When the social amoeba Dictyostelium discoideum runs out of food, thousands of formerly independent cells begin a choreographed group performance. They secrete the chemical signal cyclic AMP, relay it to their neighbors, and migrate toward the source, forming rotating clusters that spin like living pinwheels. A new study combining live-cell microscopy with computer simulations reports that the size of these clusters sets the tempo of the entire performance: larger aggregates rotate more slowly, carry slower cAMP waves, and are more likely to host spiral waves with two arms instead of one.
The research, published in BMC Molecular and Cell Biology, examined pre-mound stage aggregates — clusters that form after starvation but before cells pile up into mounds and slugs.
The cells carried Flamindo2, a fluorescent reporter whose glow dims when intracellular cAMP rises. This inverse signal let the researchers watch chemical waves sweep across each cluster while differential interference contrast imaging captured the cells themselves moving beneath the waves. Strikingly, the waves and the cells rotate in opposite directions: the spiral cAMP wave spins one way, while the chemotaxing cells, chasing the gradient at the wave front, circle the other way. The authors attribute this counter-rotation to the refractory state of cells that have just relayed the signal, combined with a persistence memory that keeps cells moving along their established heading.
To quantify the wave rotation, the team divided each cluster’s fluorescence image into 72 wedges of five degrees each, computed intensity time series for every wedge, and used autocorrelation analysis to extract the local wave period. Cell rotation was measured with optical flow analysis of the DIC images, taking the angular velocity at the radius of maximum tangential speed. Across 17 single-armed clusters from four experimental days, the wave angular velocity fell from about 1.8 radians per minute in small clusters of roughly 4,000 square micrometers to about 0.8 radians per minute in clusters near 70,000 square micrometers — a range the authors estimate corresponds to roughly 50 to 1,000 cells in these flattened aggregates.
Cell rotation followed the same downward trend, though at much lower absolute speeds, ranging from about 0.1 to 0.5 radians per minute. When the two velocities were plotted against each other, they fell on a straight line with a coefficient of determination of 0.824, meaning that slower waves accompanied proportionally slower cell rotation across nearly an order of magnitude in cluster size. Within any single cluster, rotation was also not rigid: angular velocity decreased with distance from the center, indicating that cells continuously slide past one another rather than turning as a solid disk.
Spiral arm number added a second size-dependent signature. Counting arms as stable only if they persisted for at least two full wave rotations, the researchers found that double-armed spirals were significantly more likely in larger clusters, with the size distributions of single- and double-armed clusters differing significantly at p = 0.0008. The interpretation is geometric and dynamical: in small clusters, most cells sit close to a passing wave arm and remain refractory, so secondary arms are extinguished, while in large clusters a reservoir of recovered, excitable cells and a longer perimeter can sustain and separate multiple arms.
To probe the mechanism, the team built a Cellular Potts Model in which deformable, polarized, chemotactic cells sit on a lattice and secrete cAMP treated with a reaction-diffusion scheme including diffusion and degradation by the enzyme phosphodiesterase. With all parameters set independent of cluster size, the model reproduced spiral waves, counter-rotating cells, and multi-armed spirals in large aggregates — but it failed on one crucial point: the simulated wave speed stayed roughly constant regardless of cluster size, contradicting the experiments.
The fix came from the extracellular environment. Phosphodiesterase, the enzyme that degrades cAMP, is secreted by cells and diffuses through the medium. If each cell produces it at a roughly constant rate, larger aggregates should accumulate higher extracellular enzyme concentrations, lowering external cAMP, weakening the stimulus cells receive, and slowing wave propagation. When the researchers made the degradation rate proportional to cell number in the model, the simulated wave velocity declined with cluster size from about 1.25 to 1 radian per minute, matching the experimental trend, and the linear correlation between wave and cell angular velocities reappeared, with a coefficient of determination of 0.92. Notably, the decline in cell angular velocity itself did not require this modification — it emerges from simple kinematics, since cell linear speeds are bounded and angular speed must fall as radius grows.
The authors are careful to frame the size-dependent degradation as a plausible hypothesis rather than a proven mechanism. Their experiments did not directly measure extracellular phosphodiesterase concentrations, and they list alternative contributors, including wave curvature effects, cell packing, receptor adaptation, and substrate interactions, though they argue hypoxia is unlikely given that their aggregates remained stable under confinement. Other caveats include the quasi-two-dimensional preparation, optical flow velocities that should be read as relative rather than absolute, exclusion of cluster merging events, and simulated cluster sizes limited to a narrower range than the experiments, which the authors say explains why double-armed spirals appeared only about half as often in their largest simulations whereas the largest experimental clusters always showed two arms.
Even with those limits, the study delivers a clean quantitative message: aggregation in Dictyostelium is not scale-neutral. As more cells join a rotating cluster, the collective signaling environment shifts, stretching the timescale of both the chemical wave and the cellular carousel above it. Because phosphodiesterase activity is experimentally perturbable, the authors propose that altering its production or activity and measuring the resulting wave and rotation changes offers a concrete, testable route toward confirming how cell number is coupled to the emergent patterns of this classic model of multicellular self-organization.
Subject of Research: Size-dependent cAMP signaling and collective rotation in Dictyostelium discoideum aggregates
Article Title: From microscope to model: rotating signaling dynamics and cluster size in Dictyostelium discoideum aggregates
Article References: Saitia, S., Lai, Y.-C., Tang, M.-H., Jain, J., Reiss, M., & Rappel, W.-J. (2026). From microscope to model: rotating signaling dynamics and cluster size in Dictyostelium discoideum aggregates. BMC Molecular and Cell Biology, 27(1), Article 36. https://doi.org/10.1186/s12860-026-00593-9
Image Credits: AI Generated
DOI: 10.1186/s12860-026-00593-9
Keywords: Dictyostelium discoideum, cAMP signaling, spiral waves, collective cell rotation, cluster size, phosphodiesterase, Cellular Potts Model, chemotaxis, aggregation, Flamindo2, quasi-2D confinement, mathematical modeling
News Source: Drew Townsend. (October 29, 2026). Bigger Cell Clusters Signal More Slowly in Starving Social Amoebae. Scienmag.
