Zigang Li

Pachymic acid (PA), a major bioactive component of Poria cocos, suppresses colorectal cancer (CRC) proliferation and invasion by disrupting glycolysis. However, its antitumor activity and underlying mechanisms in vitro and in vivo remain unclear. RNA-binding proteins (RBPs) are key post-transcriptional regulators in CRC, but their mechanisms of action and therapeutic potential remain poorly understood. Here, we employed limited proteolysis-mass spectrometry (LiP-MS), a target deconvolution approach that does not require prior chemical modification, to profile changes in protein conformation within complex proteomes upon PA treatment in cells. Our findings revealed that RBMX (RNA-binding motif protein X-linked) acts as a target of PA and undergoes concentration-dependent conformational changes, as supported by binding validation and functional assays performed in vitro and in vivo. Furthermore, proximity-labeling mass spectrometry (PL-MS) analysis revealed that RBMX interacts with heterogeneous nuclear ribonucleoprotein C (hnRNPC). Functional assays demonstrated that PA treatment induces RBMX-dependent degradation of hnRNPC, thereby suppressing CRC cell proliferation and tumor growth. In summary, this study identifies RBMX as a molecular target of PA through LiP-MS, providing new insights into the mechanisms underlying PA-mediated CRC suppression and highlighting the RBMX–hnRNPC axis as an important mediator of its antitumor effects.

Proteolysis targeting chimera (PROTAC) technology has received extensive attention due to its “event-driven” mechanism of action. However, existing chimera molecules tend to have higher molecular weight and higher polar surface area and exhibit lower druggable properties. In order to effectively improve the bioavailability, circulation time and other proprietary drug parameters of “chimera molecules”, we combine the characteristics of nanotechnology and chimera technology, and propose the strategy of “Split-and-Mix” proteolysis degradation (SM-PROTAC) technology. However, both peptide- and liposome-based SM-PROTAC showed suboptimal in vivo efficiency. Hence, in this study, we incorporated Polylactic acid (PLA), an FDA-approved biomedical material, as the self-assembled matrix for SM-PROTAC to enhance in vivo effectiveness. PLA-based SM-PROTACs successfully degrade model targets including BRD4, ERα, and CDK4 in cellular assays and display more obvious tumor inhibition in vivo in female mice with a lower target ligand content (~1/10 of small molecule PROTAC controls) and long-term therapeutic potential (1 injection per three days vs 1 injection every day). In summary, PLA-based SM-PROTACs demonstrate promising drug characteristics and present obvious advantages in low-dose and long-term applications, providing additional insights into the pharmaceutical potential of the SM-PROTAC strategy.

RNA molecules play critical roles in various diseases by regulating transcription, translation, and protein expression at the genetic level, positioning RNA-targeted therapeutics as a transformative frontier in precision medicine. Ribonuclease-targeting chimeras (RiboTAC) have emerged as a promising modality by leveraging endogenous RNases (e.g., RNase L) for RNA degradation, offering superior pharmacokinetics, enhanced specificity, and sustained therapeutic effects compared to conventional approaches. Despite their potential, conventional RiboTAC development remains constrained by laborious screening processes and structural complexity, limiting its clinical translation. In this study, we first proposed split-and-mix-RiboTAC based on liposome self-assembly (LipoSM-RiboTAC) to achieve RNA degradation for cancer therapy. This self-adjustable platform possesses self-optimized biomolecule spatial recognition for facile screening, with rapid cellular internalization, robust activation of endogenous RNase L, and precise degradation of oncogenic RNAs. Collectively, we have established an efficient and low-toxicity platform for targeted RNA degradation, which holds promise for cancer therapy and offers novel insights into precision gene therapy.

