- Email: [email protected]
Histone acetylation contributes to chromatin looping between the locus control region and globin gene by influencing hypersensitive site formation
Biochimica et Biophysica Acta 1829 (2013) 963–969
Contents lists available at SciVerse ScienceDirect
Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbagrm
Histone acetylation contributes to chromatin looping between the locus control region and globin gene by influencing hypersensitive site formation Yea Woon Kim, AeRi Kim ⁎ Department of Molecular Biology, College of Natural Sciences, Pusan National University, Busan 609-735, South Korea
a r t i c l e
i n f o
Article history: Received 22 February 2013 Received in revised form 27 March 2013 Accepted 11 April 2013 Available online 20 April 2013 Keywords: Histone acetylation Chromatin loop Hypersensitive site β-Globin locus Transcription
a b s t r a c t Chromatin loops are formed between enhancers and promoters and between insulators to regulate gene transcription in the eukaryotic genome. These transcription regulatory elements forming loops have highly acetylated histones. To understand the correlation between histone acetylation and chromatin loop formation, we inhibited the expression of histone acetyltransferase CBP and p300 in erythroid K562 cells and analyzed the chromatin structure of the β-globin locus. The proximity between the locus control region (LCR) and the active Gγ-globin gene was decreased in the β-globin locus when histones were hypoacetylated by the double knockdown of CBP and p300. Sensitivity to DNase I and binding of erythroid specific activators were reduced in the hypoacetylated LCR hypersensitive sites (HSs) and gene promoter. Interestingly, the chromatin loop between HS5 and 3′HS1 was formed regardless of the hypoacetylation of the β-globin locus. CTCF binding was maintained at HS5 and 3′HS1 in the hypoacetylated locus. Thus, these results indicate that histone acetylation contributes to chromatin looping through the formation of HSs in the LCR and gene promoter. However, looping between insulators appears to be independent from histone acetylation. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Transcription regulatory elements, such as promoters, enhancers and insulators are generally located in long distances on the eukaryote genome, but often positioned in close proximity in the threedimensional space of nucleus, forming chromatin loops by extruding intervening regions between them [1–3]. A chromatin loop between an enhancer and promoter is formed in actively transcribed gene loci. In the mammalian β-globin locus, the locus control region (LCR), a far upstream enhancer, is closely positioned to the transcriptionally active globin genes in erythroid cells [4–7]. The close positioning between the LCR and its target genes is not generated when the globin genes are not transcribed due to the lack of erythroid specific transcriptional activators [8–10]. Thus chromatin loop formation between the LCR and its target gene appears to be a necessary step for gene transcription that causes physical and functional interaction between them. Insulators participate in chromatin loop formation to block enhancerpromoter interactions or to establish chromatin domains [11–13]. The LCR hypersensitive site (HS) 5 and 3′HS1 that are located at the
Abbreviations: LCR, locus control region; HS, hypersensitive site; HAT, histone acetyltransferase; shRNA, short hairpin RNA; ChIP, chromatin immunoprecipitation; RT-PCR, reverse transcription-PCR; 3C, chromosome conformation capture ⁎ Corresponding author. Tel.: +82 51 510 3683; fax: +82 51 513 9258. E-mail address: [email protected] (A. Kim). 1874-9399/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbagrm.2013.04.006
boundaries of the β-globin locus are closely positioned, forming a chromatin loop in erythroid cells [4,5,14]. These sites form part of an active chromatin hub with the LCR HSs and globin gene, when the gene is highly transcribed. Close positioning between insulators is dependent on the binding of CTCF protein [15,16]. The reduction of CTCF increases repressive histone modifications and decreases gene transcription in the human β-globin locus, in addition to the loss of a chromatin loop between HS5 and 3′HS1 [15]. However, the looping between insulators does not appear to be sufficient for the globin gene transcription, because the loop is maintained in the β-globin locus when the transcription of the globin genes fails due to the lack of erythroid specific activators [8,10]. The loop between HS5 and 3′HS1 is also observed in erythroid progenitor cells [4]. Specific histone modifications are associated with gene transcription. Among the many kinds of modifications, acetylation occurring at several lysine residues of histones is highly enriched in transcriptionally active genes and transcription regulatory elements [17,18]. Hypoacetylation of histones by the lack of histone acetyltransferases (HAT) inhibits gene transcription [19,20]. In the human and mouse β-globin loci, the level of histone acetylation is increased by transcriptional induction [6,21–23]. The increase of histone acetylation precedes transcription of the β major globin gene in the mouse locus. Transcription inactivation of the β-globin gene after loss of a transcriptional activator or coactivator is accompanied by the hypoacetylation of histones in the locus [24,25]. In spite of many studies showing the correlation of histone acetylation with gene transcription, it is unclear
964
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
whether histone acetylation affects chromatin loop formation and how the acetylation contributes to the loop formation. In our previous study, the chromatin loop between the LCR and G γ-globin gene was not formed when the γ-globin transcription was severely reduced by the knockdown of the erythroid specific activators GATA-1 or NF-E2 [8]. Histone hyperacetylation in the LCR and G γ-globin gene was lost in the GATA-1 knockdown cells, but it was maintained in the NF-E2 knockdown cells. Here, to understand the relationship of histone acetylation to chromatin loop formation, we inhibited histone acetylation of the β-globin locus by reducing histone acetyltransferases CBP and p300 in K562 cells. Chromatin structure including chromatin loops, histone modifications, HS formation and activator binding was examined in the β-globin locus. This study shows a role of histone acetylation in chromatin loop formation through HS formation in the LCR and globin gene promoter. Chromatin loops between the HS5 and 3′HS1 insulator sites appear to be formed in a different manner from loops between the LCR and promoter. 2. Materials and methods 2.1. Lentiviral shRNA in K562 cells CBP and p300 knockdown cells were generated using TRC lentiviral short hairpin RNA (shRNA) vectors as previously described [8]. Briefly, lentiviral vectors (Sigma) were transfected into 293FT cells with Virapower packaging mix (Invitrogen). Viruses were harvested after 3 days and mixed with K562 cells in the presence of 6 μg/ml polybrene. K562 cells were selected in 2 μg/ml of puromycin at 3 days after virus transduction. Knockdown was confirmed by Western blot analysis. Control cells were generated using lentiviral vector not expressing shRNA (pKLO.1). 2.2. Chromatin immunoprecipitation (ChIP) ChIP was carried out as previously described [26]. K562 cells (1 × 10 7) were cross-linked with 1% formaldehyde and then nuclei were isolated by cell lysis. MNase digestion and sonication were performed with nuclei to produce chromatin at a mononucleosome size. Fragmented chromatin was incubated with antibodies after pre-clearing. Protein-DNA complexes were recovered with protein A or G agarose beads and DNA was purified after reverse cross-linking. Antibodies used in this study were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA) for CBP (sc-369), p300 (sc-8981), GATA-1 (sc-1233), TAL1 (sc-12984), and p45/NF-E2 (sc-22827), from Abcam for H3K27ac (ab4729) and from Millipore for H3ac (06–599), H3K27me1 (17–643), H3K27me2 (07–452), H3K27me3 (07–449) and CTCF (07–729). Normal rabbit IgG (sc-2027) and goat IgG (sc-2028) were purchased from Santa Cruz Biotechnology. 2.3. Reverse transcription-PCR (RT-PCR) RNA was prepared from 2 × 10 6 cells using the RNeasy Plus Mini Kit (Qiagen). One microgram of RNA was reverse transcribed with random hexamers using the Superscript III first-strand synthesis system as suggested by the manufacturer (Invitrogen). cDNA was diluted to 400 μl, and 2 μl of cDNA was amplified in a 10 μl reaction volume by quantitative PCR using TaqMan chemistry. The relative intensity of specific cDNA sequences was compared with a genomic DNA standard using the comparative Ct method. 2.4. DNase I sensitivity assay Nuclei were prepared from K562 cells (2 × 107) as previously described [26]. Aliquots of 3 × 106 nuclei were digested with 150–600 U DNase I in a 100 μl volume for 10 min at 25 °C. DNA was purified and run on a 1% agarose gel to visualize the level of digestion. DNase I
sensitivity was determined by quantitatively comparing digested DNA with undigested DNA using PCR and then by inverting the value at each concentration. 2.5. Chromosome conformation capture (3C) The 3C assay was performed as described previously with reduced number of cells [5,8,27]. K562 cells were cross-linked with 1% formaldehyde and nuclei were prepared from 2 × 10 6 cells and resuspended in 0.4 ml of 1 × restriction enzyme buffer. Chromatin was digested with 800 U of Hind III overnight and DNA fragments were ligated with 400 units of T4 ligase for 4 h at 16 °C. Ligated DNA was purified after reverse cross-linking and then amplified by SYBR green qPCR. Cutting ratio of chromatin by Hind III digestion was measured by amplifying DNA extracted from digested chromatin and comparing with DNA from undigested chromatin (Supplementary Fig. 1). Samples having cutting ratio over 70% were analyzed in this assay. Ligation efficiency between fragments and PCR efficiency between primer sets were corrected using control templates that were prepared by mixing, digesting and ligating equimolar amounts of the PCR fragments spanning the restriction enzyme sites of the β-globin locus and the same amount of genomic DNA [27]. The ligation between two fragments was analyzed using the reverse direction primer of each fragment except one in the β-globin locus. The relative cross-linking frequency was determined by comparing DNA ligated between two fragments in 3C samples with DNA ligated randomly in control templates and then by normalizing with the cross-linking frequency in the Ercc3 gene. 2.6. Quantitative PCR DNA obtained from all experiments was quantitatively analyzed using the 7300 Real-time PCR system (Applied Biosystems). PCR was carried out with 200 nmol of TaqMan probes and 900 nmol of primers in a 10 μl reaction volume for cDNA and DNA obtained from the ChIP and DNase I digestion. DNA obtained from 3C and control templates ligated randomly were amplified using SYBR green fluorescence in a 10 μl reaction volume. Data were collected at the threshold where amplification was linear. The locations of PCR amplicons in the β-globin locus are indicated in Fig. 1A. Sequences of primers, TaqMan probes and primers for 3C are provided in our previous study [8]. 3. Results 3.1. Histone acetylation of the β-globin locus is inhibited by the knockdown of p300 CBP and p300 are histone acetyltransferases [28,29] that bind to the β-globin LCR HSs (Fig. 1A) [6,25,30]. To generate histone hypoacetylation of the β-globin locus, we inhibited the expression of CBP or p300 using shRNA in erythroid K562 cells. CBP and p300 proteins were reduced in the knockdown cells without affecting the expression of each other as shown by Western blot analysis (Fig. 1B). The reduction of CBP or p300 decreases its binding at the β-globin LCR HSs, as revealed by the ChIP assay (Fig. 1C). Histone acetylation was analyzed using the ChIP assay for histone H3K9/K14 and H3K27 in the LCR HSs and γ-globin genes. Relative intensity in the assay was calculated against histone H3, instead of input, to measure acetylation level on histones. In CBP knockdown K562 cells, H3K9/K14ac and H3K27ac were not reduced across the β-globin locus compared to control cells (Fig. 1D and E). The reduction of histone acetylation was observed in p300 knockdown cells. Acetylation at H3K9/K14 was markedly reduced in the γ-globin genes, which are normally highly transcribed in K562 cells (Fig. 1D). Acetylation at H3K27ac was also decreased across the β-globin locus (Fig. 1E). These results show that histone acetylation of the β-globin locus can be inhibited by the lack of p300 in K562 cells. p300 appears to play a more major role in histone acetylation of the β-globin locus than CBP.
