HS3 Core Research Articles

Talen-Mediated Knock Outs Of Cis and Trans Elements Potentially Involved In Globin Gene Switching

Transcription activator-like effector nucleases (TALEN) are engineered proteins used for precise genome editing by generating specific DNA double strand that are repaired by homologous recombination and by non-homologous end joining. TALENs can be used to study gene regulation by deleting putative regulatory elements in the context of the native chromosome and measuring mRNA synthesis. We designed TALENs to delete individual DNAse I-hypersensitive sites (HS) of the β-globin locus control region (LCR) followed by an assessment of globin gene expression and assessment of epigenetic effects in K562 erythroleukemia cells. The β-globin LCR is composed of five HSs and functions as a powerful regulatory element responsible for appropriate levels of the five β-like globin genes during development. Introduction of plasmid DNA encoding a pair of TALENs and targeting individually the flanking region of the HS2, HS3 and HS4 core elements along with a donor 100 base single-stranded oligonucleotide resulted in the successful deletions of each of the three core elements in K562 cells. Individual K562 cells were seeded to produce clones and the mutations were screened by PCR to identify both heterozygous and homozygous clones.The TALEN-mediated 288 bp HS2 core deletion resulted 32 heterozygous (48.5%) and 6 homozygous clones (9.1%) in a total of 66 clones screened. K562 carries three copies of chromosome 11 emphasizing the robustness of TALEN technology to target each of the alleles. In the 199 bp HS3 core deletion, from 113 clones we identified 28 heterozygous (24.8%) and 3 (2.7%) homozygous clones. Lastly, the 301 bp HS4 core deletion yielded 9 homozygous (5.9%) and 12 heterozygous (7.9%) clones from 151 clones screened. Total RNA was isolated from wild-type K562 cells, and from both the heterozygous and homozygous mutant clones and subjected to RNase Protection analysis to quantitate the levels of globin mRNA. Deletion if the HS3 core in K562 cells in a ∼30% reduction in ε-globin mRNA and 2-fold reduction in γ-globin mRNA. A more dramatic effect on globin expression is observed in the HS2 core deletion, as ε- and γ-globin expression is reduced by 2- and 5-fold, respectively. These results suggest that HS2 contributes the majority of the LCR enhancer function in K562 cells. The HS4 core deletion resulted in a modest ∼20% reduction in both ε- and γ-globin expression.TALENs were designed to knockout trans-acting factors implicated to be involved in globin gene regulation and/or globin switching. TALENs bracketing the gene promoters and the first exon of 25 genes encoding either a transcription factor or histone-modifying enzyme were synthesized and post-transfection PCR screens of the transfected pool of K562 cells resulted in the successful identification of 17 gene knockouts. The 17 target genes are PRMT5, LDB1, EIF2AK3, BCL11A, HBSIL, MYB, SOX6, NFE4, NR2F2, NR2C1, NR2C2, CHTOP, NFE2, DNMT3A, RBBP4, MTA2 and MBD2. Single cell clones have been generated by limited dilution of transfected K562 pools and thus far we have identified heterozygous and homozygous clones of 8 of 17 gene knockouts, importantly all clones were identified without selection. The frequency of identifying the knockout clones, represented by the number of clones screened/ number of heterozygous clones/ number of homozygous clones, are as follows: HBS1L (63/3/0), SOX6 (68/13/2), NFE4 (56/13/7), LBD1 (300/2/0), MBD2 (301/0/1), CHTOP (288/66/6), NFE2 (712/44/5) and NR2C1 (96/40/11). The remaining nine gene knockouts and globin gene expression data will be presented at the meetings.These studies highlight a powerful TALEN-mutagenesis platform for target deletions of both cis- and trans-elements to study globin gene switching. TALENs can be synthesized in several days and the screening of the individual clones for the desired knockouts is completed within two weeks. This highly efficient mutagenesis platform will further our understanding of the molecular basis of globin switching. Disclosures:No relevant conflicts of interest to declare.

LCR 5′ hypersensitive site specificity for globin gene activation within the active chromatin hub

The DNaseI hypersensitive sites (HSs) of the human β-globin locus control region (LCR) may function as part of an LCR holocomplex within a larger active chromatin hub (ACH). Differential activation of the globin genes during development may be controlled in part by preferential interaction of each gene with specific individual HSs during globin gene switching, a change in conformation of the LCR holocomplex, or both. To distinguish between these possibilities, human β-globin locus yeast artificial chromosome (β-YAC) lines were produced in which the ε-globin gene was replaced with a second marked β-globin gene (βm), coupled to an intact LCR, a 5′HS3 complete deletion (5′ΔHS3) or a 5′HS3 core deletion (5′ΔHS3c). The 5′ΔHS3c mice expressed βm-globin throughout development; γ-globin was co-expressed in the embryonic yolk sac, but not in the fetal liver; and wild-type β-globin was co-expressed in adult mice. Although the 5′HS3 core was not required for βm-globin expression, previous work showed that the 5′HS3 core is necessary for ε-globin expression during embryonic erythropoiesis. A similar phenotype was observed in 5′HS complete deletion mice, except βm-globin expression was higher during primitive erythropoiesis and γ-globin expression continued into fetal definitive erythropoiesis. These data support a site specificity model of LCR HS-globin gene interaction.

Chicken HS4 Insulators Have Minimal Functions in Human Hematopoietic Cells That Were Transduced with An HIV1-Based Lentiviral Vector.

