Microstructural and Rheological Transitions in Bacterial Biofilms

Abstract

Biofilms are aggregated bacterial communities structured within an extracellular matrix (ECM). ECM controls biofilm architecture and confers mechanical resistance against shear forces. From a physical perspective, biofilms can be described as colloidal gels, where bacterial cells are analogous to colloidal particles distributed in the polymeric ECM. However, the influence of the ECM in altering the cellular packing fraction (ϕ) and the resulting viscoelastic behavior of biofilm remains unexplored. Using biofilms of Pantoea sp. (WT) and its mutant (ΔUDP), the correlation between biofilm structure and its viscoelastic response is investigated. Experiments show that the reduction of exopolysaccharide production in ΔUDP biofilms corresponds with a seven-fold increase in ϕ, resulting in a colloidal glass-like structure. Consequently, the rheological signatures become altered, with the WT behaving like a weak gel, whilst the ΔUDP displayed a glass-like rheological signature. By co-culturing the two strains, biofilm ϕ is modulated which allows us to explore the structural changes and capture a change in viscoelastic response from a weak to a strong gel, and to a colloidal glass-like state. The results reveal the role of exopolysaccharide in mediating a structural transition in biofilms and demonstrate a correlation between biofilm structure and viscoelastic response.

Keywords: Payne effect; biofilms; extracellular exopolysaccharides; packing fraction; viscoelasticity.

Figure 1

Influence of exopolysaccharide production on the structure of Pantoea sp. biofilms. A) Pictures of WT (top panel) and ΔUDP (bottom panel) biofilms on agar plates captured using a digital single lens reflex camera after 72 h of inoculation. B) The amount of exopolysaccharide (in nmol sugar/µg protein) obtained from WT and ΔUDP biofilms after 72 h of growth, using a modified phenol sulphuric method. (Sugar concentration was measured as glucose equivalents and protein concentration measured the A280 using the Nanodrop. Error bars represent one standard deviation, n ⩾ 3, a student's T‐test was applied to test for significance, * < 0.05, ** < 0.01, *** < 0.001). C) The evaporated water content of the WT and ΔUDP biofilms grown on 1.5% agar plates at the end of 24 h (error bars represent one standard deviation, n = 5, a student's T‐test was applied to test for significance, * < 0.05, ** < 0.01, *** < 0.001). D) Time‐lapse phase contrast images of WT (top panel) and ΔUDP (bottom panel) cells on an agar pad. The scale bar for all sub‐panels is 25 µm. E) A plot of biofilm's colony packing fraction (${\varphi }_{\mathrm{colony}}$) as a function of time for the WT and ΔUDP biofilms grown within a confined agar pad over a period of 150 minutes (error bars represent one standard deviation, n ⩾ 5 colonies).

Figure 2

Reduction of exopolysaccharide production affected the rheology of WT and ΔUDP biofilms. A) Variation in elastic modulus ($G^{{}^{\prime }}$, filled circle) and viscous modulus ($G^{{}^{\prime \prime }}$, filled triangle) as a function of applied γ at 0.5 Hz. ${\gamma }_y$ denotes the value of yield strain below which the viscoelastic response is considered to be linear. Above ${\gamma }_y$ nonlinear effects appear in the rheological response. (described in Figure S2, Supporting Information). ${\gamma }_c$ is the cross‐over strain beyond which viscous response dominates the elastic response in the biofilms. B) Variation in $G^{{}^{\prime }}$ and $G^{{}^{\prime \prime }}$ as a function of applied frequency at $\gamma =1\%$ . C) Plot of the intracyle thickening (ν₃/ν₁) (ratio of third to first order viscous Chebyshev coefficients) for WT as a function of applied strain. D) Plot of the intracyle thickening (ν₃/ν₁) for ΔUDP as a function of applied γ. E) Statistical significance of the (ν₃/ν₁) _max (maximum thickening ratio at a given frequency) for the WT biofilms (n ⩾ 3). A paired student t‐test was applied, (* < 0.05, ** < 0.01, *** < 0.001) F) Statistical significance of the (ν₃/ν₁) _max (maximum thickening ratio at a given frequency) for the ΔUDP biofilms (n ⩾ 3). A paired student t‐test was applied, (* < 0.05, ** < 0.01, *** < 0.001).

Figure 3

Ratio of WT to ΔUDP altered the structure and evaporated water content of co‐cultured biofilms. A) Representative biofilms of WT and ΔUDP co‐cultures after 6 h on an agar pad. Red arrows indicate the location of the colony in which WT cells are surrounded by densely packed ΔUDP cells. B) (top) Zoom‐in view of a mixed Pantoea sp. colony and (bottom) shows the same field of view 60 min later. ECM secreted from the center of cellular cluster reduced the local packing density of colonies. C) Calculated ϕ of the Pantoea sp. biofilms with different inoculation ratios of WT to ΔUDP grown on an agar plate for 48 h. D) Evaporated water content of the co‐cultured biofilms after 24 h of growth (error bars represent one standard deviation, n=5, a student's T‐test was applied to test for significance, * < 0.05, ** < 0.01, *** < 0.001). E) Representative confocal images of biofilms 10 µm from the coverslip surface after 48 h of growth, which were started from different inoculation ratios of WT to ΔUDP. Decreasing the ratio of WT caused reduction of exopolysaccharide, transitioning the structure of biofilm into a jammed state.

Figure 4

Reduction of exopolysaccharide correlated with changes in the viscoelastic response of biofilms. A) Variation of $G^{\prime }$as a function of the γ on a log–log plot. Color codes represent the ϕ of biofilms (error bars represent the standard deviation of the mean, n⩾3) B) $G^{\prime \prime }$as a function of γ on a log–log plot (error bars represent standard deviation of the mean, n⩾3) C) Plot of linear $G^{\prime }$ and linear $G^{\prime \prime }$as a function of ϕ. Lines with a slope of two and five are shown for reference. (error bars represent the standard deviation of the mean, n⩾3) D) Plot of cross‐over stress (σ _c ) and cross‐over strain (${\gamma }_c$) as a function of the biofilm's ϕ. (error bars represent the standard deviation of the mean, n⩾3) E) A heat‐map of mean stiffening ratio (e ₃/e ₁, ratio of third to first order elastic Chebyshev coefficients) as a function of γ and ϕ at 0.75 Hz. Stars denote the location of the maximum value of e ₃/e ₁ at a constant ϕ (n⩾3, standard deviation heat maps are included in Figure S5A, Supporting Information). F) A heat‐map of mean thickening ratio (ν₃/ν₁) as a function of γ and ϕ at 0.75 Hz. Stars denote the location of the maximum value of ν₃/ν₁ at a constant ϕ (n⩾3, standard deviation heat maps are included in Figure S5B, Supporting Information).