Repression of chimeric transcripts emanating from endogenous retrotransposons by a sequence-specific transcription factor
© Mak et al.; licensee BioMed Central Ltd. 2014
Received: 17 September 2013
Accepted: 30 April 2014
Published: 30 April 2014
The Erratum to this article has been published in Genome Biology 2016 17:119
Retroviral elements are pervasively transcribed and dynamically regulated during development. While multiple histone- and DNA-modifying enzymes have broadly been associated with their global silencing, little is known about how the many diverse retroviral families are each selectively recognized.
Here we show that the zinc finger protein Krüppel-like Factor 3 (KLF3) specifically silences transcription from the ORR1A0 long terminal repeat in murine fetal and adult erythroid cells. In the absence of KLF3, we detect widespread transcription from ORR1A0 elements driven by the master erythroid regulator KLF1. In several instances these aberrant transcripts are spliced to downstream genic exons. One such chimeric transcript produces a novel, dominant negative isoform of PU.1 that can induce erythroid differentiation.
We propose that KLF3 ensures the integrity of the murine erythroid transcriptome through the selective repression of a particular retroelement and is likely one of multiple sequence-specific factors that cooperate to achieve global silencing.
Transposable elements (TEs) are mobile segments of DNA that have integrated and spread in eukaryotic genomes. They constitute a substantial portion of the mouse and human genomes (approximately 39% and 46%, respectively ), and have been a major driving force in evolution . In mammals, the vast majority of TEs are retrotransposons: genetic elements that have integrated into host DNA following reverse transcription of an RNA template. Broadly, retrotransposons fall into two categories: those that contain long terminal repeats (LTRs) and are termed endogenous retroviruses (ERVs); and those that lack LTRs, namely, long and short interspersed elements (LINES and SINES).
The expansion of TEs has played an important role in shaping eukaryotic genomes, in part by enabling genomic shuffling by non-allelic homologous recombination but also by their capacity to remodel gene regulatory networks [3–7]. Retroviral LTRs, for instance, harbor numerous, functional regulatory elements required for the initiation and control of transcription and can thus profoundly alter the expression of proximal genes [8, 9]. Furthermore, because the many distinct classes of retrovirus differ in the regulatory sequences contained within their LTRs, they have proven highly versatile in rewiring diverse transcriptional programs. Indeed, throughout mammalian evolution, the spread and proliferation of retroelements have redistributed binding sites for a number of transcriptional regulators including the pluripotency factors OCT4 and NANOG , the insulator protein CTCF [4–6], the neural repressor NRSF/REST , the tumour suppressor p53 , and others . Analogously, expansion of the MER20 and RLTR13D5 transposable elements, which promote endometrial and trophoblast expression, have been postulated to have enabled the placental transcription of genes critical to the evolutionary development of pregnancy [13, 14].
Not only do retrotransposons provide regulatory modules that influence nearby genes, they can also directly provide promoters that dictate transcriptional initiation. A notable example of this is the murine Agouti viable yellow (A vy ) allele, in which an upstream intra-cisternal A particle (IAP) retrotransposon functions as a constitutively active promoter that drives ectopic expression of Agouti, resulting in yellow fur, obesity, and increased susceptibility to tumorigenesis [15, 16]. The prevalence of this phenomenon, whereby retrotransposons serve as alternative promoters, has recently been revealed following the advent of high-throughput RNA sequencing and shown to occur primarily in embryonic cells but also to some extent in adult tissues . During early embryonic development in particular, up to 20% of the transcriptome has been shown to initiate from within retrotransposons [17, 18]. These retroelements frequently function as alternative promoters and show a propensity for tissue-specific activity, more so in fact than non-retrotransposon promoters . In many instances, these retrotransposons have been co-opted by the host by exonization and they are transcribed and spliced to downstream genic exons [17, 19, 20]. The resulting chimeric transcripts thus potentially encode isoform variants with spatio- or temporally-restricted expression profiles . Indeed, a recent study of the Drosophila melanogaster transcriptome has revealed that several hundred LTR retrotransposons serve as promoters of annotated genes throughout development, exhibiting specific expression profiles depending on the different regulatory modules they carry .
However, while there are many reported instances of TEs being co-opted by the host for various biological functions, genomic integration of TEs can also be deleterious [23–25]. For instance, Hodgkin’s lymphoma has been shown to arise from aberrant transcription of the colony-stimulating factor 1 receptor (CSF1R) gene driven by an internal LTR element known as THE1B.
Accordingly, higher eukaryotes have developed numerous defence mechanisms to silence TEs, typically involving DNA methylation and/or histone modification [27–29]. This silencing largely occurs early in embryonic development and is dependent on epigenetic modifiers including: DNA methyltransferases (DNMTs) [27, 30]; histone modifying enzymes such as the demethylase LSD1/KDM1A, the deacetylase HDAC1, and the methyltransferases SETDB1 and G9A [31–35]; and Polycomb Group proteins . Ablation of these factors in embryonic stem cells results in widespread de-repression of retrotransposon-derived transcripts.
However, while silencing of retroelements is broadly carried out by these epigenetic modifiers, little is known about the underlying mechanisms by which the diverse classes of retroelements are each specifically recognized . Indeed the lack of sequence similarity between unrelated retroviral families suggests the existence of multiple recognition factors that participate in the silencing of retroelements. An accumulating body of evidence has pointed towards the possible role of DNA binding, tandem zinc finger proteins in providing this specificity. Thomas and Schneider have proposed a model of co-evolution between retroelements and C2H2 zinc finger proteins based on striking correlations of their expansion throughout vertebrate genomes . This model followed from the discovery that the Krüppel-associated box (KRAB)-zinc finger protein ZFP809 binds and represses a large number of retroelements in mouse embryonic stem cells . ZFP809 achieves this through the recruitment of the corepressor TRIM28 (also known as KRAB-associated protein 1, KAP1). TRIM28 in turn silences ERVs through SETDB1 mediated trimethylation of H3K9 [31, 39, 40].
The Krüppel-like factors (KLFs) are a family of DNA binding, zinc finger transcription factors . They lack a KRAB domain and are characterized by a set of three tandem C2H2 zinc fingers at their C-termini that confer specificity towards CACCC-like and GC-rich sequences in regulatory elements . While the DNA binding domain is highly conserved within the family, the N-terminal regulatory domains vary considerably such that the different KLFs recruit an assortment of co-regulators to activate or repress genes .
The founding member of the family, KLF1, is an erythroid-specific transcriptional activator that drives the expression of genes required for red blood cell maturation . One such gene is that encoding the related family member KLF3 . KLF3 and KLF1 recognize similar sequences of DNA that adhere to the consensus 5′-NCN CNC CCN-3′ [42, 46]. However, unlike KLF1, KLF3 is a transcriptional repressor that recruits the co-repressor C-terminal binding protein (CTBP) . CTBP forms part of a large repressor complex that includes the histone deacetylases HDAC1 and HDAC2, the histone methyltransferases EHMT1 and G9A/EHMT2, and the lysine-specific demethylase LSD1/KDM1A . KLF1 and KLF3 exhibit opposing activities at a number of genes in erythroid cells and serve to fine-tune their expression during erythropoiesis [49, 50]. Accordingly, loss of either factor disrupts this balance. Klf1 null mice die of severe anemia in utero while mice lacking KLF3, though viable, exhibit erythroid defects in both fetal and adult tissues [49, 51].
Here, we have further explored the interplay between KLF1 and KLF3 in regulating the erythroid transcriptome. We find that KLF1 activates, while KLF3 represses, transcription from a specific family of LTR elements known as ORR1A0. Ablation of KLF3 results in widespread, de-repressed transcription from these LTRs in erythroid cells. Because the ORR1A0 element contains an intact splice donor site, these transcripts are spliced to exons of the genes in which they reside. We show that for the spleen focus forming virus proviral integration 1 (Sfpi1) gene, an ORR1A0-driven transcript is translated into a truncated variant of PU.1 which exhibits dominant negative activity and can functionally promote erythroid differentiation. These results suggest that KLF3 ensures normal murine erythropoiesis by preventing aberrant, chimeric transcription driven from ORR1A0 LTRs by KLF1.
Increased expression of downstream Pu.1 exons in erythroid cells in the absence of KLF3
We recently identified a number of KLF3 target genes via microarray analysis of Klf3 −/− TER119+ (erythroid) fetal liver cells at embryonic day E14.5 . These genes were predominantly de-repressed in Klf3 null tissue, consistent with KLF3 being a repressor of transcription. One of the most highly de-repressed genes was that encoding the key hematopoietic regulator PU.1/SFPI1, hereafter referred to as PU.1.
