Comparison of the Genetic Features Involved in Bacillus subtilis Biofilm Formation Using Multi-Culturing Approaches

Abstract

Surface-associated multicellular assemblage is an important bacterial trait to withstand harsh environmental conditions. Bacillus subtilis is one of the most studied Gram-positive bacteria, serving as a model for the study of genetic pathways involved in the different steps of 3D biofilm formation. B. subtilis biofilm studies have mainly focused on pellicle formation at the air-liquid interface or complex macrocolonies formed on nutritive agar. However, only few studies focus on the genetic features of B. subtilis submerged biofilm formation and their link with other multicellular models at the air interface. NDmed, an undomesticated B. subtilis strain isolated from a hospital, has demonstrated the ability to produce highly structured immersed biofilms when compared to strains classically used for studying B. subtilis biofilms. In this contribution, we have conducted a multi-culturing comparison (between macrocolony, swarming, pellicle, and submerged biofilm) of B. subtilis multicellular communities using the NDmed strain and mutated derivatives for genes shown to be required for motility and biofilm formation in pellicle and macrocolony models. For the 15 mutated NDmed strains studied, all showed an altered phenotype for at least one of the different culture laboratory assays. Mutation of genes involved in matrix production (i.e., tasA, epsA-O, cap, ypqP) caused a negative impact on all biofilm phenotypes but favored swarming motility on semi-solid surfaces. Mutation of bslA, a gene coding for an amphiphilic protein, affected the stability of the pellicle at the air-liquid interface with no impact on the submerged biofilm model. Moreover, mutation of lytF, an autolysin gene required for cell separation, had a greater effect on the submerged biofilm model than that formed at aerial level, opposite to the observation for lytABC mutant. In addition, B. subtilis NDmed with sinR mutation formed wrinkled macrocolony, less than that formed by the wild type, but was unable to form neither thick pellicle nor structured submerged biofilm. The results are discussed in terms of the relevancy to determine whether genes involved in colony and pellicle formation also govern submerged biofilm formation, by regarding the specificities in each model.