Large language models (LLMs) show promise for supporting Traditional Chinese Medicine (TCM) practice, but their clinical utility is limited by domain-specific knowledge gaps, hallucinations, and weak multi-turn reasoning. We present GastroTCM, a specialised LLM assistant for TCM gastroenterology that we built by fine-tuning a Llama3-8B model and augmenting it with a Retrieval-Augmented Generation (RAG) and an agent framework. GastroTCM targets key shortcomings in current TCM diagnostic support through three components: (1) a dedicated TCM gastroenterology vector database for efficient retrieval of high-value, peer-reviewed knowledge; (2) ShareGPT-style multi-turn dialogue optimisation to preserve clinical context across rounds; and (3) an intelligent agent that dynamically adapts its responses to evolving symptom profiles and user intent.GastroTCM was trained on approximately 20 million tokens of de-identified clinical records, guideline-based content, and expert-curated TCM question-answer pairs and evaluated against strong Chinese LLM baselines (ChatGLM-6B, Qwen-2). In automatic evaluations, GastroTCM outperformed all baselines in single-turn dialogue (BLEU: 0.334 vs. 0.172-0.246) and multi-turn consultations, where it achieved a substantially higher rate of proactive, clinically appropriate interactions (27/60 vs. ≤ 2/60 cases). Expert review by TCM gastroenterologists further confirmed higher diagnostic accuracy and safety, with the RAG module markedly reducing unsupported or hallucinated statements. These findings suggest that domain-specific, retrieval-enhanced LLMs can meaningfully augment-rather than replace-TCM practitioners in gastroenterology, with the potential to improve access to high-quality, explainable decision support in real-world settings.
Renal fibrosis (RF), as a critical pathological process in the progression of chronic kidney disease (CKD) to end-stage renal disease, currently lacks clear and effective reversal measures. Disordered fatty acid metabolism is highly involved in this process. Studies have revealed that molecular crosstalk under fatty acid metabolism dysregulation-such as lipotoxicity, mitochondrial damage, immune inflammation, epigenetics, metabolic reprogramming, and the gut-kidney axis-drives damage to renal tubules, glomeruli, and the interstitium. Therapeutic strategies targeting fatty acid metabolism to ameliorate RF have gradually become a research hotspot, though most current drugs are still in the clinical exploration stage, and the predictive and therapeutic value of many biomarkers remains to be validated. Traditional Chinese Medicine (TCM), with its personalized and modernized approach combining oral administration and external treatments such as acupuncture, enema, and nanodelivery, is increasingly demonstrating advantages in targeting fatty acid metabolism to improve RF. This review outlines the mechanisms and core targets of TCM in regulating fatty acid metabolism to ameliorate RF. The multi-target advantage based on complex components presents an opportunity for TCM to improve RF, though current research still faces challenges such as mechanistic complexity, standardization of TCM, and clinical translation. Future efforts should integrate multi-omics technologies, novel drug delivery systems, and disease-syndrome combination models to advance the development of new integrated Chinese and Western medicine drugs for RF.
Background: Nonalcoholic steatohepatitis (NASH) is a complex metabolic disorder with limited effective treatments, and its pathogenesis involves intricate crosstalk between gut microbiota, metabolism, and host signaling. Yinchenhaotang plus Zexietang (YCHZX), a traditional Chinese medicine (TCM) formulation, exhibits therapeutic potential in NASH, but its underlying mechanism-especially its interactions with the gut microbiota and metabolic networks-remains unclear.
Methods: A NASH mouse model was established via a high-fat/high-fructose/high-cholesterol diet. Mice were treated with YCHZX or its individual components (YCH, ZX). Serum biochemistry and liver histopathology were used to evaluate systemic therapeutic effects. Integrated multi-omics analyses (16S rRNA microbiome, serum metabolomics, colon transcriptomics) combined with immunofluorescence, immunohistochemistry, RT‒qPCR and ELISA were employed to explore regulatory networks. Complementarily, the effects of sodium butyrate and indolelactic acid (ILA) were investigated using an LPS-stimulated Caco-2 cell model. Antibiotic-mediated gut microbiota ablation was performed to verify microbiota dependency.
