The ancient mariner sails again: transposition of the human Hsmar1 element by a reconstructed transposase and activities of the SETMAR protein on transposon ends

Hsmar1, one of the two subfamilies of mariner transposons in humans, is an ancient element that entered the primate genome lineage approximately 50 million years ago. Although Hsmar1 elements are inactive due to mutational damage, one particular copy of the transposase gene has apparently been under selection. This transposase coding region is part of the SETMAR gene, in which a histone methylatransferase SET domain is fused to an Hsmar1 transposase domain. A phylogenetic approach was taken to reconstruct the ancestral Hsmar1 transposase gene, which we named Hsmar1-Ra. The Hsmar1-Ra transposase efficiently mobilizes Hsmar1 transposons by a cut-and-paste mechanism in human cells and zebra fish embryos. Hsmar1-Ra can also mobilize short inverted-repeat transposable elements (MITEs) related to Hsmar1 (MiHsmar1), thereby establishing a functional relationship between an Hsmar1 transposase source and these MITEs. MiHsmar1 excision is 2 orders of magnitude more efficient than that of long elements, thus providing an explanation for their high copy numbers. We show that the SETMAR protein binds and introduces single-strand nicks into Hsmar1 inverted-repeat sequences in vitro. Pathway choices for DNA break repair were found to be characteristically different in response to transposon cleavage mediated by Hsmar1-Ra and SETMAR in vivo. Whereas nonhomologous end joining plays a dominant role in repairing excision sites generated by the Hsmar1-Ra transposase, DNA repair following cleavage by SETMAR predominantly follows a homology-dependent pathway. The novel transposon system can be a useful tool for genome manipulations in vertebrates and for investigations into the transpositional dynamics and the contributions of these elements to primate genome evolution.     

In vitro transposition of ISY100, a bacterial insertion sequence belonging to the Tc1/mariner family

The Synechocystis sp. PCC6803 insertion sequence ISY100 (ISTcSa) belongs to the Tc1/mariner/IS630 family of transposable elements. ISY100 transposase was purified and shown to promote transposition in vitro. Transposase binds specifically to ISY100 terminal inverted repeat sequences via an N-terminal DNA-binding domain containing two helix-turn-helix motifs. Transposase is the only protein required for excision and integration of ISY100. Transposase made double-strand breaks on a supercoiled DNA molecule containing a mini-ISY100 transposon, cleaving exactly at the transposon 3' ends and two nucleotides inside the 5' ends. Cleavage of short linear substrates containing a single transposon end was less precise. Transposase also catalysed strand transfer, covalently joining the transposon 3' end to the target DNA. When a donor plasmid carrying a mini-ISY100 was incubated with a target plasmid and transposase, the most common products were insertions of one transposon end into the target DNA, but insertions of both ends at a single target site could be recovered after transformation into Escherichia coli. Insertions were almost exclusively into TA dinucleotides, and the target TA was duplicated on insertion. Our results demonstrate that there are no fundamental differences between the transposition mechanisms of IS630 family elements in bacteria and Tc1/mariner elements in higher eukaryotes.     

Metnase/SETMAR: a domesticated primate transposase that enhances DNA repair,replication, and decatenation

Metnase is a fusion gene comprising a SET histone methyl transferase domain and a transposase domain derived from the Mariner transposase. This fusion gene appeared first in anthropoid primates. Because of its biochemical activities, both histone (protein) methylase and endonuclease, we termed the protein Metnase (also called SETMAR). Metnase methylates histone H3 lysine 36 (H3K36), improves the integration of foreign DNA, and enhances DNA double-strand break (DSB) repair by the non-homologous end joining (NHEJ) pathway, potentially dependent on its interaction with DNA Ligase IV. Metnase interacts with PCNA and enhances replication fork restart after stalling. Metnase also interacts with and stimulates TopoIIα-dependent chromosome decatenation and regulates cellular sensitivity to topoisomerase inhibitors used as cancer chemotherapeutics. Metnase has DNA nicking and endonuclease activity that linearizes but does not degrade supercoiled plasmids. Metnase has many but not all of the properties of a transposase, including Terminal Inverted Repeat (TIR) sequence-specific DNA binding, DNA looping, paired end complex formation, and cleavage of the 5′ end of a TIR, but it cannot efficiently complete transposition reactions. Interestingly, Metnase suppresses chromosomal translocations. It has been hypothesized that transposase activity would be deleterious in primates because unregulated DNA movement would predispose to malignancy. Metnase may have been selected for in primates because of its DNA repair and translocation suppression activities. Thus, its transposase activities may have been subverted to prevent deleterious DNA movement.     