Abstract 3570Poster Board III-507For the hemoglobin disorders, hematopoietic stem cell gene transfer is potentially curative, yet this strategy requires high-level β-globin gene expression among erythroid cells. Position effects, which are imparted by chromosomal position and chromatin structure, induce clonal variability of transgene expression. Recent work demonstrates that the chicken HS4 insulator element reduces position effects, resulting in consistently high-level expression of a therapeutic β-globin gene in the MEL cell line. In this study, we evaluated the effects of HS4 insulators on lentiviral vector titers and transgene expression among transduced human hematopoietic cells.We constructed various types of insulated lentiviral vectors using a reverse oriented GFP under the control of the MSCV-LTR promoter (rMpGFP) or a conventional reverse oriented β-globin expression cassette, in which the globin gene was changed to GFP (BGpGFP). A full-length HS4 insulator (1.2 kb HS4), tandem HS4 core insulator (2 × 250 b HS4), and a single core insulator (250 b HS4) were inserted into the 3′ LTR. The insulator elements were inserted in both forward (F) and reverse (R) orientations.Vector titers were significantly decreased by insertion of the 1.2 kb HS4 and 2 × 250 b HS4 in both orientations and both vector constructs, compared to uninsulated vectors (p<0.05), with the degree dependent on fragment size. Interestingly, reverse-oriented insulators showed better vector titers when compared to forward-oriented insulators for all types of insulator fragments except the 2 × 250 b HS4 in rMpGFP vectors (p<0.05).We next evaluated GFP expression from various insulated rMpGFP vectors in GPA+ human erythroid cells that originated from transduced CD34+ cells (MOI=3) (Figure). The %GFP was decreased by 1.2 kb HS4 and 2 × 250 b HS4 insulators in both orientations, compared to the uninsulated vector (p<0.05). All insulated vector constructs had a tendency to lower CVs, there was no significant difference except for the 1.2 kb HS4 F vector (p<0.05). There was no significant difference of MFIs between all types of insulated and uninsulated vectors. In order to evaluate insulator function for the BGpGFP vectors in human hematopoietic cells, we practically chose the 250 b HS4 R because it did not decrease vector titers and the 1.2 kb HS4 showed 5-fold lower transduction efficiency in human erythroid cells. During erythorid culture of transduced human erythroid cells, %GFP and MFIs decreased whereas CVs increased,showing chromosomal position effects. The 250 b HS4 R insulator showed lower %GFP and lower MFIs (MOI=20) (p<0.05 on day 13 and 20), compared to those of the uninsulated vector. There was no significant difference in CVs. After MOI escalation of BGpGFP vectors (day13), the insulated vector showed lower %GFP at MOI 10, 25, and 50 (p<0.05) and lower overall GFP expression (%GFP x MFI) at MOI 25 and 50 (p<0.05) compared to uninsulated vector. These data demonstrated that inclusion of HS4 insulator elements decreases GFP expression, which is not overcome by increasing MOI. We then performed transduced hematopoietic stem cell transplantation in a human xenograft mouse model using a 250 b HS4 R insulated rMpGFP vector. In the human CD45+ fraction of mouse peripheral blood cells, the insulator element decreased both %GFP and MFIs at 4 and 8 weeks after transplantation (p<0.05). There was no significant difference of CVs among the insulated and uninsulated vector at all time points.These data demonstrate that the inclusion of HS4 insulator elements lowers viral titers, reduces efficiency of transduction and produces minimal effects on transgene expression among human hematopoietic cells in vitro and in vivo [Display omitted] Disclosures:No relevant conflicts of interest to declare.

Ikaros Drives Human Haemoglobin Switching by Facilitating Active Chromatin Hub Formation.

The human β globin locus consists of an upstream locus control region (LCR) and five functional genes arranged sequentially in the order of their expression during development: 5′-ε-Gγ-Aγ- δ- β-3′. Haemoglobin switching entails the successive recruitment of these genes into an active chromatin hub (ACH). Although much is known about the cis elements and transcription factors involved in globin gene regulation, less is known about ACH formation. Here we show that the transcription factor Ikaros plays an essential role in both the formation of the β-globin ACH, and in haemoglobin switching. In Plastic mice, where the DNA-binding region of Ikaros is disrupted by a point mutation (H191R), there is concomitant marked (10 fold) down-regulation of human β-globin, and up-regulation of γ-globin gene expression. We show Ikaros binds to a critical cis elements in the LCR near the HS3 core and upstream of the δ-globin gene in the β-globin locus by electormobility shift assay (EMSA) and chromatin immunoprecipitation (ChIP) and that this DNA binding activity is lost in Plast mice. This latter site is implicated in deletional hereditary persistence of fetal haemoglobin (HPFH). Furthermore, chromatin conformation capture (3C) data suggest Ikaros facilitates long range looping between the LCR and a region upstream of the δ-globin gene. This study provides new insights into the mechanism of adult stage-specific assembly of the β-globin ACH. In addition the findings could lead to the development of novel drugs to reactivate HbF in adults with β-thalassemia and sickle cell disease.

Erythroid-Specific Expression of β-Globin by the Sleeping Beauty Transposon for Sickle Cell Disease

