Cell Biol. Int. (2012) 36, 47–55 (Printed in Great Britain)
Research Article
Possible interaction between B1 retrotransposon-containing sequences and bmajor globin gene transcriptional activation during MEL cell erythroid differentiation Ioannis S. Vizirianakis1, Sotirios S. Tezias, Elsa P. Amanatiadou and Asterios S. Tsiftsoglou Laboratory of Pharmacology, Department of Pharmaceutical Sciences, Aristotle University of Thessaloniki, GR-54124 Thessaloniki, Greece
Abstract Repetitive sequences consist of .50% of mammalian genomic DNAs and among these SINEs (short interspersed nuclear elements), e.g. B1 elements, account for 8% of the mouse genome. In an effort to delineate the molecular mechanism(s) involved in the blockade of the in vitro differentiation program of MEL (murine erythroleukaemia) cells by treatment with methylation inhibitors, we detected a DNA region of 559 bp in chromosome 7 located downstream of the 39-end of the bmajor globin gene (designated B1-559) with unique characteristics. We have fully characterized this B1-559 region that includes a B1 element, several repeats of ATG initiation codons and consensus DNA-binding sites for erythroid-specific transcription factors NF-E2 (nuclear factor-erythroid-derived 2), GATA-1 and EKLF (erythroid Kru¨ppel-like factor). Fragments derived from B1-559 incubated with nuclear extracts form protein complexes in both undifferentiated and differentiated MEL cells. Transient reporter-gene experiments in MEL and human erythroleukaemia K-562 cells with recombinant constructs containing B1-559 fragments linked to HS-2 (hypersensitive site-2) sequences of human b-globin gene LCR (locus control region) indicated potential cooperation upon erythropoiesis and globin gene expression. The possible interaction between the B1-559 region and bmajor globin gene transcriptional activation upon execution of erythroid MEL cell differentiation programme is discussed. Keywords: B1 retrotransposon element; differentiation; erythropoiesis; globin genes; locus control region; MEL cells; methylation inhibitors; promoter activity; silent DNA sequences
1. Introduction One of the most provocative questions in biology is the potential role of short interspersed repetitive elements in genome evolution and the expression and/or repression of mammalian genes organized as transcriptional units throughout the genome (Schumann et al., 2010). Moreover, it is not known whether possible involvement of such elements affects cell fate, i.e. self-renewal, differentiation and even cell death. Genomic DNA repeat sequences serve as mobile elements that may affect gene transcription (Kramerov and Vassetzky, 2005; Ostertang and Kazazian, 2005). Repetitive elements have been now considered to participate in mechanisms involved in the regulation of gene activation or silencing. In particular, the presence of Alu elements within the 39-UTR (39-untranslated region) of genes may lead to their targeted silencing (Chen et al., 2008). Furthermore, the creation of 2 isoforms of rodent NK (natural killer) cell-activating receptor NKG2D gene may be driven by a B1 retrotransposon insertion, leading to modulated gene regulation upon development (Lai et al., 2009). Such an example of gene transcriptional regulation by retrotransposon elements acting as alternative promoter has been proposed for both human and rodent BIRC1 (baculoviral IAP repeat-containing 1) gene coding for NAIP (neuronal apoptosis
inhibitory protein; Romanish et al., 2007). Novel transcripts of NAIP gene consistently arise from transcription initiation start sites located within an Alu retrotransposon element. Interestingly, the generated RNA transcripts of NAIP gene encode truncated protein molecules. Some of these NAIP-like proteins lack their functional caspase-sequestering motifs, suggesting a potential novel, although yet elusive, intracellular function (Romanish et al., 2009). Previous reports also suggest involvement of Alu-like repetitive sequences in the coordinated post-transcriptional control of gene expression (Vidal et al., 1993), whereas non-globin DNA sequences located in large distance from the Gc-globin and other globin genes are potentially homologous to an RNA polymerase III template (Duncan et al., 1979). B1 retrotransposon elements can recruit transcription factors like PAX6 to bind at discrete sites in the mouse genome (Zhou et al., 2000) and LINE-1 (long interspersed nuclear element-1; L1) carry p53 DNA-binding sites (Harris et al., 2009). These examples support the hypothesis that repetitive sequences in the eukaryotic genome are modulatory elements in TRCs (transcriptional regulatory circuits) ensuring coordinated expression of genes organized in transcriptional units like globin genes and many others. Our previous studies have shown that MEL (murine erythroleukaemia) cells – a model system used for erythroid differentiation studies (Tsiftsoglou et al., 2003) – treated with methylation
1
To whom correspondence should be addressed (email
[email protected]). Abbreviations: AP-1, activator protein 1; BRCA2, breast-cancer susceptibility gene 2; EKLF, erythroid Kru¨ppel-like factor; EMSA, electrophoretic mobility-shift assay; HMBA, hexamethylene-bis-acetamide; HS-2, hypersensitive site-2; LCR, locus control region; MEL, murine erythroleukaemia; miRNA, microRNA; NAIP, neuronal apoptosis inhibitory protein; NF-E2, nuclear factor-erythroid-derived 2; SINE, short interspersed nuclear element; TRC, transcriptional regulatory circuit; 39-UTR, 39-untranslated region.
E The Author(s) Journal compilation E 2012 Portland Press Limited
Volume 36 (1) N pages 47–55 N doi:10.1042/CBI20110236 N www.cellbiolint.org
47
Potential role of repetitive elements on transcription
inhibitors (DNA/RNA methylation blockers) in the presence of inducer DMSO block erythroid differentiation and accumulate relatively short polyA2 RNA transcripts in their cytoplasm, although they continue to express the bmajor globin gene (Vizirianakis and Tsiftsoglou, 1996, 2005). To delineate the molecular mechanism(s) implicated in the blockade of the in vitro differentiation programme of MEL cells by methylation inhibitors, we have detected a region of 559 bp in chromosome 7 located downstream of the 39-end of the bmajor globin gene (designated B1-559) using a 7.304 bp genomic DNA hybridization probe. Further analysis and DNA mapping indicated that this region includes a B1 element, several ATG initiation codons, and consensus binding sites for erythroid-specific transcription factors NF-E2 (nuclear factor-erythroid-derived 2), GATA-1 and EKLF (erythroid Kru¨ppel-like factor) (Vizirianakis and Tsiftsoglou, 2005). We have attempted to: (i) further map and functionally characterize the B1-559 region to demonstrate whether this domain shares structural homology with the promoter sequences of bmajor globin gene and (ii) investigate whether this distant element exerts a possible cooperative interaction between the B1-559 region and globin gene transcriptional activation during erythroid differentiation of MEL cells. To this end, fragments of 112 and 123 bp derived from this region were assessed for their capacity to recruit transacting factors and form stable complexes detected by EMSA (electrophoretic mobility-shift assay). The ability of the B1-559 region to drive transient transcriptional activation of a reporter gene (luciferase) has also been investigated in cooperation with the DNase I HS-2 (hypersensitive site-2) DNA sequences of LCR (locus control region) derived from human b-globin gene cluster. The data obtained have shown that the B1-559 DNA region can recruit transcription factors and drive the expression of luciferase gene in cooperation with HS-2, raising the possibility of potential interaction with bmajor globin gene transcriptional activation upon the execution of MEL cell erythroid differentiation programme.