Asymmetric processing of human immunodeficiency virus type 1 cDNA in vivo: implications for functional end coupling during the chemical steps of DNA transposition

Retroviral integration, like all forms of DNA transposition, proceeds through a series of DNA cutting and joining reactions. During transposition, the 3' ends of linear transposon or donor DNA are joined to the 5' phosphates of a double-stranded cut in target DNA. Single-end transposition must be avoided in vivo because such aberrant DNA products would be unstable and the transposon would therefore risk being lost from the cell. To avoid suicidal single-end integration, transposons link the activity of their transposase protein to the combined functionalities of both donor DNA ends. Although previous work suggested that this critical coupling between transposase activity and DNA ends occurred before the initial hydrolysis step of retroviral integration, work in the related Tn10 and V(D)J recombination systems had shown that end coupling regulated transposase activity after the initial hydrolysis step of DNA transposition. Here, we show that integrase efficiently hydrolyzed just the wild-type end of two different single-end mutants of human immunodeficiency virus type 1 in vivo, which, in contrast to previous results, proves that two functional DNA ends are not required to activate integrase's initial hydrolysis activity. Furthermore, despite containing bound protein at their processed DNA ends, these mutant viruses did not efficiently integrate their singly cleaved wild-type end into target DNA in vitro. By comparing our results to those of related DNA recombination systems, we propose the universal model that end coupling regulates transposase activity after the first chemical step of DNA transposition.     

Structural Basis for the Inverted Repeat Preferences of mariner Transposases

The inverted repeat (IR) sequences delimiting the left and right ends of many naturally active mariner DNA transposons are non-identical and have different affinities for their transposase. We have compared the preferences of two active mariner transposases, Mos1 and Mboumar-9, for their imperfect transposon IRs in each step of transposition: DNA binding, DNA cleavage, and DNA strand transfer. A 3.1 Å resolution crystal structure of the Mos1 paired-end complex containing the pre-cleaved left IR sequences reveals the molecular basis for the reduced affinity of the Mos1 transposase DNA-binding domain for the left IR as compared with the right IR. For both Mos1 and Mboumar-9, in vitro DNA transposition is most efficient when the preferred IR sequence is present at both transposon ends. We find that this is due to the higher efficiency of cleavage and strand transfer of the preferred transposon end. We show that the efficiency of Mboumar-9 transposition is improved almost 4-fold by changing the 3′ base of the preferred Mboumar-9 IR from guanine to adenine. This preference for adenine at the reactive 3′ end for both Mos1 and Mboumar-9 may be a general feature of mariner transposition.     

Structural role of the flanking DNA in mariner transposon excision

During cut-and-paste mariner/Tc1 transposition, transposon DNA is cut precisely at its junction with flanking DNA, ensuring the transposon is neither shortened nor lengthened with each transposition event. Each transposon end is flanked by a TpA dinucleotide: the signature target site duplication of mariner/Tc1 transposition. To establish the role of this sequence in accurate DNA cleavage, we have determined the crystal structure of a pre-second strand cleavage mariner Mos1 transpososome. The structure reveals the route of an intact DNA strand through the transposase active site before second strand cleavage. The crossed architecture of this pre-second strand cleavage paired-end complex supports our proposal that second strand cleavage occurs in trans. The conserved mariner transposase WVPHEL and YSPDL motifs position the strand for accurate DNA cleavage. Base-specific recognition of the flanking DNA by conserved amino acids is revealed, defining a new role for the WVPHEL motif in mariner transposition and providing a molecular explanation for in vitro mutagenesis data. Comparison of the pre-TS cleavage and post-cleavage Mos1 transpososomes with structures of Prototype Foamy Virus intasomes suggests a binding mode for target DNA prior to Mos1 transposon integration.     

