Biologic drugs, primarily comprising proteins and nucleic acids, have emerged as powerful therapeutic modalities; however, their discovery and optimization are often hindered by their inherent complexity. The advent of artificial intelligence (AI), particularly deep learning, is catalyzing a paradigm shift in this field, transitioning it from a process reliant on serendipity and laborious experimentation to a data-driven engineering discipline. This review systematically charts the co-evolution of AI methodologies and their transformative applications across the modern biologic drug development pipeline. We first outline AI’s methodological progression, from language models deciphering biological sequence grammar to structure prediction models like AlphaFold making macromolecular folds computationally accessible, and finally to generative models enabling de novo molecular creation. We then explore the practical impact of these technologies in two core phases: the de novo design of novel biologics with bespoke functions and the subsequent multi-parameter engineering and optimization of these candidates for clinical viability. While the potential is immense, significant strategic challenges remain, including the need to build a new AI-native experimental ecosystem and bridge the profound complexity gap between molecular-level predictions and systemic in vivo outcomes. Overcoming these obstacles will usher in a new era of AI-driven, automated closed-loop drug discovery.

Nanodrug delivery systems (NDDS) have demonstrated outstanding performance in drug delivery due to their efficient delivery capacity, targeting ability, and biocompatibility. However, the development of nanomedicines still heavily relies on the expertise of formulation scientists and extensive trial-and-error experiments. Despite the abundance of data in nanoscience, traditional biological research often struggles to effectively process, analyze, and utilize these datasets, limiting nanomedicine studies to a “one-to-one” approach. Against this backdrop, the rapid growth of artificial intelligence (AI) and machine learning (ML) offers a new paradigm for nanomedicine research. Unlike traditional statistical analyses and mathematical models, AI and ML provide deeper insights into big data, enhancing the efficiency of nanomedicine development while steering the field toward more intelligent and more precise research approaches. This review focuses on milestone studies that use ML to reshape nanomedicine research from a pharmaceutics perspective, highlighting how data-driven ML models can guide new directions in nanomedicine development.

Enzymatic reactions play an emerging role in a broad spectrum of scientific and industrial applications. The inherent complexity of enzymes, such as their substrate specificity, conformational flexibility, and the vast diversity of reactions involved, poses substantial challenges for the advanced computational prediction of enzymatic reactions with desirable accuracy. Moreover, existing approaches are mostly tailored for a specific sub-task, such as substrate prediction or binding site annotation, which limits their applicability. In this study, we introduce ERAM, a task-agnostic multimodal learning framework capable of addressing a broad range of downstream applications with both accuracy and efficiency. ERAM aligns pre-trained molecular representations from Protein Language Model with the knowledge of enzyme catalysis by modeling enzymatic reactions as multi-relational data. In enzyme retrieval tasks, ERAM achieves an improvement of 28.31% in mean average precision compared with the state-of-the-art (SOTA) method, CREEP. In substrate prediction tasks, ERAM outperforms the SOTA method ESP, achieving average improvements of 35.53% and 22.97% in Matthews correlation coefficient across two datasets. Additionally, ERAM exhibits commendable interpretability by assigning higher attention weights to binding sites, resulting in lower false-positive rates (42.36%) and higher overlap scores (70.59%) in the unsupervised binding site prediction task compared to RXNAA Mapper. By learning embeddings of substrates, enzymes, and products within a unified knowledge graph latent space, ERAM demonstrates its potential as a versatile and effective tool for enzyme catalysis research.

Accurate prediction of protein-ligand complexes and binding affinity is critical for hit identification and optimization for structure-based drug design. Traditional docking simulates binding processes with searching algorithms guided by energy-scoring functions, which are quite computationally expensive and time-intensive. In contrast, deep learning approaches offer a cost-effective alternative, yet often generate conformations with limited physicochemical validity and fail to account for protein flexibility. To address these pitfalls, we propose FlowDock, a multitask framework enhanced by Bayesian Flow Networks. FlowDock simultaneously generates accurate protein-ligand complex structures and predicts binding affinity while incorporating protein conformational flexibility. By leveraging multimodal intramolecular representations with a deep equivariant generative model, our method iteratively refines complex in latent space, ensuring rapid and stable generation. Benchmark evaluations demonstrate that FlowDock achieves state-of-the-art performance in binding pose prediction, especially physical plausibility, and virtual screening capability, alongside reliable binding affinity predictions. By providing deeper molecular insights into dynamic protein-ligand interactions, FlowDock represents a robust tool for accelerating the rational development of therapeutics.

PI3Kγ represents a promising therapeutic target for its pivotal role in macrophage recruitment and polarization and its significant association with tumor invasion and metastasis. In this study, a series of novel indole-based PI3Kγ selective inhibitors were generated by machine learning combined with molecular hybridization. Intriguingly, the representative IHA-5f displayed picomolar-level potency and highly selective inhibition to PI3Kγ relative to PI3Kα/β/δ. Moreover, IHA-5f manifested prominent anti-melanoma activity in vitro and in vivo with no detectable visceral toxicity. Mechanistically, IHA-5f efficiently suppressed tumor cell proliferation and migration, and induced apoptosis by suppressing the PI3Kγ/AKT/NF-κB signaling axis. Concurrently, it restrained the M2 polarization of tumor-associated macrophages, thereby augmenting the antitumor immune response. This study underscores the potential for PI3Kγ inhibitors as immunomodulators and direct antitumor agents.