2. Materials and methods
2.2. Cell cultures Cells employed were human erythroleukaemia K-562 cells and MEL-745PC-4A, a clone of MEL-745 cells obtained after subcloning and subsequent testing of clones derived for high degree of inducibility. All cultures were maintained in RPMI 1640 and DMEM (Dulbecco’s modified Eagle’s medium) respectively containing 10% (v/v) FCS (fetal calf serum; Gibco) and antibiotics (penicillin and streptomycin 100 mg/ml). Cells were incubated at 37uC in a humidified atmosphere containing 5% CO2 in air and maintained at densities that permitted logarithmic growth (MEL: 16105–16106 cells/ml; and K-562: 16104–16105 cells/ml). Cell growth was assessed with the use of a hemocytometer under a light microscope.
2.3. Induction and assessment of differentiation Cells were incubated with no addition and/or with the inducing agent (MEL: DMSO and/or HMBA; K-562: hemin) as indicated under each Figure. At certain time intervals during incubation, the proportion of differentiated (Hb-producing cells) was assessed cytochemically by using benzidine/H2O2 solution as described elsewhere (Vizirianakis and Tsiftsoglou, 1996).
2.4. EMSAs DNA fragments 112 bp and 123 bp generated by PCR, as mentioned earlier, were also used as probes to carry out bandshift assays. The nuclear extracts derived from MEL cells, the 59-end probe labelling and the gel-shift assay were performed as described by Ausubel et al. (1993). Briefly, each 59-end [32P]-labelled DNA probe was incubated in the presence of 1 mg poly(dI-dC)?(dI-dC) with soluble nuclear protein extracts (1–1.5 mg) derived from control-untreated or DMSO-treated MEL cells and allowed to form DNA–protein-binding complexes at 4uC for 30 min in an appropriate buffer (10 mM Hepes, pH 7.9, 50 mM KCl, 5 mM MgCl2, 1 mM EDTA and 1 mM dithiothreitol). Protein concentration was determined using Bradford reagent.
2.1. Chemicals and biochemicals DMSO was purchased from Mallinckrodt and HMBA (hexamethylenebis-acetamide) and hemin from Sigma. All restriction enzymes and T4 DNA ligase were obtained from New England Biolabs. Taq DNA polymerase was purchased from Promega. Tryptone and yeast extract were delivered by Gibco Laboratories. LipofectamineTM 2000 and CellfectinTM were purchased from Invitrogen Corporation, whereas Opti-MEM was obtained from Gibco Life Technologies. The constructs pGL3-Basic and pGL3-Control were obtained from Promega. PEV-Neo plasmid bearing the HS-2 of LCR of human b-globin genes was donated by Dr G Patrinos (Department of Cell Biology and Genetics, Erasmus Medical Centre, Rotterdam, The Netherlands), whereas the CMVb-Gal plasmid was supplied by Dr M. Arsenakis (Department of Biology, Aristotle University of Thessaloniki, Greece). The 7.304 bp mouse genomic sequence bearing the bmajor globin gene locus referred to in this work has the accession number X14061 in the GenBankj/EMBL database.
48
www.cellbiolint.org N Volume 36 (1) N pages 47–55
2.5. Cloning of recombinant plasmid constructs used in reporter gene assays The pEV-Neo plasmid bearing the LCR of human b-globin gene was used as a template in PCR for the generation of DNase I HS-2 DNA fragment by using the following pair of primers: forward primer: 59-GAGCTCCCCTTCCGCATCCTCAT-39 and reverse primer: 59-GCTAGCGTATGTGAGCATGTGTCC-39. The underlined DNA sequences are recognition sites for the restriction enzymes SacI and NheI respectively added to facilitate subsequent cloning. The generated PCR product was then cloned into pGL3-Basic vector and its sequence was verified by DNA sequencing. Similarly, the pBluescript (+7.3 kb) plasmid bearing the 7.304 bp mouse genomic fragment (representing the region 36884–44192 bp of the sequence with the accession number X14601 at GenBankj/EMBL databank) was used as a template for the generation of 112 and 123 bp DNA fragments derived from the area located within 42201–42759 bp of the above DNA region by
E The Author(s) Journal compilation E 2012 Portland Press Limited
Cell Biol. Int. (2012) 36, 47–55
Figure 1
Map of the different DNA fragments originated from the 7.304 bp genomic DNA fragment and used for mobility band shift and reporter gene promoter studies The mouse genomic 7.304 bp EcoRI/EcoRI DNA fragment bearing the bmajor globin gene at its 59-end is shown in the top (A). Furthermore, the DNA fragments derived from the vicinity of B1 retrotransposon element by PCR (designated 112, 123 and 112–123) and used in band-shift assays are depicted in the boxes at (B, C). Note that the transcription factors whose consensus-binding sites have been identified are shown by arrowheads.
