Drug discovery remains a protracted and capital-intensive process, primarily hindered by inefficiencies in drug screening. Microfluidic technology provides a promising approach for in vitro drug screening, enabling physiologically relevant, high-throughput, and cost-effective analysis by mimicking key aspects of cellular microenvironments. The synergistic integration of artificial intelligence (AI) with microfluidics constitutes a pivotal advancement in biomedical analysis. The convergence of the two facilitates automated data analysis, complex pattern recognition, and intelligent experimental control, thereby accelerating drug screening and contributing to enhanced precision. This review systematically presents the latest advancements in AI-assisted microfluidic drug screening, organized by increasing biological complexity: from single-cell analysis (1D), multicellular arrays (2D), and spheroids (3D), to sophisticated Organ-on-a-chip (OoC, 3D+) platforms. We detail how AI algorithms promote screening throughput, sensitivity, and physiological relevance at each scale. Furthermore, we critically discuss the prevailing challenges, including those related to data, model robustness, interpretability, and system integration. Finally, we outline future directions, highlighting the potential of AI-enhanced microfluidics to further advance precision drug discovery and biomedical research. We believe this timely review will offer a useful reference for researchers working in the interdisciplinary field of AI, microfluidics, and pharmacology.

Generative artificial intelligence (AI) models, a class of AI techniques that learn data distributions to synthesize novel samples, have emerged as impactful tools across scientific disciplines. In recent years, these models have found extensive applications in fields such as natural language processing and biomedical sciences. Despite their growing influence, comprehensive reviews on the application of generative models in biomolecular sciences remain limited. In this review, we provide a systematic overview of recent advances in generative models applied to biomolecular sciences. We discuss several prominent generative architectures, including variational autoencoders, generative adversarial networks, and diffusion models, highlighting their applications in molecular design and bioinformatics. Additionally, we examine how these models contribute to critical challenges such as molecular property prediction and molecular generation. Finally, we discuss key challenges that remain in this field, including model interpretability, scalability, and the need for high-quality molecular datasets. We highlight emerging research directions that aim to overcome these limitations and propose strategies for improving the reliability and applicability of generative models in biomolecular problems. Through this review, our objective is to provide researchers with a comprehensive understanding of the current landscape of generative modeling in biomolecular sciences and to inspire further advancements in this interdisciplinary area.

Phenotypic drug discovery (PDD) focuses on the observable traits or phenotype of cells or organisms in response to drug treatment, rather than relying primarily on specific molecular targets. Drugs discovered through this approach may have better therapeutic relevance, as they are tested in conditions that closely mimic human disease. In this study, we present PhenoModel, a multimodal molecular foundation model developed using our unique dual-space contrastive learning framework. This model effectively connects molecular structures with phenotypic information. PhenoModel is applicable to a range of downstream drug discovery tasks, including molecular property prediction and active molecule screening based on targets, phenotypes, and ligands. Our results demonstrate that PhenoModel outperforms baseline methods in these areas. Building from this model, PhenoScreen is developed to successfully identify several phenotypically bioactive compounds against osteosarcoma and rhabdomyosarcoma cell lines. These findings highlight the versatility of PhenoModel and its potential to accelerate drug discovery by uncovering novel therapeutic pathways and expanding the diversity of viable drug candidates.

Terpenoids exhibit diverse biological activities and thus have a wide range of pharmacological applications. In modern drug discovery, data-driven deep models play a crucial role in facilitating efficient feature representation and knowledge inference. To explore the uncharted bioactivity space of terpenoids, the construction of a multi-dimensional relational terpenoid database is essential for mapping terpenoid-bioactivity profiles. In this study, we first constructed a large-scale biological knowledge graph by integrating various data types, including terpenoid compounds, protein targets, cellular targets, genes, diseases, and their interrelationships. Subsequently, we developed a network-based disease prediction model, as well as optimized multiple compound-protein interaction prediction tools to extend the framework for activity research. These resources have been deployed on a user-friendly web platform (TeroACT) accessible at: http://terokit.qmclab.com/teroact/. Using in silico models within the TeroACT platform, we screened multiple terpenoid molecules for anti-melanoma activity. In vitro and in vivo animal models further validated the anti-migration and anti-proliferative effects of mollugin and columbianadin in melanoma. Additionally, integrated computational screening and experimental approaches identified numerous terpenoids with anti-inflammatory properties. In this sense, TeroACT fills the gap in terpenoid bioactivity study by providing a comprehensive data resource and AI-driven drug discovery tools.

