Single-Molecule Identification and Quantification of Steviol Glycosides with a Deep Learning-Powered Nanopore Sensor.

Steviol glycosides (SGs) are a class of high-potency noncalorie natural sweeteners made up of a common diterpenoid core and varying glycans. Thus, the diversity of glycans in composition, linkage, and isomerism results in the tremendous structural complexity of the SG family, which poses challenges for the precise identification and leads to the fact that SGs are frequently used in mixtures and their variances in biological activity remain largely unexplored. Here we show that a wild-type aerolysin nanopore can detect and discriminate diverse SG species through the modulable electro-osmotic flow effect at varied applied voltages. At low voltages, the neutral SG molecule was drawn and stuck in the pore entrance due to an energy barrier around R220 sites. The ensuing binding events enable the identification of the majority of SG species. Increasing the voltage can break the barrier and cause translocation events, allowing for the unambiguous identification of several pairs of SGs differing by only one hydroxyl group through recognition accumulation from multiple sensing regions and sites. Based on nanopore data of 15 SGs, a deep learning-based artificial intelligence (AI) model was created to process the individual blockage events, achieving the rapid, automated, and precise single-molecule identification and quantification of SGs in real samples. This work highlights the value of nanopore sensing for precise structural analysis of complex glycans-containing glycosides, as well as the potential for sensitive and rapid quality assurance analysis of glycoside products with the use of AI.

Precise Structural Analysis of Neutral Glycans Using Aerolysin Mutant T240R Nanopore.

Glycans play vital roles in nearly all life processes of multicellular organisms, and understanding these activities is inseparable from elucidating the biological significance of glycans. However, glycan research has lagged behind that of DNA and protein due to the challenges posed by structural heterogeneity and isomerism (i.e., structures with equal molecular weights) the lack of high-efficiency structural analysis techniques. Nanopore technology has emerged as a sensitive single-molecule biosensor, shining a light on glycan analysis. However, a significant number of glycans are small and uncharged, making it challenging to elicit identifiable nanopore signals. Here we introduce a R-binaphthyl tag into glycans, which enhances the cation-π interaction between the derivatized glycan molecules and the nanopore interface, enabling the detection of neutral glycans with an aerolysin nanopore. This approach allows for the distinction of di-, tri-, and tetrasaccharides with monosaccharide resolution and has the potential for group discrimination, the monitoring of enzymatic transglycosylation reactions. Notably, the aerolysin mutant T240R achieves unambiguous identification of six disaccharide isomers, trisaccharide and tetrasaccharide linkage isomers. Molecular docking simulations reveal that multiple noncovalent interactions occur between residues R282, K238, and R240 and the glycans and R-binaphthyl tag, significantly slowing down their translocation across the nanopore. Importantly, we provide a demonstration of the kinetic translocation process of neutral glycan isomers, establishing a solid theoretical foundation for glycan nanopore analysis. The development of our technology could promote the analysis of glycan structural isomers and has the potential for nanopore-based glycan structural determination and sequencing.

Identification of tagged glycans with a protein nanopore.

Structural complexity of glycans derived from the diversities in composition, linage, configuration, and branching considerably complicates structural analysis. Nanopore-based single-molecule sensing offers the potential to elucidate glycan structure and even sequence glycan. However, the small molecular size and low charge density of glycans have restricted direct nanopore detection of glycan. Here we show that glycan sensing can be achieved using a wild-type aerolysin nanopore by introducing a facile glycan derivatization strategy. The glycan molecule can induce impressive current blockages when moving through the nanopore after being connected with an aromatic group-containing tag (plus a carrier group for the neutral glycan). The obtained nanopore data permit the identification of glycan regio- and stereoisomers, glycans with variable monosaccharide numbers, and distinct branched glycans, either independently or with the use of machine learning methods. The presented nanopore sensing strategy for glycans paves the way towards nanopore glycan profiling and potentially sequencing.

Blending Technology Based on HPLC Fingerprint and Nonlinear Programming to Control the Quality of Ginkgo Leaves.