applying the following pair of primers: forward primer for the 112 bp DNA fragment: 59-CGTTGCTAGCTGCCCAGACTTGGCAAATTAG-39 and reverse primer 59-TCACAAGCTTGCAACACTATACCATTCATCC-39; forward primer for the 123 bp DNA fragment: 59-CGTTGCTAGCCTACAGAGTGTGTTCCAGGAC-39 and reverse primer 59-TCACAAGCTTCACTGTGATGATTGATGGCTG-39. The underlined sequence in forward primers represents the recognition sites for the restriction enzyme NheI, whereas that in reverse primers represents the recognition site for the restriction enzyme HindIII. However, an experimental difficulty was found upon trying to clone directly the DNA fragment that could include both the 112 and 123 bp DNA fragments due to existing repeat DNA sequences within these fragments. Indeed, in order to generate the full-length 112–123 bp DNA fragment, the 123 bp DNA fragment was cloned next to the 112 bp DNA fragment by generating the appropriate DNA fragments in PCR by using the following pair of primers for the 123 bp DNA fragment: forward primer 59-CCCAAGCTTCTACAGAGTGTGTTCCAGGAC-39 and reverse primer 59-CATGCCATGGCACTGTGATGATTGATGGCTG-39. The underlined sequences are recognition sites for restriction enzymes HindIII and NcoI respectively. All the generated PCR products were cloned into pGL3-Basic plasmid that bears the luciferase gene, and their sequence was verified by DNA sequencing. The following recombinant constructs were generated that either contained also the HS-2 DNA sequence (pGL3-HS2-112; pGL3-HS2-123; pGL3-HS2-112–123) or not (pGL3-112; pGL3123, pGL3-112–123). The pGL3-Basic (empty) vector was used as negative control. Furthermore, two plasmids served as positive control in promoter assays. The first was the pGL3-Control vector that contains the SV40 (simian virus 40) promoter and was obtained from Promega. The second set included the recombinant constructs pGL3-Gc and pGL3-HS2-Gc that both contain the promoter of human Gc globin gene (260 bp) either in the presence (pGL3-HS2-Gc) or the absence (pGL3-Gc) of HS-2 and they have been generated by PCR using the following pair of primers: forward, 59-GTTGCTAGCTTAAGCAGCAGTATCCTC-39 and reverse, 59-CCCAAGCTTTTGATAACCTC-39. Note that the underlined sequences are recognition sites for restriction enzymes NheI and HindIII, respectively.
E The Author(s) Journal compilation E 2012 Portland Press Limited
2.6. Transient transfection of MEL and K-562 cells with recombinant plasmid vectors and reporter gene assays The transient transfection of MEL and K-562 cells with the generated recombinant constructs (pGL3-112; pGL3-123, pGL3112–123; pGL3-HS2-112; pGL3-HS2-123; pGL3-HS2-112–123) was carried out according to the manufacturer’s protocol. Briefly, 1 mg of DNA (either from the above-mentioned pGL3 recombinant plasmids or the control vectors) was transfected in triplicate using either 8 ml LipofectamineTM (LipofectamineTM Reagent) or 10 ml of cellfectin (CellfectinTM Transfection Reagent) in MEL (26106) or K-562 (1.56106) cells grown in 60 mm plates. Transfection efficiency was normalized by co-transfection of 1 mg CMV-b-Gal vector DNA used as internal control. After 48 h, the cells were processed for luciferase and b-galactosidase activity according to manufacturer’s protocol (Promega). These experiments were carried out at least twice, and for each recombinant construct the measurements from 3 separate cell cultures (MEL and/or K-562) were used for statistical analysis.
3. Results 3.1. The B1-559 region co-exists in both BALB/c and DBA/2J mouse strains Our previous work, complemented with bioinformatic analysis, revealed a 39-end flanking bmajor globin gene DNA region containing the B1 element and consensus-binding sequences for GATA-1, AP-1 (activator protein 1)/NF-E2 and EKLF (B1-559 region) in the deposited genomic DNA sequence X14061 isolated from mouse strain BALB/c (diagrammatically shown in Figure 1A; Vizirianakis and Tsiftsoglou, 2005). However, before proceeding towards the functional characterization of B1-559, it was reasonable to confirm its existence in the genome of MEL cells that were derived from the mouse strain DBA/2J. In this regard, we cloned and characterized the genomic DNA sequence of interest from MEL cells by PCR using primers based on the genomic DNA
Volume 36 (1) N pages 47–55 N www.cellbiolint.org
49
Potential role of repetitive elements on transcription
sequence deposited in GenBankj/EMBL data bank under accession number X14061 (forward primer: 59-GAGGGGAAGTTTTAAATGCC-39 at position 42201–42220; reverse primer: 59-GTACATTGCTGAGTAAAGTG-39 at position 42600–42619). The cloning and subsequent DNA sequencing of that genomic DNA fragment derived from MEL cells verified the presence of the referred B1-559 bp DNA region, showing almost 100% structural homology with that existing in the genomic DNA sequence X14061 (see results deposited in GenBankj/EMBL databank under the accession number FR667688). The latter indicated that the structure of B1-559 DNA sequence in the 2 BALB/c and DBA/ 2J mouse strains is identical, which permitted us to investigate further the potential function of the existence of the B1 element and the consensus-binding sites for GATA-1, AP-1/NF-E2 and EKLF within this region in MEL cells.
3.2. The B1-559 bp DNA region shares transcription factor recruitment capacity The data obtained thus far indicate that DNA sequences located at distant 39-end flanking bmajor globin gene region may encode RNA transcripts accumulated upon exposure of MEL cells to methylation inhibitors. Four polyA2 RNAs of 448, 391, 174 and 151 nt were cloned and characterized as described in Tezias et al. (2011). Keeping in mind that this region contains consensus sequences recognized by the erythroid-specific transcription factors GATA-1, AP-1/NF-E2 and EKLF in the vicinity of B1 repeat element (Vizirianakis and Tsiftsoglou, 2005), assessment of the
functional characterization of this particular DNA region was considered important. To this end, 3 different DNA fragments derived from the B1-559 DNA region were generated by PCR and used in EMSAs (see section 2 and the diagram depicted in Figure 1B). The first DNA fragment contained the GATA-1 and AP-1/NF-E2 binding sites (designated as 112 bp), the second the binding site for EKLF (designated as 123 bp), and the third all 3 consensus-binding sites for GATA-1, AP-1/NF-E2 and EKLF (designated as 112–123 bp). The identification of the consensus-binding sequences of GATA-1 and NF-E2 in the vicinity of B1 element and the presence of EKLF-binding site within B1 prompted us to investigate the potential protein-binding capacity and the recruitment of transcription factors in the 559 bp DNA region. In this regard, mobility band-shift assays were carried out by using nuclear extracts derived from either control-untreated (undifferentiated) or DMSOtreated (differentiated) MEL cells to assess the recruitment of transcription factors by both 112 and 123 bp DNA fragments. The presence of only one DNA–protein complex has been seen in the case of 123 bp DNA fragment, using nuclear extracts derived from either control-untreated or DMSO-treated MEL cells (Figure 2B). The result suggests that this particular DNA fragment has the ability to bind with protein(s) at one site, although no data are available to identify such molecules. Interestingly, however, when nuclear extracts isolated from DMSO-treated MEL cells were used in band-shift assays, the DNA–protein complex has a higher molecular mass, which implies the binding of another protein(s) to the original DNA–protein complex. At least 3 DNA– protein complexes were detected using the 112 bp DNA fragment with nuclear extracts derived from either control-untreated or DMSO-treated MEL cells (Figure 2A). The data support the notion that the B1-559 DNA region located downstream of the bmajor globin gene may influence globin gene expression through recruitment of specific transcription factors.