Natural products and their derivatives have long been crucial in drug therapy, especially in traditional medicine. However, challenges in screening, isolation, characterization, and optimization have slowed their development in the pharmaceutical industry. Recent advancements in artificial intelligence (AI) and multi-omics technologies are revitalizing this field. AI offers powerful tools for understanding natural compounds, enhancing molecular representations, and supporting tasks such as binding prediction, drug repurposing, and retrosynthesis. Moreover, generative models are aiding in natural product optimization and the creation of pseudo-natural compounds. At the same time, multi-omics technologies, including genomics, transcriptomics, proteomics, and metabolomics, have enabled high-throughput studies of plant traits, synthesis, regulatory mechanisms, and quality control, providing valuable data for AI model development. These advancements help accelerate the discovery of new compounds with medicinal potential. Furthermore, in the field of traditional Chinese medicine research, which is largely based on natural plant sources, AI systems exemplified by UNIQ system, combining AI and multi-omics, have been instrumental in mechanistic studies and new drug development. This study comprehensively discusses the algorithms and applications of AI and multi-omics technologies in the drug development of natural compounds and plants, as well as summarizing relevant databases which might provide high-quality data for the future development of AI algorithms targeting natural products.

In a drug product, the major components by mass are the drug inactive ingredients (DIGs), which raises great concerns about their unwanted effects and clinical toxicities. It is demanded to unveil their proteome-wide bioactive landscape using computational methods. However, existing methods are impeded by either incapability to scan human proteome or inaccuracy in DIGs’ bioactivity prediction. Here, a cross-attention transformer model, titled TransDIG, leveraging cross-module deep transfer learning was therefore developed to map the bioactive landscape of DIGs using minimal experimental data. First, the generalizability and interpretability of this model was verified by the prediction of zero-shot proteins and identification of key atoms/residues, respectively. Then, the bioactive landscape of hundreds of DIGs was unveiled using TransDIG, and thousands of potential bioactivities were found for the DIGs currently employed in pharmaceutical industry. Finally, the bioactivities of four popular DIGs were identified based on the landscape and experimentally validated by activity assay. As a result, the colorant β-carotene was validated to inhibit a critical drug transporter, and our study presented the first in vitro evidence of the bioactivity of the antioxidant dodecyl gallate that has not previously been reported to regulate any human protein. This study might offer insights for the design of drug formulation and its clinical utilization.

Despite extensive structural and functional characterization of the α7 nicotinic acetylcholine receptor, valuable structural insights into its interactions with conopeptides remain limited, thereby hindering the rational development of peptide-based modulators for this clinically important receptor subtype. Here, we present an integrated pipeline combining deep learning, structural biology, computational modeling and electrophysiology to accelerate the discovery and optimization of α7 nAChR-targeting conopeptides. To overcome data scarcity, we developed a deep learning model using the ESM-2 protein language framework, enabling efficient screening of 689 disulfide-poor conopeptides. This approach identified SS1, a novel antagonist of α7 nAChR, which was systematically optimized via structure-activity relationship studies to yield [△QP,S8R]SS1—a minimalist peptide with nanomolar potency (IC₅₀ = 49.2 nmol/L), enhanced selectivity, and improved stability. Cryo-EM and computational modeling resolved the 3.3 Å resolution structure of α7 nAChR bound to [S8R]SS1, revealing a unique binding mode stabilized by hydrogen bonds, hydrophobic interactions, and glycan contacts, while hybrid receptor conformations (closed/desensitized) elucidated its inhibitory mechanism. This work establishes a transformative deep learning-to-experiment framework for accelerating the discovery and optimization of nature-inspired peptide therapeutics.

Triple-negative breast cancer (TNBC) is a highly aggressive and heterogeneous subtype of breast cancer characterized by early metastasis, poor prognosis, and high recurrence rates. Targeting dysregulated PI3K/Akt/mTOR signaling and triggering anti-tumor immunity represent promising strategies for TNBC therapy. In this study, we report the discovery of a series of novel chromone derivatives as potent mTOR inhibitors by artificial intelligence-assisted drug design and structure-based drug design. The optimal compound, MT-44, was a highly selective mTOR inhibitor and showed no obvious binding activity to a broad panel of 200 kinases, and it exhibited nanomolar-level mTOR inhibitory and anti-TNBC cells proliferative activities. MT-44 effectively blocked the PI3K/Akt/mTOR signaling pathway and exerted robust anti-tumor efficacy in an MDA-MB-231 xenograft mouse model. Furthermore, MT-44 activated pattern recognition receptor TLR2 and upregulated the cGAS/STING signaling pathway, and reshaped the tumor microenvironment, thereby enhancing the tumor immune landscape. Collectively, our findings highlighted MT-44 as a highly selective and potent mTOR inhibitor with dual-targeted therapeutic and immunomodulatory effects, offering an appealing strategy for TNBC.

Non-canonical kinases (NCKs) are emerging as druggable targets in oncology, yet a comprehensive map linking their molecular mechanisms and targeting strategies to small-molecule modulators is lacking. Based on the hallmarks of cancer, we explain how NCKs buffer replication stress to preserve genome integrity, reprogram metabolic and stress pathways, coordinate angiogenesis and invasion, and support durable remodeling of the immune-tumor microenvironment. We review preclinical progress from hit identification to lead optimization, highlighting exploitable ATP-site and allosteric pockets, targeted protein degradation, and rational dual-node designs, all supported by structural insights and phenotypic discovery. Early clinical signals from mTOR, ATR, and DNA-PKcs inhibitors, along with late-preclinical programs targeting eEF2K, TBK1, FAM20C, TRPM7, and WNKs, reveal context-dependent NCK vulnerabilities. With further exploration of NCK functions and structures, additional targeted drugs are likely to be developed, potentially transforming non-canonical kinase biology into durable precision oncology.

