UCSF

The complexity of the three-dimensional structures formed by RNA is essential for its function and as a result, a large number of protein co-factors are required to maintain RNA homeostasis. The largest family of enzymatic RNA chaperones are the DEAD-box proteins, which utilize ATP hydrolysis to modify RNA substrates. DEAD-box proteins are required for all stages of RNA metabolism and are implicated in diseases including cancer, viral pathogenesis, and developmental delay. Yet despite the biochemical and structural characterization of these enzymes over the past three decades, the specific functions and substrates of DEAD-box proteins remains poorly understood. Chemical inhibition would be an excellent tool for the elucidation of DEAD-box protein biology because of its rapid onset of inhibition, however specific small molecule inhibitors do not exist for most DEAD-box proteins.

To develop a generalizable strategy for the inhibition of DEAD-box proteins, we developed two chemical genetic strategies – one analog-sensitive and one electrophile-sensitive – that rely on genetic perturbations of the target of interest coupled with specific chemical inhibitors. Our efforts to develop an analog-sensitive strategy yielded novel space-creating mutations off the N6-position of the adenine of ATP that were poorly tolerated in vivo, and analog-sensitive inhibitors that could not be optimized to bind with high affinities. For the electrophile-sensitive strategy, we identified a residue of low conservation within the DEAD-box protein active site for mutation to cysteine and specifically targeted it with electrophile-containing AMP-derivatives. These small molecules specifically bind to and inhibit electrophile-sensitive, but not wild-type, DEAD-box proteins. Together, these results indicate that DEAD-box proteins can be targeted using chemical genetic tools and provide a path towards the development of specific small molecule inhibitors of any member of the DEAD-box protein family.

We additionally sought to identify novel nucleotide-competitive cell-permeable small molecules that bind and inhibit DEAD-box proteins. To this end, we identified a series of di-substituted quinazolines based on the AAA+ ATPase inhibitor DBeQ that dose-dependently inhibit DDX3 in an ATP-dependent manner.

In a separate project, we explored the limited efficacy of clinical CDK4/6 inhibitors as monotherapy. We identified that the small molecule inhibitor ribociclib has limited efficacy as a monotherapy in cell culture, and found that CDK4 is bound in high molecular weight complexes in cells, which are mostly resistant to inhibition by type I kinase inhibitors. We then developed a series of small molecule derivatives of ribociclib that attempt to expand the clinical efficacy of ribociclib by targeting these high molecular weight complexes.