|M.Sc Student||Mayer Michal|
|Subject||Structure-function Characterization of Bacterial Tyrosine|
Kinases as Antibacterial Drug Targets
|Department||Department of Biology||Supervisor||Professor Meytal Landau|
The increasing antibiotics resistance of pathogenic bacteria calls for the design of new antibacterial drugs with novel mechanisms of action. Bacterial tyrosine kinases (BY-kinase) present promising antibacterial target due to their involvement in pathways associated with virulence such as the metabolism of polysaccharides and biofilm formation. Moreover, BY-kinases share no homology with mammalian proteins, allowing the design of specific drugs.
We initially focused on the BY-kinase from Burkholderia cepacia named BceF. B. cepacia is an important human pathogen, which causes pneumonia in immunocompromised individuals. BceF is involved in the production of exopolysaccharides, which are a part of the biofilm matrix formed by B. cepacia and thus contributes to its virulence. We postulated that blocking the active site of this kinase would diminish biofilm formation. We determined the crystal structure of BceF kinase domain at 1.85Å resolution, and analyzed its enzymatic activity and kinetics. We discovered that the recombinantly expressed kinase domain of BceF shows a low enzymatic efficiency in phosphorylating a tyrosine-rich peptide substrate derived from the BceF C-terminal domain. We then found that the binding affinity of the kinase domain to the ATP analog, AMP-PNP, is low (mM range), which might explain the low enzymatic efficiency. We postulated that the most likely explanation for the observed low binding affinity is the formation of a non-physiological stable dimer observed in solution as well as in crystals. Specifically, the crystal structure of the BceF kinase domain revealed a dimer in the asymmetric unit of the crystal, while the interface between the monomers appeared to block the active site. The dimeric form in the crystal structure, which we analyzed to be non-physiological, most likely prevents the release of ADP from the catalytic pocket and the binding of ATP. We concluded that the self-assembly tendency of isolated BY-kinase domains severely interfered with the activity of BceF. This result might contribute to future analysis of other BY-kinases as well as various enzymes tested as isolated catalytic domains in-vitro.
Computational screening of drug libraries, conducted by the group of Nir Ben-Tal from Tel-Aviv University, using the crystal structure of BceF kinase domain, provided a list of potential inhibitors that we experimentally tested for their effect on kinase activity. The low enzymatic efficiency of BceF kinase domain prevented a reliable assessment of the potential inhibitors. Nevertheless, three compounds suggested by the in-silico screen showed some inhibitory effect. We thus collaborated with Prof. Ehud Banin from Bar-Ilan University to test the compounds in-vivo on Pseudomonas aeruginosa. In preliminary results one of the compounds showed an effect on reducing the biofilm biomass of the bacteria, interestingly, this compound was also the most potent in inhibiting BceF in-vitro. Since there are no known BY-kinases in P. aeruginosa, we used bioinformatics methods and found two potential target proteins in P. aeruginosa that are homologous to BceF.