3.3. The B1-559 DNA region has the potential to drive luciferase gene transient expression in haemopoietic cells in cooperation with HS-2
Figure 2
Assessment of the binding capacity of proteins in DNA fragments derived from 39-end flanking bmajor globin gene DNA sequences in control and differentiating MEL cells MEL-745PC-4A cells were incubated in culture in the absence (undifferentiated) or presence of DMSO (1.5%, v/v) (differentiated) for 72 h. Nuclear extracts (2: no extract; +: 1 mg; and/or ++: 1.5 mg) were isolated and subjected to EMSAs by using [32P]labelled DNA fragments derived from 39-end flanking bmajor globin gene DNA sequences and specifically that of 112 bp (A) or 123 bp (B), as shown in section 2. The position of the detected DNA–protein complexes is denoted by an arrowhead.
50
www.cellbiolint.org N Volume 36 (1) N pages 47–55
Because the binding sites found in B1-559 region are recognized by erythroid-specific transcriptional factors, this prompted us to demonstrate whether this region can drive the luciferase gene transient expression in erythroid cells in connection with the LCR HS-2 DNA sequences. Both the above-mentioned DNA fragments (112 and 123 bp; see also Figure 1) were further cloned into appropriate pGL3 vectors bearing the luciferase gene (see section 2) and generated the recombinant constructs, some of which contain the HS-2 sequences (pGL3-HS2-112; pGL3-HS2-123; pGL3-HS2-112–123), while others did not (pGL3-112; pGL3-123, pGL3-112–123) (Figure 3). A known globin gene promoter was included to serve as transcription control. Recombinant constructs of pGL3 vector carrying the human Gc globin gene promoter (260 bp) alone (pGL3-Gc) or in connection with HS-2 (pGL3-HS2Gc) were also generated and served as a positive control in the reporter-gene assays. The vector pGL3 (pGL3-Basic) or carrying HS-2 DNA sequences (pGL3-HS2-basic) were also included as negative controls (Figure 3).
E The Author(s) Journal compilation E 2012 Portland Press Limited
Cell Biol. Int. (2012) 36, 47–55
Figure 3
Diagrammatic illustration of pGL3 recombinant constructs used in reporter-gene assay As depicted in (A), the fragments derived from the B1-559 DNA region shown in Figure 1 were used to generate recombinant constructs by cloning into pGL3 vector bearing the reporter gene for luciferase. In particular, these fragments were cloned under the influence (constructs designated as pGL3-HS2112, pGL3-HS2-123 and pGL3-HS2-112–123) (C) or not (constructs designated as pGL3-112, pGL3-123 and pGL3-112–123) (B) of the enhancer HS-2 of LCR of human b-globin genes and used in reporter gene promoter assays, as described in section 2. Furthermore, pGL3 recombinant constructs bearing the human Gc globin gene promoter under the influence (pGL3-HS2-Gc) or not (pGL3-Gc) of HS-2 DNA sequences have also been generated. Finally, the empty pGL3 vector with (pGL3-HS2-Basic) or without (pGL3-Basic) the HS-2 was used in these experiments. Note that human Gc globin gene promoter is used as a positive and the empty pGL3 vector as a negative control.
The recombinant constructs were used to assess the transcriptional activation of the reporter-gene luciferase mediated by the capacity of fragments to act as potential promoters in haemopoietic cells. To this end, a complementary experimental approach was chosen. In particular, the transcriptional gene activation of luciferase driven by the fragments derived from the B1-559 region under the influence or not of HS-2 was measured in a homologous (mouse MEL) or heterologous (human K-562) haemopoietic cell environment. Although they represent a different environment, these 2 cell lines were chosen for being erythroid and because the two DNA fragments under examination (112 and 123 bp) contain consensus-binding sites for the main erythroid transcription factors, NF-E2, GATA-1 and EKLF. Furthermore, these experiments were performed by transfecting and transiently expressing the above-mentioned constructs in either undifferentiated (control-untreated) or differentiated cells (MEL as HMBA-treated and K-562 as hemin-treated). Hemin is a known inducer of human Gc globin gene expression in K-562 cells (Tsiftsoglou et al., 2006). Moreover, although K-562 cells fail to produce adult-type b-globin full-length mRNA transcripts and translate them into nascent b-globin chains producing the two c globin genes (Ac and Gc) (Reddy et al., 1994; Armstrong and Emerson, 1996), earlier studies have shown that hemin increases the production of immature short ended b-like globin RNA transcripts in these cells (Tsiftsoglou et al., 1989). The inducer HMBA for MEL cells grown in culture was chosen in these experiments instead of DMSO due to its known higher differentiation inducing ability in erythroid maturation (Tsiftsoglou et al., 2003). The capacity of human Gc globin gene promoter to drive luciferase gene expression was verified in both control-untreated and HMBA-treated MEL cells (Figure 4A). Interestingly, the transient
E The Author(s) Journal compilation E 2012 Portland Press Limited
