Metabolic Pathway Coupled with Fermentation Process Optimization for High-Level Production of Retinol in Yarrowia lipolytica.

Retinol is a lipid-soluble form of vitamin A that is crucial for human visual and immune functions. The production of retinol through microbial fermentation has been the focus of recent exploration. However, the obtained titer remains limited and the product is often a mixture of retinal, retinol, and retinoic acid, necessitating purification. To achieve efficient biosynthesis of retinol in Yarrowia lipolytica, we improved the metabolic flux of ß-carotene to provide sufficient precursors for retinol in this study. Coupled with the optimization of the expression level of ß-carotene 15,15'-dioxygenase, de novo production of retinol was achieved. Furthermore, Tween 80 was used as an extractant and butylated hydroxytoluene as an antioxidant to extract intracellular retinol and prevent retinol oxidation, respectively. This strategy significantly increased the level of retinol production. By optimizing the enzymes converting retinal to retinol, the proportion of extracellular retinol in the produced retinoids reached 100%, totaling 1042.3 mg/L. Finally, total retinol production reached 5.4 g/L through fed-batch fermentation in a 5 L bioreactor, comprising 4.2 g/L extracellular retinol and 1.2 g/L intracellular retinol. This achievement represents the highest reported titer so far and advances the industrial production of retinol.

High-Level De Novo Production of (2S)-Eriodictyol in Yarrowia Lipolytica by Metabolic Pathway and NADPH Regeneration Engineering.

(2S)-Eriodictyol, a polyphenolic flavonoid, has found widespread applications in health supplements and food additives. However, the limited availability of plant-derived (2S)-eriodictyol cannot meet the market demand. Microbial production of (2S)-eriodictyol faces challenges, including the low catalytic efficiency of flavone 3'-hydroxylase/cytochrome P450 reductase (F3'H/CPR), insufficient precursor supplementation, and inadequate NADPH regeneration. This study systematically engineered Yarrowia lipolytica for high-level (2S)-eriodictyol production. In doing this, the expression of F3'H/CPR was balanced, and the supply of precursors was enhanced by relieving feedback inhibition of the shikimate pathway, promoting fatty acid ß-oxidation, and increasing the copy number of synthetic pathway genes. These strategies, combined with NADPH regeneration, achieved an (2S)-eriodictyol titer of 423.6 mg/L. Finally, in fed-batch fermentation, a remarkable 6.8 g/L (2S)-eriodictyol was obtained, representing the highest de novo microbial titer reported to date and paving the way for industrial production.

YaliCMulti and YaliHMulti: Stable, efficient multi-copy integration tools for engineering Yarrowia lipolytica.

Yarrowia lipolytica is widely used in biotechnology to produce recombinant proteins, food ingredients and diverse natural products. However, unstable expression of plasmids, difficult and time-consuming integration of single and low-copy-number plasmids hampers the construction of efficient production pathways and application to industrial production. Here, by exploiting sequence diversity in the long terminal repeats (LTRs) of retrotransposons and ribosomal DNA (rDNA) sequences, a set of vectors and methods that can recycle multiple and high-copy-number plasmids was developed that can achieve stable integration of long-pathway genes in Y. lipolytica. By combining these sequences, amino acids and antibiotic tags with the Cre-LoxP system, a series of multi-copy site integration recyclable vectors were constructed and assessed using the green fluorescent protein (HrGFP) reporter system. Furthermore, by combining the consensus sequence with the vector backbone of a rapidly degrading selective marker and a weak promoter, multiple integrated high-copy-number vectors were obtained and high levels of stable HrGFP expression were achieved. To validate the universality of the tools, simple integration of essential biosynthesis modules was explored, and 7.3 g/L of L-ergothioneine and 8.3 g/L of (2S)-naringenin were achieved in a 5 L fermenter, the highest titres reported to date for Y. lipolytica. These novel multi-copy genome integration strategies provide convenient and effective tools for further metabolic engineering of Y. lipolytica.

Explore the Influence of Shallow Information on Point Cloud Registration.

Feature extraction is a key step for deep-learning-based point cloud registration. In the correspondence-free point cloud registration task, the previous work commonly aggregates deep information for global feature extraction and numerous shallow information which is positive to point cloud registration will be ignored with the deepening of the neural network. Shallow information tends to represent the structural information of the point cloud, while deep information tends to represent the semantic information of the point cloud. In addition, fusing information of different dimensions is conducive to making full use of shallow information. Inspired by this, we verify shallow information in the middle layers can bring a positive impact on the point cloud registration task. We design various architectures to combine shallow information and deep information to extract global features for point cloud registration. Experimental results on the ModelNet40 dataset illustrate that feature extractors that incorporate shallow information will bring positive performance.