Jian Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jian Liu has a total of 244 co-authors. Here are the top co-authors by number of shared publications:
- Lihong Hu
Jiangsu Key Laboratory for Functional Substance of Chinese Medicine, School of Pharmacy, Nanjing University of Chinese Medicine
11 shared publications Rupeng Dai
School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing, 210023, PR China
8 shared publicationsWeijiang Lin
Nanjing University of Chinese Medicine
8 shared publicationsZhenzhen Zhu
Nanjing University of Chinese Medicine
8 shared publicationsDongjuan Si
School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing 210023, China
7 shared publicationsHongmei Wen
School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing 210023, China
7 shared publications
Latest Publications
- 16 November 2026
- Published by American Chemical Society (ACS) in Journal of Medicinal Chemistry
Ubiquitination-Informed Reinforcement Learning Linker Generator Enables Discovery of Novel FGFR4 PROTAC Degraders
Proteolysis-targeting chimeras (PROTACs) represent a promising therapeutic strategy, yet rational linker design is limited by inadequate theoretical guidance. Herein, we report PotentialLinker, a novel reinforcement learning framework for PROTAC linker optimization that uses a normalized weighted harmonic mean (NWHM) score to predict degradation activity. Benchmarking on ternary structures confirms its performance without precomputed protein−protein complexes. In case studies of STAT3/RIPK1, it recapitulated structure−activity relationships (SAR) and recommended appropriate E3 ligases. Then, conducted by PotentialLinker, a potent FGFR4 degrader D11 was obtained (DC50 = 56.23 nM, Dmax = 89.58%). Due to its covalent interacting with the unique Cys552 in FGFR4, D11 possesses excellent FGFR4 selectivity. Furthermore, D11 also demonstrated favorable oral bioavailability (F = 21.3%) and potent antitumor efficacy (TGI = 87.07%) with acceptable tolerability in vivo. Taken together, this framework effectively bridges degradation theory and practical PROTAC design, which can be utilized for the rational design of PROTAC-based therapeutic agents.
- 16 November 2026
- Published by American Chemical Society (ACS) in Journal of Medicinal Chemistry
Ubiquitination-Informed Reinforcement Learning Linker Generator Enables Discovery of Novel FGFR4 PROTAC Degraders
Proteolysis-targeting chimeras (PROTACs) represent a promising therapeutic strategy, yet rational linker design is limited by inadequate theoretical guidance. Herein, we report PotentialLinker, a novel reinforcement learning framework for PROTAC linker optimization that uses a normalized weighted harmonic mean (NWHM) score to predict degradation activity. Benchmarking on ternary structures confirms its performance without precomputed protein−protein complexes. In case studies of STAT3/RIPK1, it recapitulated structure−activity relationships (SAR) and recommended appropriate E3 ligases. Then, conducted by PotentialLinker, a potent FGFR4 degrader D11 was obtained (DC50 = 56.23 nM, Dmax = 89.58%). Due to its covalent interacting with the unique Cys552 in FGFR4, D11 possesses excellent FGFR4 selectivity. Furthermore, D11 also demonstrated favorable oral bioavailability (F = 21.3%) and potent antitumor efficacy (TGI = 87.07%) with acceptable tolerability in vivo. Taken together, this framework effectively bridges degradation theory and practical PROTAC design, which can be utilized for the rational design of PROTAC-based therapeutic agents.
- 15 October 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 66 (20) , 13313-13326
UCEC: A Knowledge Graph-Assisted Machine Learning for Molecular-Mechanism-Interpretable Natural Product-Disease Association Prioritization
Natural products (NPs) have been considered a critical source of therapeutic drugs, but prioritizing NP-disease associations remain challenging because supporting evidence is distributed across multiple ingredients, targets, protein interactions, and pathways. Herein, we propose a knowledge graph-assisted machine learning model: Uncertainty-Calibrated Evidence-Chains (UCEC) for prioritizing NP-disease associations. UCEC separates predicted association strength from the evidence-conditioned stability of local mechanistic support. On a proxy benchmark designed to recover graph-supported indirect associations, evidence-chain integration increased Hits@50 from 0.392 to 0.509, with 68.8% of NPs indicating improved per-NP AUPRC. Source-separated validation using HERB 2.0 clinical trial records further supported candidate prioritization. Thus, UCEC provides a reliability-aware approach for prioritizing graph-supported NP-disease candidates and generating traceable mechanistic hypotheses for subsequent validation.