The liver and pancreas are metabolically intertwined organs whose bidirectional communication is critical for maintaining systemic homeostasis. Dysregulation of this inter-organ crosstalk is a central driver in the pathology of a growing list of prevalent diseases, including metabolic dysfunction-associated steatotic liver disease (MASLD), liver cancer, acute and chronic pancreatitis, and various forms of diabetes. Given the substantial global health burden of these conditions and the lack of effective, Food and Drug Administration (FDA)-approved pharmacological interventions for those diseases, understanding the intricate mechanisms is an urgent and timely endeavor. This review provides a comprehensive synthesis of recent advancements in deciphering the molecular basis of liver-pancreas communication. We explore the multifaceted signaling networks involved, including the roles of liver-derived hepatokines, pancreas-derived hormones, extracellular vesicles, and metabolic exchanges. This axis is further integrated within broader systemic networks involving the gut, neuronal system, adipose tissue, and skeletal muscle. While clinical trials targeting this communication show promise (e.g., FGF21- and bile acid-related drugs), significant challenges remain, particularly the lack of FDA-approved pharmacological treatments for alcohol-associated liver disease (ALD), and acute/chronic pancreatitis. Future research must elucidate specific signaling pathways, identify novel pancreas-derived factors, and develop innovative therapeutic strategies. In particular, small molecule drug discovery based on polypharmacology, for these complex metabolic and organ-specific diseases.

Glioblastoma (GBM), the most aggressive primary brain tumor, remains a formidable therapeutic challenge, with a median survival under 15 months. Despite the current standard of care-comprising maximal safe surgical resection, radiotherapy, and temozolomide chemotherapy-patient outcomes have seen minimal improvement over the past two decades. A key barrier to effective treatment is GBM's robust and multifaceted immunosuppressive network, which critically undermines antitumor immunity. While much of the research has focused on the local immunosuppressive tumor microenvironment, systemic immunosuppression represents an equally important yet often underappreciated obstacle, significantly impairing host immune competence. Effective immunotherapy relies on an intact and functional immune system capable of mounting durable T cell-mediated responses. However, GBM induces profound systemic immune dysfunction, manifested by severe lymphopenia and depletion of effector immune cells, which further limits immune-mediated tumor control. Therefore, a comprehensive understanding of both systemic and local immunosuppressive mechanisms is essential for the rational design of effective immunotherapeutic strategies. In this review, we examine the unique physiological features of the brain, dissect the immunosuppressive landscape of GBM at both local and systemic levels, and highlight recent insights into the underlying mechanisms. We also discuss current immunotherapeutic modalities, and emerging drug delivery strategies aimed at overcoming immunosuppression to improve therapeutic efficacy.

RNA-binding proteins (RBPs) constitute central regulators of post-transcriptional gene expression and have been increasingly recognized as critical contributors to the pathogenesis of cancer, neurodegenerative disorders, and autoimmune diseases. However, in contrast to well-established drug targets such as kinases and G protein-coupled receptors, RBPs remain largely underexploited owing to their intrinsic structural heterogeneity, dynamic RNA interactions, and paucity of canonical ligand-binding pockets. In this review, we synthesize current knowledge on the roles of RBPs in disease, outline recent advances in the design of small-molecule modulators, and highlight innovative applications of high-throughput screening and chemical biology approaches for target identification and validation. We further discuss emerging concepts and challenges in translating RBP modulators into therapeutics, providing a forward-looking perspective on how these efforts may reshape small-molecule drug discovery in this evolving field.

The in vivo performance of drug delivery systems (DDS) is profoundly dictated by their interactions with the biomechanical environment. Consequently, actively tuning the mechanical properties of DDS, such as softness and deformability, has emerged as an important design principle for enhancing their therapeutic efficacy. By intelligently adapting to the body's complex biomechanical system, these engineered DDS can orchestrate specific biological responses, such as enhanced tissue penetration, prolonged systemic circulation, and even regulated cellular signaling pathways through mechanotransduction. This review systematically explores, from a biomechanical perspective, how to optimize the behavior of DDS in the complex biological environments by actively designing their mechanical properties. We discuss how this principle was applied across diverse platforms, including coacervates, hydrogels, Pickering emulsions, extracellular vesicles, and liposomes, to achieve enhanced therapeutic behavior for treating challenging diseases like cancer, chronic wounds, and neurological disorders. By focusing on these tunable mechanical properties, this review aims to provide a theoretical framework and insights for the future development of DDS with improved adaptability and therapeutic efficacy in clinical settings.

