International Journal of
HEMATOLOGY
Molecular Mechanism of High Hemoglobin F Production in Southeast Asian–Type Hereditary Persistence of Fetal Hemoglobin Khaimuk Changsri,a,b Varaporn Akkarapathumwong,b Duangporn Jamsai,a Pranee Winichagoon,a Suthat Fucharoena a b
Thalassemia Research Center, Institute of Science and Technology for Research and Development; Institute of Molecular Biology and Genetics, Mahidol University, Salaya, Nakornpathom, Thailand Received August 31, 2005; received in revised form November 9, 2005; accepted December 23, 2005
Abstract Hereditary persistence of fetal hemoglobin (HPFH) is associated with a high level of hemoglobin F (HbF) synthesis in adult heterozygotes. In this study, 2 of 6 unrelated HPFH Thai families were found to be Southeast Asian–type HPFH (SEA-HPFH) by analyses of the hematologic data and Southern blot hybridization with polymerase chain reaction–amplified DNA probes. DNA mapping with a probe for a -globin fragment showed a 27-kb deletion of DNA that included the -globin gene and the 3 deoxyribonuclease I hypersensitive site 1 (3HS1) sequence downstream. Deletion of the insulator, 3HS1, and the juxtaposition of the HPFH-3 core enhancer downstream to the 3 breakpoint have been postulated to be the cause of high HbF production in these individuals. To test this hypothesis, we transfected K562 cells with 4 different bacterial artificial chromosome constructs containing the enhanced green fluorescent protein (EGFP) gene at the position of the A-globin gene (pEBAC/148:EGFP). Flow cytometry was used to compare EGFP expression from the pEBAC/148:EGFP construct with the HPFH-3 core enhancer immediately 5 to the SEA-HPFH breakpoint (pEnH), from the pEBAC/148:EGFP construct with 8 kb of the breakpoint sequence and the HPFH-3 core enhancer (pSEA-HPFH), and from the construct with 3HS1 followed by the pSEA-HPFH sequence (pSEA-HPFH_3pHS1). The results show that high HbF production in SEA-HPFH occurs from a deletion of the 3HS1 sequence and the juxtaposition of the HPFH-3 enhancer downstream to the -globin gene. Int J Hematol. 2006;83:229-237. doi: 10.1532/IJH97.A20518 ©2006 The Japanese Society of Hematology Key words: HPFH; SEA-HPFH; Hemoglobin F; Enhanced green fluorescent protein (EGFP) reporter system; Molecular mechanism of HPFH
thalassemia patients. The molecular defects in HPFH cases are either deletion of DNA from the -globin gene to 3 of the -globin gene (deletional HPFH) or mutation at the A -globin or G-globin gene promoter (nondeletional HPFH). Known mutations include a deletion of 13 nucleotides, a nucleotide substitution, and an insertion [2]. High HbF production in deletional HPFH may occur from loss of the -globin gene–silencing sequences [3,4], competition of the locus control region (LCR) with the -globin gene promoter [3,4], or juxtaposition of enhancer sequences [5-11]. Point mutations at the -globin gene promoter cause high HbF production by changing transcription factor binding, which elicits -globin gene expression [12-19]. The LCR consists of 5 deoxyribonuclease I hypersensitive sites (HSs) that are distributed between 6 kb and 20 kb 5 to the -globin gene. The LCR plays a critical role in -globin gene expression by maintaining an open chromatin state and acting as a powerful enhancer of globin gene transcription. Another HS (3HS1), which is approximately 20 kb 3 to the
1. Introduction Hereditary persistence of fetal hemoglobin (HPFH) and -thalassemia are associated with a high level of hemoglobin F (HbF) synthesis (5%-30%) in the adult individual. HPFH is usually characterized by normal values for hematologic parameters, whereas individuals with -thalassemia have mild to moderate hypochromia and microcytosis [1,2]. The mechanism of high HbF production in the adult has been investigated for the purposes of providing molecular modulation of HbF stimulation and gene therapy in Correspondence and reprint requests: Suthat Fucharoen, MD, Thalassemia Research Center, Institute of Science and Technology for Research and Development, Mahidol University, 25/25, Phutthamonthon 4 Rd, Salaya, Nakornpathom 73170, Thailand; 66-28892558; fax: 66-28892559 (e-mail:
[email protected];
[email protected]).
