Acta Pharmaceutica Sinica B

Deuterium modification of tyrosine kinase inhibitors contributes to reversing ferroptosis resistance through upregulation of aldehyde oxidase 1 in hepatocellular carcinoma

Acta Pharmaceutica Sinica B. 2026, 16(2): 802-819.

The susceptibility to ferroptosis partially determines the efficacy of tyrosine kinase inhibitors (TKIs) in hepatocellular carcinoma (HCC), exposing a mechanistic vulnerability that can be therapeutically exploited. The development of deuterated compounds is a promising strategy for the improvement of anti-tumor efficacy. Here, we identified HCC with higher level of ferroptosis-resistance exhibited insensitive to TKIs, which could be reversed by deuterated TKIs. Aldehyde oxidase 1 (AOX1) was screened as a critical gene mediating the responsiveness to deuterated TKIs-induced ferroptosis in HCC. The presence of a pyridyl tri-deuterated methanamide contributed to the upregulation of AOX1 in a structure-dependent manner, thereby promoting ferroptosis. Mechanistically, AOX1 inhibited sirtuin 6-mediated deacetylation of H3K9 and H3K56, leading to transcriptional activation of acyl-CoA synthetase long chain family member 5, which resulted in poly-unsaturated fatty acids hyperaccumulation-induced ferroptosis. Additionally, HCC with lower AOX1 expression conferred better efficacy to deuterated TKIs. In patient cohorts with HCC, those with lower AOX1 expression exhibited a more pronounced therapeutic response to deuterated sorafenib. Overall, the present study elucidates the mechanism by which deuterated TKIs reverse TKI resistance by promoting ferroptosis and suggests that AOX1 could serve as a biomarker to guide clinical decision-making for deuterated TKI treatment in HCC.

Activating protein kinases to treat diseases: Current understanding and future challenges

Peng Jin, Minru Liao, Huidi Liu, Kun Huang, Leilei Fu, Xin Jin

Acta Pharmaceutica Sinica B. 2026, 16(2): 746-769.

Protein kinases, as one of the most important human enzymes, are signaling molecules that regulate almost all cell activities, including growth, cell division and metabolism. Dysfunction of these cellular pathways can lead to a variety of human diseases. Accumulating evidence on the down-regulation of key protein kinases in diseases has made a big progress. The down-regulation is related to cancer, heart disease, neurodegenerative diseases and other diseases. Thus, in this review, we defined the classifications of protein kinases and demonstrated the mechanisms of protein kinase activators in the treatment of human diseases, summarized the research progress of protein kinase activators, and further discussed the development progress of protein kinase activators in clinical stage. Accordingly, activation of protein kinases has become a crucial target for drug development. With the in-depth understanding of protein kinase functions and regulation mechanisms, the development of new protein kinase activators may continue to be a rapidly growing field, which will help to develop more accurate and effective targeted therapeutic strategies in the near future.

From concept to application: Exploring the evolution and potential of DUBTAC technology

Danfeng Wang, Wenjian Min, Binjian Jiang, Haopeng Sun, Chengliang Sun, Peng Yang

Acta Pharmaceutica Sinica B. 2026, 16(2): 770-787.

Deubiquitinase-targeting chimeras (DUBTAC), as a highly promising emerging technology, can precisely remove ubiquitin chains from target proteins by recruiting deubiquitinases (DUBs), thereby enhancing the stability of the target proteins. Multiple functional proteins, such as the tumor suppressor proteins p53, RB, PTEN, upon stabilization by DUBTAC, can effectively restore or enhance their physiological functions, thus achieving therapeutic effects. Currently, the DUBTAC technology is still in its early stage of development, yet it has broad application prospects and represents a technological approach for developing various “undruggable” targets. This article delves into the design strategy of DUBTAC, and screens and recommends some candidate proteins with the potential to serve as drug targets. We aim to provide perspective in drug design, structural optimization, target selection, and related aspects.

Machine learning empowered formulation design, optimization and characterization of nanoparticulate drug delivery systems: Current applications, challenges, and future perspectives

Acta Pharmaceutica Sinica B. 2026, 16(2): 665-685.

