Microarray analyses were performed with Affymetrix GeneChip Mouse Genome 430 2.0 arrays; 3 independent EKLF +/+ and −/− RNA samples were analyzed. AHSP had the most significantly decreased expression in −/− samples, reduced to 5% of WT (p<0.0001). Other significantly down regulated genes, in addition to those identified by subtractive hybridization, included pyruvate kinase and ankyrin. Reduced expression in EKLF −/− RNA was confirmed using ribonuclease protection assay and/or real-time RT-PCR. AHSP mRNA was decreased by 75–90%; band 3 and β-spectrin were decreased by ~40%; ankyrin by 45%, and pyruvate kinase 15%. Flow cytometry of +/+ and −/ − fetal liver cells using TER119 revealed a TER119Hi population of cells absent in −/ − fetal liver cells, suggesting a block in differentiation to more mature erythroid progenitors. To ensure that the potential EKLF target genes were expressed in the TER119Lo population and were not identified because they were present only in the TER119Hi, the expression of each selected gene was analyzed in TER119Lo and TER119Hi cells by real time PCR. Target gene/α-globin ratios indicated the selected genes were expressed at levels >2 fold higher in TER119Lo than TER119Hi cells. Studies of the AHSP locus identified a strong DNaseI hypersensitive site (HS) in WT fetal liver nuclei between 5′ of the AHSP promoter, that was absent in −/ − chromatin. Chromatin immunoprecipitation analysis of WT fetal liver chromatin spanning 3.5kb of the AHSP locus using 13 primer pairs (~300bp intervals from >1kb 5′ and 3′) demonstrated 2 regions of hyperacetylation of histones H3 and H4. The 5′ region corresponded to the DNaseI HS, and the other region maps 3′ of the AHSP polyA signal. Histones H3 and H4 were also acetylated in the interval between these hyperacetylated regions, while the chromatin upstream and downstream of these regions were hypoacetylated. In chromatin from −/ − fetal liver cells, all sites were hypoacetylated, correlating with the severe reduction in AHSP gene expression. These results support the hypothesis that the anemia in EKLF −/ − mice is due to both decreased expression of the β-globin gene and other erythroid genes including those involved in membrane integrity, stabilization of α-globin protein, and glycolysis leading to defects in erythrocyte structure, function, and metabolism. They also suggest that EKLF may act as a transcription factor and a chromatin modulator for genes other than β-globin.


2007 ◽  
Vol 30 (11) ◽  
pp. 2091-2097 ◽  
Author(s):  
Masataka Maruyama ◽  
Tamihide Matsunaga ◽  
Eri Harada ◽  
Shigeru Ohmori

Blood ◽  
2019 ◽  
Vol 134 (Supplement_1) ◽  
pp. 812-812
Author(s):  
Mudit Chaand ◽  
Chris Fiore ◽  
Brian T Johnston ◽  
Diane H Moon ◽  
John P Carulli ◽  
...  