An ORR1A0 LTR element serves as an alternative promoter in the Pu.1 locus in the absence of KLF3
Searches using the RepeatMasker program showed that exon 2b lies within a 343 bp long terminal repeat (LTR) element, named ORR1A0, belonging to the MaLR (mammalian apparent LTR retrotransposon) family (Figure 2B) . The ORR1A0 element in the murine Pu.1 locus contains several hallmarks of a eukaryotic core promoter including a TATA box at −30, an initiator sequence (5′-TCAGTY-3′) at the TSS and a downstream promoter element around +30 . In addition, it contains several motifs fitting the KLF DNA-binding consensus 5′-NCN CNC CCN-3′ (Figure 2B).
In order to verify that this novel Pu.2 transcript is de-repressed in erythroid cells lacking KLF3, we performed real-time RT-PCR on Klf3 +/+, Klf3 +/− and Klf3 −/− Ter119+ E14.5 fetal liver RNA using a forward primer specific for exon 2b and a reverse primer targeting exon 3 of Pu.1. Indeed, significant up-regulation of the Pu.2 transcript (>140-fold) was observed in Klf3 −/− compared to Klf3 +/+ and Klf3 +/− samples (Figure 2D). While this transcript was not amplified from wild-type tissue by 5′ RACE (Figure 2A), we detected low amounts of it in wild-type and Klf3 +/− tissue by RT-PCR. In adult Klf3 −/− mice, we observed marked upregulation of this chimeric transcript in erythroid organs (spleen and bone marrow) (Additional file 1: Figure S1A). In contrast, canonical Pu.1 mRNA was unaltered in these and other tissues examined (Additional file 1: Figure S1B).
KLF3 can repress transcription by recruiting CTBP, a co-repressor that silences genes through a number of different histone-modifying enzymes. We therefore analyzed a series of histone marks around the Pu.2 ORR1A0 promoter in Klf3 −/− compared to Klf3 +/+ E14.5 fetal liver cells (Figure 2E). In particular, we observed a marked increase specifically of histone 3 lysine 4 tri-methylation (H3K4me3) in Klf3 −/− cells at the ORR1A0 promoter (approximately 12-fold). We found that this mark was also increased at the promoters of previously validated KLF3 target genes such as Klf8 and Fam132a/adipolin[49, 54] (Figure 2E). H3K4me3 is a mark typically found at actively transcribed promoters . Moreover, loss of this mark, rather than the acquisition of repressive modifications, has been reported during the developmental silencing of retroelements . In addition, the Pu.2 promoter displayed only a moderate level of H3K9me3 in wild-type cells (Additional file 2: Figure S2). This was not appreciably altered in Klf3 −/− cells, suggesting that H3K9 tri-methylation is not the primary mechanism through which KLF3 silences transcription at this locus. Together, these results indicate that the ORR1A0 element is a functional, alternative promoter for the Pu.1 gene in erythroid cells and is highly de-repressed in the absence of KLF3.
KLF1 and KLF3 can bind to the CACCC-like boxes in the ORR1A0 LTR and activate and repress transcription, respectively
To investigate whether KLF1 indeed drives transcription of these chimeric Pu.2 transcripts from the ORR1A0 element in vivo, we employed a KLF1-inducible erythroid cell line known as B1.6 . These cells were derived from Klf1 −/− fetal liver and have been rescued with a transgene encoding a tamoxifen-inducible KLF1-ER (estrogen receptor) fusion protein. Upon addition of tamoxifen, KLF1-ER is activated and drives expression of KLF1 target genes to induce hemoglobinization and erythroid differentiation. KLF3 protein has not been detected in these cells, although Klf3 mRNA is induced after KLF1-ER activation .
Using real-time RT-PCR we observed a low level of Pu.2 mRNA in untreated B1.6 cells; however, tamoxifen induction resulted in a dramatic increase of these transcripts (Figure 4C). Moreover, despite decreased expression of canonical Pu.1 transcripts upon KLF1-ER induction (represented by exon 2/exon 3, Figure 4D), the total level of Pu.1 plus Pu.2 transcripts increased (represented by the exon 3/exon 4 junction, Figure 4E), albeit not significantly. This suggests that the chimeric transcripts contribute substantially to the total Pu.1 plus Pu.2 mRNA pool in induced B1.6 cells. In addition, the induction of Pu.2 expression was rapid following tamoxifen addition (within 2 hours) and occurred in the presence of the translation inhibitor cycloheximide (Additional file 3: Figure S3A, B). Taken together, these data suggest that KLF1 directly activates Pu.2 transcription from the endogenous ORR1A0 promoter in erythroid cells in the absence of KLF3.
Widespread de-repression of chimeric transcripts from ORR1A0 elements in the absence of KLF3
A RepeatMasker survey revealed that there are approximately 2,130 ORR1A0 integrants in the mouse genome. The consensus sequence of ORR1A0 found in Repbase contains all of the core promoter sequences shown in Figure 2B as well as the four 5′-NCN CNC CCN-3′ motifs. In addition, there is little divergence between individual ORR1A0 elements with elements generally sharing greater than 97% sequence identity to the consensus . We therefore hypothesized that KLF3 might play a broader role in silencing aberrant transcription from ORR1A0 LTRs.
To investigate this, we performed RNA-Seq on triplicate samples of Klf3 +/+ and Klf3 −/− E14.5 TER119+ fetal liver cells. In total, 1,025 genes were found to be significantly deregulated (FDR <0.05) in the absence of KLF3 (Additional file 4: Table S1). The majority of these (76.7%) were upregulated in Klf3 −/− cells, concordant with the view from previous studies that KLF3 is predominantly a transcriptional repressor . Importantly, previously validated KLF3 targets also displayed significant upregulation by RNA-Seq including Klf8 (108-fold), Lgals3 (33-fold), Fam132a/adipolin (7.5-fold), Hba-x (2.3-fold), and Hbb-y (1.8-fold) [49, 50, 54, 58].
We next assessed whether the ORR1A0 LTR, and related ORR1A0-int, elements were enriched among the list of KLF3 target genes. We found that of the 786 significantly upregulated genes, 166 of these (21.1%) contained one or more ORR1A0 LTRs. In contrast, of the 239 downregulated genes, there was only one instance of an overlap with an ORR1A0 element (0.004%). Similarly, ORR1A0-int elements, which are typically flanked by ORR1A0 LTRs, were found in 96 upregulated genes (12.2%) and only in a single case of a downregulated gene. These results illustrate a clear enrichment of the ORR1A0 and ORR1A0-int retroelements specifically within genes that are normally repressed by KLF3 in erythroid cells.
Because of sequence conservation between ORR1A0 elements and difficulties associated with unambiguously assigning them to specific genomic loci, we instead looked for evidence of splicing between ORR1A0 transcripts and downstream genic exons. To do this, we confined our analysis to annotated genes which displayed differential isoform expression in Klf3 −/− cells (greater than 10-fold upregulated compared to Klf3 +/+). We identified 70 such genes (Additional file 5: Table S2). Of these, 34.3% contained transcribed ORR1A0 elements, and almost half of these (41.7%) were spliced to genic exons and a further 16.7% showed splicing to un-annotated exons.
Several of these chimeric transcripts have previously been reported as ESTs that have typically been detected in embryonic cells and tissues from a range of developmental stages including 4-cell (Brca2; GenBank:CN716605) and 8-cell stage embryos (Chr10: chr10:61042355–61063209 shown in Figure 6D; GenBank:CJ067427), E13 liver tissue (chr9:9049867–9081010; GenBank:CJ043932), and E17 amnion (Znrf2; GenBank: BY073363 and CJ093793). This provides further evidence that ORR1A0 LTRs are capable of functioning as bona fide promoters in vivo.
The chimeric Pu.2 transcript driven by the ORR1A0 promoter is translated in vivo
Thus far, the results presented here suggest that KLF3 silences aberrant transcription from ORR1A0 LTRs in erythroid cells. These chimeric transcripts potentially encode protein variants that might functionally impact normal murine erythropoiesis. Indeed, Klf3 null mice exhibit a number of erythroid defects including increased immature red blood cells (reticulocytes) and nuclear inclusions (Howell-Jolly bodies) in peripheral blood . We thus sought to determine whether these chimeric transcripts are in fact translated in vivo using the Pu.1 gene as an example, given the role of this transcription factor as a master regulator of hematopoietic differentiation .
To determine whether PU.2 protein is expressed in vivo, we analyzed nuclear extracts from E14.5 Klf3 −/− fetal livers. Extracts from these cells formed bands that co-migrated with both PU.1 and PU.2 (Figure 7A). The upper band is supershifted by both antisera while the lower band is only supershifted by the C-terminal antisera, confirming their identities as PU.1 and PU.2, respectively (Figure 7B, C). PU.2 protein was also detected in nuclear extracts from Klf3 +/+ fetal liver cells (Figure 7B, C) albeit at a lower level than in Klf3 −/− samples. Lastly, we also observed marked induction of PU.2 protein upon tamoxifen activation of KLF1-ER in B1.6 cells (Figure 7D) consistent with the upregulation of Pu.2 transcripts (Figure 4C). Taken together, these results indicate that PU.2 is indeed translated in erythroid cells in vivo.
