Structural studies on a 1.5-MDa bacterial adhesin reveal its adhesive and cohesive properties in biofilm formation

Abstract

Ice-binding proteins (IBPs) have long been known to help organisms resist freezing (antifreeze) or tolerate freezing (ice recrystallization inhibition). Discoveries made in my thesis research reveal a third function for an IBP, which is host adhesion to ice. The IBP found on the cell surface of an Antarctic Gram-negative bacterium, Marinomonas primoryensis, is part of an exceptionally large protein of 1.5 MDa. The highly motile M. primoryensis swims to ice and attaches to the surface to form bacterial clusters. We suggest that MpAFP facilitates the formation of bacterial communities or biofilms underneath lake or sea ice, where oxygen and nutrients are most abundant due to the photosynthetic activity of other microorganisms. MpAFP can be divided into five distinct regions. The N-terminal end (RI) attaches to the bacterial cell envelope, likely by binding to the cell wall peptidoglycan or other polysaccharides on the cell surface while spanning the outer membrane. Next are ~ 120 identical tandem repeats of a 104-aa Ig-like domain (RII) that makes up 90% of the protein. Region III, which contains a carbohydrate-binding domain, separates the highly repetitive region II (RII) from the moderately repetitive region IV (RIV). The 34-kDa RIV, which folds as a Ca2+- dependent β-solenoid, is the only ice-binding domain of MpAFP. At the C terminus RV also contains Ca2+-binding RTX repeats that may serve as the secretion sequence for the giant protein. Using a “dissect and build” approach I succeeded in piecing together >95% of the structure of MpAFP. With this knowledge, I was able to postulate a general mechanism by which RTX adhesins enable their hosts to form biofilms by using a combination of surface adhesion and cell cohesion. X-ray crystallography suggests tandem arrays of the Ig-like repeats (RII) attract and slide along each other in an antiparallel fashion until their sugar binding domains lock onto each other’s’ surface glycans. Thus MpAFP not only binds the bacteria to ice but can potentially link the bacteria together in clusters to increase the total number of adhesins binding the group to the underside of ice where the environment is most favourable for growth. Given that many bacteria produce adhesins similar to MpAFP, work described in this thesis can be extended to give insight into the formation and disruption of other bacterial biofilms, including those of human pathogens.