The mechanism of loading of the FLP recombinase onto its DNA target sequence

The FLP recombinase interacts with its target sequence with the formation of three distinct DNA-protein complexes. The first complex leaves neither a DNase footprint nor is the DNA protected from methylation by dimethyl sulfate. We have found, however, that the FLP protein is bound predominantly to only one of the three 13 base-pair (bp) symmetry elements. This asymmetric loading of the FLP site seems to require the presence of an adjacent directly repeated 13 bp element. We speculate that this asymmetric filling of the target site may be accompanied by the unique order of cleavage and exchange of DNA strands.     

Specific contacts between the FLP protein of the yeast 2-micron plasmid and its recombination site

Contact points between the FLP protein of the yeast 2-micron plasmid and its recombination site have been defined. Important features of the region previously defined as the minimal recombination site in vitro include a pair of 13-base pair inverted repeats separated by an 8-base pair spacer. The two FLP protein-binding sites within this region are 12 base pairs in length. In each case they include the internal 11 base pairs of one of the 13-base pair repeats, as well as the adjacent base pair within the spacer. The internal 6 base pairs within the spacer are not involved in binding or recognition by FLP protein. When the size of the spacer is increased or decreased by one base pair, the distance between the cleavage points is also increased or decreased correspondingly by one base pair. Points of cleavage are unaffected by changes in the spacer sequence. Specific contact points involving purine residues, identified by methylation protection and recombination interference experiments, are located in both the major and minor grooves of the DNA. Additional contact points between FLP protein and phosphate groups in the phosphate-deoxyribose backbone are clustered near the cleavage sites.     

Interaction of the FLP recombinase of the Saccharomyces cerevisiae 2 micron plasmid with mutated target sequences.

The 2 micron plasmid of Saccharomyces cerevisiae codes for a site-specific recombinase, the FLP protein, that catalyzes efficient recombination across two 599-base-pair (bp) inverted repeats of the plasmid DNA both in vivo and in vitro. We analyzed the interaction of the purified FLP protein with the target sequences of two point mutants that exhibit impaired FLP-mediated recombination in vivo. One mutation lies in one of the 13-bp repeat elements that had been previously shown to be protected from DNase digestion by the FLP protein. This mutation dramatically reduces FLP-mediated recombination in vitro and appears to act by reducing the binding of FLP protein to its target sequence. The second mutation lies within the 8-bp core region of the FLP target sequence. The FLP protein introduces staggered nicks surrounding this 8-bp region, and these nicks are thought to define the sites of strand exchange. The mutation in the core region abolishes recombination with a wild-type site. However, recombination between two mutated sites is very efficient. This result suggests that proper base pairing between the two recombining sites is an important feature of FLP-mediated recombination.     

The minimal duplex DNA sequence required for site-specific recombination promoted by the FLP protein of yeast in vitro.

The 2-micron plasmid of the yeast Saccharomyces cerevisiae codes for a site-specific recombinase ('FLP') that efficiently catalyses recombination across the plasmid's two 599 bp repeats both in vivo and in vitro. We have used the partially purified FLP protein to define the minimal duplex DNA sequence required for intra- and intermolecular recombination in vitro. Previous DNase footprinting experiments had shown that FLP protected 50 bp of DNA around the recombination site. We made BAL31 deletions and synthetic FLP sites to show that the minimal length of the site that was able to recombine with a wild-type site was 22 bp. The site consists of two 7 bp inverted repeats surrounding an 8 bp core region. We also showed that the deleted sites recombined with themselves and that one of three 13 bp repeated elements within the FLP target sequence was not necessary for efficient recombination in vitro. Mutants lacking this redundant 13 bp element required a lower amount of FLP recombinase to achieve maximal yield of recombination than the wild type site. Finally, we discuss the structure of the FLP site in relation to the proposed function of FLP recombination in copy number amplification of the 2-micron plasmid in vivo.     