229
230
Changsri et al / International Journal of Hematology 83 (2006) 229-237
-globin gene, may mark the 3 boundary of the -globin gene domain [20]. Three enhancer sequences have been identified from transient transfection and transgenic mice studies: the HPFH-1 enhancer sequence (shared in HPFH-1 and HPFH-2), the HPFH-6 enhancer sequence, and the HPFH-3 enhancer sequence (shared in HPFH-3 and HPFH-4) [5,10,21,22]. The first enhancer is marked by an HS sequence extending from 1.1 kb to 1.4 kb 3 to the breakpoint of the HPFH-1 deletion that is then brought to within 10 kb to 15 kb of the -globin gene in HPFH-1 and HPFH-2. The second enhancer, which appears to enhance only the -globin gene, lies between the breakpoint of the HPFH-6 deletion and that of the Chinese G A ( )-thalassemia deletion. The third region, the 0.7-kb HPFH-3 core enhancer sequence, has been identified to be immediately 3 to the HPFH-3 3 breakpoint and 2 kb downstream of the HPFH-4 breakpoint [5]. Its activity is restricted to the -globin and -globin gene promoters, and there is no effect on the -globin or -globin gene enhancers. Moreover, in the absence of the LCR, the HPFH-3 core enhancer sequence can enhance A-globin gene expression in the embryonic and fetal stages of life [5]. The human 3HS1 is located approximately 20 kb downstream from the -globin gene. 3HS1 has been postulated to act as an insulator to prevent downstream enhancers from activating the -globin genes; thus, the absence of a 3HS1 insulator would allow the downstream enhancers to reactivate the -globin genes.A number of naturally occurring thalassemias and deletional HPFHs (except HPFH-5) have lost the 3HS1 sequence [1]. The region is very A/T rich and contains topoisomerase II recognition sequences as well as several consensus binding motifs for GATA-1 and AP-1/ NF-E2. Functional studies have shown that the region serves as a scaffold-attachment region in erythroid and nonerythroid cells [23]. The CTCC transcription factor (CTCF)binding sites, which are located at the most distal HS of
mouse and human -globin loci, have been found to function as insulators preventing the enhancer effect of HS2 on the A -globin promoter [24]. In this study, 2 unrelated Thai families were found to exhibit Southeast Asian–type HPFH (SEA-HPFH) in analyses of the hematologic data and Southern blot hybridization data, which showed a 27-kb DNA deletion including the -globin gene and the 3HS1 sequence of the -globin gene cluster. We postulated that deletion of 3HS1 and the juxtaposition of the HPFH-3 core enhancer sequence were the mechanisms leading to high HbF production. We tested our hypothesis by using an enhanced green fluorescent protein (EGFP) reporter system in a eukaryotic bacterial artificial chromosome (EBAC) containing the human -globin gene cluster. The effects of the HPFH-3 enhancer sequence and human 3HS1 on -globin gene expression were studied by transient transfection of EGFP-modified EBACs in erythroleukemia cell line K562.
2. Materials and Methods 2.1. Subjects and Hematologic Analysis We studied 2 unrelated Thai patients, a pregnant woman aged 37 years (proband of family A) and a 25-year-old man (proband of family B) (Table 1). Informed consent was given under the guidelines of the Human Experimental Committee of Mahidol University,Thailand.The proband of family A attended the antenatal clinic for a second pregnancy, which was characterized by a slight anemia. Her first child, a 14-year-old boy, had a previous diagnosis of a mild form of -thalassemia/HbE disease because of the presence of HbE and HbF. The proband of family B entered the hospital with a history of anemia and was reported to have homozygous -thalassemia due to the presence of an abnormal hemoglobin type (A2F). A study of the family revealed that he had
Table 1. Hematologic Parameter Values, Hemoglobin (Hb) Analysis, -Globin Genotypes, and -Globin Haplotypes of Members of 2 Families with Hereditary Persistence of Fetal Hemoglobin* Analysis
Sex/Age, y
Hb, g/dL
MCV, fL
Family A Son
M/14
12.1
59
M/37 F/37
11.7 9.2
Family B Son
M/25
Father
Father Mother‡
Mother‡
Hb Type
-Globin Genotype
HbF, %
HbA2, %
EF
47.5
(E) 45.4
–/E
65 83
EFA A2FA
2.6 30.4
(E) 93.0 3.4
E/E –/N
12.8
76
A2F
96.4
3.6
–/35§
M/48
11.5
66
A2FA
2.1
5.6
N/35§
F/45
14.4
83
A2FA
15.1
3.0
–/N
Haplotype† +, – –, + ND +, – –, +
+ – + +, – + – +, + + + – + +, – – – –, + +
+, – + – + +, –, + – – – –, + + –, + – – – –, + + –, – – – – –, + + +, – + – + +, –, + – – – –, + +
*Probands are underlined. MCV indicates mean corpuscular volume; HbF, hemoglobin F; M, male; E, mutation in codon 26 of the -globin gene (GAG → AAG); ND, not determined; F, female; N, normal. †For -globin haplotype (–158 G/XmnI, /HincII, A/HindIII, G/HindIII, /HincII, 3/HincII, /AvaII, 3/BamHI), + indicates presence of the restriction site; –, absence of the restriction site; , absence of the digested fragment. ‡Heterozygote. §Mutation in codon 35 (C → A) of the -globin gene.
HbF Production in SEA-HPFH
Table 2. Primer Pairs for Synthesis of Digoxigenin-Labeled Probes and Probe Location on the -Globin Gene Cluster* Gene
Location
G
Promoter
Coding region
Exon 1
IVS 1–exon 2
231
1 minute. An aliquot of the DIG-incorporated PCR product was denatured and used directly without purification as a DNA probe in hybridization experiments.
Primer Sequence (5 → 3) F-AAATCCTGGACCTATGCCTA R-TGAAGGGTGCTTCCTTT F-GAACAGAAGTTGAGATAGAGA R-ACTCAGTGGTCTTGTGGGCT F-GAGTCAACTTGTTAAAACAT R-GGACTCACCCTGAAGTT F-ATATGTGTACACATATTG R-CAGCACACAGACCAGCACGTT
*F indicates forward; R, reverse; IVS, intervening sequence.
inherited the -thalassemia gene from the father (HbA2, 5.6% of total Hb) and a gene for high HbF production from the mother. Details of the hematologic parameter data and -globin genotyping data for the 2 families are presented in Table 1. The proband of family A and the mother of the family B proband have been shown to have hemoglobin type A2FA with elevated HbF levels of 30.4% and 15.1%, respectively. The diagnosis of HPFH was made because these patients had normal clinical features and normal values for hematologic parameters.
2.2. DNA Analysis Ten milliliters of blood were collected with EDTA as an anticoagulant. The buffy coat was isolated, and DNA was extracted by means of the phenol-chloroform method. thalassemia mutations were characterized by reverse dotblot hybridization [25], and screening for HPFH-6, Vietnamese -thalassemia, and Chinese G(A)-thalassemia was carried out by means of the polymerase chain reaction (PCR), as previously described [26]. The -globin gene haplotype was investigated with a PCR-based restriction fragment polymorphism analysis as described elsewhere [27].
