Back

High-throughput Virtual Screen of Endocrine-disrupting Chemicals Identifies Disruptors of EGFR Signaling

Jesikeiwicz, L.; Marathe, R.; Sepehri, B.; Demissie, R.; Lee, H.; Veiga-Lopez, A.; Villegas, J. A.

2026-08-11 pharmacology and toxicology
10.64898/2026.08.05.743028 bioRxiv
Show abstract

Chemical exposures during pregnancy are linked to an increased risk of pregnancy complications that contribute significantly to maternal and infant morbidity and mortality and can lead to long term health consequences for both the mother and the offspring. The placenta, a central regulator of pregnancy health, is a direct target of environmental toxicants. Epidermal growth factor receptor (EGFR), highly expressed in the placenta, regulates proliferation, migration, invasion, fusion, and cellular bioenergetics. To identify compounds of environmental concern with potential for EGFR-disrupting activity, we optimized a high-throughput virtual screening protocol for the identification of EGFR inhibitors and achieved enrichment factors of EF1% = 10.09, EF5% = 3.86, and EF10% = 3.0 in a benchmarking dataset. We applied this protocol to screen the Collaborative Estrogen Receptor Activity Prediction Project database, finding that top-scoring compounds were enriched for aromatic and fused-ring chemical classes, including dyes. Kinase activity assays revealed that two out of thirteen selected compounds, Vat Red 32 and Reactive Red 136, inhibited EGFR kinase activity with micromolar IC50 values. Additionally, pose refinement with molecular dynamics simulations characterized the binding interactions of Reactive Red 136 within the EGFR kinase domain, and functional assays in HTR-8/SVneo placental trophoblast cells showed that Reactive Red 136, but not Vat Red 32, partially attenuated EGF-mediated cell migration despite both compounds inhibiting EGFR kinase activity. Together, this study has generated an enriched dataset of candidate environmental EGFR modulators, with experimental validation confirming enrichment for EGFR-disrupting activity among the selected compounds. These results provide a valuable resource for toxicological studies.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.