Nanoparticulate drug delivery systems (NDDS) have revolutionized modern medicine by significantly improving drug targeting, bioavailability, and therapeutic efficacy. Despite the clinical success of over 90 approved nanomedicines, the development of NDDS remains challenging due to the complexity of formulation design, optimization, and characterization processes. Artificial intelligence, particularly machine learning (ML), offers powerful data analytics and predictive capabilities that can address these challenges. This review systematically summarizes recent advances in ML applications across various NDDS formulations, including polymeric nanoparticles, lipid nanoparticles, liposomes, solid lipid nanoparticles, nanostructured lipid carriers, nanoemulsions, nanosuspensions, lipid-based hybrid NDDS, self-emulsifying drug delivery systems, niosomes, and nanocrystals. We also summarize how ML algorithms could help predict critical quality attributes of NDDS, such as particle size, shape, surface properties, drug encapsulation efficiency, drug loading efficiency, drug release behavior, and stability. Furthermore, we discuss existing challenges and prospects for the formulation development empowered by ML in NDDS. In conclusion, this review provides a comprehensive overview of the transformative potential of ML in improving the formulation development of nanomedicines, ultimately accelerating their clinical translation.

Applications of AI/ML in accelerating the development of pulmonary drug delivery system

Acta Pharmaceutica Sinica B. 2026, 16(2): 686-708.

Artificial intelligence (AI) is a transformative technique for drug development, and it has been widely applied in pharmaceutical industry and academia. Pulmonary drug delivery systems (PDDS) are preferred for treating respiratory diseases due to their ability to provide localized and rapid action with fewer side effects. The integration of AI and Machine Learning (ML) has significantly accelerated the development of PDDS by enhancing both respiratory disease detection, and different stages during PDDS development. This paper provides an overview of the present landscape by literature analysis of the key areas of research. This review first introduces the fundamental principles of AI/ML and how they are applied in respiratory disease detection and diagnostics, highlighting FDA-approved software used in this field. Furthermore, we examine the role of AI in different stages during the development of PDDS, from identifying novel drug candidates to optimizing formulations and drug delivery mechanisms. The review also discusses regulatory and ethical considerations, along with existing challenges during AI-driven PDDS development. By addressing these key aspects, we provide insights into the revolutionary potential of AI/ML in advancing pulmonary drug delivery and improving therapeutic outcomes.

Artificial intelligence-guided design of lipid nanoparticles for mRNA delivery

Kexin Su, Junjie Qiu, Tengfei Xu, Shuai Liu

Acta Pharmaceutica Sinica B. 2026, 16(2): 709-727.

Lipid nanoparticles (LNPs) hold significant potential for mRNA-based therapeutics, as evidenced by their successful use in SARS-CoV-2 mRNA vaccines. LNPs effectively protect and transport mRNA to target sites, thereby ensuring its stability and efficient transfection. Despite the progress, some challenges remain in the development of mRNA-LNP delivery systems, such as limited targeting specificity, the complexity of formulations, and the time-consuming and high-throughput screening process. Artificial intelligence (AI) has emerged as a powerful tool to address these challenges, accelerating the design and optimization process of LNPs. AI-guided approaches can improve the efficiency of lipid structure and formulation screening by rapidly identifying key design parameters and employing predictive modeling to optimize LNP properties. The combination of AI and LNP technology offers significant advantages, including enabling the design of more personalized and precise delivery systems, streamlining the development process, and reducing the cost. This review discusses recent advancements in AI-guided mRNA-LNP delivery systems and highlights their potential to revolutionize mRNA therapeutics.

Unraveling the FGFR-RNA splicing axis: Mechanisms, oncogenic crosstalks and innovations for therapeutic purpose

Acta Pharmaceutica Sinica B. 2026, 16(1): 35-61.

Fibroblast growth factor receptor (FGFR) signaling is a pivotal regulator of tumor progression, driving cell proliferation, survival, metastasis, and therapeutic resistance across diverse cancer types. RNA alternative splicing profoundly shapes FGFR isoform diversity, endowing tumors with heterogeneity and adaptability to targeted interventions. While significant progress has been made in identifying splicing regulators that govern FGFR pre-mRNA processing, the extracellular cues influencing this process and the reciprocal impact of FGFR signaling pathway on global splicing networks remain underexplored. This review provides a comprehensive overview of the bidirectional interplay linking FGFR signaling and RNA splicing in cancer. Mechanistically, we first detail how FGFR mutations, epigenetic modifications, and crosstalks with oncogenic pathways reprogram splicing to generate tumor-specific FGFR splice variants. We then systematically classify distinct FGFR isoforms and delineate how they contribute to main cancer hallmarks, underscoring the central role of the FGFR-splicing axis in driving tumor plasticity, heterogeneity and adaptive progression. Conversely, we also examine how FGFR signaling modulates RNA splicing programs beyond FGFR itself, reshaping global splicing events that contribute to tumorigenesis, an emerging and still largely unexplored area of cancer biology. From therapeutic perspective, we highlight emerging strategies targeting the axis. Notably, FGFR splicing isoform-directed radiopharmaceuticals hold great promise for patient stratification and biomarker-directed theranostics, providing a precise approach to identify aggressive tumors and guide tailored interventions. As well, complementary approaches, including CRISPR/Cas9-based splicing modulation and long non-coding RNAs-targeted therapies, further expand the toolbox for isoform-specific intervention. Moreover, integrating splicing modulators with FGFR TKIs may overcome drug resistance. Understanding the intricate interplay between FGFR signaling and RNA splicing will not only advance biomarker-guided therapeutic development but also provide a novel framework to counteract tumor adaptability, ultimately improving outcomes in FGFR-driven malignancies.