2.3. Preparation of Nonradioactively Labeled DNA Probes by the PCR The amplified DNA probe was prepared by incorporating digoxigenin (DIG)-labeled uridine (Roche Diagnostics, Mannheim, Germany) during the PCR reaction and using specific pairs of primers located at the -globin gene promoter, at the -globin gene, at exon 1 of the -globin gene, and between intervening sequence 1 and exon 2 of the -globin gene (Table 2). The PCR reaction consisted of 200 ng normal DNA, 5 pmol of each primer, 1.5 U AmpliTaq Gold DNA polymerase (PerkinElmer, Norwalk, CT, USA), 1 mM MgCl2, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, and 0.2 mM of each deoxynucleoside triphosphate in a final volume of 50 L. Half of the deoxythymidine triphosphate (0.1 mM) was replaced by its homologue, DIG-11–deoxyuridine triphosphate (DIGdUTP). Amplification was performed in a DNA thermal cycler (GeneAmp PCR System 2400; PerkinElmer). The initial denaturation step performed at 94C for 12 minutes was followed by 30 cycles of denaturation at 94C for 30 seconds, annealing at 55C for 45 seconds, and extension at 72C for
2.4. Nonradioactive Southern Blot Hybridization Nonradioactive Southern blot hybridization was developed to detect the deletion in the -globin gene cluster. Ten micrograms of DNA were digested with various restriction endonucleases (New England Biolabs, Beverly, MA, USA). The digested DNA was fractionated on a 0.8% agarose gel, denatured, blotted onto a nylon membrane (Zeta-Probe, Bio-Rad, Hercules, CA, USA), and hybridized with the denatured DIG-labeled DNA probe for 16 to 20 hours at 61C. The membranes were subjected to autoradiography with x-ray film (Fuji Photo Film, Tokyo, Japan) for 5 to 15 minutes.
2.5. Gap PCR and DNA Sequencing To confirm the type of SEA-HPFH deletion, we carried out PCR analyses of DNA from the affected individuals with primers FH-1, FH-4, and FH-5 to produce a 565-bp normal fragment and a 376-bp fragment specific to the SEA-HPFH genotype, as has previously been described [28]. The amplified fragment encompassing the breakpoint of the deletional chromosome was obtained by using primers FH1 and FH4 and then was subjected to DNA sequencing. The sequencing reaction was carried out with the ABI Prism Dye Primer Cycle Sequencing Ready Reaction kit (Applied Biosystems, Foster City, CA, USA) and analyzed on an ABI Prism 377 DNA sequencer (Applied Biosystems).
2.6. EGFP Expression Analysis in EBACs Plasmids pEBAC/148 and pEBAC/148:GAEGFP were kindly provided by the Cell and Gene Therapy Research Group, The Murdoch Children’s Research Institute, Royal Children’s Hospital, Melbourne, Australia. pEBAC/148 is a BAC clone that harbors a 185-kb genomic fragment containing the 73-kb sequence of the -globin locus (GenBank accession no. U10317) flanked by 92 kb of sequence at the 5 end and 20 kb of sequence at the 3 end [29]. pEBAC/148:GAEGFP containing the EGFP gene under control of the G-globin promoter was used as a positive control of transfection into K562 cells.
2.7. Construction of EBAC Vectors To construct the EGFP-modified EBAC/148 (pEBAC/ 148:EGFP), we used a modification of a previously described technique [30,31]. The pEGFP-N22 vector was used as a template in the amplification reaction to generate a 2.8-kb EGFP/Kan cassette flanked by 50 nucleotides with homology to the promoter and to the poly-A tail sequence of the A-globin gene at the 5 and 3 ends, respectively.The cassette was introduced into pEBAC/148 by homologous recombination in Escherichia coli strain DY380. The EGFP/ Kan cassette replaced the A-globin gene in frame from the start codon of the A-globin gene.
232
Changsri et al / International Journal of Hematology 83 (2006) 229-237
pEBAC/148:EGFP was used as the backbone plasmid to produce 3 more EGFP-modified EBAC constructs by homologous recombination (Figure 1). Plasmid pEnH contains the 0.7-kb HPFH-3 core enhancer sequence, plasmid pSEA-HPFH contains 8.0 kb of DNA downstream of the 3 breakpoint of SEA-HPFH plus the 0.7-kb HPFH-3 core enhancer sequence, and plasmid pSEA-HPFH_3pHS1 contains the 1-kb -globin 3HS1 sequence 5 to the 8 kb of DNA downstream of the 3 breakpoint plus the 0.7-kb HPFH-3 core enhancer sequence. Plasmids pEnH and pSEA-HPFH were constructed from normal DNA by individually amplifying the 0.7-kb HPFH-3 core enhancer sequence and the 8.0 kb of DNA downstream of the 3 breakpoint of SEA-HPFH plus the 0.7-kb HPFH-3 core enhancer sequence. The amplified fragments were inserted into pEBAC/148:EGFP by homologous recombination to replace sites between the 5 breakpoint of SEAHPFH (GenBank accession no. AF042277) and 17 kb 3 (downstream) of the -globin cluster. Plasmid pSEAHPFH_3pHS1 was produced by homologous recombination of the amplified fragment of the 1-kb 3HS1 sequence, which contained the CTCF sequence, to the 5 sequence of the DNA fragment of pSEA-HPFH (8.0 kb of DNA downstream of the 3 breakpoint of SEA-HPFH plus the 0.7-kb HPFH-3 core enhancer sequence).