- 15 October 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 66 (20) , 13313-13326
UCEC: A Knowledge Graph-Assisted Machine Learning for Molecular-Mechanism-Interpretable Natural Product-Disease Association Prioritization
Natural products (NPs) have been considered a critical source of therapeutic drugs, but prioritizing NP-disease associations remain challenging because supporting evidence is distributed across multiple ingredients, targets, protein interactions, and pathways. Herein, we propose a knowledge graph-assisted machine learning model: Uncertainty-Calibrated Evidence-Chains (UCEC) for prioritizing NP-disease associations. UCEC separates predicted association strength from the evidence-conditioned stability of local mechanistic support. On a proxy benchmark designed to recover graph-supported indirect associations, evidence-chain integration increased Hits@50 from 0.392 to 0.509, with 68.8% of NPs indicating improved per-NP AUPRC. Source-separated validation using HERB 2.0 clinical trial records further supported candidate prioritization. Thus, UCEC provides a reliability-aware approach for prioritizing graph-supported NP-disease candidates and generating traceable mechanistic hypotheses for subsequent validation.
- Open Access
- 3 February 2026
- Published by Wiley in Advanced Science
- Vol. 13 (20) , e20277
Multimodal AI‐Driven Identification of Dehydrocostus Lactone as a Potent Renal Fibrosis Attenuator Targeting IQGAP1
Renal fibrosis, a hallmark of chronic kidney disease (CKD), remains a critical therapeutic challenge with limited effective interventions. Herein, we proposed a multimodal AI‐driven Traditional Chinese Medicine (TCM) symptom prediction model (TCM‐SPred) and predicted potential herb‐symptom associations between herbs and symptoms to obtain agents for treating renal fibrosis. The prediction results of TCM‐SPred revealed that a natural guaianolide sesquiterpene lactone derivative dehydrocostus lactone (DCL, a main chemical constituent of Aucklandiae Radix) demonstrated significant anti‐fibrotic effects in vivo (unilateral ureteral obstruction) and in vitro (TGF‐β1‐induced epithelial‐mesenchymal transition). DCL directly targeted IQGAP1 to inhibit the Wnt signaling pathway by blocking the interaction between IQGAP1 and CCT3. These findings highlight the potential of DCL as a promising therapeutic candidate for renal fibrosis, providing novel insights into the IQGAP1‐CCT3‐Wnt signaling axis as a potential target for renal fibrosis intervention.
- 3 February 2026
- Published by Wiley in Advanced Science
- Vol. 13 (20) , e20277
Open AccessMultimodal AI‐Driven Identification of Dehydrocostus Lactone as a Potent Renal Fibrosis Attenuator Targeting IQGAP1
Renal fibrosis, a hallmark of chronic kidney disease (CKD), remains a critical therapeutic challenge with limited effective interventions. Herein, we proposed a multimodal AI‐driven Traditional Chinese Medicine (TCM) symptom prediction model (TCM‐SPred) and predicted potential herb‐symptom associations between herbs and symptoms to obtain agents for treating renal fibrosis. The prediction results of TCM‐SPred revealed that a natural guaianolide sesquiterpene lactone derivative dehydrocostus lactone (DCL, a main chemical constituent of Aucklandiae Radix) demonstrated significant anti‐fibrotic effects in vivo (unilateral ureteral obstruction) and in vitro (TGF‐β1‐induced epithelial‐mesenchymal transition). DCL directly targeted IQGAP1 to inhibit the Wnt signaling pathway by blocking the interaction between IQGAP1 and CCT3. These findings highlight the potential of DCL as a promising therapeutic candidate for renal fibrosis, providing novel insights into the IQGAP1‐CCT3‐Wnt signaling axis as a potential target for renal fibrosis intervention.
