A Novel Pseudoautosomal Human Gene Encodes A Putative Protein Similar to Ac-like Transposases

Human Molecular Genetics, Jan 1999

We report the cloning of a novel gene, called Tramp, in the Xp/Yp PAR region that has a functional homologue on the Y chromosome and escapes X-inactivation. This gene encodes, within a single exon, a putative protein that has amino acid similarity with transposases of the Ac family. Flanking this gene we have identified putative terminal inverted repeats (TIRs) and a duplicate target site, suggesting that it may be an ancient transposable element. The nucleotide differences in these sites and the TIR-binding inactivity of the putative Tramp protein suggest that this element is not an autonomous transposon. In the human genome, the Tramp protein may be involved in the transposition of other transposable elements, like medium reiterated frequency repeats, or it could be specialized in the acquisition of a new cellular function.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

http://hmg.oxfordjournals.org/content/8/1/61.full.pdf

A Novel Pseudoautosomal Human Gene Encodes A Putative Protein Similar to Ac-like Transposases

Teresa Esposito 1 2 Fernando Gianfrancesco 2 Alfredo Ciccodicola 1 Luisa Montanini 2 Steven Mumm 0 Michele D'Urso 1 Antonino Forabosco 2 0 Division of Bone and Mineral Diseases, Washington University School of Medicine , St Louis, MO 63110, USA 1 International Institute of Genetics and Biophysics (IIGB) , CNR, 80125 Naples, Italy 2 Dipartimento di Scienze Morfologiche e Medico-Legali, Sezione di Istologia, Embriologia e Genetica, Universit di Modena , Via del Pozzo 71, 41100 Modena, Italy We report the cloning of a novel gene, called Tramp, in the Xp/Yp PAR region that has a functional homologue on the Y chromosome and escapes X-inactivation. This gene encodes, within a single exon, a putative protein that has amino acid similarity with transposases of the Ac family. Flanking this gene we have identified putative terminal inverted repeats (TIRs) and a duplicate target site, suggesting that it may be an ancient transposable element. The nucleotide differences in these sites and the TIR-binding inactivity of the putative Tramp protein suggest that this element is not an autonomous transposon. In the human genome, the Tramp protein may be involved in the transposition of other transposable elements, like medium reiterated frequency repeats, or it could be specialized in the acquisition of a new cellular function. - Transposable elements are discrete DNA segments that are mobile and able to transport themselves to other locations within the genome at non-homologous insertion sites. Such elements are widespread in nature, having been identified in virtually all organisms examined (1). The transposable elements with a DNA intermediate are called transposons. They are characterized by terminal inverted repeats (TIRs) and duplication of the target site (visible as direct repeats flanking the element). Moreover, the autonomous elements (able to promote their transposition) encode an active transposase that binds specifically to the TIRs and through a cut-and-paste mechanism catalyses excision of the transposon from its original location and promotes its re-integration elsewhere in the genome (2). Many transposons have been studied in invertebrate and plant species (1) and recently they have been described in vertebrate genomes (3). In the eukaryotes, the best studied groups are Tc1/mariner and Ac-like elements, defined on the basis of similarities between the sequences encoding the transposase. In general, for a given member of the family, a genome contains few copies of the autonomous full-length elements and numerous copies of internally deleted defective elements. Until now, not a single autonomous element has been isolated from vertebrates, and the two Mariner transposons found in humans are also defective (4). They seem to be mutated in the transposase gene, as a result of a process called vertical inactivation (5). For this reason, and to avoid extinction, the transposons must find ways to establish themselves in a new host (horizontal transmission). The introduction into vegetal heterologous species has been successful (6), whereas in animals a transposable species-specific action is present because factors produced by the host are required (79). Because of the enormous power of transposons, to find an active element or an active transposase is an important step for the development of potential vectors containing exogenous DNA. In our search for genes which escape X-inactivation we have identified a novel pseudoautosomal human gene called