Rongchun Han

Coprinellus radians has been identified as an important germination-associated fungus of Cremastra appendiculata. However, no reference genome was previously available for this species. In this study, we generated a complete telomere-to-telomere (T2T) genome assembly for C. radians by integrating PacBio HiFi, Oxford Nanopore, BGI short-read, and Hi-C sequencing data. Hi-C interaction mapping anchored 99.36% of the assembled sequences onto 13 pseudochromosomes. Following contaminant removal, gap closure using TGS-GapCloser, and telomere extension, the final assembly consists of 13 gap-free, single-contig chromosomes with telomeric repeats (CCCTAA/TTAGGG) at both ends, yielding a total of 26 telomeres. The resulting high-quality genome is 41.09 Mb (41,092,644 bp) in size, with a contig N50 of 3.36 Mb and a BUSCO completeness score of 96.4%. This gap-free, T2T genome provides a valuable resource for investigating the molecular mechanisms of orchid-fungus symbiosis and the ecological adaptation of wood-decaying fungi.

Psychobiotics are emerging microbiota-targeted interventions with potential relevance for mood, stress resilience, anxiety, sleep, cognition, and neuroimmune regulation. Although initially viewed mainly as probiotic strains acting through the microbiota–gut–brain axis, the concept is now expanding toward food-associated systems integrating live microorganisms, fermented matrices, prebiotic substrates, postbiotic fractions, and neuroactive metabolites. This review examines psychobiotic foods as integrated strain–matrix–metabolite–host systems, emphasizing fermentation as both a source of candidate microorganisms and a biotechnological process generating gut–brain-relevant bioactives. Fermented dairy products, kefir, sourdough, cereals, legumes, vegetables, fruits, tea, cocoa, coffee, and plant-based products may provide strains and metabolites such as γ-aminobutyric acid, short-chain fatty acids, tryptophan derivatives, bioactive peptides, exopolysaccharides, organic acids, vitamins, and phenolic metabolites. These compounds may influence gut–brain communication through microbial ecology, vagal signaling, HPA-axis regulation, immune modulation, intestinal and blood–brain barrier integrity, oxidative-stress control, and enteroendocrine pathways. However, evidence remains heterogeneous, with stronger support for depressive symptoms than for anxiety, sleep, cognition, or other psychiatric outcomes. Future development requires strain authentication, genome-based safety screening, controlled fermentation, metabolomics, simulated digestion, undesirable metabolite monitoring, and well-powered human trials linking food composition to validated mental-health-related endpoints.

Haizhixiaolin Capsule (HZXL), a traditional Chinese medicine hospital preparation for treating chronic prostatitis and urinary disorders, consists of ten herbal components. Due to the complexity of its multi‐component nature, establishing a robust quality control system remains challenging. This study presents the first systematic quality evaluation of HZXL using high‐performance liquid chromatography (HPLC) fingerprinting combined with ultra‐high‐performance liquid chromatography–Orbitrap mass spectrometry (UHPLC–Orbitrap MS) untargeted metabolomics. Based on high‐performance liquid chromatography, seven reference standards (caffeic acid, calceolarioside B, chlorogenic acid, geniposide, glycyrrhetinic acid, luteolin, and myricitrin) were used as markers for the fingerprint analysis of ten batches of HZXL, and the results demonstrated excellent similarity (> 0.930). Using liquid chromatography–mass spectrometry (LC–MS), untargeted metabolomics identified 152 compounds, predominantly flavonoids, iridoid glycosides, phenolic acids, and organic acids. The confidence of structural annotations was rigorously evaluated according to the metabolomics standards initiative (MSI) framework. Chemometric analysis, including hierarchical cluster analysis (HCA), and principal component analysis (PCA) on all 10 batches, revealed high overall batch consistency with only minor inter‐batch variations. The technique for order preference by similarity to ideal solution (TOPSIS) combined with entropy weighting was applied to screen core quality markers from the 152 identified compounds. Chlorogenic acid, geniposide, neochlorogenic acid, caffeic acid, myricitrin, and ursolic acid were selected as core quality markers after entropy‐weighted TOPSIS ranking combined with pharmacological evaluation. This multi‐component quality evaluation framework, integrating fingerprinting, high‐resolution metabolomics, and advanced chemometrics, provides a scientific basis for establishing comprehensive multi‐indicator quality standards, improving batch‐to‐batch reproducibility, and supporting the modernization and internationalization of HZXL.

Insulin resistance is a core pathological hallmark of metabolic syndromes like type 2 diabetes mellitus. Peroxisome proliferator-activated receptor gamma (PPAR-γ) is a classical therapeutic target for improving insulin sensitivity, yet the clinical utility of its synthetic agonists faces diverse adverse reactions. We previously showed that phillyrin, an important component of Forsythia suspensa, could improve insulin resistance in obesity. However, its direct molecular targets remain not fully understood. Herein, we adopted a combined in silico and experimental approach to determine whether phillyrin could function as a ligand of PPAR-γ. We implemented a multi-scale strategy integrating network pharmacology, molecular modeling, with experimental verification. Network pharmacology analysis was employed to predict system-level targets and pathways associated with phillyrin. Subsequently, molecular docking and molecular dynamics simulations were conducted using computational chemistry methods to characterize the binding mode, dynamic stability, and binding free energy (MM-GBSA) of phillyrin within the PPAR-γ ligand-binding domain. Finally, key computational predictions were experimentally validated in insulin-resistant 3T3-L1 adipocytes and in a high-fat diet-induced murine model of insulin resistance.

