搜索结果：Biochimica et biophysica acta

Early zebrafish embryos rely on maternally supplied yolk lipids to fuel growth before the onset of feeding; yet, how these lipids are mobilized and redistributed remains poorly understood. Here, we combine live imaging with the solvatochromic dye Nile Red to map lipid composition in space and time through changes in the emission peak, a readout of local polarity. We show that lipid droplets (LDs) in the blastodisc are highly heterogeneous in size and polarity at the "one-cell stage" but progressively homogenize as development proceeds. LDs originate at the yolk-blastodisc interface, where localized lipase activity drives their biogenesis and initial composition. As LDs migrate toward the animal pole, their polarity increases, reflecting continuous lipolysis within a spatially confined metabolic zone. Smaller LDs display greater lipolytic efficiency than larger ones, linking droplet geometry to metabolic turnover. Inhibition of lipase activity disrupts both LD formation and lipid cycling, demonstrating that shared enzymatic machinery underlies droplet synthesis and degradation. Together, our findings reveal a spatially organized and developmentally regulated lipid metabolism in the zebrafish blastodisc, where local enzyme activity, droplet mechanics, and lipid composition are dynamically coupled. This live-imaging approach establishes a framework for studying lipid regulation in vertebrate development and disease. DIGEST SUMMARY: By visualizing lipids in living zebrafish embryos, this study reveals that lipid droplets are not passive fat stores but dynamic organelles whose size, composition, and behavior are shaped by local enzyme activity at the yolk-blastodisc interface, offering new insight into how developing cells control their energy reserves.

Immune checkpoint blockade (ICB) provides limited benefit in triple-negative breast cancer (TNBC), partly due to a lactate-rich, immunosuppressive tumor microenvironment (TME). Integrative bulk and single-cell transcriptomics, corroborated by tissue analyses, identify MCT4 as the predominant lactate transporter in TNBC, while supporting a context-dependent contribution of MCT1, and reveal an inverse association between MCT4 and intratumoral CD8+ T cells. We therefore tested whether inhibition of monocarboxylate transporters could recondition the TME. The repurposed agent syrosingopine, a dual MCT1/MCT4 inhibitor, consistently blocked lactate export across MCT1+ and MCT4+ models without altering transporter/LDH abundance, and reduced tumor cell PD-L1 while restoring T-cell effector function. Syrosingopine also sensitized tumor cells to NK-cell cytotoxicity without detectable toxicity in immune cells. In immunocompetent TNBC models, syrosingopine synergized with anti-PD-1 to suppress tumor growth, remodel the immune landscape, and prolong survival, while genetic perturbation of MCTs recapitulated key metabolic and immunologic phenotypes. These findings identify tumor lactate export as an important metabolic barrier to antitumor immunity in TNBC and support MCT-targeted lactate transport inhibition as a promising strategy to reprogram the TME and improve immunotherapeutic efficacy.

Venlafaxine (VLX) and its active metabolite O-desmethylvenlafaxine (ODV) must partition into lipid membranes before reaching their membrane-embedded transporter targets. The principal active metabolite, ODV, differs from VLX by the removal of a methyl group and is itself marketed as an antidepressant. Here, we combined long-timescale all-atom molecular dynamics simulations with experimental membrane partitioning measurements to investigate how demethylation, stereochemistry, and drug concentration influence their interactions with phosphatidylcholine bilayers. All drug species spontaneously partitioned into the lipid-water interface and stabilized at ~1.4 nm from the bilayer center while maintaining a conserved amphipathic orientation. Free-energy calculations revealed similar interfacial free-energy minima of approximately -20 kJ mol-1 for both compounds, whereas the energetic penalty for translocation toward the bilayer core exceeded 40 kJ mol-1. Demethylation lowered the energetic barrier for membrane insertion without altering the thermodynamically favored interfacial state, indicating that ODV accesses the membrane interface more efficiently while preserving comparable equilibrium membrane affinity. Experimental second-derivative UV-Vis spectroscopy further confirmed similar membrane partitioning for VLX and ODV. At low drug concentrations, membrane structure remained largely unperturbed following insertion. In contrast, elevated drug loading induced cooperative membrane perturbation characterized by lateral bilayer expansion, membrane thinning, and reduced acyl-chain order, with these effects consistently more pronounced for ODV. Together, these findings identify the lipid bilayer under studied conditions as a dynamically adaptive kinetic interface rather than a passive equilibrium reservoir, suggesting that functional differences between VLX and ODV arise primarily from differences in interfacial accessibility and subsequent protein-specific interactions rather than nonspecific membrane affinity alone.