Drug-induced liver injury (DILI) is an adverse hepatic reaction with a complex etiopathogenesis, involving direct cellular damage by the drug, immune-mediated mechanisms, and host and environmental factors. The gut microbiota has recently been recognized as a key player in human physiopathology. In DILI, animal models and patients have shown significant alterations in gut microbiota composition. Evidence indicates that patients with DILI often present intestinal barrier dysfunction, characterized by increased permeability and the translocation of pathogen-associated molecular patterns (PAMPs) to the liver. This process may contribute to the onset or aggravation of liver injury by triggering harmful immune responses. Furthermore, dysbiosis can alter bacterial metabolite production, thereby affecting intestinal and hepatic homeostasis. The gut microbiota can also modify the efficacy and toxicity of drugs on an individual level, thereby increasing the risk of DILI. This review provides an overview of the current evidence on the mechanisms by which the gut microbiota contributes to inflammation, immune recruitment, and exacerbation of liver damage in DILI. A more in-depth understanding of how the gut microbiota alters intestinal and hepatic homeostasis is essential for advancing our knowledge of DILI pathogenesis, integrating novel biomarkers into clinical practice, and developing microbiota-based interventions.

Molecular glue degraders (MGDs) have emerged as a transformative modality in the field of targeted protein degradation (TPD), enabling the selective elimination of disease-relevant proteins, including those traditionally considered undruggable. Unlike bifunctional proteolysis-targeting chimeras (PROTACs), MGDs operate through monovalent architectures that induce protein-protein interactions (PPIs) between E3 ligases and neosubstrates, offering advantages in chemical simplicity, cell permeability, and target scope. However, MGD discovery remains serendipitously, and a translational framework that links rational design to predictable selectivity and tissue exposure is still lacking. In this review, we present an integrated framework for advancing next-generation MGDs through three critical dimensions: rational design, specificity optimization, and delivery systems. First, we examined cutting-edge strategies in MGD design, including covalent handle-based reprogramming, PPI-driven stabilization, and multi-site, multi-functional constructs. Second, we explored structure-guided engineering and chemoinformatic models, such as cereblon degron motifs, zone-based design and multiparameter optimization, to improve neosubstrate selectivity while minimizing off-target liabilities. Third, we summarized delivery platforms, including antibody-drug conjugates, nanoparticle-enabled systems, and folate-mediated targeting, which are primarily intended to improve tissue selectivity and targeted distribution, thereby promoting local tissue accumulation. Finally, we discussed emerging opportunities at the intersection of artificial intelligence, structural biology, and systems pharmacology for accelerating MGD discovery and clinical translation. Collectively, these interdisciplinary insights underscore the therapeutic promise of MGDs and lay the groundwork for their next-generation evolution in precision medicine.

RNA therapy represents an innovative approach for cancer treatment, with several RNA-based therapeutics having received approval from the US Food and Drug Administration. Circular RNA (circRNA), a closed-loop RNA molecule characterized by its high stability, plays a significant role in regulating biological processes by modulating gene expression and facilitating protein translation. Given its unique structure and diverse functionalities, the delivery of exogenous circRNA has emerged as a novel strategy for cancer therapy. This review examines the mechanism underlying circRNA-mediated tumor therapy, emphasizing its various biological roles, including that of an RNA sponge, aptamer, gene editing tool, and facilitator of protein translation, and explores the therapeutic potential in oncology. The review provides a comprehensive discussion on the synthesis strategies of exogenous circRNA, based on T4 DNA ligase and the permuted introns-exons (PIE) method, as well as elucidating the purification techniques. This article also reviews prominent carriers currently employed for circRNA delivery, such as lipid nanoparticle (LNP), exosomes, and virus-like particle (VLP), with a particular emphasis on their application in cancer-targeted therapies. Finally, the review summarizes key challenges currently in the field along with viable solutions. It highlights the prospective role of artificial intelligence in enhancing circRNA delivery to facilitate precise cancer treatment based on exogenous circRNA.

In recent years, China's new round of institution reform has further optimized the drug regulatory system. Relevant departments and institutions involved in traditional Chinese medicine (TCM) regulation have been strengthened. TCM regulatory science, as an emerging interdisciplinary field, has received high regard and experienced rapid development, significantly enhancing TCM regulatory capabilities. Simultaneously, accelerated progress in emerging technologies and production innovation for TCM drug discovery, coupled with the implementation of the National Major Scientific and Technological Special Project for "Significant New Drugs Development" and its translational achievements, have led to a historic turning point in the development of innovative natural TCM drugs over the past five years. Driven by the dual engines of "regulatory science" and "policy restructuring", the development of new TCM drugs has entered a fast lane. Both the quantity and quality of investigational new drug (IND) and new drug application (NDA) registrations and approvals for new natural TCM drugs have shown rapid growth, effectively meeting the public's health demands for TCM products and unmet clinical needs of patients. This study focuses on the development of new TCM drugs during the significant historical phase from 2021 to 2025. It provides a comprehensive overview of new TCM and natural drug applications and regulatory reviews over the past five years, delves into the implementation of the National Drug Regulatory Science Action Plan, and highlights the importance of TCM regulatory science as an emerging interdisciplinary field in accelerating the creation of new TCM drugs. It systematically summarizes the effects of regulatory policies and regulations, the reform of TCM registration classification, specialized TCM registration provisions, and incentive measures such as the National Major Scientific and Technological Special Project for "Significant New Drugs Development". Based on an international perspective, it provides a focused review of recent highlights in TCM new drug development and regulation. This holds significant importance for promoting breakthroughs in TCM new drugs across more disease areas and advancing the international coordination of TCM regulation. The challenge faced in managing the registration of new TCM drugs lies in resolving the conflict between TCM theory and modern drug attributes, while balancing the rapid advancement of traditional medical theory and emerging technologies with the robustness of the drug regulatory framework. In the future, actively advancing research and translation in TCM regulatory science, innovatively establishing benefit-risk assessment systems and standards for new TCM drugs, and accelerating the development of a globally leading regulatory system with Chinese characteristics that aligns with the unique nature of TCM will be particularly crucial for global coordination of TCM regulatory policies, and the modernization and internationalization of TCM.

