Why Cells Divide Unequally: It's All About Crowding ✛ ✎

Original: "Cell size and confinement drive asymmetric cell division through a cortical instability"
· Da Gao, Guoye Guan, Chao Tang, Rui Ma
arXiv:2609.01393v2 · 2026-09-01 · CC BY-NC-ND 4.0 · independent analysis · 1 min
When a cell is cramped or small, its division can become asymmetric on its own, without chemical labels.

A tightly stretched membrane, confined in a narrow case, decides on its own where to burst. So does a cell: when it's crowded, the rupture point shifts from the center, and the two halves end up unequal. Previously, such asymmetry was thought to be set in advance by chemical labels, but new work shows that mechanics alone suffices.

The cell cortex, a thin contractile layer beneath the membrane, was modeled as a closed viscous surface inside an ellipsoidal shell. The equations included active stresses that depend linearly on curvature, because the cortex breaks the 'up-down' symmetry. The dynamics of shape and flow were solved numerically, varying cell size and the degree of confinement. For validation, published three-dimensional data from C. elegans embryogenesis were used: eight uncompressed and seventeen compressed embryos.

Three modes of division were found: symmetric, asymmetric, and division arrest. The transitions between them are governed by a pitchfork bifurcation: as cell size decreases or confinement increases, the instability threshold rises, and above the threshold, asymmetry grows.

In experiments, division asymmetry increased statistically significantly with decreasing mother cell volume (p < 0.001 for late germline divisions). Mechanical compression of the embryo increased asymmetry by an average of 15% (regression slope 1.151, R² = 0.917). This is the first quantitative test of how size and external confinement jointly trigger cortical instability.

In C. elegans, the germline divides asymmetrically from the start, while somatic cells initially divide almost equally but become asymmetric as they shrink. This was reproduced with a single model without fitting to each lineage.

t^a_{ij} = \xi^s \Delta\mu \left( C_{ij} - \frac{1}{2} C^k_k g_{ij} \right) + \xi^b \Delta\mu C^k_k g_{ij}
Active cortical stresses are proportional to the local surface curvature
\kappa(u) = \kappa_0 \left( 1 + \Delta\kappa \exp\left( -\frac{(u-u_0)^2}{\omega^2} \right) \right)
Cortical stiffness is enhanced near a given position u0
Concepts
cell division asymmetry stem cell embryonic development polarity charge separation spontaneous symmetry breaking alzheimers disease tumor brain tumor mutation cancer instability polar molecule outbreak matter antimatter asymmetry epilepsy mirror symmetry cell Mendel's Law of Segregation