PU.2 can act as a dominant negative protein in erythroid cells
The PU.2 protein lacks the N-terminal activation domain of PU.1, a region that interacts with the general transcription factor TFIID . We therefore postulated that PU.2 might not function as a transcriptional activator and might antagonize the activity of PU.1 at its target genes. To investigate this, we first conducted reporter assays using the promoter of a previously characterized PU.1 target gene, CLEC5A. We found that while PU.1 robustly activated expression, PU.2 repressed this promoter in a dose-dependent manner (Additional file 8: Figure S6A, B).
It has recently been shown that TEs frequently act as promoters of genic transcription and are dynamically transcribed during ontogeny [17, 18, 20, 22]. Typically, the majority of retroelement silencing occurs early in gestation  and numerous studies have demonstrated the role of histone- and DNA-modifying enzymes in this process [27, 30–34, 36]. KRAB domain zinc finger proteins, which interact with the co-repressor TRIM28, have been proposed to play a role in the specific recognition and repression of distinct retroelement families [31, 35, 37, 39, 40]. This has indeed shown to be the case for ZFP809 in embryonic stem cells . Other DNA-binding zinc finger proteins, including REX1/ZFP42, have also been implicated, but their mechanism of action remains unclear and they appear to affect multiple, unrelated retroviral families .
Here we show that the zinc finger protein KLF3, which lacks a KRAB domain, is required for the silencing of transcription from ORR1A0 elements of the MaLR family. In the absence of KLF3, there is a pronounced increase in chimeric transcripts generated from these LTRs. The phenomenon of deregulated chimeric transcription has previously been observed upon ablation of epigenetic modifiers such as SETDB1 and LSD1 in embryonic stem cells [27, 34]. For example, in cells lacking SETDB1, 15% of de-repressed genes arise due to failed silencing of promoter-proximal ERVs and half of these genes exhibit chimeric transcripts . However, in both of these cases the effects described were more global than we observe for KLF3 and covered diverse retroelement families.
Silencing of the ORR1A0 LTR by KLF3 appears to occur largely independently of the TRIM28/SETDB1 pathway and DNA methylation. Analysis of over 23,000 TRIM28 binding sites in ES cells compiled from two studies [40, 64] revealed that only eight lie within 100 bp of an ORR1A0 element. This may partially reflect the difficulties of detecting ChIP peaks that extend beyond the boundaries of repeat elements. However, a separate examination of regions of TRIM28-dependent H3K9 tri-methylation, which typically spread beyond repeat borders, revealed that only 62 of the 2,140 ORR1A0 LTRs (that is, 2.9%) lie within 100 bp . Similarly, analysis of the DNA methylation status of hematopoietic stem cells and erythroblasts revealed that only 1.0% (22) and 0.3% (7) of ORR1A0s, respectively, lie proximal to regions of DNA-methylation as determined by MBD-Seq .
From a number of Klf3 −/− tissues examined, the upregulation of ORR1A0 transcription appeared to be restricted to erythroid cells. This is of particular interest given that most retroelement silencing has been demonstrated in embryonic stem and germ cells [27, 30, 31, 33]. This suggests that KLF3 plays an active role in somatic repression of retroviral transcription, both in fetal and adult erythroid tissues. Consistent with its erythroid restricted profile, we found that the master erythroid regulator KLF1 drives expression of ORR1A0-originating transcripts in the absence of KLF3.
The specific recognition of the ORR1A0 element by KLF3 and KLF1 appears to involve two 5′-CACNCCC-3′ boxes upstream of the TSS (Figures 2B and 3). The similar DNA-binding specificities of KLF1 and KLF3 have previously been noted and indeed, approximately 50% of KLF3 target genes in erythroid cells are also regulated by KLF1 . Interestingly, the ORR1A0-related LTR ORR1A1 lacks one of the 5′-CACNCCC-3′ boxes and additionally lacks the TATA box. Despite being 97% homologous with ORR1A0 and occurring at twice the frequency in the mouse genome, we did not observe enrichment of ORR1A1 in KLF3 repressed transcripts, alluding to the functional importance of these two promoter motifs and the extraordinary specificity of KLF3 for the ORR1A0 LTR.
Moreover, although the ORR1 retroelement family is abundantly represented throughout diverse rodent species, the ORR1A0 LTR is specifically only found in the mouse. It is interesting to note that the DNA-binding domain of KLF3 shows complete sequence conservation between Mus musculus, Rattus norvegicus, and Homo sapiens. This suggests that in the mouse, the DNA-binding specificity of KLF3 has not altered in response to the emergence of the ORR1A0 retroelement. Rather, it appears that KLF3-mediated repression of ORR1A0 occurred intrinsically from the initial appearance of the retroelement, and in this context, the spread of the ORR1A0 LTR may have been tolerated without deleterious impacts.
While KLF3 appears to efficiently silence ORR1A0 transcription in spleen and bone marrow cells, it is possible that at particular stages of development or cellular maturation, KLF1 predominates and these chimeric transcripts are expressed at higher levels, as observed in the B1.6 erythroblast line. These chimeric transcripts potentially encode functional protein isoforms and indeed, in the case of the Pu.1 locus, we have detected a truncated isoform expressed in fetal liver. PU.2 counters the normal activity of PU.1 and instead promotes erythroid differentiation when ectopically expressed in myeloid leukemic K562 cells (Figure 8). This is consistent with the role of its transcriptional activator, KLF1, in driving terminal erythroid differentiation . From this study, we cannot discount the possibility that the chimeric transcripts driven by KLF1 may be biologically functional in some circumstances. Indeed, host exaptation of TEs by exonization or through the remodelling of expression programs is a phenomenon of which multiple instances have been described [5, 13, 14, 19]. Incidentally, although the consensus sequence of the ORR1A0 exon contains a number of short open reading frames (see Figure 2B), each ATG is ultimately succeeded by an in-frame stop codon. Thus ORR1A0 promoters are predicted to drive expression of either full-length endogenous proteins (where a canonical translation start site lies in a downstream, spliced exon), or N-terminal truncated protein isoforms, in cases where internal ATGs are suitable start sites of translation, as for PU.2.
The importance of appropriate retrotransposon control is apparent from numerous examples in which dysregulation results in disease [25, 26]. It is unclear as to the extent to which the dysregulation of ORR1A0 transcription affects the physiology of the Klf3 null mice. However, these animals do display an erythroid phenotype, with impaired maturation, reticulocytosis, increased Howell-Jolly bodies and decreased lifespan of erythrocytes . These defects have not been attributed to any single gene and it is more likely that they arise as a complicated result of multiple defects including the widespread expression of aberrant, chimeric transcripts that we have presented here.
In summary, these findings emphasize the non-redundant role that KLF3 plays in preventing widespread, promiscuous transcription specifically from the ORR1A0 LTR. We suggest that KLF3 is likely one of a patchwork of zinc finger proteins including ZFP809 that together collaborate to silence the diverse collection of TEs that constitute such a large fraction of vertebrate genomes.
Materials and methods
Klf3 −/− mice
Generation and maintenance of the Klf3 −/− mouse line has been described previously . Ethical approval for animal use was obtained from the appropriate Animal Care and Ethics Committees (University of Sydney, approval number L02/7-2009/3/5079; and University of New South Wales, approval number 09/128A). Genomic DNA was extracted from tail snips using DirectPCR Lysis Reagent (Viagen Biotech Inc, Los Angeles, CA, USA) as per the manufacturer’s instructions and genotyping was performed as described previously .
Sorting of TER119+ fetal liver cells
TER119+ cells were sorted from whole fetal livers (E13.5 or E14.5) using anti-TER119 Microbeads with MS or LS columns (Miltenyi Biotec Australia, North Ryde, NSW, Australia) as per the manufacturer’s protocol. After eluting, cells were centrifuged at 300 g for 10 min at 4°C and RNA was extracted from the cell pellet as described below.
RNA extraction and quantitative real-time RT-PCR
RNA was extracted, purified and subjected to DNase treatment as described previously [45, 49]. Extracted RNA was then used as a template for cDNA synthesis using the SuperScript® VILO™ cDNA Synthesis Kit (Invitrogen, Carlsbad, CA, USA) as instructed by the supplier. Quantitative real-time RT-PCR reactions were set up as described previously [45, 49] but using FastStart Universal SYBR Green Master (ROX) (Roche Diagnostics Australia Pty Ltd, Castle Hill, NSW, Australia) or Power SYBR® Green PCR Master Mix (Life Technologies, Gladesville, NSW, Australia). Reactions were run using the Applied Biosystems 7500 Fast Real-Time PCR System (Life Technologies) and data were analyzed using 7500 Software v2.0.4 (Life Technologies). Relative quantification was calculated using 18S rRNA levels and standard curves derived from serial dilutions of amplicon as described previously .