Liver fibrosis is a pathological process primarily driven by activated hepatic stellate cell (HSC). Single-cell transcriptomics of human fibrotic livers identified ASPN (Asporin) as highly expressed in inflammatory and fibrogenic HSC subsets. Clinically, Asporin was markedly elevated in liver tissue and serum, correlating with fibrosis stage across datasets and cohorts, supporting its potential as a non-invasive biomarker. Functionally, Asporin promoted HSC activation and extracellular matrix (ECM) remodeling, whereas its depletion reduced fibrosis in CCl₄-induced mouse models. Mechanistically, Asporin bound directly to ERH and stabilized it by preventing ubiquitin-mediated degradation. Structural modeling showed Asporin masked ERH's K12 ubiquitination site via hydrogen bonds and hydrophobic interactions. ERH overexpression in HSC activated fibrogenic genes and IL-17 signaling, converging on MAPK11 as a common downstream effector. Notably, ERH knockdown abrogated Asporin-driven profibrotic responses. High throughput screening identified prasugrel, a clinically approved drug, as a potent Asporin suppressor. In CCl₄ and high-fat diet induced fibrosis models, prasugrel alleviated fibrosis, inflammation, lipid accumulation, and portal hypertension by suppressing Asporin and ERH expression. Collectively, these findings define the Asporin/ERH/IL-17/MAPK11 axis as a key mediator of HSC activation and fibrogenesis and highlight prasugrel as a promising anti-fibrotic therapy.

Aberrant metabolic alterations underlie microglial dysfunction, which plays an important role during neurodegenerative progression. However, the role of aberrant glycogen metabolism remains elusive. Here, we identified glycogen accumulation and upregulated glycogenolytic enzymes in brain microglia from patients with Alzheimer's disease (AD) and transgenic animal models. Particularly, the principal microglial glycogenolytic enzyme PYGL exhibited the most notable spatiotemporal upregulation during disease progression. Specific knockdown of microglial PYGL ameliorated neuropathological changes and cognitive deficits in AD mice. Bioinformatics analysis and experimental validation confirmed that enhancing microglial autophagic flux-dependent Aβ clearance was the underlying mechanism. Furthermore, among all possible glycogenolytic pathways, PYGL downregulation primarily reduced hexosamine biosynthesis pathway activity, diminished UDP-GlcNAc and O-GlcNAcylation of the autophagy key protein SNAP29, and thereby facilitated formation of the SNARE complex, which is essential for autophagosome-lysosome fusion. These findings reveal a glycogenolysis-driven post-translational pathway regulating microglial autophagy, establishing PYGL as a therapeutic target for AD.

While adeno-associated virus (AAV)-mediated gene delivery has emerged as a promising therapeutic modality for neurological disorders, dose-dependent immune responses remain a critical barrier to clinical translation. Here we reveal the cyclic GMP-AMP synthase-stimulator of interferon genes (cGAS-STING) pathway as a key mediator of innate immune activation following intracranial AAV administration. Through comparative analyses in genetic and pharmacological intervention models, we demonstrate that STING signaling mediates key neuroinflammatory sequelae including glia reactivation, cytotoxic T cell infiltration, and neuronal injury. Mechanistically, microglia serve as the predominant sentinels detecting AAV immunogenicity via cGAS-STING activation. Therapeutic inhibition of this pathway by either microglia depletion or antagonism of STING by small molecules significantly mitigates high-dose AAV9-induced neurotoxicity while enhancing transgene delivery efficacy. Our work delineates a unified mechanistic framework linking AAV-triggered DNA sensing to neuroinflammatory pathology, and provides two clinically actionable approaches to decouple therapeutic gene delivery from detrimental immune activation in nervous system targeted gene therapy.

Nonstructural protein 12 (NSP12), the RNA-dependent RNA polymerase (RdRp) of SARS-CoV-2, serves as the catalytic core of the viral replication-transcription complex and is pivotal for viral RNA synthesis. Given its essential role, SARS-CoV-2 must precisely regulate intracellular NSP12 levels to support efficient viral replication. However, the mechanisms governing its stability remain poorly understood. Here, we reveal a previously unrecognized viral strategy in which NSP12 hijacks the host chaperone Hsc70 to modulate its own stability. Mechanistically, Hsc70 plays a dual role: mediating NSP12 degradation via chaperone-mediated autophagy (CMA) while also promoting its accumulation. The dynamic balance between these opposing functions determines NSP12 fate. NSP12 can evade CMA-mediated degradation by binding Hsc70 with higher affinity. This disrupts the Hsc70-LAMP2a interaction and shifts Hsc70’s role toward primarily facilitating NSP12 accumulation, thereby enhancing viral replication. We identified Hlyc41 as a potential antiviral agent that disrupts this hijacking mechanism. Hlyc41 competes with NSP12 for binding to the F428 residue of Hsc70, thereby promoting NSP12 degradation and suppressing viral replication. These findings reveal a novel host hijacking mechanism that regulates NSP12 levels and support a promising therapeutic strategy targeting this process.