Human beta-like globin gene expression is developmentally regulated. Erythroblasts (EBs) derived from fetal tissues, such as umbilical cord blood (CB), primarily express gamma globin mRNA (HBG) and HbF, while EBs derived from adult tissues, such as bone marrow (BM), predominantly express beta globin mRNA (HBB) and adult hemoglobin. Human genetics has validated de-repression of HBG in adult EBs as a powerful therapeutic paradigm in diseases involving defective HBB, such as sickle cell anemia. To identify novel factors involved in the switch from HBG to HBB expression, and to better understand the global regulatory networks driving the fetal and adult cell states, we performed transcriptome profiling (RNA-seq) and chromatin accessibility profiling (ATAC-seq) on sorted EB cell populations from CB or BM. This approach improves upon previous studies that used unsorted cells (Huang J, Dev Cell 2016) or that did not measure chromatin accessibility (Yan H, Am J Hematol 2018). CD34+ cells from CB and BM were differentiated using a 3-phase in vitro culture system (Giarratana M, Blood 2011). Fluorescence-activated cell sorting and the cell surface markers CD36 and GYPA were used to isolate 7 discrete populations, with each sorting gate representing increasingly mature, stage-matched EBs from CB or BM (Fig 1A, B). RNA-seq analysis revealed expected expression patterns of the beta-like globins, with total levels increasing during erythroid maturation and primarily composed of HBB or HBG transcripts in BM or CB, respectively (Fig 1C). Erythroid maturation led to progressive increases in chromatin accessibility at the HBB promoter in BM populations. In CB-derived cells, erythroid maturation led to progressive increases in chromatin accessibility at the HBG promoters through the CD36+GYPA+ stage (Pops 1-5). Chromatin accessibility shifted from the HBG promoters to the HBB promoter during the final stages of differentiation (Pops 6-7), suggesting that HBG gene activation is transient in CB EBs (Fig 1D). Hierarchical clustering and principal component analysis of ATAC-seq data revealed that cell populations cluster based on differentiation stage rather than by BM or CB lineage, suggesting most molecular changes are stage-specific, not lineage-specific (Fig 2A, B). To identify transcription factors driving cell state, and potentially beta-like globin expression preference, we searched for DNA binding motifs within regions of differential chromatin accessibility and found NFI factor motifs enriched under peaks that were larger in BM relative to CB (Fig 2C). Transcription factor footprinting analysis showed that both flanking accessibility and footprint depth at NFI motifs were also increased in BM relative to CB (Fig 2D). Increased chromatin accessibility was observed at the NFIX promoter in BM relative to CB populations, and in HUDEP-2 relative to HUDEP-1 cell lines (Fig 2E). Furthermore, accessibility at the NFIX promoter correlated with elevated NFIX mRNA in BM and HUDEP-2 relative to CB and HUDEP-1, respectively. Together these data implicated NFIX in HbF repression, a finding consistent with previous genome-wide association and DNA methylation studies that suggested a possible role for NFIX in regulating beta-like globin gene expression (Fabrice D, Nat Genet 2016; Lessard S, Genome Med 2015). To directly test the hypothesis that NFIX represses HbF, short hairpin RNAs were used to knockdown (KD) NFIX in primary erythroblasts derived from human CD34+ BM cells (Fig 3A). NFIX KD led to a time-dependent induction of HBG mRNA, HbF, and F-cells comparable to KD of the known HbF repressor BCL11A (Fig 3B-D). A similar effect on HbF was observed in HUDEP-2 cells following NFIX KD (Fig 3E). Consistent with HbF induction, NFIX KD also increased chromatin accessibility and decreased DNA methylation at the HBG promoters in primary EBs (Fig 3F, G). NFIX KD led to a delay in erythroid differentiation as measured by CD36 and GYPA expression (Fig 3H). Despite this delay, by day 14 a high proportion of fully enucleated erythroblasts was observed, suggesting NFIX KD cells are capable of terminal differentiation (Fig 3H). Collectively, these data have enabled identification and validation of NFIX as a novel repressor of HbF, a finding that enhances the understanding of beta-like globin gene regulation and has potential implications in the development of therapeutics for sickle cell disease. Disclosures Chaand: Syros Pharmaceuticals: Employment, Equity Ownership. Fiore:Syros Pharmaceuticals: Employment, Equity Ownership. Johnston:Syros Pharmaceuticals: Employment, Equity Ownership. Moon:Syros Pharmaceuticals: Employment, Equity Ownership. Carulli:Syros Pharmaceuticals: Employment, Equity Ownership. Shearstone:Syros Pharmaceuticals: Employment, Equity Ownership.


2012 ◽  
Vol 2012 ◽  
pp. 1-5 ◽  
Author(s):  
Supachai Ekwattanakit ◽  
Yuwarat Monteerarat ◽  
Suchada Riolueang ◽  
Kalaya Tachavanich ◽  
Vip Viprakasit

Background and Objectives. To explore the role ofcis-regulatory sequences within theβglobin gene cluster at chromosome 11 on humanγglobin gene expression related to Hb E allele, we analyze baseline hematological data and Hb F values together withβglobin haplotypes in homozygous Hb E.Patients and Methods. 80 individuals with molecularly confirmed homozygous Hb E were analyzed for theβglobin haplotypes andXmnI polymorphism using PCR-RFLPs. 74 individuals with complete laboratory data were further studied for association analyses.Results. Eight differentβglobin haplotypes were found linked to Hb E alleles; three major haplotypes were (a) (III), (b) (V), and (c) (IV) accounting for 94% of Hb E chromosomes. A new haplotype (Th-1) was identified and most likely converted from the major ones. The majority of individuals had Hb F < 5%; only 10.8% of homozygous Hb E had high Hb F (average 10.5%, range 5.8–14.3%). No association was found on a specific haplotype orXmnI in these individuals with high Hb F, measured by alkaline denaturation. Conclusion. The cis-regulation ofγglobin gene expression might not be apparent under a milder condition with lesser globin imbalance such as homozygous Hb E.


