Search

Compatible solutes are key for the ability of halophilic bacteria to resist high osmotic stress. They have received wide attention from researchers for their excellent osmotic protection properties. Hydroxyectoine is a particularly important compatible solute, but its production by microbes faces several challenges, including low titer/yield, the presence of the byproduct ectoine, and the requirement of high salinity. Here, we aimed to metabolically engineer Escherichia coli to efficiently produce hydroxyectoine in the absence of osmotic stress without accumulating the byproduct ectoine. First, combinatorial optimization of the expression strength of key genes in the ectoine synthesis module and hydroxyectoine synthesis module was conducted. After optimization of the expression of these genes, 12.12 g/L hydroxyectoine and 0.24 g/L ectoine were obtained at 36 h in shake-flask fermentation with the addition of the co-substrate α-ketoglutarate. Further optimization of the addition of α-ketoglutarate achieved the sole production of hydroxyectoine (i.e., no ectoine accumulation), indicating that the supply of α-ketoglutarate is critically important for sole hydroxyectoine production. Finally, quorum sensing-based auto-regulation of intracellular α-ketoglutarate pool was implemented as an alternative to α-ketoglutarate addition by coupling the expression of sucA with the esaI/esaR circuit, which led to 14.93 g/L hydroxyectoine with a unit cell yield of 1.678 g/g and no ectoine accumulation in the absence of osmotic stress. This is the highest reported titer of sole hydroxyectoine production under salinity-free fermentation to date.

The task of text-to-image synthesis is a new challenge in the field of image synthesis. In the earlier research, the task of text-to-image synthesis is mainly to achieve the alignment of words and images by the way of retrieval based on the sentences or keywords. With the development of deep learning, especially the application of deep generative models in image synthesis, image synthesis achieves promising progress. The Generative adversarial networks (GANs) are one of the most significant generative models, and GANs have been successfully applied in computer vision, natural language processing and so on. In this paper, we review and summarize the recent research in GANs-based text-to-image synthesis, and provide a summary of the development of classic and advanced models. The input of the GANs-based text-to-image synthesis is not only the general text description as earlier studies, also includes scene layout and dialog text. The typical structure of each categories is elaborated. The general text-based image synthesis is the most commonly in the text-to-image synthesis, and it is subdivided into three groups based on the improvements of text information utilization, network structure and output control conditions. Through the survey, the detailed and logical overview of the evolution of GANs-based text-to-image synthesis is presented. Finally, the challenged problems and the future development of text-to-image synthesis are discussed.

The dual stream-based human action recognition model offers the advantage of high recognition accuracy, but the algorithm is less robust in case of lighting changes. The human skeleton has a strong ability to express human behavior and actions; however, the scene information is ignored. Drawing on the idea of the dual-stream model, this paper proposes a human skeleton and scene image-based dual-stream model for human action recognition. The motion features are extracted through the spatio-temporal graph convolution of the human skeleton, and a scene recognition model is proposed based on the sparse frame sampling of video and video-level consensus strategy to process the scene video and gather the visual scene information. The proposed model exploits the advantages of skeleton information in motion expression and the superiority of the image in scene presentation. The scene information and spatio-temporal graph convolution-based human skeleton limbs are fused complementarily to achieve human action recognition. Compared to the conventional optical flow-based dual-stream action recognition method, this model is verified by experimenting under unstable light conditions, and the performance of human action recognition is robust and promising.

