Abstract
CDK5 is an atypical protein kinase belonging to the family of CDKs discovered in the central nervous system where it is predominantly expressed and has long been considered as neurospecific. CDK5 is associated with its partner p35/p25 and allows the regulation of many neuronal functions such as neuronal migration, synapse formation and neuronal plasticity. However, we now know that this kinase is expressed ubiquitously and also has functions outside the brain. The deregulation of its kinase activity contributes to the development of neurodegenerative diseases such as Alzheimer's or Parkinson's. But CDK5 is also involved in the development of cancers and has been identified in lung cancer for its important role in the proliferation and migration of tumor cells that lead to the development of metastases.Currently, no compound is approved by the FDA or the EMA for the therapeutic targeting of CDK5 and the only inhibitors designed to target CDK5 are small molecules targeting the ATP-binding pocket such as Roscovitine, as well as some peptides targeting the CDK5/p25 interface. However, the latter have severe drawbacks such as lack of specificity, development of resistance and poor pharmacokinetic properties. An alternative strategy is to develop allosteric modulators that disrupt the conformational transitions leading to CDK5 activation. Our laboratory has developed the CDKCONF5 biosensor which has allowed the identification of allosteric modulators of CDK5, including ethaverine and papaverine. These molecules were initially described as inhibitors of PDE4 and PDE10 in the neuronal system. During my thesis I characterized the molecular mechanism by which these molecules target and inhibit CDK5 functions. I first sought to understand how these inhibitors affect the interactions of CDK5 between its partners and substrates through in vitro assays and in glioblastoma-derived cell extracts. I then characterized the inhibitory potential of these compounds in human glioblastoma and lung cancer cell lines, which allowed us to show that these compounds inhibit cell proliferation and migration. Finally, we highlighted a p53-dependence on their CDK5 inhibitory potential. The identification of allosteric inhibitors of CDK5 by repositioning compounds offers attractive perspectives to propose effective therapies for lung cancer.