2.8. Homologous Recombination PCR fragment recombination was performed within E coli strain DY380 (kindly provided by the Cell and Gene Therapy Research Group, The Murdoch Children’s Research Institute, Royal Children’s Hospital) as described previously [32]. DY380 is a bacteriophage recombination system host in which the red , red, and gam genes are under the control of a temperature-sensitive promoter. Briefly, 1 g of purified PCR product was electroporated (Gene-Pulser II; Bio-Rad, Hercules, CA, USA) into 50 L of electrocompetent E coli DY380 cells, as previously described.
2.9. Transfection of EGFP-Modified EBACs into K562 Cells K562 cells were maintained continuously in Dulbecco modified Eagle medium (Gibco BRL, Gaithersburg, MD, USA) supplemented with 10% fetal calf serum (Gibco BRL), 100 U/mL penicillin, and 100 U/mL streptomycin. The cell density was maintained between 2 105 cells/mL and 8 105 cells/mL, and cells were incubated at 37C in a humidified incubator with 5% carbon dioxide. Plasmid pEBAC/148:GAEGFP was used as positive control for transient transfection. The transfection of each vector was carried out for 4 hours with DMRIE-C (Invitrogen, Carlsbad, CA, USA). A ratio of 2:1 was used for EGFPmodified EBAC and pEBAC/148:GAEGFP constructs. For the establishment of an episomal-maintenance culture, hygromycin was added to 400 g/mL after 48 hours for cells transfected with EGFP-modified EBACs, and this concentration was maintained throughout the growth of the cell cultures.
2.10. EGFP Expression Analysis The percentage of EGFP-expressing K562 cells and the median peak fluorescence (MPF) of EGFP expression were measured by flow cytometry using a FACSCalibur instrument (BD Medical Systems, Los Angeles, CA, USA). Transfected K562 cells (10,000-12,000 cells) were assayed at 2 days posttransfection and subsequently grown for 40 days in media containing hygromycin before further analysis. Data were acquired and analyzed with FACSCalibur instrument software or with WinMDI software, version 2.8 (The Scripps Research Institute, La Jolla, CA, USA; http://facs.scripps. edu/software.html).
3. Results 3.1. -Globin Genotyping and Haplotype Analysis Reverse dot-blot hybridization analysis showed that the husband and son of the family A proband had the E-globin gene and that the proband in family B carried a -thalassemia mutation in codon 35 (C → A), which was inherited from his father. The results of screening for common HPFHs and -thalassemias were negative for both families.A haplotype analysis of the family members for DNA polymorphic sites on the -globin gene cluster clearly showed that there were no deletions around the -globin, -globin, and -globin genes, because the allelic loci had different restriction patterns, and a chromosome-harboring haplotype (–158 G/XmnI, +; /HincII, –; G/HindIII, +; A/HindIII, –; /HincII, +; 3/HincII, +) was associated with high HbF production. However, deletion might have occurred at the -globin gene, because only 1 pattern (+ +) was detected for /AvaII and 3/BamHI sites (Table 1).
3.2. Detection of Gene Deletion by Southern Blot Hybridization and PCR-Based Techniques To detect the deletion fragment in the -globin gene cluster of the proband’s DNA, we performed Southern blotting with PCR-amplified fragments of the -globin, -globin, globin, and -globin genes as DNA probes to hybridize with DNA digested with various restriction enzymes. Restriction maps for BglII, XbaI, and EcoRI sites of the normal -globin gene cluster that are within the area of the deletion breakpoint of SEA-HPFH are shown in Figure 2. Although the normal-sized fragment was found in the EcoRI-digested DNA of the proband after hybridization with the -globin gene probe, the intensity of the 5.2-kb -globin gene fragment was much less than that of the normal control (Figure 2A), suggesting that this DNA region is deleted from one chromosome. The coexistence of the normal and abnormal bands in the DNA samples from proband A indicates a heterozygous state for this region. Abnormal fragments of 8.5 kb and 10.3 kb detected when the DNA of proband A was digested with XbaI and BglII and hybridized with the DIG-labeled -globin gene probe confirmed the heterozygosity of the 2 alleles (Figures 2B and 2C). Similar results were found with Southern blotting of DNA from proband B (data not shown). Positive bands for the -globin
HbF Production in SEA-HPFH
Figure 1.
233
Structure of enhanced green fluorescent protein (EGFP)-modified eukaryotic bacterial artificial chromosome (EBAC) constructs and EGFP-expressing control plasmid. A, EBAC/148 contains 185 kb of the human -globin cluster flanked by NotI restriction sites. The A-globin gene within the -globin cluster was replaced by the EGFP gene to yield pEBAC/148:EGFP (backbone) (construct 1). The backbone was used to produce another 3 constructs: pEnH (construct 2), pSEA-HPFH (construct 3), and pSEA-HPFH_3pHS1 (construct 4). The insert fragment of each construct was flanked by NotI restriction sites (illustration not to scale). The vertical dotted arrow locates the 5 breakpoint of Southeast Asian–type hereditary persistence of fetal hemoglobin (SEA-HPFH), which is approximately 2 kb upstream of the -globin gene and approximately 18 kb downstream of the A-globin promoter. pEnH contains the 0.7-kb HPFH-3 core enhancer sequence located immediately 5 to the breakpoint of SEA-HPFH. pSEAHPFH contains the 8-kb 3 breakpoint sequence followed by the 0.7-kb HPFH-3 core enhancer sequence. pSEA-HPFH_3pHS1 was modified from the backbone such that the 1-kb 3HS1 sequence was followed by the 8-kb 3 breakpoint sequence and the 0.7-kb HPFH-3 core enhancer sequence. B, pEBAC/148:GAEGFP is the EGFP-expressing control plasmid. It was modified from EBAC/148 in that the EGFP gene replaced the G-globin gene and downstream sequence until the -globin gene.