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
A 5
4
3
2
1
Aγ
Gγ
LCR
ε
965
β
δ
3’HS1
10kb
C
p300 β -tubulin
1.2
0.6
Con IgG CBPi IgG p300i IgG
Con CBP CBPi CBP p300i CBP
0.4 0.2 0.0
E Con IgG CBPi IgG p300i IgG
Con H3ac CBPi H3ac p300i H3ac
0.8
0.4
0.0
0.25 0.20
Con IgG CBPi IgG p300i IgG
Con p300 CBPi p300 p300i p300
0.15 0.10 0.05 0.00
Relative intensity
Relative intensity
D
0.8
Relative intensity
CBP
Relative intensity
B
5.0 4.0
Con IgG CBPi IgG p300i IgG
Con K27ac CBPi K27ac p300i K27ac
3.0 2.0 1.0 0.0
Fig. 1. Knockdown of CBP or p300 in erythroid K562 cells. (A) The human β-globin locus is presented. Vertical arrows indicate DNase I hypersensitive sites in LCR and 3′HS1. The exons of the globin genes are represented by squares. Vertical bars named below the diagram denote the locations of amplicons used in quantitative PCR. (B) Western blotting was performed using antibodies specific to CBP or p300 in protein extract from K562 cells expressing a control, CBP or p300 shRNA (con, CBPi and p300i). Blotting with β-tubulin antibody was used as an experimental control. ChIP was performed with antibodies specific to CBP or p300 (C) in K562 cells expressing each shRNA. Relative intensity was determined by quantitatively comparing input with immunoprecipitated DNA for the indicated amplicons. In ChIP assay using antibodies specific to H3K9/K14ac (D) or H3K27ac (E), relative intensity was calculated by dividing immunoprecipitated DNA intensity into histone H3 intensity to measure histone acetylation level in present histones. Normal rabbit IgG (IgG) served as experimental control. Bars representing IgG are almost non-detectable in graphs, because the relative intensity is very low. The results of three (C) or two (D) independent experiments ± SEM are graphed. *P b 0.05.
3.2. Histone hypoacetylation decreases transcription of the human γ-globin genes To inhibit histone acetylation in the β-globin locus and to remove any possible redundancy effects between CBP and p300, we performed the double knockdown of CBP and p300 in K562 cells. The expression of CBP and p300 was reduced in the double knockdown cells (Fig. 2A) and their binding at the LCR HSs was reduced to less than half compared to control cells (Fig. 2B). The knockdown of CBP and p300 decreased H3K9/K14ac and H3K27ac in the LCR HSs and the active γ-globin genes (Fig. 2C and D). The decrease was not larger than the decrease obtained by knockdown of p300 alone. Next, the transcription of the globin genes was analyzed by quantitative RT-PCR. RNA transcripts at exons 2 and 3 of the γ-globin genes were reduced to about 30% in the CBP/p300 knockdown cells compared to control cells (Fig. 2E). The δ- and β-globin genes remained in a silent status. The transcription of ε-globin gene, which occurs at low level in erythroid K562 cells, was slightly reduced in the HAT knockdown cells. These results show that the hyperacetylation of histones is required for the transcription of the γ-globin genes in the β-globin locus. 3.3. Chromatin looping of the Gγ-globin gene with LCR HSs and 3′HS1 is disrupted by histone hypoacetylation The active Gγ-globin gene is closely positioned with the LCR and 3′ HS1 in K562 cells, forming chromatin loops [8]. To explain the correlation
between histone acetylation and chromatin loop formation, we measured the relative cross-linking frequency between the Gγ-globin gene and LCR HSs by the 3C assay in control cells and CBP/p300 knockdown K562 cells (Fig. 3A). When a fragment containing LCR HS2 was used as an anchor in the 3C assay, the cross-linking frequency to a fragment containing the Gγ-globin gene was reduced to less than half in the knockdown cells (Fig. 3B). The cross-linking frequency between HS2 and 3′HS1 also decreased in the hypoacetylated β-globin locus. A fragment containing the Gγ-globin gene was less ligated with LCR HSs and 3′HS1 in the knockdown cells (Fig. 3C). No new strong ligation between any fragments was established by CBP/p300 knockdown in the β-globin locus. Thus these results indicate that histone hyperacetylation supports chromatin looping among the LCR, Gγ-globin gene and 3′HS1 in the β-globin locus, generating the active chromatin hub [5]. 3.4. Loss of acetylation at H3K27 increases mono- and di-methylation in the LCR and γ-globin genes Histone H3K27 is subject to four kinds of covalent modifications including acetylation. The modifications at H3K27, mono-, di- and tri-methylation and acetylation, affect each other [31–33]. To examine changes of modifications at H3K27 depending on the loss of acetylation, we analyzed the methylation status of the β-globin locus in control cells and CBP/p300 knockdown cells. Mono- and di-methylation were increased in the LCR and the γ-globin gene where acetylation was lost, but the increases were not observed in the inactive β-globin gene
966
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
B
E
CBP p300 β -tubulin
0.8
IgG CBP p300 x 5
0.6 0.4 0.2
C/pi
Con
HS4
Con
HS2
2.0
0.0
C/pi
HS1
D
1.2
Con IgG C/pi IgG
Con H3ac C/pi H3ac
Relative intensity
Relative intensity
C/pi
Con RT+ C/pi RT+
4.0
0.0 Con
C
6.0
Relative intensity
Relataaaive intensity
A
0.8
0.4
0.0
5.0
Con IgG C/pi IgG
4.0
Con K27ac C/pi K27ac
3.0 2.0 1.0 0.0
Fig. 2. Double knockdown of CBP and p300 in K562 cells. (A) Western blotting was performed using antibodies specific to CBP or p300 in protein extract from K562 cells where transduction of control shRNA (Con) or transduction of both CBP and p300 shRNA (C/pi) was performed. ChIP was performed with antibodies specific to CBP or p300 (B) in K562 cells expressing shRNA and analyzed as described in Fig. 1C. The relative intensity of p300 was multiplied by five to use same Y scale with CBP. In ChIP assay using antibodies specific to H3K9/K14ac (C) or H3K27ac (D), relative intensity was calculated as described in Fig. 1D and E. Normal rabbit IgG (IgG) served as experimental control. (E) cDNA was prepared from RNA isolated from K562 cells expressing shRNA and then amplified by quantitative PCR using primers and probes for exons of the ε-, γ-, δ- and β-globin genes. Relative intensity was calculated by comparing cDNA with genomic DNA for the indicated amplicons. The results of three (B, C, D) or two (E) independent experiments ± SEM are graphed. *P b 0.05.
where histones are hypoacetylated (Fig. 4A and B). Tri-methylation at H3K27, established at a very low level in control cells, was unchanged throughout the locus by hypoacetylation (Fig. 4C) [31]. Taken together, the loss of acetylation at H3K27 results in increasing mono and dimethylation. The increase of mono and di-methylation at H3K27, as well as the decrease of acetylation, could affect chromatin loop formation in the β-globin locus.
A
LCR 5
4
3
2
ε
1
Gγ
3.5. Histone hypoacetylation inhibits the formation of HSs in the β-globin locus HS formation in the LCR and gene promoter might be important for chromatin looping in the β-globin locus, because it allows the binding of transcriptional activators and chromatin loop mediating factors. Here, the effect of reduced histone acetylation on HS formation was studied
Aγ
β
δ
3’HS1
HindIII
C
LCR 5 43 21 ε
G
γ
A
γ
δ
β
3'HS1
3.0 Con C/pi
2.5 2.0 1.5 1.0 0.5 0.0 -20
0
20 G
40
60
80
Relative cross-linking frequency
Relative cross-linking frequency
B
10kb LCR 5
4
3
2
1
ε
Gγ
δ β
3'HS1
5.0 Con C/pi
4.0 3.0 2.0 1.0 0.0 -20
-10
0
40
60
80
Fig. 3. Relative proximity between HSs and the γ-globin gene in the β-globin locus in CBP/p300 knockdown cells. The 3C assay was performed with Hind III restriction enzyme in K562 cells expressing shRNA. (A) Hind III sites and PCR primers in the β-globin locus were represented by vertical bars and horizontal arrows, respectively. The black shading represents the anchor fragment for HS2 (B) and Gγ-globin gene (C) in PCR. The gray shadings are fragments generated by Hind III digestion. Relative cross-linking frequency was determined by quantitatively comparing ligated DNA in cross-linked chromatin with control DNA and then normalizing to the cross-linking frequency at the Ercc3 gene. The results are averages of four to five independent experiments ± SEM.