Real-time RT-PCR primers
Gene-specific primer pairs were designed using the Primer-BLAST tool  or PrimerExpress v3.0 (Applied Biosystems, Foster City, CA, USA) software to cross exon boundaries where possible. Primers were included in reactions at a concentration of 400 nM. The sequences of forward and reverse primers for each gene are: 18S, 5′-CACGGCCGGTACAGTGAAAC-3′ and 5′-AGAGGAGCGAGCGACCAA-3′; Pu.1 exons 2/3, 5′-CCTTCGTGGGCAGCGATGGA-3′ and 5′-GAGCTGCTGTAGCTGCGGGG-3′; Pu.1 exons 3/4, 5′-GAGCTGGAACAGATGCACGTCCT-3′ and 5′-GTGGGCTGGGGACAAGGTTTGAT-3′; Pu.1 exons 4/5, 5′-TGGAGAAGCTGATGGCTTGGAGC-3′ and 5′-CAGCAGGAACTGGTACAGGCGAA-3′; Pu.2 exons 2b/3, 5′-CCTGGATGCTGTCATGCTCCCAA-3′ and 5′-GTTGTTGTGGACATGGTGTGCGG-3′; Znrf2, 5′-CCACTAGCAGCCTTTGGATGAAGACA-3′ and 5′-GGGTACAAATTTTGAGCATACCGGGC-3′; Brca2, 5′-CTGCCATGCTCCCACCTTGATGA-3′ and 5′-ACCGTGGGGCTTATACTCAGATTCCT-3′; Pqlc3, 5′-ACTCTCAGCTCTGCCTGTGCCAT-3′ and 5′-GGTGAGCAGTGGGTTCCCATAGT-3′; Bzw2, 5′-GCCATGCCTGCCTGGATA-3′ and 5′-ACTGGCTTCTGATGCTTATTCATAAA-3′; Cpe, 5′-GCTCCTCCTGTGCCATGCT-3′ and 5′-CATGCATGTTCCCAATGTATTTAAA-3′; Cd59b, 5′-GCTCCCACCTTGATGATAATGG-3′ and 5′-TGAGTCCCCTCTGAGCTCTCA-3′; Tmx4, 5′-CCACCTTGATGATAATGGACTGAA-3′ and 5′-TGCTGGGAGAGTGGTGACAA-3′; Tcfl5, 5′-GCCATGCCTGCCTGGAT-3′ and 5′-TGGATGTCGAATAAGAGTTACCAAAAG-3′; HBZ, 5′-GAGGACCATCATTGTGTCCA-3′ and 5′-AGTGCGGGAAGTAGGTCTTG-3′; HBD, 5′-AACCTCAAGGGCACTTTTTCT-3′ and 5′-GGAAACAGTCCAGGATCTCAA-3′; ALAS2, 5′-TAAGGCAACAAAGGCTGGAG-3′ and 5′-GCCTTCACATCTTCCTGGAC-3′; and ERMAP, 5′-GCTGTCTGTGCATGTGTCAG-3′ and 5′-CCACCTCACCTCCTTGGGTA-3′.
5′ RACE (Rapid Amplification of cDNA Ends)
600 ng total RNA from TER119+ fetal liver cells obtained from Klf3 +/+ and Klf3 −/− embryos (E13.5) was used as a template for first strand cDNA synthesis using the SMARTer™ RACE cDNA Amplification Kit (Clontech, Mountain View, CA, USA). The RACE PCR was performed as directed by the supplier but using 0.05 unit/μL REDTaq DNA Polymerase (Sigma Aldrich, St Louis, MO, USA). Thermal cycler settings were 24 cycles of 94°C/30 s, 64°C/30 s, and 72°C/1 min. Amplified products were resolved by electrophoresis through a 1.2% agarose gel and were purified using the Wizard SV Gel and PCR Clean-up System (Promega Corporation, Madison, WI, USA) as per the manufacturer’s manual. Nested RACE PCR was then performed as described above with slight alterations to the thermal cycler parameters: 24 cycles of 94°C/30 s, 60°C/30 s, and 72°C/1 min. First round and nested RACE PCR primers are listed respectively: reverse primer targeting exon 4 of Pu.1, 5′-GTGGGCTGGGGACAAGGTTTGAT-3′; reverse primer targeting exon 3 of Pu.1, 5′- GCTGTAGCTGCGGGGGCTGCACACT-3′. Amplicons were resolved and purified as described above and were sequenced by the Australian Genome Research Facility Ltd, Brisbane, QLD, Australia.
Chromatin immunoprecipitation (ChIP)
ChIP was performed with slight modifications based on Schmidt et al. . Briefly, one cross-linked E14.5 liver was used per IP with antibodies as follows: IgG, 5 μg sc-2027 (Santa Cruz Biotechnology, Santa Cruz, CA, USA); anti-H3K4me3, 5 μg C42D8 (Cell Signaling Technology, Danvers, MA, USA), or 5 μg ab12209 (abcam, Cambridge, MA, USA); anti-H3K9me3, 5 μg ab8898 (abcam); anti-H3K27me3, 3 μg C36B11 (Cell Signaling Technology) or 5 μg ab6002 (abcam); anti-H3K27ac, 0.5 μg #4353 (Cell Signaling Technology) or 5 μg ab4729 (abcam). Real-time PCR quantification of chromatin pull-down was performed as described above and amounts were normalized to the level of input material prior to immunoprecipitation. ChIP primer sequences have been described previously for Klf8 promoter 1a and Fam132a[49, 50]. Other primers used are as follows: Pu.2 ORR1A0 -250 bp, 5′-GAAGTCCTTCTGGCTTCTGCAT-3′ and 5′-CTGACCTTGTCTAACCCTTTTGTTTA-3′; Pu.2 ORR1A0 -120 bp, 5′-GACAAGGTCAGGAGAGGTTT-3′ and 5′-CCACAGTGACACACCCAT-3′; Myod1, 5′-TCCTATGCTTTGCCTGGTCT-3′ and 5′-GGAAGGAGGGCAGAGAGACT-3′; Serpina9, 5′-TGTGCTGGACCTGGTTTGTA-3′ and 5′-CTGGCAGCTCTCACCTCTCT-3′, and; Gapdh, 5′-GACAGTCGGAAACTGGGAAG-3′ and 5′-CATCACGTCCTCCATCATCC-3′. Base positions refer to where amplicons are centered relative to the TSS.
Vectors and cloning of PU.1 and PU.2
The vectors pPac and pPac-Klf1 were provided by Menie Merika and Stuart Orkin (Harvard Medical School, Boston, MA, USA). The plasmid pPac-Klf3 and the mammalian expression vectors pMT3 and pMT3-Klf3 were gifts from Dr José Perdomo (St George Clinical School, Sydney, Australia). The mammalian expression vector pSG5-Klf1 was supplied by James Bieker (Mount Sinai School of Medicine, New York, NY, USA). Firefly and Renilla luciferase vectors used were pGL4.10[luc2], and pGL4.23[luc2/minP] and pGL4.74[hRluc/TK] as transfection controls (Promega Corporation). The vector pEF-IRES-puro5 (pEF1α) was kindly provided by Dr Daniel Peet (University of Adelaide, Adelaide, Australia). The Renilla luciferase reporter construct pLightSwitch-Clec5a, containing approximately 1.1 kb of the human CLEC5A promoter, was purchased from Switchgear Genomics (Menlo Park, CA, USA). The Pu.2 ORR1A0 promoter (−140 to +23) was synthesized by GeneArt® (Life Technologies) and subcloned into KpnI/XhoI pGL4.10[luc2] to create pGL4.10-ORR1A0.
Full length Pu.1 and Pu.2 were cloned from the cDNA generated during the 5’̕′ RACE described above. PCRs were set up using Phusion High-Fidelity DNA Polymerase (Finnzymes OY, Espoo, Finland) as directed by the supplier. Primers used in the first round of amplification include: 10x Universal Primer A Mix (UPM) supplied from SMARTer™ RACE cDNA Amplification Kit (Clontech) as a forward primer, and a reverse primer specific for exon 5 of Pu.1; 5′-TCCGGGCCGGGCGACGGGTTAATGCTAT-3′. Thermal cycler parameters were 98°C/30 s, followed by 25 cycles of 98°C/10 s, 59°C/30 s, and 72°C/1 min, and a final cycle of 72°C/5 min. Amplified Pu.1 and Pu.2 products were resolved by electrophoresis, purified as described above and subjected to nested PCR using foward primers covering the start points of translation for PU.1 (5′-ATTACTCGAGGCTCAGCTGGATGTTACAGGCGTGCAAA-3′) and PU.2 (5′-ATTACTCGAGGCCACCATGGAGCTGGAACAGATGCAC-3′) together with the common reverse primer 5′-TAATGAATTCAGCCTGGCGGTCTCTGCGGGCGATCAGT-3′ (which includes the Pu.1 stop codon in exon 5). Parameter settings were 98°C/30 s, followed by 25 cycles of 98°C/10 s, 69°C/30 s, and 72°C/1 min, and a final cycle of 72°C/5 min. The fragments were subsequently cloned into XhoI/EcoRI pMT3 to form pMT3-Pu.1 and pMT3-Pu.2, respectively. Pu.1 and Pu.2 were then subcloned into XhoI/EcoRI pEF1α to generate pEF1α-Pu.1 and pEF1α-Pu.2. Similarly, Pu.1 and Pu.2 were cloned into BglII/HpaI pMSCVpuro and pMSCVhyg (Clontech), respectively, using the forward primers 5′-ATTAAGATCTGCTCAGCTGGATGTTACAGGCGTGCAAA-3′ and 5′- ATTAAGATCTGCCACCATGGAGCTGGAACAGATGCAC -3′ and the reverse primer 5′-TAATGTTAACAGCCTGGCGGTCTCTGCGGGCGATCAGT-3′.
