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

Nicolas Pasquier^1,2,#, Meri Pelkonen^1,#, Elise Carraz-Billat^2,§, Aleksi Isomorsu^1,§,**, Raphaël Merand ^2,§, Hellyeh Hamidi¹, Jacques R.R. Mathieu², Jouni Härkönen¹, Gautier Follain¹, Christophe Desterke³, Zoé Fusilier^4,5, Junel Solis⁶, Irina Belaya⁶, Pasi Kankaanpää⁶, Valeria Barresi⁷, Mohammed-Amine Bani^8,9, Johanna Protin², Jérôme Cartry², Sabrina Bedja², Klaus Elenius^1,10-12, Florent Peglion², Fanny Jaulin^{*, ꟸ,2} & Johanna Ivaska^*,ꟸ,1,8-11

See affiliations

Abstract

Mucinous colorectal carcinoma (MUC CRC) metastasis to multiple organs, and to the peritoneum, is associated with poor prognosis. Disseminating MUC CRCs exhibit either conventional (apical-in) or inverted (apical-out) polarity that influence patient outcomes. Therefore, it is critical to identify how MUC CRC polarity is regulated. Here, we analyze patient-derived MUC CRC xenografts with either apical-in or apical-out polarity. Single-cell analyses reveal α2β1-integrin as a key collagen-binding receptor in these models. Collagen–α2β1-integrin interaction activates Src and upregulates SorLA, an endosomal sorting receptor. SorLA supports apical-in polarity by promoting integrin recycling and HER2/HER3 expression. We observe positive correlation between HER2, HER3 and SorLA in patient samples and higher HER2 expression in apical-in-presenting tissues. Clinically relevant HER2/HER3-targeting antibodies revert tumor sphere polarity, and impede collagen remodeling and adhesion to mouse peritoneum. This SorLA—integrin—HER2/HER3 axis could represent a MUC CRC-patient stratification approach and be relevant for other carcinomas with apical-out phenotypes.

Image data

[Update with links to Fairdata.fi]

Image processing workflow

This repository contains the image processing code used in the manuscript. Download the required image data and metadata (see Image data section).

• Clone this repository
• Configure a Python environment using environment.yml
• Run ./util/convert_to_zarr.py to generate a Zarr dataset of all the input images.
• Run ./main.py

Organizational affiliations

¹Turku Bioscience Centre, University of Turku and Åbo Akademi University, Turku, Finland
²Université Paris-Saclay, Gustave Roussy, Inserm, UMR 1279 Tumor Cell Dynamics, F-94805, Villejuif, France
³INSERM UMR-S1310, Université Paris-Saclay, Hôpital Paul Brousse, Villejuif F-94805, France
⁴INSERM-U932, Immunity and Cancer, Institut Curie, Paris-Cité University, Paris, France
⁵INSERM-U932, Immunity and Cancer, Institut Curie, PSL University, Paris, France
⁶Turku BioImaging, Åbo Akademi University and University of Turku, Turku, Finland
⁷Department of Diagnostics and Public Health, University of Verona, Verona 37134, Italy
⁸Medical Biology and Pathology Department, Pathology Laboratory, Gustave Roussy Cancer Campus, Villejuif, France
⁹Platform for Experimental and Translational Pathology (PETRA), Gustave Roussy Cancer Campus, Villejuif, France
¹⁰Institute of Biomedicine, and MediCity Research Laboratory, University of Turku, Turku, Finland
¹¹Department of Oncology, Turku University Hospital, Turku, Finland
¹²InFLAMES Research Flagship Center, University of Turku, Turku, Finland
¹³Department of Life Technologies, University of Turku, Turku, Finland
¹⁴Foundation for the Finnish Cancer Institute, Helsinki, Finland
¹⁵Western Finnish Cancer Center, University of Turku, Turku FI-20520, Finland

^#Equal contribution
^§Equal contribution
^ꟸ Equal contribution
^*Correspondence: johanna.ivaska@utu.fi and fanny.jaulin@gustaveroussy.fr
^**Current address: Cell Polarity, Migration and Cancer Unit, Institut Pasteur, CNRS UMR3691, Université Paris Cité, Équipe Labellisée Ligue Contre le Cancer, F-75015, Paris, France