CRC is one of the leading causes of cancer mortality worldwide. Chemotherapy is widely used for the treatment of CRC, but its efficacy remains unsatisfactory, mainly due to drug resistance. Therefore, it is urgent to develop new strategies to overcome drug resistance. Combination therapy that aims to achieve additive or synergistic therapeutic effects is an effective approach to tackle the development of drug resistance. Given its established roles in tumor development, progression and metastasis, IGF-1R is a promising drug target for combination therapy against CRC. In this study, we revealed that the novel IGF-1R inhibitor PB-020 can act synergistically with mebendazole (MBZ) to reduce the viability of CRC cells and block xenograft CRC progression. Moreover, the PB-020/anti-PD-1 combination synergistically blocked CRC propagation in the MC38 murine colon carcinoma model. Both combination therapies potently suppressed the PI3K/AKT signaling pathway genes in CRC that may be associated with the development of drug resistance. Our findings establish a preclinical proof-of-concept for combating CRC using combined multi-target treatment with PB-020 and clinical anticancer drugs, which may provide useful clues for clinical trials to evaluate the efficacy and safety of these drug combinations in CRC patients.

Deciphering the regulatory network for human naive and primed pluripotency is of fundamental theoretical and applicable significance. Here, by combining quantitative proteomics, phosphoproteomics, and acetylproteomics analyses, we revealed RNA processing and translation as the most differentially regulated processes between naive and primed human embryonic stem cells (hESCs). Although glycolytic primed hESCs rely predominantly on the eukaryotic initiation factor 4E (eIF4E)-mediated cap-dependent pathway for protein translation, naive hESCs with reduced mammalian target of rapamycin complex (mTORC1) activity are more tolerant to eIF4E inhibition, and their bivalent metabolism allows for translating selective mRNAs via both eIF4E-dependent and eIF4E-independent/eIF4A2-dependent pathways to form a more compact naive proteome. Globally up-regulated proteostasis and down-regulated post-translational modifications help to further refine the naive proteome that is compatible with the more rapid cycling of naive hESCs, where CDK1 plays an indispensable coordinative role. These findings may assist in better understanding the unrestricted lineage potential of naive hESCs and in further optimizing conditions for future clinical applications

Naive human embryonic stem cells (hESCs) that resemble the pre-implantation epiblasts are fueled by a combination of aerobic glycolysis and oxidative phosphorylation, but their mitochondrial regulators are poorly understood. Here we report that, proline dehydrogenase (PRODH), a mitochondria-localized proline metabolism enzyme, is dramatically upregulated in naive hESCs compared to their primed counterparts. The upregulation of PRODH is induced by a reduction in c-Myc expression that is dependent on PD0325901, a MEK inhibitor routinely present in naive hESC culture media. PRODH knockdown in naive hESCs significantly promoted mitochondrial oxidative phosphorylation (mtOXPHOS) and reactive oxygen species (ROS) production that triggered autophagy, DNA damage, and apoptosis. Remarkably, MitoQ, a mitochondria-targeted antioxidant, effectively restored the pluripotency and proliferation of PRODH-knockdown naive hESCs, indicating that PRODH maintains naive pluripotency by preventing excessive ROS production. Concomitantly, PRODH knockdown significantly slowed down the proteolytic degradation of multiple key mitochondrial electron transport chain complex proteins. Thus, we revealed a crucial role of PRODH in limiting mtOXPHOS and ROS production, and thereby safeguarding naive pluripotency of hESCs.

The transcription factor OCT4A, a well-established master pluripotency factor, exerts regulatory effects on cell fate determination that are closely associated with its protein levels. This study aims to uncover the downstream gene profile features relevant to tumorigenic potential mediated by OCT4A under varying protein abundance in somatic cancer cells (SCCs). CRISPR-Cas9-mediated knockout and doxycycline-inducible OCT4A expression systems were established in cervical (HeLa) and hepatocellular (HepG2, Huh7) cancer cells. Single-cell sequencing, spatial transcriptomic and survival analysis data were used to elucidate the expression pattern of OCT4 in somatic cancer tissues and its prognostic relevance. The plate colony formation assay was performed to assess the tumorigenic capacity of SCCs, and Bulk RNA sequencing coupled with weighted gene co-expression network analysis (WGCNA) identified dose-relevant downstream pathways. Functional enrichment, survival modeling, and RT-qPCR validation were used to construct OCT4A-dose-dependent transcriptional regulatory networks. OCT4 transcript, is heterogeneously present and confined to a small subset of tumor cells within somatic cancer tissues, with a significantly higher proportion of OCT4-positive cells in tumor tissues compared to paired paraneoplastic tissues and is significantly correlated with poor prognosis in SCCs. Endogenous low-level OCT4A positively regulates tumorigenic capacity predominantly through targeting non-coding genes, whereas high-level OCT4A suppresses tumorigenic capacity primarily via protein-coding genes in SCCs. A prognostic model based on high-level OCT4A-regulated protein-coding genes was associated with favorable clinical outcomes, aligning with in vitro phenotypic results. Our findings further confirm in SCCs that the functional pleiotropy of OCT4A is closely linked to its protein abundance, and further systematically elucidate the molecular signatures of OCT4A-regulated downstream gene networks associated with tumorigenic phenotypes at differential protein levels, providing novel insights for its translational exploitation in both oncological intervention and regenerative medicine strategies.

Breast cancer stem cells (BCSCs) drive therapeutic resistance, recurrence, and metastasis via self-renewal, phenotypic plasticity, and adaptive tolerance to treatment and microenvironmental stress. Extracellular vesicles (EVs) mediate bidirectional intercellular communication between stem-like tumor cells and tumor, stromal, immune and vascular compartments, but their translation is blocked by critical unaddressed gaps: BCSC rarity and heterogeneity, incomplete donor-cell validation, uncertain vesicle-source attribution, and nonstandardized assays. This review outlines a unifying bioengineering framework built around source-resolved analysis to integrate BCSC enrichment, EV characterization, functional validation and translational development and evaluate current platforms for BCSC isolation (including flow cytometry, magnetic enrichment, microfluidic sorting, affinity capture, and single-cell platforms), EV purification, biosensing, single-vesicle profiling, and multimarker detection, alongside organoid and tumor-on-chip models for dissecting EV-driven resistance and metastatic niche remodeling. BCSC-associated EV signatures complement circulating tumor cells and circulating tumor DNA for noninvasive monitoring of stemness dynamics, treatment adaptation, and recurrence risk, while engineered EVs offer a unique targeted delivery route to resistant tumor states. Real-world deployment will require direct head-to-head comparison with clinically established nanocarriers for efficacy, manufacturability, biodistribution and safety. The eventual clinical translation of this field hinges on validated BCSC identity, reliable EV source attribution, standardized detection protocols, causal mechanistic validation, and scalable manufacturing.