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
Relative intensity
A
1.0
Con IgG C/pi IgG
0.8
by quantitatively measuring sensitivity to DNase I attack (Fig. 5A). DNase I sensitivity was severely decreased at the LCR HS4, HS2, HS1 and the γ-globin promoter in CBP/p300 knockdown cells (Fig. 5B). In addition, the binding of erythroid specific activator GATA-1 and TAL1 was markedly reduced at the LCR HSs and the γ-globin promoter in the knockdown cells as shown by the ChIP assay (Fig. 5C and D). NF-E2 was less detected at the LCR HS2 (Fig. 5E). However, the transcription of the genes coding these activators was largely maintained in the knockdown cells (Supplementary Fig. 2). Thus, these results might indicate that the hypoacetylation of histones affects HS formation in the LCR and gene promoter, resulting in the reduction of transcription activator binding. The failure of activator binding could be a reason for the disruption of chromatin loops in the β-globin locus.
Con K27me1 C/pi K27me1
0.6 0.4 0.2 0.0
Relative intensity
B 0.20
Con IgG C/pi IgG
0.15
Con K27me2 C/pi K27me2
3.6. Chromatin loop between insulators is maintained in the hypoacetylated β-globin locus
0.10 0.05
Chromatin looping between the LCR HS5 and 3′HS1 is observed in the β-globin locus even the globin genes are not highly transcribed [4,8]. To analyze the relationship of histone acetylation and loop formation between insulators, the relative cross-linking frequency between HS5 and 3′HS1 was measured by the 3C assay in the β-globin locus. When HS5 was used as an anchor in the 3C, the cross-linking frequency with the Gγ-globin gene was reduced by CBP/p300 knockdown, but the frequency with 3′HS1 was almost unaffected (Fig. 6A). In agreement with this result, CTCF was detected at HS5 and 3′HS1 without reduction in the HAT knockdown cells (Fig. 6B). 3′HS1 maintains a hypersensitive chromatin structure in the knockdown cells, even though the sensitivity to DNase I was lower compared to the control cells (Fig. 5B). Thus, these results indicate that histone hyperacetylation is not necessary for chromatin loop formation between the LCR HS5 and 3′HS1 in the β-globin locus.
0.00
C 0.03 Relative intensity
967
Con IgG C/pi IgG
Con K27me3 C/pi K27me3
0.02
0.01
0.00
Fig. 4. Histone modifications of the β-globin locus in CBP/p300 knockdown cells. ChIP was performed with antibodies specific to H3K27me1 (A), H3K27me2 (B) and H3K27me3 (C) in K562 cells expressing shRNA. Relative intensity was calculated as described in Fig. 1D, E. Normal rabbit IgG (IgG) served as experimental control. The results of three independent experiments ± SEM are graphed. *P b 0.05.
Con
M
0
B
C/pi
0
DNase I
DNase I sensitivity
A
4. Discussion This study shows that histone acetylation is required for the chromatin loop formation of the active gene with the LCR and insulators in
120 150U
300U
600U
90 60 30 0 Con C/pi Con C/pi Con C/pi Con C/pi Con C/pi Con C/pi Con C/pi HS5
7.5 5.0 2.5 0.0
D Con IgG C/pi IgG
Con GATA-1 C/pi GATA-1
12.0 9.0 6.0 3.0 0.0
HS3
HS2
E Con IgG C/pi IgG
Con TAL1 C/pi TAL1
Relative intensity
10.0
Relative intensity
Relative intensity
C
HS4
HS1
3.0
γ pro
3'HS1
Con IgG
Con NFE2
C/pi IgG
C/pi NFE2
2.0
1.0
0.0
Fig. 5. DNase I sensitivity at the β-globin LCR HSs, γ-globin promoter and 3′HS1 in CBP/p300 knockdown cells. (A) Nuclei of K562 cells expressing shRNA were digested with 150 to 600 U DNase I. DNA extracted from the digest was run on 1.2% agarose gel. (B) DNase I sensitivity was determined by quantitatively comparing digested DNA with undigested DNA. The results are averages of two independent nuclei preparations ± SEM. ChIP was performed using GATA-1 (C), TAL1 (D) and NF-E2 (E) antibody in K562 cells expressing shRNA and analyzed in described Fig. 1C. Normal rabbit and goat IgG (IgG) served as experimental control. The results of two (C, D) or three (E) independent experiments ± SEM are graphed.
968
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
B
LCR 5 43 21 ε
Gγ Aγ
1.5
δ
β
3'HS1
Relative intensity
Relative cross-linking frequency
A
Con C/pi
1.0
10.0 7.5
Con IgG
Con CTCF
C/pi IgG
C/pi CTCF
5.0 2.5 0.0
0.5
0.0
-20
0
20
40
60
80
Fig. 6. Relative proximity between HS5 and 3′HS1 in the β-globin locus in CBP/p300 knockdown cells. (A) The 3C assay was performed and analyzed as described in Fig. 3. The black shading represents the anchor fragment for HS5 in PCR. The results are averages of four to five independent experiments ± SEM. (B) ChIP was performed using CTCF antibody in K562 cells expressing shRNA and analyzed as described in Fig. 1C. Normal rabbit IgG (IgG) served as experimental control. The results of two independent experiments ± SEM are graphed.
the β-globin locus (Fig. 7). Hyperacetylation of histone appears to play a role in generating HSs in the LCR and promoter. Failure of HS formation might be a reason for the reduced binding of erythroid specific activators. However, the looping between insulators, HS5 and 3′HS1, is not affected by the loss of histone acetylation. Retention of CTCF binding at HS5 and 3′HS1 supports the maintenance of a loop between them. Therefore, histone acetylation might play a role in chromatin loop formation by generating hypersensitive chromatin structure in the LCR/enhancer and promoter, which might allow the binding of activators and coactivators mediating chromatin loop between them. When the chromatin loop is finally formed, the gene will be transcribed.
promoter. When the LCR HSs lose sensitivity after the loss of EKLF or NF-E2, chromatin looping between the LCR and globin gene is reduced in the β-globin locus [8,10,40]. The loop is also reduced when the LCR HSs are not generated after deletion of HS3 and the β-globin promoter sequences [41]. Therefore histone acetylation appears to be required for the formation of the active chromatin hub where the gene promoter, LCR and insulators are closely positioned by looping out intervening
4.1. Histone acetyltransferases for the β-globin locus In this study, the loss of p300 decreases histone acetylation in the β-globin locus, while the reduction of CBP does not affect this modification. CBP and p300 belong to same family of HAT, KAT3 [34]. However, CBP does not appear to be necessary for histone acetylation in the β-globin locus, at least at H3K9/14 and H3K27. It does not provide a redundant role for p300 in this modification. The similar reduction of histone acetylation by the CBP/p300 double knockdown compared to the p300 single knockdown suggests that p300 has a more critical role in histone acetylation at the β-globin locus than CBP, even though both CBP and p300 bind to the LCR. A more powerful role of p300 compared to CBP in H3K27ac is observed at a global protein level in mouse ES cells, and CBP and p300 are not redundant for this modification [32]. Although some studies show that CBP and p300 are functionally redundant for H3K27ac and H3K18ac at a global protein level [19,35], these HAT are likely to have distinct roles in histone acetylation at the β-globin locus.
TA TA HAT G
5’
γ
3’
Aγ K27me2
δ
β
C/pi ε 5’
4
3
2
1
CT 5 CT 1’
3’
4.2. Histone acetylation is required for active chromatin hub formation in the β-globin locus Histone hypoacetylation by CBP/p300 knockdown inhibits the close positioning of the active Gγ-globin gene with the LCR and 3′HS1 in the β-globin locus as shown in Fig. 3. Hypersensitivity at the LCR HSs and gene promoter is severely decreased in the hypoacetylated locus (Fig. 5). Although CBP/p300 loss could have other effects, histone hypoacetylation might be a primary reason for the failure of HS formation in the β-globin locus, because histone acetylation makes chromatin accessible by weakening interactions between histones and DNA and depleting histones [36–39]. As a result, transcriptional activators mediating a chromatin loop might be unable to bind to the LCR and gene
Ac
Ac
TA
K27me2
β TA
δ
Aγ
Gγ
Transcriptional activators : GATA-1, TAL1, NF-E2
HAT Histone acetyltransferases : CBP, p300 CT
CTCF active HSs inactive HSs
Fig. 7. Summary of the changes by CBP/p300 knockdown in the human β-globin locus in K562 cells. The results presented in Figs. 2–6 were combined and represented in model of the human β-globin locus.