234
Changsri et al / International Journal of Hematology 83 (2006) 229-237
Figure 2. Southern blot hybridization of DNA from the proband of family A. Ten-microgram samples of proband A DNA and normal DNA were digested with EcoRI and hybridized with the -globin gene probe (A). The -globin gene probe was used to hybridize XbaI-digested DNA (B) and BglII-digested DNA (C). The positions and sizes of abnormal bands from XbaI and BglII digestion are indicated (arrows). The restriction map of BglII, XbaI, and EcoRI sites of normal DNA around the 5 breakpoint of the Southeast Asian–type hereditary persistence of fetal hemoglobin (SEAHPFH) deletion fragment on the -globin gene cluster is illustrated (D). The sizes of normal bands hybridized with the -globin gene probe are shown for each restriction enzyme digestion. M indicates DIGmarker II (sizes correspond to those of HindIII markers); P, proband DNA; N, normal DNA. gene fragment appeared after hybridization with the -globin gene probe (Figure 2C) because cross-hybridization occurred between the similar sequences of the -globin and -globin genes. However, these signals can be differentiated by the sizes of the restriction fragments. The results of DNA mapping indicate that the 5 breakpoint is located between the -globin and -globin genes. This type of deletion is similar to the previously reported SEA-HPFH deletion [33]. PCR analysis of the DNA from the probands of families A and B also showed the 27-kb deletion in 1 allele of the heterozygote and compoundheterozygote members of both families (Figure 3). To identify the deletion breakpoint, we amplified the PCR product from the abnormal allele and directly sequenced it by using primer F1 as a sequencing primer. We demonstrated the breakpoint sequence, as previously reported for the SEAHPFH deletion (data not shown). The 5 breakpoint located approximately 2 kb upstream of the -globin gene and the 3 breakpoint located approximately 5 kb downstream of the 3HS1 sequence indicate a deletion that covers 27 kb of the -globin gene and downstream sequences. High HbF production in SEA-HPFH possibly results from juxtaposition of a 3 enhancer downstream to the -globin gene and/ or removal of the 3HS1 sequence.
3.3. EGFP Expression in Transient Transfection into K562 Cells The mechanism of high HbF production in SEA-HPFH was investigated by transient transfection of EGFP-modified EBAC constructs into human erythroleukemia cells. The EGFP gene replaced the A-globin gene in the modified
Figure 3.
Results of gap polymerase chain reaction (PCR) analysis for detection of Southeast Asian–type hereditary persistence of fetal hemoglobin (SEA-HPFH) in 2 unrelated families. The 565-bp PCR fragment was amplified from the normal allele with the FH4/FH5 primer pair. The 376-bp PCR fragment was amplified from the SEAHPFH deletion allele with the FH1/FH5 primer pair. Shown are PCR products from the DNA of the probands and their families separated in a 1.0% agarose gel. The mothers (Mo) and sons (S) of both families show a heterozygous state and show a -globin genotype indicating a heterozygote for SEA-HPFH and a compound heterozygote for SEAHPFH and -thalassemia, respectively. Each of the fathers (F) had a -thalassemia trait and showed the same band as normal DNA (N). Lane M, 100-bp DNA markers; lane 1, normal DNA; lanes 2-4, DNA samples from family A; lanes 5-7, DNA samples from family B.
HbF Production in SEA-HPFH
235
Table 3. Enhanced Green Fluorescent Protein (EGFP) Expression of the Construct-Transfected K562 Cells* Construct pEBAC/148:EGFP (backbone) pEnH pSEA-HPFH pSEA-HPFH_3pHS1 pEBAC/148:GAEGFP (control)
2 Days Posttransfection EGFP-Expressing Cells, % 2.9 3.3 2.6 2.4 3.1
0.1 0.3 0.1 0.2 0.3
MPF
157.3 378.0 367.7 147.0 268.3
7.5 8.2 15.4 7.5 36.2
40 Days Posttransfection EGFP-Expressing Cells, % MPF 31.0 37.7 30.3 26.3 27.3
3.6 3.5 2.5 2.1 3.5
174.3 395.0 372.7 167.3 275.0
5.1 13.2 8.0 10.7 17.2
*Data are expressed as the mean SD of 3 independent experiments. MPF indicates median peak fluorescence.
EBAC construct, with its expression coming under the control of the A-globin gene promoter. Four EGFP-modified EBAC constructs (Figure 1) were transiently transfected into K562 cells. Transfection with plasmid pEBAC/148:GAEGFP was used as a positive control to represent EGFP-expressing K562 cells with EGFP-modified EBAC constructs, and EGFP expression was under the control of the G-globin promoter. Table 3 presents the levels of EGFP expression for K562 cells transfected with the 4 constructs and the control at 2 days and 40 days after transient transfection. Three independent experiments were performed with each construct. At 2 days posttransfection, the percentage of EGFPexpressing cells of the pEBAC/148:GAEGFP control was 2% to 3%, the same as for the 4 EGFP-modified EBAC constructs. The MPF was 2 times higher than that of the pEBAC/148:EGFP backbone. This result indicates that EGFP expression under the control of the G-globin gene promoter within an EBAC construct was better than that of the A-globin promoter in K562 cells. The K562 cells transfected with the pSEA-HPFH plasmid, as well as with the pEnH plasmid, exhibited an MPF that was 2 to 3 times higher than that of the backbone, indicating that the HPFH-3 core enhancer sequence can enhance A-globin gene expression with or without the 8-kb downstream 3 breakpoint sequence in the pSEA-HPFH plasmid. The MPF of the cells transfected with pSEA-HPFH_3pHS1 showed the same level of EGFP expression as that of the backbone, indicating that the presence of the 3HS1 sequence interrupts the enhancer effect of the downstream HPFH-3 core enhancer. The number of EGFP-expressing cells for the constructs increased to between 26% and 38% at 40 days posttransfection with selection with 400 g/mL hygromycin (Table 3). This result reflects the constant episomal maintenance of the constructs in K562 cells. The MPF for each construct at 40 days posttransfection was not different from that at 2 days posttransfection (Figure 4). The results confirmed the effects of the HPFH-3 core enhancer and 3HS1 on -globin gene expression. These results also indicate that the juxtaposition of the HPFH-3 core enhancer to the SEA-HPFH breakpoint and the removal of 3HS1 in SEA-HPFH elicited the elevation in -globin gene expression.