Sickle cell disease (SCD) results predominately from a single monogenic mutation that affects thousands of individuals worldwide. Gene therapy approaches have focused on using viral vectors to transfer wild-type beta- or gamma-globin transgenes into hematopoietic stem cells for long-term expression of the recombinant globins. In this study, we investigated the use of a novel nonviral vector system, the Sleeping Beauty (SB) transposon (Tn) to insert a wild-type beta-globin expression cassette into the human genome for sustained expression of beta-globin. We initially constructed a beta-globin expression vector composed of the hybrid cytomegalovirus (CMV) enhancer chicken beta-actin promoter (CAGGS) and full-length beta-globin cDNA, as well as truncated forms lacking either the 3' or 3' and 5' untranslated regions (UTRs), to optimize expression of beta-globin. Beta-globin with its 5' UTR was efficiently expressed from its cDNA in K-562 cells induced with hemin. However, expression was constitutive and not erythroid-specific. We then constructed cis SB-Tn-beta-globin plasmids using a minimal beta-globin gene driven by hybrid promoter IHK (human ALAS2 intron 8 erythroid-specific enhancer, HS40 core element from human alphaLCR, ankyrin-1 promoter), IHbetap (human ALAS2 intron 8 erythroid-specific enhancer, HS40 core element from human alphaLCR, beta-globin promoter), or HS3betap (HS3 core element from human betaLCR, beta-globin promoter) to establish erythroid-specific expression of beta-globin. Stable genomic insertion of the minimal gene and expression of the beta-globin transgene for >5 months at a level comparable to that of the endogenous gamma-globin gene were achieved using a SB-Tn beta-globin cis construct. Interestingly, erythroid-specific expression of beta-globin driven by IHK was regulated primarily at the translational level, in contrast to post-transcriptional regulation in non-erythroid cells. The SB-Tn system is a promising nonviral vector for efficient genomic insertion conferring stable, persistent erythroid-specific expression of beta-globin.

Human β-globin locus control region hypersensitive site specificity for globin gene activation during erythropoiesis

Although the human β-globin locus control region (LCR) functions as a holocomplex within an active chromatin hub, we provide evidence that within the aggregate hypersensitive site (HS) activation domain of the holocomplex, the individual HSs still mediate preferential activation of the globin genes during development. A 2.9 Kb deletion of 5′HS3 (Δ5′HS3) or a 234 bp deletion of the 5′HS3 core (Δ5′HS3c) in a 213 Kb human β-globin locus yeast artificial chromosome (β-YAC) abrogate e-globin gene expression during primitive erythropoiesis in β-YAC transgenic mice, suggesting that 5′HS3 sequences of the LCR are involved directly in e-globin gene activation. The reduction of e-globin gene transcription in Δ5′HS3 or Δ5′HS3c β-YAC transgenics can be explained by two hypotheses. The first is site-specificity. The interaction between the LCR and the e-globin gene promoter involves specific sequences of 5′HS3 and specific sequences of the e-globin gene promoter. When 5′HS3 or its core is deleted, these interactions do not take place and e-globin gene transcription is diminished. The second hypothesis is change in conformation of the LCR. Normally, in the embryonic stage, the LCR achieves a three-dimensional conformation that favors interaction with the first gene in the complex, the e-globin gene. When 5′HS3 is deleted, an alternate conformation is assumed that decreases the chance that there will be an interaction between the LCR and the e-globin gene. However, the LCR interacts with the next gene in order, the γ-globin gene. In Δ5′HS3c β-YAC mice, γ-globin gene expression is normal during primitive erythropoiesis, but is extinguished in the fetal stage of definitive erythropoiesis. These data suggest that a conformational change occurs in the Δ5′HS3c LCR during the switch from embryonic to definitive erythropoiesis, from one that supports γ-globin gene expression to one that does not. Alternately, the embryonic trans-acting environment may allow the mutant LCR to interact with and activate the γ-globin genes, but the fetal trans-acting environment may not support this interaction in the absence of the 5′HS3 core. To distinguish between these possibilities, β-YAC lines were produced in which the e-globin gene was replaced with a second marked β-globin gene (βm), coupled to either an intact LCR, a 2.9 Kb 5′HS3 deletion or a 234 bp 5′HS3 core deletion. Δ5′HS3c Δe::βm β-YAC mice expressed βm-globin throughout development beginning at day 10 in the yolk sac. γ-globin was expressed in the embryonic yolk sac, but not in the fetal liver. Some wild-type β-globin was expressed in addition to βm-globin in adult mice. The γ-globin phenotype is consistent with published data on Δ5′HS3c β-YAC mice. Although e-globin was not expressed in Δ5′HS3c β-YAC mice, βm-globin was expressed in Δ5′HS3c Δe::βm β-YAC embryos, demonstrating that the 5′HS3 core was necessary for e-globin expression during embryonic erythropoiesis, but not for βm-globin expression. A similar phenotype was observed in Δ5′HS3 Δe::βm β-YAC mice, except βm-globin expression was higher in the day 10 yolk sac and γ-globin expression continued into the fetal liver stage of definitive erythropoiesis consistent with results published on Δ5′HS3 β-YAC mice. These data support a site specificity model of LCR HS-globin gene interaction.

In Vivo GATA-1 Binding to an LCR Core Element Requires Flanking Krupple-Like Factor Biniding Sites Which Direct the Formation of a Chromatin Mini-Domain.

