Back

Signaling downstream of tumor-stroma interaction regulates mucinous colorectal adenocarcinoma apicobasal polarity

Pasquier, N.; Isomursu, A.; Hamidi, H.; Mathieu, J. R. R.; Härkönen, J.; Follain, G.; Desterke, C.; Fusilier, Z.; Solis, J.; Belaya, I.; Kankaanpää, P.; Barresi, V.; Cartry, J.; Bedja, S.; Jaulin, F.; Ivaska, J.

2024-11-21 cancer biology
10.1101/2024.11.20.624298 bioRxiv
Show abstract

Mucinous colorectal carcinoma (MUC CRC) dissemination into the tumor stroma and metastasis to multiple organs, including the peritoneum, is associated with poor prognosis. Disseminating MUC CRCs exhibit either a conventional apical-in or an inverted apical-out polarity phenotype that influence patient outcome. Identifying the mechanisms controlling MUC CRC polarity is critical to understand disease progression. Here, we analyze patient-derived MUC CRC xenografts, with apical-in or apical-out polarity, ex vivo or within collagen gels to mimic the peritumoral stroma. Single-cell analyses reveal 2{beta}1-integrin as a key collagen-binding receptor in these models. Collagen-2{beta}1-integrin interaction activates Src and ERK/MAPK signaling and upregulates the expression of SorLA, an endosomal sorting receptor. SorLA supports apical-in polarity and carcinoma-stroma interactions by promoting integrin recycling to the plasma membrane and HER2/HER3 expression through a positive feedback mechanism. Accordingly, we observe positive correlation between HER2, HER3 and SorLA in patient samples with the highest HER2 expression in apical-in-presenting tissues. Treatment of tumor spheres with clinically relevant HER2/HER3-targeting antibodies reverts sphere polarity and impedes collagen remodeling and adhesion to mouse peritoneum. This SorLA--integrin--HER2/HER3 signaling axis may represent a basis for MUC CRC-patient stratification and shed light on other carcinomas with similar apical-out phenotypes.

Published in Nature Communications (predicted rank #1) · training set

Matching journals

The top 6 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.