Efficient production of mycosporine-like amino acids, natural sunscreens, in Yarrowia lipolytica.

BACKGROUND: Mycosporine-like amino acids (MAAs), including shinorine and porphyra-334, are gaining attention as safe natural sunscreens. The production of MAAs has been achieved in diverse microbial hosts, including Saccharomyces cerevisiae. While S. cerevisiae is the most extensively studied model yeast, the oleaginous yeast Yarrowia lipolytica has emerged as a promising candidate for the synthesis of valuable products. In this study, we explored the potential of Y. lipolytica as a host for producing MAAs, utilizing its advantages such as a robust pentose phosphate pathway flux and versatile carbon source utilization. RESULTS: We produced MAAs in Y. lipolytica by introducing the MAA biosynthetic genes from cyanobacteria Nostoc punctiforme and Anabaena variabilis. These genes include mysA, mysB, and mysC responsible for producing mycosporine-glycine (MG) from sedoheptulose 7-phosphate (S7P). The two strains utilize different enzymes, D-Ala-D-Ala ligase homologue (MysD) in N. punctiforme and NRPS-like enzyme (MysE) in A. variabilis, for amino acid conjugation to MG. MysE specifically generated shinorine, a serine conjugate of MG, while MysD exhibited substrate promiscuity, yielding both shinorine and a small amount of porphyra-334, a threonine conjugate of MG. We enhanced MAAs production by selecting mysA, mysB, and mysC from A. variabilis and mysD from N. punctiforme based on their activities. We further improved production by strengthening promoters, increasing gene copies, and introducing the xylose utilization pathway. Co-utilization of xylose with glucose or glycerol increased MAAs production by boosting the S7P pool through the pentose phosphate pathway. Overexpressing GND1 and ZWF1, key genes in the pentose phosphate pathway, further enhanced MAAs production. The highest achieved MAAs level was 249.0 mg/L (207.4 mg/L shinorine and 41.6 mg/L of porphyra-334) in YP medium containing 10 g/L glucose and 10 g/L xylose. CONCLUSIONS: Y. lipolytica was successfully engineered to produce MAAs, primarily shinorine. This achievement involved the introduction of MAA biosynthetic genes from cyanobacteria, establishing xylose utilizing pathway, and overexpressing the pentose phosphate pathway genes. These results highlight the potential of Y. lipolytica as a promising yeast chassis strain for MAAs production, notably attributed to its proficient expression of MysE enzyme, which remains non-functional in S. cerevisiae, and versatile utilization of carbon sources like glycerol.

Efficient production of natural sunscreens shinorine, porphyra-334, and mycosporine-2-glycine in Saccharomyces cerevisiae.

Mycosporine-like amino acids (MAAs) are promising natural sunscreens mainly produced in marine organisms. Until now, metabolic engineering efforts to produce MAAs in heterologous hosts have mainly focused on shinorine production, and the low production levels are still not suitable for industrial applications. In this study, we successfully developed Saccharomyces cerevisiae strains that can efficiently produce various disubstituted MAAs, including shinorine, porphyra-334, and mycosporine-2-glycine (M2G), which are formed by conjugating serine, threonine, and glycine to mycosporine-glycine (MG), respectively. We first generated an MG-producing strain by multiple integration of the biosynthetic genes from cyanobacteria and applying metabolic engineering strategies to increase sedoheptulose-7-phosphate pool, a substrate for MG production. Next, five mysD genes from cyanobacteria, which encode D-Ala-D-Ala ligase homologues that conjugate an amino acid to MG, were introduced into the MG-producing strain to determine the substrate preference of each MysD enzyme. MysDs from Lyngbya sp., Nostoclinckia, and Euhalothece sp. showed high specificity toward serine, threonine, and glycine, resulting in efficient production of shinorine, porphyra-334, and M2G, respectively. This is the first report on the production of porphyra-334 and M2G in S. cerevisiae. Furthermore, we identified that the substrate specificity of MysD was determined by the omega loop region of 43-45 amino acids predicted based on its structural homology to a D-Ala-D-Ala ligase from Thermus thermophilus involved in peptidoglycan biosynthesis. The substrate specificities of two MysD enzymes were interchangeable by swapping the omega loop region. Using the engineered strain expressing mysD from Lyngbya sp. or N. linckia, up to 1.53 g/L shinorine or 1.21 g/L porphyra-334 was produced by fed-batch fermentation in a 5-L bioreactor, the highest titer reported so far. These results suggest that S. cerevisiae is a promising host for industrial production of different types of MAAs, providing a sustainable and eco-friendly alternative for the development of natural sunscreens.