Maintaining skeletal muscle mass is crucial for health, as muscle atrophy caused by drugs, cancer, or aging poses serious risks. However, there are few effective pharmacological interventions targeting muscle atrophy, highlighting the need for new therapeutic strategies. In this study, in vivo self‐assembled siRNA is designed to silence myostatin (MSTN), a key regulator of muscle growth and atrophy, aiming to prevent muscle atrophy. Using synthetic constructs and the host liver as a scaffold, the assembly of MSTN‐siRNA is guided into muscle‐specific peptide MSP‐tagged small extracellular vesicles (sEVs). These MSP‐tagged sEVs selectively deliver MSTN‐siRNA to muscle tissue. Treatment significantly reduces MSTN protein levels in skeletal muscle, promotes muscle mass gain in healthy mice, and protectes skeletal muscles from atrophy in cancer‐ and dexamethasone‐induced muscle atrophy models. Notably, the sEV‐encapsulated MSTN‐siRNA is produced in a nontoxic, nonimmunogenic, and biocompatible manner. This study offers a promising therapeutic approach for muscle atrophy, addressing a key gap in current treatment options and potentially improving outcomes for patients with muscle‐wasting conditions.

A concise enantioselective total synthesis of leptosphine C (1) has been achieved. Our synthetic strategy features an unprecedented diastereo-/enantioselective Hajos-Parrish-Eder-Sauer-Wiechert (HPESW) reaction for construction of the AB ring system with continuous stereogenic centers (76% yield, 90% ee, dr = 19:1) and an uncommon Pauson-Khand (PK) reaction of geminal dimethyl alkenyl side chains for establishing the CD ring system with the quaternary carbon at C14.

PROTAC, as a highly valued bifunctional molecule, provides a new approach for the drug development of traditional undruggable targets, which holds significant potential in cancer therapy and other disease treatments. However, due to the complicated design process and hard synthesis task, the development of traditional PROTAC drugs is both complex and costly, especially for the design of multitargeting PROTACs. Herein, considering the flexibility and efficacy of the split-and-mix strategy, we proposed a novel Multi-SM-PROTAC platform capable of stable performance with over three ligands simultaneously. In this study, the self-assembly advantage of the split-and-mix strategy was utilized to easily achieve: (1) the simultaneous degradation of dual targets, (2) multi-E3 ligase-mediated degradation of a single target, and (3) multi-E3 ligase-mediated degradation of dual targets. By successfully degrading multiple proteins, it was proven that the Multi-SM-PROTAC platform could achieve multimodule selection and programming for specific therapeutic goals, demonstrating its broad application prospects in drug discovery.

Bioactive peptides have become strong candidates for a variety of clinical therapies due to their diverse advantages, which promote the development of deep generative models for peptide generation. Considering that existing methods cannot effectively deal with the conformational flexibility of peptides and find it difficult to capture accurate residue-to-residue interaction dependencies, we propose a unified peptide generative framework, PepGenWL, via a weakly order-dependent autoregressive language model and lifelong learning. PepGenWL introduces tolerance for out-of-order input as an inductive bias into the autoregressive language model, coupled with Mixture-of-Experts-style plugins to maintain the optimal trade-off between memory stability and learning plasticity across multiple rounds of fine-tuning. The superiority of PepGenWL was demonstrated by generating three classes of therapeutic peptides, including antimicrobial peptides, anticancer peptides, and peptide binders. Under performance evaluation on raw and permuted peptide sequences, PepGenWL not only surpassed state-of-the-art baseline models across the board but also exhibited a significant propensity to incorporate specific residues that are beneficial for antimicrobial or anticancer bioactivity. Furthermore, the property-guided peptide binder generation, screening, and in vitro experimental validation pipeline was presented, achieving a target binding rate of 28.6% with binding specificity. More importantly, the applicability of PepGenWL can be broadened to encompass the peptide SMILES chemical space, thereby facilitating the generation of chemically modified peptides as well as cyclic peptides. Overall, PepGenWL is a unified framework for general-purpose peptide generation that can be flexibly customized for different task requirements.