Membrane-derived biomimetic nanovesicles have emerged as a promising platform in cancer immunotherapy due to their intrinsic biocompatibility, functional plasticity, and capability to modulate immune responses. By integrating various immunotherapeutic agents, including immune checkpoint inhibitors, tumor antigens, and immunostimulatory adjuvants, these vesicles can be engineered to mimic natural immune communication and overcome key barriers in the tumor immune microenvironment. This review summarizes recent advances in the design, functionalization, and application of biomimetic nanovesicles for anti-tumor immunity. We particularly highlight strategies that harness these vesicles to enhance innate and adaptive immune responses, reverse immune suppression, and synergize with existing immunotherapy modalities. Furthermore, we discuss the challenges associated with biosafety, large-scale manufacturing, and clinical translation. Continued innovation in vesicle engineering and immunological modulation will be crucial for transforming biomimetic nanovesicles into viable next-generation cancer immunotherapeutics.

Metal-polyphenol networks (MPNs), a novel class of nano-biomaterials, have recently emerged as promising candidates for tumor diagnosis and therapy due to their unique chemical tunability, excellent biocompatibility, and synergistic multifunctionality. Notably, MPNs can be synthesized via one-step or multi-step approaches, allowing precise control over their morphology, size, and drug-loading capacity. The versatility of MPNs is further demonstrated by their ability to integrate multiple therapeutic modalities, including chemotherapy, photothermal therapy, photodynamic therapy, and chemical dynamic therapy. Furthermore, through surface modification with targeted molecules, MPNs enable tumor-specific targeting while facilitating real-time therapeutic monitoring via multimodal imaging. Additionally, MPNs exhibit excellent biocompatibility and superior biodegradability, making them highly suitable for biomedical applications. This review systematically explores MPN synthesis strategies and physicochemical properties. It then comprehensively analyzes MPN-based biomaterials and their tumor therapeutic mechanisms. Furthermore, we evaluate the challenges in MPN clinical translation and propose future perspectives for precise tumor treatment using MPN-based platforms. Ultimately, this review highlights the transformative potential of MPNs in advancing tumor theranostics and lays the foundation for their future clinical applications.

Phosphoglycerate kinase 1 (PGK1) is traditionally recognized for its pivotal role in glycolysis. Our findings reveal that PGK1 also functions as a protein kinase phosphorylating valosin-containing protein (VCP) at S746, which subsequently reduces Beclin 1 deubiquitination and impairs autophagy. Inhibition of PGK1 initiates autophagy in T315I-mutant chronic myeloid leukemia (CML) cells, thereby enhancing their sensitivity to first-generation Tyrosine Kinase Inhibitor (TKI) imatinib and third-generation TKI ponatinib. Despite the significant clinical implications, few PGK1-targeting inhibitors have been approved for clinical use to date. Through a comprehensive high-throughput screening of ∼20,000 natural compounds, we identified flavonoid as potent inhibitors of the enzymatic activity of PGK1. Subsequent structural optimization of these flavonoid derivatives led to the development of CPU-216, a compound that binds to the GLU344 and PHE292 residues of PGK1, effectively inhibiting its enzymatic and kinase activity. Notably, CPU-216 induces autophagy via VCP and Beclin 1 in CML-T315I cells, enhancing their responsiveness to TKIs. These discoveries propose a novel therapeutic strategy for T315I-mutant CML, underscoring the potential to develop targeted treatments that leverage the kinase functions of PGK1.

