Transposition of bacteriophage Mu uses two DNA cleavage sites and six transposase recognition sites, with each recognition site divided into two half-sites. The recognition sites can activate transposition of non-Mu DNA sequences if a complete set of Mu sequences is not available. We have analyzed 18 sequences from a non-Mu DNA molecule, selected in a functional assay for the ability to be transposed by MuA transposase. These sequences are remarkably diverse. Nonetheless, when viewed as a group they resemble a Mu DNA end, with a cleavage site and a single recognition site. Analysis of these "pseudo-Mu ends" indicates that most positions in the cleavage and recognition sites contribute sequence-specific information that helps drive transposition, though only the strongest contributors are apparent from mutagenesis data. The sequence analysis also suggests variability in the alignment of recognition half-sites. Transposition assays of specifically designed DNA substrates support the conclusion that the transposition machinery is flexible enough to permit variability in half-site spacing and also perhaps variability in the placement of the recognition site with respect to the cleavage site. This variability causes only local perturbations in the protein-DNA complex, as indicated by experiments in which altered and unaltered DNA substrates are paired.Some of the most fundamental of biological processes occur within nucleoprotein complexes. These processes include recombination, replication, transcription, and RNA splicing. The nucleoprotein complexes are often large and elaborate, containing features that permit sophisticated regulation. As a result, dissecting the chemistry of interactions within a complex is a challenging task, but one that is critical to understanding biological function.Transposition of bacteriophage Mu, like that of other transposons, occurs within complexes called transpososomes. Transpososomes mediate at least two sequential chemical reactions, on a pathway toward transferring the transposon DNA from one site to another. The two reactions are: (i) DNA cleavage, in which a nick is introduced precisely at the end of the transposon, on the 3Ј strand and (ii) DNA strand transfer, in which the nicked strand is joined to a separate DNA molecule called the target (1, 2). The reaction sites on the transposon DNA, or "donor DNA," are defined by specific DNA sequences, but Mu target sites are not very sequence-specific (3).Transpososomes contain multiple subunits of a transposase protein, bound to DNA sequences from both of the transposon's ends (1). These protein-DNA complexes are also called "synaptic complexes" because they bring together the two ends of the transposon DNA. The phage Mu transposase, MuA, is monomeric in solution but forms a tetramer upon binding to specific DNA recognition sites near the transposon ends (4 -6).Each end of the Mu transposon has three MuA recognition sites: R1, R2, and R3 on the right end and L1, L2, and L3 on the left, not all of which are essential for transposition...

Members of the transposase͞retroviral-integrase superfamily use a single active site to perform at least two reactions during transposition of a DNA transposon or a retroviral cDNA. They hydrolyze a DNA sequence at the end of the mobile DNA and then join this DNA end to a target DNA (a reaction called DNA strand transfer). Critical to understanding the mechanism of recombination is elucidating how these distinct reactions are orchestrated by the same active site. Here we find that DNA substrates terminating in a dideoxynucleotide allow Mu transposase to hydrolyze a target DNA, combining aspects of both natural reactions. Analyses of the sequence preferences for target hydrolysis and of the structure of the cleaved product indicate that this reaction is promoted by the active site in the conformation that normally promotes DNA strand transfer. Dissecting the DNA requirements for target hydrolysis reveals that the ribose of the last nucleotide of the Mu DNA activates transposase's catalytic potential, even when this residue is not a direct chemical participant. These findings provide insight into the molecular mechanism insuring that DNA strand transfer ordinarily occurs rather than inappropriate DNA cleavage. The required presence of the terminal nucleotide in the transposase active site creates a great advantage for the attached 3OH to serve as nucleophile.

Confronted with thousands of potential DNA substrates, a site-specific enzyme must restrict itself to the correct DNA sequence. The MuA transposase protein performs site-specific DNA cleavage and joining reactions, resulting in DNA transposition-a specialized form of genetic recombination. To determine how sequence information is used to restrict transposition to the proper DNA sites, we performed kinetic analyses of transposition with DNA substrates containing either wild-type transposon sequences or sequences carrying mutations in specific DNA recognition modules. As expected, mutations near the DNA cleavage site reduce the rate of cleavage; the observed effect is about 10-fold. In contrast, mutations within the MuA recognition sequences do not directly affect the DNA cleavage or joining steps of transposition. It is well established that the recognition sequences are necessary for assembly of stable, multimeric MuA-DNA complexes, and we find that recognition site mutations severely reduce both the extent and the rate of this assembly process. Yet if the MuA-DNA complexes are preassembled, the first-order rate constants for both DNA cleavage and DNA strand transfer (the joining reaction) are unaffected by the mutations. Furthermore, most of the mutant DNA molecules that are cleaved also complete DNA strand transfer. We conclude that the sequence-specific contacts within the recognition sites contribute energetically to complex assembly, but not directly to catalysis. These results contrast with studies of more orthodox enzymes, such as EcoRI and some other type II restriction enzymes. We propose that the strategy employed by MuA may serve as an example for how recombinases and modular restriction enzymes solve the DNA specificity problem, in that they, too, may separate substrate recognition from catalysis.