COS cells were cultured as described previously . K562 and HL60 cells were maintained similarly but in RPMI 1640 culture medium (Gibco-BRL Life Technologies, Grand Island, NY, USA). Culture conditions for B1.6 erythroblast cells have been described elsewhere . B1.6 cells were induced with tamoxifen as described previously  and, with the exception of the time-course and cycloheximide experiments, were harvested for RNA or nuclear extracts after 48 h. SL2 cells were cultured in Schneider’s Drosophila medium (Gibco-BRL Life Technologies, Grand Island, NY, USA) supplemented with 10% heat-inactivated FCS and 1% penicillin/streptomycin/glutamine solution at 24°C.
Transfections and retroviral infections
COS cells were transfected with 5 μg pMT3-Pu.1, pMT3-Pu.2 or pMT3 empty using FuGENE6 (Roche Diagnostics) as instructed by the supplier. Cells were harvested after 48 h for nuclear extracts. For retroviral infection of K562 cells, Phoenix A packaging cells were transfected with 12 μg total vector DNA (12 μg pMSCVpuro-Pu.1, 12 μg pMSCVhyg-Pu.2, 6 μg pMSCVpuro plus 6 μg pMSCVhyg, or 6 μg pMSCVpuro-Pu.1 plus 6 μg pMSCVhyg-Pu.2) using Lipofectamine 2000 (Life Technologies) according to the manufacturer’s instructions. Target K562 cells were seeded at 1 × 105 cells/mL in 6-well Plates 24 h prior to infection. Forty-eight hours following infection of packaging cells, virus-containing media (VCM) were collected and passed through a 0.45 μm low protein binding filter. Non-tissue culture treated 6-well dishes were then coated with RetroNectin® (Clontech) as instructed by the manufacturer. Half the volume of VCM was then added to the 6-well dishes for 30 min at room temperature. The target K562 cells were resuspended in the remaining VCM with the addition of protamine to a final concentration of 8 μg/mL. After 30 min, the VCM was aspirated from the coated wells and was seeded with the K562 cells. The plates were subsequently centrifuged at 400 g for 1.5 h at 30°C and then incubated at 37°C with 5% CO2 overnight. After 12 h, VCM was collected from the packaging cells for a second round of infection as described above. Eight hours after the second spinoculation, the K562 cells were replenished with additional VCM and incubated for 48 h before replating in RPMI 1640 culture medium with 10% (v/v) heat-inactivated FCS (Gibco) and 1% (v/v) penicillin/streptomycin/glutamine solution (Gibco). After another 48 h, cells were subjected to antibiotic selection and maintained in 1 μg/mL puromycin dihydrochloride (Sigma) and 200 μg/mL hygromycin B (Life Technologies) as appropriate.
Separately, K562 cells were also transfected with pEF1α-Pu.2 or pEF1α empty using the Gene Pulser Xcell electroporation system (Bio-Rad, Hercules, CA, USA). Briefly, 106 cells and 20 μg plasmid were resuspended in Dulbecco’s modified eagle medium (DMEM) low glucose (LG) (Gibco-BRL Life Technologies) without serum and in a total volume of 400 μL. Cells were electroporated at 200 V, 950 μF and subsequently cultured and maintained in 2 μg/mL puromycin dihydrochloride to generate monoclonal lines.
Nuclear extracts and electrophoretic mobility shift assays (EMSAs)
Nuclear extracts were obtained and EMSAs were performed as described previously . The radiolabelled probe containing the PU.1 consensus binding site comprises sense and antisense oligonucleotides for the sequence 5′-GCTCGAGGACTTCCTCTTTCCAGTGC-3′ as described elsewhere . ORR1A0 double stranded CACCC probe sequences are as follows: CACCC#1, 5′-ATGGTGCCACTCCCTGGTCC-3′; CACCC#2, 5′-CAGTGACACACCCATTCCA-3′; CACCC#3, 5′-TTATACCCACACCCACAGTG-3′; CACCC#4, 5′-TAAGATCCTCACCCTAGTTG-3′. The positive control KLF3 binding site in the Fam132a promoter is ‘Probe C’ from . The antibodies used that were specific for the N-terminus and C-terminus of Pu.1 were 9G7 (Cell Signaling Technology) and T-21 (Santa Cruz Biotechnology), respectively. KLF1- and KLF3-specific antisera have been described previously .
Western blots were performed by standard methods. Briefly, nuclear extracts were separated by SDS-PAGE and were electrotransferred to PVDF membrane, which was then blocked with 5% skim milk in 50 mM Tris–HCl (pH 7.4), 150 mM NaCl, and 0.05% Tween 20. PU.2 protein was probed by overnight incubation of membrane in 5% skim milk with 0.2 μg/mL PU.1 antibody (T-21) (Santa Cruz Biotechnology) at 4°C. Detection was achieved using Immobilon Western Chemiluminescent HRP Substrate (Millipore Corporation, Billerica, MA, USA) and subsequently, membranes were stripped in 0.2 M NaOH for 10 min and were probed with β-actin antibody (Sigma).
SL2 cells were split into 6-well plates at a concentration of 5 × 105/mL and 24 h later were transfected with pPac-Klf1 (0, 50, or 250 ng) and pPac-Klf3 (0, 12.5, 25, 50, or 100 ng) supplemented to equal loads with pPac empty vector, together with 100 ng pGL4.74 [hRLuc/TK] and 1 μg pGL4.10 [luc2] or pGL4.10-ORR1A0 using FuGene6 (Roche Diagnostics) as instructed by the supplier. In competition assays, the pGL4.10-ORR1A0 vector (and pGL4.10 [luc2]) were driven by co-transfection of 200 ng pPac-Klf1. After 48 h, cells were harvested and lysates analyzed using the Dual-luciferase® Reporter Assay System (Promega Corporation) and a TD20/20 luminometer (Turner Biosystems, Sunnyvale, CA, USA). HEK293 cells were similarly transfected but with 0, 10, 100, and 1,000 ng pEF1α-Pu.1 (supplemented with pEF1α to a total of 1,000 ng vector) together with 1 μg pLightSwitch (empty vector) or pLightSwitch-Clec5a. As a control, 100 ng pGL4.23 [luc2/minP] Firefly luciferase vector was co-transfected. In competition assays, HEK293 cells were transfected and analyzed as above but with 1 μg pEF1α-Pu.1 together with 0, 10, 100 and 1,000 ng pEF1α-Pu.2.
Microarray data from TER119+ E14.5 fetal liver cells from Klf3 +/+ and Klf3 −/− embryos have previously been described . Monoclonal K562 cell lines stably transfected with pEF1α or pEF1α-Pu.2 (n = 3 each) were harvested for total RNA which was subsequently hybridized to Affymetrix Human Gene 1.0 ST arrays according to the manufacturer’s instructions (Affymetrix, Santa Clara, CA, USA). Hybridization and processing were performed by the Ramaciotti Centre for Gene Function Analysis (University of New South Wales, Sydney, Australia). Data were analyzed using Affymetrix® Expression Console™ software (Affymetrix). Microarray data are available in the Gene Expression Omnibus database  under accession number GSE50083.
High throughput RNA-sequencing (RNA-Seq)
RNA was extracted from TER119+-sorted liver cells from three Klf3 +/+ and three Klf3 −/− litter-matched E14.5 embryos (two litters total). Libraries were prepared using 1 μg total RNA using the TruSeq RNA Sample Prep Kit v2 (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions. The six libraries were multiplexed into two lanes using sample specific adapters such that there were three samples per lane. 100 bp paired end reads were sequenced using TruSeq v3 SBS reagents on the Hiseq 2000 (Illumina, San Diego, CA, USA). Library preparation and sequencing were performed by the Ramaciotti Centre, University of New South Wales, New South Wales, Australia. Quality control on the reads was performed using FastQC v0.10.1 available from .
