There is a lack of effective therapies for cholestatic liver injury. Recent studies have shown that pan-inhibitors of histone deacetylases (HDACs) play a protective role in cholestasis and fibrosis, but the dynamic changes and role of histone deacetylase 3 (HDAC3) under these conditions remain unclear. In the current study, the expression of acetylation-modifying enzymes, including HDAC3, in livers from lithocholic acid (LCA)-induced intrahepatic and bile duct ligation (BDL)-induced extrahepatic cholestatic mice, as well as in cholestatic patients, was examined. HDAC3 was aberrantly upregulated in the livers of cholestatic mice and patients. Functional studies in LCA- and BDL-induced cholestatic mouse models using RGFP966, a specific HDAC3 inhibitor, were then performed. Pharmacological inhibition of HDAC3 significantly alleviated liver injury in LCA-induced cholestasis but had no protective effect on BDL-induced cholestasis, as assessed by survival tests, gross examination, H&E staining, serum biochemical assays, cytokine expression, and immunohistochemical staining of immune cell biomarkers. Mechanistically, HDAC3 inhibition upregulated genes involved in bile acid and lipid homeostasis in LCA-induced cholestatic mice, as determined by RNA-Seq and validated by RT‒qPCR and Western blotting. The upregulation of these genes, together with increased levels of acetylation of histone H3 at lysine 9 (H3K9Ac) and lysine 27 (H3K27Ac), was validated in transformed human liver epithelial-2 (THLE-2) cells treated with RGFP966. In summary, this study demonstrated that HDAC3 is a potential therapeutic target for LCA-induced intrahepatic cholestasis, with inhibition exerting protective effects through genes involved in bile acid and lipid homeostasis, possibly via epigenetic regulation.
ERα36, a membrane-associated splice variant of estrogen receptor α, mediates rapid non-genomic estrogen signaling and is frequently overexpressed in aggressive and therapy-resistant breast cancers. Although ERα36 represents a promising therapeutic target, its clinical exploitation remains limited due to the lack of high-affinity and highly specific targeting ligands capable of recognizing its extracellular domain. In this paper, we constructed a shark-derived variable new antigen receptor (VNAR) phage display immune library and isolated two high-affinity VNAR single-domain antibodies (FR6 and TH4) targeting distinct epitopes on ERα36. The TH4 VNAR was further engineered by fusion with a truncated Pseudomonas aeruginosa exotoxin (PE24) to generate a recombinant immunotoxin (TH4-PE24), characterized by (1) high specificity and affinity for ERα36 (KD = 0.751 nmol/L), (2) potent and selective cytotoxicity against ERα36-positive breast cancer cells (IC50 = 1.54 nmol/L), (3) a dual mechanism involving ERα36 degradation and suppression of PI3K/AKT and MAPK/ERK signaling pathways, and (4) significant in vivo antitumor efficacy in xenograft models without observable systemic toxicity. With this immunotoxin, we further investigated how ERα36 targeting could remodel tumor metabolism. ERα36-mediated metabolic regulation in breast cancer is a new research direction with many aspects unknown. We found that TH4-PE24 treatment led to extensive metabolic reprogramming, including disruption of nucleotide metabolism, ABC transporters, and central carbon metabolism. Overall, we report a novel shark VNAR-based immunotoxin platform that can achieve precise and potent antitumor effects by targeting the previously intractable membrane receptor ERα36. The modular design features make it translational for various other studies, such as targeting additional immune checkpoint molecules and developing next-generation precision therapeutics against membrane receptors that are inaccessible to conventional antibody modalities.
Platycodin D (PD), a triterpenoid saponin from Platycodon grandiflorus, has been reported to have cardiovascular activity, but its actions in vascular smooth muscle cells (VSMCs) and atherosclerosis remain unclear. We integrated network pharmacology, molecular docking, 100-ns molecular dynamics simulation, VSMC assays, and two mouse carotid models. In VSMCs cultured under basal conditions, PD reduced viability and BrdU incorporation, downregulated cyclin D1/E1, and induced TUNEL positivity and cleaved caspase-3. At the cellular level, PD activated mTORC1-associated signaling, increased p70S6K and eIF2α phosphorylation, increased ATF4 protein and activated ATF6, impaired autophagic flux, and induced CHOP-associated endoplasmic reticulum stress. Rapamycin and 4-phenylbutyric acid attenuated PD-induced apoptosis, whereas MHY1485 and thapsigargin exacerbated it. Perivascular PD reduced neointimal hyperplasia after complete carotid ligation in C57BL/6J mice. In contrast, PD accelerated early carotid atherosclerosis after partial ligation in ApoE–/– mice fed a high-fat diet. Under ox-LDL challenge, PD increased VSMC lipid accumulation, Olr1/Cd36 expression, mTORC1-p70S6K signaling, and BrdU incorporation; rapamycin limited these effects. These findings suggest that PD activates an mTORC1-associated stress program whose vascular outcome depends on lipid and flow context. PD may have local anti-neointimal potential but warrants caution in hyperlipidemic atherosclerosis.
Acquired resistance to osimertinib remains a critical clinical challenge in EGFR-mutant non-small cell lung cancer (NSCLC), with 40%–60% of progressing patients lacking an identifiable genetic resistance driver. Through quantitative secretomic profiling of paired osimertinib-sensitive and -resistant NSCLC cell lines, we identified laminin subunit γ2 (LAMC2) as markedly enriched in resistant cell secretomes. Single-cell RNA sequencing reanalysis revealed a LAMC2-high drug-tolerant persister subpopulation that expanded under osimertinib pressure, with pseudotime analysis suggesting progressive co-upregulation of LAMC2 and matrix metalloproteinase 14 (MMP14) along the treatment-to-resistance trajectory. Across four independent cohorts, MMP14 exhibited the strongest co-expression with LAMC2 among all metalloproteinases; the two proteins also physically interacted, and their concurrent high expression predicted the worst overall survival. Mechanistically, MMP14 cleaved LAMC2 to release a bioactive Domain III (DIII) fragment harboring EGF-like repeats, which reactivated EGFR/AKT/ERK signaling despite continued osimertinib exposure. Exogenous DIII conferred resistance in sensitive cells, whereas an anti-DIII blocking antibody reversed MMP14-driven resistance in cell lines and patient-derived organoids. The selective MMP14 inhibitor NSC405020 restored osimertinib sensitivity in vitro and suppressed xenograft growth. In residual tumor tissues from neoadjuvant osimertinib-treated patients, non-responders exhibited higher LAMC2/MMP14 co-expression and more prominent spatial exclusion of CD8⁺ T-cells from the tumor core compared with major pathological responders. These findings identify the MMP14–LAMC2–DIII proteolytic axis as a non-genetic resistance mechanism and suggest that MMP14 inhibition or DIII neutralization may represent a rational strategy to overcome acquired osimertinib resistance.
