Mechanical Interactions Govern Self-Organized Ordering in Bacterial Colonies on Surfaces

TL;DR

This study provides a physical framework for understanding how mechanical interactions shape the self-organization of bacterial communities under surface confinement.

Abstract

Bacterial colonies growing on surfaces are shaped by mechanical stresses transmitted through the community, governed by the balance between cell growth and steric and cell-substrate interactions. Using overdamped dynamics simulations of nonmotile, stress-responsive bacteria, we examine how purely mechanical interactions determine colony morphology and internal organization. Growth-induced extensile stresses compete with steric constraints, giving rise to the spontaneous formation of microdomains composed of highly aligned cells. We characterize this self-organization through the distribution of microdomain areas and a nematic order parameter that quantifies colony-wide alignment. Mechanosensitivity does not systematically alter domain structure, but increasing substrate friction reduces the mean domain size and broadens the diversity of orientations. Shifting the balance toward steric interactions, by lengthening the cell division size, slows the relaxation of colony shape toward isotropy and broadens the distribution of contact forces, producing a slower exponential decay. In dense colonies, strong forces are transmitted anisotropically through chains of aligned neighbors within microdomains. These findings demonstrate that colony-level morphology and stress organization can emerge from local mechanical interactions alone, even without requiring biochemical signaling. By linking microscopic force transmission to macroscopic growth dynamics, our study provides a physical framework for understanding how mechanical interactions shape the self-organization of bacterial communities under surface confinement.

Figures (6)

• Figure 1: (a) Schematic of capsule geometry and cell-cell contact. (b) Cell division into two daughter cells when the division length $l_{_\text{d}}$ is reached. (c) Example of a final colony configuration showing two microdomains $i$ and $j$ with mean orientations $\Psi_i$ and $\Psi_j$ relative to the $x$ axis. Colors correspond to domain orientation as indicated by the color wheel.
• Figure 2: (a) Final configurations of bacterial colonies growing within circular confinement for two substrate friction coefficients $\zeta\,{=}\,200$ and $1000\,\text{Pa.h}$. Other parameters: $l_{_\text{d}}\,{=}\,3\,\mu\text{m}$, $\beta\,{=}\,0.02\,(\mu\text{m.kPa.h})^{-1}$. (b) Probability distribution of rescaled microdomain areas for different friction values. (c) Mean microdomain area as a function of substrate friction. Error bars indicate the standard error of the mean. The dotted line indicates the approximate mean microdomain size in frictionless packings of passive particles whose elongation matches the maximum elongation of the bacteria; see text.
• Figure 3: (a,b) Probability distribution of domain orientations, $P(\Psi)$, and probability distribution of the magnitude of the cross product between the unit orientation vectors of contacting bacteria, $P(|\hat{l}_i{\times}\hat{l}_j|)$, for the final colony configurations and different values of the substrate friction coefficient $\zeta$. Lower friction promotes stronger local alignment, whereas higher friction leads to a more isotropic orientation distribution. Other parameters: $l_{_\text{d}}\,{=}\,3\,\mu\text{m}$, $\beta\,{=}\,0.02\, (\mu\text{m.kPa.h})^{-1}$. (c) Nematic order parameter versus substrate friction coefficient for the final configurations (full circles) and freely growing colonies before reaching the boundaries (open circles). (d) Evolution of the minor-to-major axis ratio of the colony as a function of bacterial population size for both friction coefficients. (e,f) $P(|\hat{l}_i{\times}\hat{l}_j|)$ and $P(\Psi)$ for freely growing colonies and different values of $\zeta$.
• Figure 4: (a) Final configurations of bacterial colonies growing under circular confinement for two different values of mechanosensitivity $\beta\,{=}\,0.002$ and $0.2\,(\mu\text{m.kPa.h})^{-1}$. Other parameters: $l_{_\text{d}}\,{=}\,3\, \mu\text{m}$, $\zeta\,{=}\,200\,\text{Pa.h}$. (b) Probability distribution of microdomain areas for different values of $\beta$. (c) Mean microdomain area $\langle S \rangle$ and global nematic order parameter $\sigma$ vs $\beta$.
• Figure 5: (a) Simulated configurations at concurrent time steps for $\zeta\,{=}\,200 \,\text{Pa.h}$, $\beta\,{=}\,0.02\,(\mu\text{m.kPa.h})^{-1}$, and different division lengths $l_{_\text{d}}\,{=}\,2\,\mu\text{m}$ (left), $3\,\mu\text{m}$ (middle), and $4\,\mu\text{m}$ (right). The corresponding number of bacteria are $N\,{=}\,2853$, $1265$, and $505$, respectively. Color circle shows the orientational distribution. (b) Evolution of the minor-to-major axis ratio of the bacterial colony as a function of the number of bacteria within the colony for different division lengths. Insets show how the global shape anisotropy of the colony evolves towards the final isotropic one. Black arrows mark the moment when the expanding colony reaches the confining walls.