Holliday junctions — seeing single

Abstract

Crossing over is believed to involve an intermediate in which homologues are linked by two crossed-strand structures called Holliday junctions. This model, based on observations in Saccharomyces cerevisiae, was thought to be universal. But new work from Gerald Smith and colleagues shows that, in Schizosaccharomyces pombe, crossing over predominantly occurs between sister chromatids, rather than between homologues, and that it involves not double but single Holliday junctions. The authors used a natural hotspot of recombination — the mbs1 locus — and marked it with flanking restriction sites to allow for physical assays of recombination intermediates and crossovers. These markers also allowed them to determine the parental origin of the DNA molecules that underwent the crossovers — restriction enzyme digestion followed by 2D-gel analysis revealed that, in fission yeast, crossovers between sister chromatids outnumber those between homologues by about 3 to 1. To investigate the structure of the meiotic recombination intermediates, the authors tested their sensitivity to resolvase and to high temperature; they also examined them by electron microscopy. The results confirmed that the intermediate involves Holliday junctions and, together with 2D-gel data, revealed that most of them are not double but single Holliday junctions. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, or to obtain a text description, please contact npg@nature.com The authors propose a model to explain how recombination initiated by a double-strand break might involve a single Holliday junction intermediate. Similar to the current conventional model, the first DNA end generates a so-called D-loop with a Holliday junction at one end. However, the other end of the D-loop is cut before the second DNA end is captured (see figure). The authors provide evidence that Mus81–Eme1 — which cleaves branched DNA structures — is required for meiotic Holliday junction resolution; they propose that although it resolves the Holliday junction the D-loop is cut by a different enzyme. The long-standing model of how meiotic crossing over occurs has been challenged. Although the work in budding yeast is not itself being doubted — indeed, by electron microscopy the authors confirmed double Holliday junctions in budding yeast — the widespread applicability of these findings must now be re-examined. Depending on the aspect of biology that is under investigation, either the budding yeast or the fission yeast is a better model. The jury is out on whether most organisms resemble the budding or the fission yeast way of doing crossing over.