Pseudomonas aeruginosa type IV pili

Hazes et al., J. Mol. Biol. 299, p1005-17 (2000). Model of 1.5 turn of a type IV pilus based on the PAK pilin structure and fiber assembly parameters from Parge et al. (1995). The core of the fiber is formed by conserved highly hydrophobic a-helices surrounded by a variable sheet of b-sheet structure and loops. The proposed receptor binding loop is shown in red with a disaccharide receptor fragment modeled in to show its relative size.

P. aeruginosa and several other human pathogens use type IV pili to adhere to human tissue. Bacterial adhesion is critical for infection and blocking adhesion is therefore an interesting target to fight disease. We are using protein crystallography to characterize the mechanism of binding in order to develop anti-adhesin therapeutics.

Apart from the applied interest, the pilin research also aims to solve another puzzle. Amongst clinical isolates strains with extremely different pilin sequences can be found. The sequence of the helical residues (blue in the figure) are very to moderately well conserved and we believe that they are hidden from the immune system inside the fiber (as depicted). In contrast the remainder of the sequence, including the receptor binding loop (shown in red), can have below 20% sequence identity, apparently to evade detection by antibodies. In spite of the sequence divergence, all strains are believed to bind the same glycolipid asialo-GM1 and antibodies that bind the receptor binding site recognize multiple strains (and even a completely unrelated lectin in the yeast Candida albicans that also binds asialo-GM1). In the PAK structure we observed that the receptor binding loop displays a shallow pocket made exclusively from mainchain atoms. We have therefore proposed that conservation of binding specificity and antibody recognition without conserving the amino acid sequence can be due to conservation of the tertiary structure of the binding loop. Sequence variation is thus allowed as long as the mainchain structure of the residues that form the binding pocket remains the same. Our ongoing studies should reveal if this is true or yet some other mechanism exists.

Structural studies of the F-plasmid conjugation apparatus

Several bacterial plasmids can be transmitted between compatible bacteria by the process of conjugation. This is one of the mechanisms for the spread of antibiotic resistance and other virulence factors. A related system is also used by several pathogens to secrete virulence factors into the medium (for instance pertussis toxin) or to directly inject them into host cells (for instance H. pylori CagA).

The F-plasmid is the prototype system but after five decades of study little is known about the organisation of the membrane spanning protein complex that transfers the plasmid between donor and recipient cell. In a collaboration with Dr. Laura Frost (Dept. of Biological Sciences, UofA) we are targetting about a dozen proteins that are part of this conjugation apparatus (or transferosome). We will look at protein function, localisation, protein-protein interactions, and protein structure. Our ultimate goal is to understand not only the structures and functions of the components that form the conjugation apparatus, but also the way they interact to form a complex molecular pump.

At this moment there is a vacancy for a postdoctoral fellow to work in my laboratory on the structural biology part of the project, using X-ray diffraction as the main experimental technique. Funding is available immediately and applicants will be considered until the position is filled.

Ly-49 natural killer cell receptors

Tormo et al., Nature 402, p623 (1999)

Cells use MHC-I molecules to "inform" the immune system that they have been infected, for instance by a virus. Cytotoxic lymphocytes can then kill these cells and prevent further spread of the virus. However, several clever viruses are able to prevent the formation of MHC-I molecules and therefore make themselves invisible. This is where the natural killer (NK) cells come in as they recognize and kill cells that have no or only few MHC-I molecules. The receptors involved in this process are KIR receptors in humans and Ly-49 receptors in mice and rats. We are looking at the Ly-49 family to understand how they detect the presence of MHC-I.