Background 5-hydroxytryptophan (5-HTP), the direct biosynthetic precursor of the neurotransmitter 5-hydroxytryptamine, has been shown to have unique efficacy in the treatment of a variety of disorders, including depression, insomnia, and chronic headaches, and is one of the most commercially valuable amino acid derivatives. However, microbial fermentation for 5-HTP production continues to face many challenges, including low titer/yield and the presence of the intermediate L-tryptophan (L-Trp), owing to the complexity and low activity of heterologous expression in prokaryotes. Therefore, there is a need to construct an efficient microbial cell factory for 5-HTP production. Results We describe the systematic modular engineering of wild-type Escherichia coli for the efficient fermentation of 5-HTP from sugars. First, a xylose-induced T7 RNA polymerase-PT7 promoter system was constructed to ensure the efficient expression of each key heterologous pathway in E. coli. Next, a new tryptophan hydroxylase mutant was used to construct an efficient tryptophan hydroxylation module, and the cofactor tetrahydrobiopterin synthesis and regeneration pathway was expressed in combination. The L-Trp synthesis module was constructed by modifying the key metabolic nodes of tryptophan biosynthesis, and the de novo synthesis of 5-HTP was achieved. Finally, the NAD(P)H regeneration module was constructed by the moderate expression of the heterologous GDHesi pathway, which successfully reduced the surplus of the intermediate L-Trp. The final engineered strain HTP11 was able to produce 8.58 g/L 5-HTP in a 5-L bioreactor with a yield of 0.095 g/g glucose and a maximum real-time productivity of 0.48 g/L/h, the highest values reported by microbial fermentation. Conclusions In this study, we demonstrate the successful design of a cell factory for high-level 5-HTP production, combined with simple processes that have potential for use in industrial applications in the future. Thus, this study provides a reference for the production of high-value amino acid derivatives using a systematic modular engineering strategy and a basis for an efficient engineered strain development of 5-HTP high-value derivatives.

Currently, microbial production is becoming a competitive method for N-acetyl-glucosamine production. As the biosynthesis of N-acetyl-glucosamine originating from fructose-6-P directly competes with central carbon metabolism for precursor supply, the consumption of glucose for cell growth and cellular metabolism severely limits the yield of N-acetyl-glucosamine. In this study, appropriate catabolic division of labor in the utilization of mixed carbon sources was achieved by deleting the pfkA gene and enhancing the utilization of glycerol by introducing the glpK mutant. Glycerol thus mainly contributed to cell growth and cellular metabolism, and more glucose was saved for efficient N-acetyl-glucosamine synthesis. By optimizing the ratio of glycerol to glucose, the balancing of cell growth/cellular metabolism and N-acetyl-glucosamine synthesis was achieved. The resulting strain GLALD-7 produced 179.7 g/L N-acetyl-glucosamine using mixed glycerol/glucose (1:8, m/m) carbon sources in a 5 L bioreactor, with a yield of 0.458 g/g total carbon sources (0.529 g/g glucose) and a productivity of 2.57 g/L/h. Coherent high titer/yield/productivity was obtained, with the highest values ever reported, suggesting that an appropriate catabolic division of labor using mixed glycerol/glucose carbon sources is a useful strategy for facilitating the microbial production of chemicals originating from glucose or metabolites upstream of glycolysis.

Aristolochic acid I (AA-I), presenting in a variety of natural medicinal plants, which could cause tubular epithelial cell injury. Curcumin (CUR), a polyphenolic substance isolated from turmeric, is a natural antioxidant. The aim of this experiment was to investigate whether CUR attenuated AA-I-induced renal injury in rats through the SIRT1/Nrf2/HO-1 signaling pathway. SD rats were treated with AA-I (10 mg/kg) or/and CUR (200 mg/kg) for 28 days to assess the protective effect of CUR on AA-I-induced renal injury in vivo. NRK-52E cells were treated with AA-I (40 μ M) or/and CUR (20 μ M) for 24 h in vitro. The intervention pathway of CUR against oxidative stress injury induced by AA-I was assessed by observing pathological changes, oxidative stress status, apoptosis and the expression of SIRT1/Nrf2/HO-1 signaling pathway-related factors. The results showed that AA-I exposure increased the contents of BUN, Cr, KIM-1, NGAL, ALT and AST in serum. It increased the content of MDA, decreased the activities of SOD, GST, GSH and the content of ATP in renal tissue. Pathological changes such as inflammatory cell infiltration and mitochondrial injury occurred in renal tissue. AA-I exposure resulted in a substantial rise in the levels of BAX, Ccaspase-9, Cleaved Caspase-9, Caspase-3, Cleaved Caspase-3 and a significant decrease in mRNA and protein expression levels of Bcl-2, SIRT1, Nrf2, NQO1, HO-1 and Keap1. However, these changes were reversed by CUR intervention. In summary, AA-I exposure caused mitochondrial dysfunction and triggered apoptosis through the oxidative stress pathway. However, CUR could reduce AA-I-induced renal injury by activating the SIRT1/Nrf2/HO-1 signaling pathway.