The pre-metastatic niche (PMN) serves as a catalyst for tumor metastasis and colonization, involving communication between immune cells and stromal cells. However, less is known about the specific cell-type and their organ-specific functions in PMN formation, with available therapeutic strategies still limited. Here, we identified a significant expression of fibroblast activation protein alpha (FAPα) in hepatic stellate cells (HSCs) associated with the formation of liver PMN, which was dramatically attenuated in HSC-specific conditional Fap-knockout mice. Mechanistically, tumor cell-derived exosomal miR-2467-3p upregulated FAPα expression in HSCs. FAPα⁺ HSCs promoted IL-18 secretion via NF-κB/NLRP3/caspase-1 signaling pathway, which facilitated extracellular matrix (ECM) remodeling and macrophage recruitment. By targeting FAPα⁺ HSCs, the FAPα-activated prodrug Z-GP-DAVLBH disrupted the PMN and suppressed tumor liver metastasis. Collectively, our study emphasizes the crucial role of FAPα⁺ HSCs in the liver PMN and provides a promising therapeutic strategy for tumor metastasis.

Pharmacoresistance to anti-seizure medications (ASMs) remains a major unmet challenge in temporal lobe epilepsy (TLE), and occurs diversely, as classified to primary or acquired manner. The pathophysiological underpinnings of acquired pharmacoresistance remain elusive. Here, using a hippocampal kindling mouse model, we established that prolonged lamotrigine (LTG) treatment—either during or after kindling—induces broad-spectrum resistance to multiple ASMs, effectively recapitulating clinical patterns of acquired pharmacoresistance. Multimodal interrogation revealed hyperexcitability of subicular pyramidal neurons as a critical factor in pharmacoresistance, characterized by elevated c-Fos expression specifically within the subiculum, as well as hyperexcitability of subicular glutamatergic pyramidal neurons. This hyperexcitability phenotype stemmed from Nav1.6 upregulation, driving both enhanced persistent sodium current (INaP) and a pro-excitatory shift in voltage-dependent activation kinetics of voltage-gated sodium channel (VGSC). Crucially, pharmacological activation of subicular Nav1.6 sufficed to induce acquired pharmacoresistance in pharmaco-responsive mice. Conversely, subiculum-specific Nav1.6 knockdown in pyramidal neurons (but not GABAergic neurons) prevented or reversed pharmacoresistance, while analogous genetic manipulation in the CA1 had no such impact. Chemogenetic inhibition of subicular pyramidal neurons (mimicking ASM effects) restored drug responsiveness, directly implicating compensatory increases in Nav1.6 offsetting ASM’s inhibitory function on subicular excitability in acquired pharmacoresistance. These findings collectively identify Nav1.6 upregulation in subicular pyramidal neurons as a critical driver of acquired pharmacoresistance in TLE, highlighting a novel therapeutic target for refractory epilepsy.

Considering the exceptional anti-HIV-1 potency of rilpivirine (RPV) against diverse mutant strains and its remarkable human ether-a-go-go related gene (hERG) potassium channel inhibition (IC₅₀ = 0.50 μmol/L) as well as low selectivity (SI = 3989), a series of novel furo-[3,2-d]pyrimidine derivatives were rationally designed through a scaffold hopping strategy. Encouragingly, compound 10 revealed a striking reduction in hERG channel inhibition (IC₅₀ > 30 μmol/L) and significant increase in selectivity (SI = 161580). Notably, 10 exhibited excellent antiviral activity against various HIV-1 strains (EC₅₀ = 1.9-46.3 nmol/L). In particular, 10 displayed prominent inhibitory potency against Y188L (EC₅₀ = 15.5 nmol/L) and F227L + V106A strain (EC₅₀ = 8.7 nmol/L), which was superior to those of RPV (EC₅₀ ₍Y₁₈₈L₎ = 79.4 nmol/L, EC₅₀ ₍F₂₂₇L ₊ V₁₀₆A₎ = 81.6 nmol/L). Besides, no apparent cytotoxicity (CC₅₀ = 314.8 μmol/L) and negligible suppression of CYP isoenzymes were detected. Overall, these findings illustrated that 10 was a potentially promising NNRTI for HIV-1 therapy.

Excessive incorporation of long-chain fatty acids (LCFAs) into triglycerides in adipose tissue is a key contributor to obesity and related metabolic disorders, and pharmacologically modulating this process remains challenging. Here, we synthesized a library of β-indoquinazolinone derivatives via palladium-catalyzed oxidative addition complex chemistry and identified compound b2b as a potent and selective anti-obesity candidate. b2b inhibited triglyceride accumulation in adipocytes in vitro and significantly reduced adiposity, body weight, and lipid metabolic disturbances in diet-induced obese mice without observable toxicity. Mechanistic studies revealed that b2b directly activates nicotinamide phosphoribosyltransferase (NAMPT) and elevates intracellular NAD⁺ levels to enhance the NAD⁺-dependent regulatory protein SIRT1 activity. This activation leads to transcriptional repression of acyl-CoA synthetase long-chain family member 1 (ACSL1), thereby inhibiting LCFAs incorporation into triglycerides. These findings demonstrate that pharmacological activation of the NAMPT-NAD⁺-SIRT1 axis by b2b offers a novel strategy for obesity treatment.

