Targeting allosteric scaffolding functions of Aurora kinase A in cancer

ABSTRACT Neuroblastoma is the most common solid tumor in infants. About 25% of patients have high-risk neuroblastoma, a devastating disease with poor prognosis and few treatment options. The primary driver of high-risk neuroblastoma is the oncogene MYCN, a MYC-family transcription factor that has no druggable

pockets and has long eluded drug development efforts. Recently, the protein kinase Aurora A (AurA) was shown to bind to the N-Myc protein in neuroblastoma cells and interfere with its ubiquitination by the SCF ubiquitin ligase complex, preventing N-Myc from being degraded by the proteosome. Blocking complex formation between AurA and N-Myc results in rapid N-Myc

degradation and cell death in neuroblastoma cell lines. The same AurA/N-Myc complex has now been shown to drive neuroendocrine prostate cancer (NEPC), and AurA also forms a similar complex with the closely- related c-Myc protein in liver cancer. These recent discoveries point to a new paradigm for targeting Myc-

family transcription factors in cancer using inhibitors that trigger structural changes in AurA that block Myc protein binding and promote Myc degradation. Our lab has recently shown that most existing AurA inhibitors, including the current clinical candidate alisertib, do not have a strong enough allosteric effect on AurA to be effective at weakening N-Myc binding. In

agreement with this, alisertib has inconsistent effects on N-Myc levels in cell lines, and has not performed well in ongoing clinical trials in neuroblastoma and NEPC. The weakness in our current understanding of how AurA binds to c-Myc and N-Myc and how these interactions are affected by inhibitors represents a major impediment

to this therapeutic strategy for targeting Myc-driven cancers. The goal of this project is to provide the missing molecular picture of the interactions between AurA and Myc transcription factors and how they can be modulated by inhibitor binding. We plan to use new experimental tools and approaches to define how the binding of c-Myc and N-Myc alters the conformation

(shape) and dynamics (protein motion) of AurA, and to delineate the specific structural changes an inhibitor must trigger to efficiently destabilize these complexes. We will a) define the structure of the AurA/Myc complexes at atomic resolution using x-ray crystallography, magnetic resonance spectroscopies and molecular

modeling, b) determine how these interactions alter AurA conformation and dynamics by tracking key structural elements of the kinase in solution, c) correlate the effects of a large panel of kinase inhibitors on AurA conformation with their ability to alter the binding affinities of N-Myc and c-Myc, and d) test the efficiency of the

strongest AurA allosteric modulators in a series of N-Myc- and c-Myc-dependent cancer cell lines including neuroblastoma, NEPC and liver cancer cells. The insights will pave the way for the repurposing of existing kinase inhibitors and the development of new inhibitors as a new treatment modality for Myc-driven cancers.