Post-translational modifications (PTMs) play a key role in regulating protein/protein interactions and protein stability, thus influencing protein expression and function. HSPs and p53, including its wtp53 and mutp53 forms, make no exception to this rule, although the impact of PTMs on the regulation of these proteins has not yet been fully elucidated, particularly in the case of mutp53. These proteins, unlike wtp53, can behave as oncogenes, making their targeting an important step for successful anticancer therapy. We previously reported that mutp53 is degraded, preferentially via CMA, in colon cancer cells stressed by long-term TG treatment. Whether TG could induce PTMs and how they could contribute to mutp53 degradation has not yet been investigated and will be explored in this study. Acetylation of mutp53, as well as HSP90, has been reported to promote mutp53 degradation. However, we found that TG promoted deacetylation of mutp53 and HSP90, due to the sustained activity of HDAC6, a PTM that protected mutp53 from degradation. We then found that mutp53 was progressively demethylated by KDM1 at lysine K370 during TG treatment, which facilitated the interaction with HSC70 involved in mutp53 protein degradation via CMA. In conclusion, this study suggests that, in colon cancer cells subjected to stress by TG, mutp53 was degraded as a consequence of demethylation at lysine K370. Therefore, specific epigenetic drugs capable of reducing constitutive methylation and/or increasing acetylation could preemptively target mutp53 and improve the outcome of endoplasmic reticulum stress-inducing treatments in tumors harboring these proteins.

Tumor heterogeneity is a fundamental feature of malignant tumors and a major driver of treatment resistance, ultimately contributing to treatment resistance, therapeutic failure, disease recurrence, and poor outcome. Rather than being static, heterogeneity arises early during tumorigenesis and continues to evolve during progression, metastasis, and therapeutic selection. Advances in single-cell and spatial omics, integrative multi-omics profiling, liquid biopsy, and computational modeling have markedly improved the characterization of heterogeneity across genetic, epigenetic, transcriptomic, phenotypic, and microenvironmental dimensions. However, translating heterogeneity profiling into routine clinical decision-making remains challenging due to sampling bias, technical variability, data-integration complexity, and the need for prospectively validated, actionable metrics. In this review, we summarize the biological origins and major classifications of tumor heterogeneity, discuss current approaches for its detection and longitudinal monitoring, examine its implications for established and emerging therapies, and highlight precision-oncology strategies aimed at anticipating, tracking, and ultimately exploiting heterogeneity to achieve durable cancer control.

Komagataella phaffii, formerly known as Pichia pastoris, is a methylotrophic yeast employed as a recombinant protein factory for academic and industrial purposes. Despite its advantages, some recombinant proteins, which are engineered to be secreted, are retained intracellularly and subject to degradation. Serendipitously, we discovered that fetal bovine serum (FBS) improved the secretion of several proteins. When a P. pastoris strain expressing enhanced green fluorescent protein (eGFP) was grown with FBS, western analysis revealed a 2-3 fold increased level of eGFP secretion as well as the induction of an upper molecular band (30 kD) in addition to the expected band (26.8 kD). FBS protected the C-terminus of eGFP from degradation, and the alterations to this upper molecular weight variant were triggered intracellularly by an FBS component that was most likely not a protein. To elucidate the mechanisms behind the production of the novel recombinant protein variant, we utilized site-directed mutagenesis and mass spectrometry. Through these strategies, we were able to localize posttranslational modifications to the C-terminus associated with FBS treatment. Analysis of the intracellular lysate revealed proteomic alterations, suggesting that genes involved in translation, trafficking, stress response and energetics were key players in FBS's effect on the P. pastoris secretion mechanism through interaction with eGFP's C-terminus. CONCLUSIONS/. Our study suggests that C-terminal degradation may affect other recombinant proteins produced in P. pastoris, a problem that may be resolved by FBS.

Most deaths from pancreatic ductal adenocarcinoma (PDAC) result from metastatic organ failure rather than primary tumor burden. Across the metastatic cascade, tumor cells encounter hypoxia, nutrient limitation, oxidative stress, immune pressure, and changing stromal conditions, and only those capable of dynamic metabolic adaptation successfully disseminate and colonize distant sites. In this review, we summarize how metabolic plasticity shapes PDAC dissemination, organ-specific colonization, and metastatic outgrowth. We further discuss how this framework may shift therapeutic strategy from broad metabolic blockade toward precision interventions that target metastatic fitness.