Hypobaric hypoxia-induced lung injury can exacerbate the incidence of plateau pulmonary edema, but relevant pharmacologic measures are relatively limited. Here, we investigate the possible role of paeoniflorin (Pae) in hypobaric hypoxia-induced lung injury. Through in vivo and in vitro experiments, we observed that Pae significantly ameliorated hypobaric hypoxia-triggered oxidative stress, inflammatory response, mitochondrial dysfunction and ferroptosis. Using limited proteolysis-mass spectrometry (LiP-MS), molecular docking, and molecular dynamics simulations, we identified that Pae directly binds to three amino acid residues (K101, D156, and S198) of the MEK2 protein. Knockdown of MEK2 expression in vivo and in vitro abrogated the protective effect of Pae. It was also observed that Pae promotes the binding of MEK2 and ERK2 and increases the phosphorylation level of ERK2, leading to its activation. This process induced upregulation of SGK1 and the protective effect of Pae against hypobaric hypoxic lung injury was dependent on SGK1. Collectively, these findings provide pharmacological evidence that Pae activates SGK1 by targeting MEK2 and mediating MEK2-ERK2 crosstalk, highlighting Pae's potential as a promising therapeutic agent for hypoxic lung injury-related diseases.

The increased stiffness of the extracellular matrix (ECM) is known to promote the progression of hepatocellular carcinoma (HCC). Currently, there are no approved therapies for targeting ECM sensors and remodelers. The objective of this study was to identify the molecular mechanisms underlying the role of Pleomorphic adenoma gene-like 2 (PLAGL2) in HCC ECM remodeling and to formulate compounds that effectively inhibit PLAGL2 transcriptional regulation. Our work revealed that PLAGL2 remodeled the ECM produced by HCC cells via an autocrine mechanism and activated HSCs via a paracrine pathway. Mechanistically, PLAGL2 functioned as a transcriptional regulator of insulin-like growth factor 2 (IGF2) and insulin-like growth factor 1 receptor (IGF1R). IGF2 enhanced ECM remodeling by HCC cells and activated HSCs through the IGF1R-PI3K-Akt signaling pathway. Furthermore, using a computer-aided drug design strategy, a novel compound, DC218, derived from the chemical evolution of cytisine, has been developed for the first time to exhibit specificity as an inhibitor of the PLAGL2 DNA binding domain. DC218 significantly degraded ECM, overcame lenvatinib resistance, and synergistically inhibited HCC. These findings provide mechanistic insight into the role of PLAGL2 in HCC ECM remodeling, as well as suggest a novel strategy for inhibiting ECM and treating HCC.

In our screening campaign for novel ferroptosis inhibitors, we identified that vitamin A (VA) and its metabolite all-trans retinoic acid (ATRA) exhibited potent ferroptosis-suppressing activity. Notably, through a combination of biochemical and pharmacological assays, we demonstrated that the anti-ferroptotic effects of VA and ATRA are independent of both antioxidative mechanisms and the canonical RAR/RXR signaling pathway. This conclusion was corroborated by a series of newly synthesized VA analogues. Furthermore, VA and its structural derivatives significantly alleviated ferroptosis-associated pathological phenotypes in murine models. Intriguingly, we discovered a novel function of VA and its analogues, which directly target acyl-CoA synthetase long-chain family member 3 (ACSL3) and enhance its enzymatic activity. This ACSL3-dependent mechanism increases the MUFA/PUFA ratio in phospholipids, thereby preventing lipid peroxidation. Strikingly, we further demonstrated that VA and its analogue D3 [(2E,4E,6E,8E)-N,3,7-trimethyl-9-(2,6,6-trimethylcyclohex-1-en-1-yl)nona-2,4,6,8-tetraenamide] extend the lifespan of C. elegans in a manner dependent on ACSL3, highlighting the physiological relevance of this pathway in aging. Collectively, our findings unveil a previously unrecognized role for VA and its analogues in modulating lipid metabolism, thereby providing a theoretical basis for their potential application in treating ferroptosis-related diseases and possibly enhancing longevity.

Ligand selectivity between μ-opioid receptors (μ-OR) and δ-opioid receptors (δ-OR) is key for improving opioid analgesics. While the “message-address” hypothesis has been central to explaining this selectivity, we present evidence for an additional mechanism. Chimeric receptor and mutagenesis studies identify residue 2.63 in the orthosteric pocket as a key determinant of morphine's μ/δ-OR selectivity, while EM-1's selectivity involves a combination of residues at positions 2.63 and 3.29, the N-terminus, and extracellular loop 3 (ECL3). Approaches like voltage-clamp fluorometry, engineered zinc-bridge and microscale thermophoresis show that EM-1's Y¹P²W³F⁴ sequence confers over 1500-fold μ/δ-OR selectivity, driven by steric hindrance and a β-turn structure. β-Endorphin, with a more flexible sequence, binds non-selectively to both receptors. Morphine's rigid isoquinoline scaffold and broader geometry restrict it to binding through the wide ECL2-transmembrane (TM) 5 cleft, explaining its modest μ/δ-OR selectivity. These findings reveal that μ/δ-OR selectivity is driven by both “message-address” interactions and receptor-specific binding pathways, advancing opioid drug design.