Determination of DNA sequences essential for FLP-mediated recombination by a novel method

The yeast 2-micron circle plasmid encodes a protein, FLP, that mediates site-specific recombination across the two FLP-binding sites of the plasmid. We have used a novel technique, "exonuclease-treated substrate analysis," to determine the minimal duplex DNA sequence needed for this recombination event. A linear DNA containing two FLP sites in a direct orientation was treated with the double-strand specific 3'-exonuclease, exonuclease III, to generate molecules with a nested set of single-strand deletions that extended into one of the FLP sites. The DNA was then end-labeled at the sites of the deletions and used as a substrate for recombination in vitro. FLP-mediated recombination between two FLP sites excised a restriction endonuclease cleavage site from the DNA. Comparison of the fragments produced by restriction enzyme digestion of untreated and FLP-treated DNA showed to the nucleotide the duplex DNA sequence required for FLP-mediated recombination. To examine essential sequences in the opposite DNA strand, similar experiments were done using the 5'-exonuclease encoded by phage T7. The minimal essential duplex DNA sequence lies within the region of the FLP site that was previously shown to be protected from nuclease digestion in the presence of FLP. A modified form of this technique can be used to study the minimal sequence requirements of site-specific DNA binding proteins.     

Site-specific interaction of vaccinia virus topoisomerase I with duplex DNA. Minimal DNA substrate for strand cleavage in vitro.

Purified vaccinia virus DNA topoisomerase I forms a cleavable complex with duplex DNA at a conserved sequence element 5'(C/T)CCTTdecreases in the incised DNA strand. DNase I footprint studies show that vaccinia topoisomerase protects the region around the site of covalent adduct formation from nuclease digestion. On the cleaved DNA strand, the protected region extends from +13 to -13 (+1 being the site of cleavage). On the noncleaved strand, the protected region extends from +13 to -9. Similar nuclease protection is observed for a mutant topoisomerase (containing a Tyr ---- Phe substitution at the active site amino acid 274) that is catalytically inert and does not form the covalent intermediate. Thus, vaccinia topoisomerase is a specific DNA binding protein independent of its competence in transesterification. By studying the cleavage of a series of 12-mer DNA duplexes in which the position of the CCCTTdecreases motif within the substrate is systematically phased, the "minimal" substrate for cleavage has been defined; cleavage requires six nucleotides upstream of the cleavage site and two nucleotides downstream of the site. An analysis of the cleavage of oligomer substrates mutated singly in the CCCTT sequence reveals a hierarchy of mutational effects based on position within the pentamer motif and the nature of the sequence alteration.     

FLP protein of 2 mu circle plasmid of yeast induces multiple bends in the FLP recognition target site

The FLP recombinase of the 2 mu plasmid of Saccharomyces cerevisiae binds to a target containing three 13 base-pair symmetry elements called a, b and c. The symmetry elements b and c are in direct orientation while the a element is in inverted orientation with respect to b and c on the opposite side of an eight base-pair core region. Each symmetry element acts as a binding site for the FLP protein. The FLP protein can form three different complexes with the FLP recognition target (FRT site) according to the number of elements within the site that are occupied by the FLP protein. Binding of FLP to the FRT site induces DNA bending. We have measured the angles of bends caused by the binding of the FLP protein to full and partial FRT sites. We find that FLP induces three types of bend in the FRT-containing DNA. The type I bend is approximately 60 degrees and results from a molecule of FLP bound to one symmetry element. The type II bend is greater than 144 degrees and results from FLP molecules bound to symmetry elements a and b. The type III bend is approximately 65 degrees and results from FLP proteins bound to symmetry elements b and c. Certain FLP proteins that are defective in recombination can generate the type I and type III bends but are impaired in their ability to induce the type II bend. We discuss the role of bending in FLP-mediated recombination.     

The role of the loxP spacer region in P1 site-specific recombination.