Macrophage lipid accumulation is a hallmark of atherosclerosis and other metabolic diseases. Toll-like receptor 4 (TLR4) signaling and cytoskeletal remodeling have each been implicated in this process, yet the mechanistic connections among these three elements remain incompletely defined. This review synthesizes current evidence for pathways in which TLR4 activation, via bifurcated signaling through MyD88/TRAM, SYK/Src, and integrins, engages small GTPases (Rac, Cdc42, Rho) and downstream effectors including cofilin and paxillin to drive cytoskeletal remodeling. These cytoskeletal changes facilitate lipid internalization through multiple routes, including macropinocytosis, receptor redistribution, and lysosomal synapse formation, and regulate lipid droplet dynamics, ultimately promoting foam cell formation. We also discuss the therapeutic potential and safety challenges of targeting cytoskeletal regulators in diseases such as atherosclerosis. The convergence of TLR4 signaling, cytoskeletal remodeling, and lipid metabolism represents a compelling mechanistic nexus for understanding foam cell formation. Elucidating the molecular details of this pathway-particularly the SYK/Src-cytoskeleton axis-may reveal macrophage-selective therapeutic targets that suppress pathological lipid accumulation while preserving host defense functions.

Platycodon grandiflorus is a widely used medicinal and edible species in Asia. With ongoing global climate warming, the rising incidence of extreme heat episodes has emerged as a major constraint on its productivity. To dissect the molecular regulatory network governing the response of P. grandiflorus to high-temperature stress, this study measured physiological indices and performed transcriptome sequencing (RNA-seq) on seedlings treated at 40 °C for 0, 1, 6, 12, 24, and 48 h. This was combined with microRNA (miRNA) sequencing data from the 0 h and 48 h time points for an integrated multi-omics analysis. Under high-temperature (HT) stress, the activities of superoxide dismutase (SOD), peroxidase (POD), and ascorbate peroxidase (APX) were markedly elevated, whereas malondialdehyde (MDA) and hydrogen peroxide (H ₂ O ₂ ) levels exhibited a transient increase followed by a decline. A comparison between the control and five HT treatment groups identified 1019 common significantly differentially expressed genes. miRNA sequencing identified 62 HT-responsive differentially expressed miRNAs targeting 157 predicted genes, and the integrated analysis further screened 32 differentially expressed target genes. Based on this, the study focused on elucidating the expression profiles of structural enzyme genes ( BASs, CYPs, UGTs, BGLUs ) in the platycodin biosynthesis pathway. It was found that upstream genes involved in terpenoid backbone synthesis were generally down-regulated, whereas downstream genes related to oxidative modification and glycosylation dynamics exhibited differential response patterns. Integrated analysis further revealed that ent-kaurene synthase (KS), the major rate-limiting enzyme in the gibberellin (GA) biosynthesis pathway, is potentially co-targeted by pgy-miR408-3p-3 and pgy-miR408d at the post-transcriptional level and exhibits significantly up-regulated expression. Conversely, BGLU11 , a β-glucosidase gene involved in saponin deglycosylation, was negatively regulated by pgy-miR395b-3 and exhibited significantly down-regulated expression. These results suggest that under HT stress, P. grandiflorus employs a complex miRNA-mRNA regulatory network to mediate a potential redistribution of terpenoid metabolic flux between the saponin and gibberellin pathways, while dynamically modifying saponin structures to maintain metabolic homeostasis. This study constructs a transcriptional and post-transcriptional regulatory framework for the response of P. grandiflorus to HT stress, providing key genetic resources and molecular clues for in-depth analysis of its thermotolerance mechanisms and for stress-resistance breeding.

Natural products from medicinal plants are vital sources for medicines, but understanding their complex production pathways within the plant is challenging. This review explores how artificial intelligence (AI), defined here as a suite of computational techniques including Machine Learning (ML), Deep Learning (DL), Natural Language Processing (NLP), and network analysis—is transforming this field of research. We describe how AI technologies, particularly machine and deep learning, are used to integrate large, heterogeneous biological datasets, extract features and identify key components in the biosynthesis of valuable compounds, and model how these metabolic networks behave over time. The review demonstrates that AI technologies effectively integrate large biological datasets to model dynamic metabolic behaviors. Furthermore, AI facilitates the optimization of the entire production chain, from cultivation conditions to extraction parameters. Ultimately, these technologies are shifting the research paradigm from conventional methods to precise, data-driven approaches, accelerating the sustainable bioproduction of plant-based natural products.