While the global distribution of microplastics (MP) in the marine environment is currently being critically evaluated, the potential role of MP as a vector for distinct microbial assemblages or even pathogenic bacteria is hardly understood. To gain a deeper understanding, we investigated how different in situ conditions contribute to the composition and specificity of MP-associated bacterial communities in relation to communities on natural particles. Polystyrene (PS), polyethylene (PE), and wooden pellets were incubated for 2 weeks along an environmental gradient, ranging from marine (coastal Baltic Sea) to freshwater (waste water treatment plant, WWTP) conditions. The associated assemblages as well as the water communities were investigated applying high-throughput 16S rRNA gene sequencing. Our setup allowed for the first time to determine MP-dependent and -independent assemblage factors as subject to different environmental conditions in one system. Most importantly, plastic-specific assemblages were found to develop solely under certain conditions, such as lower nutrient concentration and higher salinity, while the bacterial genus Erythrobacter, known for the ability to utilize polycyclic aromatic hydrocarbons (PAH), was found specifically on MP across a broader section of the gradient. We discovered no enrichment of potential pathogens on PE or PS; however, the abundant colonization of MP in a WWTP by certain bacteria commonly associated with antibiotic resistance suggests MP as a possible hotspot for horizontal gene transfer. Taken together, our study clarifies that the surrounding environment prevailingly shapes the biofilm communities, but that MP-specific assemblage factors exist. These findings point to the ecological significance of specific MP-promoted bacterial populations in aquatic environments and particularly in plastic accumulation zones.

Summary: Plastids (chloroplasts) possess an enormous capacity to synthesize and accumulate foreign proteins. Here we have maximized chloroplast protein production by over‐expressing a proteinaceous antibiotic against pathogenic group A and group B streptococci from the plastid genome. The antibiotic, a phage lytic protein, accumulated to enormously high levels (>70% of the plant’s total soluble protein), and proved to be extremely stable in chloroplasts. This massive over‐expression exhausted the protein synthesis capacity of the chloroplast such that the production of endogenous plastid‐encoded proteins was severely compromised. Our data suggest that this is due to translational rather than transcriptional limitation of gene expression. We also show that the chloroplast‐produced protein antibiotic efficiently kills the target bacteria. These unrivaled expression levels, together with the chloroplast’s insensitivity to enzymes that degrade bacterial cell walls and the elimination of the need to remove bacterial endotoxins by costly purification procedures, indicate that this is an effective plant‐based production platform for next‐generation antibiotics, which are urgently required to keep pace with rapidly emerging bacterial resistance.

A putative collagen-binding MSCRAMM, Ace, of Enterococcus faecalis was identified by searching bacterial genome data bases for proteins containing domains homologous to the ligand-binding region of Cna, the collagen-binding MSCRAMM fromStaphylococcus aureus. Ace was predicted to have a molecular mass of 71 kDa and contains features characteristic of cell surface proteins on Gram-positive bacteria, including a LPXTG motif for cross-linking to the cell wall. The N-terminal region of Ace contained a region (residues 174–319) in which 56% of the residues are identical or similar when compared with the minimal ligand-binding region of Cna (Cna 151–318); the remainder of the Ace A domain has 46% similarity with the corresponding region of the Cna A domain. Antibodies raised against recombinant Ace A domain were used to verify the cell surface expression of Ace on E. faecalis. These antibodies also effectively inhibited the adhesion of enterococcal cells to a collagen substrate, suggesting that Ace is a functional collagen-binding MSCRAMM. Structural modeling of the conserved region in Ace (residues 174–319) suggested a structure very similar to that reported for residues 151–318 of the Cna collagen-binding domain in which the ligand-binding site was identified as a trench transversing a β-sheet face (Symersky, J., Patti, J. M., Carson, M., House-Pompeo, K., Teale, M., Moore, D., Jin, L., DeLucas, L. J., Höök, M., and Narayana, S. V. L. (1997) Nat. Struct. Biol. 10, 833–838). Biochemical analyses of recombinant Ace and Cna A domains supported the modeling data in that the secondary structures were similar as determined by CD spectroscopy and both proteins bound at multiple sites in type I collagen with micromolar affinities, but with different apparent kinetics. We conclude that Ace is a collagen-binding MSCRAMM on enterococci and is structurally and functionally related to the staphylococcal Cna protein.