Oncolytic peptides, which possess synergistic oncolytic-immunotherapy effects, have exhibited advantages including unique anticancer mechanism, overcoming drug resistance and broad anticancer spectrum in clinic. Nevertheless, conventional oncolytic peptides often suffer from poor stability, limited efficacy, and moderate anticancer selectivity. In this study, the stability-, potency-, and selectivity-guided optimizations were conducted on the first-in-class oncolytic peptide Clip-71. The robust synthetic strategy, in vitro and in vivo anticancer activity, and anticancer mechanism especially immune activation ability were systematically investigated for obtained peptides. The first round of stability-guided optimization identified the mirror-image peptides represented by QY-8, which possessed strikingly high stability, as well as outstanding in vitro and in vivo anticancer activities. Subsequently, to address the limited selectivity and nonspecific distribution of oncolytic peptides, the second round of selectivity-guided and PPCs-strategy based optimization was conducted on QY-8. Strikingly, the novel PPC QY-13, which was obtained through conjugation of the SSTR2-targeting peptide Tyr³-octreotate to QY-8, exhibited most potent anticancer activities both in vitro and in vivo, enhanced immunostimulatory activity, and superior biosafety as well as tumor targeting potency. Collectively, this study not only established promising strategies to significantly improve the anticancer potential of cytotoxic peptides, but also provided the new paradigm for targeted oncolytic-immunotherapy.

Target-based drug screening typically relies on biochemical or affinity-based assays to identify compounds that modulate or bind to purified target proteins in vitro. However, additional cellular validation is essential to confirm genuine drug-target engagements. Integrating screening and validation within a single cellular assay could greatly expedite the drug discovery process. Herein, we developed a cellular ligand discovery method called CPSEA (cellular protein stability enhancement assay), which leverages the biophysical principle of ligand-induced stabilization of target proteins containing destabilizing mutations. Using CPSEA, we identified arteannuin B and colchicine as novel ligands for FKBP12 and KRASG₁₂S, respectively. Importantly, we introduced both experimental and computational strategies to identify destabilizing mutations, thereby broadening the applicability of CPSEA for target proteins with and without known stabilizing ligands. Overall, CPSEA represents a powerful cell-based screening strategy with significant potential in target-based drug discovery.

Hydrogen persulfide/polysulfides (H₂S₂/H₂Sn), as an oxidized derivative of hydrogen sulfide (H₂S), is capable of directly inducing the S-persulfidation of cysteine residues, thereby modulating the activity of relevant enzymes. Owing to its unique reactive properties, H₂S₂/H₂Sn is emerging as a central focus in the study of reactive sulfur species. Therefore, the precise detection of H₂S₂/H₂Sn in vivo is critical for elucidating their roles in redox signaling and cellular regulation. However, conventional probes face challenges such as poor sensitivity, cross-reactivity, and instability. Here, we report a bioorthogonal ether linkage fluorescent probe toolkit (Cyne-1-5) with a cyclooctyne warhead, enabling ultra-sensitive (LOD = 3.3 nmol/L), selective, and real-time tracking of H₂S₂/H₂Sn in living systems. These probes feature rapid activation (>1018-fold fluorescence activation in 5 min), broad spectral coverage (blue to NIR), and exceptional enzymatic stability. Using this toolkit, we uncovered the spontaneous oxidation of H₂S to trace H₂S₂/H₂Sn and demonstrated steric hindrance-driven self-disproportionation of persulfides, where less bulky persulfides efficiently yield H₂S₂/H₂Sn. Furthermore, we achieved the cellular-level visualization of protein S-persulfidation dynamics. This work advances persulfide chemical biology and offers transformative tools for probing H₂S₂/H₂Sn in disease mechanisms and therapeutic development.

Duchenne muscular dystrophy (DMD), a fatal X-linked disorder, features progressive muscle fibrosis as a key driver of mortality. While CTGF represents a therapeutic target for DMD, its VWC-domain-targeting antibody (FG-3019) failed in clinical trials. Through experimental validation, we identified CT-domain as a superior target domain, as it contributed more fibrosis activity of CTGF than VWC-domain without elevating compensatory TGF-β1 level. Aptamers are synthetic oligonucleotides identified through SELEX, which can specifically bind to flexible protein domains through their unique 3D conformations. Their small molecular size enables effective tissue penetration while maintaining high target specificity, making them ideal CT-domain inhibitors. Nevertheless, conventional SELEX involves time-consuming and inefficient multiple screening rounds. Here, we employed our generative AI model, AptGEN, to rapidly discover a potent CT-domain specific aptamer within 42 days. This chemically modified aptamer (Apc003OA) distributed and remained in muscle tissues for an extended period, whereas FG-3019 could not. Importantly, it demonstrated better fibrosis inhibitory activity in vitro and in mdx mice when compared to FG-3019. Furthermore, Apc003OA demonstrated a favorable safety profile in mdx mice. Within 10 months, we progressed from target domain discovery, aptamer drug discovery, and then obtained both Orphan Drug Designation and Pediatric Rare Disease Designation by US Food and Drug Administration.

Antimicrobial resistance (AMR) poses a significant challenge to public health and human security, with plasmid-mediated horizontal gene transfer (HGT) being a primary driver for its dissemination. Here, we identify indole analogs containing electron-withdrawing groups as a new category of HGT inhibitors. Using indole-3-acetic acid (IAA) as a representative scaffold, we demonstrate that IAA targets glycolytic enolase to deplete phosphoenolpyruvate (PEP) in donor bacteria; this suppresses phosphotransferase system (PTS) activity, blocks PtsI phosphorylation, reduces cAMP synthesis and prevents catabolite repressor protein (CRP) activation, ultimately limiting intracellular ATP availability. Concurrently, IAA attenuates reactive oxygen species (ROS) generation by inhibiting FADH₂ oxidation and riboflavin biosynthesis. The concerted reduction in ATP and ROS arrests conjugative plasmid transfer. Overall, our work suggests the potential of indole analogs as a new class of conjugative transfer inhibitors and highlights that bacterial phosphotransferase system represents a promising target to prevent the propagation of AMR.