Hairpin formation in Tn5 transposition

The initial chemical steps in Tn5 transposition result in blunt end cleavage of the transposon from the donor DNA. We demonstrate that this cleavage occurs via a hairpin intermediate. The first step is a 3' hydrolytic nick by transposase. The free 3'OH then attacks the phosphodiester bond on the opposite strand, forming a hairpin at the transposon end. In addition to forming precise hairpins, Tn5 transposase can form imprecise hairpins. This is the first example of imprecise hairpin formation on transposon end DNA. To undergo strand transfer, the hairpin must to be resolved by a transposase-catalyzed hydrolytic cleavage. We show that both precise and imprecise hairpins are opened by transposase. A transposition mechanism utilizing a hairpin intermediate allows a single transposase active site to cleave both 3' and 5' strands without massive protein/DNA rearrangements.     

Human Pso4 is a metnase (SETMAR)-binding partner that regulates metnase function in DNA repair

Metnase, also known as SETMAR, is a SET and transposase fusion protein with an undefined role in mammalian DNA repair. The SET domain is responsible for histone lysine methyltransferase activity at histone 3 K4 and K36, whereas the transposase domain possesses 5'-terminal inverted repeat (TIR)-specific DNA binding, DNA looping, and DNA cleavage activities. Although the transposase domain is essential for Metnase function in DNA repair, it is not clear how a protein with sequence-specific DNA binding activity plays a role in DNA repair. Here, we show that human homolog of the ScPSO4/PRP19 (hPso4) forms a stable complex with Metnase on both TIR and non-TIR DNA. The transposase domain essential for Metnase-TIR interaction is not sufficient for its interaction with non-TIR DNA in the presence of hPso4. In vivo, hPso4 is induced and co-localized with Metnase following ionizing radiation treatment. Cells treated with hPso4-siRNA failed to show Metnase localization at DSB sites and Metnase-mediated stimulation of DNA end joining coupled to genomic integration, suggesting that hPso4 is necessary to bring Metnase to the DSB sites for its function(s) in DNA repair.     

The DDN Catalytic Motif Is Required for Metnase Functions in Non-homologous End Joining (NHEJ) Repair and Replication Restart

Metnase (or SETMAR) arose from a chimeric fusion of the Hsmar1 transposase downstream of a protein methylase in anthropoid primates. Although the Metnase transposase domain has been largely conserved, its catalytic motif (DDN) differs from the DDD motif of related transposases, which may be important for its role as a DNA repair factor and its enzymatic activities. Here, we show that substitution of DDN⁶¹⁰ with either DDD⁶¹⁰ or DDE⁶¹⁰ significantly reduced in vivo functions of Metnase in NHEJ repair and accelerated restart of replication forks. We next tested whether the DDD or DDE mutants cleave single-strand extensions and flaps in partial duplex DNA and pseudo-Tyr structures that mimic stalled replication forks. Neither substrate is cleaved by the DDD or DDE mutant, under the conditions where wild-type Metnase effectively cleaves ssDNA overhangs. We then characterized the ssDNA-binding activity of the Metnase transposase domain and found that the catalytic domain binds ssDNA but not dsDNA, whereas dsDNA binding activity resides in the helix-turn-helix DNA binding domain. Substitution of Asn-610 with either Asp or Glu within the transposase domain significantly reduces ssDNA binding activity. Collectively, our results suggest that a single mutation DDN⁶¹⁰ → DDD⁶¹⁰, which restores the ancestral catalytic site, results in loss of function in Metnase.     

Mechanism of IS200/IS605 family DNA transposases: activation and transposon-directed target site selection

The smallest known DNA transposases are those from the IS200/IS605 family. Here we show how the interplay of protein and DNA activates TnpA, the Helicobacter pylori IS608 transposase, for catalysis. First, transposon end binding causes a conformational change that aligns catalytically important protein residues within the active site. Subsequent precise cleavage at the left and right ends, the steps that liberate the transposon from its donor site, does not involve a site-specific DNA-binding domain. Rather, cleavage site recognition occurs by complementary base pairing with a TnpA-bound subterminal transposon DNA segment. Thus, the enzyme active site is constructed from elements of both protein and DNA, reminiscent of the interdependence of protein and RNA in the ribosome. Our structural results explain why the transposon ends are asymmetric and how the transposon selects a target site for integration, and they allow us to propose a molecular model for the entire transposition reaction.     