Blood ◽  
2004 ◽  
Vol 104 (11) ◽  
pp. 1210-1210
Author(s):  
Lauren Sterner ◽  
Toru Miyazaki ◽  
Larry Swift ◽  
Ann Dean ◽  
Jane Little

Abstract We examined the effects of short chain fatty acids (SCFAs) on globin gene expression during development. We studied globin gene expression in transgenic mice that have endogenous elevations in the SCFA propionate due to a knockout (KO) of the gene for propionyl CoA carboxylase subunit A (PCCA, Miyazaki et al. JBC, 2001 Sep 21;276(38):35995–9). Serum propionate levels measured by gas chromatography were 2.5 to 3.6 mgms/ml in 2 adult PCCA KO mice and were undetectable in 2 wild type (wt) or heterozygous control adult mice. Embryonic PCCA KO offspring had propionate levels of 2.3 and 5.0 μgms/100 mgms of fetal liver, at day 16.5 (E16.5), while wt or heterozygotes at E14.5 had levels &lt;1 μgm/100 mgms. Analysis of expression from alpha (α), beta major (βmaj), embryonic beta-type epsilon-y (εy), embryonic beta-type beta H1 (βH1) and embryonic alpha-type zeta (ζ) globin genes plus 18S ribosomal RNA as a control was undertaken using real-time PCR with gene-specific primers and taqman probes. cDNA was reverse-transcribed from the mRNA of yolk sac (YS) and fetal liver of PCCA KO and wt progeny of more than one litter from timed pregnancies. Individual PCCA embryos at E10 (n=10), E12 (n=9), and E14 (n=7) were analyzed for globin gene expression, normalized to18S expression and were compared to age-matched wt embryos (n&gt;=4 for each time point). As expected, embryonic alpha- and beta-type globin gene expression (ζ and βH1 plus εy) predominated in E 10 YS, and definitive globin gene expression, α and βmaj, predominated in E12 or E14 fetal liver. Expression from embryonic alpha-type globin was calculated as normalized ζ/(ζ+α) and from embryonic beta-type globins as normalized (βH1+εy)/(βH1+εy+βmaj), see table. Embryonic globin gene expression was statistically significantly increased in PCCA KO E12 YS at 1.3 fold relative to wt ζ and in PCCA KO E14 YS at 1.8 fold and 2.1 fold relative to wt ζ or βH1 and εy respectively (p&lt;.05). No increase in embryonic globin mRNA was seen in adult PCCA KO animals. We conclude that elevations of SCFAs during normal murine development causes a persistence of both embryonic alpha-type and embryonic beta-type globin gene expression during primitive, but not definitive, erythropoiesis, suggesting that SCFAs cannot reactivate silenced murine embryonic globin genes in the absence of erythroid stress. Embryonic Globin Gene Expression in Mice with Endogenous Elevations of SCFAs % Expression PCCA KO wild type p value, t test E10 ζ Yolk Sac 53+/− 2 nd E10 βH1 & ε y Yolk Sac 99 +/− 0.3 nd E12 ζ Yolk Sac 32 +/− 3 25 +/− 1 p &lt; .05 E12 βH1 & ε y Yolk Sac 77 +/− 6 74 +/− 3 ns E14 ζ Yolk Sac 7 +/− 1.5 4 +/− 1.4 p &lt; .05 E14 βH1 & ε y Yolk Sac 13 +/− 6 6 +/− 0.5 p &lt; .05 E12 ζ Fetal Liver 11 +/− 4 9 +/− 2 ns E12 βH1 & ε y Fetal Liver 13 +/− 5 13+/− 3 ns E14 ζ Fetal Liver 1 +/− 0.4 0.7 +/− 0.2 ns E14 βH1 & εy Fetal Liver 6 +/− 1.8 4 +/− 1 ns


Blood ◽  
2013 ◽  
Vol 122 (21) ◽  
pp. 1020-1020
Author(s):  
Kenneth R Peterson ◽  
Zhen Zhang ◽  
Ee Phie Tan ◽  
Anish Potnis ◽  
Nathan Bushue ◽  
...  

