Efficient gene targeting in non-homologous end-joining-deficient Lipomyces starkeyi strains.

Microbial lipids are sustainable feedstock for the production of oleochemicals and biodiesel. Oleaginous yeasts have recently been proposed as alternative lipid producers to plants and animals to promote sustainability in the chemical and fuel industries. The oleaginous yeast Lipomyces starkeyi has great industrial potential as an excellent lipid producer. However, improvement of its lipid productivity is essential for the cost-effective production of oleochemicals and fuels. Genetic and metabolic engineering of L. starkeyi via gene manipulation techniques may result in improvements in lipid production and our understanding of the mechanisms behind lipid biosynthesis pathways. We previously described an integrative transformation system using a drug-resistant marker for L. starkeyi. However, gene-targeting frequencies were very low because non-homologous recombination is probably predominant in L. starkeyi. Genetic engineering tools for L. starkeyi have not been sufficiently developed. In this study, we describe a new genetic tool and its application in L. starkeyi. To develop a highly efficient gene-targeting system for L. starkeyi, we constructed a series of mutants by disrupting genes for LsKu70p, LsKu80p, and/or LsLig4p, which share homology with other yeasts Ku70p, Ku80p, and Lig4p, respectively, being involved in non-homologous end-joining pathway. Deletion of the LsLIG4 gene dramatically improved the homologous recombination efficiency (80.0%) at the LsURA3 locus compared with that in the wild-type strain (1.4%), when 2000-bp homologous flanking regions were used. The homologous recombination efficiencies of the double mutant ∆l sku70∆lslig4 and the triple mutant ∆lsku70∆lsku80∆lslig4 were also markedly enhanced. Therefore, the L. starkeyi ∆lslig4 background strains have promise as efficient recipient strains for genetic and metabolic engineering approaches in this yeast.