Pulcherrimin formation controls growth arrest of the Bacillus subtilis biofilm

Abstract

Biofilm formation by Bacillus subtilis is a communal process that culminates in the formation of architecturally complex multicellular communities. Here we reveal that the transition of the biofilm into a nonexpanding phase constitutes a distinct step in the process of biofilm development. Using genetic analysis we show that B. subtilis strains lacking the ability to synthesize pulcherriminic acid form biofilms that sustain the expansion phase, thereby linking pulcherriminic acid to growth arrest. However, production of pulcherriminic acid is not sufficient to block expansion of the biofilm. It needs to be secreted into the extracellular environment where it chelates Fe3+ from the growth medium in a nonenzymatic reaction. Utilizing mathematical modeling and a series of experimental methodologies we show that when the level of freely available iron in the environment drops below a critical threshold, expansion of the biofilm stops. Bioinformatics analysis allows us to identify the genes required for pulcherriminic acid synthesis in other Firmicutes but the patchwork presence both within and across closely related species suggests loss of these genes through multiple independent recombination events. The seemingly counterintuitive self-restriction of growth led us to explore if there were any benefits associated with pulcherriminic acid production. We identified that pulcherriminic acid producers can prevent invasion by neighboring communities through the generation of an “iron-free” zone, thereby addressing the paradox of pulcherriminic acid production by B. subtilis.

Machine learning uncovers independently regulated modules in the Bacillus subtilis transcriptome

“…We also see PchR activation in late-stage biofilm, glucose exhaustion, and phosphate starvation experiments. These results agree with a recent study that found pulcherrimin to be an important intercellular signal for the stationary phase that also helps exclude competing bacteria from established biofilms 47 . The regulation mechanisms of iModulons like this one can be the subject of future research.…”

Machine learning uncovers independently regulated modules in the Bacillus subtilis transcriptome

“…We also see PchR activation in late stage biofilm, glucose exhaustion, and phosphate starvation experiments. These results agree with a recent study that found pulcherrimin to be an important intercellular signal for the stationary phase that also helps exclude competing bacteria from established biofilms 38 . The regulation mechanisms of i-modulons like this one can be the subject of future research.…”

iModulonDB: a knowledgebase of microbial transcriptional regulation derived from machine learning

“…For instance, we observed that the iron chelator pulcherrimin was active in stationary phase signaling in B. subtilis ( 23 ). This observation was recently externally validated ( 24 ). More broadly, each decomposition explained 65–80% of the variance in the transcriptomic data sets, indicating that the functions and regulators identified captured most of the transcriptional functions of the TRNs under the conditions where the data was obtained.…”