Two separate strategies were used for alignment and RNA-seq analysis. The first of these sought to uniquely map reads from repeat elements by using high stringency alignment cut-offs and was primarily employed to visualize chimeric splicing events using Integrative Genomics Viewer . Reads were aligned to the mm10 Mus musculus genome using tophat2 (v2.0.4) using the default settings except for the following (-r -40 --segment-length 50 --coverage-search --segment-mismatches 0 -g 50 --genome-read-mismatches 0 --read-mismatches 0 -I 200000 --no-discordant --no-mixed --b2-L 30 --b2-D 10000 --b2-R 100 -n 0 -M) . The second approach used more relaxed cutoffs to better measure expression levels of non-repeat exons. This latter approach was employed to determine differential gene and isoform expression (Additional file 4: Tables S1 and Additional file 5: Table S2). Again, reads were aligned to the mm10 Mus musculus genome using the default settings of tophat2 (v2.0.4), except for the following (--no-discordant -M --no-coverage-search --microexon-search -n 3 --genome-read-mismatches 3 --read-mismatches 3 --b2-sensitive -G) . Transcripts were assembled using Cufflinks v2.0.2 and the mm10 annotations were included . Transcripts across all replicates were merged using cuffmerge and differential expression analysis was performed pair-wise on the groups using cuffdiff. A q-value (FDR) threshold of <0.05 was used to determine significantly differentially expressed genes. RNA-Seq data have been deposited in the Gene Expression Omnibus under the accession number GSE50554.
The sequence and genomic positions of ORR1A0 LTR elements were determined using the RepeatMasker program  in conjunction with the University of California Santa Cruz (UCSC) Genome Browser . Sequence information of retroelement families was obtained using Repbase Update [57, 77]. The overlap between differentially expressed loci as measured by RNA-Seq and ORR1A0 and ORR1A0-int elements was established using the intersect command in Bedtools v2.17.0 . Similarly, overlaps (using 100 bp windows) were determined between ORR1A0 elements and: TRIM28 ChIP peaks (from [40, 64]); TRIM28-dependent H3K9me3 peaks (from ), and regions of DNA methylation in hematopoietic stem cells and erythroblasts (from ). Gene Expression Omnibus accession numbers for the data sets analyzed are GSM1032198, GSM773067, GSM1032190, and GSE38354, respectively.
Description of additional files
The following additional data are available with the online version of this paper. Additional file 1: Figure S1 shows that the chimeric Pu.1 transcript (Pu.2) is predominantly upregulated in erythroid tissues in the absence of KLF3. Additional file 2: Figure S2 shows levels of H3K9 tri-methylation at the Pu.2 promoter in Klf3 −/− and wild-type E14.5 fetal liver cells. Additional file 3: Figure S3 shows that Pu.2 and other ORR1A0 chimeric transcripts are rapidly activated by KLF1 and in the presence of cycloheximide, suggesting that they are direct targets. Additional file 6: Figure S4 provides validation of genes to which the ORR1A0 exon is spliced in Klf3 −/− E14.5 TER119+ fetal liver cells. Additional file 7: Figure S5 gives further examples of de-repressed ORR1A0 transcripts in the absence of KLF3. Additional file 8: Figure S6 contains reporter assay data demonstrating the opposing transcriptional activities of PU.1 and PU.2. Additional file 4: Table S1 shows the list of genes that are significantly, differentially expressed in Klf3 −/− TER119+ fetal liver cells by RNA-Seq. Additional file 5: Table S2 shows the list of genes that have significantly de-repressed isoforms. Additional file 9: Table S3 shows microarray results from K562 cell stably expressing PU.2.
This work has been supported by funding from the Australian Research Council and the National Health and Medical Research Council. We would like to thank Natalie Twine for advice with bioinformatics.
- Waterston RH, Lindblad-Toh K, Birney E, Rogers J, Abril JF, Agarwal P, Agarwala R, Ainscough R, Alexandersson M, An P, Antonarakis SE, Attwood J, Baertsch R, Bailey J, Barlow K, Beck S, Berry E, Birren B, Bloom T, Bork P, Botcherby M, Bray N, Brent MR, Brown DG, Brown SD, Bult C, Buton J, Butler J, Campbell RD, Mouse Genome Sequencing Consortium, et al.: Initial sequencing and comparative analysis of the mouse genome. Nature 2002, 420:520–562.View ArticlePubMedGoogle Scholar
- Kazazian HH Jr: Mobile elements: drivers of genome evolution. Science 2004, 303:1626–1632.View ArticlePubMedGoogle Scholar
- Bourque G: Transposable elements in gene regulation and in the evolution of vertebrate genomes. Curr Opin Genet Dev 2009, 19:607–612.View ArticlePubMedGoogle Scholar
- Bourque G, Leong B, Vega VB, Chen X, Lee YL, Srinivasan KG, Chew JL, Ruan Y, Wei CL, Ng HH, Liu ET: Evolution of the mammalian transcription factor binding repertoire via transposable elements. Genome Res 2008, 18:1752–1762.View ArticlePubMedPubMed CentralGoogle Scholar
- Kunarso G, Chia NY, Jeyakani J, Hwang C, Lu XY, Chan YS, Ng HH, Bourque G: Transposable elements have rewired the core regulatory network of human embryonic stem cells. Nat Genet 2010, 42:631–634.View ArticlePubMedGoogle Scholar
- Schmidt D, Schwalie PC, Wilson MD, Ballester B, Goncalves A, Kutter C, Brown GD, Marshall A, Flicek P, Odom DT: Waves of retrotransposon expansion remodel genome organization and CTCF binding in multiple mammalian lineages. Cell 2012, 148:335–348.View ArticlePubMedPubMed CentralGoogle Scholar
- Baillie JK, Barnett MW, Upton KR, Gerhardt DJ, Richmond TA, De Sapio F, Brennan PM, Rizzu P, Smith S, Fell M, Talbot RT, Gustincich S, Freeman TC, Mattick JS, Hume DA, Heutink P, Carninci P, Jeddeloh JA, Faulkner GJ: Somatic retrotransposition alters the genetic landscape of the human brain. Nature 2011, 479:534–537.View ArticlePubMedPubMed CentralGoogle Scholar
- Ward MC, Wilson MD, Barbosa-Morais NL, Schmidt D, Stark R, Pan Q, Schwalie PC, Menon S, Lukk M, Watt S, Thybert D, Kutter C, Kirschner K, Flicek P, Blencowe BJ, Odon DT: Latent regulatory potential of human-specific repetitive elements. Mol Cell 2013, 49:262–272.View ArticlePubMedPubMed CentralGoogle Scholar
- Thornburg BG, Gotea V, Makalowski W: Transposable elements as a significant source of transcription regulating signals. Gene 2006, 365:104–110.View ArticlePubMedGoogle Scholar
- Johnson R, Gamblin RJ, Ooi L, Bruce AW, Donaldson IJ, Westhead DR, Wood IC, Jackson RM, Buckley NJ: Identification of the REST regulon reveals extensive transposable element-mediated binding site duplication. Nucleic Acids Res 2006, 34:3862–3877.View ArticlePubMedPubMed CentralGoogle Scholar
- Wang T, Zeng J, Lowe CB, Sellers RG, Salama SR, Yang M, Burgess SM, Brachmann RK, Haussler D: Species-specific endogenous retroviruses shape the transcriptional network of the human tumor suppressor protein p53. Proc Natl Acad Sci USA 2007, 104:18613–18618.View ArticlePubMedPubMed CentralGoogle Scholar
- Wang J, Zhuang J, Iyer S, Lin X, Whitfield TW, Greven MC, Pierce BG, Dong X, Kundaje A, Cheng Y, Rando OJ, Birney E, Myers RM, Noble WS, Snyder M, Weng Z: Sequence features and chromatin structure around the genomic regions bound by 119 human transcription factors. Genome Res 2012, 22:1798–1812.View ArticlePubMedPubMed CentralGoogle Scholar
- Lynch VJ, Leclerc RD, May G, Wagner GP: Transposon-mediated rewiring of gene regulatory networks contributed to the evolution of pregnancy in mammals. Nat Genet 2011, 43:1154–1159.View ArticlePubMedGoogle Scholar
