Fig. 5: Active fluctuations at the periphery of colonies.
a Simulated setup for quantifying boundary fluctuations by measuring the radial distance rCMS between a fixed angular position on the surface and the colony center. b Active pilus retraction results in a slower decay of the velocity autocorrelation function (VACF) of the surface point. Main plot: colonies are first grown from cells with retracting pili and the role of pili is studied after growth is switched off. Inset: colonies are grown with retraction-deficient cells. c Active pilus retraction produces a power spectral density of fluctuations characteristic for a visco-elastic material with an elastic behavior at low frequencies. For passive colonies, \({v}_{{{{{{{{\rm{re}}}}}}}}}=0\), bond rupture results in a viscous material behavior. Colony size is 4000 cells. For comparison with experimental data, a small colony with 600 mutant cells is simulated having a higher binding rate \({k}_{{{{{{{{\rm{bind}}}}}}}}}^{{{{{{{{\rm{mutant}}}}}}}}}=50\,{{{{{{{{\rm{s}}}}}}}}}^{-1}\), lower rupture rate \({k}_{{{{{{{{\rm{rupt}}}}}}}}}^{{{{{{{{\rm{mutant}}}}}}}}}=1\,{{{{{{{{\rm{s}}}}}}}}}^{-1}\) and lower retraction velocity \({v}_{{{{{{{{\rm{re}}}}}}}}}=0.5\,\upmu {{{{{{{\rm{m}}}}}}}}/{{{{{{{\rm{s}}}}}}}}\). Inset: experimental data for wild-type cells and a ΔpptA strain, error bars show standard deviations with three samples. d Simulated setup for colony-shape perturbation. e Simulated mutant colonies with retraction-deficient pili that form permanent bonds. The equilibrium fluctuation-response relationship holds. f For simulated wild-type cells, the equilibrium fluctuation-response relationship is strongly violated (\({v}_{{{{{{{{\rm{re}}}}}}}}}=1.0\,\upmu {{{{{{{\rm{m}}}}}}}}/{{{{{{{\rm{s}}}}}}}}\), dbind = 3.0 μm).
