Dynamics of bacterial population growth in biofilms resemble spatial and structural aspects of urbanization

Abstract

Biofilms develop from bacteria bound on surfaces that grow into structured communities (microcolonies). Although surface topography is known to affect bacterial colonization, how multiple individual settlers develop into microcolonies simultaneously remains underexplored. Here, we use multiscale population-growth and 3D-morphometric analyses to assess the spatiotemporal development of hundreds of bacterial colonizers towards submillimeter-scale microcolony communities. Using an oral bacterium (Streptococcus mutans), we find that microbial cells settle on the surface randomly under sucrose-rich conditions, regardless of surface topography. However, only a subset of colonizers display clustering behavior and growth following a power law. These active colonizers expand three-dimensionally by amalgamating neighboring bacteria into densely populated microcolonies. Clustering and microcolony assembly are dependent on exopolysaccharides, while population growth dynamics and spatial structure are affected by cooperative or antagonistic microbes. Our work suggests that biofilm assembly resembles certain spatial-structural features of urbanization, where population growth and expansion can be influenced by type of settlers, neighboring cells, and further community merging and scaffolding occurring at various scales.

Fig. 1. Setup of the confocal laser scanning microscope (CLSM) and the image-processing software (BioSPA) to perform the spatial population-growth analysis over time.

a Experimental setup for analyzing microbial colonization and further growth under flow in situ using confocal laser scanning microscopy-surface topography imaging approach. b Schematics of biofilm constituents detected by independent signal detection of bacteria (green), exopolysaccharides substances (EPS; red), and hydroxyapatite disc (HAD) surface (gray). The HAD is placed inside the flow cell in the position as shown in the diagram. c Representative Z-projection (max intensity) of the CLSM images depicting the evolution of a surface colonizer (single cells, cluster, aggregate) into structured communities (microcolony). White bars: 5 µm. d Schematics of the image-processing method to capture the population units (i.e., single cells, clusters, aggregates, and microcolonies) across time and space, i.e., three dimensions and at multiple scales. The data were analyzed using biofilm spatiotemporal population analysis (BioSPA) software.

Fig. 2. Attachment and stability of the bacterial colonizers as a function of the surface topography.

a In situ visualization of surface colonizers (at t₀) on hydroxyapatite discs (HAD), overlapped with the surface topography (heightmap, left panel). Surface colonizers are shown in blue in the top view of a 3D representation of the bacterial cells on HAD (right panel). Z-axis (red line in right panel) represents a scale of 48 μm. Initial colonization of the surface consisted of single cells, clusters, and aggregates. b–d Topographical parameters average roughness (S_a), root-mean-square roughness (S_q), and skewness (S_sk) were determined at t₀, and after the nutrient flow started in the chamber (flow). e–g Normalized histograms of S_a, S_q, and S_sk values. For non-colonized areas (n.c.a.), the parameters were calculated for the whole-scanned area and at different length scales (multiscale area selection—MAS; shown in a). The results represent a large population of bacterial surface colonizers (C > 800), captured from independent experiments (n = 3). Source data are provided as a Source Data file.

Fig. 3. Dynamics of spatial population growth of surface-attached bacteria.

a Three-dimensional representation of the S. mutans growth under the flow of 1%-w/v sucrose at two different time points (0 and 510 min). Z-axis (red lines in (a)) represents a scale of 48 μm. b Time-lapsed heatmap of bacteria attachment and growth. Colored boxes indicate the time and location where bacteria attached. Elements shown in gray represent a projection of the bacteria signal at t = 300 min. Bacteria without colored boxes were not included in the population analysis. c Histogram with the initial volume values (V(0 min)) for surface colonizers at t₀ (in pink) and volume values at 420 min (V(420 min); in gray). Flow indicates the time that the culture medium starts to flow. d The evolution of the colonizers into microcolonies was assessed by biofilm growth curves (in gray-colored data points for each time point) and fitted curves (red-colored continuous lines) (n = 3). Source data are provided as a Source Data file.

Fig. 4. Multidimensional population-growth analysis of colonizers during biofilm development.

a V(420) as a function of P₀ for all bacterial population units (fitted and non-fitted). Bacterial growth followed the power law V(t) = a·t^b (for V(420 min) »V(0 min)), which represented around 40% of C (dark gray rectangle), termed dynamic colonizers (DC). b Distribution of V(420) as a function of V(0) for the non-fitted population and also for surface colonizers that do not grow (V(t) ≈V(0)). Non-fitted curves represented less than 0.1% of C. The results represent a large population of bacterial surface colonizers (C > 800), captured from independent experiments (n = 3). c, d Distribution values of the normalization constant (a) and the exponent (“b“) for the fitted curves. e Histogram depicting bacterial cell number of the surface colonizers at t₀ (P₀). f, g Distribution of “a” and “b” values as a function of P₀. Median is represented with red lines, and 1st and 3rd quartiles with dark gray lines (n = 3). Source data are provided as a Source Data file.

Fig. 5. Merging behavior of neighboring microcolonies.

a Time-lapsed confocal images of microcolonies development. Scale bars indicate 10 µm. b Projection map of binarized stacks from 540 min showing elements merging of multiple S. mutans microcolonies along time. c 3D scalar-field analysis (represented as Z-projection) performed for each marginal element merging to the central one. The volumes were calculated for three colored regions of the marginal elements as shown in c. Black dots represent the centroids of each element at 540 min, determined on the Z-projection. d Growth and e growth rate curve of the respective marginal elements. Colored lines indicate the moment of merging to the central element (n = 3). Source data are provided as a Source Data file.

Fig. 6. EPS production in situ mediates cell aggregation and microcolony structuring.

Z-projection (max intensity) of the CLSM images stack showing S. mutans growth on HAD in (a) 0.5 %-w/v fructose + 0.5 %-w/v glucose (0.5% Fru + 0.5% Glu) for bacteria (green), in (b) 1 %-w/v sucrose (1% Suc) for bacteria (green) and extracellular polymeric matrix (red), and in (c) 1 %-w/v sucrose with EPS-degrading enzymes (1% Suc, Dex/Mut) for bacteria (green) and extracellular polymeric matrix (red) over time. Scale bar indicates 2 μm. d Morphological analysis of the biofilm-forming elements formed over time, evaluated by the ratio between the volume and the convex hull volume (Volume/qhull volume). e Curves showing volume growth and surface area occupation of S. mutans growing on HAD in different conditions (n = 3). Source data are provided as a Source Data file.

Fig. 7. Growth dynamics in mixed communities.

a Growth curve (dots) and fitting (lines) for S. mutans in single-species biofilm (Sm; dark purple dots), in cross-kingdom biofilm (Sm-Ca; green dots), and in mixed-bacterial species biofilm (Sm-So; yellow dots). b Localized S. mutans and C. albicans growth curves at distinct sites (R1, R2, R3, and R4 as shown in c). Exponent b values are in parenthesis. c Z-projection (max intensity) of the CLSM images stack showing growth of S. mutans and C. albicans on HAD, with bacteria (green) and fungi (purple) channels, (e, f) zoom-in images of S. mutans superstructures formed in the presence of C. albicans (also highlighted by dotted-white line in f). d Z-projection of the CLSM image of C. albicans single-species biofilms. Scale bars indicate 10 µm. Source data are provided as a Source Data file.

Fig. 8. Schematic diagram of spatiotemporal population growth during biofilm development.

Dynamics of bacterial population growth during biofilm development resemble spatial and structural aspects of urbanization.