Highly Efficient Biosynthesis of Rebaudioside M8 through Structure-Guided Engineering of Glycosyltransferase UGT94E13.

Given the low-calorie, high-sweetness characteristics of steviol glycosides (SGs), developing SGs with improved taste profiles is a key focus. Rebaudioside M8 (Reb M8), a novel non-natural SG derivative obtained through glycosylation at the C-13 position of rebaudioside D (Reb D) using glycosyltransferase UGT94E13, holds promise for further development due to its enhanced sweetness. However, the low catalytic activity of UGT94E13 hampers further research and commercialization. This study aimed to improve the enzymatic activity of UGT94E13 through semirational design, and a variant UGT94E13-F169G/I185G was obtained with the catalytic activity improved by 13.90 times. A cascade reaction involving UGT94E13-F169G/I185G and sucrose synthase AtSuSy was established to recycle uridine diphosphate glucose, resulting in an efficient preparation of Reb M8 with a yield of 98%. Moreover, according to the analysis of the distances between the substrate Reb D and enzymes as well as between Reb D and the glucose donor through molecular dynamics simulations, it is found that the positive effect of shortening the distance on glycosylation reaction activity accounts for the improved catalytic activity of UGT94E13-F169G/I185G. Therefore, this study addresses the bottleneck in the efficient production of Reb M8 and provides a foundation for its widespread application in the food industry.

[Determination of 12 prohibited veterinary drug residues in pig urine by ultra high performance liquid chromatography-tandem mass spectrometry].

A method was established for the simultaneous detection of 12 prohibited veterinary drugs, including ß2-receptor agonists, nitrofuran metabolites, nitroimidazoles, chlorpromazine, and chloramphenicol, in pig urine. The sample was pretreated by enzymolysis, acid hydrolysis/derivatization, and liquid-liquid extraction combined with solid-phase extraction. Detection was performed using ultra high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). Ammonium acetate solution (0.2 mol/L, 4.5 mL) and ß-glucuronidase/aryl sulfatase (40 µL) were added to the sample, which was subsequently enzymolized at 37 ℃ for 2 h. Then, 1.5 mL of 1.0 mol/L hydrochloric acid solution and 100 µL of 0.1 mol/L o-nitrobenzaldehyde solution were added to the sample. The mixture was incubated at 37 ℃ for 16 h, and the analytes were extracted with 8 mL of ethyl acetate by liquid-liquid extraction. The lower aqueous phase obtained after extraction was extracted and purified using a mixed cation-exchange solid-phase extraction column. The extracts were combined, the extraction solution was blow-dried with nitrogen, and the residue was redissolved for determination. The samples were analyzed under multiple-reaction monitoring mode with both positive and negative electrospray ionization, and quantified using an isotope internal standard method. The correlation coefficients (r) of the 12 compounds were >0.99. The limits of detection (LODs) and quantification (LOQs) of chloramphenicol were 0.05 and 0.1 µg/L, respectively, and the LODs and LOQs of the other compounds were 0.25 and 0.5 µg/L, respectively. The mean recoveries and RSDs at 1, 2, and 10 times the LOQ were 83.6%-115.3% and 2.20%-12.34%, respectively. The proposed method has the advantages of high sensitivity, good stability, and accurate quantification; thus, it is suitable for the simultaneous determination of the 12 prohibited veterinary drug residues in pig urine.

Discovery of a Fungal P450 with an Unusual Two-Step Mechanism for Constructing a Bicyclo[3.2.2]nonane Skeleton.

The successful biomimetic or chemoenzymatic synthesis of target natural products (NPs) and their derivatives relies on enzyme discovery. Herein, we discover a fungal P450 BTG5 that can catalyze the formation of a bicyclo[3.2.2]nonane structure through an unusual two-step mechanism of dimerization and cyclization in the biosynthesis of beticolin 1, whose bicyclo[3.2.2]nonane skeleton connects an anthraquinone moiety and a xanthone moiety. Further investigation reveals that BTG5-T318 not only determines the substrate selectivity but also alters the catalytic reactions, which allows the separation of the reaction to two individual steps, thereby understanding its catalytic mechanism. It reveals that the first heterodimerization undergoes the common oxidation process for P450s, while the second uncommon formal redox-neutral cyclization step is proved as a redox-mediated reaction, which has never been reported. Therefore, this work advances our understanding of P450-catalyzed reactions and paves the way for expansion of the diversity of this class of NPs through synthetic biology.