Results: YCHZX outperformed YCH and ZX in improving TC, LDL-C and hepatic pathology. Integrated multi-omics analysis demonstrated that the efficacy of YCHZX was associated with a distinct restructuring of the gut microbiota, specifically enriching butyrate-producing genera such as Lachnospiraceae_NK4A136_group. Concomitantly, YCHZX intervention suppressed colonic indoleamine 2,3-dioxygenase 1 (IDO1) and significantly elevated serum levels of ILA, a shift validated in vitro by the direct inhibitory effect of sodium butyrate on IDO1. The elevated ILA was shown to strengthen the gut barrier by upregulating occludin expression in LPS-stimulated Caco-2 cells via an aryl hydrocarbon receptor (AhR)-dependent mechanism. Further, YCHZX activated the AhR, upregulating tight-junction proteins (occludin) to reduce lipopolysaccharide (LPS) translocation, and inhibiting hepatic LPS/TLR4 signaling, TG accumulation, and IL-1β inflammation. All these effects of YCHZX were diminished by antibiotic-induced gut microbiota depletion.
Conclusion: Our findings demonstrate that YCHZX alleviates NASH in a gut microbiota-dependent manner. We propose a mechanism whereby YCHZX enriches butyrate-producing bacteria, which is associated with the suppression of colonic IDO1 and a shift in tryptophan metabolism toward ILA production. The increased ILA, in turn, contributes to the activation of the AhR, thereby restoring gut barrier integrity and mitigating liver inflammation.
Background: Gout is an inflammatory arthritis caused by purine metabolism disorders. The gout with the dampness-heat syndrome (GDHS) is a common Traditional Chinese Medicine (TCM) syndrome in this kind of disease, yet its modern scientific basis remains poorly understood. Simiao Pill (SMP), a classic formula in treating GDHS, has an unclear mechanism of action.
Methods: We conducted a targeted Nuclear Magnetic Resonance (NMR)-based metabolomic analysis on serum and urine samples from 197 GDHS patients and 101 healthy controls. Multiple machine learning algorithms, including support vector machine (SVM), random forest (RF), and least absolute shrinkage and selection operator (LASSO), were employed to identify potential biomarkers for GDHS. The Apriori algorithm was applied to uncover associations between TCM syndrome manifestations and metabolomic biomarkers. A subgroup of 50 GDHS patients received a 4-week SMP treatment, and their metabolomic profiles were compared pre- and post- intervention.
Results: GDHS patients exhibited a significant remodeled metabolome, characterized by disruptions in pyruvate, amino acid metabolism, and energy metabolism. A panel of 12 biomarkers with high diagnostic power was identified. Association rule mining further highlighted triglycerides and glycine as central nodes showing extensive connections to TCM syndromes. SMP intervention significantly reversed the level of 10 biomarkers (e.g., citrate, glycine, lactate), effectively normalizing perturbations in glycolysis/gluconeogenesis, the tricarboxylic acid cycle, and glycine/serine/threonine metabolism, and lipid homeostasis.
Conclusion: This real-world clinical study systematically delineates the metabolic features of GDHS, innovatively linking TCM syndromes to specific metabolic disturbances. It confirms that SMP exerts its therapeutic effects through multi-targeted modulation of the metabolic network. This work provides a new scientific paradigm for the study of "disease-syndrome-treatment" in TCM.
Background: Metabolic associated fatty liver disease (MAFLD) has currently emerged as the predominant form of chronic liver diseases nowadays, with a high morbidity. Nevertheless, the general remission rate of current treatments for MAFLD remains unsatisfactory. The traditional Chinese medicine formulation Xiaochaihu decoction (XCHD) alleviates triglyceride (TG) and total cholesterol (TC) in patients with fatty liver disease, but the precise mechanisms remain to be determined.
Purpose: This study aimed to investigate the effects and underlying mechanisms of XCHD on MAFLD.