- Chuong EB, Rumi MAK, Soares MJ, Baker JC: Endogenous retroviruses function as species-specific enhancer elements in the placenta. Nat Genet 2013, 45:325–329.View ArticlePubMedPubMed CentralGoogle Scholar
- Morgan HD, Sutherland HG, Martin DI, Whitelaw E: Epigenetic inheritance at the agouti locus in the mouse. Nat Genet 1999, 23:314–318.View ArticlePubMedGoogle Scholar
- Duhl DM, Vrieling H, Miller KA, Wolff GL, Barsh GS: Neomorphic agouti mutations in obese yellow mice. Nat Genet 1994, 8:59–65.View ArticlePubMedGoogle Scholar
- Faulkner GJ, Kimura Y, Daub CO, Wani S, Plessy C, Irvine KM, Schroder K, Cloonan N, Steptoe AL, Lassmann T, Waki K, Horniq N, Arakawa T, Takahashi H, Kawai J, Forrest AR, Suzuki H, Hayashizaki Y, Hume DA, Orlando V, Grimmond SM, Carninci P: The regulated retrotransposon transcriptome of mammalian cells. Nat Genet 2009, 41:563–571.View ArticlePubMedGoogle Scholar
- Fadloun A, Le Gras S, Jost B, Ziegler-Birling C, Takahashi H, Gorab E, Carninci P, Torres-Padilla ME: Chromatin signatures and retrotransposon profiling in mouse embryos reveal regulation of LINE-1 by RNA. Nat Struct Mol Biol 2013, 20:332–338.View ArticlePubMedGoogle Scholar
- Schmitz J, Brosius J: Exonization of transposed elements: a challenge and opportunity for evolution. Biochimie 1928–1934, 2011:93.Google Scholar
- Macfarlan TS, Gifford WD, Driscoll S, Lettieri K, Rowe HM, Bonanomi D, Firth A, Singer O, Trono D, Pfaff SL: Embryonic stem cell potency fluctuates with endogenous retrovirus activity. Nature 2012, 487:57–63.PubMedPubMed CentralGoogle Scholar
- Cohen CJ, Lock WM, Mager DL: Endogenous retroviral LTRs as promoters for human genes: a critical assessment. Gene 2009, 448:105–114.View ArticlePubMedGoogle Scholar
- Batut P, Dobin A, Plessy C, Carninci P, Gingeras TR: High-fidelity promoter profiling reveals widespread alternative promoter usage and transposon-driven developmental gene expression. Genome Res 2013, 23:169–180.View ArticlePubMedPubMed CentralGoogle Scholar
- Shukla R, Upton KR, Munoz-Lopez M, Gerhardt DJ, Fisher ME, Nguyen T, Brennan PM, Baillie JK, Collino A, Ghisletti S, Sinha S, Ianelli F, Radaelli E, Dos Santos A, Rapoud D, Guettier C, Samuel D, Natoli G, Carninci P, Ciccarelli FD, Garcia-Perez JL, Faivre J, Faulkner GJ: Endogenous retrotransposition activates oncogenic pathways in hepatocellular carcinoma. Cell 2013, 153:101–111.View ArticlePubMedPubMed CentralGoogle Scholar
- Faulkner GJ: Retrotransposons: mobile and mutagenic from conception to death. FEBS Lett 2011, 585:1589–1594.View ArticlePubMedGoogle Scholar
- Hancks DC, Kazazian HH Jr: Active human retrotransposons: variation and disease. Curr Opin Genet Dev 2012, 22:191–203.View ArticlePubMedPubMed CentralGoogle Scholar
- Lamprecht B, Walter K, Kreher S, Kumar R, Hummel M, Lenze D, Kochert K, Bouhlel MA, Richter J, Soler E, Stadhouders R, Johrens K, Wurster KD, Callen DF, Harte MF, Giefing M, Barlow R, Stein H, Anagnostopoulos I, Janz M, Cockerill PN, Siebert R, Dorken B, Bonifer C, Mathas S: Derepression of an endogenous long terminal repeat activates the CSF1R proto-oncogene in human lymphoma. Nat Med 2010, 16:571–579.View ArticlePubMedGoogle Scholar
- Karimi MM, Goyal P, Maksakova IA, Bilenky M, Leung D, Tang JX, Shinkai Y, Mager DL, Jones S, Hirst M, Lorincz MC: DNA methylation and SETDB1/H3K9me3 regulate predominantly distinct sets of genes, retroelements, and chimeric transcripts in mESCs. Cell Stem Cell 2011, 8:676–687.View ArticlePubMedGoogle Scholar
- Rowe HM, Trono D: Dynamic control of endogenous retroviruses during development. Virology 2011, 411:273–287.View ArticlePubMedGoogle Scholar
- Leung DC, Lorincz MC: Silencing of endogenous retroviruses: when and why do histone marks predominate? Trends Biochem Sci 2012, 37:127–133.View ArticlePubMedGoogle Scholar
- Bourc'his D, Bestor TH: Meiotic catastrophe and retrotransposon reactivation in male germ cells lacking Dnmt3L. Nature 2004, 431:96–99.View ArticlePubMedGoogle Scholar
- Matsui T, Leung D, Miyashita H, Maksakova IA, Miyachi H, Kimura H, Tachibana M, Lorincz MC, Shinkai Y: Proviral silencing in embryonic stem cells requires the histone methyltransferase ESET. Nature 2010, 464:927–931.View ArticlePubMedGoogle Scholar
- Rowe HM, Friedli M, Offner S, Verp S, Mesnard D, Marquis J, Aktas T, Trono D: De novo DNA methylation of endogenous retroviruses is shaped by KRAB-ZFPs/KAP1 and ESET. Development 2013, 140:519–529.View ArticlePubMedPubMed CentralGoogle Scholar
- Reichmann J, Crichton JH, Madej MJ, Taggart M, Gautier P, Garcia-Perez JL, Meehan RR, Adams IR: Microarray analysis of LTR retrotransposon silencing identifies Hdac1 as a regulator of retrotransposon expression in mouse embryonic stem cells. PLoS Comput Biol 2012, 8:e1002486.View ArticlePubMedPubMed CentralGoogle Scholar
- Macfarlan TS, Gifford WD, Agarwal S, Driscoll S, Lettieri K, Wang J, Andrews SE, Franco L, Rosenfeld MG, Ren B, Pfaff SL: Endogenous retroviruses and neighboring genes are coordinately repressed by LSD1/KDM1A. Genes Dev 2011, 25:594–607.View ArticlePubMedPubMed CentralGoogle Scholar
- Maksakova IA, Thompson PJ, Goyal P, Jones SJ, Singh PB, Karimi MM, Lorincz MC: Distinct roles of KAP1, HP1 and G9a/GLP in silencing of the two-cell-specific retrotransposon MERVL in mouse ES cells. Epigenetics Chromatin 2013, 6:15.View ArticlePubMedPubMed CentralGoogle Scholar
- Leeb M, Pasini D, Novatchkova M, Jaritz M, Helin K, Wutz A: Polycomb complexes act redundantly to repress genomic repeats and genes. Genes Dev 2010, 24:265–276.View ArticlePubMedPubMed CentralGoogle Scholar
- Thomas JH, Schneider S: Coevolution of retroelements and tandem zinc finger genes. Genome Res 2011, 21:1800–1812.View ArticlePubMedPubMed CentralGoogle Scholar
- Wolf D, Goff SP: Embryonic stem cells use ZFP809 to silence retroviral DNAs. Nature 2009, 458:1201–1204.View ArticlePubMedPubMed CentralGoogle Scholar
- Rowe HM, Jakobsson J, Mesnard D, Rougemont J, Reynard S, Aktas T, Maillard PV, Layard-Liesching H, Verp S, Marquis J, Spitz F, Constam DB, Trono D: KAP1 controls endogenous retroviruses in embryonic stem cells. Nature 2010, 463:237–240.View ArticlePubMedGoogle Scholar
- Rowe HM, Kapopoulou A, Corsinotti A, Fasching L, Macfarlan TS, Tarabay Y, Viville S, Jakobsson J, Pfaff SL, Trono D: TRIM28 repression of retrotransposon-based enhancers is necessary to preserve transcriptional dynamics in embryonic stem cells. Genome Res 2013, 23:452–461.View ArticlePubMedPubMed CentralGoogle Scholar
- McConnell BB, Yang VW: Mammalian Kruppel-like factors in health and diseases. Physiol Rev 2010, 90:1337–1381.View ArticlePubMedPubMed CentralGoogle Scholar
- Feng WC, Southwood CM, Bieker JJ: Analyses of beta-thalassemia mutant DNA interactions with erythroid Kruppel-like factor (EKLF), an erythroid cell-specific transcription factor. J Biol Chem 1994, 269:1493–1500.PubMedGoogle Scholar
- Pearson RC, Jack BH, Lee SHY, Funnell AP, Crossley M: Co-regulator interactions in Kruppel-like factor transcriptional programs. In The Biology of Kruppel-like Factors. Edited by: Nagai R, Kasuga M, Friedman MD. Tokyo: Springer; 2009:51–64.View ArticleGoogle Scholar
- Tallack MR, Perkins AC: KLF1 directly coordinates almost all aspects of terminal erythroid differentiation. IUBMB Life 2010, 62:886–890.View ArticlePubMedGoogle Scholar