Fluorophore Label-Free Light-up Near Infrared Deoxyribonucleic Acid Nanosensor for Monitoring Extracellular Potassium Levels.

Fluorescent DNA nanosensors have been widely used due to their unique advantages, among which the near-infrared (NIR) imaging mode can provide deeper penetration depth and lower biological background for the nanosensors. However, efficient NIR quenchers require ingenious design, complex synthesis, and modification, which severely limit the development of NIR DNA nanosensors. Label-free strategies based on G-quadruplex (G4) and NIR G4 dyes were first introduced into in situ extracellular imaging, and a novel NIR sensing strategy for the specific detection of extracellular targets is proposed. The strategy avoids complex synthesis and site-specific modification by controlling the change of the NIR signal through the formation of a G4 nanostructure. A light-up NIR DNA nanosensor based on potassium ion (K+)-sensitive G4 chain PS2.M was constructed to verify the strategy. PS2.M forms a stable G4 nanostructure in the presence of K+ and activates the NIR G4 dye CSTS, thus outputting NIR signals. The nanosensor can rapidly respond to K+ with a linear range of 5-50 mM and has good resistance to interference. The nanosensor with cholesterol can provide feedback on the changes in extracellular K+ concentration in many kinds of cells, serving as a potential tool for the study of diseases such as epilepsy and cancer, as well as the development of related drugs. The strategy can be potentially applied to the NIR detection of a variety of extracellular targets with the help of functional DNAs such as aptamer and DNAzyme.

Selective synthesis of rebaudioside M2 through structure-guided engineering of glycosyltransferase UGT94D1.

Rebaudioside M2 (Reb M2), a novel steviol glycoside derivative, has limited industrial applications due to its low synthetic yield and selectivity. Herein, we identify UGT94D1 as a selective glycosyltransferase for rebaudioside D (Reb D), leading to the production of a mono ß-1,6-glycosylated derivative, Reb M2. A variant UGT94D1-F119I/D188P was developed through protein engineering. This mutant exhibited a 6.33-fold improvement in catalytic efficiency, and produced Reb M2 with 92% yield. Moreover, molecular dynamics simulations demonstrated that UGT94D1-F119I/D188P exhibited a shorter distance between the nucleophilic oxygen (OH6) of the substrate Reb D and uridine diphosphate glucose, along with an increased Ophosphate-C1-Oacceptor angle, thus improving the catalytic activity of the enzyme. Therefore, this study provides an efficient method for the selective synthesis of Reb M2 and paves the way for its applications in various fields.

Modular assembly of an artificially concise biocatalytic cascade for the manufacture of phenethylisoquinoline alkaloids.

Plant-derived alkaloids are an important class of pharmaceuticals. However, they still rely on phytoextraction to meet their diverse market demands. Since multistep biocatalytic cascades have begun to revolutionize the manufacture of natural or unnatural products, to address the synthetic challenges of alkaloids, herein we establish an artificially concise four-enzyme biocatalytic cascade with avoiding plant-derived P450 modification for synthesizing phenethylisoquinoline alkaloids (PEIAs) after enzyme discovery and enzyme engineering. Efficient biosynthesis of diverse natural and unnatural PEIAs is realized from readily available substrates. Most importantly, the scale-up preparation of the colchicine precursor (S)-autumnaline with a high titer is achieved after replacing the rate-limiting O-methylation by the plug-and-play strategy. This study not only streamlines future engineering endeavors for colchicine biosynthesis, but also provides a paradigm for constructing more artificial biocatalytic cascades for the manufacture of diverse alkaloids through synthetic biology.