N6-methyladenosine (m6A) is an essential RNA modification that regulates gene expression and influences diverse cellular processes. Yet, fully characterizing its transcriptome-wide landscape and biogenesis mechanisms remains challenging. Traditional next-generation sequencing (NGS) methods rely on short-reads aggregation, overlooking the inherent heterogeneity of RNA transcripts. Third-generation sequencing (TGS) platforms offer direct RNA sequencing (DRS) at the resolution of individual RNA molecules, enabling simultaneous detection of RNA modifications and RNA processing events. In this study, we introduce SingleMod, a deep learning model tailored for precise detection of m6A modification on individual RNA molecules from DRS data. SingleMod innovatively employs a multiple instance regression (MIR) framework, leveraging extensive methylation-rate labels provided by the quantitative NGS-based method, and achieves ROC AUC and PR AUC of ~0.95 for single-molecule m6A prediction. Applying SingleMod to human cell lines, we systematically dissect the transcriptome-wide m6A landscape at single-molecule and single-base resolution, characterizing m6A heterogeneity in RNA molecules from the same transcript. Through comparative analyzes across eight diverse species, we quantitatively elucidate three distinct m6A distribution patterns correlated with phylogenetic relationships and suggest divergent regulatory mechanisms. This study provides a framework for understanding the shaping of epitranscriptome in a single-molecule perspective.

Proteolysis targeting chimera (PROTAC) technology has received extensive attention due to its "event-driven" mechanism of action. However, existing chimera molecules tend to have higher molecular weight and higher polar surface area and exhibit lower druggable properties. In order to effectively improve the bioavailability, circulation time and other proprietary drug parameters of "chimera molecules", we combine the characteristics of nanotechnology and chimera technology, and propose the strategy of "Split-and-Mix" proteolysis degradation (SM-PROTAC) technology. However, both peptide- and liposome-based SM-PROTAC showed suboptimal in vivo efficiency. Hence, in this study, we incorporated Polylactic acid (PLA), an FDA-approved biomedical material, as the self-assembled matrix for SM-PROTAC to enhance in vivo effectiveness. PLA-based SM-PROTACs successfully degrade model targets including BRD4, ERα, and CDK4 in cellular assays and display more obvious tumor inhibition in vivo with a significantly lower dosage (~ 1/10 of small molecule PROTAC controls) and long-term therapeutic potential (1 injection per three days vs 1 injection every day). In summary, PLA-based SM-PROTACs demonstrate promising drug characteristics and present obvious advantages in low-dose and long-term applications, providing additional insights into the pharmaceutical potential of SM-PROTAC strategy.

Drug conjugates have emerged as a promising alternative delivery system designed to deliver an ultra-toxic payload directly to the target cancer cells, maximizing therapeutic efficacy while minimizing toxicity. Among these, antibody-drug conjugates (ADCs) have garnered significant attention from both academia and industry due to their great potential for cancer therapy. However, peptide-drug conjugates (PDCs) offer several advantages over ADCs, including more accessible industrial synthesis, versatile functionalization, high tissue penetration, and rapid clearance with low immunotoxicity. These factors position PDCs as up-and-coming drug candidates for future cancer therapy. Despite their potential, PDCs face challenges such as poor pharmacokinetic properties and low bioactivity, which hinder their clinical development. How to design PDCs to meet clinical needs is a big challenge and urgent to resolve. In this review, we first carefully analyzed the general consideration of successful PDC design learning from ADCs. Then, we summarised the basic functions of each component of a PDC construct, comprising of peptides, linkers and payloads. The peptides in PDCs were categorized into three types: tumor targeting peptides, cell penetrating peptide and self-assembling peptide. We then analyzed the potential of these peptides for drug delivery, such as overcoming drug resistance, controlling drug release and improving therapeutic efficacy with reduced non-specific toxicity. To better understand the potential druggability of PDCs, we discussed the pharmacokinetics of PDCs and also briefly introduced the current PDCs in clinical trials. Lastly, we discussed the future perspectives for the successful development of an oncology PDC. This review aimed to provide useful information for better construction of PDCs in future clinical applications.