Tramp, probably originating from an ancient transposon, encoding a protein similar to transposases of the Ac family. Transposons of this family have similar transposases, TIRs of 11 bp and a duplicated target site of 8 bp. This family includes at least four members: Ac from Zea mays (10,11); Hobo from Drosophila melanogaster (12,13); Tam3 from Antirrhinum majus (14,15); and the recently isolated Tol2 element from Oryzias latipes (16). Isolation and characterization of Tramp Previously, EST 832 was isolated from human skeletal muscle (17). By PCR-based screening (18), using the EST primers, a cDNA of almost 3000 bp was recovered from a cDNA library derived from a human uninduced male teratocarcinoma cell line, NT2/D1 (19). A PCR product of 456 bp made from the 3 end of this cDNA (3 probe) was used as a probe to examine the levels of expression in various tissue types. Northern analysis detected only one RNA species of ~ 5000 bp in all human adult tissues tested, with higher levels of expression in muscle and heart (Fig. 1). It was apparent that a full-length cDNA would be longer than the initially isolated 3000 bp clone. The remainder of the cDNA clone was determined by PCR assay on cDNA pools from a previously described teratocarcinoma library (18,19) using a primer from the transcript (75 bp from the 5 end) and primer R of the l gt10 vector. A clone of 1076 bp that overlapped with the clone of 3000 bp was isolated and sequenced. Finally, another primer (37 bp from the 5 end of this new clone) was used to repeat the previous experiment and to obtain an additional clone of 312 bp that was sequenced. From the assembly of these three clones we have obtained a consensus cDNA sequence of 4357 bp. It contains an open reading frame (ORF) that encodes a putative protein of 694 amino acids preceded by an in-frame stop codon (TGA). The 3-UTR contains two Alu repetitive elements (nt 23042595 and 31513411) and a putative polyadenylation signal (AATAAA) at 21 nt from the poly(A) tail. Using the deduced Tramp protein sequence as a query, some significant sequence similarities were detected in a search of the GenBank and EMBL databases. The best similarities were with the Ac transposase (X05424) of Zea mays (10,11), Hobo transposase (M69216) of Drosophila melanogaster (12,13), Tam3 transposase (X55078) of Antirrhinum majus (14,15), Tol2 transposase (D84375) of Oryzias latipes (16) and with the Hobo transposase-like proteins C10A4.1 (U23454), K09A11.1 (Z50742) and C53D6.6 (Z70270) of Caenorhabditis elegans (20). As expected, Tramp protein shows higher identity (27%) with and similarity (49%) to Tol2, the only vertebrate transposase described of this group (Fig. 2A), and lower identity (2022%) with and similarity (3540%) to the other transposases. Notably, Tol2 transposase lacks the C-terminal region that is essential for transposase activity (13), while this region is highly homologous between Tramp, Ac, Hobo and Tam3. The alignment of this region does not require the introduction of gaps (Fig. 2B). Moreover, the amino acid sequence of Tramp shows similarity with DREF, a transcriptional regulatory factor which specifically Figure 2. Comparison of the amino acid sequences. Amino acids are given in the single letter code. The vertical lines represent identical amino acids and the colons represent conservative amino acid changes. (A) Comparison of the amino acid sequences between Tramp and Tol2 transposases. The C terminal region, essential for transposase activity, is indicated by solid lines. (B) Alignment of the amino acid sequences of a C-terminal region for Tramp, Ac, Hobo and Tam3. binds the promoter-activating element DNA replication-related element (DRE) of Drosophila DNA replication-related genes (21). The sequence similarity is stronger (31% identity, 50% similarity) in the N-terminal region. Genomic organization of Tramp A set of primers based on the cDNA sequence was used for direct sequencing of the cosmid U231H7 (positive for the Tramp STS) to verify the genomic organization of the gene. The genomic sequence, compared with that of the cDNA, showed that no introns were present. In fact, the two sequences were almost identical. There were 13 nucleotide differences along the entire length: all nucleotide differences were outside the coding region. Using the Promoter Prediction program (http://www-hgc.lbl.gov/ projects/promoter.html ), a putative promoter sequence (nt 2878) was identified. In order to verify that Tramp was part of a human autonomous transposable element (by the presence of TIRs and a duplicated target site), we extended the sequence by synthesizing new primers and walking into undetermined