Stem cell-derived extracellular vehicles (EVs) hold great therapeutic potential for myocardial infarction (MI). However, the efficient production of EVs with high bioactivity remains a critical bottleneck limiting their clinical translation. Here, we demonstrate that conditioned photobiomodulation (PBM) with green light is capable of activating human embryonic stem cells (hESCs) to secrete more EVs with superior cardioprotective activity. These PBM-reprogrammed hESC-EVs improve cardiac recovery in a murine MI model by promoting cardiomyocyte proliferation and angiogenesis while inhibiting apoptosis. Notably, we validate that these EVs similarly enhance the survival and proliferation of human cardiomyocytes, underscoring their translational potential. Further analysis reveals that this benefit is due to the higher miR-423-3p content in reprogrammed hESC-EVs, which enhances glycolytic metabolism and restores mitochondrial function by regulating the ZBTB7A/PKM2 axis. Moreover, we synthesize a methacryloyl hydrogel microneedle patch with superior biocompatibility, biodegradability, and mechanical strength for loading hESC-EVs, and convey the patch to the infarcted heart via a modified delivery apparatus. This system ensures the precise and sustained delivery of EVs to ischemic myocardium, offering a potent treatment for MI. Collectively, this optical and biomaterials-based approach efficiently prepares EVs with higher cardioprotective activity, providing new therapeutic strategies for heart disease.

Although immunotherapy has revolutionized cancer treatment, antitumor immunological responses remain limited by insufficient tumor immunogenicity and immunosuppressive tumor microenvironment. Herein, pyroptosis induction is integrated into a photosensitizer to potentiate tumor immunogenicity. In this part, artesunate is disclosed to increase GSDME and modified to synthesize its ROS-cleavable prodrug, which is then installed into the self-assembled photosensitizer, resulting in a novel oil-in-water nanoplatform (BDP-pATS). The cytotoxicity, pyroptosis feature and potentiated GSDME induced by BDP-pATS are well confirmed. Subsequently, a novel acid-activatable adenosine-A2AR inhibitor is synthesized and further installed into the aforementioned platform to obtain BDP-pATS-aA2Ai, manipulating immunometabolic strategy to counterbalance the enhanced adenosine caused by pyroptosis. Such a photo-controlled and acid-activatable nanoprodrug enhances cytotoxic T cell functions while restrains regulatory T cell activities, leading to potent effects toward primary and abscopal tumor inhibition. In addition, BDP-pATS-aA2Ai also manifests desirable performance on pulmonary metastasis and tumor recurrence mouse model. To the best of our knowledge, this study presents the first concept of blocking adenosine-A2AR pathway during pyroptosis occurrence. Collectively, this work strategically combines immunometabolic interception, pyroptosis induction, photodynamic therapy and epigenetic regulation, emphasizing the significance of comprehensive therapy, which should open up a new viewpoint for cancer immunotherapy.

Cerebral ischemia-reperfusion (I/R) injury is exacerbated by the infiltration of splenic monocytes/macrophages (Mo/Mϕ) via the spleen-brain axis, where splenic-derived Mo/Mϕ migrate to cerebral lesions through C-C chemokine ligand 2/receptor 2 (CCL2/CCR2) chemotaxis, thereby amplifying oxidative stress and the neuroinflammatory cascade. Building on this endogenous pathway, we devised a delivery strategy that utilizes splenic Mo/Mϕ as “living vehicles” for targeted drug delivery. To this end, we developed a spleen-targeted magnolol liposome (Mag-PEG₅K) through optimized PEGylation, ensuring its spleen-specific accumulation and uptake by splenic Mo/Mϕ. After cerebral I/R injury, these nanoparticle-laden cells migrate to the ischemic brain via the CCR2/CCL2 axis to remodel the immunomodulatory microenvironment. This targeted system orchestrates dual therapeutic mechanisms within the lesion: mitochondria-directed reactive oxygen species (ROS) scavenging mitigates oxidative stress and peroxisome proliferator-activated receptor gamma (PPARγ) activation reprograms macrophage polarization, suppressing pro-inflammatory M1 differentiation and curtailing tumor necrosis factor-alpha (TNF-α) and interleukin-1beta (IL-1β) secretion. The attenuated cytokine release suppresses neuronal inflammatory cascades, thereby reducing apoptosis. In vivo, Mag-PEG₅K showed superior efficacy to free magnolol, effectively reducing infarct volume and improving long-term neurological outcomes. Supported by favorable biosafety, this work proposes spleen-targeted nanotherapy as an innovative strategy for reprogramming peripheral immunity via the spleen-brain axis, highlighting the translational potential of Mag-PEG₅K for addressing neuroinflammation and oxidative damage in ischemic stroke.

To explore a therapeutic approach with dual functions of activating cytotoxic CD8⁺ T cells and remodeling immunosuppressive tumor-associated macrophages (TAMs), the engineered macrophage-derived nanovesicles (MAC-PNV) were developed in this study. The MAC-PNV were derived from activated DC-like M1 macrophages after reprogramming macrophages by transcription factors PIB (PU.1, IRF8, and BATF3) and further stimulated with antigenic peptide, lipopolysaccharide (LPS), and interferon-γ (IFN-γ). In vitro results demonstrated that the enriched antigen-presenting complexes and co-stimulatory molecules were displayed on the surface of pro-inflammatory cargo-contained MAC-PNV, which enabled significant activation of CD8⁺ T cells and repolarization of M2 macrophages towards the M1 phenotype. After peritumoral administration, MAC-PNV alone significantly inhibited tumor growth by promoting the activation and intra-tumoral infiltration of CD8⁺ T cells, and remodeling the immunosuppressive tumor microenvironment (TME) in B16-OVA-bearing mouse models. More importantly, MAC-PNV remarkably enhanced the anti-tumor efficacy of low-dose liposomal doxorubicin (DOX-Lipo, 1 mg/kg), along with reducing its dose-limiting toxicities in B16-F10-bearing mouse models. This study highlights that the MAC-PNV would be a potential and effective immunomodulatory enhancer for providing a promising combination strategy with clinical chemotherapeutics.