KL4, a 21-residue mimetic of the C-terminus of surfactant protein B, is effective in treating infant respiratory distress syndrome (RDS). It works by reducing alveolar surface tension and improving oxygen exchange. Here, we use steady-state fluorescence measurements to probe how KL4 modulates bilayer order and organization in pulmonary surfactant-like lipid mixtures with varying acyl chain saturation. We show how fluorescence-based methods can distinguish between a more surface-associated and a more deeply inserted mode of interaction of KL4 in phospholipid membranes with varying acyl chain saturation. Specifically, in saturated liposomes with a tail-labeled pyrene analog, KL4 addition decreases excimer-to-monomer (Ie/Im) ratios and produces larger changes in the steady-state anisotropy of rhodamine-labeled lipids, consistent with reduced lipid mobility and peptide-induced changes in bilayer packing, and is qualitatively consistent with previous ssNMR/EPR studies of KL4. Conversely, fluorescence results for unsaturated liposomes show smaller changes in Ie/Im and anisotropy, suggesting greater interfacial interaction between KL4 and a more modest impact on bilayer order. Increasing concentrations of KL4 in unsaturated lipids result in a notable decrease in rhodamine headgroup emission intensity, suggesting local environmental shifts or collisional quenching, consistent with a more surface-proximal peptide. Overall, these fluorescence readouts are sensitive to KL4-induced changes in bilayer order, lipid dynamics, and lateral organization, but do not, by themselves, define peptide orientation or insertion depth. These assays require minimal sample and simple equipment, making them practical for monitoring how KL4 and similar surfactant mimics influence bilayer behavior across different lipid compositions and for screening lipid and peptide combinations in future models of healthy and diseased surfactant relevant to neonatal RDS and adult ARDS, including COVID-19. ABBREVIATIONS.

Trans-10, cis-12 conjugated linoleic acid (t10c12-CLA) can affect lipid metabolism, leading to weight loss and attracting widespread attention. However, its suitability for use in lactating mothers remains unclear. A transgenic mouse, named Pai (homozygous) and Pai/wt (heterozygous), capable of producing endogenous t10c12-CLA was successfully established. In this study, embryo transfer technology was utilized to transplant embryos of wild-type (wt) mice into the transgenic mice, in order to investigate the effects of t10c12-CLA on biometric and reproductive parameters. Embryo transfer efficiency remained unchanged following the procedure. The male-to-female ratio of offspring in Pai/wt group was significantly higher than that in wt and Pai groups. Additionally, the birth weight, 21-day body weight, serum glucose level and triglyceride level of pups in Pai group were significantly lower than those in wt group. Simultaneously, the Pai pups exhibited modulated gut microbiota composition and regulated metabolic profiles in white adipose tissue and liver. Further investigations in liver tissue and HepG2 cells revealed that these effects featured suppressed lipogenesis (FASN, ChREBP), enhanced fatty acid oxidation (CPT1A) through activation of AMPK pathway in liver. Maternal t10c12-CLA could suppress lipogenesis and lower lipid levels by activating the AMPK pathway, and alter the composition of gut microbiota, thereby impacting offspring lipid metabolism.

Immunoglobulin light chains (LCs) exhibit diverse aggregation behaviours that depend sensitively on sequence composition and intermolecular interactions. Understanding how specific residues modulate aggregation kinetics remains a key challenge in elucidating the molecular basis of light-chain amyloidosis. Here, we investigate sequence-dependent aggregation using recombinant λ LCs derived from the IGLV2 gene family. Comparison of two closely related LCs differing by only 16 amino acids revealed striking differences in aggregation behaviour under thermal stress. Bioinformatic analysis identified an additional aggregation-prone segment in the CDR1 region of the aggregation-prone M10 variant, associated with residues Ser33 and Tyr34. Rational substitution of these residues (S33D/Y34S) markedly reduced aggregation while leaving the thermal transition temperature largely unchanged (∼53 °C). Differential scanning calorimetry revealed that the wild-type M10 LC unfolds with a significantly lower apparent activation energy (∼290 kJ/mol) compared with the non-aggregating H9 (∼605 kJ/mol) and the stabilised double mutant (∼560 kJ/mol), indicating reduced kinetic stability. Aggregation of unfolded species showed much weaker temperature dependence (Ea ≈ 10-70 kJ/mol) and exhibited strong concentration dependence consistent with a multimolecular association process. Additional experiments suggest that aromatic interactions involving Tyr34 contribute to the stabilisation of intermolecular assemblies. Together, these results establish a quantitative link between local sequence variation in the CDR1 region, kinetic stability of the LC fold, and aggregation propensity, highlighting how targeted mutations can modulate aggregation behaviour in immunoglobulin light chains.