expression of luciferase gene was more intense in HMBA-treated MEL cells, especially those transfected with recombinant vectors bearing the HS-2 control element, as expected, based on the known influence of LCR on globin gene promoters. Similar results were obtained with K-562 cells (Figure 5A). In this case, however, the transient activation of luciferase gene expression mediated by human Gc globin gene promoter was stronger than in MEL cells. The latter can be attributed to the fact that, in K-562 culture, both the cells and the structural elements (Gc globin gene promoter and HS-2) are of human origin, thus the cooperation of cis- and trans-elements could be more efficient. As far as the DNA fragments derived from the examined B1-559 DNA region are concerned, the data obtained shows the following: (i) only the 112–123 bp DNA fragment containing all 3 consensus-binding sites for GATA-1, AP-1/NF-E2 and EKLF have a high capacity to drive luciferase gene expression in control untreated MEL cells, that, however, can be suppressed by the presence of HS-2 (Figure 4B); (ii) in contrast to that seen in untreated MEL cells, no such suppression by HS-2 was seen for the 112–123 bp DNA fragment in MEL cells treated with the HMBA inducer. Furthermore, in HMBA-treated MEL cells, the ability of HS-2 to cooperate with the 112 bp DNA fragment that contains only the GATA-1 and AP-1/ NF-E2 binding sequences occurs (Figure 4B); (iii) in control untreated K-562 cells, the capacity of the 112–123 bp DNA fragment to drive luciferase gene expression only occurred in cooperation with HS-2 (Figure 5B); (iv) interestingly, in hemintreated K-562 cells all 3 DNA fragments (112, 123 and 112–123 bp) had the potential to drive luciferase gene expression in cooperation with HS-2, but to different degrees. Comparatively, the 112–123 bp DNA fragment possesses a higher ability (Figures 4B and 5B). The data indicate the capacity of the 559 bp DNA region to drive
Volume 36 (1) N pages 47–55 N www.cellbiolint.org
51
Potential role of repetitive elements on transcription
Figure 4
Assessment of promoter-like activity of pGL3 recombinant constructs in undifferentiated and differentiated MEL cells The generated pGL3 constructs were separately transfected as indicated in section 2 in MEL, with conditions where the cells were in undifferentiated or differentiated state. In particular, MEL cells were grown in culture in the absence (undifferentiated) or presence of the inducing agent HMBA (5 mM) (differentiated). The assays for luciferase activity in transfected MEL cells were performed as shown in section 2 and the results obtained were expressed in relative luciferase units. (A) Indicates the luciferase activity data obtained in MEL cells transfected with the control constructs, the empty pGL3 vector alone (basic) or carrying the human Gc globin gene promoter (Gc) either in the absence or presence of HS-2. Similarly (B) presents the data obtained by using the pGL3 constructs carrying the 112, 123 and 112–123 bp DNA fragments in the absence or presence of HS-2, as shown in Figure 3.
luciferase gene expression in both MEL and K-562 cells, although in much less degree compared with that seen for the human Gc globin gene promoter (Figures 4A and 5A).
4. Discussion The work presented in this paper extends earlier observations (Vizirianakis and Tsiftsoglou, 1996, 2005) and attempts to demonstrate the capacity of B1-559 DNA fragment located downstream from the bmajor globin gene to recruit binding of specific transcriptional factors and activate transient expression of a reporter gene (e.g. luciferase gene) in haemopoietic cells either alone, or in cooperation with the element HS-2 of globin gene LCR region. We also showed in another study that nonpoly(A)2 RNA transcripts isolated and cloned from the cytoplasm of MEL cells exposed to both differentiation inducers and
Figure 5
52
methylation inhibitors have a high structural similarity with the B1 retrotransposon element located within the B1-559 region (Tezias et al., 2011). The fact that the B1-559 DNA fragment described here exhibits potential promoter-like activity to drive luciferase gene transient expression in haemopoietic cells in cooperation with HS-2 suggests that the identified 3 consensus sequences for GATA-1, AP-1/NF-E2 and EKLF located within this fragment may contribute to transcriptional activation. Data obtained from EMSA experiments by using DNA fragments derived from this region (fragments 112 and 123 bp; Figure 1) and nuclear extracts derived from MEL cells confirmed the formation of DNA–protein complexes created through the binding of transacting proteins and DNA. The co-existence of the B1 repeat element that is the mouse equivalent of the Alu element in human DNA (Labuda et al., 1991; Kramerov and Vassetzky, 2005) with several ATG initiation codons (Vizirianakis and Tsiftsoglou, 2005) within the B1-559, suggests a potential function of gene transcriptional activity. Furthermore, the
Assessment of promoter-like activity of pGL3 recombinant constructs in undifferentiated and differentiated K-562 cells The generated pGL3 constructs were separately transfected in K-562 cells under conditions where the cells are in undifferentiated or differentiated state. To this end, K-562 cells were grown in culture in the absence (undifferentiated) or presence of inducing agent hemin (60 mM) (differentiated). Note that hemin is a known inducer of human Gc globin gene expression in K-562 cells (Tsiftsoglou et al., 2006). The assays for luciferase activity in transfected K562 cells were performed as shown in Figure 4. (A, B) Present the luciferase activity data obtained in K-562 cells transfected with the pGL3 constructs shown in Figure 3.