Intramolecular transposition by Tn10

Transposon Tn10 promotes the formation of a circular product containing only transposon sequences. We show that these circles result from an intramolecular transposition reaction in which all of the strand cleavage and ligation events have occurred but newly created transposon/target junctions have not undergone repair. The unligated strand termini at these junctions are those expected according to a simple model in which the target DNA is cleaved by a pair of staggered nicks 9 bp apart, transposon sequences are separated from flanking donor DNA by cleavage at the terminal nucleotides on both strands (at both ends) of the element, and 3' transposon strand ends are ligated to 5' target strand ends. The stability of the unligated junctions suggests that they are protected from cellular processing by transposase and/or host proteins. We propose that the nonreplicative nature of Tn10 transposition is determined by the efficiency with which the nontransferred transposon strand is separated from flanking donor DNA and by the nature of the protein-DNA complexes present at the strand transfer junctions.     

Transposition of the human Hsmar1 transposon: rate-limiting steps and the importance of the flanking TA dinucleotide in second strand cleavage

Hsmar1 is a member of the mariner family of DNA transposons. Although widespread in nature, their molecular mechanism remains obscure. Many other cut-and-paste elements use a hairpin intermediate to cleave the two strands of DNA at each transposon end. However, this intermediate is absent in mariner, suggesting that these elements use a fundamentally different mechanism for second-strand cleavage. We have taken advantage of the faithful and efficient in vitro reaction provided by Hsmar1 to characterize the products and intermediates of transposition. We report different factors that particularly affect the reaction, which are the reaction pH and the transposase concentration. Kinetic analysis revealed that first-strand nicking and integration are rapid. The rate of the reaction is limited in part by the divalent metal ion-dependent assembly of a complex between transposase and the transposon end(s) prior to the first catalytic step. Second-strand cleavage is the rate-limiting catalytic step of the reaction. We discuss our data in light of a model for the two metal ion catalytic mechanism and propose that mariner excision involves a significant conformational change between first- and second-strand cleavage at each transposon end. Furthermore, this conformational change requires specific contacts between transposase and the flanking TA dinucleotide.     

Early intermediates of mariner transposition: catalysis without synapsis of the transposon ends suggests a novel architecture of the synaptic complex

The mariner family is probably the most widely distributed family of transposons in nature. Although these transposons are related to the well-studied bacterial insertion elements, there is evidence for major differences in their reaction mechanisms. We report the identification and characterization of complexes that contain the Himar1 transposase bound to a single transposon end. Titrations and mixing experiments with the native transposase and transposase fusions suggested that they contain different numbers of transposase monomers. However, the DNA protection footprints of the two most abundant single-end complexes are identical. This indicates that some transposase monomers may be bound to the transposon end solely by protein-protein interactions. This would mean that the Himar1 transposase can dimerize independently of the second transposon end and that the architecture of the synaptic complex has more in common with V(D)J recombination than with bacterial insertion elements. Like V(D)J recombination and in contrast to the case for bacterial elements, Himar1 catalysis does not appear to depend on synapsis of the transposon ends, and the single-end complexes are active for nicking and probably for cleavage. We discuss the role of this single-end activity in generating the mutations that inactivate the vast majority of mariner elements in eukaryotes.     

Mariner Mos1 transposase dimerizes prior to ITR binding

The mariner Mos1 synaptic complex consists of a tetramer of transposase molecules that bring together the two ends of the element. Such an assembly requires at least two kinds of protein-protein interfaces. The first is involved in "cis" dimerization, and consists of transposase molecules bound side-by-side on a single DNA molecule. The second, which is involved in "trans" dimerization, consists of transposase molecules bound to two different DNA molecules. Here, we used biochemical and genetic methods to enhance the definition of the regions involved in cis and trans-dimerization in the mariner Mos1 transposase. The cis and trans-dimerization interfaces were both found within the first 143 amino acid residues of the protein. The cis-dimerization activity was mainly contained in amino acids 1-20. The region spanning from amino acid residues 116-143, and containing the WVPHEL motif, was involved in the cis- to trans-shift as well as in trans-dimerization stabilization. Although the transposase exists mainly as a monomer in solution, we provide evidence that the transposase cis-dimer is the active species in inverted terminal repeat (ITR) binding. We also observed that the catalytic domain of the mariner Mos1 transposase modulates efficient transposase-transposase interactions in the absence of the transposon ends.     