The lox-Cre site-specific recombination system of bacteriophage P1 is comprised of a site on the DNA where recombination occurs called loxP, and a protein, Cre, which mediates the reaction. The loxP site is 34 base pairs (bp) in length and consists of two 13 bp inverted repeats separated by an 8 bp spacer region. Previously it has been shown that the cleavage and strand exchange of recombining loxP sites occurs within this spacer region. We report here an analysis of various base substitution mutations within the spacer region of loxP, and conclude the following: Homology is a requirement for efficient recombination between recombining loxP sites. There is at least one position within the spacer where a base change drastically reduces recombination even when there is homology between the two recombining loxP sites. When two loxP sites containing symmetric spacer regions undergo Cre-mediated recombination in vitro, the DNA between the sites undergoes both excision and inversion with equal frequency.     

The functional significance of DNA sequence structure in a site-specific genetic recombination reaction.

A series of sequence changes in the spacer region of the FLP recombination target (FRT) site are presented which drastically reduce site function without affecting recognition by the FLP protein. The effects follow a pattern which indicates that two structural features of the FRT site are essential for site function: a pair of pyrimidine tracts arranged in a palindrome and a predominance of AT base pairs in the spacer. The FRT site represents a sequence that serves to facilitate unwinding of the DNA within the spacer region during recombination. The results provide a clear demonstration of a role for a DNA sequence element that is distinct from protein recognition.     

An altered DNA conformation in origin region I is a determinant for the binding of SV40 large T antigen

Seventeen base pairs of DNA from SV40 origin region I encode a tripartite binding site for a dimeric mass of SV40 large T antigen. Two binding components are the directly repeated pentanucleotide sequences 5'-GAGGC-3'/5'-GCCTC-3'. The third component is the asymmetric sequence 5'-TTTTTTG-3'/5'-CAAAAAA-3' that separates the pentanucleotides. Nucleotide-specific features of this spacer element stabilize binding to the adjacent pentanucleotides. We report here that the spacer sequence determines a DNA conformation that correlates with high affinity binding of T antigen. The nature of the spacer sequence suggests that the DNA is bent. We propose that binding of T antigen to region I proceeds through monomer-pentanucleotide interactions and either protein-protein or protein-spacer interactions directed by the spacer-encoded structure.     

Stimulation of V(D)J cleavage by high mobility group proteins.

V(D)J recombination requires a pair of signal sequences with spacer lengths of 12 and 23 bp between the conserved heptamer and nonamer elements. The RAG1 and RAG2 proteins initiate the reaction by making double-strand DNA breaks at both signals, and must thus be able to operate on these two different spatial arrangements. We show that the DNA-bending proteins HMG1 and HMG2 stimulate cleavage and RAG protein binding at the 23 bp spacer signal. These findings suggest that DNA bending is important for bridging the longer spacer, and explain how a similar array of RAG proteins could accommodate a signal with either a 12 or a 23 bp spacer. An additional effect of HMG proteins is to stimulate coupled cleavage greatly when both signal sequences are present, suggesting that these proteins also aid the formation of a synaptic complex.     

Chemical probe and missing nucleoside analysis of Flp recombinase bound to the recombination target sequence.

The Flp protein catalyzes a site-specific recombination reaction between two 47 bp DNA sites without the assistance of any other protein or cofactor. The Flp recognition target (FRT) site consists of three nearly identical sequences, two of which are separated by an 8 bp spacer sequence. In order to gain insight into this remarkable protein-DNA interaction we used a variety of chemical probe methods and the missing nucleoside experiment to examine Flp binding. Hydroxyl radical footprints of Flp bound to a recombinationally-competent site fall on opposite faces of canonical B-DNA. The 8 bp spacer region between the two Flp binding sites becomes reactive towards 5-phenyl-1,10-phenanthroline.copper upon Flp binding, indicating that once bound by Flp, this segment of DNA is not in the B-form. Missing nucleoside analysis reveals that within each binding site the presence of two nucleosides on the top strand and four on the bottom, are required for formation of a fully-occupied FRT site. In contrast, loss of any nucleoside in the three binding sites in the FRT interferes with formation of lower-occupancy complexes. DNA molecules with gaps in the 8 bp spacer region are over-represented in complexes with either two or three binding sites occupied by Flp, evidence that DNA flexibility facilitates the cooperative interaction of Flp protomers bound to a recombinationally-active site.     

