We tested the ability of the recombinant yeast strain to synthesize flavanone compounds by feeding it with phenylpropanoid acids, such as cinnamic acid, p-coumaric acid, caffeic acid, and ferulic acid. In order to reduce cell growth inhibition, phenylpropanoid precursors were added every 13 h to the cultures in five equal doses, reaching a final concentration of 1 mM. In all cases, cultures were terminated after 65 h of incubation and flavonoid substances were extracted from the culture broth with an equal volume of ethyl acetate. They were further analyzed by reverse-phase high-performance liquid chromatography (HPLC) using an acetonitrile-water gradient, at a flow rate of 1.0 ml/min. The HPLC conditions were as follows: 10 to 40% for 10 min, 40 to 60% for 5 min, and 60 to 10% for 2 min. The retention times under these HPLC conditions for the standard authentic samples are presented in Table .
Retention times of standard (authentic) compounds used in the present study
When cinnamic acid was used as a precursor metabolite and galactose as the sole carbon source and inducer, a large amount (16.3 mg/liter) of the corresponding unhydroxylated flavanone pinocembrin accumulated in the medium. This is a 22-fold increase compared to the amount of pinocembrin produced by the most efficient E. coli
recombinant strain (7
). However, only a relatively low concentration of naringenin (0.2 mg/liter) was detected, demonstrating that although C4H was functionally expressed in yeast, it is still a rate-limiting step enzyme in the four-enzyme hybrid pathway. It is possible that increasing the activity of CPR1, the yeast P450 reductase that is required for P450 monooxygenase function through episomal overexpression, could lead to increased C4H activity (6
). When p
-coumaric acid was used as a precursor, a large amount of naringenin (28.3 mg/liter) accumulated in the culture, which is 62 times higher compared to the amount of naringenin produced by the most efficient recombinant E. coli
strain. Similarly, when caffeic acid was used as a precursor, natural (2S
)-eriodictyol was produced in significant amounts (6.5 mg/liter). This is the first time eriodictyol biosynthesis has been achieved through microbial fermentation. Finally, ferulic acid, which carries a methoxy group on the aromatic ring, failed to be metabolized by the recombinant yeast strain. This result is in agreement with recent data obtained by Schroeder et al. demonstrating that O
-methylations on the B ring of flavonoid substrates result in complete loss of enzymatic activity. It is therefore possible that B-ring methylations occur later in the complex flavonoid pathway to the end products (i.e., after flavanones have been synthesized) (13
). The biotransformation results are summarized in Table .
Production of natural flavanones by recombinant yeast S. cerevisiae INVSCI harboring plasmid Ycc4c181a
In conclusion, we describe the biosynthesis of milligram quantities of flavanone substances from an S. cerevisiae recombinant strain that carries a plant-derived gene cluster. Since many flavonoid substances are formed through the action of P450 monooxygenases that cannot be readily expressed in E. coli, our success in producing a variety of flavanone skeletons from recombinant yeast will allow us to proceed in the future with the biosynthesis of several other high-value flavonoid molecules, such as genistein and quercetin.