A taste for ‘sour’ sugars: characterization of a highly efficient D-galacturonic acid metabolism in the basidiomycete yeast genus Rhodosporidium

Saccharides released during the degradation of lignocellulosic material (e.g. for biorefinery purposes) are traditionally fermented using strains of Saccharomyces cerevisiae, which however has only a very limited ability to metabolize all available sugars. Thus, major efforts have to be undertaken to engineer catabolic pathways for sugars such as D-xylose, L-arabinose and D-galacturonic acid (GalA) in this ascomycete yeast. Still, the metabolism of some of these monosaccharides remains slow. Better catabolic pathways have to be transferred from fungal strains with a high ability to utilize these monosaccharides.

The oleaginous basidiomycete yeasts Rhodosporidium toruloides and Rhodotorula mucilaginosa were discovered as saprophytes colonizing for example pectin-rich olives and wine grapes and are known to be able to produce large quantities of carotenoids in short time. However, they also thrive on GalA, the main component of the pectin backbone. This is remarkable, since GalA is an oxidized sugar and therefore energetically difficult to catabolize by fungi. In addition, the molecular mechanisms behind the pectinolytic capabilities of basidiomycete fungi are almost entirely unknown.

In this work, we characterized the GalA metabolism pathway of these special yeasts on multiple levels. Following physiological growth assays with multiple strains, RNAseq analyses were performed in R. toruloides to identify genes that are differentially expressed on GalA and pectin. Moreover, RB-TDNAseq, a recently reported novel method for fitness scoring in barcoded TDNA-transformed populations, was used to identify targets affecting fitness during growth on GalA. This way, we identified the enzymes being involved in the GalA catabolism, which were heterologously expressed and their kinetics determined. Moreover, sugar transporters and a novel transcription factor putatively regulating the GalA metabolism were identified. Taken together, our results demonstrate that the genes from R. toruloides and R. mucilaginosa are promising candidates for rational engineering of an efficient GalA metabolism for example in S. cerevisiae.