Solution conformations of early intermediates in Mos1 transposition

DNA transposases facilitate genome rearrangements by moving DNA transposons around and between genomes by a cut-and-paste mechanism. DNA transposition proceeds in an ordered series of nucleoprotein complexes that coordinate pairing and cleavage of the transposon ends and integration of the cleaved ends at a new genomic site. Transposition is initiated by transposase recognition of the inverted repeat sequences marking each transposon end. Using a combination of solution scattering and biochemical techniques, we have determined the solution conformations and stoichiometries of DNA-free Mos1 transposase and of the transposase bound to a single transposon end. We show that Mos1 transposase is an elongated homodimer in the absence of DNA and that the N-terminal 55 residues, containing the first helix-turn-helix motif, are required for dimerization. This arrangement is remarkably different from the compact, crossed architecture of the dimer in the Mos1 paired-end complex (PEC). The transposase remains elongated when bound to a single-transposon end in a pre-cleavage complex, and the DNA is bound predominantly to one transposase monomer. We propose that a conformational change in the single-end complex, involving rotation of one half of the transposase along with binding of a second transposon end, could facilitate PEC assembly.     

The Folding of the Specific DNA Recognition Subdomain of the Sleeping Beauty Transposase Is Temperature-Dependent and Is Required for Its Binding to the Transposon DNA

The reaction of DNA transposition begins when the transposase enzyme binds to the transposon DNA. Sleeping Beauty is a member of the mariner family of DNA transposons. Although it is an important tool in genetic applications and has been adapted for human gene therapy, its molecular mechanism remains obscure. Here, we show that only the folded conformation of the specific DNA recognition subdomain of the Sleeping Beauty transposase, the PAI subdomain, binds to the transposon DNA. Furthermore, we show that the PAI subdomain is well folded at low temperatures, but the presence of unfolded conformation gradually increases at temperatures above 15°C, suggesting that the choice of temperature may be important for the optimal transposase activity. Overall, the results provide a molecular-level insight into the DNA recognition by the Sleeping Beauty transposase.     

Alternative conformations of a nucleic acid four-way junction

Sleeping Beauty (SB), a member of the Tc1/mariner superfamily of transposable elements, is the only active DNA-based transposon system of vertebrate origin that is available for experimental manipulation. We have been using the SB element as a research tool to investigate some of the cis and trans-requirements of element mobilization, and mechanisms that regulate transposition in vertebrate species. In contrast to mariner transposons, which are regulated by overexpression inhibition, the frequency of SB transposition was found to be roughly proportional to the amount of transposase present in cells. Unlike Tc1 and mariner elements, SB contains two binding sites within each of its terminal inverted repeats, and we found that the presence of both of these sites is a strict requirement for mobilization. In addition to the size of the transposon itself, the length as well as sequence of the DNA outside the transposon have significant effects on transposition. As a general rule, the closer the transposon ends are, the more efficient transposition is from a donor molecule. We have found that SB can transform a wide range of vertebrate cells from fish to human. However, the efficiency and precision of transposition varied significantly among cell lines, suggesting potential involvement of host factors in SB transposition. A positive-negative selection assay was devised to enrich populations of cells harboring inserted transposons in their chromosomes. Using this assay, of the order of 10,000 independent transposon insertions can be generated in human cells in a single transfection experiment. Sleeping Beauty can be a powerful alternative to other vectors that are currently used for the production of transgenic animals and for human gene therapy.     

Two repeated motifs enriched within some enhancers and origins of replication are bound by SETMAR isoforms in human colon cells

Setmar is a gene specific to simian genomes. The function(s) of its isoforms are poorly understood and their existence in healthy tissues remains to be validated. Here we profiled SETMAR expression and its genome-wide binding landscape in colon tissue. We found isoforms V3 and V6 in healthy and tumour colon tissues as well as incell lines. In two colorectal cell lines SETMAR binds to several thousand Hsmar1 and MADE1 terminal ends, transposons mostly located in non-genic regions of active chromatin including in enhancers. It also binds to a 12-bp motifs similar to an inner motif in Hsmar1 and MADE1 terminal ends. This motif is interspersed throughout the genome and is enriched in GC-rich regions as well as in CpG islands that contain constitutive replication origins. It is also found in enhancers other than those associated with Hsmar1 and MADE1. The role of SETMAR in the expression of genes, DNA replication and in DNA repair are discussed.