Methods: The components of XCHD and XCHD-containing serum were characterized by UPLC/MS. High-fat diet and glucose-fructose water-induced MAFLD model in mice was established to evaluate the effects of XCHD. Non-targeted metabolomics, RNA-sequencing, and Network pharmacology were performed and integrated in the mice liver. Molecular biology experiments, like Western blot, were taken to investigate its potential mechanisms. Finally, the effects of PPARγ, perilipin 2 (PLIN2), and perilipin 3 (PLIN3) were detected by over-expressed PPARγ or suppressed PPARγ.
Results: XCHD markedly alleviates MAFLD by reducing lipid droplets, TC, and TG accumulation in MAFLD mice and free fatty acid induced HepG2 and AML12 cells. Non-targeted metabolomics suggested that XCHD reduced hepatic lysophospholipid, and RNA-sequencing revealed that the underlying mechanism would be related to lipid droplet formation. Molecular biology experiments indicated that XCHD decreased PLIN2 and PLIN3 in vivo and in vitro. Network pharmacology analysis revealed that the mechanism of XCHD against MAFLD involves the PPARγ signaling pathway. Molecular biology experiments indicated that XCHD decreased PPARγ in vivo and in vitro. Overexpression of PPARγ indicated that XCHD exert regulatory effects through PPARγ-mediated inhibition of PLIN2 and PLIN3. However, the inhibitory effects of XCHD on PLIN2 and PLIN3 are not solely dependent on this pathway.
Conclusion: XCHD alleviates MAFLD by decreasing lysophospholipid, regulating liver lipid droplets formation, and targeting PLIN2/PLIN3. PPARγ was one of the cross in for PLIN2/PLIN3 pathway, which provides novel insights for MAFLD therapy.
Clinical practice guideline adaptation is critical for standardizing care, yet existing frameworks like the ADAPTE process lack specificity for Traditional Chinese Medicine (TCM), whose principles diverge fundamentally from western medicine. To address this gap, we modified the ADAPTE process to align with the unique characteristics of TCM. A dual-tiered working group (core team and expert panel) simulated tested the original ADAPTE process on Traditional Chinese Medicine clinical practice guidelines (TCM CPGs), identifying challenges such as incompatibility with TCM diagnostic frameworks and holistic treatment paradigms and formed the initial version of ADAPTE-TCM. 21 experts were invited to evaluate the initial version of ADAPTE-TCM and reached a high consensus, while also providing suggestions for modifications to the initial version. Ultimately, 20 steps were modified (e.g., evidence synthesis, consensus-building) and integration of TCM CPGs specific evaluation criteria and tool. The resulting ADAPTE-TCM framework provides methodological rigor for adapting TCM CPGs. This adaptation not only bridges a critical gap in TCM CPGs adaptation but also exemplifies how evidence-based methodologies can evolve to respect diverse medical traditions.
Background: Cerebral ischemia/reperfusion injury (CI/RI) remains a critical barrier to effective ischemic stroke (IS) treatment. While mitophagy activation has been shown to attenuate apoptosis and pyroptosis, thereby ameliorating CI/RI, the therapeutic potential of natural compounds targeting this pathway remains underexplored. Penthorum chinense Pursh (PCP), a traditional hepatoprotective herb, contains Thonningianin A (TA), a bioactive compound with reported autophagic properties. However, the role and mechanisms of TA in CI/RI mitigation remain unclear.
Methods: In vivo, a middle cerebral artery occlusion/reperfusion (MCAO/R) rat model was established to evaluate TA's neuroprotective effects via TTC staining, Longa neurological scoring, and immunofluorescence staining. In vitro, oxygen-glucose deprivation/reoxygenation (OGD/R)-treated HT22 and BV2 cells were used to assess TA's impact on cell viability (MTT, Hoechst/PI staining), mitochondrial oxidative stress (DHE, TMRM, JC-1, Mito-Tracker staining and Western blot), apoptosis (flow cytometry, immunofluorescence staining, Hochest and PI staining and Western blot), and pyroptosis (EthD-2/YO-PRO-1 staining and Western blot). Autophagy and mitophagy modulation was investigated using rapamycin (Rap), 3-MA (autophagy inhibitor), CCCP (mitophagy inducer), and AC220 (mitophagy inhibitor) in EGFP-LC3-U87 and mCherry-GFP-FIS1-293T cells. Co-localization immunofluorescence and Western blotting were employed to validate PINK1/Parkin pathway involvement.
