Back

Engineered trophoblast organoids recapitulate molecular and functional features of preeclampsia

Arthurs, A. L.; Lushington, C.; Medina Garcia, D. L.; Merriman, A. L.; Mora-Roldan, G. A.; Parry, L.; Boparai, A.; Polo, J. M.; Adikusuma, F. L.; Thomas, P. Q.; Roberts, C. T.

2026-07-22 cell biology
10.64898/2026.07.22.739976 bioRxiv
Show abstract

Preeclampsia is a major pregnancy complication driven by placental dysfunction, yet research is limited by reliance on patient-derived tissues and models that do not fully capture human disease. Here, we develop a genetically engineered human trophoblast organoid model of preeclampsia that can be generated without access to placental tissue. Using a CRISPR-based Prime Integrase strategy, we engineered induced trophoblast stem cells to express the preeclampsia-associated soluble fms-like tyrosine kinase-1 (sFlt-1) exon 15a isoform. Engineered organoids showed a transcriptional shift towards primary preeclamptic placentae and developed several features of disease. These included reduced PlGF, increased IL-6 and soluble endoglin, oxidative stress, impaired growth and an elevated sFlt-1/PlGF ratio comparable to primary preeclamptic trophoblast organoids. These broader changes were not reproduced by adding recombinant human sFlt-1 to control organoids. Conditioned media from engineered organoids impaired endothelial network formation, demonstrating a functional effect of the altered trophoblast secretome. Treatment with sulfasalazine and metformin also restored angiogenic balance and organoid growth. Together, these findings establish a tractable human model that reproduces molecular and functional features of preeclampsia and provides a platform to study disease mechanisms and test potential therapies.

Matching journals

The top 12 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.