Abstract Patients with sickle cell disease (SCD), caused by mutation of the adult β-globin gene, are phenotypically normal if they carry compensatory mutations that result in continued expression of the fetal γ-globin genes, a condition termed hereditary persistence of fetal hemoglobin (HPFH). Thus, a logical clinical goal for treatment of SCD is to up-regulate γ-globin synthesis using compounds that are specific for increasing fetal hemoglobin (HbF) without pleiotropic effects on cellular homeostasis. Developmental regulation of the γ-globin genes is complex and normal silencing during the adult stage of erythropoiesis likely results from a combination of the loss of transcriptional activators and the gain of transcriptional repressor complexes. One mode of γ-globin silencing occurs at the GATA binding sites located at -566 or -567 relative to the Aγ-globin or Gγ-globin CAP sites respectively, and is mediated through the DNA binding moiety of GATA-1 and its recruitment of co-repressor partners, FOG-1 and Mi-2 (NuRD complex). Modifications of repressor complexes can regulate gene transcription; one such modification is O-GlcNAcylation. The O-GlcNAc post-translational modification is the attachment of a single N-acetyl-glucosamine moiety to either a serine or threonine residue on nuclear and cytoplasmic proteins. O-GlcNAc is added to proteins by O-GlcNAc transferase (OGT) and removed by O-GlcNAcase (OGA) in response to changes in extracellular signals and nutrients. A dynamic balance in protein levels also exists between these two enzymes; an increase or decrease of one results in a like compensatory change in the other. Thus, the rate of O-GlcNAc addition and removal is a dynamic cycling event that is exquisitely controlled for a given target molecule, which may offer a point of intervention in the turning off or on of gene expression. O-GlcNAcylation is involved in the regulation of many cellular processes such as stress response, cell cycle progression, and transcription. Potentially, O-GlcNAc plays a pivotal role in regulating transcription of the human γ-globin genes. We induced human erythroleukemia cell line K562 with sodium butyrate to differentiate toward the erythroid lineage and observed the expected increase of γ-globin gene expression. A robust increase of γ-globin gene expression was measured after pharmacological inhibition of OGA using Thiamet-G (TMG). Using chromatin immunoprecipitation (ChIP), we demonstrated that OGT and OGA are recruited to the -566 region of the Aγ-globin promoter, the same region occupied by the GATA-1-FOG-1-Mi-2 (NuRD) repressor complex. However, OGT recruitment to this region was decreased when O-GlcNAc levels were artificially elevated by OGA inhibition with TMG. When γ-globin expression was not induced, Mi-2 was modified with O-GlcNAc and interacted with both OGT and OGA. After induction, O-GlcNAcylation of Mi-2 was reduced and Mi2 no longer interacted with OGT. Stable K562 cells were generated in which OGA was knocked down using shRNA. Following induction of these cells with sodium butyrate, γ-globin gene expression was higher compared to control cells. These data suggest that the dynamic cycling of O-GlcNAc on the Mi-2 (NuRD) moiety contributes towards regulation of γ-globin transcription. Concurrent ChIP experiments in human β-globin locus yeast artificial chromosome (β-YAC) transgenic mice demonstrated that GATA-1, Mi2 and OGT were recruited to the -566 Aγ-globin GATA silencer site in day E18 fetal liver when γ-globin is repressed, but not in day E12 fetal liver when γ-globin is expressed. These data demonstrate that O-GlcNAc cycling is a novel mechanism regulating γ-globin gene expression and will provide new avenues to explore in how alterations in gene regulation lead to the onset, progression, and severity of hematological disease. Disclosures: No relevant conflicts of interest to declare.


Blood ◽  
1991 ◽  
Vol 78 (6) ◽  
pp. 1589-1595
Author(s):  
L Romao ◽  
L Osorio-Almeida ◽  
DR Higgs ◽  
J Lavinha ◽  
SA Liebhaber

We describe an alpha-thalassemia determinant in which alpha-globin expression is silenced by a deletion located 27 kb 5′ to the transcription start site of the alpha 2-globin gene. This alpha- thalassemic determinant, (alpha alpha)MM, is a member of a newly described group of thalassemic mutations resulting from deletion of locus-controlling sequences critical to globin gene expression.


Sign in / Sign up

Export Citation Format

Share Document