Investigating bacterial adhesion proteins as tools for controlling bacteria-surface interactions

Abstract

Adhesins are cell-surface proteins that initiate target binding and surface colonization by bacteria on the path to infection and biofilm formation. A deeper understanding of the interactions between adhesins and their ligands may lead to the development of strategies for controlling infection and biofilm formation. Using two ligand-binding domains from a model Repeats-in-toxin (RTX) adhesin in the Gram-negative marine bacterium Marinomonas primoryensis, I was able to prevent their binding to a photosynthetic diatom with which the bacteria form symbiotic biofilms underneath sea ice. The sugar-binding domain was optimally blocked with the monosaccharide fucose, and the peptide-binding domain was efficiently blocked with a peptide that ended in -Tyr-Thr-Asp. Homologs of these two ligand-binding domains are widespread in Gram-negative bacteria, including many pathogens. The same approach was employed to block the highly similar peptide-binding domain of FrhA adhesin of the pathogenic bacterium Vibrio cholerae from binding to the diatoms. Two adhesin engineering systems were explored as ways of altering the targeting of bacterial colonization. In one, redesigned and shortened versions of the M. primoryensis ice- and diatom-binding adhesin were tested for expression in Escherichia coli. In particular, the number of extender domains was reduced from 120 in the wild type to 15 or fewer. Although there was no evidence of ice-binding domain expression, antibodies to the extender domain and the diatom-binding domains detected these proteins on the outside of intact cells. A set of novel adhesins based on the beta-intimin system were designed using protein engineering techniques. In this series, a bacterial cellulose-binding module, and the peptide-binding domain of the M. primoryensis ice- and diatom-binding adhesin were separately fused to the end of the beta-intimin adhesin. In neither case was there any sign of external presentation of the binding domain. I suggest that these domains might have folded prior to passage through the adhesin pore, thus blocking export.