Synthetic Biology-based Construction of Unnatural Perylenequinones with Improved Photodynamic Anticancer Activities.

The construction of structural complexity and diversity of natural products is crucial for drug discovery and development. To overcome high dark toxicity and poor photostability of natural photosensitizer perylenequinones (PQs) for photodynamic therapy, herein, we aim to introduce the structural complexity and diversity to biosynthesize the desired unnatural PQs in fungus Cercospora through synthetic biology-based strategy. Thus, we first elucidate the intricate biosynthetic pathways of class B PQs and reveal how the branching enzymes create their structural complexity and diversity from a common ancestor. This enables the rational reprogramming of cercosporin biosynthetic pathway in Cercospora to generate diverse unnatural PQs without chemical modification. Among them, unnatural cercosporin A displays remarkably low dark toxicity and high photostability with retention of great photodynamic anticancer and antimicrobial activities. Moreover, it is found that, unlike cercosporin, unnatural cercosporin A could be selectively accumulated in cancer cells, providing potential targets for drug development. Therefore, this work provides a comprehensive foundation for preparing unnatural products with customized functions through synthetic biology-based strategies, thus facilitating drug discovery pipelines from nature.

A Parallel Framework for Streaming Dimensionality Reduction.

The visualization of streaming high-dimensional data often needs to consider the speed in dimensionality reduction algorithms, the quality of visualized data patterns, and the stability of view graphs that usually change over time with new data. Existing methods of streaming high-dimensional data visualization primarily line up essential modules in a serial manner and often face challenges in satisfying all these design considerations. In this research, we propose a novel parallel framework for streaming high-dimensional data visualization to achieve high data processing speed, high quality in data patterns, and good stability in visual presentations. This framework arranges all essential modules in parallel to mitigate the delays caused by module waiting in serial setups. In addition, to facilitate the parallel pipeline, we redesign these modules with a parametric non-linear embedding method for new data embedding, an incremental learning method for online embedding function updating, and a hybrid strategy for optimized embedding updating. We also improve the coordination mechanism among these modules. Our experiments show that our method has advantages in embedding speed, quality, and stability over other existing methods to visualize streaming high-dimensional data.

Photoenzymatic Enantioselective Synthesis of Oxygen-Containing Benzo-Fused Heterocycles.

New-to-nature biocatalysis in organic synthesis has recently emerged as a green and powerful strategy for the preparation of valuable chiral products, among which chiral oxygen-containing benzo-fused heterocycles are important structural motifs in pharmaceutical industry. However, the asymmetric synthesis of these compounds through radical-mediated methods is challenging. Herein, a novel asymmetric radical-mediated photoenzymatic synthesis strategy is developed to realize the efficient enantioselective synthesis of oxygen-containing benzo-fused heterocycles through structure-guided engineering of a flavin-dependent 'ene'-reductase GluER. It shows that variant GluER-W100H could efficiently produce various benzoxepinones, chromanone and indanone with different benzo-fused rings in high yields with great stereoselectivities under visible light. Moreover, these results are well supported by mechanistic experiments, revealing that this photoenzymatic process involves electron donor-acceptor complex formation, single electron transfer and hydrogen atom transfer. Therefore, we provide an alternative green approach for efficient chemoenzymatic synthesis of important chiral skeletons of bioactive pharmaceuticals.

Hairpin Switches-Based Isothermal Transcription Amplification for Simple, Sensitivity Detection of MicroRNA.

The ability to simply, selectively, and sensitively detect low numbers of miRNAs in clinical samples is highly valuable but remains a challenge. In this work, we present a novel miRNA detection system by using the elaborately designed hairpin switch, where the T7 primer, template, target recognize sequence, and light-up RNA aptamer template are edited and embedded in one single-stranded DNA hairpin structure. In the beginning, the hairpin switch maintained the hairpin structure 1, in which the ds promoter of T7 polymerase was disrupted, thus the transcription reaction of T7 polymerase was inhibited. After binding to the target, the hairpin switch 1 was unfolded and turned to the hairpin structure 2. This switch initiates the in vitro T7 transcription reaction, producing plenty of RNA transcripts containing RNA aptamers. Consequently, transcribed tremendous RNA aptamers lighted up the fluorophore for quantitative analysis. Compared with the existing T7 polymerase-based amplification system, this strategy exhibits several advantages, including simplicity, convenience, and high selectivity and sensitivity. The experimental results demonstrated that we could achieve the quantification of miRNA in buffer and complex biological samples.

