Rapid localization and characterization of random mutations within the 2 micron circle site-specific recombinase: a general strategy for analysis of protein function

A dual reporter screening system identifies the amino acid at position 82 in Flp site-specific recombinase as a determinant for target specificity

We have developed a dual reporter screen in Escherichia coli for identifying variants of the Flp site-specific recombinase that have acquired reactivity at an altered target site (mFRT). In one reporter, the lacZalpha gene segment is flanked by mFRTs in direct orientation. In the other, the red fluorescence protein (RFP) gene is flanked by the native FRTs. Hence, the color of a colony on an X-gal indicator plate indicates the recombination potential of the variant Flp protein expressed in it: blue if no recombination or only FRT recombination occurs, red if only mFRT recombination occurs and white if both FRT and mFRT recombinations occur. The scheme was validated by identification and in vivo characterization of Flp variants that show either relaxed specificity (active on FRT and mFRT) or moderately shifted specificity toward mFRT. We find that alteration of Lys-82 to Met, Thr, Arg or His enables the corresponding Flp variants to recombine FRT sites as well as altered FRT sites containing a substitution of G-C by C-G at position 1 of the Flp binding element (mFRT11). In contrast, wild-type Flp has no detectable activity on mFRT11. When Lys-82 is replaced by Tyr, the resulting Flp variant shows a small but reproducible preference for mFRT11 over FRT. However, this preference for mFRT11 is nearly lost when Tyr-82 is substituted by Phe.

Different thermostabilities of FLP and Cre recombinases: implications for applied site-specific recombination

Genomic manipulations using site-specific recombinases rely on their applied characteristics in living systems. To understand their applied properties so that they can be optimally deployed, we compared the recombinases FLP and Cre in two assays. In both Escherichia coli and in vitro, FLP shows a different temperature optimum than Cre. FLP is more thermolabile, having an optimum near 30 degrees C and little detectable activity above 39 degrees C. Cre is optimally efficient at 37 degrees C and above. Consistent with FLP thermolability, recombination in a mammalian cell line mediated by a ligand- regulated FLP-androgen receptor fusion protein is more efficient at 35 degrees C than at higher temperatures. We also document a mutation in a commercially available FLP plasmid (FLP-F70L) which renders this recombinase even more thermolabile. The different temperature optima of FLP, FLP-F70L and Cre influence their strategies of usage. Our results recommend the use of Cre for applications in mice that require efficient recombination. The thermolabilities of FLP and FLP-F70L can be usefully exploited for gain of function and cell culture applications.

Activity of the yeast FLP recombinase in Arabidopsis

The coding sequence for FLP recombinase, originally from the 2 mu plasmid of Saccharomyces cerevisiae, was introduced into Arabidopsis behind the cauliflower mosaic virus 35S promoter. FLP activity was monitored by the glucuronidase activity resulting from inversion of an antisense-oriented GUS reporter gene flanked by a pair of FRT target sites in inverted repeat. FLP-dependent Gus activity was observed in both transient assays and transgenic plants. The FLP system will be useful for a variety of in planta genetic manipulations.

Domain of a yeast site-specific recombinase (Flp) that recognizes its target site

Binding of a partial proteolytic digest by V8 enzyme of the yeast site-specific recombinase Flp to its target site gives rise to DNA-protein complexes that are smaller than those produced by the full-sized protein. The smallest of these complexes (occupancy of one peptide monomer per site) contains either one of two polypeptides (32 and 28 kDa) of the V8 digestion mixture. The amino termini of both polypeptides map to Ser-129 of Flp, corresponding to V8 cleavage at Glu-128. The relative mobilities of the complexes formed by the V8 peptides indicate that they lack the sharp substrate bend that is characteristic of Flp-derived complexes. A hybrid protein consisting of the amino-terminal one-third of the R recombinase (from Zygosaccharomyces rouxii) and the carboxyl-terminal two-thirds of Flp recognizes the Flp target site.

Tyr60 variants of Flp recombinase generate conformationally altered protein-DNA complexes. Differential activity in full-site and half-site recombinations

The tyrosine at position 60 of the Flp recombinase of the Saccharomyces cerevisiae plasmid, 2 mu circle, is invariant among site-specific recombinases of the "yeast plasmid family". Alterations of this residue give rise to Flp variants that show no recombination activity when assayed in vivo in Escherichia coli. Upon purification, they bind substrate, execute DNA cleavage and catalyze recombination. The efficiency of strand cleavage follows the order: Flp(Y60F) greater than Flp greater than Flp(Y60S) greater than Flp(Y60D); efficiency of recombination between Flp sites on a linear substrate and a circular one follows the order: Flp greater than Flp(Y60F) greater than Flp(Y60S) greater than Flp(Y60D). Methylation footprints of the DNA-protein complexes formed by two of the Flp variants, Flp(Y60S) and Flp(Y60D), do not show hypermethylation of the G residues within the substrate core that is characteristic of complexes formed by wild-type Flp. The third variant, Flp(Y60F), causes significant distortion (although less than wild-type Flp) of the substrate core, as indicated by enhanced G-methylation. Binding profiles with circularly permuted substrates indicate that Flp(Y60S) and Flp(Y60D), but not Flp(Y60F), are defective in bending substrate DNA. In recombination between two Flp half-sites, the variant proteins are significantly more active than in normal full-site recombination.