The GATA-1 transcription factor is involved in the expression of most, if not all erythroid genes. In vitro, GATA-1 binds the WGATAR consensus and related sequences. If GATA-1 binding to DNA in nuclear chromatin were only dependent on these recognition sequences, there would be roughly 3 million sites per genome – or approximately 1 site per 1000 bp. The alternative is that mechanisms beyond the consensus sequence are involved in targeting GATA-1 (and other transcription factors) to sites in chromatin. To investigate which model is correct, we performed GATA-1 ChIP assays in MEL cells on 9 WGATAR sites in the murine β-globin gene locus that were not associated with known regulatory elements. GATA-1 bound to none of these sites, while it did bind to positive control sites in LCR HS2 and the GATA-1 promoter. This implied that factors other than DNA sequence are involved in GATA-1 targeting. We hypothesized that factor binding sites flanking the WGATAR motifs are necessary for GATA-1 binding. To test this idea, we chose the human LCR HS4 core as a model. Here, a cluster of 6 binding sites, contained within ~100 bp, is found in the following arrangement: 5′-K-N-K-GG-K-3′ (K=Krupple-like factor, N=NF-E2, G=GATA). To determine which sites were required for GATA-1 binding, we used site-directed mutagenesis to insert combinations of the sites into a non-expressed fragment of DNA in a pUC-based plasmid. Constructs were then stably transfected into MEL cells and pools of 25 clones were analyzed for GATA-1 binding to the integrated test sites. Each ChIP assay was performed in triplicate and analyzed by quantitative RT-PCR, also in triplicate. Internal positive and negative control sites for GATA-1 binding were included for each ChIP assay. GATA-1 did not bind the following constructs in vivo: no sites, N alone, GG alone, N-GG, N-K-GG, N-K-GG-K, and K-N-K-GG. Only when all sites were present, K-N-K-GG-K, did GATA-1 bind. These results showed that both the 5′ and 3′ flanking KLF sites were necessary for GATA-1 binding. To determine whether they were also sufficient for GATA-1 binding, we next evaluated K-K and K-GG-K constructs, where the K sites were the 5′ and 3′ flanking sites in their normal locations. While both the K-K and K-GG-K constructs formed DNAse I HSs and directed H3 hyperacetylation in the region (the control without binding sites showed neither characteristic), indicating that the site was accessible for factor binding, the K-GG-K construct still did not bind GATA-1. We next treated cells containing the GG construct with trichostatin A to verify that histone hyperacetylation alone was insufficient to allow GATA-1 binding. Despite a large increase in total H3 acetylation (Western blot) and a doubling of local H3 acetylation (ChIP), GATA-1 still did not bind the GG construct. Based on these results we propose a mechanism for GATA-1 binding to the LCR HS4 core in which flanking KLF sites initially bind factors that form a localized “mini-domain” characterized by histone hyperacetylation, HS formation and trans-factor accessibility, but that factors binding to the NF-E2 and/or the internal KLF site are required for stabilizing GATA-1 binding within this open chromatin domain. These results are relevant not only to GATA-1 binding, but to understanding how transcription factors, in general, are targeted to specific sites in nuclear chromatin.

Specificity Protein 1 (Sp1) Consensus Binding Sequences Flank Many Functional GATA-1 Binding Sites and Are Sufficient To Allow GATA-1 Binding to a WGATAR Sequence in Nuclear Chromatin.

GATA-1 is an hematopoietic transcription factor known for its roles in erythropoiesis and erythroid gene expression. In vitro, GATA-1 readily binds to WGATAR and closely related sequences. However, in vivo, GATA-1 appears to bind only a small subset of potential sites - those associated with gene regulatory elements. It is currently unclear what allows stable binding of GATA-1 at these sites in nuclear chromatin. We hypothesized that the binding of factors adjacent to WGATAR sites may “open” the local chromatin region to allow GATA-1 access to binding sites, while other factors may stabilize GATA-1 binding. To begin to test this model, we examined 11 well-characterized regulatory elements which are known to bind GATA-1 to determine whether there were any common associated binding sites. To do this we used the Transcription Element Search Software (TESS). These sites were from the human β-globin LCR HSs 2–4, human and mouse β-globin promoters, human γ-globin, NF-E2, EKLF and SCl/TALI (proximal and distal) promoters and the chicken 3′ β-globin enhancer. This analysis revealed that in each case, Sp1 consensus binding sites were found flanking the WGATAR elements. While this association between Sp1 and GATA binding motifs has been noted before, the reason for this association has not been determined. On average, the 5′ Sp1 motifs were located 18 +/− 8 bp upstream of the WGATAR elements, and the 3′ Sp1 motifs were located 30 +/− 17 bp downstream. To determine whether these flanking Sp1 elements were sufficient to allow GATA-1 binding in chromatin, we developed a new approach based on the Bouhassira lab's recombinase mediated cassette exchange (RMCE) system. This system allows us to design oligonucleotides up to ~120 bp in length containing clusters of test binding sites and then insert them as single copies into a defined genomic location in MEL cells. ChIP experiments, analyzed by quantitative RT-PCR, can then be used to characterize factor binding and local histone modification. The positive internal control for GATA-1 binding and histone H3 acetylation was the −2.8G region of the mGATA-1 promoter. The negative control was the murine necdin gene promoter. As an additional positive control, we inserted the 100 bp LCR HS4 core element and found, as expected, that it was able to mediate GATA-1 binding. We then inserted 3 different GATA-1 binding elements and two derivative sites which have been shown to bind GATA-1 with high affinity in vitro. None of these sites alone was able to bind GATA-1 in our test system. We next placed one of the high affinity sites, a GATA-PAL motif from the murine gata-1 gene promoter, between two exact Sp1 consensus sequences (GGGGCGGGG) located 18 bp upstream and 30 bp downstream of the WGATAR element. When tested in our system, the flanking Sp1 elements were sufficient to allow GATA-1 binding to the test site at levels equivalent to those seen for the positive control site. As expected, no GATA-1 bound when the Sp1 sites were used alone, but H3 histones in the region did become hyperacetylated, indicating that factors interacting with the Sp1 direct localized histone acetylation and mediate GATA-1 binding. These results begin to provide a mechanism for GATA-1 targeting to sites in nuclear chromatin and provide an explanation for the observed association of WGATAR and Sp1 binding elements within erythroid gene regulatory elements.

Cooperativeness of the Higher Chromatin Structure of the β-Globin Locus Revealed by the Deletion Mutations of DNase I Hypersensitive site 3 of the LCR

High-level transcription of the globin genes requires the enhancement by a distant element, the locus control region (LCR). Such long-range regulation in vivo involves spatial interaction between transcriptional elements, with intervening chromatin looping out. It has been proposed that the clustering of the HS sites of the LCR, the active globin genes, as well as the remote 5′ hypersensitive sites (HSs) (HS-60/-62 in mouse, HS-110 in human) and 3′HS1 forms a specific spatial chromatin structure, termed active chromatin hub (ACH). Here we report the effects of the HS3 deletions of the LCR on the spatial chromatin structure of the β-globin locus as revealed by the chromatin conformation capture (3C) technology. The small HS3 core deletion (0.23 kb), but not the large HS3 deletion (2.3 kb), disrupted the spatial interactions among all the HS sites of the LCR, the β-globin gene and 3′HS1. We have previously demonstrated that the large HS3 deletion barely impairs the structure of the LCR holocomplex, while the structure is significantly disrupted by the HS3 core deletion. Taken together, these results suggest that the formation of the ACH is dependent on a largely intact LCR structure. We propose that the ACH indeed is an extension of the LCR holocomplex.

