On the nucleotide sequence recognized by a eukaryotic site-specific endonuclease, Endo.SceI from yeast

Abstract

Endo.SceI which is isolated from cells of Saccharomyces cerevisiae is a eukaryotic site-specific endonuclease active on double-stranded DNA. At each cleavage site, Endo.SceI cuts only a defined phosphodiester bond in each strand of the double helix. We compared nucleotide sequences around five cleavage sites for Endo.SceI using a computer. We could not find any common specific sequence consisting of five base pairs or more among them. However, we found a 26-base pair consensus sequence which included 15 conserved nucleotides, allowing any of the five sequences to include a few nucleotides deviated from the consensus sequence. The consensus sequence is 5'-CAn*PYnnAnnCYYGTTnnnPnYnnYA-3', where P, Y, n, and * denote purine, pyrimidine, any nucleotide, and the center of the cleavage site, respectively. The numbers of sites at which the consensus sequence appears in pBR322 DNA, phi X174 replicative form DNA, fd replicative form DNA, or SV40 DNA are close to those of the cleavage sites for Endo.SceI. We found that a 33-base pair fragment was efficiently cut at the defined phosphodiester bonds by Endo.SceI. This 33-base pair fragment included 25 base pairs out of the 26-base pair consensus sequence. The fragments in which a part of the consensus sequence was missing were not cut by Endo.SceI. These observations suggest that the consensus sequence described above is the major characteristic around the cleavage sites recognized by Endo.SceI and that the mode of recognition of cleavage sites by Endo.SceI is different from that by restriction endonucleases. We found homology between the consensus sequence for Endo.SceI and the sequences around the cleavage sites for two other site-specific endonucleases of S. cerevisiae: Endo.SceII and YZ-Endo which is involved in mating type switching.