Abstract

Data from thousands of human genomes is currently available, revealing several thousand variations in our genes. A grand challenge is to determine what, if any, impact each missense mutation has on human health. The fact that many human genes can functionally replace their corresponding yeast equivalents provides an opportunity to use humanized yeast to model human genetic variation. Swapping essential yeast genes with their human counterparts allows us to link the fitness of the human protein variant with the fitness of the yeast cell. Therefore, mutations that have functional consequences will show phenotypic differences in the humanized yeast, such as slower growth rate.
We demonstrate this approach on the rare human genetic disorder known as mevalonate kinase deficiency by devising a strategy to conditionally replace yeast ScERG12 gene with the orthologous human mevalonate kinase gene, Hs-MVK. Next, we demonstrate that yeast growth acts as an easily measured proxy for the proper functioning of the human gene. By humanizing yeast with gene variants followed by sequencing as a readout to deconvolute each strain’s relative growth, we can score the functional impact of human gene variation. Thus, neutral variants are easily distinguished from deleterious variants. This work, in addition to previous similar strategies, establishes a platform for using humanized yeast to model human genetic variation at scale.