Human β-Globin Locus Control Region Hypersensitive Site Specificity for Globin Gene Activation during Erythropoiesis.

Although the human β-globin locus control region (LCR) functions as a holocomplex within an active chromatin hub, we provide evidence that within the aggregate hypersensitive site (HS) activation domain of the holocomplex, the individual HSs still mediate preferential activation of the globin genes during development. A 2.9 Kb deletion of 5′HS3 (Δ5′HS3) or a 234 bp deletion of the 5′HS3 core (Δ5′HS3c) in a 213 Kb human β-globin locus yeast artificial chromosome (β-YAC) abrogate ε-globin gene expression during primitive erythropoiesis in β-YAC transgenic mice, suggesting that 5′HS3 sequences of the LCR are involved directly in ε-globin gene activation. The reduction of ε-globin gene transcription in Δ5′HS3 or Δ5′HS3c β-YAC transgenics can be explained by two hypotheses. The first is site-specificity. The interaction between the LCR and the ε-globin gene promoter involves specific sequences of 5′HS3 and specific sequences of the ε-globin gene promoter. When 5′HS3 or its core is deleted, these interactions do not take place and ε-globin gene transcription is diminished. The second hypothesis is change in conformation of the LCR. Normally, in the embryonic stage, the LCR achieves a three-dimensional conformation that favors interaction with the first gene in the complex, the ε-globin gene. When 5′HS3 is deleted, an alternate conformation is assumed that decreases the chance that there will be an interaction between the LCR and the ε-globin gene. However, the LCR interacts with the next gene in order, the γ-globin gene. In Δ5′HS3c β-YAC mice, γ-globin gene expression is normal during primitive erythropoiesis, but is extinguished in the fetal stage of definitive erythropoiesis. These data suggest that a conformational change occurs in the Δ5′HS3c LCR during the switch from embryonic to definitive erythropoiesis, from one that supports γ-globin gene expression to one that does not. Alternately, the embryonic trans-acting environment may allow the mutant LCR to interact with and activate the γ-globin genes, but the fetal trans-acting environment may not support this interaction in the absence of the 5′HS3 core. To distinguish between these possibilities, β-YAC lines were produced in which the ε-globin gene was replaced with a second marked β-globin gene (βm), coupled to either an intact LCR, a 2.9 Kb 5′HS3 deletion or a 234 bp 5′HS3 core deletion. Δ5′HS3c Δε::βm β-YAC mice expressed βm-globin throughout development beginning at day 10 in the yolk sac. γ-globin was expressed in the embryonic yolk sac, but not in the fetal liver. Some wild-type β-globin was expressed in addition to βm-globin in adult mice. The γ-globin phenotype is consistent with published data on Δ5′HS3c β-YAC mice. Although ε-globin was not expressed in Δ5′HS3c β-YAC mice, βm-globin was expressed in Δ5′HS3c Δε::βm β-YAC embryos, demonstrating that the 5′HS3 core was necessary for ε-globin expression during embryonic erythropoiesis, but not for βm-globin expression. A similar phenotype was observed in Δ5′HS3 Δε::βm β-YAC mice, except βm-globin expression was higher in the day 10 yolk sac and γ-globin expression continued into the fetal liver stage of definitive erythropoiesis consistent with results published on Δ5′HS3 β-YAC mice. These data support a site specificity model of LCR HS-globin gene interaction.

The Binding of GATA-1 to a β-Globin LCR HS Core Element in Nuclear Chromatin Requires Multiple Associated Factor Binding Sites.

The GATA-1 transcription factor has been shown to interact in vitro with many gene regulatory elements including promoters, enhancers, silencers and locus control region elements. The GATA-1 consensus binding sequence is generally considered to be WGATAR, yet it can also bind to other related sequences that do not perfectly fit this consensus. This raises the question of whether all potential GATA-1 binding sites identified by in vitro DNA binding assays or sequence analysis are actually bound by GATA-1 in nuclear chromatin. We hypothesized that it was unlikely that all, or even most, predicted GATA-1 sites were bound in vivo. If correct, this would imply that additional mechanisms, beyond DNA sequence, are required to target GATA-1 to specific genetic regulatory elements. To address this hypothesis we first performed GATA-1 ChIP assays in MEL cells at 10 WGATAR sites which spanned the mouse β-globin locus but were not associated with known regulatory elements. Positive controls for GATA-1 binding were the core region of LCR HS2 and the -2.8 mGATA-1 gene upstream enhancer. Negative controls included two regions from the β-globin locus which do not contain WGATAR elements and a WGATAR element from the proximal mGATA-1 gene promoter previously shown by the Bresnick lab to not bind GATA-1. While enrichment of the positive control sites using anti-GATA-1 antibody was 10-20-fold more than with IgG (p<0.001), binding at none of the 10 test WGATAR sites was enriched. This result implied that conditions other than the presence of the WGATAR consensus sequence are required to specify GATA-1 binding in nuclear chromatin. To better understand this process we decided to more closely evaluate the binding of GATA-1 to β-globin LCR HS core elements. We first tested the idea that GATA-1 was able to bind these sites due to high-level affinity based on the presence of tandem-inverted WGATAR elements and the known ability of GATA-1 molecules to self-interact. However, when tested in gel shift assays, GATA-1 binding to the LCR HS core elements had a very rapid off-rate which was much faster than a known high-affinity control site from the murine α-globin promoter. This led us to next test the idea that surrounding factor binding sites were required for GATA-1 binding to the LCR HS cores. Using human LCR HS4 as a model, we systematically assembled combinations of binding sites (3 KLF sites and 1 NF-E2 site) known to flank the tandem-inverted WGATAR elements and tested these artificial HS cores for GATA-1 binding after integration into MEL cell chromatin. The wild-type HS4 core served as a positive control and a fragment of luciferase cDNA served as a negative control for GATA-1 binding. Results for each cell line were normalized to the internal positive control -2.8 mGATA-1 gene site. Our experiments showed that only when all known flanking sites were present was GATA-1 binding detected. Similarly, DNase HS formation and histone H3 acetylation only occurred when all of the surrounding sites were present. Finally, we used ChIP to show that the assembled complex, with all sites present, includes BRG1 a component of the swi/snf nucleosome remodeling complex. These results indicate that GATA-1 targeting to the LCR HS cores involves complex, multi-protein interactions including histone acetylation and nucleosome displacement. Similar mechanisms are likely to be involved in targeting this and other transcription factors to functionally important sites in the genome.

