Back

PAK4 promotes vertex remodeling to maintain epithelial integrity and barrier function

Adhikary, B.; Chang, A.; Higashi, A. Y.; Chiba, H.; Higashi, T.

2025-10-04 cell biology
10.1101/2025.10.02.680170 bioRxiv
Show abstract

Cell-cell junctions are essential for epithelial integrity and barrier function, but the mechanisms regulating their remodeling remain unclear. Here, we investigated the role of the junctional kinase PAK4 in vertex remodeling. PAK4 localized to apical junctions and accumulated at multicellular vertices in MCDK cells and Xenopus embryos. Inhibition or knockout of PAK4 increased higher-order vertices, caused junctional discontinuities, and impaired barrier function in MDCK cells. PAK4 recruitment required the scaffolding protein Afadin. Afdn-KO cells exhibited severe junctional defects and reduced barrier function. Artificial targeting of PAK4 in Afdn-KO cells partially restored junctional continuity and barrier function. In Xenopus embryos, live imaging showed dynamic PAK4 accumulation at remodeling vertices, and PAK4 inhibition hindered resolution of multicellular vertices. Expression of an amino-terminal fragment (PAK4-NT) impaired remodeling and induced cytokinetic failure. Live imaging in Xenopus revealed barrier leakage at multicellular vertices upon PAK4 inhibition. These findings indicate that PAK4 and Afadin cooperate to maintain epithelial integrity and barrier function by promoting vertex remodeling. SummaryAdhikary et al. show that PAK4, recruited to epithelial junctions by Afadin, drives dynamic vertex remodeling in cells and embryos. Inhibiting PAK4 disrupts tissue packing and compromises junctional continuity and barrier function, implicating PAK4 and Afadin as critical partners in vertex remodeling to preserve tissue integrity.

Published in Journal of Cell Biology (predicted rank #1) · training set

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.