regions towards the 5 and 3 ends of cosmid U231H7. Following this method, we identified the putative TIRs and the 8 bp of the duplicated target site. These points, that represented the ends of the transposon, had some differences, suggesting that Tramp may not be able to move itself in the genome (Fig. 3). In order to test the TIR-binding activity of our protein, two Tramp gene segments of 1020 and 2082 bp, encoding the N-terminal 340 amino acids and the full ORF of 694 amino acids, respectively, were cloned into a GST gene fusion vector system. These two proteins were expressed in Escherichia coli and purified using the GST Purification Module (Pharmacia Biotech). Upon incubation of these proteins with a radiolabelled 300 bp DNA fragment containing either the putative left or right Tramp TIR, no nucleoprotein complex was observed in a mobility shift assay (data not shown). The same result was confirmed when the two proteins were incubated with radiolabelled 30 bp oligonucleotides containing the putative Tramp TIRs (data not shown). Chromosomal localization of Tramp Previously, EST 832 was localized on the X and Y chromosomes (17). The sub-regional Tramp gene localization on the X chromosome was obtained by PCR on a panel of somatic cell hybrids that subdivides the X chromosome into 31 intervals (22). The gene was mapped to Xp22.33, telomeric to marker DXS1233 (data not shown). When YAC collections of overlapping clones across XpPAR (23) were screened, six positive clones (yWXD1974, yWXD1976, A0101, D1116, yOX222 and yOX223) were recovered. This result localized the Tramp gene to 350450 kb distal of the boundary of PAR1 (Fig. 4). By Southern blot hybridization assay, the 3 probe detected a 4 kb EcoRI fragment in the YAC yWXD1974 (which was used as a positive control) as well as in DNAs from hybrid cells containing a human X chromosome or a human Y chromosome. The same fragment was detected in male and female DNA. Surprisingly, at longer exposure, other EcoRI fragments whose sizes varied between 0.5 and 4 kb were observed only in male and female DNAs (Fig. 5, left). This result suggests the presence of other copies (probably deleted) of this element in the human genome. The same filter was reprobed with a fragment of ~ 1200 bp containing part of the coding region (coding region probe) and it showed, as expected, only a fragment of ~ 2600 bp in YAC, somatic cell hybrid and human DNAs at short exposure. At least two other fragments of 1700 and 1200 bp were detected in human DNAs at longer exposures (Fig. 5, right). X-inactivation studies To assess the transcriptional activity of Tramp on the X chromosome in various cell lines with active and inactive X chromosomes, RNA was amplified by reverse transcription (RT) and PCR using specific primer pairs (see Materials and Methods). RTPCR amplification of Tramp, with tested primers, was present in both the active and the inactive X- as well as in the Y-containing hybrid cell lines, 4X and male and female cell lines (Fig. 6). Because the primer pairs (832F/832R) amplified identical products from genomic DNA, the RTPCR assay was performed with and without reverse transcriptase to confirm that amplification was from RNA and not from contaminating DNA in the RNA preparations. Control genes showed the expected amplification products from the active X for HPRT (hypoxanthine phosphoribosyl transferase, a gene known to undergo HNuumclaenicMAocliedcsuRlaesreGarecnhe,t1ic9s9, 41,9V9o9l,. V2o2l,.N8o, .N1o. 1 X-inactivation), from the inactive X for XIST (24) and from the active X, the inactive X and the Y for MIC2 (25; Fig. 6). The escape from inactivation for the Tramp gene was confirmed by RTPCR assays with two other primer pairs (832V/832Y and 832N/832H) (see Materials and Methods) based on the cDNA and genomic sequences (data not shown). In our search for genes which escape X-inactivation, we identified a novel pseudoautosomal gene that has a functional homologue on the Y chromosome and escapes X-inactivation. This gene is an ancient transposon that encodes a putative transposase of the Ac family. In a mobile cut-and-paste type transposon system, there are two fundamental components: a source of an active transposase and two DNA sequences (TIRs) that are recognized and mobilized by the transposase. Indeed, the Tramp gene contains an intact ORF, found in both the cDNA and genomic clones, encoding a putative protein of 694 amino acids. Comparative analysis of our putative protein with the transposases of the other organisms showed that it is more homologous to Tol2 (the only known vertebrate transposase of this group) than Ac, Hobo and Tam3. The best matches