The isomeric preference of Holliday junctions influences resolution bias by lambda integrase.

Lambda site-specific recombination proceeds by a pair of sequential strand exchanges that first generate and then resolve a Holliday junction intermediate. A family of synthetic Holliday junctions with the branch point constrained to the center of the 7 bp overlap region was used to show that resolution of the top strands and resolution of the bottom strands are symmetrical but stereochemically distinct processes. Lambda integrase is sensitive to isomeric structure, preferentially resolving the pair of strands that are crossed in the protein-free Holliday junction. At the branch point of stacked immobile Holliday junctions, the number of purines is preferentially maximized in the crossed (versus continuous) strands if there is an inequality of purines between strands of opposite polarity. This stacking preference was used to anticipate the resolution bias of freely mobile junctions and thereby to reinforce the conclusions with monomobile junctions. The results provide a strong indication that in the complete recombination reaction a restacking of helices occurs between the top and bottom strand exchanges.

Swapping DNA strands and sensing homology without branch migration in lambda site-specific recombination.

BACKGROUND: Many site-specific recombinases act by forming and resolving branched Holliday junction intermediates. Previous findings have been consistent with models involving branch migration across the 'overlap region' of obligate homology, located between the staggered sites where the two single-strand exchanges occur. We have investigated the validity of such models in the case of bacteriophage lambda site-specific recombination. RESULTS: By using synthetic lambda att-site Holliday junctions, incorporating sequence heterologies that impose constraints on branch migration, we have found that the optimal position of the junction for either top-strand or bottom-strand resolution by lambda integrase (Int) is not at the ends, but close to the middle of the seven base-pair overlap region. A minor shift of the branch point around the central base pair caused a remarkable switch in resolution bias. Our findings suggest that branch migration is limited to the central one to three base pairs of the overlap region. They lead to a new model for lambda site-specific recombination, in which there are two symmetrical swaps of two to three nucleotides each, linked by a central isomerization step that causes a change of the stacking interactions between the four junction arms. On the basis of isolated strand-joining reactions carried out by Int in the presence or absence of base complementarity, we propose that sequence homology is sensed during the annealing step prior to strand joining. The new model eliminates mechanistic complications associated with large helical rotations required by branch-migration models. CONCLUSIONS: The results reported here suggest that the recognition of sequence homology in Int-dependent site-specific recombination does not rely primarily on branch migration. The property of cleaving Holliday junctions a few base pairs away from the crossover puts lambda Int into the same category as endonucleases that cleave Holliday junctions in homologous recombination.