Back

Higher-Order Mechanical Neighbourhood Sensing Governs Cell Fate in Complex Tissues

John, A.; Ma, T.; Mackay, D.; Connolly, E. C.; Colombani, J.; Doostmohammadi, A.; Andersen, D. S.

2026-08-18 cell biology
10.64898/2026.08.13.744616 bioRxiv
Show abstract

Adult tissues maintain precise architectures composed of multiple interspersed cell types, yet prevailing models of cell fate patterning rely on pairwise interactions between neighbouring cells that generate binary decisions. How local interactions encode higher-order, beyond pairwise, multicellular organization remains unclear. Here, using the Drosophila midgut as a model of adult tissue self-organization, we identify a higher-order neighbourhood-sensing mechanism that couples cell identity, mechanics and stem cell (SC) fate. We show that the receptors Cirl/Latrophilin and Toll-8 are expressed in complementary cellular compartments, restricting their engagement to heterotypic interfaces between progenitors and differentiated enterocytes (ECs). These interactions generate interface-specific myosin II-dependent tension that depends on EC neighbourhood composition and mechanically gates Delta-Notch signalling. Thus, although Delta-Notch acts through pairwise interactions, these fate decisions are modulated by a higher-order mechanical state integrating information from surrounding cell contacts. A mathematical model incorporating this coupling reproduces multicellular tissue architecture from first principles and predicts that loss of higher-order coupling destabilizes progenitor organization. This prediction is consistent with the excessive fate transitions, topological disorganization and regenerative defects observed after Cirl-Toll-8 disruption. These findings establish higher-order mechanical neighbourhood sensing as a general principle by which tissues integrate local cellular identities to maintain and restore complex architectures.

Matching journals

The top 5 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.