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
regions. This modification might contribute to generate accessible chromatin structure in the regulatory regions for the binding of transcriptional activators and loop-mediating coactivators. 4.3. Chromatin loop formation between insulators is independent from histone acetylation The close positioning between HS5 and 3′HS1 is maintained in the hypoacetylated β-globin locus (Fig. 6), suggesting that histone hyperacetylation is not necessary for generating chromatin loop between insulators. In mouse erythroid progenitor cells where the LCR is hypoacetylated and the β-globin gene is not transcribed yet, HS5 is closely positioned to 3′HS1 [4,42]. Hyperacetylation of histone H3 is established after differentiation of these progenitor cells, which induces the transcription of the globin gene. These observations from erythroid progenitor cells and HAT knockdowns in K562 cells indicate that the loop formation between insulators is independent from histone acetylation and gene transcription and might precede these events. The relatively hypersensitive chromatin structure at 3′HS1 in the hypoacetylated β-globin locus, although not at HS5, might contribute to loop formation between insulators by allowing CTCF binding. Acknowledgements We are grateful to Ann Dean for critical reading and comments. This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology, Republic of Korea (2009-0067060 and 2012R1A1B5001946). Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.bbagrm.2013.04.006. References [1] M. Bartkuhn, R. Renkawitz, Long range chromatin interactions involved in gene regulation, Biochim. Biophys. Acta 1783 (2008) 2161–2166. [2] A. Dean, In the loop: long range chromatin interactions and gene regulation, Brief. Funct. Genomics 10 (2011) 3–10. [3] S. Kadauke, G.A. Blobel, Chromatin loops in gene regulation, Biochim. Biophys. Acta 1789 (2009) 17–25. [4] R.J. Palstra, B. Tolhuis, E. Splinter, R. Nijmeijer, F. Grosveld, W. de Laat, The β-globin nuclear compartment in development and erythroid differentiation, Nat. Genet. 35 (2003) 190–194. [5] B. Tolhuis, R.J. Palstra, E. Splinter, F. Grosveld, W. de Laat, Looping and interaction between hypersensitive sites in the active β-globin locus, Mol. Cell 10 (2002) 1453–1465. [6] S. Kim, Y.W. Kim, S.H. Shim, C.G. Kim, A. Kim, Chromatin structure of the LCR in the human β-globin locus transcribing the adult δ- and β-globin genes, Int. J. Biochem. Cell Biol. 44 (2012) 505–513. [7] J. Dostie, T.A. Richmond, R.A. Arnaout, R.R. Selzer, W.L. Lee, T.A. Honan, E.D. Rubio, A. Krumm, J. Lamb, C. Nusbaum, R.D. Green, J. Dekker, Chromosome Conformation Capture Carbon Copy (5C): a massively parallel solution for mapping interactions between genomic elements, Genome Res. 16 (2006) 1299–1309. [8] Y.W. Kim, S. Kim, C.G. Kim, A. Kim, The distinctive roles of erythroid specific activator GATA-1 and NF-E2 in transcription of the human fetal γ-globin genes, Nucleic Acids Res. 39 (2011) 6944–6955. [9] C.R. Vakoc, D.L. Letting, N. Gheldof, T. Sawado, M.A. Bender, M. Groudine, M.J. Weiss, J. Dekker, G.A. Blobel, Proximity among distant regulatory elements at the β-globin locus requires GATA-1 and FOG-1, Mol. Cell 17 (2005) 453–462. [10] R. Drissen, R.J. Palstra, N. Gillemans, E. Splinter, F. Grosveld, S. Philipsen, W. de Laat, The active spatial organization of the β-globin locus requires the transcription factor EKLF, Genes Dev. 18 (2004) 2485–2490. [11] J. Yang, V.G. Corces, Insulators, long-range interactions, and genome function, Curr. Opin. Genet. Dev. 22 (2012) 86–92. [12] L. Handoko, H. Xu, G. Li, C.Y. Ngan, E. Chew, M. Schnapp, C.W. Lee, C. Ye, J.L. Ping, F. Mulawadi, E. Wong, J. Sheng, Y. Zhang, T. Poh, C.S. Chan, G. Kunarso, A. Shahab, G. Bourque, V. Cacheux-Rataboul, W.K. Sung, Y. Ruan, C.L. Wei, CTCF-mediated functional chromatin interactome in pluripotent cells, Nat. Genet. 43 (2011) 630–638.
969
[13] C. Hou, H. Zhao, K. Tanimoto, A. Dean, CTCF-dependent enhancer-blocking by alternative chromatin loop formation, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 20398–20403. [14] X. Fang, P. Xiang, W. Yin, G. Stamatoyannopoulos, Q. Li, 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, J. Mol. Biol. 365 (2007) 31–37. [15] C. Hou, R. Dale, A. Dean, Cell type specificity of chromatin organization mediated by CTCF and cohesin, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 3651–3656. [16] E. Splinter, H. Heath, J. Kooren, R.J. Palstra, P. Klous, F. Grosveld, N. Galjart, W. de Laat, CTCF mediates long-range chromatin looping and local histone modification in the β-globin locus, Genes Dev. 20 (2006) 2349–2354. [17] N. Rybtsova, E. Leimgruber, Q. Seguin-Estevez, I. Dunand-Sauthier, M. Krawczyk, W. Reith, Transcription-coupled deposition of histone modifications during MHC class II gene activation, Nucleic Acids Res. 35 (2007) 3431–3441. [18] Z. Wang, C. Zang, J.A. Rosenfeld, D.E. Schones, A. Barski, S. Cuddapah, K. Cui, T.Y. Roh, W. Peng, M.Q. Zhang, K. Zhao, Combinatorial patterns of histone acetylations and methylations in the human genome, Nat. Genet. 40 (2008) 897–903. [19] Q. Jin, L.R. Yu, L. Wang, Z. Zhang, L.H. Kasper, J.E. Lee, C. Wang, P.K. Brindle, S.Y. Dent, K. Ge, Distinct roles of GCN5/PCAF-mediated H3K9ac and CBP/p300-mediated H3K18/27 ac in nuclear receptor transactivation, EMBO J. 30 (2011) 249–262. [20] A. Kristjuhan, J. Walker, N. Suka, M. Grunstein, D. Roberts, B.R. Cairns, J.Q. Svejstrup, Transcriptional inhibition of genes with severe histone h3 hypoacetylation in the coding region, Mol. Cell 10 (2002) 925–933. [21] K. Kim, A. Kim, Sequential changes in chromatin structure during transcriptional activation in the β globin LCR and its target gene, Int. J. Biochem. Cell Biol. 42 (2010) 1517–1524. [22] O. Sripichai, C.M. Kiefer, N.V. Bhanu, T. Tanno, S.J. Noh, S.H. Goh, J.E. Russell, C.L. Rognerud, C.N. Ou, P.A. Oneal, E.R. Meier, N.M. Gantt, C. Byrnes, Y.T. Lee, A. Dean, J.L. Miller, Cytokine-mediated increases in fetal hemoglobin are associated with globin gene histone modification and transcription factor reprogramming, Blood 114 (2009) 2299–2306. [23] A. Kim, A. Dean, Developmental stage differences in chromatin sub-domains of the β-globin locus, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 7028–7033. [24] S.J. Bultman, T.C. Gebuhr, T. Magnuson, A Brg1 mutation that uncouples ATPase activity from chromatin remodeling reveals an essential role for SWI/SNF-related complexes in β-globin expression and erythroid development, Genes Dev. 19 (2005) 2849–2861. [25] D.L. Letting, C. Rakowski, M.J. Weiss, G.A. Blobel, Formation of a tissue-specific histone acetylation pattern by the hematopoietic transcription factor GATA-1, Mol. Cell. Biol. 23 (2003) 1334–1340. [26] Y. Cho, S.H. Song, J.J. Lee, N. Choi, C.G. Kim, A. Dean, A. Kim, The role of transcriptional activator GATA-1 at human β-globin HS2, Nucleic Acids Res. 36 (2008) 4521–4528. [27] H. Hagège, P. Klous, C. Braem, E. Splinter, J. Dekker, G. Cathala, W. de Laat, T. Forné, Quantitative analysis of chromosome conformation capture assays (3C-qPCR), Nat. Protoc. 2 (2007) 1722–1733. [28] V.V. Ogryzko, R.L. Schiltz, V. Russanova, B.H. Howard, Y. Nakatani, The transcriptional coactivators p300 and CBP are histone acetyltransferases, Cell 87 (1996) 953–959. [29] A.J. Bannister, T. Kouzarides, The CBP co-activator is a histone acetyltransferase, Nature 384 (1996) 641–643. [30] A. Kim, S.H. Song, M. Brand, A. Dean, Nucleosome and transcription activator antagonism at human β-globin locus control region DNase I hypersensitive sites, Nucleic Acids Res. 35 (2007) 5831–5838. [31] Y.W. Kim, A. Kim, Characterization of histone H3K27 modifications in the β-globin locus, Biochem. Biophys. Res. Commun. 405 (2011) 210–215. [32] D. Pasini, M. Malatesta, H.R. Jung, J. Walfridsson, A. Willer, L. Olsson, J. Skotte, A. Wutz, B. Porse, O.N. Jensen, K. Helin, Characterization of an antagonistic switch between histone H3 lysine 27 methylation and acetylation in the transcriptional regulation of Polycomb group target genes, Nucleic Acids Res. 38 (2010) 4958–4969. [33] F. Tie, R. Banerjee, C.A. Stratton, J. Prasad-Sinha, V. Stepanik, A. Zlobin, M.O. Diaz, P.C. Scacheri, P.J. Harte, CBP-mediated acetylation of histone H3 lysine 27 antagonizes Drosophila Polycomb silencing, Development 136 (2009) 3131–3141. [34] D.C. Bedford, L.H. Kasper, T. Fukuyama, P.K. Brindle, Target gene context influences the transcriptional requirement for the KAT3 family of CBP and p300 histone acetyltransferases, Epigenetics 5 (2010) 9–15. [35] G.A. Horwitz, K. Zhang, M.A. McBrian, M. Grunstein, S.K. Kurdistani, A.J. Berk, Adenovirus small e1a alters global patterns of histone modification, Science 321 (2008) 1084–1085. [36] J.D. Anderson, P.T. Lowary, J. Widom, Effects of histone acetylation on the equilibrium accessibility of nucleosomal DNA target sites, J. Mol. Biol. 307 (2001) 977–985. [37] W.A. Krajewski, P.B. Becker, Reconstitution of hyperacetylated, DNase I-sensitive chromatin characterized by high conformational flexibility of nucleosomal DNA, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 1540–1545. [38] H. Reinke, W. Hörz, Histones are first hyperacetylated and then lose contact with the activated PHO5 promoter, Mol. Cell 11 (2003) 1599–1607. [39] T. Kouzarides, Chromatin modifications and their function, Cell 128 (2007) 693–705. [40] M. Wijgerde, J. Gribnau, T. Trimborn, B. Nuez, S. Philipsen, F. Grosveld, P. Fraser, The role of EKLF in human β-globin gene competition, Genes Dev. 10 (1996) 2894–2902. [41] G.P. Patrinos, M. de Krom, E. de Boer, A. Langeveld, A.M. Imam, J. Strouboulis, W. de Laat, F.G. Grosveld, Multiple interactions between regulatory regions are required to stabilize an active chromatin hub, Genes Dev. 18 (2004) 1495–1509. [42] J. Kooren, R.J. Palstra, P. Klous, E. Splinter, M. von Lindern, F. Grosveld, W. de Laat, β-globin active chromatin Hub formation in differentiating erythroid cells and in p45 NF-E2 knock-out mice, J. Biol. Chem. 282 (2007) 16544–16552.
