Vortex Defects in Cell Sheets Build Orderly 3D Tissues Layer by Layer

A flat sheet of cells seems like the least likely place to watch a three-dimensional organ take shape, yet that is precisely what a team of physicists and biologists in Paris has achieved. In a study published in Nature Physics, researchers led by Hsiang-Ying Chen and Pascal Silberzan of the Institut Curie show that a simple swirl in the orientation of a cell monolayer can act as a blueprint for an entire multicellular structure. When spindle-shaped myoblasts were grown on microscopic surfaces patterned with concentric circular tracks, the cells aligned themselves along the rings, creating a vortex-like topological defect at the centre. Rather than remaining a purely two-dimensional curiosity, this vortex became the engine of a remarkable construction project: cells converged on the defect core, piled on top of one another, and assembled a bell-shaped mound of striking internal order.

Topological defects are familiar characters in the physics of liquid crystals and active matter. They are points or lines where the local orientation field of a material cannot be defined smoothly, forcing the system into characteristic patterns such as vortices, spirals and asters. Over the past decade, dense sheets of cells have been shown to behave like active nematic liquid crystals, with elongated cells playing the role of rod-like molecules that constantly burn chemical energy to move and exert forces. Earlier work had revealed that defects in such monolayers can act as mechanical organizing centres: they concentrate stresses, attract cell death and extrusion in epithelia, and even mark the sites where the freshwater polyp Hydra sprouts new tentacles. What remained unclear was how a defect could coordinate cell motion on large scales and drive the assembly of genuinely three-dimensional tissue.

The French team solved this problem by taking control of the defect itself. Instead of waiting for vortices to appear spontaneously in a disordered monolayer, they fabricated polydimethylsiloxane substrates engraved with concentric micro-rails. Myoblasts migrating on these tracks follow contact guidance, the tendency of elongated cells to align with underlying grooves, and so the pattern imposes a well-defined vortex of cell orientation with an integer topological charge of +1 at its centre. This experimental trick transformed a fleeting, hard-to-predict feature of active matter into a stationary, reproducible target that could be imaged for days with phase-contrast and confocal microscopy, and quantified with particle image velocimetry and custom orientation analysis.

The first surprise came before any three-dimensional structure appeared. For roughly the first twelve hours after the monolayer reached confluence, the cells did not simply circle the centre along the rails. Their velocity field underwent a dramatic reorganization: the azimuthal, ring-following motion gave way to a radial flow converging on the defect core, with cells migrating perpendicular to their long axis, along their short axis, toward the centre of the pattern. The researchers tracked this transition with an order parameter that distinguishes azimuthal from radial flow, and found that the convergent regime emerged across millimetre scales, a huge distance compared with the size of a single cell. A swirl in orientation, in other words, was broadcasting a coordinated traffic instruction to thousands of cells simultaneously.

That convergent traffic had immediate architectural consequences. Cells arriving at the core began to extrude out of the plane of the monolayer, climbing over their neighbours and forming a second layer roughly six micrometres thick. Crucially, this new layer was not a random heap. Confocal sections revealed that the second layer inherited the vortex orientation of the layer beneath it, and when a third layer appeared a day later, it too was vortex-aligned. The team found that the cells deposit their own extracellular matrix, including collagen IV, in patterns that mirror the orientation of the cells that laid it down. This secreted scaffold then guides the orientation of subsequent arrivals through contact guidance, creating a feedback loop in which each layer templates the one above it. The mound, in effect, was printing itself layer by layer, with the topological defect supplying the pattern.

As the structure matured over tens of hours, its internal architecture became even more sophisticated. Confocal reconstructions of fixed mounds showed that the nematic order varied systematically with altitude. Near the base, close to the core, cells adopted an aster-like arrangement pointing radially outward, while farther from the centre they retained the vortex configuration. At the apex of the mound, the orientation twisted into a spiral. Remarkably, despite these changes in appearance, the in-plane topological charge remained invariant and equal to +1 along the entire vertical axis of the mound, from the first layer to the summit. When the researchers compared mounds from different experiments and different ages, the profiles of their orientation order parameter collapsed onto a single master curve once rescaled by the maximum height, indicating that the mound possesses a self-similar, reproducible internal blueprint.

The growth mechanism itself proved equally elegant. By tracking fluorescently labelled cells shuttling between layers, the team documented a process they describe as interlayer permeation: cells do not simply stack passively at the core but percolate through the existing structure, moving between adjacent layers like fluid seeping through a porous medium. The flux between layers peaked at a specific radius, the boundary between the inner aster-like region and the outer vortex region, where the orientation order parameter crosses zero. Even more striking was a reversal of flow direction as the mound aged. Around thirty hours after confluence, cells near the axis of the mound, which had initially migrated inward, began to migrate outward, and this switch from convergent to divergent flow was mirrored by a local change in orientation from vortex to aster. The growing structure was not static; it continuously reorganized its own traffic patterns as it expanded.

To make sense of these observations, Carles Blanch-Mercader and Jacques Prost developed a theoretical model treating the mound as a stack of active nematic fluid layers. The framework incorporates the active anisotropic stresses generated by the cells, the stresses in the external single layer, and an interfacial stress between layers, and it reproduces the measured radial velocity profiles and permeation fluxes with remarkable fidelity. The radial velocity in the monolayer follows a one-over-radius power law characteristic of a converging flow, while the flux from the mound to the external medium decays exponentially. In the late stages, the bell-shaped profile of the mound can be modelled entirely by dynamics driven by the converging flows, without invoking cell proliferation as the dominant driver. Drug perturbations confirmed this hierarchy: when proliferation was blocked with mitomycin C, the mounds retained their wild-type organization, but when contractility was inhibited with blebbistatin or the ROCK inhibitor Y-27632, the ordered alignment was almost completely destroyed. The architecture, it turns out, is a mechanical phenomenon built on cellular forces, not a by-product of cell division.

The implications reach well beyond the culture dish. Cellular whorls and swirls have long been noted by pathologists as diagnostic features in brain tumours, ovarian cancer surfaces and fine-needle aspirates of papillary thyroid carcinoma, and defects in the mesothelium have been shown to suppress the clearance of ovarian cancer cells. The new results suggest a physical mechanism by which such swirling patterns could actively organize cells into multilayered, tumour-like masses, with convergent defect-driven flows feeding cells into the core and permeation spreading them through the structure. The same physics may operate in normal morphogenesis, where embryonic tissues fold, bud and layer themselves with exquisite precision, and recent work on lumen nucleation in embryos has already implicated topological defect transitions controlled by boundary geometry. By demonstrating that an integer topological defect alone, imposed by nothing more than a patterned surface, can orchestrate millimetre-scale flows and the assembly of a highly ordered three-dimensional tissue from a featureless monolayer, the study provides both a conceptual framework and an experimental platform. It suggests that the grammar of tissue construction may be written partly in the language of topology, and that learning to read and perhaps rewrite that grammar could one day inform strategies in regenerative medicine, tissue engineering and cancer therapy.

Subject of Research: Topological-defect-driven morphogenesis of three-dimensional cell mounds in active nematic monolayers

Article Title: Morphogenesis of a stratified cell mound at a vortex defect

Article References: Morphogenesis of a stratified cell mound at a vortex defect. (n.d.). https://doi.org/10.1038/s41567-026-03427-9

Image Credits: AI Generated

DOI: 10.1038/s41567-026-03427-9

Keywords: topological defects, active nematics, cell monolayers, morphogenesis, collective cell migration, myoblasts, interlayer permeation, biological physics, tissue engineering, tumorigenesis, contact guidance, Institut Curie