Synergistic and Additive Properties of the Beta-Globin Locus Control Region (LCR) Revealed by 5′HS3 Deletion Mutations: Implication for LCR Chromatin Architecture

Deletion of the 234-bp core element of the DNase I hypersensitive site 3 (5'HS3) of the locus control region (LCR) in the context of a human beta-globin locus yeast artificial chromosome (beta-YAC) results in profound effects on globin gene expression in transgenic mice. In contrast, deletion of a 2.3-kb 5'HS3 region, which includes the 234-bp core sequence, has a much milder phenotype. Here we report the effects of these deletions on chromatin structure in the beta-globin locus of adult erythroblasts. The 234-bp 5'HS3 deletion abolished histone acetylation throughout the beta-globin locus; recruitment of RNA polymerase II (pol II) to the LCR and beta-globin gene promoter was reduced to a basal level; and formation of all the 5' DNase I hypersensitive sites of the LCR was disrupted. The 2.3-kb 5'HS3 deletion mildly reduced the level of histone acetylation but did not change the profile across the whole locus; the 5' DNase I hypersensitive sites of the LCR were formed, but to a lesser extent; and recruitment of pol II was reduced, but only marginally. These data support the hypothesis that the LCR forms a specific chromatin structure and acts as a single entity. Based on these results we elaborate on a model of LCR chromatin architecture which accommodates the distinct phenotypes of the 5'HS3 and HS3 core deletions.

LCR 5′HS3 Displays Specificity for ε-Globin Gene Activation during Primitive Erythropoiesis and γ-Globin Gene Activation during Fetal Definitive Erythropoiesis.

A 2.9 Kb deletion of 5′HS3 (Δ5′HS3) or a 234 bp deletion of the 5′HS3 core (Δ5′HS3c) in a 213 Kb human β-globin locus yeast artificial chromosome (β-YAC) abrogate ε-globin gene expression during primitive erythropoiesis in β-YAC transgenic mice, suggesting that HS sequences of the LCR are involved directly in ε-globin gene activation. The reduction of ε-globin gene transcription in Δ5′HS3 or Δ5′HS3c β-YAC transgenics can be explained by two hypotheses. The first is site-specificity. The interaction between the LCR and the ε-globin gene promoter involves specific sequences of 5′HS3 and specific sequences of the ε-globin gene promoter. When 5′HS3 or its core is deleted, these interactions do not take place and ε-globin gene transcription is diminished. The second hypothesis is change in conformation of the LCR. Normally, in the embryonic stage, the LCR achieves a three-dimensional conformation that favors interaction with the first gene in the complex, i.e., the ε-globin gene. When 5′HS3 is deleted, an alternate conformation is assumed that decreases the chance that there will be an interaction between the LCR and the ε-globin gene. However, the LCR interacts with the next gene in order, the γ-globin gene. In Δ5′HS3c β-YAC mice, γ-globin gene expression is normal during primitive erythropoiesis, but is extinguished in the fetal stage of definitive erythropoiesis. These data suggest that a conformational change occurs in the Δ5′HS3c LCR during the switch from embryonic to definitive erythropoiesis, from one that supports γ-globin gene expression to one that does not. Alternately, the embryonic trans-acting environment may allow the mutant LCR to interact with and activate the γ-globin genes, but the fetal trans-acting environment may not support this interaction in the absence of the 5′HS3 core. To distinguish between these possibilities, β-YAC lines were produced in which the ε-globin gene was replaced with a second marked β-globin gene (βm), coupled to either an intact LCR, a 2.9 Kb 5′HS3 deletion or a 234 bp 5′HS3 core deletion. Δ5′HS3c Δε::βm β-YAC mice expressed βm-globin throughout development beginning at day 10 in the yolk sac. γ-globin was expressed in the embryonic yolk sac, but not in the fetal liver. Some wild-type β-globin was expressed in addition to βm-globin in adult mice. The γ-globin phenotype is consistent with published data on Δ5′HS3c β-YAC mice. Although ε-globin was not expressed in Δ5′HS3c β-YAC mice, βm-globin was expressed in Δ5′HS3c Δε::βm β-YAC embryos, demonstrating that the 5′HS3 core was necessary for ε-globin expression during embryonic erythropoiesis, but not for βm-globin expression. These data support a site specificity model of LCR HS-globin gene interaction. In addition, nuclear ligation experiments provided evidence for a specific physical interaction between 5′HS3 and the γ-globin promoter during fetal definitive erythropoiesis, further supporting a site specificity model.