Structural Insight into the Catalytic Mechanisms of an L-Sorbosone Dehydrogenase.

L-Sorbosone dehydrogenase (SNDH) is a key enzyme involved in the biosynthesis of 2-keto-L-gulonic acid , which is a direct precursor for the industrial scale production of vitamin C. Elucidating the structure and the catalytic mechanism is essential for improving SNDH performance. By solving the crystal structures of SNDH from Gluconobacter oxydans WSH-004, a reversible disulfide bond between Cys295 and the catalytic Cys296 residues is discovered. It allowed SNDH to switch between oxidation and reduction states, resulting in opening or closing the substrate pocket. Moreover, the Cys296 is found to affect the NADP+ binding pose with SNDH. Combining the in vitro biochemical and site-directed mutagenesis studies, the redox-based dynamic regulation and the catalytic mechanisms of SNDH are proposed. Moreover, the mutants with enhanced activity are obtained by extending substrate channels. This study not only elucidates the physiological control mechanism of the dehydrogenase, but also provides a theoretical basis for engineering similar enzymes.

CRISPR-Cas12a coupled with cyclic reverse transcription for amplified detection of miRNA.

In this work, we present a highly sensitive, specific, and versatile method to quantify miRNA expression by coupling CRISPR-Cas12a with cyclic reverse transcription (CRT), termed as CRISPR-CRT. Each miRNA target was first converted and amplified into multiple hairpin RT products via CRT. Afterward, the hairpin RT products could serve as activators to initiate the collateral cleavage activity of CRISPR-Cas12a. Due to the above two-stage amplification, this assay could detect miRNA at sub-femtomolar level (LOD, 0.201 fM). Since the sequence of target miRNA is double checked: first in the CRT and then in the CRISPR system, the proposed assay also shows an excellent specificity in detecting miR-21. Finally, with the usage of this assay, the sensitive assessment of miR-21 levels in human serum samples has been achieved and the disease human serum has been detected. Conclusively, CRISPR-CRT holds a great application prospective in the field of clinical molecular diagnosis.

Bone marrow mesenchymal stem cell-derived exosomes attenuate the maturation of dendritic cells and reduce the rejection of allogeneic transplantation.

BACKGROUND: Bone mesenchymal stem cell (BMSC)-derived exosomes (B-exos) are attractive for applications in enabling alloantigen tolerance. An in-depth mechanistic understanding of the interaction between B-exos and dendritic cells (DCs) could lead to novel cell-based therapies for allogeneic transplantation. OBJECTIVES: To examine whether B-exos exert immunomodulatory effects on DC function and maturation. MATERIAL AND METHODS: After mixed culture of BMSCs and DCs for 48 h, DCs from the upper layer were collected to analyze the expression levels of surface markers and mRNAs of inflammation-related cytokines. Then, before being collected to detect the mRNA and protein expression levels of indoleamine 2,3-dioxygenase (IDO), the DCs were co-cultured with B-exos. Then, the treated DCs from different groups were co-cultured with naïve CD4+ T cells from the mouse spleen. The proliferation of CD4+ T cells and the proportion of CD4+CD25+Foxp3+ T cells were analyzed. Finally, the skins of BALB/c mice were transplanted to the back of C57 mice in order to establish a mouse allogeneic skin transplantation model. RESULTS: The co-culture of DCs with BMSCs downregulated the expression of the major histocompatibility complex class II (MHC-II) and CD80/86 costimulatory molecules on DCs. Moreover, B-exos increased the expression of IDO in DCs treated with lipopolysaccharide (LPS). The proliferation of CD4+CD25+Foxp3+ T cells increased when cultured with B-exos-exposed DCs. Finally, mice recipients injected with B-exos-treated DCs had significantly prolonged survival after receiving the skin allograft. CONCLUSIONS: Taken together, these data suggest that the B-exos suppress the maturation of DCs and increase the expression of IDO, which might shed light on the role of B-exos in inducing alloantigen tolerance.