were in the C-terminal region of these proteins, although Tol2 lacks this region. It was previously demonstrated, by mutational analysis of the Cterminal region, that this part is essential for transposase activity (13). This observation supports the hypothesis that Tol2 transposase may not be able to promote transposition, whereas Tramp transposase, which contains all the domains required to promote transposition, represents the first active candidate of the Ac family of the vertebrates transposases. Smit and Riggs (26), analysing 40 fragments of human medium reiterated frequency repeats (MERs), report that 13 MERs resemble fossils of DNA transposons. They divided them into groups MER1 and MER2. The MER2 group (~ 35 000 in the human genome) has TIRs and a duplicated target site (TA) typical of the Tc/mariner family of DNA transposons. Looking for sources of transposases responsible for accumulation of this MER2 group, they found Tigger1 and Tigger2. The MER1 group (>100 000 MERs in the human genome) has TIRs and a duplicated target site (8 bp) typical of the Ac family. The authors did not find the sources (transposase) for this group in the human genome and suggested that somewhere in the genome there is an autonomous/non-autonomous element that has been responsible for spread of the MER1 group. However, at this stage, we do not know the relationship between Tramp and the MER1 group, but since Tramp transposase is the only member of the Ac family found in the human genome, it may be, directly or indirectly, responsible for the spread of ~ 100 000 MER elements. In the Tramp element, we have also identified the putative TIRs and the duplicated target site. Comparative analysis does not show a perfect match of the nucleotide sequences in the TIRs and duplicated target site, suggesting that this transposon may be unable to move. To verify if Tramp transposase can recognize its flanking TIRs, two gene segments encoding the N-terminal and the full Tramp transposase were expressed in E.coli and used in a mobility shift assay. In this prokaryotic expression system the human Tramp transposase is unable to bind its flanking TIRs, supporting the hypothesis that this element is not autonomous. Further analysis is required to define the functional activity of Tramp as a transposase in eukaryotic cells. HNuumclaenicMAocliedcsuRlaesreGarecnhe,t1ic9s9, 41,9V9o9l,. V2o2l,.N8o, .N1o. 1 Figure 5. Southern blot analysis of genomic DNAs digested with EcoRI. The bars on the left indicate the position of migration of the DNA markers. (Left) The filter was hybridized with the 3 probe. (Right) The same filter was reprobed with the coding region probe. Moreover, Tramp transposase could be involved in other important biological functions because this protein is homologous to DREF, a promoter-activating factor for Drosophila DNA replication-related genes. An example of the acquisition of a cellular function by a transposable element has already been reported in the human genome. This protein, CENP-B, has a central function in the assembly of centromere structures (2729) and it is derived from a pogo-like transposase. In conclusion, looking to the future, Tramp may prove useful as an efficient vector for transposon tagging, enhancer trapping and transgenesis in species in which DNA transposon technology is currently not available. MATERIALS AND METHODS Southern blot hybridization was carried out in 5 SSPE, 5 Denhardts, 0.5% SDS at 65 C. The filter was washed once in 2 SSC (300 mM NaCl, 30 mM sodium citrate, pH 7.5), 1% SDS at 65 C and twice in 0.2 SSC, 0.2% SDS at 65 C. The membrane was exposed to XAR-5 film at 80 C for 24 h for a short exposure and 5 days for a long exposure. cDNA and sequence analysis The cDNA library used in this work was from a human uninduced male teratocarcinoma cell line, NT2/D1 (19). The cDNA was subcloned into vector pGEM-4Z (Promega Biotech) and analysed by Dye-Terminator cycle sequencing on an Applied Biosystems 373A automated sequencer. The cDNA and genomic sequences were deposited in the EMBL database with accession nos Y16947 and Y17156, respectively. Sequence databases were searched using the BLAST sequence alignment program (30). Northern blot analysis Two human multiple tissue northern blots (Clontech) were hybridized with the 3 and coding region probes and with a b -actin probe to verify the relative normalization of mRNA amounts. The northern blots were prehybridized, hybridized and washed by an ExpessHyb hybridization solution protocol (Clontech) (31). The filters were exposed for 16 h for Tramp probes and 2 h for b -actin probe. The panel of somatic cell hybrid lines (32) used in RTPCR