4. Discussion Previously described Southern blot methods for determining the deletion of DNA fragments have required the use of radioactively labeled probes, as well as the use of plasmids
containing the target gene for the synthesis of the specific DNA probes [34,35]. In this study, we developed the use of nonradioactive probes by incorporating DIG-dUTP during the PCR with DNA fragments of the -globin gene cluster and using normal DNA as a template. With this approach, we were able to detect the SEA-HPFH deletion by using a DIGlabeled -globin gene probe. Our methodology of using DIG-labeled amplified DNA has proved useful for the detection of large DNA deletions. The SEA-HPFH genotype has been observed in Cambodians, Vietnamese, Chinese in Guangdong and Taiwan, and the Karen tribe at the border of Thailand and Myanmar [28,33,36,37]. In this study, the -globin haplotype (– + – + +) in the 2 unrelated families suggests that the mutation has the same origin. The hematologic data, including red blood cell indices, were normal for the 2 families’ mothers, who are heterozygous for HPFH, as has previously been described [1,20]. A
Figure 4. Comparative expression of enhanced green fluorescent protein (EGFP) for the EGFP-modified eukaryotic bacterial artificial chromosome (EBAC) constructs in K562 cell lines at 2 days and 40 days posttransfection. The median peak fluorescence (MPF) represents the flow cytometric quantitation of EGFP expression of EGFP-expressing cells. The EGFP-modified EBAC constructs were pEBAC/148:EGFP (backbone), pEnH, pSEA-HPFH, and pSEA-HPFH_3pHS1; pEBAC/ 148:GAEGFP (control) was the EGFP-expressing control plasmid. Data are presented as the mean SD of 3 independent experiments. TF indicates after transient transfection.
236
Changsri et al / International Journal of Hematology 83 (2006) 229-237
high level of HbF (15%-30% of the total hemoglobin) was detected, and a pancellular distribution of HbF was observed when red blood cells were exposed to acid-elution staining (data not shown). We also report, for the first time, a compound heterozygote of SEA-HPFH with HbE in the son of family A and with a nonsense mutation at codon 35 (C → A) in the proband of family B (Table 1). The patients are asymptomatic, without anemia or jaundice, and their hemoglobin levels are within the normal range. In contrast, compound heterozygotes with HbE for cases of other types of deletional HPFH or -thalassemia commonly exhibit thalassemia intermedia of variable severity [35,38-40]. The persistence of HbF production in deletional HPFH is generally explained by the juxtaposition of the enhancer sequence, which is brought into the vicinity of the fetal -globin gene. In the HPFH-3 deletion, the 0.7-kb core enhancer sequence, consisting of the consensus binding site for the erythroid-specific GATA-1 transcription factor (AGATAA motif) and 7 consensus sites for the ubiquitous CP1 transcription factor (CCAAT motif), is moved to within 5 kb of the A-globin gene. In SEA-HPFH, the effect of the imported HPFH-3 enhancer is persistent, even though it is located more than 25 kb downstream of the A-globin gene. EGFP expression was observed at 2 days posttransfection, although the number of K562 cells transfected with EGFPmodified EBACs was rather low (2%-3%). One possible explanation is that the erythroleukemic K562 cell line has functional proteins for fetal hemoglobin expression, especially the G-globin chain. The production of the A-globin chain is less than that of the G-globin chain in K562 cells, as demonstrated by a G/(A + G) percentage of approximately 60% in the fetal heterozygote. However, the use of K562 cells as a host for expressing adult -globin chains has been shown by Anagnou et al and Haley et al [5,41]. The A-globin gene promoter with the HPFH-3 core enhancer has been evaluated by transient transfection and shown to have the ability to activate a fusion reporter gene by 3-fold in K562 cells [5]. Approaches in which the A-globin promoter with the luciferase gene in a plasmid construct or the A-globin promoter in the context of the -globin cluster within a yeast artificial chromosome is transfected into K562 have also been used for screening pharmaceutical HbF stimulation [41]. Our experiment used K562 cells for A-globin gene expression because K562 cells can be transfected with BACs with high efficiency and because elevated levels of A-globin gene expression, which identify -globin gene stimulation in adults, can be demonstrated with these cells. A BAC is normally maintained at only 1 or 2 copies within K562 cells because its size is approximately 200 kb. To increase the number of EGFP-expressing K562 cells that contained an EGFP-modified EBAC and maintained an episomal status, we used hygromycin for selection and continuous subculture until 40 days posttransfection. K562 cells transfected with an EGFP-modified EBAC construct were selected with hygromycin because the construct contains the hygromycin-resistance gene. The number of EGFP-expressing cells then increased (26%-38%) after selection and continued growing until 40 days posttransfection. A higher proportion, up to 30%, at 2 days after transient transfection with EGFP-modified EBACs was reported after transfection was
carried out in K562-EBNA-1 cells [31]. However, the culture medium of K562-EBNA-1 cells requires special supplements, such as 20% fetal calf serum and antioxidants. The MPFs at 2 days and at 40 days posttransfection were the same for each construct. The higher MPF for the control than for the backbone indicated that EGFP expression under the control of the G-globin gene promoter within an EBAC construct was higher than that of the A-globin gene promoter in K562 cells. The MPF from pEnH-expressing cells confirmed the enhancer effect of the HPFH-3 core enhancer on the A-globin gene. In addition, the cells expressing pSEA-HPFH, which contained the 8 kb of DNA downstream of the 3 breakpoint in the construct, still demonstrated a high MPF, indicating the enhancer effect of the HPFH-3 core enhancer, even though the distance between the HPFH-3 core enhancer and the A-globin promoter is approximately 28 kb. The decrease in A-globin gene expression observed when 3HS1 was present between the A-globin promoter and the HPFH-3 core enhancer in the context of -globin gene cluster confirmed the role of 3HS1 as a 3 insulator of the -globin gene cluster. In summary, the results showed that the 3 HPFH-3 enhancer elevated A-globin gene expression and that the insulator effect of 3HS1 in the environment of the -globin gene cluster can interrupt A-globin gene expression signals from 3 enhancer sequences. We conclude that the mechanism of high HbF levels in SEA-HPFH is caused by coexistence of the 3 HPFH-3 enhancer and the loss of 3HS1. This work also provides evidence for cases of deletional HPFH and some types of -thalassemia that are characterized by high HbF production in the adult due to the loss of 3HS1 and juxtaposition of the 3 enhancer.