- Open Access
- 1 January 2026
- Published by American Association for the Advancement of Science (AAAS) in Research
- Vol. 9
Inhibition of DDX5-Mediated G-Quadruplex Unwinding in the β-Catenin 5′-UTR by Magnesium Isoglycyrrhizinate Alleviates Chemotherapy-Induced Intestinal Injury
Chemotherapy-induced gastrointestinal toxicity, particularly intestinal barrier disruption and diarrhea, is a major dose-limiting adverse effect with unclear mechanisms. Here, we identify magnesium isoglycyrrhizinate (MIG) as a novel therapeutic agent that ameliorates 5-fluorouracil (5-FU)- and irinotecan-induced intestinal injury by enhancing epithelial barrier integrity. Using MIG as a molecular probe, we revealed a pathogenic mechanism underlying chemotherapy-induced barrier damage. Integrated chemoproteomics (limited proteolysis–mass spectrometry and thermal proteome profiling) analyses revealed direct binding of MIG to the RNA helicase DDX5 (DEAD-box helicase 5), whose expression is markedly increased upon 5-FU-induced injury. Mechanistically, DDX5 destabilizes the G-quadruplex (G4) structure in the CTNNB1 (β-catenin) 5′ untranslated region, suppressing β-catenin production and compromising barrier function. Crucially, MIG acts as a novel DDX5 inhibitor that blocks G4 unwinding, thereby restoring β-catenin expression and barrier integrity in a dose-dependent manner. In vivo, MIG outperforms the canonical DDX5 inhibitor supinoxin in mitigating intestinal damage. Taken together, the results of our study not only establish MIG as a promising therapeutic candidate but also delineate the DDX5/ CTNNB1 axis as a targetable pathway for treating chemotherapy-induced barrier dysfunction.
- 1 January 2026
- Published by American Association for the Advancement of Science (AAAS) in Research
- Vol. 9
Open AccessInhibition of DDX5-Mediated G-Quadruplex Unwinding in the β-Catenin 5′-UTR by Magnesium Isoglycyrrhizinate Alleviates Chemotherapy-Induced Intestinal Injury
Chemotherapy-induced gastrointestinal toxicity, particularly intestinal barrier disruption and diarrhea, is a major dose-limiting adverse effect with unclear mechanisms. Here, we identify magnesium isoglycyrrhizinate (MIG) as a novel therapeutic agent that ameliorates 5-fluorouracil (5-FU)- and irinotecan-induced intestinal injury by enhancing epithelial barrier integrity. Using MIG as a molecular probe, we revealed a pathogenic mechanism underlying chemotherapy-induced barrier damage. Integrated chemoproteomics (limited proteolysis–mass spectrometry and thermal proteome profiling) analyses revealed direct binding of MIG to the RNA helicase DDX5 (DEAD-box helicase 5), whose expression is markedly increased upon 5-FU-induced injury. Mechanistically, DDX5 destabilizes the G-quadruplex (G4) structure in the CTNNB1 (β-catenin) 5′ untranslated region, suppressing β-catenin production and compromising barrier function. Crucially, MIG acts as a novel DDX5 inhibitor that blocks G4 unwinding, thereby restoring β-catenin expression and barrier integrity in a dose-dependent manner. In vivo, MIG outperforms the canonical DDX5 inhibitor supinoxin in mitigating intestinal damage. Taken together, the results of our study not only establish MIG as a promising therapeutic candidate but also delineate the DDX5/ CTNNB1 axis as a targetable pathway for treating chemotherapy-induced barrier dysfunction.
- 18 October 2025
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 65 (21) , 11701-11715
Machine Learning-Guided Generation of Novel Pseudo-Natural Products: Applications to Accelerate Drug Discovery
Natural products (NPs) are a critical source for drug discovery, and artificial intelligence (AI) is utilized to improve the efficiency of NP-based drug discovery. However, the existing AI-driven models typically generate a library of pseudo-natural products that only covers a small portion of the chemical space and the compounds were also restricted by poor drug-likeness profiles. Herein, the GPT1 is developed to generate diverse pseudo-natural products with excellent validity, uniqueness, and novelty while retaining molecular features similar to the training set. Subsequently, the Augmented Hill-Climb (AHC) strategy is employed to generate synthetically accessible compounds with enhanced drug-likeness. Using the integrated NPDL-GEN model (GPT1 + AHC), compounds G1-G5 were obtained, exhibiting significantly improved drug-likeness profiles. Furthermore, the pseudo-natural products H1-H3 generated via transfer learning also possess potent anti-inflammatory activities. Thus, our developed machine learning models can accelerate NP-based drug discovery.
