Molecular architecture of the Mos1 paired-end complex: the structural basis of DNA transposition in a eukaryote

Abstract

A key step in cut-and-paste DNA transposition is the pairing of transposon ends before the element is excised and inserted at a new site in its host genome. Crystallographic analyses of the paired-end complex (PEC) formed from precleaved transposon ends and the transposase of the eukaryotic element Mos1 reveals two parallel ends bound to a dimeric enzyme. The complex has a trans arrangement, with each transposon end recognized by the DNA binding region of one transposase monomer and by the active site of the other monomer. Two additional DNA duplexes in the crystal indicate likely binding sites for flanking DNA. Biochemical data provide support for a model of the target capture complex and identify Arg186 to be critical for target binding. Mixing experiments indicate that a transposase dimer initiates first-strand cleavage and suggest a pathway for PEC formation.

Figure 1. Mechanisms of DNA Transposition

A Mos1 transposition pathway: the 1.3kB Mos1 transposon (light blue) has 28 bp imperfect IRs at both ends (orange triangles) and encodes a transposase (blue circle), the sole requirement for Mos1 transposition. Transposase binds to a single end as a monomer (SEC1) or a dimer (SEC2). The ends are brought together to form a paired-end complex (PEC) and the transposon is excised from flanking DNA. Subsequently target DNA binds, forming the target capture complex (TCC), and the transposon integrates at a TA sequence. B DNA excision proceeds via a hairpin intermediate in transposition of some prokaryotic (Tn5) and eukaryotic (Hermes) elements and flipped bases (shown as thick lines) facilitate hairpin formation (Ason and Reznikoff, 2002; Grundy et al., 2007). By contrast Tc1/mariner transposons (e.g. Mos1 and Sleeping Beauty) do not form DNA hairpin intermediates and both strands are presumed to be cleaved by hydrolysis. The 3′OH on the TS attacks target DNA in the strand transfer reaction. Tn5 inserts into a 9 bp consensus sequence, Hermes an 8bp sequence and Tc1/mariner elements always integrate into TA di-nucleotides. In the two metal ion mechanism it is proposed that the roles of the two active site metal ions swap in successive reactions, so that the metal which stabilizes the substrate in one step activates the nucleophile in the next step, and vice versa. C The DNA duplex used for crystallisation has the sequence of the right Mos1 IR after DNA excision, with a 3 nucleotide protruding 3′ end.

Figure 2. Architecture of the Mos1 PEC

A and B: orthogonal views of the PEC crystal structure. Transposase monomer A is coloured orange and monomer B blue. The two major-groove DNA-binding motifs contain HTH1 (residues 24 to 55) and HTH2 (residues 89 to 110). The minor-groove binding motif comprises residues 63 to 71. The two DNA duplexes bound by the DNA-binding domains are labelled IR DNA and the two extra DNA duplexes are labelled FL DNA. C Schematic diagram of the structure. An arrow indicates the 3′ end of each DNA strand and a black dot indicates the 5′ phosphate of the NTS. The purple sphere indicates the metal ion in active site A.

Figure 3. Protein-DNA and protein-protein interfaces

A Cis Protein-DNA interactions between transposase monomer A and IR DNA_A. Transposase is coloured orange and shown in ribbon representation. The TS and NTS of IR DNA_A are numbered and coloured red and beige respectively. The side-chains of key residues involved in DNA interactions are labelled and shown as sticks. B Interactions between the linker of monomer A (orange) and the clamp loop of monomer B (blue), with the short β-strands (β1-4) labelled. There are also symmetry related interactions between the linker of monomer B and the clamp loop of monomer A (not shown). The staggered ends of IR DNA are lodged on the α11 helices of the catalytic domains and key residues involved in trans protein-DNA interactions are labelled. C Dimerisation of the two HTH1 motifs is mediated by hydrophobic contacts (green dotted lines) between α-helices 1 and 2 of each transposase monomer. Residues involved in this interface are labelled and shown as sticks.

Figure 4. The Active Sites

A The active site of monomer A has one metal ion bound in site 1: the position of the Mg²⁺ ion is shown as a purple sphere. B Active site of transposase monomer B does not contain a metal ion. The 2Fo-Fc electron density map (grey mesh) is contoured at 1.7σ. C Key interactions between residues in the linker, the clamp loop and the single-stranded bases at the 3′ end of the TS. A Mn anomalous difference density map (pink mesh, contoured at 3.5σ) confirms the position of the single Mn²⁺ ion in active site A in the Mn²⁺ bound PEC structure (pink sphere). The position of the second (unoccupied) metal binding site (Richardson et al., 2006) is shown as a grey sphere. D Mutation of residues R118 and W119 in the linker region diminished transposase activity in second strand cleavage assays. Cleavage of the fluorescein labelled 100nt TS was detected by observation of the 70nt reaction product on an 8% denaturing polyacrylamide gel. Lanes 6 and 7 contained fluorescein labelled DNA markers of 70 and 67nts respectively.

Figure 5. Target DNA Binding Model

A Proposed target DNA (black) binding site. The TS terminal 3′OHs are marked. B View from the underside of the PEC, with the target TA sequence (purple) and the TS 3′OHs highlighted. The 2-fold axis of symmetry is marked (red dot). C Strand transfer assay. The 50mer target DNA substrates contained one TA dinucleotide, 10 and 11 bases from one end, and a 3′ fluorescein tag on either the top or the bottom strand. Integration of the 28mer TS of IRR DNA yields a 68mer or 40mer labelled product depending on whether the top or bottom strand of target DNA was labelled. D Dissolved crystals integrate into target DNA (lanes 1-3 and 6-8) to give identical integration products to T216A transposase plus IRR DNA in solution (lanes 5 and 10). Negative controls: without crystals (lanes 4 and 9) or with T216A but without IRR DNA (lane 11). The band labelled # corresponds to an alternative secondary structure of the target DNA substrate which was not fully denatured on this gel. E. View of the target TA sequence (purple) from the topside of the PEC, with the positions of R186 and R183 in each monomer highlighted. F Target integration assay of double mutant transposases (lanes 3-10) compared with the activity of T216A transposase (lanes 2 and 11). Lanes 1 and 12 correspond to reaction without transposase and lanes 13 and 14 are 67nt and 66nt DNA markers, respectively. Percentage strand transfer is indicated at the bottom of each lane.

Figure 6. First strand cleavage assays

A Schematic of the possible trans or cis arrangements of transposase heterodimers in SEC2 complexes and their predicted cleavage activity. B Cleavage assay of T216A transposase and R48Q/Q100R and D284A mutants at 60nM, 30nM and 15nM. C Graph of percentage cleavage of 1:1 mixtures of mutant transposases in assays where the total protein concentration was constant at 30nM. Data are represented as mean +/-SD. D Mixed mutant cleavage assay in which mutants were titrated into fixed concentrations of T216A transposase (30nM or 5nM). The percentage cleavage activity is shown at the bottom of each lane.