The Janus kinase (JAK)/STAT signaling pathway plays a pivotal role in cancer biology as well as in inflammatory and autoimmune disorders such as psoriasis. Recent advances in biomedical research and targeted therapies have highlighted the importance of computational approaches for accelerating the discovery of selective kinase inhibitors. This study aimed to develop a robust computational framework for predicting the inhibitory potency of JAK2 ligands and for analyzing their binding interactions using structure-based methods. A curated dataset of 1869 chemically valid JAK2 ligands with experimentally reported Ki values was compiled, standardized, and converted to pKi. Using this dataset, a bond-aware graph neural network (GNN) was trained and evaluated for pKi prediction. Top-ranked predicted ligands were further examined via molecular docking, pharmacophore modeling, and molecular dynamics simulations to assess their interactions within the JAK2 ATP-binding site. The proposed model achieved strong predictive performance, yielding an average test-set R2 of 0.91 ± 0.01, MAE of 0.14 ± 0.01, and RMSE of 0.26 ± 0.02 across repeated data splits. Structure-based analyses supported the predicted binding poses and identified key stabilizing interactions within the JAK2 ATP-binding site. Overall, this integrative computational framework provides a reliable approach for predicting JAK2 inhibitory potency and offers mechanistic insights that may support the computational prioritization of candidate molecules for future experimental evaluation.

Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by the pathological accumulation of amyloid-β (Aβ) plaques and neurofibrillary tangles composed of hyperphosphorylated Tau protein. While microRNAs (miRNAs) have emerged as critical regulators of gene expression and potential biomarkers for AD, their role in Tau pathology remains incompletely understood. In this study, we investigated the role of the miR-92 family and its downstream targets in modulating Tau-induced neurodegeneration using the Drosophila model. Overexpression of miR-311, miR-312, and miR-313, members of the miR-92 family, exacerbated Tau-induced phenotypes. We identified deltex (dx) and engrailed (en) as miR-92 target genes with genetic screening with in silico target prediction. Overexpression of en or EN2, a human ortholog of en, reduced Tau protein levels in Drosophila and human neuroblastoma cells, respectively. Furthermore, overexpression of en attenuates Tau-associated memory deficits in Drosophila. These findings suggest that the engrailed gene is an evolutionarily conserved regulator of Tauopathy and highlight the utility of the Drosophila AD model for identifying genetic modulators with therapeutic potential in AD.

Dysregulated lipid metabolism drives atherosclerosis (AS). Yacon, an Andean lipid-modulating tuber, exerts anti-AS potential, but mechanisms remain unclear. We integrated network pharmacology, machine learning, single-cell RNA sequencing (scRNA-seq), and in vivo validation to explore its anti-AS effects and targets. Active constituents and targets were curated from literature, TCMSP, and SwissTargetPrediction; lipid/AS genes from GeneCards, OMIM, and GEO were filtered via limma, WGCNA, LASSO, randomForest, and SVM-RFE. Immune infiltration and external validation confirmed hub gene relevance. scRNA-seq prioritized FABP5; docking and dynamics quantified compound-FABP5 interactions. In vivo efficacy was tested in high-fat diet (HFD)-fed ApoE-/- mice via histology (Oil Red O, H&E, and Masson) and molecular assays (RT-qPCR, Western blot, and immunofluorescence). We identified 12 constituents, 384 targets, and seven core targets (AURKA, MMP9, FABP5, etc.), with FABP5 top-ranked. Docking and dynamics identified Enhydrin as the strongest FABP5 binder. Enhydrin administration was associated with reduced hepatic lipid accumulation, decreased serum triacylglycerol (TG), total cholesterol (TC), and low-density lipoprotein cholesterol (LDL-C) levels, and increased high-density lipoprotein cholesterol (HDLC) levels. Histopathological analysis of arterial tissues revealed attenuated vascular lipid deposition and delayed atherosclerotic lesion progression. Across assays, Enhydrin downregulated FABP5, reduced abnormal fatty acid trafficking, and upregulated PPARγ and ABCA1, with markedly reduced vascular lipid deposition and improved serum lipid profiles, reflecting enhanced cholesterol efflux. Conclusion: Our integrative multi-omics analysis pinpointed FABP5 as a promising novel target for yacon-derived Enhydrin in atherosclerosis. In vivo, Enhydrin markedly downregulated FABP5 and upregulated PPARγ and ABCA1, suggesting this axis mediates its anti-atherosclerotic activity. SIGNIFICANCE STATEMENT: 1: This study identified FABP5 as a candidate target for atherosclerosis through integrative multi-omics, and its association with the anti-atherosclerotic effects of Enhydrin suggests therapeutic potential. 2: The anti-atherosclerotic effects of yacon's active components and their underlying molecular pathways were systematically screened and preliminarily characterized by integrating bioinformatic prediction with in vivo validation, laying a preliminary theoretical foundation for further pharmacological investigation and clinical translation.