Results: TA administration significantly improved neurological function, reduced cerebral infarct volume, and attenuated neuronal damage in MCAO/R rats. In vitro, TA suppressed OGD/R-induced mitochondrial oxidative stress and apoptosis in HT22 cells while mitigating pyroptosis in BV2 microglia. Mechanistically, TA activated PINK1/Parkin-dependent mitophagy, as evidenced by enhanced LC3-II/I ratio, and increased mitochondrial-autophagosome co-localization. Crucially, TA's anti-apoptotic and anti-pyroptotic effects were abolished upon mitophagy inhibition. These findings were corroborated in the MCAO/R model, where TA upregulated PINK1/Parkin signaling and mitigated cell damage.
Conclusion: This study identifies TA as a novel natural agent alleviating CI/RI by activating PINK1/Parkin-mediated mitophagy, thereby concurrently suppressing apoptosis and pyroptosis. These findings provide the first elucidating the molecular mechanis underlying TA's potential as a therapeutic candidate for IS.
Background: Traditional Chinese Medicines (TCM) has long relied on bioactive compounds derived from natural sources, but conventional extraction and separation methods often involve violate/hazardous organic solvents, posing environmental and health risks. Deep Eutectic Solvents (DESs) have emerged as a sustainable alternative, offering tunable physicochemical properties, biodegradability, and enhanced extraction efficiency for TCM constituents such as alkaloids, flavonoids, and polysaccharides.
Methods: This review comprehensively summarizes the synergistic integration of DESs with TCM, highlighting their applications in green extraction, purification, and stabilization of bioactive compounds. It investigates a series of separation techniques, including liquid/solid-liquid (micro) extraction, chromatographic systems and others, where DESs enhance efficiency and recyclability. Environmental-Health-Safety (EHS) analyses, such as life cycle assessments and related tools, are also discussed.
Results: DESs demonstrate superior performance in preserving heat-sensitive compounds, improving solubility, and enabling selective extraction as well as isolation while aligning with green chemistry principles. However, challenges such as high viscosity, scalability, and toxicological assessments remain. Despite these limitations, DESs show significant eco-friendly potential, and future opportunities in policy support and AI-driven design could further advance their role in modernizing TCM for safer, more efficient, and sustainable therapeutic development.
Depression is a widespread mental disorder with profound effects on both physical and psychological health. Aberrant signal transduction contributes to depression by impairing neuronal function, reducing synaptic plasticity, and disrupting neurotransmitter transmission. Aberrant signal transduction can impair neuronal function, reduce synaptic plasticity, and disrupt neurotransmitter transmission, thereby contributing to the development of depression. Xiao-Yao-San (XYS), a traditional Chinese medicine (TCM) formula, has been extensively employed in the treatment of depression, with a broad therapeutic profile. This review aims to critically assess current scientific evidence on the antidepressant effects of XYS, focusing on its modulation of key signaling pathways. The goal is to provide a more robust mechanistic foundation for XYS-based therapies and offer insights for developing novel antidepressants that target signaling pathways. We systematically searched PubMed, Web of Science, ScienceDirect, CNKI, and Wanfang databases from inception to May 1, 2025, using relevant terms to identify studies on XYS and its active components, particularly those elucidating its regulation of signaling pathways involved in depression. XYS and its active ingredients modulate several crucial signaling pathways implicated in depression, including the phosphatidylinositol 3-kinase/protein kinase B (PI3K/Akt), NOD-like receptor family pyrin domain containing 3 (NLRP3), nuclear factor-kappa B (NF-κB), and mitogen-activated protein kinase (MAPK) pathways. The antidepressant effects of XYS are primarily mediated through anti-inflammatory actions, improved hippocampal architecture, suppression of neuronal and mitochondrial apoptosis, reduction of oxidative stress, and enhancement of synaptic plasticity. Despite current limitations, this review identifies future research directions focusing on unexplored and cross-linked signalling pathways, which may support the development of signalling-targeted antidepressant agents based on XYS.