To improve fermentation efficiency during L-tyrosine production, we established a novel method to produce L-tyrosine by repeated batch fermentation through cell recycling. According to the characteristics of the fermented broth, the cells were recovered using two dish centrifuges in series. First, L-tyrosine crystals were removed from the previous batch of fermented broth by selective centrifugation. The cells were then concentrated in another centrifuge and recycled for use in the next cycle of batch fermentation. L-tyrosine removal and cell recovery rates were 86% and 67%, respectively, in a centrifuge bioreactor. Through repeated batch fermentation, L-tyrosine productivity was increased by up to 44.1% compared with that for fed-batch fermentation, and the coupling of fermentation and separation was achieved, which is useful in reducing production costs. This study provides a rationale for the fermentation of low-solubility products by repeated batch or cell-recycle fermentation.

Most of the current industrial processes for L-leucine production are based on fermentation, usually in fed-batch operation mode. Although the culture technology has advanced in recent decades, the process still has significant drawbacks. To solve these problems, we investigated the effects of chemostat culture conditions on the production of L-leucine by Corynebacterium glutamicum CP. The dilution rate, the nitrogen source, and the carbon–nitrogen ratio of the medium were optimized. With the addition of ammonium acetate to the chemostat medium, the initial C/N ratio was adjusted to 57.6, and the L-leucine titer reached the highest level at the optimal dilution rate of 0.04 h−1. Compared with fed-batch culture, the L-leucine titer was reduced (from 53.0 to 24.8 g L−1), but the yield from glucose was increased by 10.0% (from 0.30 to 0.33 mol mol−1) and productivity was increased by 58.3% (from 1.2 to 1.9 g L−1 h−1). Moreover, the titer of the by-product L-alanine was significantly reduced (from 8.9 to 0.8 g L−1). In addition, gene expression levels and activity of key enzymes in the synthesis of L-leucine and L-alanine were analyzed to explain the difference of production performance between chemostat culture and fed-batch culture. The results indicate that chemostat culture has great potential to increase the industrial production of L-leucine compared to current fed-batch approaches., Graphical abstract

The development of organic room-temperature phosphorescence (RTP) materials has been a rich research field. However, there are still limited practical and directive strategies to regulate the phosphorescence emission. Herein, six naphthalene-based thiochromanone analogs with different heterocycles were synthesized. And all these molecules exhibited bright phosphorescence when embedded into a polyvinyl alcohol matrix under ambient conditions. The absorption spectra showed a distinct red-shift as the size of the heterocycle shrunk, which resulted from the facilitated p–π conjugation. In particular, the absorption spectra of molecules with a five-membered heterocycle were extended to visible light, which is rarely reported for RTP materials. These results might present a new strategy to adjust the photophysical properties of phosphors to obtain phosphorescence materials with the desired excitation and emission wavelength.

There have been several attempts to integrate a spatial visual attention mechanism into an image caption model and introduce semantic concepts as the guidance of image caption generation. High-level semantic information consists of the abstractedness and generality indication of an image, which is beneficial to improve the model performance. However, the high-level information is always static representation without considering the salient elements. Therefore, a semantic attention mechanism is used for the high-level information instead of conventional of static representation in this article. The salient high-level semantic information can be considered as redundant semantic information for image caption generation. Additionally, the generation of visual words and non-visual words can be separated, and an adaptive attention mechanism is employed to realize the guidance information of image caption generation switching between new fusion information (fusion of image feature and high-level semantics) and a language model. Therefore, visual words can be generated according to the image features and high-level semantic information, and non-visual words can be predicted by the language model. The semantics attention, adaptive attention, and previous generated words are fused to construct a special attention module for the input and output of long short-term memory. An image caption can be generated as a concise sentence on the basis of accurately grasping the rich content of the image. The experimental results show that the performance of the proposed model is promising for the evaluation metrics, and the captions can achieve logical and rich descriptions.