Divergent Functions of ERK2 Substrate Binding Domains in Pathogenesis of KRAS-Driven Pancreatic Cancer

PROJECT SUMMARY/ABSTRACT The central KRAS-ERK signaling axis is activated in most human pancreatic adenocarcinomas (PDAC). Attempts to target ERK1 and ERK2 signaling with non-selective kinase inhibitors have produced only limited efficacy in the clinic at the cost of dose-limiting toxicity. Our breakthrough discovery for this proposal is that

ERK2, and not ERK1, is an essential driver of pancreatic carcinogenesis but is dispensable in non-transformed pancreatic epithelial cells. Based on this, ERK2 represents a new potent target for therapeutic intervention in PDAC. This proposal is built on the results of our recent studies of genetic models of myeloid neoplasms in which

we have established ERK2 as a central oncogenic effector via its unique substrate interaction domains. In our ERK2 knock-in mouse genetic models, the D domain of ERK2 promotes cancer progression, while the ERK2 DBP domain opposes progression and induces senescence in oncogene-transformed cells. However, the role

of the ERK2-D and DBP domains in driving KRAS-initiated PDAC tumorigenesis has not been investigated. Hence, our central hypothesis is that pancreatic carcinogenesis driven by mutant KRAS is dependent on domain- specific functions of ERK2. We propose the following aims: Aim 1. To assess the opposing functions of the ERK2-D and DBP domains in pancreatic carcinogenesis. Our

preliminary results indicate that conditional deletion of ERK2 (Mapk1 gene) in the KPC model dramatically impedes PDAC development. We will intercross KPC mice in which the wild type Mapk1 gene is replaced with the D- and DBP-domain mutants (knock-in models already available in our lab) to pinpoint which of substrate

binding pockets is critical for pancreatic cancer development and progression. Aim 2. Determine the mechanistic basis by which the ERK2-D and DBP domains exert their distinct functions. We will generate in vitro models of ERK2 deletion and replacement with D- and DBP-domains mutations by RNAi and CRISPR-targeting of ERK2 in PDAC cell lines and assess the effects on cell growth, and MAPK pathway

signaling. These models will be essential discovery tools for future proteomic and transcriptome analyses to pinpoint the ERK2 domains interactors that are essential for exerting the KRAS malignant phenotype and will allow us to decipher the molecular basis for the opposing functions of ERK2 D and DBP domains.

Through these efforts, we expect to bring new insights into the role of ERK2 substrate binding modules in regulating pancreatic cancer progression, and how to exploit this information therapeutically.