Contents lists available at SciVerse ScienceDirect
Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbagrm
Histone acetylation contributes to chromatin looping between the locus control region and globin gene by influencing hypersensitive site formation Yea Woon Kim, AeRi Kim ⁎ Department of Molecular Biology, College of Natural Sciences, Pusan National University, Busan 609-735, South Korea
a r t i c l e
i n f o
Article history: Received 22 February 2013 Received in revised form 27 March 2013 Accepted 11 April 2013 Available online 20 April 2013 Keywords: Histone acetylation Chromatin loop Hypersensitive site β-Globin locus Transcription
a b s t r a c t Chromatin loops are formed between enhancers and promoters and between insulators to regulate gene transcription in the eukaryotic genome. These transcription regulatory elements forming loops have highly acetylated histones. To understand the correlation between histone acetylation and chromatin loop formation, we inhibited the expression of histone acetyltransferase CBP and p300 in erythroid K562 cells and analyzed the chromatin structure of the β-globin locus. The proximity between the locus control region (LCR) and the active Gγ-globin gene was decreased in the β-globin locus when histones were hypoacetylated by the double knockdown of CBP and p300. Sensitivity to DNase I and binding of erythroid specific activators were reduced in the hypoacetylated LCR hypersensitive sites (HSs) and gene promoter. Interestingly, the chromatin loop between HS5 and 3′HS1 was formed regardless of the hypoacetylation of the β-globin locus. CTCF binding was maintained at HS5 and 3′HS1 in the hypoacetylated locus. Thus, these results indicate that histone acetylation contributes to chromatin looping through the formation of HSs in the LCR and gene promoter. However, looping between insulators appears to be independent from histone acetylation. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Transcription regulatory elements, such as promoters, enhancers and insulators are generally located in long distances on the eukaryote genome, but often positioned in close proximity in the threedimensional space of nucleus, forming chromatin loops by extruding intervening regions between them [1–3]. A chromatin loop between an enhancer and promoter is formed in actively transcribed gene loci. In the mammalian β-globin locus, the locus control region (LCR), a far upstream enhancer, is closely positioned to the transcriptionally active globin genes in erythroid cells [4–7]. The close positioning between the LCR and its target genes is not generated when the globin genes are not transcribed due to the lack of erythroid specific transcriptional activators [8–10]. Thus chromatin loop formation between the LCR and its target gene appears to be a necessary step for gene transcription that causes physical and functional interaction between them. Insulators participate in chromatin loop formation to block enhancerpromoter interactions or to establish chromatin domains [11–13]. The LCR hypersensitive site (HS) 5 and 3′HS1 that are located at the
Abbreviations: LCR, locus control region; HS, hypersensitive site; HAT, histone acetyltransferase; shRNA, short hairpin RNA; ChIP, chromatin immunoprecipitation; RT-PCR, reverse transcription-PCR; 3C, chromosome conformation capture ⁎ Corresponding author. Tel.: +82 51 510 3683; fax: +82 51 513 9258. E-mail address: [email protected] (A. Kim). 1874-9399/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbagrm.2013.04.006
boundaries of the β-globin locus are closely positioned, forming a chromatin loop in erythroid cells [4,5,14]. These sites form part of an active chromatin hub with the LCR HSs and globin gene, when the gene is highly transcribed. Close positioning between insulators is dependent on the binding of CTCF protein [15,16]. The reduction of CTCF increases repressive histone modifications and decreases gene transcription in the human β-globin locus, in addition to the loss of a chromatin loop between HS5 and 3′HS1 [15]. However, the looping between insulators does not appear to be sufficient for the globin gene transcription, because the loop is maintained in the β-globin locus when the transcription of the globin genes fails due to the lack of erythroid specific activators [8,10]. The loop between HS5 and 3′HS1 is also observed in erythroid progenitor cells [4]. Specific histone modifications are associated with gene transcription. Among the many kinds of modifications, acetylation occurring at several lysine residues of histones is highly enriched in transcriptionally active genes and transcription regulatory elements [17,18]. Hypoacetylation of histones by the lack of histone acetyltransferases (HAT) inhibits gene transcription [19,20]. In the human and mouse β-globin loci, the level of histone acetylation is increased by transcriptional induction [6,21–23]. The increase of histone acetylation precedes transcription of the β major globin gene in the mouse locus. Transcription inactivation of the β-globin gene after loss of a transcriptional activator or coactivator is accompanied by the hypoacetylation of histones in the locus [24,25]. In spite of many studies showing the correlation of histone acetylation with gene transcription, it is unclear
964
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
whether histone acetylation affects chromatin loop formation and how the acetylation contributes to the loop formation. In our previous study, the chromatin loop between the LCR and G γ-globin gene was not formed when the γ-globin transcription was severely reduced by the knockdown of the erythroid specific activators GATA-1 or NF-E2 [8]. Histone hyperacetylation in the LCR and G γ-globin gene was lost in the GATA-1 knockdown cells, but it was maintained in the NF-E2 knockdown cells. Here, to understand the relationship of histone acetylation to chromatin loop formation, we inhibited histone acetylation of the β-globin locus by reducing histone acetyltransferases CBP and p300 in K562 cells. Chromatin structure including chromatin loops, histone modifications, HS formation and activator binding was examined in the β-globin locus. This study shows a role of histone acetylation in chromatin loop formation through HS formation in the LCR and globin gene promoter. Chromatin loops between the HS5 and 3′HS1 insulator sites appear to be formed in a different manner from loops between the LCR and promoter. 2. Materials and methods 2.1. Lentiviral shRNA in K562 cells CBP and p300 knockdown cells were generated using TRC lentiviral short hairpin RNA (shRNA) vectors as previously described [8]. Briefly, lentiviral vectors (Sigma) were transfected into 293FT cells with Virapower packaging mix (Invitrogen). Viruses were harvested after 3 days and mixed with K562 cells in the presence of 6 μg/ml polybrene. K562 cells were selected in 2 μg/ml of puromycin at 3 days after virus transduction. Knockdown was confirmed by Western blot analysis. Control cells were generated using lentiviral vector not expressing shRNA (pKLO.1). 2.2. Chromatin immunoprecipitation (ChIP) ChIP was carried out as previously described [26]. K562 cells (1 × 10 7) were cross-linked with 1% formaldehyde and then nuclei were isolated by cell lysis. MNase digestion and sonication were performed with nuclei to produce chromatin at a mononucleosome size. Fragmented chromatin was incubated with antibodies after pre-clearing. Protein-DNA complexes were recovered with protein A or G agarose beads and DNA was purified after reverse cross-linking. Antibodies used in this study were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA) for CBP (sc-369), p300 (sc-8981), GATA-1 (sc-1233), TAL1 (sc-12984), and p45/NF-E2 (sc-22827), from Abcam for H3K27ac (ab4729) and from Millipore for H3ac (06–599), H3K27me1 (17–643), H3K27me2 (07–452), H3K27me3 (07–449) and CTCF (07–729). Normal rabbit IgG (sc-2027) and goat IgG (sc-2028) were purchased from Santa Cruz Biotechnology. 2.3. Reverse transcription-PCR (RT-PCR) RNA was prepared from 2 × 10 6 cells using the RNeasy Plus Mini Kit (Qiagen). One microgram of RNA was reverse transcribed with random hexamers using the Superscript III first-strand synthesis system as suggested by the manufacturer (Invitrogen). cDNA was diluted to 400 μl, and 2 μl of cDNA was amplified in a 10 μl reaction volume by quantitative PCR using TaqMan chemistry. The relative intensity of specific cDNA sequences was compared with a genomic DNA standard using the comparative Ct method. 2.4. DNase I sensitivity assay Nuclei were prepared from K562 cells (2 × 107) as previously described [26]. Aliquots of 3 × 106 nuclei were digested with 150–600 U DNase I in a 100 μl volume for 10 min at 25 °C. DNA was purified and run on a 1% agarose gel to visualize the level of digestion. DNase I
sensitivity was determined by quantitatively comparing digested DNA with undigested DNA using PCR and then by inverting the value at each concentration. 2.5. Chromosome conformation capture (3C) The 3C assay was performed as described previously with reduced number of cells [5,8,27]. K562 cells were cross-linked with 1% formaldehyde and nuclei were prepared from 2 × 10 6 cells and resuspended in 0.4 ml of 1 × restriction enzyme buffer. Chromatin was digested with 800 U of Hind III overnight and DNA fragments were ligated with 400 units of T4 ligase for 4 h at 16 °C. Ligated DNA was purified after reverse cross-linking and then amplified by SYBR green qPCR. Cutting ratio of chromatin by Hind III digestion was measured by amplifying DNA extracted from digested chromatin and comparing with DNA from undigested chromatin (Supplementary Fig. 1). Samples having cutting ratio over 70% were analyzed in this assay. Ligation efficiency between fragments and PCR efficiency between primer sets were corrected using control templates that were prepared by mixing, digesting and ligating equimolar amounts of the PCR fragments spanning the restriction enzyme sites of the β-globin locus and the same amount of genomic DNA [27]. The ligation between two fragments was analyzed using the reverse direction primer of each fragment except one in the β-globin locus. The relative cross-linking frequency was determined by comparing DNA ligated between two fragments in 3C samples with DNA ligated randomly in control templates and then by normalizing with the cross-linking frequency in the Ercc3 gene. 2.6. Quantitative PCR DNA obtained from all experiments was quantitatively analyzed using the 7300 Real-time PCR system (Applied Biosystems). PCR was carried out with 200 nmol of TaqMan probes and 900 nmol of primers in a 10 μl reaction volume for cDNA and DNA obtained from the ChIP and DNase I digestion. DNA obtained from 3C and control templates ligated randomly were amplified using SYBR green fluorescence in a 10 μl reaction volume. Data were collected at the threshold where amplification was linear. The locations of PCR amplicons in the β-globin locus are indicated in Fig. 1A. Sequences of primers, TaqMan probes and primers for 3C are provided in our previous study [8]. 3. Results 3.1. Histone acetylation of the β-globin locus is inhibited by the knockdown of p300 CBP and p300 are histone acetyltransferases [28,29] that bind to the β-globin LCR HSs (Fig. 1A) [6,25,30]. To generate histone hypoacetylation of the β-globin locus, we inhibited the expression of CBP or p300 using shRNA in erythroid K562 cells. CBP and p300 proteins were reduced in the knockdown cells without affecting the expression of each other as shown by Western blot analysis (Fig. 1B). The reduction of CBP or p300 decreases its binding at the β-globin LCR HSs, as revealed by the ChIP assay (Fig. 1C). Histone acetylation was analyzed using the ChIP assay for histone H3K9/K14 and H3K27 in the LCR HSs and γ-globin genes. Relative intensity in the assay was calculated against histone H3, instead of input, to measure acetylation level on histones. In CBP knockdown K562 cells, H3K9/K14ac and H3K27ac were not reduced across the β-globin locus compared to control cells (Fig. 1D and E). The reduction of histone acetylation was observed in p300 knockdown cells. Acetylation at H3K9/K14 was markedly reduced in the γ-globin genes, which are normally highly transcribed in K562 cells (Fig. 1D). Acetylation at H3K27ac was also decreased across the β-globin locus (Fig. 1E). These results show that histone acetylation of the β-globin locus can be inhibited by the lack of p300 in K562 cells. p300 appears to play a more major role in histone acetylation of the β-globin locus than CBP.