Purification and characterization of the Ner repressor of bacteriophage Mu

The Ner protein of bacteriophage Mu acts as a lambda cro-like negative regulator of the phage's early (transposase) operon. Using the band retardation assay to monitor ner-operator-specific DNA-binding activity, the 8 kDa Ner protein was purified to homogeneity. DNase I footprinting revealed that the purified protein bound and protected a specific DNA operator that contains two 12 bp sites with the consensus sequence 5'-ANPyTAPuCTAAGT-3', separated by a 6 bp spacer region. Moreover, regions corresponding to a turn of the DNA helix flanking these 12 bp repeats are also protected by Ner. Unlike the functionally similar lambda cro protein, gel filtration experiments show that the native molecular mass of Mu Ner to be approx. 8 kDa. These results, plus the pattern of DNase I protection, suggest that the protein may bind as a monomer to each of its specific DNA substrates.     

Evidence for a conformational change in the DNA gyrase-DNA complex from hydroxyl radical footprinting.

We have used the technique of hydroxyl radical footprinting to probe the complex between DNA gyrase and a 198 bp DNA fragment containing the preferred gyrase cleavage site from plasmid pBR322. We find that gyrase protects 128 bp from the hydroxyl radical with the central 13 bp (adjacent to the gyrase cleavage site) being most strongly protected. Flanking the central region are arms showing periodic protection from the reagent suggesting a helical repeat of 10.6 bp, consistent with the DNA being wrapped upon the enzyme surface. The presence of 5'-adenylyl-beta,gamma-imidodiphosphate or a quinolone drug causes alteration of the protection pattern consistent with a conformational change in the complex involving one arm of the wrapped DNA. The significance of these results for the mechanism of DNA supercoiling by gyrase is discussed.     

DNA binding specificity of the bovine papillomavirus E1 protein is determined by sequences contained within an 18-base-pair inverted repeat element at the origin of replication.

Bovine papillomavirus type 1 (BPV-1) DNA replicates episomally and requires two virally expressed proteins, E1 and E2, for this process. Both proteins bind to the BPV-1 genome in the region that functions as the origin of replication. The binding sequences for the E2 protein have been characterized previously, but little is known about critical sequence requirements for E1 binding. Using a bacterially expressed E1 fusion protein, we examined binding of the BPV-1 E1 protein to the origin region. E1 strongly protected a 28-bp segment of the origin (nucleotides 7932 to 15) from both DNase I and exonuclease III digestion. Additional exonuclease III protection was observed beyond the core region on both the 5' and 3' sides, suggesting that E1 interacted with more distal sequences as well. Within the 28-bp protected core, there were two overlapping imperfect inverted repeats (IR), one of 27 bp and one of 18 bp. We show that sequences within the smaller, 18-bp IR element were sufficient for specific recognition of DNA by E1 and that additional BPV-1 sequences beyond the 18-bp IR element did not significantly increase origin binding by E1 protein. While the 18-bp IR element contained sequences sufficient for specific binding by E1, E1 did not form a stable complex with just the isolated 18-bp element. Formation of a detectable E1-DNA complex required that the 18-bp IR be flanked by additional DNA sequences. Furthermore, binding of E1 to DNA containing the 18-bp IR increased as a function of overall increasing fragment length. We conclude that E1-DNA interactions outside the boundaries of the 18-bp IR are important for thermodynamic stabilization of the E1-DNA complex. However, since the flanking sequences need not be derived from BPV-1, these distal E1-DNA interactions are not sequence specific. Comparison of the 18-bp IR from BPV-1 with the corresponding region from other papillomaviruses revealed a symmetric conserved consensus sequence, T-RY--TTAA--RY-A, that may reflect the specific nucleotides critical for E1-DNA recognition.     