- 18 October 2025
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 65 (21) , 11701-11715
Machine Learning-Guided Generation of Novel Pseudo-Natural Products: Applications to Accelerate Drug Discovery
Natural products (NPs) are a critical source for drug discovery, and artificial intelligence (AI) is utilized to improve the efficiency of NP-based drug discovery. However, the existing AI-driven models typically generate a library of pseudo-natural products that only covers a small portion of the chemical space and the compounds were also restricted by poor drug-likeness profiles. Herein, the GPT1 is developed to generate diverse pseudo-natural products with excellent validity, uniqueness, and novelty while retaining molecular features similar to the training set. Subsequently, the Augmented Hill-Climb (AHC) strategy is employed to generate synthetically accessible compounds with enhanced drug-likeness. Using the integrated NPDL-GEN model (GPT1 + AHC), compounds G1-G5 were obtained, exhibiting significantly improved drug-likeness profiles. Furthermore, the pseudo-natural products H1-H3 generated via transfer learning also possess potent anti-inflammatory activities. Thus, our developed machine learning models can accelerate NP-based drug discovery.
- 10 October 2025
- Published by American Chemical Society (ACS) in Journal of Medicinal Chemistry
- Vol. 68 (20) , 21534-21559
AI-Driven Transfer Learning and Generative Model (TransGenGRU) Enables the Drug Discovery of Novel Natural Guaianolide Sesquiterpene Derivatives as Potent NLRP3 Inhibitors
The NLRP3 inflammasome is recognized as a critical mediator of innate immunity, which can regulate the maturation of proinflammatory cytokines. Nowadays, several natural products have been confirmed to exhibit potent NLRP3 inhibitory effects and possess novel binding mechanisms with NLRP3. Herein, an AI-driven model (TransGenGRU) is proposed to generate novel natural products with NLRP3 inhibitory activities. Through the modeling of TransGenGRU, two guaianolide sesquiterpenoids (A3 and A8) are identified to possess moderate NLRP3 inhibitory activities. Then, through detailed structure optimization, E1 demonstrates the most potent NLRP3 inhibitory activity (IC50 = 24.42 nM), and the inhibitory effect on the NLRP3 inflammasome is correlated to the assembly of NLRP3/pro-caspase-1/ASC. Notably, E1 is confirmed to covalent-irreversibly interact with Cys280 that is totally different from MCC950. Besides, E1 also demonstrates potent anti-inflammatory activity in vivo, favorable DMPK profiles, and low hERG toxicity. Thus, E1 has been considered a novel and potent NLRP3 inhibitor.
- 10 October 2025
- Published by American Chemical Society (ACS) in Journal of Medicinal Chemistry
- Vol. 68 (20) , 21534-21559
AI-Driven Transfer Learning and Generative Model (TransGenGRU) Enables the Drug Discovery of Novel Natural Guaianolide Sesquiterpene Derivatives as Potent NLRP3 Inhibitors
The NLRP3 inflammasome is recognized as a critical mediator of innate immunity, which can regulate the maturation of proinflammatory cytokines. Nowadays, several natural products have been confirmed to exhibit potent NLRP3 inhibitory effects and possess novel binding mechanisms with NLRP3. Herein, an AI-driven model (TransGenGRU) is proposed to generate novel natural products with NLRP3 inhibitory activities. Through the modeling of TransGenGRU, two guaianolide sesquiterpenoids (A3 and A8) are identified to possess moderate NLRP3 inhibitory activities. Then, through detailed structure optimization, E1 demonstrates the most potent NLRP3 inhibitory activity (IC50 = 24.42 nM), and the inhibitory effect on the NLRP3 inflammasome is correlated to the assembly of NLRP3/pro-caspase-1/ASC. Notably, E1 is confirmed to covalent-irreversibly interact with Cys280 that is totally different from MCC950. Besides, E1 also demonstrates potent anti-inflammatory activity in vivo, favorable DMPK profiles, and low hERG toxicity. Thus, E1 has been considered a novel and potent NLRP3 inhibitor.