Efficient production of shinorine, a natural sunscreen material, from glucose and xylose by deleting HXK2 encoding hexokinase in Saccharomyces cerevisiae.

Mycosporine-like amino acids (MAAs), microbial secondary metabolites with ultraviolet (UV) absorption properties, are promising natural sunscreen materials. Due to the low efficiency of extracting MAAs from natural producers, production in heterologous hosts has recently received attention. Shinorine is a well characterized MAA with strong UV-A absorption property. Previous, we developed Saccharomyces cerevisiae strain producing shinorine by introducing four shinorine biosynthetic genes from cyanobacterium Nostoc punctiforme. Shinorine is produced from sedoheptulose 7-phosphate (S7P), an intermediate in the pentose phosphate pathway. Shinorine production was greatly improved by using xylose as a co-substrate, which can increase the S7P pool. However, due to a limited xylose-utilizing capacity of the engineered strain, glucose was used as a co-substrate to support cell growth. In this study, we further improved shinorine production by attenuating glucose catabolism via glycolysis, which can redirect the carbon flux from glucose to the pentose phosphate pathway favoring shinorine production. Of the strategies we examined to reduce glycolytic flux, deletion of HXK2, encoding hexokinase, was most effective in increasing shinorine production. Furthermore, by additional expression of Ava3858 from Anabaena variabilis, encoding a rate-limiting enzyme 2-demethyl 4-deoxygadusol synthase, 68.4 mg/L of shinorine was produced in an optimized medium containing 14 g/L glucose and 6 g/L xylose, achieving a 2.2-fold increase compared with the previous strain.

High-yield production of (R)-acetoin in Saccharomyces cerevisiae by deleting genes for NAD(P)H-dependent ketone reductases producing meso-2,3-butanediol and 2,3-dimethylglycerate.

Acetoin is widely used in food and cosmetics industries as a taste and fragrance enhancer. To produce (R)-acetoin in Saccharomyces cerevisiae, acetoin biosynthetic genes encoding α-acetolactate synthase (AlsS) and α-acetolactate decarboxylase (AlsD) from Bacillus subtilis and water-forming NADH oxidase (NoxE) from Lactococcus lactis were integrated into delta-sequences in JHY605 strain, where the production of ethanol, glycerol, and (R,R)-2,3-butanediol (BDO) was largely eliminated. We further improved acetoin production by increasing acetoin tolerance by adaptive laboratory evolution, and eliminating other byproducts including meso-2,3-BDO and 2,3-dimethylglycerate, a newly identified byproduct. Ara1, Ypr1, and Ymr226c (named Ora1) were identified as (S)-alcohol-forming reductases, which can reduce (R)-acetoin to meso-2,3-BDO in vitro. However, only Ara1 and Ypr1 contributed to meso-2,3-BDO production in vivo. We elucidate that Ora1, having a substrate preference for (S)-acetoin, reduces (S)-α-acetolactate to 2,3-dimethylglycerate, thus competing with AlsD-mediated (R)-acetoin production. By deleting ARA1, YPR1, and ORA1, 101.3 g/L of (R)-acetoin was produced with a high yield (96% of the maximum theoretical yield) and high stereospecificity (98.2%).