Metabolic engineering consistently demands to produce the maximum carbon and energy flux to target chemicals. To balance metabolic flux, gene expression levels of artificially synthesized pathways usually fine-tuned using multimodular optimization strategy. However, forward construction is an engineering conundrum because a vast number of possible pathway combinations need to be constructed and analyzed.Here, an iterative high-throughput balancing (IHTB) strategy was established to thoroughly fine-tune the (2S)-naringenin biosynthetic pathway. A series of gradient constitutive promoters from Escherichia coli were randomly cloned upstream of pathway genes, and the resulting library was screened using an ultraviolet spectrophotometry-fluorescence spectrophotometry high-throughput method, which was established based on the interactions between AlCl 3 and (2S)-naringenin. The metabolic flux of the screened high-titer strains was analyzed and iterative rounds of screening were performed based on the analysis results. After several rounds, the metabolic flux of the (2S)-naringenin synthetic pathway was balanced, reaching a final titer of 191.9 mg/L with 29.2 mg/L p-coumaric acid accumulation. Chalcone synthase was speculated to be the rate-limiting enzyme because its expression level was closely related to the production of both (2S)-naringenin and p-coumaric acid. The established IHTB strategy can be used to efficiently balance multigene pathways, which will accelerate the development of efficient recombinant strains. K E Y W O R D S flavonoids, metabolic engineering, modular optimization, promoter Biotechnology and Bioengineering. 2019;116:1392-1404. wileyonlinelibrary.com/journal/bit 1392 |

(2S)-Naringenin, a (2S)-flavanone, is widely used in the food, chemical, and pharmaceutical industries because of its diverse physiological activities. The production of (2S)-naringenin in microorganisms provides an ideal source that reduces the cost of the flavonoid. To achieve efficient production of (2S)-naringenin in Saccharomyces cerevisiae (S. cerevisiae), we constructed a biosynthetic pathway from p-coumaric acid, a cost-effective and more efficient precursor. The (2S)-naringenin synthesis pathway genes were integrated into the yeast genome to obtain a (2S)-naringenin production strain. After gene dosage experiments, the genes negatively regulating the shikimate pathway and inefficient chalcone synthase activity were verified as factors limiting (2S)-naringenin biosynthesis. With fed-batch process optimization of the engineered strain, the titer of (2S)-naringenin reached 648.63 mg/L from 2.5 g/L p-coumaric acid. Our results indicate that the constitutive production of (2S)-naringenin from p-coumaric acid in S. cerevisiae is highly promising.

Dihydromyricetin (DHM) is a traditional plant-extracted flavonoid with some health benefits. This study aimed to metabolically engineer the strains for DHM bioproduction. Two strains of BK-11 and BQ-21 were integrated with flavonoid 3hydroxylase (F3H) or both F3H and flavonoid 3′-hydroxylase (F3′H). The resulting strains have expressed the enzymes of GmCPR and SlF3′5′H, and then, the promoters of INO1p and TDH1p were used to enhance further the DHM production from naringenin in Saccharomyces cerevisiae. Through multiple-copy integration, 709.6 mg/L DHM was obtained by adding 2.5 g/L naringenin in a 5 L bioreactor, implying that the synergistic effect between F3′H and flavonoid 3′5′-hydroxylase is likely to promote the DHM production. An yield of 246.4 mg/L DHM was obtained from glucose by deleting genes for branch pathways and integrating PhCHS, MsCHI, Pc4CL, and FjTAL. To our knowledge, this is the highest production reported for the de novo biosynthesis of DHM.

The (2S)-naringenin is an important natural flavonoid with several bioactive effects on human health. It is also a key precursor in the biosynthesis of other high value compounds. The production of (2S)-naringenin is significantly influenced by the acetyl-CoA available in the cytosol. In this study, we increased the acetyl-CoA supply via the β-oxidation of fatty acids in the peroxisomes of Saccharomyces cerevisiae. Several lipases from different sources and PEX11, FOX1, FOX2, and FOX3, the key genes of the fatty acid β-oxidation pathway, were overexpressed during the production of (2S)-naringenin in yeast. The level of acetyl-CoA was 0.205 nmol higher than that in the original strain and the production of (2S)-naringenin increased to 286.62 mg/g dry cell weight when PEX11 was overexpressed in S. cerevisiae strain L07. Remarkable (2S)-naringenin production (1129.44 mg/L) was achieved with fed-batch fermentation, with the highest titer reported in any microorganism. Our results demonstrated the use of fatty acid β-oxidation to increase the level of cytoplasmic acetyl-CoA and the production of its derivatives.

Flavan-3-ols are a group of flavonoids that exert beneficial effects. This study aimed to enhance key metabolic processes related to flavan-3-ols biosynthesis. The engineered Saccharomyces cerevisiae strain E32 that produces naringenin from glucose was further engineered for de novo production of two basic flavan-3-ols, afzelechin (AFZ) and catechin (CAT). Through introduction of flavonoid 3-hydroxylase, flavonoid 3′-hydroxylase, dihydroflavonol 4-reductase (DFR), and leucoanthocyanidin reductase (LAR), de novo production of AFZ and CAT can be achieved. The combination of FaDFR from Fragaria × ananassa and VvLAR from Vitis vinifera was optimal. (GGGGS)2 and (EAAAK)2 linkers between DFR and LAR proved optimal for the production of AFZ and CAT, respectively. Optimization of promoters and the enhanced supply of NADPH further increased the production. By combining the best engineering strategies, the optimum strains produced 500.5 mg/L AFZ and 321.3 mg/L CAT, respectively, after fermentation for 90 h in a 5 L bioreactor. The strategies presented could be applied for a more efficient production of flavan-3-ols by various microorganisms.