By employing a targeted strategy integrating a building blocks-based molecular network (BBMN), network annotation propagation (NAP), ultraviolet spectroscopy (UV), and mass spectrometry (MS), twelve novel spirocyclic heterodimeric alkaloid flavescensines A-L (1-12) with 5/6/6/6/5 and 6/5/6/6/5 pentacyclic skeletons were isolated from Sophora flavescens. Their structures were unambiguously elucidated through comprehensive spectroscopic data, quantum chemical calculations, and single-crystal X-ray diffraction. Structurally, these compounds represent the first examples of azaspiro[4.4] alkaloids formed through the inert ring A or B of the C₁₅ matrine-type alkaloid and C₉ units. Notably, the NMR signals of C-9′ methylene can serve as diagnostic indicators to determine the absolute configuration of the spiro carbon. A plausible biosynthetic pathway involving an unusual pattern of [3 + 2] cycloaddition was proposed. The hepatoprotective activities of 1-12 were evaluated in vitro, and 10 exhibited the most significant activity. Further in vivo experiments demonstrated that 10 dramatically inhibited the APAP-induced increase in the serum ALT, AST, and LDH levels, reversed the depletion of hepatic GSH, and attenuated hepatic centrilobular necrosis and hemorrhage. Mechanistically, 10 exhibits a potential interaction with DUSP2 and inhibits its expression, thereby suppressing DUSP2-mediated mitochondrial apoptosis via PI3K/Akt/JNK pathway. This represents the first discovery of DUSP2 involving in the hepatoprotection.

Tanshinones (TAs), well-known specialized diterpenoid metabolites in Salvia plants, exhibit distinct tissue-specific production in the root periderm. However, the mechanisms regulating this accumulation pattern remain unknown. Here, we employed a multi-omics analysis strategy to uncover the transcriptional regulatory network responsible for TA biosynthesis in Salvia miltiorrhiza roots. By integrating metabolic profiling, RNA-seq, and ATAC-seq, we profile the temporo-spatial dynamics of metabolic, transcriptional, and chromatin landscapes during early root development. Our results demonstrate that TAs biosynthesis and accumulation in S. miltiorrhiza roots display spatiotemporal patterns, marked by periderm-specific accumulation and initiation exclusively at specific developmental stages, tightly coordinated with dynamic changes in chromatin accessibility and transcriptional regulation. The constructed transcriptional regulatory network driving TA biosynthesis was found to be dominated by 211 key transcription factors (TFs). Experimental validations highlighted SmERF105 as a key positive regulator of TA, activating the transcription of KSL1, CYP76AH3, and the TA transporter ABCG1 to modulate the TA production. Our study uncovers novel, high-confidence regulators and offers an effective strategy for dissecting the genetic basis of plant specialized metabolites, offering value for advancing TA metabolic engineering.

The effective treatment of nasopharyngeal carcinoma (NPC) is challenged by an immunosuppressive tumor microenvironment (TME) and insufficient immune effector cell activation. Herein, we design a synergistic tri-modal therapeutic strategy to overcome these barriers. This platform integrates: (1) a CD109-targeted liposomal doxorubicin (S3-Lip-DOX) for precise chemotherapy and induction of immunogenic cell death (ICD); (2) non-genetically engineered natural killer (NK) cells armed with dual aptamers (targeting CD109 and PD-L1) via bio-orthogonal chemistry for enhanced tumor recognition (S3-P-NK); and (3) an Fc-engineered anti-PD-L1 antibody (Atezolizumab/IgG1) that restores antibody-dependent cellular cytotoxicity (ADCC). Crucially, we uncovered a key mechanistic synergy: S3-Lip-DOX treatment, as a stress-adaptive response, upregulates PD-L1 expression on NPC cells. This finding provides a compelling rationale for the integration, turning a potential immune escape mechanism into a therapeutic vulnerability. The complete regimen, comprising S3-Lip-DOX, S3-P-NK, and Atezolizumab/IgG1, demonstrated potent synergistic antitumor effects in vitro and in vivo. This triple-combination therapy not only achieved significant tumor regression but also robustly reprogrammed the innate tumor microenvironment, evidenced by enhanced dendritic cell (DC) maturation and pro-inflammatory macrophage activation. This work establishes a mechanism-driven, modular therapeutic platform that effectively coordinates targeted chemotherapy with innate immunotherapy, holding significant translational potential for solid tumors.