www.cellbiolint.org N Volume 36 (1) N pages 47–55
E The Author(s) Journal compilation E 2012 Portland Press Limited
Cell Biol. Int. (2012) 36, 47–55
capacity of B1-559 region to drive luciferase gene expression in cooperation with LCR HS-2 in a different manner in controluntreated (gene suppression) and HMBA-treated (gene activation) MEL cells implies recruitment of different proteins to a developmentally regulated transcriptional control mechanism. Such a direction is further supported by the fact that B1-559 sequences cooperate with HS-2 by activating luciferase gene expression in K-562 cells both in control-untreated and hemin-treated, i.e. the environment (homologous for MEL and heterologous for K-562 cells) contributes to the observed different cooperative capacity of B1-559 region with HS-2. In this regard, it is likely that B1-559 DNA, due to possession of consensus-binding sequences for GATA-1, NF-E2 and EKLF, has a potential transcriptional role and affects the expression pattern of bmajor gene during mouse development (Sawado et al., 2001; Cantor and Orkin, 2002; van de Lagemaat et al., 2003; Tsiftsoglou et al., 2009). On the other hand, the human Gc globin promoter used as control in these studies exhibited the same cooperative transcriptional activity for luciferase gene both in MEL (untreated and HMBA-treated) and K-562 (untreated and hemin-treated) cells, i.e. irrespective of the environment (in this case homologous for K-562 and heterologous for MEL cells). Although no firm data exist to support this notion, recent findings implicate involvement of B1 repeat family in complex regulatory elements in the control of gene expression in the mouse genome (Vidal et al., 1993). Briefly, insertion of a B1 in the 39-UTR of the rabbit b-globin gene generated a construct that conferred transcriptional regulation of this recombinant gene upon its transfection into the cells. The observation that genomic repeats act as regulatory elements, like promoters and/or enhancers that actively reshape the cellular transcriptome machinery has been an interesting development (Weiner, 2002; Roman-Gomez et al., 2005; Schumann et al., 2010). Approximately 25% of all human promoters contain retrotransposons in their sequence, whereas 7–10% of experimentally characterized transcription factor binding sites have been proposed to be derived from repetitive sequences (van de Lagemaat et al., 2003; Conley et al., 2008a, 2008b; Polavarapu et al., 2008). In this regard, the integration of a retrotransposon into CYP19 gene encoding for aromatase, the key enzyme in oestrogen biosynthesis, has created an alternative promoter that plays an important role in controlling oestrogen levels during pregnancy by allowing placental-specific CYP19 expression. Furthermore, antisense L1 and Alu sequences act as the unique promoter for HYAL-4 gene that encodes for hyaluronoglucosaminidase-4, a protein similar in structure to hyaluronidases and necessary for hyaluronan catabolism (van de Lagemaat et al., 2003). Using genome-wide screening, over 50% of human-specific endogenous retroviruses (HML-2) LTRs (long terminal repeat) possess promoter activity in vivo for host non-repetitive DNA transcription (Buzdin et al., 2006a, 2006b). In contrast, some retrotransposons may function as transcriptional silencers by down-regulating transcription of specific genes (Schumann et al., 2010). This is the case for tumour suppressor protein BRCA2 (breast-cancer susceptibility gene 2), whose gene expression is negatively regulated in a tissue-specific manner by the involvement of a 221 bp sequence of an Alu element located at the distal part of the human BRCA2 gene promoter (Sharan et al., 1999).
E The Author(s) Journal compilation E 2012 Portland Press Limited
Repetitive elements, like L1s and SINEs (sort interspersed nuclear elements), can also serve as transcriptional enhancers (Schumann et al., 2010). The enhancer of human apoA resides within the L1 element (Yang et al., 1998). In addition, an Alu sequence is a part of enhancer element located in the last intron of the human CD8 alpha gene whose expression is restricted to cells of lymphoid lineage and is developmentally regulated during thymopoesis (Hambor et al., 1993). Another recent example is the new SINE family identified in the genomes of Amniota, where these repetitive elements may act as distal transcriptional enhancers for FGF8 (fibroblast growth factor 8) and SATB2 (special AT-rich sequence-binding protein 2) genes in developing mouse forebrain (Sasaki et al., 2008). Also, evidence indicates that the origin of a number of miRNAs (microRNAs) lies within Alu and other repetitive elements in the human genome, implying the potential interaction of Alu-repetitive elements with specific miRNAs (Borchert et al., 2006). This interaction may be correlated with self-controlling cellular mechanisms towards the unrestricted multiplication of such repeat sequences within the genome (Djupedal and Ekwall, 2009; Lehnert et al., 2009). The possible interaction between the B1-559 region and bmajor globin gene transcriptional activation upon the execution of erythroid MEL cell differentiation programme seen in this study is consistent with other observations suggesting the ability of B1 retrotransposon elements to recruit transcription factors, like PAX6, at discrete binding sites in the mouse genome (Zhou et al., 2000). Similar observations were also shown for retrotransposon L1s, another family of repetitive elements. For example, p53 DNAbinding sites are detected within L1 that are involved in the regulation of their expression in the mouse genome (Harris et al., 2009). Interestingly, data derived from human neural progenitor cells suggests that L1 de novo retrotransposition may occur in the human brain and has the potential to contribute to individual somatic mosaicism (Coufal et al., 2009). In trying to understand the molecular mechanisms that regulate such an activity of L1s during brain development and contribute to neuronal function by modulating gene expression and thereby increasing brain-specific genetic mosaicism, the pivotal role of MeCP2 (methyl-CpGbinding protein 2) as a modulator of L1s neuronal transcription and retrotransposition has recently been unravelled (Muotri et al., 2010). The transcriptional activation of L1 elements is closely related to hypomethylation of their promoter region. Interestingly, this activation of L1 contributes to the progression and clinical behaviour of chronic myeloid leukaemia (Roman-Gomez et al., 2005, 2008). The latter data add new insights into the role of DNA methylation on the complexity of the molecular events that can lead to leukaemia and neurological disorders, a fact that also imposes a potential pharmacological exploitation for the treatment of such diseases. Overall, the data obtained with the transient expression of luciferase gene suggest that B1-559 DNA region cooperates with LCR [HS-2 (hypersensitive site-2)] to activate transcription. The structural homology found between B1-559 and bmajor globin gene promoter proposes that B1-559 fused region may influence bmajor globin gene expression from distance (diagrammatically depicted in Figure 6). Previous studies have proposed that upon evolution repetitive elements insert functional transcription factor-binding
Volume 36 (1) N pages 47–55 N www.cellbiolint.org
53
Potential role of repetitive elements on transcription
Figure 6
Diagrammatic outline indicating the influence of close and distant elements on the expression of bmajor globin gene trans-regulation The proposed TRC based on the data presented in this study is depicted above. Although the positive cooperative effect of LCR sequences is already known (Armstrong and Emerson, 1996; Tsiftsoglou et al., 2009), the impact of potential co-operativity of HS-2 and B1 element on bmajor globin gene transcription is suggested by the findings presented in this paper. However, the precise mechanism(s) via which the B1-559 DNA region may exert such an effect through HS-2 cooperation and bmajor globin gene promoter remains unknown.