- Funnell AP, Maloney CA, Thompson LJ, Keys J, Tallack M, Perkins AC, Crossley M: Erythroid Kruppel-like factor directly activates the basic Kruppel-like factor gene in erythroid cells. Mol Cell Biol 2007, 27:2777–2790.View ArticlePubMedPubMed CentralGoogle Scholar
- Crossley M, Whitelaw E, Perkins A, Williams G, Fujiwara Y, Orkin SH: Isolation and characterization of the cDNA encoding BKLF/TEF-2, a major CACCC-box-binding protein in erythroid cells and selected other cells. Mol Cell Biol 1996, 16:1695–1705.View ArticlePubMedPubMed CentralGoogle Scholar
- Turner J, Crossley M: Cloning and characterization of mCtBP2, a co-repressor that associates with basic Kruppel-like factor and other mammalian transcriptional regulators. EMBO J 1998, 17:5129–5140.View ArticlePubMedPubMed CentralGoogle Scholar
- Shi Y, Sawada J, Sui G, Affar EB, Whetstine JR, Lan F, Ogawa H, Luke MP, Nakatani Y, Shi Y: Coordinated histone modifications mediated by a CtBP co-repressor complex. Nature 2003, 422:735–738.View ArticlePubMedGoogle Scholar
- Funnell AP, Norton LJ, Mak KS, Burdach J, Artuz CM, Twine NA, Wilkins MR, Power CA, Hung TT, Perdomo J, Koh P, Bell-Anderson KS, Orkin SH, Fraser ST, Perkins AC, Pearson RC, Crossley M: The CACCC-binding protein KLF3/BKLF represses a subset of KLF1/EKLF target genes and is required for proper erythroid maturation in vivo. Mol Cell Biol 2012, 32:3281–3292.View ArticlePubMedPubMed CentralGoogle Scholar
- Eaton SA, Funnell AP, Sue N, Nicholas H, Pearson RC, Crossley M: A network of Kruppel-like factors (Klfs): Klf8 is repressed by Klf3 and activated by Klf1 in vivo. J Biol Chem 2008, 283:26937–26947.View ArticlePubMedPubMed CentralGoogle Scholar
- Perkins AC, Sharpe AH, Orkin SH: Lethal beta-thalassaemia in mice lacking the erythroid CACCC-transcription factor EKLF. Nature 1995, 375:318–322.View ArticlePubMedGoogle Scholar
- Smit AFA: Identification of a new, abundant superfamily of mammalian LTR-transposons. Nucleic Acids Res 1863–1872, 1993:21.Google Scholar
- FitzGerald PC, Sturgill D, Shyakhtenko A, Oliver B, Vinson C: Comparative genomics of Drosophila and human core promoters. Genome Biol 2006, 7:R53.View ArticlePubMedPubMed CentralGoogle Scholar
- Bell-Anderson KS, Funnell AP, Williams H, Mat Jusoh H, Scully T, Lim WF, Burdach JG, Mak KS, Knights AJ, Hoy AJ, Nicholas HR, Sainsbury A, Turner N, Pearson RC, Crossley M: Loss of Kruppel-like factor 3 (KLF3/BKLF) leads to upregulation of the insulin-sensitizing factor adipolin (FAM132A/CTRP12/C1qdc2). Diabetes 2013, 62:2728–2737.View ArticlePubMedPubMed CentralGoogle Scholar
- Kooistra SM, Helin K: Molecular mechanisms and potential functions of histone demethylases. Nat Rev Mol Cell Biol 2012, 13:297–311.PubMedGoogle Scholar
- Coghill E, Eccleston S, Fox V, Cerruti L, Brown C, Cunningham J, Jane S, Perkins A: Erythroid Kruppel-like factor (EKLF) coordinates erythroid cell proliferation and hemoglobinization in cell lines derived from EKLF null mice. Blood 1861–1868, 2001:97.Google Scholar
- Jurka J, Kapitonov VV, Pavlicek A, Klonowski P, Kohany O, Walichiewicz J: Repbase update, a database of eukaryotic repetitive elements. Cytogenet Genome Res 2005, 110:462–467.View ArticlePubMedGoogle Scholar
- Funnell AP, Mak KS, Twine NA, Pelka GJ, Norton LJ, Radziewic T, Power M, Wilkins MR, Bell-Anderson KS, Fraser ST, Perkins AC, Tam PP, Pearson RC: Generation of mice deficient in both KLF3/BKLF and KLF8 reveals a genetic interaction and a role for these factors in embryonic globin gene silencing. Mol Cell Biol 2013, 33:2976–2987.View ArticlePubMedPubMed CentralGoogle Scholar
- Mak KS, Funnell AP, Pearson RC, Crossley M: PU.1 and haematopoietic cell fate: dosage matters. Int J Cell Biol 2011, 2011:808524.View ArticlePubMedPubMed CentralGoogle Scholar
- Hagemeier C, Bannister AJ, Cook A, Kouzarides T: The activation domain of transcription factor PU.1 binds the retinoblastoma (RB) protein and the transcription factor TFIID in vitro: RB shows sequence similarity to TFIID and TFIIB. Proc Natl Acad Sci USA 1993, 90:1580–1584.View ArticlePubMedPubMed CentralGoogle Scholar
- Batliner J, Mancarelli MM, Jenal M, Reddy VA, Fey MF, Torbett BE, Tschan MP: CLEC5A (MDL-1) is a novel PU.1 transcriptional target during myeloid differentiation. Mol Immunol 2011, 48:714–719.View ArticlePubMedGoogle Scholar
- Delgado MD, Gutierrez P, Richard C, Cuadrado MA, Moreau-Gachelin F, Leon J: Spi-1/PU.1 proto-oncogene induces opposite effects on monocytic and erythroid differentiation of K562 cells. Biochem Biophys Res Commun 1998, 252:383–391.View ArticlePubMedGoogle Scholar
- Guallar D, Perez-Palacios R, Climent M, Martinez-Abadia I, Larraga A, Fernandez-Juan M, Vallejo C, Muniesa P, Schoorlemmer J: Expression of endogenous retroviruses is negatively regulated by the pluripotency marker Rex1/Zfp42. Nucleic Acids Res 2012, 40:8993–9007.View ArticlePubMedPubMed CentralGoogle Scholar
- Quenneville S, Verde G, Corsinotti A, Kapopoulou A, Jakobsson J, Offner S, Baglivo I, Pedone PV, Grimaldi G, Riccio A, Trono D: In embryonic stem cells, ZFP57/KAP1 recognize a methylated hexanucleotide to affect chromatin and DNA methylation of imprinting control regions. Mol Cell 2011, 44:361–372.View ArticlePubMedPubMed CentralGoogle Scholar
- Hogart A, Lichtenberg J, Ajay SS, Anderson S, Margulies EH, Bodine DM, Center NIHIS: Genome-wide DNA methylation profiles in hematopoietic stem and progenitor cells reveal overrepresentation of ETS transcription factor binding sites. Genome Res 2012, 22:1407–1418.View ArticlePubMedPubMed CentralGoogle Scholar
- Sue N, Jack BH, Eaton SA, Pearson RC, Funnell AP, Turner J, Czolij R, Denyer G, Bao S, Molero-Navajas JC, Perkins A, Fujiwara Y, Orkin SH, Bell-Anderson K, Crossley M: Targeted disruption of the basic Kruppel-like factor gene (Klf3) reveals a role in adipogenesis. Mol Cell Biol 2008, 28:3967–3978.View ArticlePubMedPubMed CentralGoogle Scholar
- Primer-BLAST. http://0-www.ncbi.nlm.nih.gov.brum.beds.ac.uk/tools/primer-blast/
- Schmidt D, Wilson MD, Spyrou C, Brown GD, Hadfield J, Odom DT: ChIP-seq: using high-throughput sequencing to discover protein-DNA interactions. Methods 2009, 48:240–248.View ArticlePubMedPubMed CentralGoogle Scholar
- Li SL, Schlegel W, Valente AJ, Clark RA: Critical flanking sequences of PU.1 binding sites in myeloid-specific promoters. J Biol Chem 1999, 274:32453–32460.View ArticlePubMedGoogle Scholar
- Gene expression omnibus. http://0-www.ncbi.nlm.nih.gov.brum.beds.ac.uk/geo/
- FastQC. http://www.bioinformatics.babraham.ac.uk/projects/fastqc/
- Robinson JT, Thorvaldsdottir H, Winckler W, Guttman M, Lander ES, Getz G, Mesirov JP: Integrative genomics viewer. Nat Biotechnol 2011, 29:24–26.View ArticlePubMedPubMed CentralGoogle Scholar
- Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, Salzberg SL: TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol 2013, 14:R36.View ArticlePubMedPubMed CentralGoogle Scholar
- Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ, Salzberg SL, Wold BJ, Pachter L: Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol 2010, 28:511–515.View ArticlePubMedPubMed CentralGoogle Scholar
- RepeatMasker. http://www.repeatmasker.org/
- UCSC genome bioinformatics. http://genome.ucsc.edu/
- Repbase Update. http://www.girinst.org/repbase/index.html
- Quinlan AR, Hall IM: BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics 2010, 26:841–842.View ArticlePubMedPubMed CentralGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.