Polyethylene glycol (PEG) carriers can improve drug circulation, but encounter biocompatibility and tumor penetration challenges. The CD47-SIRPα interaction on macrophages can initiate a “don't-eat-me” signal, inhibiting phagocytosis. This study elucidates the dichotomous role of the CD47-SIRPα axis in conferring phagocytosis resistance and transport assistance for enhanced nanocarrier biocompatibility and tumor penetration. Using CD47-functional peptide, we elucidated the capacity of this axis to preserve carrier-cell membrane accessibility, impede macrophage-mediated nanocarrier endocytosis, reduce the secretion of IgG and IgM antibodies, and attenuate complement cascade activation. These mechanisms collectively neutralize the accelerated blood clearance of PEGylated liposomes. Notably, we identified the presence of SIRPα in endothelial vasculature and, for the first time, verified its pivotal role in orchestrating liposomal transit across the endothelial barrier. Moreover, within the tumor region, the CD47-SIRPα axis facilitated carrier hitchhiking on macrophages, enabling deep penetration into the tumor parenchyma and regulating the tumor microenvironment through the differential recognition of M1/M2-type tumor-associated macrophages. This study presents the first evidence of the dichotomous role of the CD47-SIRPα axis in regulating carrier biocompatibility, offering insights into its function to overcome the tumor permeability barrier challenge.

Intranasal vaccines specifically eliciting mucosal immunity in the upper respiratory tract have shown advantages in protecting against respiratory virus invasion. Yet, no clinically licensed intranasal adjuvant remains a major hurdle for the development of intranasal vaccines with low immunogenic antigens like subunit vaccines. Here, we show that liposomes loading simvastatin (Lipo-SV) serve as potent mucosal adjuvants for the intranasal liposomal subunit vaccine encapsulating the hemagglutinin 1 (HA1) glycoprotein of A/PR/8/34 (PR8) H1N1 influenza (Lipo-HA1), providing robust protection against the lethal PR8 H1N1 infection. Compared to cholera toxin subunit B (CTB), the only mucosal adjuvant used in humans, the Lipo-SV substantiate intranasal Lipo-HA1 vaccines to elicit robust systemic and local mucosal immune responses. The underlying mechanism of the adjuvanticity of Lipo-SV involves the increased transcytosis of antigens by inhibiting the geranylgeranylation of RAB5 and RAB7B GTPases in nasal epithelial cells. Moreover, Lipo-SV enhance the submucosal recruitment of dendritic cell for antigen uptake via the Toll-like receptor 4-dependent pathway. Unlike CTB, intranasal Lipo-SV do not induce inflammation in the lung or the inflammatory cytokines in the central nervous system. Our results present a paradigm of design of mucosal adjuvant to target the mucosal epithelial cells in addition to the antigen-presenting cells.

Bacterial enteritis is a specific gastrointestinal tract disorder caused by pathogenic bacterial infection, which not only disrupts the commensal microbiota but also contributes to cascaded complications. Here, we prepared polyethyleneimine (PEI)-based mesoporous silica nanostructures, co-modified with -SH and -S-S- groups, to simultaneously eradicate the pathogenic bacteria, regulate the immune response, and reprogram the inflammatory microenvironment in the infected intestine. Referring to the multivalent sulfur modification, the -S-S- group, with its oxidizability, perturbs the glutathione balance within bacteria, while the combined reductive capacity of -SH and -S-S- scavenges excessive reactive oxygen species and mitigates inflammation-induced damage. Additionally, the well-developed nanopores with a positively charged PEI network facilitate the absorption of bacterial lipopolysaccharide, lipopeptides, flagella and cell-free DNA through hydrogen bonding and electrostatic interactions. Furthermore, the biosilica nanostructures enable the efficient encapsulation of conventional antibacterial agents, such as berberine chloride and norfloxacin, thereby achieving targeted delivery and reducing side effects, which represents a promising strategy for next-generation antimicrobial therapies.