Resistance to molecular-targeted agents, represented by sorafenib (SOR), poses a major challenge in the treatment of hepatocellular carcinoma (HCC). Huaier (HUA), a traditional Chinese medicine used as an adjuvant cancer therapy, has demonstrated favorable efficacy in clinical practice for advanced HCC. This study demonstrates that HUA reverses SOR resistance in HCC and exhibits synergistic effects with SOR by modulating NCOA4-mediated ferritinophagy, thereby influencing iron and lipid metabolic pathways associated with ferroptotic vulnerability. In SOR-resistant Huh7R cells and mouse xenograft models, HUA alone or combined with SOR disrupted iron homeostasis through ferritinophagy induction, promoted Fe3+ reduction to Fe2+, enhanced ROS generation, induced lipid peroxidation, and concurrently reduced lipid droplet storage and triglyceride accumulation. Mechanistically, HUA upregulated NCOA4 and increased the LC3B-II/I ratio, leading to FTH1 degradation. Accumulated Fe2+ drove robust ROS production via the Fenton reaction, exacerbated lipid peroxidation, upregulated ACSL4, and downregulated SCD1 and GPX4, collectively inducing ferroptosis, a form of regulated cell death specifically rescued by ferrostatin-1 or deferoxamine but not by Z-VAD-FMK or necrostatin-1, with minimal caspase activation. Furthermore, this cascade markedly downregulated PLIN2, resulting in energy depletion in Huh7R cells and tumor tissues, thereby impairing adaptive survival and ultimately reversing SOR resistance. Notably, NCOA4 silencing substantially attenuated HUA's efficacy in overcoming SOR resistance. These findings highlight the role of HUA extract in promoting NCOA4-dependent ferritinophagy and modulating iron and lipid metabolism to enhance ferroptotic sensitivity, providing a rationale for the clinical development of HUA + SOR combination therapy in HCC.

Cellular organelles are uniquely specialized membrane-bound structures that enable cells to organize and coordinate biochemical processes. Specifically, mitochondria are essential organelles for cellular metabolism, coordinating energy production, and connecting signaling networks for cellular homeostasis. 99% of mitochondrial proteins are encoded by nuclear genes that require precise and efficient translation and import into mitochondria for biological processes. This process is mediated by coordinated pathways involving the mitochondrial specific translocation complexes, chaperones, and specialized targeting routes. Tight regulation of these import mechanisms allows for proper protein localization, folding, and assembly. Disruptions in the mitochondrial protein import pathway compromise organelle homeostasis and activate proteostatic stress and quality control pathways. Such defects have been observed in a wide range of pathophysiological conditions, including cardiovascular disease, neurodegeneration, and cancer. The import defects destabilizing mitochondrial proteins can impair oxidative phosphorylation and metabolic signaling. In sum, defects to mitochondrial function can highlight a central role of mitochondrial protein import beyond maintaining cellular function and how defects at distinct stages of import contribute to disease, underscoring opportunities for therapeutic intervention targeting mitochondrial proteostasis.

