Propionibacterium acnes inhibits FOXM1 and induces cell cycle alterations in human primary prostate cells.

Propionibacterium acnes has been detected in diseased human prostate tissue, and cell culture experiments suggest that the bacterium can establish a low-grade inflammation. Here, we investigated its impact on human primary prostate epithelial cells. Microarray analysis confirmed the inflammation-inducing capability of P. acnes but also showed deregulation of genes involved in the cell cycle. qPCR experiments showed that viable P. acnes downregulates a master regulator of cell cycle progression, FOXM1. Flow cytometry experiments revealed that P. acnes increases the number of cells in S-phase. We tested the hypothesis that a P. acnes-produced berninamycin-like thiopeptide is responsible for this effect, since it is related to the FOXM1 inhibitor siomycin. The thiopeptide biosynthesis gene cluster was strongly expressed; it is present in subtype IB of P. acnes, but absent from type IA, which is most abundant on human skin. A knock-out mutant lacking the gene encoding the berninamycin-like peptide precursor was unable to downregulate FOXM1 and to halt the cell cycle. Our study reveals a novel host cell-interacting activity of P. acnes.

Antagonism between Staphylococcus epidermidis and Propionibacterium acnes and its genomic basis.

BACKGROUND: Propionibacterium acnes and Staphylococcus epidermidis live in close proximity on human skin, and both bacterial species can be isolated from normal and acne vulgaris-affected skin sites. The antagonistic interactions between the two species are poorly understood, as well as the potential significance of bacterial interferences for the skin microbiota. Here, we performed simultaneous antagonism assays to detect inhibitory activities between multiple isolates of the two species. Selected strains were sequenced to identify the genomic basis of their antimicrobial phenotypes. RESULTS: First, we screened 77 P. acnes strains isolated from healthy and acne-affected skin, and representing all known phylogenetic clades (I, II, and III), for their antimicrobial activities against 12 S. epidermidis isolates. One particular phylogroup (I-2) exhibited a higher antimicrobial activity than other P. acnes phylogroups. All genomes of type I-2 strains carry an island encoding the biosynthesis of a thiopeptide with possible antimicrobial activity against S. epidermidis. Second, 20 S. epidermidis isolates were examined for inhibitory activity against 25 P. acnes strains. The majority of S. epidermidis strains were able to inhibit P. acnes. Genomes of S. epidermidis strains with strong, medium and no inhibitory activities against P. acnes were sequenced. Genome comparison underlined the diversity of S. epidermidis and detected multiple clade- or strain-specific mobile genetic elements encoding a variety of functions important in antibiotic and stress resistance, biofilm formation and interbacterial competition, including bacteriocins such as epidermin. One isolate with an extraordinary antimicrobial activity against P. acnes harbors a functional ESAT-6 secretion system that might be involved in the antimicrobial activity against P. acnes via the secretion of polymorphic toxins. CONCLUSIONS: Taken together, our study suggests that interspecies interactions could potentially jeopardize balances in the skin microbiota. In particular, S. epidermidis strains possess an arsenal of different mechanisms to inhibit P. acnes. However, if such interactions are relevant in skin disorders such as acne vulgaris remains questionable, since no difference in the antimicrobial activity against, or the sensitivity towards S. epidermidis could be detected between health- and acne-associated strains of P. acnes.