Acknowledgments This work was supported by grants from the Thailand Research Fund to S.F. as a Senior Research Scholar and from the Thailand University Development Program. We thank Prof. Dr. Duncan R. Smith for constructive criticisms of this manuscript.
References 1. Wood WG. Hereditary persistence of fetal hemoglobin and -thalassemia. In: Steinberg MH, Forget BG, Higgs DR, et al, eds. Disorders of Hemoglobin: Genetics, Pathophysiology, and Clinical Management. Cambridge, UK: Cambridge University Press; 2001: 356-388. 2. Forget BG. Molecular basis of hereditary persistence of fetal hemoglobin. Ann N Y Acad Sci. 1998;850:38-44. 3. Vitale M, Di Marzo R, Calzolari R, et al. Evidence for a globin promoter-specific silencer element located upstream of the human globin gene. Biochem Biophys Res Commun. 1994;204:413-418. 4. Vitale M, Calzolari R, Di Marzo R, Acuto S, Maggio A. A region upstream of the human -globin gene shows a stage-specific interaction with globin promoters in erythroid cell lines. Blood Cells Mol Dis. 2001;27:874-881. 5. Anagnou NP, Perez-Stable C, Gelinas R, et al. Sequences located 3 to the breakpoint of the hereditary persistence of fetal hemoglobin-3 deletion exhibit enhancer activity and can modify the developmental expression of the human fetal A-globin gene in transgenic mice. J Biol Chem. 1995;270:10256-10263. 6. Arcasoy MO, Romana M, Fabry ME, Skarpidi E, Nagel RL,
HbF Production in SEA-HPFH
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
Forget BG. High levels of human -globin gene expression in adult mice carrying a transgene of deletion-type hereditary persistence of fetal hemoglobin. Mol Cell Biol. 1997;17:2076-2089. Camaschella C, Serra A, Saglio G, et al. The 3 ends of the deletions of Spanish -thalassemia and black HPFH 1 and 2 lie within 17 kilobases. Blood. 1987;70:593-596. Camaschella C, Serra A, Gottardi E, et al. A new hereditary persistence of fetal hemoglobin deletion has the breakpoint within the 3 -globin gene enhancer. Blood. 1990;75:1000-1005. Feingold EA, Forget BG. The breakpoint of a large deletion causing hereditary persistence of fetal hemoglobin occurs within an erythroid DNA domain remote from the -globin gene cluster. Blood. 1989;74:2178-2186. Kosteas T, Palena A, Anagnou NP. Molecular cloning of the breakpoints of the hereditary persistence of fetal hemoglobin type-6 (HPFH-6) deletion and sequence analysis of the novel juxtaposed region from the 3 end of the -globin gene cluster. Hum Genet. 1997;100:441-445. Stolle CA, Penny LA, Ivory S, Forget BG, Benz EJ Jr. Sequence analysis of the -globin gene locus from a patient with the deletion form of hereditary persistence of fetal hemoglobin. Blood. 1990;75: 499-504. Fucharoen S, Shimizu K, Fukumaki Y. A novel C-T transition within the distal CCAAT motif of the G-globin gene in the Japanese HPFH: implication of factor binding in elevated fetal globin expression. Nucleic Acids Res. 1990;18:5245-5253. Nicolis S, Ronchi A, Malgaretti N, Mantovani R, Giglioni B, Ottolenghi S. Increased erythroid-specific expression of a mutated HPFH -globin promoter requires the erythroid factor NFE-1. Nucleic Acids Res. 1989;17:5509-5516. Liu LR, Du ZW, Zhao HL, et al. T to C substitution at –175 or –173 of the -globin promoter affects GATA-1 and Oct-1 binding in vitro differently, but can independently reproduce HPFH phenotype in transgenic mice. J Biol Chem. 2005;280:7452-7459. Jane SM, Gumucio DL, Ney PA, Cunningham JM, Nienhuis AW. Methylation-enhanced binding of Sp1 to the stage selector element of the human -globin gene promoter may regulate development specificity of expression. Mol Cell Biol. 1993;13:3272-3281. Lloyd JA, Lee RF, Lingrel JB. Mutations in two regions upstream of the A globin gene canonical promoter affect gene expression. Nucleic Acids Res. 1989;17:4339-4352. Zhou W, Clouston DR,Wang X, Cerruti L, Cunningham JM, Jane SM. Induction of human fetal globin gene expression by a novel erythroid factor, NF-E4. Mol Cell Biol. 2000;20:7662-7672. Mantovani R, Malgaretti N, Nicolis S, Ronchi A, Giglioni B, Ottolenghi S. The effects of HPFH mutations in the human -globin promoter on binding of ubiquitous and erythroid specific nuclear factors. Nucleic Acids Res. 1988;16:7783-7797. Mantovani R, Superti-Furga G, Gilman J, Ottolenghi S. The deletion of the distal CCAAT box region of the A-globin gene in black HPFH