Effect of LCR 5′HS4 and 5′HS1 Deletions on β-Like Globin Gene Expression in β-Yac Transgenic Mice.

A 213 Kb human β-globin locus yeast artificial chromosome (β-YAC) was modified by homologous recombination to delete 2.9 Kb of cross-species conserved sequence similarity encompassing the LCR 5′HS4 (Δ5′HS4 β-YAC). Three transgenic mouse lines were established; each contained two intact copies of the β-globin locus as determined by long range restriction enzyme mapping (LRRM) and Southern blot hybridization analyses. Human ε-, γ- and β-globin, and mouse α- and ζ-globin mRNAs were measured by RNAse protection in hematopoietic tissues derived from staged embryos, fetuses and adult mice. No difference in the temporal pattern of globin transgene expression was observed between Δ5′HS4 β-YAC mice and wild-type β-YAC mice. In addition, quantitative per-copy human β-like globin mRNA levels were similar between Δ5′HS4 and wild-type β-YAC transgenic lines, although γ-globin gene expression was slightly increased in the fetal liver, while β-globin gene expression was slightly decreased in Δ5′HS4 β-YAC mice. These data are in contrast to data obtained from β-YAC mice containing a deletion of the 280 bp 5′HS4 core. In these mice, γ- and β-globin gene expression was significantly decreased during fetal definitive erythropoiesis and β-globin gene expression was decreased during adult definitive erythropoiesis. However, these data are consistent with the observation that deletion of the 5′HS core elements is more deleterious than large deletions of the 5′HSs. Together, the compiled deletion data supports the hypothesis that the LCR exists as a holocomplex in which the 5′HS cores form an active site and the flanking 5′HS regions constrain the holocomplex conformation. In this model, 5′HS core mutations are dominant negative, whereas larger deletions allow the LCR to fold into alternate holocomplex structures that function normally, albeit less efficiently.To complete the study on the contribution of the individual 5′HSs to LCR function, a 0.8 Kb 5′HS1 fragment was deleted in the 213 Kb β-YAC by homologous recombination. Two ΔHS1 β-YAC transgenic lines have been established; four additional founders were recently identified. Of the two lines, one contains two intact copies of the globin locus; the other contains four deleted copies, one of which extends from the LCR through just 5′ to the β-globin gene. For both lines, ε-globin gene expression was markedly reduced, approximately 5–10 fold, during primitive erythropoiesis. Developmental expression profiles and levels of the γ- and β-globin genes (in the line that contains loci including the β-globin gene) were unaffected by deletion of 5′HS1. Breeding of the remaining four founders to obtain F1 and F2 progeny for similar structure/function studies is in progress. Decreased expression of the β-globin gene is the first phenotype ascribed to a 5′HS1 mutation, suggesting that this HS does indeed have a role in LCR function.

Long-Range Enhancement of β-Globin Gene Expression Is Dependent on Gt Motifs Residing in the HS3 Core Element of the Locus Control Region.

Normal expression of the human β-globin genes is dependent on a powerful regulatory element residing upstream of the globin gene cluster referred as the locus control region (LCR), physically characterized by five DNAseI-hypersensitive sites (HS1 to HS5). Of particular interest is HS3, which is characterized by a 225 bp core element region containing seven GT motifs (GT1 to GT7) that alternate with four GATA-1 binding sites. The GT motifs are bound by a family of Kruppel-like factors and are important for proper expression of many housekeeping and tissue-specific genes. We previously demonstrated in transgenic mice carrying a 248 kb human β-globin yeast artificial chromosome in which the GT6 motif (GT6m β-YAC) was mutated resulted in a decrease in ε- and γ-globin gene expression during embryonic erythropoiesis and a decrease in γ-globin expression during definitive erythropoiesis, thus, providing evidence that a single transcriptional motif distantly located can have profound effects on gene expression. We have used the same β-YAC transgenic mouse model system to analyze the function of the remaining six GT motifs of the HS3 core element. Currently we have produced transgenic mice carrying either a mutation of GT1/2 (GT1 and GT2 motifs overlap) or GT3. Total RNA was isolated from yolk sac, liver and blood samples of transgenic F2 embryos at 12- and 14-day postconception and adult blood, and subjected to RNAse protection analysis. The GT1/2m β-YAC transgenic mice exhibited normal ε- and γ-globin gene expression during embryonic erythropoiesis, but during definitive erythropoiesis γ-globin gene expression was reduced 4 to 5 fold in day 12- and day 14- fetal livers. In the two GT1/2m β-YAC transgenic lines produced, β-globin gene expression was reduced and levels varied from 12.3% ± 2.9% to 63.2% ± 5.8% of endogenouse murine α-globin indicating that β-globin expression is strongly influenced by the position of integration of the transgene. In contrast, the GT3m β-YAC transgenic mice the mutation to GT3 had no effect on globin gene expression during development. Our results suggest that multiple GT motifs within the same highly complex regulatory element contribute differently to the enhancement of the downstream genes.

Chromatin structure at the flanking regions of the human beta-globin locus control region DNase I hypersensitive site-2: proposed nucleosome positioning by DNA-binding proteins including GATA-1

The human beta-globin locus control region DNase I hypersensitive site-2 (LCR HS-2) is erythroid-specific and is located 10.9 kb upstream of the epsilon-globin gene. Most studies have only examined the core region of HS-2. However, previous studies in this laboratory indicate that positioned nucleosomes are present at the 5′- and 3′-flanking regions of HS-2. In addition, footprints were observed that indicated the involvement of DNA-binding proteins in positioning the nucleosome cores. A consensus GATA-1 site exists in the region of the 3′-footprint. In this study, using an electrophoretic mobility shift assay (EMSA) and DNase I footprinting, we confirmed that GATA-1 binds in vitro at the 3′-end of HS-2. An additional GATA-1 site was found to bind GATA-1 in vitro at a site positioned 40 bp upstream. At the 5′-end of HS-2, DNase I footprinting revealed a series of footprints showing a marked correlation with the in vivo footprints. EMSA indicated the presence of several erythroid-specific complexes in this region including GATA-1 binding. Sequence alignment for 12 mammalian species in HS-2 confirmed that the highest conservation to be in the HS-2 core. However, a second level of conservation extends from the core to the sites of the proposed positioning proteins at the HS-2 flanking regions, before declining rapidly. This indicates the importance of the HS-2 flanking regions and supports the proposal of nucleosome positioning proteins in these regions.

