Benzenoid-derived metabolites act as precursors for a wide variety of products involved in essential metabolic roles in eukaryotic cells. They are synthesised in plants and some fungi through the phenylalanine ammonia lyase (PAL) or tyrosine ammonia lyase (TAL) pathways. Ascomycete yeasts and animals both lack the capacity for PAL/TAL pathways, and metabolic reactions leading to benzenoid synthesis in these organisms have remained incompletely known for decades. Here we show genomic, transcriptomic and metabolomic evidence that yeasts use a mandelate pathway to synthesise benzenoids, with some similarities to pathways used by bacteria. We conducted feeding experiments using a synthetic fermentation medium that contained either 13C-phenylalanine or 13C-tyrosine and, using methylbenzoylphosphonate (MBP) to inhibit benzoylformate decarboxylase, we were able to accumulate intracellular intermediates in the yeast Hanseniaspora vineae. To further confirm this pathway, we tested three mutants in separate fermentation experiments with deletions in the key genes putatively proposed to form benzenoids (Saccharomyces cerevisiae aro10Δ, dld1Δ and dld2Δ). Our results elucidate the mechanism of benzenoid synthesis in yeast through phenylpyruvate linked with the mandelate pathway to produce benzyl alcohol and 4-hydroxybenzaldehyde from the aromatic amino acids, phenylalanine and tyrosine, as well as sugars. These results provide an explanation for the origin of the benzoquinone ring, 4-hydroxybenzoate, and suggest Aro10p has benzoylformate and 4-hydroxybenzoylformate decarboxylase functions in yeast.
Importance We present here evidence of the existence of the mandelate pathway in yeast for the synthesis of benzenoids. The link between phenylpyruvate and 4-hdroxyphenlypyruvate-derived compounds with the corresponding synthesis of benzaldehydes through benzoylformate decarboxylation is demonstrated. Hanseniaspora vineae was used in these studies because of a high capacity to produce benzenoid derivatives of two orders of magnitude higher than that produced by Saccharomyces. Contrary to what was hypothesised, β-oxidation derivatives or 4-coumaric acid are not intermediates in the synthesis of yeast benzenoids. Our results might offer an answer to the long standing question of the origin of 4-hydroxybenzoate to the synthesis of Q10 in humans.