motifs into promoter regions, thus playing a role in the regulation of biological processes, such as suppression of cell proliferation during differentiation (Polak and Domany, 2006). Furthermore, previous data support the notion that Alu repeats bearing functional NF-kB (nuclear factor kB) binding sites as integral functional component of the repeats required for the stochastic IFNb (interferon b) gene expression participates in enhanceosome assembly modulating transcription through interchromosomal interactions (Apostolou and Thanos, 2008). The existence of dynamic chromosomal associations in transcriptionally active genomic sites has also been shown in mouse. For example, such murine interchromosomal associations between alternatively expressed loci implicating the promoter region of the IFNc (interferon c) gene on chromosome 10 and the regulatory regions of the T-helper-cell 2 cytokine locus on chromosome 11 were also reported (Spilianakis et al., 2005). In addition, as shown earlier, upon transcription, the bmajor globin gene is in proximity to LCR in murine nuclei, suggesting a functional long-range chromatin regulatory interaction and potentially the base of a chromosomal loop (Carter et al., 2002). Indeed, bmajor globin gene interacts with the LCR in mouse liver cells where both are transcriptionally active, but not in the brain where both are silent (Tolhuis et al., 2002). Interestingly, a significant proportion of such chromatin interactions of b-globin locus in mice are interchromosomal. Moreover, in the liver where the gene is active, bmajor globin is more likely to interact with other chromosomes compared with the brain, where the gene is silent (Pink et al., 2010). The capacity of the characterized B1-559 region to act as a potential promoter in erythroid haemopoietic cells has been shown in our work. Whether this region plays any role, and how transcriptional regulation of b-globin gene locus occur remain to be elucidated. Alternatively, however, we might postulate that B1-559 unmasked by chromatin remodelling can act developmentally upon b-globin gene transcriptional regulation by participating in enhanceosome assembly, affecting LCR-promoter interactions. The impact of such a potential cooperation between LCR (HS-2) and B1-559 bp region is very interesting but remains to be proved. In addition, the possibility that LCR (HS-2) distant elements and DNA domains like B1-559 exert a regulatory effect on bmajor globin gene expression via a ‘trans-regulated circuit’ in haemopoietic cells upon development and under the influence of DNA methylation state is challenging.
54
www.cellbiolint.org N Volume 36 (1) N pages 47–55
Author contribution Ioannis Vizirianakis and Asterios Tsiftsoglou conceived and designed the experiments. Sotirios Tezias and Elsa Amanatiadou performed the experiments. All authors analysed the data and wrote the paper.
Acknowledgement We thank Mrs Anna Sidiropoulou for her help in EMSAs.
Funding This work was in part supported by interdepartmental funds of the Aristotle University of Thessaloniki and in part by PENED and EPEAK grants awarded by the Greek government.
References Apostolou E, Thanos D. Virus infection induces NF-kappaB-dependent interchromosomal associations mediating monoallelic IFN-beta gene expression. Cell 2008;134:85–96. Armstrong JA, Emerson BM. NF-E2 disrupts chromatin structure at human beta-globin locus control region hypersensitive site 2 in vitro. Mol Cell Biol 1996;16:5634–44. Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA et al. Current protocols in molecular biology. New York: John Wiley & Sons; 1993. Borchert GM, Lanier W, Davidson BL. RNA polymerase III transcribes human microRNAs. Nat Rev Strut Mol Biol 2006;13:1097–101. Buzdin A, Kovalskaya-Alexandrova E, Gogvadze E, Sverdlov E. At least 50% of human-specific HERV-K (HML-2) long terminal repeats serve in vivo as active promoters for host nonrepetitive DNA transcription. J Virol 2006a;80:10752–62. Buzdin A, Kovalskaya-Alexandrova E, Gogvadze E, Sverdlov E. GREM, a technique for genome-wide isolation and quantitative analysis of promoter active repeats. Nucleic Acids Res 2006b;34:e67. Cantor AB, Orkin SH. Transcriptional regulation of erythropoiesis: an affair involving multiple partners. Oncogene 2002;21:3368–76. Carter D, Chakalova L, Osborne C, Dai Y, Fraser P. Long-range chromatin regulatory interactions in vivo. Nat Genet 2002;32:623–6. Chen L-L, DeCerbo JN, Carmichael GG. Alu element-mediated gene silencing. EMBO J 2008;27:1694–705. Conley AB, Miller WJ, Jordan IK. Human cis natural antisense transcripts initiated by transposable elements. Trends Genet 2008a;24:53–6. Conley AB, Piriyapongsa J, Jordan IK. Retroviral promoters in the human genome. Bioinformatics 2008b;24:1563–7.
E The Author(s) Journal compilation E 2012 Portland Press Limited
Cell Biol. Int. (2012) 36, 47–55
Coufal NG, Garcia-Perez JL, Peng GE, Yeo GW, Mu Y, Lovci MT et al. L1 retrotransposition in human neural progenitor cells. Nature 2009;460:1127–31. Djupedal I, Ekwall K. Epigenetics: heterochromatin meets RNAi. Cell Res 2009;19:282–95. Duncan C, Biro PA, Choudary PV, Elder JT, Wang RRC, Forget BG et al. RNA polymerase III transcriptional units are interspersed among human non-a-globin genes. Proc Natl Acad Sci USA 1979;76:5095–9. Hambor JE, Mennone J, Coon ME, Hanke JH, Kavathas P. Identification and characterization of an Alu-containing, T-cell-specific enhancer located in the last intron of the human CD8 alpha gene. Mol Cell Biol 1993;13:7056–70. Harris CR, Dewan A, Zupnick A, Normart R, Gabriel A, Prives C et al. p53 responsive elements in human retrotransposons. Oncogene 2009;28:3857–65. Kramerov DA, Vassetzky NS. Short retroposons in eukaryotic genomes. Int Rev Cytol 2005;247:165–221. Labuda D, Sinnett D, Richer C, Deragon