Multiple sclerosis (MS) is a chronic inflammatory demyelinating disease of the central nervous system (CNS). Epstein-Barr virus (EBV)-induced B-cell overactivation could lead to inflammatory injury to the CNS, which is thought to underlie the initiation and progression of MS. To specifically eradicate these B cells, we report in situ EBNA1-specific chimeric antigen receptor (CAR)-T cells that were transiently programmed with circular RNA (circRNA)-laden CD7-targeted lipid nanoparticles (CD7-LNP). We demonstrate that systematic injection of CD7-LNP can efficiently introduce CAR circRNA to T lymphocytes and yield in vivo CAR-T cells. These in situ CAR-T cells were able to specifically clear EBNA1-specific B cells and significantly mitigate the progression of MS in a MS mouse model. Thus, in situ generation of EBNA1-specific CAR-T cells hold promise as a therapeutic strategy for MS that avoids the risks of general immunosuppression, and warrant further clinical trials.

Sepsis is a comprehensive ailment of systemic inflammatory response syndrome arising from infection. Activation of CASPASE-1 plays a central role in initiating the inflammatory cascade during sepsis. Herein, we construct optogenetically engineered extracellular vesicles (EVs) that achieve the specific degradation of CASPASE-1 and inhibit sepsis-associated inflammation. Specifically, blue light (460 nm)-induced CRY2/CIBN heterodimerization was applied during the EVs production stage to selectively load GCE-CTM fusion proteins into EVs by EXPLORs technology, yielding EVsGCE⁻CTM loading efficiency compared to conventional methods. Upon systemic delivery, EVsGCE⁻CTM preferentially accumulated in macrophages, where the GCE domain selectively bound activated CASPASE-1. The CTM motif then facilitated its lysosomal degradation by chaperone-mediated autophagy, resulting in potent inhibition of CASPASE-1 activity. In a murine model of sepsis, treatment with EVsGCE⁻CTM effectively attenuated systemic inflammation, reduced multi-organ damage, and significantly improved survival outcomes. This approach enables highly efficient, ubiquitin-independent degradation of intracellular target proteins through macrophage-directed EVs delivery, offering a potential therapeutic approach to address sepsis and other inflammation-related diseases.

Urinary tract infections (UTIs), especially complex or recurrent cases, present a significant challenge. Systemic antibiotics often fail to provide adequate local drug concentrations, leading to limited efficacy and relapse. Local delivery is hindered by the bladder’s dynamic environment, causing rapid drug clearance. To address this, a hydrogel scaffold, DRIVER (dual release, repair of tissues, immunomodulation, vesical adaptation, elimination of pathogens, and regulation of autophagy), is designed for sustained release in sync with the infection cycle. DRIVER forms a self-regulating 3D network through dynamic bonding and metal ion coordination, ensuring stable release and adaptability. It delivers a biphasic release profile comprising an initial antimicrobial burst that rapidly suppresses acute infection, followed by a 7-day sustained release phase. In vitro, DRIVER retained potent antibacterial and antifungal activity against planktonic, biofilm-embedded, and intracellular pathogens, including drug-resistant strains. Notably, the hydrogel also promoted bladder and kidney repair by modulating mitochondrial activity and autophagy, pathways essential for restoring urothelial integrity. DRIVER achieved >3 log reductions in bacterial and fungal burdens, normalized urinary function, and markedly attenuated systemic and tissue inflammation. Histological analyses confirmed robust architectural recovery in both the bladder and kidney. Together, these findings establish DRIVER as a compelling therapeutic strategy for complex UTIs.

Pyroptosis, a highly pro-inflammatory form of immunogenic cell death, holds great promise for cancer treatment. However, its efficacy in cancer cells is often limited due to low efficiency and cellular complex pro-survival mechanisms. In this study, we address this challenge by an integrated nanoplatform simultaneously activating two pathways of pyroptosis. Manganese ions and imidazole serve as a framework to coordinate glucose oxidase (GOx) and epigallocatechin gallate (EGCG) into stable biomineralized-like nanoparticles. We hypothesize that EGCG, as an inhibitor of DNA methyltransferase, may restore the expression of Gasdermin E (GSDME), a crucial component of pyroptosis activated by cleaved caspase-3. Through glucose consumption, GOx triggers both the caspase-1/Gasdermin D (GSDMD)-mediated and caspase-3/GSDME-mediated pathways of pyroptosis simultaneously, leading to efficient pyroptosis in cancer cells and a robust anti-tumor immune response, accompanied by the upregulated expression of PD-L1. Our results reveal that integrating this strategy with immune checkpoint inhibitors results in a tumor inhibition rate exceeding 80% across several “cold” tumor models, a 20% cure rate in the CT26 unilateral tumor model, and 5/8 distant tumors remaining free of recurrence upon re-challenge. In conclusion, this dual-pathway induction of pyroptosis offers a novel and promising strategy for enhancing cancer immunotherapy.