Identification of the DNA-binding domain of the FLP recombinase

We have subjected the FLP protein of the 2-micron plasmid to partial proteolysis by proteinase K and have found that FLP can be digested into two major proteinase K-resistant peptides of 21 and 13 kDa, respectively. The 21-kDa peptide contains a site-specific DNA-binding domain that binds to the FLP recognition target (FRT) site with an affinity similar to that observed for the native FLP protein. This peptide can induce DNA bending upon binding to a DNA fragment containing the FRT site, but the angle of the bend (approximately 24 degrees) is smaller in magnitude than that induced by the native FLP protein (60 degrees). The additional DNA bending induced by the interaction between two native FLP molecules bound to the FRT site is not observed with the 21-kDa DNA-binding peptide. Amino-terminal sequencing has been used to map this peptide to an internal region of FLP that begins at residue Leu-148. It is likely that the DNA-binding peptide includes the catalytic site of the FLP protein.     

DNA recognition by the FLP recombinase of the yeast 2 mu plasmid. A mutational analysis of the FLP binding site

The 2 mu plasmid of the yeast Saccharomyces cerevisiae encodes a site-specific recombination system consisting of the FLP protein and two inverted recombination sites on the plasmid. The minimal fully functional substrate for in-vitro recombination in this system consists of two FLP protein binding sites separated by an eight base-pair spacer sequence. We have used site-directed mutagenesis to generate every possible mutation (36 in all) within 11 base-pairs of one FLP protein binding site and the base-pair immediately flanking it. The base-pairs within the binding site can be separated into three classes on the basis of these results. Thirty of the 36 sequence changes, including all three at seven different positions (class I) produce a negligible or modest effect on FLP protein-promoted recombination. In particular, most transition mutations are well-tolerated in this system. In only one case do all three possible mutations produce large effects (class II). At three positions, clustered near the site at which DNA is cleaved by FLP protein, one of the two possible transversions produces a large effect on recombination, while the other two changes produce modest effects (class III). For seven mutants for which FLP protein binding was measured, a direct correlation between decreases in recombination activity and in binding was observed. Positive effects on the reaction potential of mutant sites are observed when the other FLP binding site in a single recombination site is unaltered or when the second recombination site in a reaction is wild-type. This suggests a functional interaction between FLP binding sites both in cis and in trans. When two mutant recombination sites (each with 1 altered FLP binding site) are recombined, the relative orientation of the mutations (parallel or antiparallel) has no effect on the result. These results provide an extensive substrate catalog to complement future studies in this system.     

Substrate recognition by the 2 micron circle site-specific recombinase: effect of mutations within the symmetry elements of the minimal substrate.

The minimal substrate for the 2 microns circle site-specific recombinase FLP consists of a nearly perfect 13-base-pair dyad symmetry with an 8-base-pair core. By using a series of chemically synthesized FLP substrates in in vitro FLP recombination and FLP-binding assays, we have identified four positions within each of the symmetry elements that are important contact points for the FLP protein. Furthermore, the binding and recombination data provide evidence for cooperativity between the two symmetry elements of a substrate and between the symmetry elements of two partner substrates during FLP recombination.     

Site-specific DNA binding of nuclear factor I: effect of the spacer region.

Nuclear factor I (NFI) is a site-specific DNA binding protein required for the replication of adenovirus type 2 DNA in vitro and in vivo. To study sequence requirements for the interaction of NFI with DNA, we have measured the binding of the protein to a variety of synthetic sites. Binding sites for NFI (FIB sites) were previously shown to contain a consensus sequence composed of 2 motifs, TGG (Motif 1), and GCCAA (Motif 2), separated by a 6 or 7bp spacer region. To assess conserved sequences in the spacer region and flanking sequences which affect NFI binding, we have isolated clones from oligonucleotide libraries that contain the two motifs flanked by 3 degenerate nucleotides and separated by degenerate spacer regions of 6 or 7 nucleotides. With a 6bp spacer region, a strong bias exists for a C or A residue in the first position of the spacer. Sites with a 7bp spacer region contain a G and C or A residue at the first and second positions, respectively, of the spacer, but also possess conserved residues at other positions of the site.