JM, Striker G. Evolution of mouse B1 repeats: 7SL RNA folding pattern conserved. J Mol Evol 1991;32:405–14. Lai CB, Zhang Y, Rogers SL, Mager DL. Creation of the two isoforms of rodent NKG2D was driven by a B1 retrotransposon insertion. Nucleic Acids Res 2009;37:3032–43. Lehnert S, Van Loo P, Thilakarathne PJ, Marynen P, Verbeke G, Schuit FC. Evidence for co-evolution between human microRNAs and Alurepeats. PLoS ONE 2009;4:e4456. Muotri AR, Marchetto MC, Coufal NG, Oefner R, Yeo G, Nakashima K et al. L1 retrotransposition in neurons is modulated by MeCP2. Nature 2010;468:443–6. Ostertang EM, Kazazian HH. LINEs in mind. Nature 2005;435:890–1. Pink RC, Eskiw CH, Caley DP, Carter DR. Analysis of b-globin chromatin micro-environment using a novel 3C variant, 4Cv. PLoS ONE 2010;5:e13045. Polak P, Domany E. Alu elements contain many binding sites for transcription factors and may play a role in regulation of developmental processes. BMC Genomics 2006;7:133. Polavarapu N, Marino-Ramirez L, Landsman D, McDonald JF, Jordan IK. Evolutionary rates and patterns for human transcription factor binding sites derived from repetitive DNA. BMC Genomics 2008;9:226. Reddy PM., Stamatoyannopoulos G, Papayannopoulou T, Shen CK. Genomic footprinting and sequencing of human beta-globin locus: tissue specificity and cell line artifact. J Biol Chem 1994;269:8287–95. Roman-Gomez J, Jimenez-Velasco A, Agirre X, Castillejo JA, Navarro G, San Jose-Eneriz E et al. Repetitive DNA hypomethylation in the advanced phase of chronic myeloid leukemia. Leuk Res 2008;32:487–90. Roman-Gomez J, Jimenez-Velasco A, Agirre X, Cervantes F, Sanchez J, Garate L et al. Promoter hypomethylation of the LINE-1 retrotransposable elements activates sense/antisense transcription and marks the progression of chronic myeloid leukemia. Oncogene 2005;24:7213–23. Romanish MT, Lock WM, van de Lagemaat LN, Dunn CA, Mager DL. Repeated recruitment of LTR retrotransposons as promoters by the anti-apoptotic locus NAIP during mammalian evolution. PLoS Genet 2007;3:e10. Romanish MT, Nakamura H, Lai CB, Wang Y, Mager DL. A novel protein isoform of the multicopy human NAIP gene derives from intragenic Alu SINE promoters. PLoS One 2009;4:e5761.
Sasaki T, Nishihara H, Hirakawa M, Fujimura K, Tanaka M, Kokubo N et al. Possible involvement of SINEs in mammalian-specific brain formation. Proc Natl Acad Sci USA 2008;105:4220–5. Sawado T, Igarashi K, Groudine M. Activation of beta-major globin gene transcription is associated with recruitment of NF-E2 to the betaglobin LCR and gene promoter. Proc Natl Acad Sci USA 2001;98:10226–31. Schumann GG, Gogvadze EV, Osanai-Futahashi M, Kuroki A, Mu¨nk C, Fujiwara H et al. Unique functions of repetitive transcriptomes. Int Rev Cell Mol Biol 2010;285:115–88. Sharan C, Hamilton NM, Parl AK, Singh PK, Chaudhuri G. Identification and characterization of a transcriptional silencer upstream of the human BRCA2 gene. Biochem Biophys Res Commun 1999;265:285–90. Spilianakis CG, Lalioti MD, Town T, Lee GR, Flavell RA. Interchromosomal associations between alternatively expressed loci. Nature 2005;435:637–45. Tezias SS, Tsiftsoglou AS, Amanatiadou EP, Vizirianakis IS. Cloning and characterization of polyA- RNA transcripts encoded by activated B1-like retrotransposons in mouse erythroleukemia MEL cells exposed to methylation inhibitors. BMB Reports 2011; in press. Tolhuis B, Palstra R, Splinter E, Grosveld F, de Laat W. Looping and interaction between hypersensitive sites in the active beta-globin locus. Mol Cell 2002;10:1453–65. Tsiftsoglou AS, Pappas IS, Vizirianakis IS. Mechanisms involved in the induced differentiation of leukemia cells. Pharmacol Ther 2003;100:257–90. Tsiftsoglou AS, Tsamadou AI, Papadopoulou LC. Heme as key regulator of major mammalian cellular functions: molecular, cellular, and pharmacological aspects. Pharmacol Ther 2006;111:327–45. Tsiftsoglou AS, Vizirianakis IS, Strouboulis J. Erythropoiesis: model systems, molecular regulators, and developmental programs. IUBMB Life 2009;61:800–30. Tsiftsoglou AS, Wong W, Robinson SH, Hensold J. Hemin increase production of b-like globin RNA transcripts in human erythroleukemia K-562 cells. Dev Genet 1989;10:311–7. van de Lagemaat LN, Landry JR, Mager DL, Medstrand P. Transposable elements in mammals promote regulatory variation and diversification of genes with specialized functions. Trends Genet 2003;19:530–6. Vidal F, Mougneau E, Glaichenhaus N, Vaigot P, Darmon M, Cuzin F. Coordinated posttranscriptional control of gene expression by modular elements including Alu-like repetitive sequences. Proc Natl Acad Sci USA 1993;90:208–12. Vizirianakis IS, Tsiftsoglou AS. Induction of murine erythroleukemia cell differentiation is associated with methylation and differential stability of polyA+ RNA transcripts. Biochim Biophys Acta 1996;1312:8–20. Vizirianakis IS, Tsiftsoglou AS. Blockade of murine erythroleukemia cell differentiation by hypomethylating agents causes accumulation of discrete small poly(A)2 RNAs hybridized to 39-end flanking sequences of beta(major) globin gene. Biochim Biophys Acta 2005;1743:101–14. Weiner AM. SINEs and LINEs: the art of biting the hand that feeds you. Curr Opin Cell Biol 2002;14:343–50. Yang Z, Boffelli D, Boonmark N, Schwartz K, Lawn R. Apolipoprotein(a) gene enhancer resides within a LINE element. J Biol Chem 1998;273:891–7. Zhou Y, Zheng JB, Gu X, Li W, Saunders GF. A novel Pax-6 binding site in rodent B1 repetitive elements: coevolution between developmental regulation and repeated elements? Gene 2000;245:319–28.
Received 14 April 2011/ 20 July 2011; accepted 4 October 2011 Published as Immediate Publication 4 October 2011, doi 10.1042/CBI20110236
E The Author(s) Journal compilation E 2012 Portland Press Limited
Volume 36 (1) N pages 47–55 N www.cellbiolint.org
55