abolishes the binding of the erythroid specific protein NFE3 and of the CCAAT displacement protein. Nucleic Acids Res. 1989; 17:6681-6691. Stamatoyannopoulos G, Grosveld F. Hemoglobin switching. In: Varmus H, ed. The Molecular Basis of Blood Diseases. Philadelphia, Pa: W.B. Saunders; 2001:135-182. Elder JT, Forrester WC, Thomson C, et al. Translocation of an erythroid-specific hypersensitive site in deletion-type hereditary persistence of fetal hemoglobin. Mol Cell Biol. 1990;10:1382-1389. Kosteas T, Pavlou O, Palina A. Complete sequencing and functional analysis of the HPFH-6 enhancer: detection of multiple motifs for transcription factors and identification of an open reading frame [abstract]. Blood. 1996;88:150a. Fleenor DE, Kaufman RE. Characterization of the DNase I hypersensitive site 3 of the human -globin gene domain. Blood. 1993; 81:2781-2790. Farrell CM, West AG, Felsenfeld G. Conserved CTCF insulator ele-
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
237
ments flank the mouse and human -globin loci. Mol Cell Biol. 2002;22:3820-3831. Winichagoon P, Saechan V, Sripanich R, et al. Prenatal diagnosis of -thalassaemia by reverse dot-blot hybridization. Prenat Diagn. 1999;19:428-435. Craig JE, Barnetson RA, Prior J, Raven JL, Thein SL. Rapid detection of deletions causing -thalassemia and hereditary persistence of fetal hemoglobin by enzymatic amplification. Blood. 1994;83: 1673-1682. Sutton M, Bouhassira EE, Nagel RL. Polymerase chain reaction amplification applied to the determination of -like globin gene cluster haplotypes. Am J Hematol. 1989;32:66-69. Xu XM, Li ZQ, Liu ZY, Zhong XL, Zhao YZ, Mo QH. Molecular characterization and PCR detection of a deletional HPFH: application to rapid prenatal diagnosis for compound heterozygotes of this defect with -thalassemia in a Chinese family. Am J Hematol. 2000;65:183-188. Narayanan K, Williamson R, Zhang Y, Stewart AF, Ioannou PA. Efficient and precise engineering of a 200 kb -globin human/ bacterial artificial chromosome in E. coli DH10B using an inducible homologous recombination system. Gene Ther. 1999;6: 442-447. Orford M, Nefedov M,Vadolas J, Zaibak F,Williamson R, Ioannou PA. Engineering EGFP reporter constructs into a 200 kb human -globin BAC clone using GET recombination. Nucleic Acids Res. 2000; 28:E84. Vadolas J, Wardan H, Orford M, et al. Development of sensitive fluorescent assays for embryonic and fetal hemoglobin inducers using the human -globin locus in erythropoietic cells. Blood. 2002;100: 4209-4216. Lee CH, Yu D, Martinez de Velasco J, et al. A highly efficient Escherichia coli-based chromosome engineering system adapted for recombinogenic targeting and subcloning of BAC DNA. Genomics. 2001;73:56-65. Dimovski AJ, Divoky V, Adekile AD, et al. A novel deletion of ~27 kb including the -globin gene and the locus control region 3HS-1 regulatory sequence: -thalassemia or hereditary persistence of fetal hemoglobin? Blood. 1994;83:822-827. Fucharoen S, Winichagoon P, Chaicharoen S, Wasi P. Different molecular defects of G(A)-thalassaemia in Thailand. Eur J Haematol. 1987;39:154-160. Winichagoon P, Fucharoen S, Thonglairoam V, Wasi P. Thai G(A)-thalassemia and its interaction with a single -globin gene on a chromosome carrying -thalassemia. Hemoglobin. 1990;14:185-197. Trachoo O, Sura T, Sakuntabhai A, et al. Molecular characterization of hereditary persistence of fetal hemoglobin in the Karen people of Thailand. Hemoglobin. 2003;27:97-104. Motum PI, Hamilton TJ, Lindeman R, Le H, Trent RJ. Molecular characterisation of Vietnamese HPFH. Hum Mutat. 1993;2: 179-184. Fucharoen S, Pengjam Y, Surapot S, Fucharoen G, Sanchaisuriya K. Molecular characterization of ()/-thalassemia and ()-thalassemia/hemoglobin E in Thai patients. Eur J Haematol. 2001;67:258-262. Fucharoen S, Pengjam Y, Surapot S, Fucharoen G, Sanchaisuriya K. Molecular and hematological characterization of HPFH-6/Indian deletion-inversion G(A)-thalassemia/HbE in Thai patients. Am J Hematol. 2002;71:109-113. Fucharoen S, Fucharoen G, Sanchaisuriya K, Surapot S. Molecular characterization of thalassemia intermedia associated with HPFH6/-thalassemia and HPFH-6/Hb E in Thai patients. Acta Haematol. 2002;108:157-161. Haley JD, Smith DE, Schwedes J, et al. Identification and characterization of mechanistically distinct inducers of -globin transcription. Biochem Pharmacol. 2003;66:1755-1768.