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
A 5
4
3
2
1
Aγ
Gγ
LCR
ε
965
β
δ
3’HS1
10kb
C
p300 β -tubulin
1.2
0.6
Con IgG CBPi IgG p300i IgG
Con CBP CBPi CBP p300i CBP
0.4 0.2 0.0
E Con IgG CBPi IgG p300i IgG
Con H3ac CBPi H3ac p300i H3ac
0.8
0.4
0.0
0.25 0.20
Con IgG CBPi IgG p300i IgG
Con p300 CBPi p300 p300i p300
0.15 0.10 0.05 0.00
Relative intensity
Relative intensity
D
0.8
Relative intensity
CBP
Relative intensity
B
5.0 4.0
Con IgG CBPi IgG p300i IgG
Con K27ac CBPi K27ac p300i K27ac
3.0 2.0 1.0 0.0
Fig. 1. Knockdown of CBP or p300 in erythroid K562 cells. (A) The human β-globin locus is presented. Vertical arrows indicate DNase I hypersensitive sites in LCR and 3′HS1. The exons of the globin genes are represented by squares. Vertical bars named below the diagram denote the locations of amplicons used in quantitative PCR. (B) Western blotting was performed using antibodies specific to CBP or p300 in protein extract from K562 cells expressing a control, CBP or p300 shRNA (con, CBPi and p300i). Blotting with β-tubulin antibody was used as an experimental control. ChIP was performed with antibodies specific to CBP or p300 (C) in K562 cells expressing each shRNA. Relative intensity was determined by quantitatively comparing input with immunoprecipitated DNA for the indicated amplicons. In ChIP assay using antibodies specific to H3K9/K14ac (D) or H3K27ac (E), relative intensity was calculated by dividing immunoprecipitated DNA intensity into histone H3 intensity to measure histone acetylation level in present histones. Normal rabbit IgG (IgG) served as experimental control. Bars representing IgG are almost non-detectable in graphs, because the relative intensity is very low. The results of three (C) or two (D) independent experiments ± SEM are graphed. *P b 0.05.
3.2. Histone hypoacetylation decreases transcription of the human γ-globin genes To inhibit histone acetylation in the β-globin locus and to remove any possible redundancy effects between CBP and p300, we performed the double knockdown of CBP and p300 in K562 cells. The expression of CBP and p300 was reduced in the double knockdown cells (Fig. 2A) and their binding at the LCR HSs was reduced to less than half compared to control cells (Fig. 2B). The knockdown of CBP and p300 decreased H3K9/K14ac and H3K27ac in the LCR HSs and the active γ-globin genes (Fig. 2C and D). The decrease was not larger than the decrease obtained by knockdown of p300 alone. Next, the transcription of the globin genes was analyzed by quantitative RT-PCR. RNA transcripts at exons 2 and 3 of the γ-globin genes were reduced to about 30% in the CBP/p300 knockdown cells compared to control cells (Fig. 2E). The δ- and β-globin genes remained in a silent status. The transcription of ε-globin gene, which occurs at low level in erythroid K562 cells, was slightly reduced in the HAT knockdown cells. These results show that the hyperacetylation of histones is required for the transcription of the γ-globin genes in the β-globin locus. 3.3. Chromatin looping of the Gγ-globin gene with LCR HSs and 3′HS1 is disrupted by histone hypoacetylation The active Gγ-globin gene is closely positioned with the LCR and 3′ HS1 in K562 cells, forming chromatin loops [8]. To explain the correlation
between histone acetylation and chromatin loop formation, we measured the relative cross-linking frequency between the Gγ-globin gene and LCR HSs by the 3C assay in control cells and CBP/p300 knockdown K562 cells (Fig. 3A). When a fragment containing LCR HS2 was used as an anchor in the 3C assay, the cross-linking frequency to a fragment containing the Gγ-globin gene was reduced to less than half in the knockdown cells (Fig. 3B). The cross-linking frequency between HS2 and 3′HS1 also decreased in the hypoacetylated β-globin locus. A fragment containing the Gγ-globin gene was less ligated with LCR HSs and 3′HS1 in the knockdown cells (Fig. 3C). No new strong ligation between any fragments was established by CBP/p300 knockdown in the β-globin locus. Thus these results indicate that histone hyperacetylation supports chromatin looping among the LCR, Gγ-globin gene and 3′HS1 in the β-globin locus, generating the active chromatin hub [5]. 3.4. Loss of acetylation at H3K27 increases mono- and di-methylation in the LCR and γ-globin genes Histone H3K27 is subject to four kinds of covalent modifications including acetylation. The modifications at H3K27, mono-, di- and tri-methylation and acetylation, affect each other [31–33]. To examine changes of modifications at H3K27 depending on the loss of acetylation, we analyzed the methylation status of the β-globin locus in control cells and CBP/p300 knockdown cells. Mono- and di-methylation were increased in the LCR and the γ-globin gene where acetylation was lost, but the increases were not observed in the inactive β-globin gene
966
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
B
E
CBP p300 β -tubulin
0.8
IgG CBP p300 x 5
0.6 0.4 0.2
C/pi
Con
HS4
Con
HS2
2.0
0.0
C/pi
HS1
D
1.2
Con IgG C/pi IgG
Con H3ac C/pi H3ac
Relative intensity
Relative intensity
C/pi
Con RT+ C/pi RT+
4.0
0.0 Con
C
6.0
Relative intensity
Relataaaive intensity
A
0.8
0.4
0.0
5.0
Con IgG C/pi IgG
4.0
Con K27ac C/pi K27ac
3.0 2.0 1.0 0.0
Fig. 2. Double knockdown of CBP and p300 in K562 cells. (A) Western blotting was performed using antibodies specific to CBP or p300 in protein extract from K562 cells where transduction of control shRNA (Con) or transduction of both CBP and p300 shRNA (C/pi) was performed. ChIP was performed with antibodies specific to CBP or p300 (B) in K562 cells expressing shRNA and analyzed as described in Fig. 1C. The relative intensity of p300 was multiplied by five to use same Y scale with CBP. In ChIP assay using antibodies specific to H3K9/K14ac (C) or H3K27ac (D), relative intensity was calculated as described in Fig. 1D and E. Normal rabbit IgG (IgG) served as experimental control. (E) cDNA was prepared from RNA isolated from K562 cells expressing shRNA and then amplified by quantitative PCR using primers and probes for exons of the ε-, γ-, δ- and β-globin genes. Relative intensity was calculated by comparing cDNA with genomic DNA for the indicated amplicons. The results of three (B, C, D) or two (E) independent experiments ± SEM are graphed. *P b 0.05.
where histones are hypoacetylated (Fig. 4A and B). Tri-methylation at H3K27, established at a very low level in control cells, was unchanged throughout the locus by hypoacetylation (Fig. 4C) [31]. Taken together, the loss of acetylation at H3K27 results in increasing mono and dimethylation. The increase of mono and di-methylation at H3K27, as well as the decrease of acetylation, could affect chromatin loop formation in the β-globin locus.
A
LCR 5
4
3
2
ε
1
Gγ
3.5. Histone hypoacetylation inhibits the formation of HSs in the β-globin locus HS formation in the LCR and gene promoter might be important for chromatin looping in the β-globin locus, because it allows the binding of transcriptional activators and chromatin loop mediating factors. Here, the effect of reduced histone acetylation on HS formation was studied
Aγ
β
δ
3’HS1
HindIII
C
LCR 5 43 21 ε
G
γ
A
γ
δ
β
3'HS1
3.0 Con C/pi
2.5 2.0 1.5 1.0 0.5 0.0 -20
0
20 G
40
60
80
Relative cross-linking frequency
Relative cross-linking frequency
B
10kb LCR 5
4
3
2
1
ε
Gγ
δ β
3'HS1
5.0 Con C/pi
4.0 3.0 2.0 1.0 0.0 -20
-10
0
40
60
80
Fig. 3. Relative proximity between HSs and the γ-globin gene in the β-globin locus in CBP/p300 knockdown cells. The 3C assay was performed with Hind III restriction enzyme in K562 cells expressing shRNA. (A) Hind III sites and PCR primers in the β-globin locus were represented by vertical bars and horizontal arrows, respectively. The black shading represents the anchor fragment for HS2 (B) and Gγ-globin gene (C) in PCR. The gray shadings are fragments generated by Hind III digestion. Relative cross-linking frequency was determined by quantitatively comparing ligated DNA in cross-linked chromatin with control DNA and then normalizing to the cross-linking frequency at the Ercc3 gene. The results are averages of four to five independent experiments ± SEM.
