Control of Oncogenic Signaling Through Spatial Organization of Kinases and mRNAs

Division Of Basic Sciences - Nci

Investigators

Linked publications & trials

Abstract

We have continued our studies on 3'UTR regulation of C/EBPbeta activity in tumor cells. We have verified our observations that CEBPB mRNA decay at or near perinuclear signaling endosomes prevents C/EBPbeta phosphorylation/activation in tumor cells. This mechanism involves the RNA decay proteins UPF1 and Staufen (STAU1/2). STAU is a sequence/structure-specific dsRNA binding protein and we have identified a potential STAU binding site (SBS) in the CEBPB 3'UTR. We have used deletion mutants expressed in cells to show that STAU2 binds to this SBS motif. We also analyzed mRNA localization to demonstrate that SBS mutant mRNAs accumulate in the kinase-rich perinuclear region. This coincides with C/EBPbeta phosphorylation on an activating CK2 site, leading to increased senescence in the host cells. We further showed that a mouse mutant lacking part of the SBS and an adjacent G/U-rich region (GRE) that binds HuR are resistant to Kras-driven lung adenocarcinomas. Immortalized mouse embryo fibroblasts (MEFs) derived from the delGRE mice undergo oncogenic RAS-induced senescence rather than neoplastic transformation. Nevertheless, these cells do not express senescence-associated secretory phenotype (SASP genes) characteristic of senescent cells. However, the delGRE cells up-regulate the pro-senescent cytokine S100a9 in response to oncogenic RAS, and S100a9 is required for their precocious senescence. Thus, one key function of the CEBPB 3'UTR is to suppress expression of S100a9 and thus prevent senescence of tumor cells. Another research focus is to characterize perinuclear ER-tethered signaling endosomes that carry oncogenic kinases. These perinuclear signaling centers (PSCs) are observed in all cancer cells and tumor tissues examined but not in normal cells, suggesting that PSCs function as key oncogenic signaling platforms. Oncogenic RAS induces perinuclear translocation of the effector kinases ERK1/2 and CK2 and their signaling scaffold, KSR1. PSCs containing these proteins occur on endosomes embedded within the perinuclear ER network. We showed that an endosomal adapter, TOLLIP, is required for perinuclear localization of RAB11A+ endosomes harboring CK2 and KSR1 but not ERK. Accordingly, we found that ERK is present on a distinct endosomal population. TOLLIP binds to KSR1 through conserved domains on each protein, recruiting signaling complexes to endosomes. Immunoprecipitation experiments with endogenous proteins confirmed that TOLLIP forms a complex with KSR1 and CK2 (and possibly RAS) in tumor cells. Phospho-proteomic studies revealed that TOLLIP-dependent perinuclear CK2 phosphorylates distinct substrates, including proteins involved translation and ribosome biogenesis. Thus, perinuclear CK2 plays an important role in generating onco-phosphoproteomes in tumor cells. Cancer cell lines carrying KRAS or NRAS mutations, but not mutant HRAS, BRAF, PTEN or other oncogenes, require TOLLIP for proliferation/survival. Accordingly, KRasG12V-driven lung tumors in Tollip deficient mice displayed reduced progression to malignant adenocarcinomas. TOLLIP is therefore a key signaling adaptor selectively required by K/NRAS tumor cells and whose inhibition is a potential liability of these cancers. We are currently engaged in drug discovery studies using TOLLIP as a target for small molecule inhibitors. Although cancer cells carrying oncogenes other than K/NRAS are TOLLIP independent, they nevertheless display perinuclear CK2. These findings suggest that another adapter provides a redundant function in other types of tumor cells. Moreover, ERK does not co-localize with TOLLIP and remains perinuclear in its absence. Initial studies indicate that ERK2 associates with another endosomal adapter that anchors ERK-bearing endosomes to the ER. Future studies will focus on identifying these additional adapters for ERK, CK2 and other kinases, as these may represent dependency genes for specific cancers and could be targets for novel anti-cancer drugs.