Fbw7 is the substrate recognition component of an SCF-ubiquitin ligase that targets a remarkable group of proteins for degradation by the proteasome. Many Fbw7 substrates are critical oncoproteins, and Fbw7 is one of the most frequently mutated human tumor suppressor genes. Our lab has made numerous key contributions to understanding this complex pathway. One area involves our discovery of new substrates, including Myc, Notch, cyclin E, and Mediator, and their mechanisms of degradation. Another area has entailed the use of physiologic human and mouse gene targeting models to understand the normal and neoplastic functions of the various Fbw7 isoforms. Recently, we described the fundamental importance of Fbw7 dimerization to its substrate interactions, and how dimerization may explain Fbw7’s unusual mutational spectrum in cancer and lead to novel therapeutic strategies. Finally, we have developed new mouse models of Fbw7-associated cancer, including a new model of metastatic and chromosomally unstable colon cancer.
The cell cycle field exploded during my postdoc and early independent career, during which time a basic framework of cell division emerged, including cyclin E’s central role in G0 to S-phase progression and tumorigenesis. We contributed substantially to this field and established new paradigms in cyclin E regulation and function. One area is cyclin E-degradation by the ubiquitin proteasome system, wherein we first made the surprising discovery that cyclin E autophosphorylation triggers its own ubiquitylation and destruction. We subsequently detailed the mechanisms through which cyclin E phosphorylations regulate its ubiquitylation by Fbw7 and establish its normal periodicity during the cell cycle. Another surprising discovery was our finding that cyclin E has a CDK-independent function in licensing replication origins, which was the first description of a CDK-independent cyclin function. We have also described the genomic instability caused by cyclin E deregulation in cancer and a homeostatic response that protects cells against this genomic instability. Finally, we have developed mouse models with targeted mutations of cyclin E phosphorylation sites to demonstrate the roles of phosphorylation-dependent cyclin E control in differentiation, genome stability, cell division, and tumorigenesis.
We have made many contributions to this field that have had a sustained impact. The first was the finding that cyclin E-CDK2 inhibited p27 by targeting it for proteolysis. The idea that a CDK inhibitor could be antagonized by the very kinase it inhibits was completely unexpected. Nonetheless, these data are now widely accepted and set the stage for many future studies of CDK inhibitor regulation. Another example is our work on the ubiquitin-independent degradation of the p21 CDK inhibitor by the proteasome. Although p21 is ubiquitylated, we found that its proteasomal degradation does not require ubiquitylation. Although controversial at the time, these data also established a new mechanistic paradigm which we subsequently fully characterized. We have also studied normal and neoplastic p27 functions models in novel mouse models, including a high throughput insertional mutagenesis screen, which revealed p27-collaborating oncogenes.
One continuing aspect of our work on CDK2 has focused on identifying CDK2 substrates, which began with our studies of cyclin E autophosphorylation. We subsequently developed novel proteomic methods using substrate thiophosphorylation and ATP analog substrate-sensitive CDKs, to discover more than a hundred new CDK2 substrates. We have now adapted this method to highly physiologic contexts, which has revealed even more new substrates, including the discovery that CDK2 regulates a host of chromatin-modifying enzymes to coordinate cell division and gene expression. In light of new therapies targeting CDK4/6, we have now adapted these methods to identify new CDK4/6 substrates. We have more recently focused on the importance of CDK2 inhibitory phosphorylation by Wee1, and made a human knockin cell line to show that CDK2 inhibitory phosphorylation has essential roles in cell cycle control, in DNA replication dynamics, and in genome instability. Most importantly, we found that the inability to inhibit CDK2 during stalled S-phase leads to massive DNA damage, which has formed the basis for an exciting new chemotherapy strategy.