Self-assembly of DNA nanospheres with controllable size and self-degradable property for enhanced antitumor chemotherapy.

Controllable size, self-degradability and targeting property are important for a precise improvement of anticancer effects and reduction of side effects of drug vehicles. Here, a series of DNA nanospheres with controllable size and self-degradation ability were constructed through the hybridization of two i-motif strands and two linker strands for targeted cancer therapy. DNA nanospheres with different sizes were fabricated by regulating the linker sequence, and their pH-responsive self-degradation property was realized by the introduction of the i-motif strand. Moreover, the ZY11 aptamer was introduced to endow the DNA nanospheres with targeting property toward SMMC-7721 cancer cells. The results revealed that the appropriate size of DNA nanospheres (80 nm) highly promoted the internalization by mammalian cells. The results of DLS, AFM and CD spectra showed that the DNA nanospheres were stable in a physiological environment but they self-degraded in a slightly acidic environment due to the existence of the i-motif strand. Moreover, the fluorescence of DOX@AP-NSs2 was triple at pH = 5.0 than at pH = 7.4, which further confirmed the pH-responsive drug release performance. The above results proved that the use of DOX@AP-NSs2 is a promising approach to accelerate the rapid release of drugs into the tumors and avoid drug leakage into the normal tissue. The results at a cellular level and in vivo confirmed the pH-responsive targeted antitumor effect. Hence, the novel DNA nanospheres with controllable size and self-degradable property represent a potential tool for targeted drug delivery and cancer therapy.

Structural analysis of an anthrol reductase inspires enantioselective synthesis of enantiopure hydroxycycloketones and ß-halohydrins.

Asymmetric reduction of prochiral ketones, particularly, reductive desymmetrization of 2,2-disubstituted prochiral 1,3-cyclodiketones to produce enantiopure chiral alcohols is challenging. Herein, an anthrol reductase CbAR with the ability to accommodate diverse bulky substrates, like emodin, for asymmetric reduction is identified. We firstly solve crystal structures of CbAR and CbAR-Emodin complex. It reveals that Tyr210 is critical for emodin recognition and binding, as it forms a hydrogen-bond interaction with His162 and π-π stacking interactions with emodin. This ensures the correct orientation for the stereoselectivity. Then, through structure-guided engineering, variant CbAR-H162F can convert various 2,2-disubstituted 1,3-cyclodiketones and α-haloacetophenones to optically pure (2S, 3S)-ketols and (R)-ß-halohydrins, respectively. More importantly, their stereoselectivity mechanisms are also well explained by the respective crystal structures of CbAR-H162F-substrate complex. Therefore, this study demonstrates that an in-depth understanding of catalytic mechanism is valuable for exploiting the promiscuity of anthrol reductases to prepare diverse enantiopure chiral alcohols.

EEG-Based Seizure Prediction via Model Uncertainty Learning.

Deep neural networks (DNNs) have the powerful ability to automatically extract efficient features, which makes them prominent in electroencephalogram (EEG) based seizure prediction tasks. However, current research in this field cannot take the model uncertainty into account, causing the prediction less credible. To this end, we introduce a novel end-to-end patient-specific seizure prediction framework via model uncertainty learning. Specifically, we propose a reparameterized EEG-based lightweight CNN architecture and a modified Monte Carlo dropout (RepNet-MMCD) strategy to improve the reliability of the DNNs-based model. In RepNet, we obtain multi-scale feature representations by applying depthwise separable convolutions of different kernels. After training, depthwise convolutions with different scales are equivalently converted into a single convolution layer, which can greatly reduce computational budgets without losing model performance. In addition, we propose a modified Monte Carlo (MMCD) strategy, leveraging the samples-based temporal information in EEG signals to simulate the Monte Carlo dropout sampling. Sensitivity, false-positive rate (FPR), and area under curve (AUC) of the proposed RepNet-MMCD achieve 93.1%, 0.033/h, 0.950 and 81.6%, 0.056/h, 0.903 on two public datasets, respectively. We further extend the MMCD strategy to the other baseline methods, which can improve the performance of seizure prediction by a clear margin.