The breadth and depth of traditional Chinese medicine (TCM) applications have been expanding in recent years, yet the problem of quality control has arisen in the application process. It is essential to design an algorithm to provide blending ratios that ensure a high overall product similarity to the target with controlled deviations in individual ingredient content. We developed a new blending algorithm and scheme by comparing different samples of ginkgo leaves. High-consistency samples were used to establish the blending target, and qualified samples were used for blending. Principal component analysis (PCA) was used as the sample screening method. A nonlinear programming algorithm was applied to calculate the blending ratio under different blending constraints. In one set of calculation experiments, the result was blended by the same samples under different conditions. Its relative deviation coefficients (RDCs) were controlled within ±10%. In another set of calculations, the RDCs of more component blending by different samples were controlled within ±20%. Finally, the near-critical calculation ratio was used for the actual experiments. The experimental results met the initial setting requirements. The results show that our algorithm can flexibly control the content of TCMs. The quality control of the production process of TCMs was achieved by improving the content stability of raw materials using blending. The algorithm provides a groundbreaking idea for quality control of TCMs.

Rapid Discrimination and Prediction of Ginsengs from Three Origins Based on UHPLC-Q-TOF-MS Combined with SVM.

Ginseng, which contains abundant ginsenosides, grows mainly in the Jilin, Liaoning, and Heilongjiang in China. It has been reported that the quality and traits of ginsengs from different origins were greatly different. To date, the accurate prediction of the origins of ginseng samples is still a challenge. Here, we integrated ultra-high-performance liquid chromatography quadrupole time-of-flight mass spectrometry (UHPLC-Q-TOF-MS) with a support vector machine (SVM) for rapid discrimination and prediction of ginseng from the three main regions where it is cultivated in China. Firstly, we develop a stable and reliable UHPLC-Q-TOF-MS method to obtain robust information for 31 batches of ginseng samples after reasonable optimization. Subsequently, a rapid pre-processing method was established for the rapid screening and identification of 69 characteristic ginsenosides in 31 batches ginseng samples from three different origins. The SVM model successfully distinguished ginseng origin, and the accuracy of SVM model was improved from 83% to 100% by optimizing the normalization method. Six crucial quality markers for different origins of ginseng were screened using a permutation importance algorithm in the SVM model. In addition, in order to validate the method, eight batches of test samples were used to predict the regions of cultivation of ginseng using the SVM model based on the six selected quality markers. As a result, the proposed strategy was suitable for the discrimination and prediction of the origin of ginseng samples.

Mechanism deconvolution of Qing Fei Pai Du decoction for treatment of Coronavirus Disease 2019 (COVID-19) by label-free integrative pharmacology assays.

ETHNOPHARMACOLOGICAL RELEVANCE: Traditional Chinese medicine (TCM) has a long history in the prevention and treatment of pandemics. The TCM formula Lung Cleansing and Detoxifying Decoction (LCDD), also known as Qing Fei Pai Du Decoction, has been demonstrated effective against Coronavirus Disease 2019 (COVID-19). AIM OF THE STUDY: This work aimed to elucidate the active ingredients, targets and pathway mechanism of LCDD related to suppression of inflammatory, immunity regulation and relaxation of airway smooth muscle for the treatment of COVID-19. MATERIALS AND METHODS: Mining chemical ingredients reported in LCDD, 144 compounds covering all herbs were selected and screened against inflammatory-, immunity- and respiratory-related GPCRs including GPR35, H1, CB2, B2, M3 and ß2-adrenoceptor receptor using a label-free integrative pharmacology method. Further, all active compounds were detected using liquid chromatography-tandem mass spectrometry, and an herb-compound-target network based on potency and content of compounds was constructed to elucidate the multi-target and synergistic effect. RESULTS: Thirteen compounds were identified as GPR35 agonists, including licochalcone B, isoliquiritigenin, etc. Licochalcone B, isoliquiritigenin and alisol A exhibited bradykinin receptor B2 antagonism activities. Atractyline and shogaol showed as a cannabinoid receptor CB2 agonist and a histamine receptor H1 antagonist, respectively. Tectorigenin and aristofone acted as muscarinic receptor M3 antagonists, while synephrine, ephedrine and pseudoephedrine were ß2-adrenoceptor agonists. Pathway deconvolution assays suggested activation of GPR35 triggered PI3K, MEK, JNK pathways and EGFR transactivation, and the activation of ß2-adrenoceptor mediated MEK and Ca2+. The herb-compound-target network analysis found that some compounds such as licochalcone B acted on multiple targets, and multiple components interacted with the same target such as GPR35, reflecting the synergistic mechanism of Chinese medicine. At the same time, some low-abundance compounds displayed high target activity, meaning its important role in LCDD for anti-COVID-19. CONCLUSIONS: This study elucidates the active ingredients, targets and pathways of LCDD. This is useful for elucidating multitarget synergistic action for its clinical therapeutic efficacy.