Endometrial carcinoma (EC) is a prevalent gynecologic malignancy. Ferroptosis, a type of programmed cell death, holds promise as a therapeutic strategy for cancer treatment. However, the specific role of ferroptosis and the underlying mechanisms in EC cells remain largely unexplored. In this study, we identified ubiquitin-specific protease 1 (USP1) as a key inhibitor of ferroptosis in EC cells. Our findings show that USP1 expression is reduced in EC cells undergoing ferroptosis but is elevated in EC tissues. Overexpressing USP1 significantly enhanced cell viability, colony formation, cell cycle progression, and migration/invasion capabilities, while strongly suppressing ferroptosis induced by the ferroptosis activator erastin. Through bioinformatic analysis and functional validation, we identified estrogen receptor 1 (ESR1) as a USP1-interacting protein involved in ferroptosis regulation. USP1 was shown to interact with ESR1, reduce its polyubiquitination, and enhance its protein stability. Rescue experiments revealed that ESR1 overexpression counteracted the pro-ferroptotic effects of USP1 depletion in erastin-treated EC cells. Finally, in vivo studies using an EC xenograft model confirmed the anti-ferroptotic and tumor-promoting role of the USP1-ESR1 axis. Together, our findings reveal a novel mechanism by which USP1 promotes EC progression through suppression of ferroptosis via ESR1 stabilization. Thus, targeting the USP1-ESR1 signaling axis may offer a new therapeutic strategy for EC.

The existence of multiple molecular forms of enzymes, genetic polymorphism and functional promiscuity raise the question of the identity of active center(s) responsible for several activities. In the present review, we recapitulate the general strategy, implementing the simple and rigorous inhibition kinetic method for probing the existence of single or multiple active sites on enzyme molecules. The model enzyme we chose to illustrate this approach is human butyrylcholinesterase, an enzyme that shows a complex functional (promiscuity), structural (multiple oligomeric forms) and genetic polymorphisms (numerous allelozymes and isoenzymes). This classical active site discrimination method is based on the analysis of enzyme irreversible inhibition profiles of enzymes under first-order conditions by monitoring the progressive enzyme activity decay with two reporter substrates of different specificity. The use of chiral irreversible inhibitors and/or chiral reporter substrates provides additional kinetic information about preferential enantioselectivity or binding complementarity of the target enzyme, allowing selection of the best inhibitors or substrates. Then, additional investigations, using structural methods (X-ray structure analysis, mass spectrometry), in silico simulations and classical biochemical methods (electrophoresis, PCR) provide definitive answers.

Liver metastasis is an intractable clinical challenge due to profound immunosuppression within the liver metastatic niche (LMN), which fundamentally limits immunotherapy efficacy. The unique immune landscape of LMN is shaped by inter-organ, primary tumor-origin, and intra-lesional heterogeneity, driven by multilayered mechanisms including poor immunogenicity, antigen presentation deficiency, antitumor immune cell dysfunction, hyperactive immunosuppressive cells, spatial immune remodeling and host systemic factors. Temporally, LMN evolves from immune surveillance to immune escape. Notably, distinct primary tumors share certain common LMN features, enabling a potential paradigm shift from primary tumor-based classification to LMN-guided subtyping. Precision strategies integrating multi-omics, dynamic biomarkers, and combinatorial interventions hold promise to overcome LMN-driven resistance. This review delineates the spatiotemporal immune remodeling of the LMN, summarizes current therapeutic strategies, and highlights future translational priorities for precision immunotherapy.

Endothelial cells (EC) form the inner lining of the blood vessels and are essential for the vascular homeostasis. EC death has been implicated in the pathology of vascular diseases. Our previous studies indicated that docosahexaenoic acid (DHA) induces EA.hy926 EC death by upregulating lipid droplet (LD) biogenesis and activating p38 MAPK, however, how LDs contribute to this process remains unclear. DHA belongs to the polyunsaturated fatty acids that are susceptible to lipid peroxidation, which is a common mechanism in lipid-induced cell injury and ferroptosis. Therefore, the current study is to investigate the mechanism of LD in the regulation of lipid peroxidation in relation to p38 MAPK in DHA-induced EA.hy926 endothelial ferroptosis. Results showed that DHA induced a time-dependent, bidirectional modulation of GPX4 expression accompanied by a progressive GSH and NADPH depletion and increased lipid peroxidation in ECs. Suppression of LD formation through diacylglycerol acyltransferases inhibition, siRNA, or promotion of LD lipolysis reduced lipid peroxidation and restored the GPX4 expression. Inhibiting lipid peroxidation chemically, activating GPX4 or supplementing GSH all prevented the cell death by DHA. Our findings reveal a novel role of LD in regulating GPX4-mediated lipid peroxidation and highlight the key role of p38 MAPK in downregulating GPX4 in lipid peroxidation-induced EC ferroptosis by DHA. In conclusion, the present study suggested that DHA-induced LD formation in ECs increases the membrane area, which could facilitate the lipid peroxidation reaction, disrupt the endogenous GPX4 antioxidant system, and thereby exacerbate the detrimental effects of lipid peroxidation, ultimately leading to cell ferroptosis.