Confinement inhibits surficial attachment and induces collective behaviors in bacterial colonies

TL;DR

The effects of confinement to 2D on the collective behaviors of pathogenic bacteria are described using novel imaging and computational analysis techniques to demonstrate the remarkable breadth of collective behaviors exhibited by bacteria in different environments.

Abstract

Bacterial colonies are a well-known example of living active matter, exhibiting collective behaviors such as nematic alignment and collective motion that play an important role in the spread of microbial infections. While the underlying mechanics of these behaviors have been described in model systems, many open questions remain about how microbial self-organization adapts to the variety of different environments bacteria encounter in natural and clinical settings. Here, using novel imaging and computational analysis techniques, the effects of confinement to 2D on the collective behaviors of pathogenic bacteria are described. Biofilm-forming Pseudomonas aeruginosa are grown on different substrates, either open to the surrounding fluid or confined to a single monolayer between two surfaces. Orientational ordering in the colony, cell morphologies, and trajectories are measured using single-cell segmentation and tracking. Surprisingly, confinement inhibits permanent attachment and induces twitching motility, giving rise to multiple coexisting collective behaviors. This effect is shown to be independent of the confining material and the presence of liquid medium. The nematic alignment and degree of correlation in the cells' trajectories determines how effectively bacteria can invade the space between two surfaces and the 3D structure of the colony after several days. Confinement causes the formation of dynamic cell layers driven by collective motion as well as collective verticalization leading to the formation of densely packed crystalline structures exhibiting long-range order. These results demonstrate the remarkable breadth of collective behaviors exhibited by bacteria in different environments, which must be considered to better understand bacterial colonization of surfaces.

Figures (7)

• Figure 1: Unconfined monolayers. (A) Diagram of microfluidic flow chamber for bacterial growth on unconfined surfaces in liquid growth medium tryptic soy broth (TSB). (B) Representative microcolony of unconfined P. aeruginosa on glass in liquid growth medium. Cells on the right side are colored by their nematic order parameter $S_i$. (C) Diagram of unconfined bacteria colony growing at a solid-air interface on solid growth medium tryptic soy agar (TSA) (D) Representative microcolony of unconfined P. a. on solid TSA, cells on the right are colored by $S_i$. (E) Single-cell nematic order parameter $S_i$ as a function of cell aspect ratio $\eta$ for unconfined P. a. colonies. (F) Nematic order parameter as a function of cell packing fraction $\phi$ for unconfined P. a..
• Figure 2: Modes of collective motion under confinement. (A) Experimental conditions leading to distinct collective behaviors under 2D confinement: confinement between PDMS and glass or between TSA and glass. (B)-(C) representative single-cell trajectories for chaotic twitching and collective twitching, respectively. Colored dots and lines are subsets of trajectories over 1-3 minutes (see supporting videos 5 and 6). (D) Representative image of bacteria exhibiting chaotic twitching. Right half shows single-cell nematic order parameters. (E) Bacteria exhibiting collective twitching, along with nematic order parameters. (F) Distributions of single-cell aspect ratios from different modes of collective behaviors for unconfined and confined bacteria. Black dots represent mean values. (G) Distributions of single-cell nematic order parameters from different modes of collective behaviors and experimental conditions. Black dots represent mean values. For each distribution in F and G, $N>20000$ cells. Scale bars are 20 µ m.
• Figure 3: Quantitative description of collective behaviors. (A) Single-cell nematic order parameter vs. cell aspect ratio for different modes and conditions. Error bars represent standard errors, legend applies to A and B. (B) Single-cell nematic order parameter vs. cell packing fraction. (C) Mean square displacements as a function of the lag. Lines represent best power law fits, with exponents given in the legend. (D) Spatial velocity correlation functions $C(r)$ vs. cell-cell distance $r$. Lines represent best power law (chaotic twitching) or exponential (collective twitching) fits. Legend applies to both C and D.
• Figure 4: Chaotic twitching leads to mass verticalization. (A) Edge of colony undergoing chaotic twitching. Right side: segmented cells colored by their projected aspect ratio (the aspect ratio of each cell in the focal plane). (B) Projected aspect ratios $\eta_P$ as a function of the distance from the edge of the colony, measured only from colonies undergoing chaotic twitching. (C) Snapshots from z-stack of colony undergoing chaotic twitching after about 20 hrs, with focal planes $1$ µ m apart. Verticalized cells are shown breaking out of the monolayer. Right: representative orthogonal slice from a z-stack showing verticalization near the edge of the colony. (D) Representative image of colony after 36-48 hrs, with nearly all cells having tilted vertically upwards. Right side shows single-cell segmentation masks (random colors). Inset: closeup of densely-packed vertical cells exhibiting highly ordered, hexagonal ordering. (E) Probability density function of projected aspect ratio $\eta_P$ in unconfined biofilms grown on glass (in liquid medium) or TSA (solid medium) and confined colonies between glass and TSA. Confined colonies exhibit mass verticalization resulting in smaller values of $\eta_P$. See Fig S2 for representative images.
• Figure 5: Collective twitching leads to layer formation. (A) Top: schematic of confined cells pushing against the glass-TSA interface during colony expansion. Bottom: diagram of $+1/2$ and $-1/2$ topological defects. (B) Representative confocal image of colony edge undergoing collective twitching, with $\pm1/2$ topological defects shown in red and blue, respectively. (C) Maximum intensity projection over time obtained from a video of colony expansion under collective twitching. Arrows represent major collective flows in the colony (see supporting video 7). (D) Outlines of the colony at successive time points $30$ s apart showing outward colony expansion. (E) Widefield fluorescence image of colony during layer formation. Brighter areas correspond to more cell layers. (F) Widefield image colored by the number of cell layers. Inset: histogram of pixel intensities, showing 3 distinct peaks corresponding to different numbers of layers.