Marina Svetec Miklenic

Sgs1 and Exo1 suppress targeted chromosome duplication during ends-in and ends-out gene targeting.

Gene targeting is extremely efficient in the yeast Saccharomyces cerevisiae. It is performed by transformation with a linear, non-replicative DNA fragment carrying a selectable marker and containing ends homologous to the particular locus in a genome. However, even in S. cerevisiae, transformation can result in unwanted (aberrant) integration events, the frequency and spectra of which are quite different for ends-out and ends-in transformation assays. It has been observed that gene replacement (ends-out gene targeting) can result in illegitimate integration, integration of the transforming DNA fragment next to the target sequence and duplication of a targeted chromosome. By contrast, plasmid integration (ends-in gene targeting) is often associated with multiple targeted integration events but illegitimate integration is extremely rare and a targeted chromosome duplication has not been reported. Here we systematically investigated the influence of design of the ends-out assay on the success of targeted genetic modification. We have determined transformation efficiency, fidelity of gene targeting and spectra of all aberrant events in several ends-out gene targeting assays designed to insert, delete or replace a particular sequence in the targeted region of the yeast genome. Furthermore, we have demonstrated for the first time that targeted chromosome duplications occur even during ends-in gene targeting. Most importantly, the whole chromosome duplication is POL32 dependent pointing to break-induced replication (BIR) as the underlying mechanism. Moreover, the occurrence of duplication of the targeted chromosome was strikingly increased in the exo1Δ sgs1Δ double mutant but not in the respective single mutants demonstrating that the Exo1 and Sgs1 proteins independently suppress whole chromosome duplication during gene targeting.

In Saccharomyces cerevisiae gene targeting fidelity depends on a transformation method and proportion of the overall length of the transforming and targeted DNA

Gene replacement is one of the most essential approaches in construction of the genetically modified yeast strains. However, the fidelity of gene targeting and the effort needed for construction of a particular strain can vary significantly. We investigated the influence of two important factors-the choice of the transformation method and the design of the transforming DNA fragment, which can vary in overall length (including flanking regions and selectable marker) compared to the length of the targeted region in the genome. Gene replacement fidelity was determined in several assays using electroporation and spheroplast transformation, and compared with our previous results obtained by lithium acetate. We have demonstrated clearly that gene targeting fidelity depends on the transformation protocol, being highest for lithium acetate method. In contrast, lower fidelity was observed with electroporation and spheroplast transformation. Additionally, the fidelity also depends on a design of the transformation assay, since a higher overall length ratio of the transforming DNA and targeted region results in higher fidelity. Moreover, the karyotype analysis of the aberrant transformants by qPCR demonstrates that gene targeting can result in diploidisation of haploid strains, most likely via targeted chromosome duplication followed by subsequent duplication of other chromosomes.