Prodrug nanoassemblies offer an innovative approach to drug delivery, but their lysosomal entrapment often impairs drug release. Notably, tertiary amine structures can undergo protonation reactions, thereby facilitating lysosomal escape through the proton sponge effect. In this study, we developed three novel paclitaxel prodrug nanoassemblies (PTX-SS-NO NPs, PTX-SS-CC NPs and PTX-SS-NC NPs) featuring distinct heterocyclic tertiary amine structures to investigate structure-activity relationships in lysosomal escape and drug delivery. Among them, PTX-SS-NC NPs demonstrated excellent lysosomal escape capability, enabling rapid drug release into the cytosol. Systematic evaluation revealed that the PTX-SS-NC NPs exhibited optimized pharmacokinetics and significant tumor accumulation, further contributing to their strong antitumor efficacy. Our findings establish heterocyclic tertiary amines as crucial design elements for overcoming lysosomal entrapment and optimizing chemotherapeutic prodrug nanoassemblies.

The blood-brain barrier (BBB) is a crucial biological interface between the central nervous system and circulation, playing a key role in maintaining the brain's homeostasis and function, but presenting a challenge for drug delivery. Although enhancing the paracellular permeability of the BBB using graphene-based materials (GBMs) is a promising strategy, detailed biological effects and molecular mechanisms remain poorly understood and understudied. In this study, we prepared gold nanoparticle-modified reduced graphene oxide (Au-rGO) sheets as representative GBMs to enhance BBB paracellular permeability transiently. Remarkably, the induced permeability modulation exhibited both significant efficacy and complete reversibility. Biological experiments and molecular dynamics simulations provided full insights into the molecular mechanism of this process, confirming that Au-rGO influences the dynamic equilibrium of MTs by adsorbing soluble tubulins to hinder MT polymerization, and disassembling MTs to promote depolymerization, thus mediating the unlocking of the BBB. Moreover, the subsequent closure of the BBB is mediated by increased MT polymerization, regulated by the activation of the ERK-stathmin pathway. Collectively, this present study comprehensively elucidates the phenomenon of reversible BBB opening induced by Au-rGO and its intricate molecular mechanism, and supports the safety and efficacy of Au-rGO as a paracellular penetration enhancer for enhancing BBB drug delivery efficiency.

Pyroptosis is a unique programmed cell death pattern, and targeting it is an effective strategy against cancer therapy by overcoming apoptosis resistance. However, Golgi apparatus-targeted aggregation induced emission (AIE) photosensitizer as pyroptosis inducer for efficient antitumor treatment has not been reported. In this study, we successfully synthesized three new AIEgens, including TMN, TBN and TCN, by changing functional groups through a reasonable molecular design strategy, which targeted mitochondria, lysosome and Golgi apparatus (GA), respectively. In vitro experiments demonstrated that TCN exhibited the strongest reactive oxygen species (ROS) production ability and significant phototoxicity. Therefore, TCN as the GA-targeted AIE photosensitizer wore biomimetic hybrid extracellular vehicles (EVs) and M1-type macrophage membranes (denoted as EM@TCN) as pyroptosis inducer were rationally designed and engineered to trigger the production of GA cytotoxic ROS in situ. EM@TCN plus white light irradiation caused GA oxidative stress and induced pyroptosis synergistic photoimmunotherapeutic, which could rebuild tumor microenvironment and improve tumor immunogenicity. Combined with αPD-L1 (anti-mouse PD-L1 antibody), the biomimetic hybrid delivery system EM@TCN significantly inhibit the both primary and distant tumors, and effectively suppress the orthotopic breast tumor. This is the first report on a hybrid nanovesicle coated GA-targeted AIE photosensitizer to induce pyroptosis for combination with αPD-L1 to enhance antitumor photoimmunotherapy.