assays comprises: two hybrids retaining the active human X chromosome, Hy136c and GM06318B; three hybrids retaining an inactive human X chromosome, THX88, Hy70C4T3 and Y.162.5E1T2 [in addition to an inactive human X chromosome, this hybrid retains a portion of chromosome 5 (5pter5cen) and a fragment of chromosome 12 (12q24.3qter)]; a hybrid retaining two inactive human X chromosomes, X8/6T2; two hybrids retaining the human Y chromosome, GM06317 and Hy853. The human lymphoblastoid cell line GM1416 (48,XXXX) (NIGMS, Camden, NJ) and normal male and female cell lines were used as positive controls. The hamster cell line YH21 was used as a negative control. Analysis of expression from active and inactive X chromosomes RNAs from cell lines were obtained by extraction in guanidinium thiocyanate followed by centrifugation in caesium chloride solution (33). Aliquots of 10 m g of total RNA, 1 reaction buffer (40 mM TrisHCl, pH 7.9, 10 mM NaCl, 6 mM MgCl2), 16.5 U RNasin (Promega) and 7 U RQ1 DNase (Promega) were incubated in 50 m l total volume at 37C for 30 min. After incubation, the RNA was purified on Strataclean resin (Stratagene). Approximately 1 m g of total RNA was reverse transcribed in a 50 m l reaction mixture containing 1 RT buffer (20 mM TrisHCl, pH 8.4, 50 mM KCl, 2.5 mM MgCl2), 10 mM DTT, 0.5 mM dNTP, 0.2 m g random hexamers (Boehringer Mannheim) and 200 U SuperScript reverse transcriptase (Gibco BRL). After a 60 min incubation at 37C, 1 gm of DNase-free RNase was added and incubated for 10 min at 37 C. The cDNA formed was extracted twice with phenol/chloroform/isoamyl alcohol (25:24:1). The cDNA was then precipitated overnight with a 0.1 vol of 3 M sodium acetate, pH 5.2, and 2.5 vol of ethanol. Approximately 1 m g of cDNA was recovered. RTPCR experiments were carried out using 100 ng of RNA and 5 ng of cDNA as the template in a 10 m l PCR reaction containing 1 TNK 100 buffer (34), 0.2 mM dNTP, 0.35 U AmpliTaq polymerase (Boehringer Mannheim) and 0.5 m M each primer sequence derived from the EST. Using an MJR DNA Thermal Cycler (M.J.Research), we carried out 35 cycles of amplification using a step program: 1 min at 94 C, 2 min at 55 C, 2 min at 72C. The Tramp primers used in the RTPCR assay were: 832 F, 5-CAGCGTTCTCTTCACGTCTCTAACA-3; 832 R, 5-GTCCAACTTTACAGCATTAAATAAG-3; 832 V, 5-ATCTCTACCAAGGCCATCCC-3; 832 Y, 5-ATGGACAGGCAGTTGGGGGC-3; 832 N, 5-CTCAACGAAGACCCCCTCAA-3; 832 H, 5-CAATGGGCTGCTGCGGGGAT-3. Recombinant protein and gel retardation We cloned the N-terminal fragment and the full Tramp protein in a GST gene fusion vector pGEX-2T (Pharmacia Biotech). Induction of these proteins was in E.coli strain HB101 by the addition of 0.1 mM IPTG at 0.6 OD600 and continued for 3 h at 37 C. Cells were sonicated in 1 PBS, after which we added 1% Triton X-100 to the sonicated cells. The soluble fraction was added to a GSTglutathione affinity column (Pharmacia Biotech) according to the recommendations of the manufacturer. The resin was washed with 1 PBS; bound proteins were eluted with elution buffer (10 mM reduced glutathione, 50 mM TrisHCl, pH 8) and dialysed overnight at 4 C against 1 PBS, 20% glycerol, 0.02 mM DTT, 0.2 mM PMSF. Two 300 bp fragments containing the putative left and right Tramp TIRs were labelled using [a -32P]dCTP and Klenow fragment. Nucleoprotein complexes were formed in 20 mM HEPES, pH 7.5, 0.1 mM EDTA, 0.1 mg/ml BSA, 150 mM NaCl, 1 mM DTT in a total volume of 10 m l. Reactions contained 100 pg of labelled probe, 2 m g of poly(dI-dC) and 2 m g of the two Tramp proteins. After 15 min incubation on ice, 5 m l of loading dye containing 50% glycerol and bromophenol blue was added and the samples were loaded onto a 5% polyacrylamide gel. Aliquots of 1 pmol double-stranded oligonucleotide probes were endlabelled with g -ATP and kinase. The probes were purified on a G50 column. Nucleoprotein complexes were formed in 20 mM HEPES, pH 7.9, 0.1 mM EDTA, 50 mM NaCl, 0.5 mM DTT, 0.5 mM PMSF, 20% glycerol, 2 m g of poly(dI-dC), 2 mM Mg spermidine, 5 m g of BSA, 750 ng of protein and 10 00020 000 c.p.m. of probe. After 15 min incubation on ice and 20 min incubation at room temperature, 5 m l of loading dye containing 50% glycerol and bromophenol blue were added and the samples were loaded onto a 5% polyacrylamide gel.


This is a preview of a remote PDF: http://hmg.oxfordjournals.org/content/8/1/61.full.pdf

Teresa Esposito, Fernando Gianfrancesco, Alfredo Ciccodicola, Luisa Montanini, Steven Mumm, Michele D 'Urso, Antonino Forabosco. A Novel Pseudoautosomal Human Gene Encodes A Putative Protein Similar to Ac-like Transposases, Human Molecular Genetics, 1999, 61-67, DOI: 10.1093/hmg/8.1.61