A human globin enhancer causes both discrete and widespread alterations in chromatin structure.

Gene activation requires alteration of chromatin structure to facilitate active transcription complex formation at a gene promoter. Nucleosome remodeling complexes and histone modifying complexes each play unique and interdependent roles in bringing about these changes. The role of distant enhancers in these structural alterations is not well understood. We studied nucleosome remodeling and covalent histone modification mediated by the beta-globin locus control region HS2 enhancer at nucleosome-level resolution throughout a 5.5-kb globin gene model locus in vivo in K562 cells. We compared the transcriptionally active locus to one in which HS2 was inactivated by mutations in the core NF-E2 sites. In contrast to inactive templates, nucleosomes were mobilized in discrete areas of the active locus, including the HS2 core and the proximal promoter. Large differences in restriction enzyme accessibility between the active and inactive templates were limited to the regions of nucleosome mobilization, which subsumed the DNase I hypersensitive sites. In contrast to this discrete pattern, histone H3 and H4 acetylation and H3 K4 methylation were elevated across the entire active locus, accompanied by depletion of linker histone H1. The coding region of the gene differed from the regulatory regions, demonstrating both nucleosome mobilization and histone hyperacetylation, but lacked differences in restriction enzyme accessibility between transcriptionally active and inactive genes. Thus, although the histone modification pattern we observe is consistent with the spreading of histone modifying activity from the distant enhancer, the pattern of nucleosome mobilization is more compatible with direct contact between an enhancer and promoter.

The interaction between the human beta-globin locus control region and nuclear matrix.

Our previous study showed that hydroxyurea (Hu) could induce HEL cells to express human beta-globin gene. However the molecular mechanisms by which the expression of beta-globin gene is activated and regulated are poorly understood. Here we show that the binding patterns between the core DNA sequences (HS2 core sequence -10681 approximately -10971 bp, HS3 core sequence -14991 approximately -14716 bp and HS4 core sequence -18586 approximately -18306 bp) of DNase I hypersensitive sites in the human beta-globin LCR and nuclear matrix proteins isolated from Hu induced and uninduced HEL cells are quite different. Results demonstrated that nuclear matrix proteins might play important roles in regulating the expression of human beta-like globin genes through their interaction with HSs (HS2, HS3 and HS4 core sequences) in the LCR. Moreover, the results obtained from the in vitro DNA-matrix binding assay showed that the core DNA sequences of DNase I hypersensitive sites (HS2, HS3 and HS4) were unable to bind to the nuclear matrix isolated from uninduced HEL cells; in addition, HS2 core DNA sequence was capable of binding to the nuclear matrix prepared from Hu-induced HEL cells, while both HS3 and HS4 core DNA sequences could not do so. Results indicated that the HS2 core DNA sequence may be a functional MAR (matrix attachment region). We suggest that the HS2 core DNA sequence binding to the nuclear matrix in Hu-induced HEL cells may open the structure of chromatin to make the LCR accessible to the promoter of beta-globin gene and to promote its transcription.

Activation of the beta-like globin genes in transgenic mice is dependent on the presence of the beta-locus control region.

The beta-globin locus control region (LCR) is a powerful regulatory element required for high-level globin gene expression. We have generated transgenic mouse lines carrying a beta-globin locus yeast artificial chromosome lacking the LCR to determine if the LCR is required for globin gene activation. beta-Globin gene expression was analyzed by RNase protection, but no detectable levels of epsilon-, gamma- and beta-globin gene transcripts were produced at any stage of development. These findings suggest that the presence of the LCR is a minimum requirement for globin gene expression. Next, we tested whether the LCR is necessary to activate globin gene expression in a gamma-globin promoter mutant that causes hereditary persistence of fetal hemoglobin (HPFH). beta-YAC transgenic mice carrying the -117 HPFH mutation and a HS3 core deletion that specifically abolishes gamma-globin gene expression during definitive erythropoiesis were produced to test whether the -117 (A)gamma promoter is activated in the absence of interaction with the LCR. In four transgenic mouse lines, gamma-globin gene expression was absent in adult erythrocytes, suggesting that an interaction between the gamma-globin gene promoter and the LCR is required for gamma gene activation even when the promoter contains an HPFH mutation.

GATA-1 binding sites mapped in the beta-globin locus by using mammalian chIp-chip analysis.

The expression of the beta-like globin genes is intricately regulated by a series of both general and tissue-restricted transcription factors. The hemapoietic lineage-specific transcription factor GATA-1 is important for erythroid differentiation and has been implicated in regulating the expression of the erythroid-specific genes including the genes of the beta-globin locus. In the human erythroleukemic K562 cell line, only one DNA region has been identified previously as a putative site of GATA-1 interaction by in vivo footprinting studies. We mapped GATA-1 binding throughout the beta-globin locus by using chIp-chip analysis of K562 cells. We found that GATA-1 binds in a region encompassing the HS2 core element, as was previously identified, and an additional region of GATA-1 binding upstream of the gammaG gene. This approach will be of general utility for mapping transcription factor binding sites within the beta-globin locus and throughout the genome.