Abnormal tumor vasculature greatly accelerates tumor progression and diminishes antitumor treatments. Restoring perivascular NO gradients is available to maintain tumor vessel homeostasis and promote tumor vascular normalization. However, exogenously delivering NO strategies lacks the durability to maintain precise NO localization around tumor vessels. Herein, we design a lipid nano delivery system (MC@L) and exploit endothelial transcytosis to deliver metformin (Met) and CaO₂ into tumor vascular endothelial cells (ECs) and tumor cells for achieving tumor vascular normalization-boosted antitumor immunotherapies. The Ca²⁺ and Met released in ECs could restore perivascular localization of NO by activating endothelial NOS (eNOS). Additionally, MC@L internalized by tumor cells could cause CaO₂-induced immunogenic cell death (ICD), together with hypoxia relief and acid neutralization mediated by O₂ generation and H⁺ consumption during CaO₂ degradation, thus further improving the immune effector cell functions under the accompaniment of Met-mediated inhibition of tryptophane uptake in tumor cells. Such a lipid nano delivery system greatly increases the susceptibility of 4T1 tumor-bearing mice to PD-L1 blockade efficacy.

Intratumoral bacteria, especially Gram-positive bacteria (G⁺), have a unique bacterial niche in breast cancer that promoted tumor progression. However, the effects of G⁺ have so far been overlooked, serving an “invisible driver” of breast cancer. Moreover, due to the altered biological structure of G⁺ in tumor cells and the penetration barrier of antibiotics, the effect of antibiotic-mediated eradication of G⁺ in tumors is limited. Here, to simultaneously inhibit intratumoral G⁺ and tumor cells via ferroptosis therapy, an amorphous nano-assembly (DFTV) was constructed by assembling doxorubicin (DOX), tannic acid (TA), FeSO₄, and vancomycin (Van). DFTV treatment effectively targets intratumoral G⁺, thereby inhibiting the growth of the breast tumor and postoperative recurrence by downregulating the expression of inflammatory cytokines, including interleukin-6 (IL-6), interleukin-1β (IL-1β), and tumor necrosis factor-alpha (TNF-α). Moreover, inhibiting intracellular G⁺ also restrains the reorganization of F-actin to form pseudopodia, thereby impairing tumor cell motility and blocking metastasis. Collectively, DFTV improves the antitumor efficacy by targeting G⁺ in breast tumors, offering novel insights into overcoming the limitations associated with the lack of intratumoral antibacterial therapy in clinical breast cancer treatment protocols.

Salvia miltiorrhiza is a prominent traditional Chinese medicinal (TCM) herb with notable therapeutic applications, especially in treatment of cardiovascular and cerebrovascular diseases. A crucial role in promoting tanshinone biosynthesis is attributed to the gibberellic acid (GA) signaling pathway; however, its precise regulatory mechanisms remain incompletely understood. In the current investigation, we identified bHLH130, a GA-responsive bHLH transcription factor (TF), via transcriptomic analysis of GA-treated hairy roots from S. miltiorrhiza. Functional analysis demonstrated that bHLH130 overexpression markedly suppressed tanshinone biosynthesis, whereas its knockout resulted in elevated tanshinone accumulation. Additionally, it was confirmed that bHLH130 directly targets E-box motifs within the promoters of biosynthetic genes, including DXS2, CPS1, KSL1, and CYP76AH1, thereby negatively regulating their transcriptional activities. Moreover, bHLH130 interacts with DELLA4, forming a regulatory complex implicated in the modulation of tanshinone biosynthesis. Co-overexpression assays revealed that DELLA4 attenuated the inhibitory effects of bHLH130 on tanshinone accumulation. Collectively, our data propose that the bHLH130-DELLA4 interaction constitutes a critical regulatory node, balancing GA signaling with secondary metabolite production and offering novel strategies for the metabolic engineering of tanshinones. In conclusion, this research delineates the regulatory role of bHLH130 in tanshinone synthesis, providing valuable insights into the GA-mediated modulation of secondary metabolism.

Respiratory syncytial virus (RSV) is a major global health threat, causing severe respiratory disease in infants, the elderly, and immunocompromised individuals—often surpassing the impact of influenza. Yet, effective RSV therapies remain limited. We describe anti-RSV ssDNA aptamers, selected via Systematic Evolution of Ligands by Exponential Enrichment (SELEX), that bind the RSV glycoprotein (G) with nanomolar affinity. These aptamer-based therapeutics demonstrated potent antiviral activity against RSV. We further established that the binding domains of these aptamers are critical for their antiviral function. Our findings highlight highly active aptamers as a promising strategy to combat RSV infections.