Computational approaches to druggable site identification: Current status and future perspective

Acta Pharmaceutica Sinica B. 2026, 16(1): 62-92.

With the rapid advancements in computer technology and bioinformatics, the prediction of protein-ligand-binding sites has become a central component of modern drug discovery and development. Traditional experimental methods are often constrained by long experimental cycles and high costs; therefore, the development of accurate and efficient computational methods is of paramount significance for conserving time and cost. This review comprehensively summarizes the methodological advancements and current applications in the field of screening for druggable protein target sites, systematically comparing the fundamental principles, advantages, and disadvantages of four main categories of methods: structure- and sequence-based methods, machine learning-based methods, binding site feature analysis methods, and druggability assessment methods. Subsequently, by integrating classic case studies, this paper elaborately discusses the technical support and theoretical guidance afforded by the screening of protein druggable target sites for drug discovery and drug repositioning. Finally, this paper thoroughly explores the current challenges inherent in the field of protein-ligand binding site prediction, with a particular focus on future technological trends, systematically elucidating the developmental prospects and potential applications of these predictive methods.

Targeting glial cells: Unveiling the neuroprotective mechanisms of Ginseng in the brain microenvironment

Acta Pharmaceutica Sinica B. 2026, 16(1): 13-34.

The burden imposed by central nervous system disorders (CNSD) on global health is substantial, characterized by a significant impact on quality of life, increased mortality rates, and escalating economic costs. Glial cells, primarily comprising astrocytes, microglia, oligodendrocytes, and oligodendrocyte precursor cells (OPCs, also known as NG2 cells), play crucial and diverse roles in neurological health and disease. In the treatment of CNSD with traditional herbal medicines, ginseng and its active components have made a notable impression. This comprehensive review investigates the interaction between ginseng and these essential glial cells, detailing their contributions to neurological well-being and disease states. Additionally, it thoroughly assesses the effects of ginseng on glial function, highlighting its neuroprotective potential through anti-inflammatory, antioxidative, and other restorative actions via complex molecular pathways. Moreover, the review analyzes how ginseng can facilitate neuronal viability and regeneration, as well as modulate signaling cascades, thereby highlighting the therapeutic potential of ginseng in the management of CNSD.

Structure-guided design of picomolar-level macrocyclic TRPC5 channel inhibitors with antidepressant activity

Acta Pharmaceutica Sinica B. 2026, 16(1): 371-386.

Recent advances in ion channel structural biology have enhanced structure-based drug design, yet lipid-occupied binding pockets—often large and flat—remain a major hurdle for developing selective small molecules. TRPC5, a brain-enriched channel regulating depression and anxiety, is a promising therapeutic target, but current preclinical candidates suffer from moderate off-target effects. To address this, we designed macrocyclic TRPC5 inhibitors using structure-guided macrocyclization, overcoming lipid-binding site challenges. Among these, JDIC-127 exhibited unprecedented potency with IC₅₀ of 374 pmol/L—200-fold more potent than HC-070—and exceptional selectivity. Its specificity arises from interactions with unique structural features near the S5 and S6 helices of TRPC5, minimizing activity against related TRPC channels and other ion channels. This selective inhibition aligns with preclinical evidence supporting JDIC-127's potential in treating neuropsychiatric disorders. The study demonstrates how macrocycles stabilize ligand conformations, enhance affinity, and achieve selectivity in lipid-dominated binding sites. It also highlights the synergy between macrocyclic design, cryo-EM, and computational modeling to address longstanding obstacles in ion channel drug discovery. JDIC-127 serves as a proof-of-concept for the application of macrocyclization in ion channel pharmacology, offering a roadmap for developing innovative therapeutics targeting TRP channels and beyond, with implications for a wide range of diseases.