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
Relative intensity
A
1.0
Con IgG C/pi IgG
0.8
by quantitatively measuring sensitivity to DNase I attack (Fig. 5A). DNase I sensitivity was severely decreased at the LCR HS4, HS2, HS1 and the γ-globin promoter in CBP/p300 knockdown cells (Fig. 5B). In addition, the binding of erythroid specific activator GATA-1 and TAL1 was markedly reduced at the LCR HSs and the γ-globin promoter in the knockdown cells as shown by the ChIP assay (Fig. 5C and D). NF-E2 was less detected at the LCR HS2 (Fig. 5E). However, the transcription of the genes coding these activators was largely maintained in the knockdown cells (Supplementary Fig. 2). Thus, these results might indicate that the hypoacetylation of histones affects HS formation in the LCR and gene promoter, resulting in the reduction of transcription activator binding. The failure of activator binding could be a reason for the disruption of chromatin loops in the β-globin locus.
Con K27me1 C/pi K27me1
0.6 0.4 0.2 0.0
Relative intensity
B 0.20
Con IgG C/pi IgG
0.15
Con K27me2 C/pi K27me2
3.6. Chromatin loop between insulators is maintained in the hypoacetylated β-globin locus
0.10 0.05
Chromatin looping between the LCR HS5 and 3′HS1 is observed in the β-globin locus even the globin genes are not highly transcribed [4,8]. To analyze the relationship of histone acetylation and loop formation between insulators, the relative cross-linking frequency between HS5 and 3′HS1 was measured by the 3C assay in the β-globin locus. When HS5 was used as an anchor in the 3C, the cross-linking frequency with the Gγ-globin gene was reduced by CBP/p300 knockdown, but the frequency with 3′HS1 was almost unaffected (Fig. 6A). In agreement with this result, CTCF was detected at HS5 and 3′HS1 without reduction in the HAT knockdown cells (Fig. 6B). 3′HS1 maintains a hypersensitive chromatin structure in the knockdown cells, even though the sensitivity to DNase I was lower compared to the control cells (Fig. 5B). Thus, these results indicate that histone hyperacetylation is not necessary for chromatin loop formation between the LCR HS5 and 3′HS1 in the β-globin locus.
0.00
C 0.03 Relative intensity
967
Con IgG C/pi IgG
Con K27me3 C/pi K27me3
0.02
0.01
0.00
Fig. 4. Histone modifications of the β-globin locus in CBP/p300 knockdown cells. ChIP was performed with antibodies specific to H3K27me1 (A), H3K27me2 (B) and H3K27me3 (C) in K562 cells expressing shRNA. Relative intensity was calculated as described in Fig. 1D, E. Normal rabbit IgG (IgG) served as experimental control. The results of three independent experiments ± SEM are graphed. *P b 0.05.
Con
M
0
B
C/pi
0
DNase I
DNase I sensitivity
A
4. Discussion This study shows that histone acetylation is required for the chromatin loop formation of the active gene with the LCR and insulators in
120 150U
300U
600U
90 60 30 0 Con C/pi Con C/pi Con C/pi Con C/pi Con C/pi Con C/pi Con C/pi HS5
7.5 5.0 2.5 0.0
D Con IgG C/pi IgG
Con GATA-1 C/pi GATA-1
12.0 9.0 6.0 3.0 0.0
HS3
HS2
E Con IgG C/pi IgG
Con TAL1 C/pi TAL1
Relative intensity
10.0
Relative intensity
Relative intensity
C
HS4
HS1
3.0
γ pro
3'HS1
Con IgG
Con NFE2
C/pi IgG
C/pi NFE2
2.0
1.0
0.0
Fig. 5. DNase I sensitivity at the β-globin LCR HSs, γ-globin promoter and 3′HS1 in CBP/p300 knockdown cells. (A) Nuclei of K562 cells expressing shRNA were digested with 150 to 600 U DNase I. DNA extracted from the digest was run on 1.2% agarose gel. (B) DNase I sensitivity was determined by quantitatively comparing digested DNA with undigested DNA. The results are averages of two independent nuclei preparations ± SEM. ChIP was performed using GATA-1 (C), TAL1 (D) and NF-E2 (E) antibody in K562 cells expressing shRNA and analyzed in described Fig. 1C. Normal rabbit and goat IgG (IgG) served as experimental control. The results of two (C, D) or three (E) independent experiments ± SEM are graphed.
968
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
B
LCR 5 43 21 ε
Gγ Aγ
1.5
δ
β
3'HS1
Relative intensity
Relative cross-linking frequency
A
Con C/pi
1.0
10.0 7.5
Con IgG
Con CTCF
C/pi IgG
C/pi CTCF
5.0 2.5 0.0
0.5
0.0
-20
0
20
40
60
80
Fig. 6. Relative proximity between HS5 and 3′HS1 in the β-globin locus in CBP/p300 knockdown cells. (A) The 3C assay was performed and analyzed as described in Fig. 3. The black shading represents the anchor fragment for HS5 in PCR. The results are averages of four to five independent experiments ± SEM. (B) ChIP was performed using CTCF antibody in K562 cells expressing shRNA and analyzed as described in Fig. 1C. Normal rabbit IgG (IgG) served as experimental control. The results of two independent experiments ± SEM are graphed.
the β-globin locus (Fig. 7). Hyperacetylation of histone appears to play a role in generating HSs in the LCR and promoter. Failure of HS formation might be a reason for the reduced binding of erythroid specific activators. However, the looping between insulators, HS5 and 3′HS1, is not affected by the loss of histone acetylation. Retention of CTCF binding at HS5 and 3′HS1 supports the maintenance of a loop between them. Therefore, histone acetylation might play a role in chromatin loop formation by generating hypersensitive chromatin structure in the LCR/enhancer and promoter, which might allow the binding of activators and coactivators mediating chromatin loop between them. When the chromatin loop is finally formed, the gene will be transcribed.
promoter. When the LCR HSs lose sensitivity after the loss of EKLF or NF-E2, chromatin looping between the LCR and globin gene is reduced in the β-globin locus [8,10,40]. The loop is also reduced when the LCR HSs are not generated after deletion of HS3 and the β-globin promoter sequences [41]. Therefore histone acetylation appears to be required for the formation of the active chromatin hub where the gene promoter, LCR and insulators are closely positioned by looping out intervening
4.1. Histone acetyltransferases for the β-globin locus In this study, the loss of p300 decreases histone acetylation in the β-globin locus, while the reduction of CBP does not affect this modification. CBP and p300 belong to same family of HAT, KAT3 [34]. However, CBP does not appear to be necessary for histone acetylation in the β-globin locus, at least at H3K9/14 and H3K27. It does not provide a redundant role for p300 in this modification. The similar reduction of histone acetylation by the CBP/p300 double knockdown compared to the p300 single knockdown suggests that p300 has a more critical role in histone acetylation at the β-globin locus than CBP, even though both CBP and p300 bind to the LCR. A more powerful role of p300 compared to CBP in H3K27ac is observed at a global protein level in mouse ES cells, and CBP and p300 are not redundant for this modification [32]. Although some studies show that CBP and p300 are functionally redundant for H3K27ac and H3K18ac at a global protein level [19,35], these HAT are likely to have distinct roles in histone acetylation at the β-globin locus.
TA TA HAT G
5’
γ
3’
Aγ K27me2
δ
β
C/pi ε 5’
4
3
2
1
CT 5 CT 1’
3’
4.2. Histone acetylation is required for active chromatin hub formation in the β-globin locus Histone hypoacetylation by CBP/p300 knockdown inhibits the close positioning of the active Gγ-globin gene with the LCR and 3′HS1 in the β-globin locus as shown in Fig. 3. Hypersensitivity at the LCR HSs and gene promoter is severely decreased in the hypoacetylated locus (Fig. 5). Although CBP/p300 loss could have other effects, histone hypoacetylation might be a primary reason for the failure of HS formation in the β-globin locus, because histone acetylation makes chromatin accessible by weakening interactions between histones and DNA and depleting histones [36–39]. As a result, transcriptional activators mediating a chromatin loop might be unable to bind to the LCR and gene
Ac
Ac
TA
K27me2
β TA
δ
Aγ
Gγ
Transcriptional activators : GATA-1, TAL1, NF-E2
HAT Histone acetyltransferases : CBP, p300 CT
CTCF active HSs inactive HSs
Fig. 7. Summary of the changes by CBP/p300 knockdown in the human β-globin locus in K562 cells. The results presented in Figs. 2–6 were combined and represented in model of the human β-globin locus.