Shape-Aware 3D Small Vessel Segmentation with Local Contrast Guided Attention.

The automated segmentation and analysis of small vessels from in vivo imaging data is an important task for many clinical applications. While current filtering and learning methods have achieved good performance on the segmentation of large vessels, they are sub-optimal for small vessel detection due to their apparent geometric irregularity and weak contrast given the relatively limited resolution of existing imaging techniques. In addition, for supervised learning approaches, the acquisition of accurate pixel-wise annotations in these small vascular regions heavily relies on skilled experts. In this work, we propose a novel self-supervised network to tackle these challenges and improve the detection of small vessels from 3D imaging data. First, our network maximizes a novel shape-aware flux-based measure to enhance the estimation of small vasculature with non-circular and irregular appearances. Then, we develop novel local contrast guided attention(LCA) and enhancement(LCE) modules to boost the vesselness responses of vascular regions of low contrast. In our experiments, we compare with four filtering-based methods and a state-of-the-art self-supervised deep learning method in multiple 3D datasets to demonstrate that our method achieves significant improvement in all datasets. Further analysis and ablation studies have also been performed to assess the contributions of various modules to the improved performance in 3D small vessel segmentation. Our code is available at https://github.com/dengchihwei/LCNetVesselSeg.

Highly efficient production of rebaudioside D enabled by structure-guided engineering of bacterial glycosyltransferase YojK.

Owing to zero-calorie, high-intensity sweetness and good taste profile, the plant-derived sweetener rebaudioside D (Reb D) has attracted great interest to replace sugars. However, low content of Reb D in stevia rebaudiana Bertoni as well as low soluble expression and enzymatic activity of plant-derived glycosyltransferase in Reb D preparation restrict its commercial usage. To address these problems, a novel glycosyltransferase YojK from Bacillus subtilis 168 with the ability to glycosylate Reb A to produce Reb D was identified. Then, structure-guided engineering was performed after solving its crystal structure. A variant YojK-I241T/G327N with 7.35-fold increase of the catalytic activity was obtained, which allowed to produce Reb D on a scale preparation with a great yield of 91.29%. Moreover, based on the results from molecular docking and molecular dynamics simulations, the improvement of enzymatic activity of YojK-I241T/G327N was ascribed to the formation of new hydrogen bonds between the enzyme and substrate or uridine diphosphate glucose. Therefore, this study provides an engineered bacterial glycosyltransferase YojK-I241T/G327N with high solubility and catalytic efficiency for potential industrial scale-production of Reb D.

Full-Length Transcriptome Sequencing Combined with RNA-Seq to Analyze Genes Related to Terpenoid Biosynthesis in Cinnamomum burmannii.

Cinnamomum burmannii is a cinnamomum plant rich in natural D-borneol. Natural D-borneol is a bicycle monoterpenoid compound widely used in the food, pharmaceutical, and cosmetic industries. Therefore, analyzing the biosynthesis mechanism of natural D-borneol in C. burmannii at the molecular level is helpful for directional breeding in the future and further development and utilization of C. burmannii and its related gene resources. In our study, 76 genes related to terpene metabolism were analyzed through third-generation sequencing and second-generation sequencing. Of these genes, 57 were associated with the synthesis of the terpenoid skeleton, and 19 belonged to terpenoid synthase, including four monoterpenoid synthases, seven sesquiterpenoid synthases, and eight diterpenoid synthases. Two genes in diterpenoid synthase were differentially expressed in high D-borneol and low D-borneol plants. It was speculated that these two genes might be related to D-borneol synthesis. How these two genes participate in the synthesis of D-borneol needs further study.