Premature ovarian failure (POF) is characterized by oxidative stress, cellular senescence, and impaired steroidogenesis, yet current therapies remain limited in effectiveness. Thymol, a natural monoterpene, exhibits antioxidant and anti-inflammatory properties. Network pharmacology and molecular docking identified multiple potential targets, notably the Janus kinase 1 (JAK1)-signal transducer and activator of transcription 3 (STAT3) pathway. In tert-butyl hydroperoxide (t-BHP)-induced human granulosa-like tumor cells (n = 3), 40 μg/mL thymol increased cell viability by approximately 45%, restored superoxide dismutase, catalase, and glutathione peroxidase activities to nearly twice those of the model group, and reduced reactive oxygen species accumulation by about 35% (p < 0.05). It also decreased senescence markers p53, p21, and p16 by 40–60% and inhibited JAK1–STAT3 phosphorylation (n = 3, p < 0.05). In aged pregnant mice (n = 4 per group), thymol increased viable fetus numbers by about 40%, elevated serum estradiol and progesterone levels to 1.6–1.8-fold of aged controls, and downregulated ovarian aging markers (p < 0.05). Collectively, these findings indicate that thymol mitigates oxidative stress-induced ovarian aging by modulating JAK1–STAT3 signaling and restoring steroidogenic function, supporting its potential as a natural candidate for delaying ovarian senescence.

Cytokine storm is a central pathogenic mechanism underlying sepsis-induced acute lung injury (SALI) and severe coronavirus disease 2019 (COVID-19), yet effective therapeutic strategies remain limited. Eupatorium lindleyanum DC. (EL), a traditional Chinese medicinal herb, has been reported to possess anti-inflammatory, antioxidant, and antiviral-related activities; however, its protective mechanisms in SALI and virus-associated inflammatory lung injury remain incompletely understood. In this study, an integrated strategy combining computational prediction and experimental validation was employed to investigate the therapeutic potential and underlying mechanisms of EL. The chemical constituents of EL were characterized by UPLC–Q–TOF/MS, followed by network pharmacology, molecular docking, and molecular dynamics analyses to predict key targets and signaling pathways. A cecal ligation and puncture (CLP)-induced SALI rat model was used to evaluate lung histopathology, pulmonary edema, cytokine production, and inflammatory signaling activation. In parallel, LPS-stimulated RAW264.7 macrophages were used to assess cytokine secretion and pathway regulation in vitro. In addition, a SARS-CoV-2 pseudovirus-induced mouse model was employed to further evaluate the in vivo relevance of the representative bioactive compound hyperoside in pseudovirus-associated lung injury. A total of 32 active compounds and 697 putative targets were identified, among which 116 were associated with sepsis and COVID-19. In vivo, EL markedly alleviated lung injury, reduced the lung coefficient and wet/dry ratio, and suppressed excessive production of proinflammatory cytokines and activation of key signaling proteins. In vitro, EL dose-dependently inhibited TNF-α and IL-6 secretion and regulated the PI3K–Akt and NF-κB signaling pathways. Notably, hyperoside showed favorable predicted interactions with PI3K–Akt pathway-related targets (EGFR, PI3K, and Akt), while molecular dynamics simulations supported stable interactions with several COVID-19-related targets, including ACE2, Mpro, and RdRp. Furthermore, hyperoside significantly alleviated SARS-CoV-2 pseudovirus-associated lung injury, reduced ACE2 protein expression, and downregulated EGFR, PI3K, and Akt mRNA levels in vivo. Collectively, these findings indicate that EL exerts protective effects through multi-component, multi-target, and multi-pathway mechanisms, and support its potential value for further investigation in SALI and virus-associated inflammatory lung injury.

Advanced maternal age (≥35 years) is increasingly common and associated with a higher risk of threatened abortion, yet its molecular basis remains unclear. To elucidate the underlying mechanisms of advanced maternal age-related threatened abortion (AMA-TA) by integrating metabolomics and transcriptomics. Untargeted serum metabolomics was performed in nine AMA-TA patients and seven age-matched healthy pregnant women. An AMA-TA mouse model was established using mifepristone (4 mg/kg) to assess embryo resorption, placental morphology, and serum hormone levels. Serum metabolomics and placental transcriptomic profiling (RNA-seq) were then conducted in AMA-TA mice to characterize metabolic and gene expression alterations. Cross-species and multi-omics integration was performed using HomoloGene and MetaboAnalyst 5.0. Key genes were validated by real-time quantitative polymerase chain reaction. Human serum metabolomics revealed the differential metabolites were mainly enriched in steroid hormone biosynthesis, lipid metabolism, and amino-acid metabolism. The AMA-TA model showed higher embryo resorption, abnormal placental architecture, and reduced progesterone and chorionic gonadotropin. RNA-seq revealed 111 up- and 1337 downregulated genes enriched in 68 pathways. Consistently, serum metabolomics in AMA-TA mice also showed significant metabolic disturbances, prominently involving steroid hormone biosynthesis. Integrated analysis converged on steroid hormone biosynthesis as a shared key dysregulated pathway. Real-time quantitative polymerase chain reaction further confirmed aberrant expression of steroid metabolism–related genes, including upregulation of Akr1d1 and Ugt family genes. Disruption of steroid hormone biosynthesis represents central molecular feature of AMA-TA. Integrated multi-omics analysis offers mechanistic insight and supports the development of biomarkers and therapeutic targets for AMA-TA.