Y.W. Kim, A. Kim / Biochimica et Biophysica Acta 1829 (2013) 963–969
regions. This modification might contribute to generate accessible chromatin structure in the regulatory regions for the binding of transcriptional activators and loop-mediating coactivators. 4.3. Chromatin loop formation between insulators is independent from histone acetylation The close positioning between HS5 and 3′HS1 is maintained in the hypoacetylated β-globin locus (Fig. 6), suggesting that histone hyperacetylation is not necessary for generating chromatin loop between insulators. In mouse erythroid progenitor cells where the LCR is hypoacetylated and the β-globin gene is not transcribed yet, HS5 is closely positioned to 3′HS1 [4,42]. Hyperacetylation of histone H3 is established after differentiation of these progenitor cells, which induces the transcription of the globin gene. These observations from erythroid progenitor cells and HAT knockdowns in K562 cells indicate that the loop formation between insulators is independent from histone acetylation and gene transcription and might precede these events. The relatively hypersensitive chromatin structure at 3′HS1 in the hypoacetylated β-globin locus, although not at HS5, might contribute to loop formation between insulators by allowing CTCF binding. Acknowledgements We are grateful to Ann Dean for critical reading and comments. This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology, Republic of Korea (2009-0067060 and 2012R1A1B5001946). Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.bbagrm.2013.04.006. References [1] M. Bartkuhn, R. Renkawitz, Long range chromatin interactions involved in gene regulation, Biochim. Biophys. Acta 1783 (2008) 2161–2166. [2] A. Dean, In the loop: long range chromatin interactions and gene regulation, Brief. Funct. Genomics 10 (2011) 3–10. [3] S. Kadauke, G.A. Blobel, Chromatin loops in gene regulation, Biochim. Biophys. Acta 1789 (2009) 17–25. [4] R.J. Palstra, B. Tolhuis, E. Splinter, R. Nijmeijer, F. Grosveld, W. de Laat, The β-globin nuclear compartment in development and erythroid differentiation, Nat. Genet. 35 (2003) 190–194. [5] B. Tolhuis, R.J. Palstra, E. Splinter, F. Grosveld, W. de Laat, Looping and interaction between hypersensitive sites in the active β-globin locus, Mol. Cell 10 (2002) 1453–1465. [6] S. Kim, Y.W. Kim, S.H. Shim, C.G. Kim, A. Kim, Chromatin structure of the LCR in the human β-globin locus transcribing the adult δ- and β-globin genes, Int. J. Biochem. Cell Biol. 44 (2012) 505–513. [7] J. Dostie, T.A. Richmond, R.A. Arnaout, R.R. Selzer, W.L. Lee, T.A. Honan, E.D. Rubio, A. Krumm, J. Lamb, C. Nusbaum, R.D. Green, J. Dekker, Chromosome Conformation Capture Carbon Copy (5C): a massively parallel solution for mapping interactions between genomic elements, Genome Res. 16 (2006) 1299–1309. [8] Y.W. Kim, S. Kim, C.G. Kim, A. Kim, The distinctive roles of erythroid specific activator GATA-1 and NF-E2 in transcription of the human fetal γ-globin genes, Nucleic Acids Res. 39 (2011) 6944–6955. [9] C.R. Vakoc, D.L. Letting, N. Gheldof, T. Sawado, M.A. Bender, M. Groudine, M.J. Weiss, J. Dekker, G.A. Blobel, Proximity among distant regulatory elements at the β-globin locus requires GATA-1 and FOG-1, Mol. Cell 17 (2005) 453–462. [10] R. Drissen, R.J. Palstra, N. Gillemans, E. Splinter, F. Grosveld, S. Philipsen, W. de Laat, The active spatial organization of the β-globin locus requires the transcription factor EKLF, Genes Dev. 18 (2004) 2485–2490. [11] J. Yang, V.G. Corces, Insulators, long-range interactions, and genome function, Curr. Opin. Genet. Dev. 22 (2012) 86–92. [12] L. Handoko, H. Xu, G. Li, C.Y. Ngan, E. Chew, M. Schnapp, C.W. Lee, C. Ye, J.L. Ping, F. Mulawadi, E. Wong, J. Sheng, Y. Zhang, T. Poh, C.S. Chan, G. Kunarso, A. Shahab, G. Bourque, V. Cacheux-Rataboul, W.K. Sung, Y. Ruan, C.L. Wei, CTCF-mediated functional chromatin interactome in pluripotent cells, Nat. Genet. 43 (2011) 630–638.
969
[13] C. Hou, H. Zhao, K. Tanimoto, A. Dean, CTCF-dependent enhancer-blocking by alternative chromatin loop formation, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 20398–20403. [14] X. Fang, P. Xiang, W. Yin, G. Stamatoyannopoulos, Q. Li, 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, J. Mol. Biol. 365 (2007) 31–37. [15] C. Hou, R. Dale, A. Dean, Cell type specificity of chromatin organization mediated by CTCF and cohesin, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 3651–3656. [16] E. Splinter, H. Heath, J. Kooren, R.J. Palstra, P. Klous, F. Grosveld, N. Galjart, W. de Laat, CTCF mediates long-range chromatin looping and local histone modification in the β-globin locus, Genes Dev. 20 (2006) 2349–2354. [17] N. Rybtsova, E. Leimgruber, Q. Seguin-Estevez, I. Dunand-Sauthier, M. Krawczyk, W. Reith, Transcription-coupled deposition of histone modifications during MHC class II gene activation, Nucleic Acids Res. 35 (2007) 3431–3441. [18] Z. Wang, C. Zang, J.A. Rosenfeld, D.E. Schones, A. Barski, S. Cuddapah, K. Cui, T.Y. Roh, W. Peng, M.Q. Zhang, K. Zhao, Combinatorial patterns of histone acetylations and methylations in the human genome, Nat. Genet. 40 (2008) 897–903. [19] Q. Jin, L.R. Yu, L. Wang, Z. Zhang, L.H. Kasper, J.E. Lee, C. Wang, P.K. Brindle, S.Y. Dent, K. Ge, Distinct roles of GCN5/PCAF-mediated H3K9ac and CBP/p300-mediated H3K18/27 ac in nuclear receptor transactivation, EMBO J. 30 (2011) 249–262. [20] A. Kristjuhan, J. Walker, N. Suka, M. Grunstein, D. Roberts, B.R. Cairns, J.Q. Svejstrup, Transcriptional inhibition of genes with severe histone h3 hypoacetylation in the coding region, Mol. Cell 10 (2002) 925–933. [21] K. Kim, A. Kim, Sequential changes in chromatin structure during transcriptional activation in the β globin LCR and its target gene, Int. J. Biochem. Cell Biol. 42 (2010) 1517–1524. [22] O. Sripichai, C.M. Kiefer, N.V. Bhanu, T. Tanno, S.J. Noh, S.H. Goh, J.E. Russell, C.L. Rognerud, C.N. Ou, P.A. Oneal, E.R. Meier, N.M. Gantt, C. Byrnes, Y.T. Lee, A. Dean, J.L. Miller, Cytokine-mediated increases in fetal hemoglobin are associated with globin gene histone modification and transcription factor reprogramming, Blood 114 (2009) 2299–2306. [23] A. Kim, A. Dean, Developmental stage differences in chromatin sub-domains of the β-globin locus, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 7028–7033. [24] S.J. Bultman, T.C. Gebuhr, T. Magnuson, A Brg1 mutation that uncouples ATPase activity from chromatin remodeling reveals an essential role for SWI/SNF-related complexes in β-globin expression and erythroid development, Genes Dev. 19 (2005) 2849–2861. [25] D.L. Letting, C. Rakowski, M.J. Weiss, G.A. Blobel, Formation of a tissue-specific histone acetylation pattern by the hematopoietic transcription factor GATA-1, Mol. Cell. Biol. 23 (2003) 1334–1340. [26] Y. Cho, S.H. Song, J.J. Lee, N. Choi, C.G. Kim, A. Dean, A. Kim, The role of transcriptional activator GATA-1 at human β-globin HS2, Nucleic Acids Res. 36 (2008) 4521–4528. [27] H. Hagège, P. Klous, C. Braem, E. Splinter, J. Dekker, G. Cathala, W. de Laat, T. Forné, Quantitative analysis of chromosome conformation capture assays (3C-qPCR), Nat. Protoc. 2 (2007) 1722–1733. [28] V.V. Ogryzko, R.L. Schiltz, V. Russanova, B.H. Howard, Y. Nakatani, The transcriptional coactivators p300 and CBP are histone acetyltransferases, Cell 87 (1996) 953–959. [29] A.J. Bannister, T. Kouzarides, The CBP co-activator is a histone acetyltransferase, Nature 384 (1996) 641–643. [30] A. Kim, S.H. Song, M. Brand, A. Dean, Nucleosome and transcription activator antagonism at human β-globin locus control region DNase I hypersensitive sites, Nucleic Acids Res. 35 (2007) 5831–5838. [31] Y.W. Kim, A. Kim, Characterization of histone H3K27 modifications in the β-globin locus, Biochem. Biophys. Res. Commun. 405 (2011) 210–215. [32] D. Pasini, M. Malatesta, H.R. Jung, J. Walfridsson, A. Willer, L. Olsson, J. Skotte, A. Wutz, B. Porse, O.N. Jensen, K. Helin, Characterization of an antagonistic switch between histone H3 lysine 27 methylation and acetylation in the transcriptional regulation of Polycomb group target genes, Nucleic Acids Res. 38 (2010) 4958–4969. [33] F. Tie, R. Banerjee, C.A. Stratton, J. Prasad-Sinha, V. Stepanik, A. Zlobin, M.O. Diaz, P.C. Scacheri, P.J. Harte, CBP-mediated acetylation of histone H3 lysine 27 antagonizes Drosophila Polycomb silencing, Development 136 (2009) 3131–3141. [34] D.C. Bedford, L.H. Kasper, T. Fukuyama, P.K. Brindle, Target gene context influences the transcriptional requirement for the KAT3 family of CBP and p300 histone acetyltransferases, Epigenetics 5 (2010) 9–15. [35] G.A. Horwitz, K. Zhang, M.A. McBrian, M. Grunstein, S.K. Kurdistani, A.J. Berk, Adenovirus small e1a alters global patterns of histone modification, Science 321 (2008) 1084–1085. [36] J.D. Anderson, P.T. Lowary, J. Widom, Effects of histone acetylation on the equilibrium accessibility of nucleosomal DNA target sites, J. Mol. Biol. 307 (2001) 977–985. [37] W.A. Krajewski, P.B. Becker, Reconstitution of hyperacetylated, DNase I-sensitive chromatin characterized by high conformational flexibility of nucleosomal DNA, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 1540–1545. [38] H. Reinke, W. Hörz, Histones are first hyperacetylated and then lose contact with the activated PHO5 promoter, Mol. Cell 11 (2003) 1599–1607. [39] T. Kouzarides, Chromatin modifications and their function, Cell 128 (2007) 693–705. [40] M. Wijgerde, J. Gribnau, T. Trimborn, B. Nuez, S. Philipsen, F. Grosveld, P. Fraser, The role of EKLF in human β-globin gene competition, Genes Dev. 10 (1996) 2894–2902. [41] G.P. Patrinos, M. de Krom, E. de Boer, A. Langeveld, A.M. Imam, J. Strouboulis, W. de Laat, F.G. Grosveld, Multiple interactions between regulatory regions are required to stabilize an active chromatin hub, Genes Dev. 18 (2004) 1495–1509. [42] J. Kooren, R.J. Palstra, P. Klous, E. Splinter, M. von Lindern, F. Grosveld, W. de Laat, β-globin active chromatin Hub formation in differentiating erythroid cells and in p45 NF-E2 knock-out mice, J. Biol. Chem. 282 (2007) 16544–16552.