A Biomarker‐Driven Ovary–Endometrium Organ‐on‐a‐Chip Mimicking 3D Multicellular Complexity and Menstrual Cyclicity for Predicting Reproductive Toxicity

Reliable prediction of reproductive toxicity remains a critical challenge in drug development and environmental safety. Here, a biomarker-integrated, fluorescent reporter-based reproductive organ-on-a-chip platform that recapitulates the multicellular composition, 3D architecture, endocrine signaling, and cyclic dynamics of the human menstrual cycle, is presented. The system is constructed using primary human theca, granulosa, endometrial stromal and stem cells, vascular endothelial cells, uterine macrophages, and myometrial smooth muscle cells, compartmentalized within collagen-hyaluronic acid hydrogels. Early-response toxicity biomarkers-ANGPTL4 (ovary) and SERPINB2 (endometrium)-are genetically linked to mCherry or GFP fluorescent reporters, enabling real-time, cell-type-specific visualization of toxicant-induced stress. Transcriptomic profiling, KEGG pathway enrichment, and gene knockdown studies confirm ANGPTL4 and SERPINB2 as functional mediators of toxic injury, not just passive indicators. Upon exposure to dioxin and other reproductive toxicants, the platform shows strong, region-specific fluorescent responses that preceded changes detected by conventional cytotoxicity assays. This system demonstrates high sensitivity, temporal precision, and mechanistic insight, offering a scalable and physiologically relevant tool for high-content reproductive toxicology screening. Furthermore, it supports endocrine crosstalk between the ovary and uterus, and dynamic responses across the menstrual cycle, enabling future applications in personalized toxicity prediction and preclinical safety evaluation.