- 27 September 2025
- Published by Elsevier in European Journal of Medicinal Chemistry
- Vol. 301, 118199
Highly potent and selective degrader targeting ERα to improve the treatment outcomes of breast cancer
- 27 September 2025
- Published by Elsevier in European Journal of Medicinal Chemistry
- Vol. 301, 118199
Highly potent and selective degrader targeting ERα to improve the treatment outcomes of breast cancer
- Open Access
Dehydrocostus Lactone Effectively Alleviates Inflammatory Diseases by Covalently and Irreversibly Targeting NLRP3
The activation of nucleotide oligomerization domain‐like receptor (NLR) family, pyrin domain‐containing protein 3 (NLRP3) inflammasome is implicated in the pathogenesis of various inflammatory diseases. The natural product oridonin possesses a novel mechanism for NLRP3 inhibition and a unique binding mode with NLRP3, but its poor anti‐inflammatory activity limits further application. After virtual screening of diverse natural product libraries, dehydrocostus lactone (DCL) was considered as a potential NLRP3 inhibitor. DCL effectively inhibited caspase‐1 cleavage and release of IL‐1β in mouse and human macrophages at an extremely low concentration of 10 nM, comparable to MCC950. Mechanistically, our study assigned DCL a novel role in disrupting NLRP3 inflammasome assembly and ASC oligomerization. Excluding the influence on potassium/chloride ion efflux, calcium ion influx, and production of mitochondrial ROS, DCL formed a covalent bond with cysteine 280 in NACHT domain of NLRP3, thereby inhibiting the interaction between NLRP3 and NEK7. Furthermore, DCL exhibited protective effects in mouse models of NLRP3 inflammasome‐mediated diseases, including dextran sulfate sodium‐induced colitis, 2,4,6‐trinitrobenzenesulfonic acid‐induced Crohn's disease, LPS‐induced septic shock, and monosodium urate‐induced peritonitis. Our findings identify NLRP3 as the direct target of DCL, positioning DCL as a promising lead compound for treatment of NLRP3 inflammasome‐related diseases.
Open AccessDehydrocostus Lactone Effectively Alleviates Inflammatory Diseases by Covalently and Irreversibly Targeting NLRP3
The activation of nucleotide oligomerization domain‐like receptor (NLR) family, pyrin domain‐containing protein 3 (NLRP3) inflammasome is implicated in the pathogenesis of various inflammatory diseases. The natural product oridonin possesses a novel mechanism for NLRP3 inhibition and a unique binding mode with NLRP3, but its poor anti‐inflammatory activity limits further application. After virtual screening of diverse natural product libraries, dehydrocostus lactone (DCL) was considered as a potential NLRP3 inhibitor. DCL effectively inhibited caspase‐1 cleavage and release of IL‐1β in mouse and human macrophages at an extremely low concentration of 10 nM, comparable to MCC950. Mechanistically, our study assigned DCL a novel role in disrupting NLRP3 inflammasome assembly and ASC oligomerization. Excluding the influence on potassium/chloride ion efflux, calcium ion influx, and production of mitochondrial ROS, DCL formed a covalent bond with cysteine 280 in NACHT domain of NLRP3, thereby inhibiting the interaction between NLRP3 and NEK7. Furthermore, DCL exhibited protective effects in mouse models of NLRP3 inflammasome‐mediated diseases, including dextran sulfate sodium‐induced colitis, 2,4,6‐trinitrobenzenesulfonic acid‐induced Crohn's disease, LPS‐induced septic shock, and monosodium urate‐induced peritonitis. Our findings identify NLRP3 as the direct target of DCL, positioning DCL as a promising lead compound for treatment of NLRP3 inflammasome‐related diseases.
- 21 May 2025
- Published by Elsevier in Developmental Cell
- Vol. 60 (19) , 2558-2575.e6
Stress-induced pro-inflammatory glioblastoma stem cells secrete TNFAIP6 to enhance tumor growth and induce suppressive macrophages
- 21 May 2025
- Published by Elsevier in Developmental Cell
- Vol. 60 (19) , 2558-2575.e6
Stress-induced pro-inflammatory glioblastoma stem cells secrete TNFAIP6 to enhance tumor growth and induce suppressive macrophages