Cancer involves aberrant control of cellular proliferation due to activation of oncogenes and inactivation of tumor suppressors. Tumor suppressors provide an intrinsic barrier to cell growth and cancer by promoting cell death or inducing permanent growth arrest (senescence) in pre-malignant cells. RAS proto-oncogenes are often mutationally activated in cancer cells, while the p53 or RB tumor suppressor pathways are nearly universally disabled. Loss of tumor suppressor pathways renders cells susceptible to transformation by RAS and other oncogenes, since such mutations disrupt apoptotic or senescence responses to oncogenic stress. Acquiring detailed knowledge of the various oncogenic and anti-oncogenic pathways is essential for understanding how cancers develop and to identify unique vulnerabilities of tumor cells that can be targeted by novel anti-cancer strategies. Our laboratory studies the C/EBP (CCAAT/enhancer binding protein) family of transcription factors and their roles in cell proliferation and tumorigenesis. Our research focuses primarily on C/EBPbeta and its role as a downstream target of RAS signaling. Analysis of Cebpb null mice and human and rodent tumor cells have shown that C/EBPb has pro-oncogenic functions and is essential for the development of many cancers. However, in primary mouse fibroblasts (MEFs), C/EBPb is also required for oncogene-induced senescence (OIS). In senescing cells, C/EBPb arrests cellular proliferation through a pathway requiring RB:E2F. Thus, C/EBPb possesses both pro- and anti-tumorigenic activities. Because it plays a key role in cellular responses to RAS, we are studying the mechanisms by which C/EBPb expression and activity are controlled by oncogenic RAS signaling and the molecular basis for its dual functions in suppressing and promoting cancer. C/EBPb is an intrinsically repressed (auto-inhibited) protein whose activity can be stimulated by oncogenic RAS or growth factor signaling through the RAF-MEK-ERK cascade. C/EBPb forms a folded hydrophobic core that sequesters the basic region and transactivation domains, inhibiting its DNA-binding and transactivation functions. However, C/EBPb becomes activated by RAS signaling through multiple inducible post-translational modifications (PTMs). These include a site phosphorylated by ERK1/2, a RSK site in the leucine zipper that regulates C/EBPb DNA binding and homodimerization, and a recently identified CK2 site that is required for RAS-induced DNA binding. The activated form of C/EBPb exhibits increased DNA binding and homodimerization and contributes to cell cycle arrest and senescence. It also induces expression of senescence-associated secretory phenotype (SASP) genes, such as pro-inflammatory mediators, in senescent cells. An important finding from our lab was the discovery that the Cebpb 3' untranslated region (3'UTR) inhibits RAS-induced post-translational activation of the C/EBPb protein, thereby suppressing its pro-senescence activity in tumor cells. This mechanism, termed 3'UTR regulation of protein activity or UPA, suppresses the DNA-binding and transcriptional activities of C/EBPb. The 3'UTR effect was mapped to a sequence containing several G/U rich elements (GREs), and required the ARE/GRE-binding protein, HuR. These components act by directing Cebpb mRNA transcripts to the peripheral cytoplasm, excluding them from a perinuclear region where the C/EBPb kinases p-ERK1/2 and CK2 reside in tumor cells. In this location, newly-translated C/EBPb is uncoupled from RAS signaling and fails to undergo phosphorylation and activation by ERK and CK2. We infer that the intracellular site of C/EBPb translation is critical for RAS-induced activation of the protein via effector kinases such as p-ERK. Notably, 3'UTR inhibition and Cebpb mRNA compartmentalization are not observed in senescent primary mouse and human cells. Consequently, RAS-induced activation of C/EBPb is supported and OIS can be implemented to suppress tumorigenesis. We predict that analogous RNA-dependent mechanisms may control the activities of many proteins to coordinate cellular responses to RAS signaling. We are currently investigating whether other pro- and anti-oncogenic transcription factors, including p53, are regulated by UPA-like processes. We hypothesized that other proteins besides HuR may be involved in regulating Cebpb mRNA trafficking and UPA. Therefore, we have used a proteomics approach to identify cellular proteins that specifically associate with the GRE RNA sequence. One candidate protein identified from these studies was Upf1, an RNA helicase that plays a critical role in nonsense-mediated mRNA decay (NMD). NMD eliminates faulty transcripts that contain premature stop codons. However, Upf1 also regulates degradation of 10% of normal mRNAs. Interestingly, we observed that Upf1 is localized to the perinuclear region of tumor cells, corresponding to the site of Cebpb mRNA exclusion. Depletion of Upf1 in human lung adenocarcinoma cells disrupted peripheral localization of Cebpb transcripts and increased their accumulation in the perinuclear cytoplasm. Accordingly, the cells became senescent and expressed SASP genes such as IL-6 and IL-1 in a C/EBPb-dependent manner. We also identified the RNA-binding protein, staufen, as a GRE interactor. Stau1 and Stau2 have been linked to mRNA localization and decay, and Staufen-mediated mRNA decay (SMD) is known to require Upf1, functionally linking these candidate UPA factors. Collectively, our data indicate that perinuclear Upf1 and staufen, together with cytoplasmic HuR, promote perinuclear Cebpb mRNA degradation to prevent C/EBPb activation and senescence in tumor cells. In cells expressing oncogenic RAS or BRAF, the C/EBPb kinases p-ERK and CK2 become re-localized to nuclear-proximal structures we call perinuclear signaling complexes or PSCs. PSCs are associated with endosomes and require the MAPK scaffold KSR1 (kinase suppressor of Ras 1) for their formation. KSR1 also undergoes RAS-induced targeting to PSCs and colocalizes with p-ERK and CK2. Thus, besides its ability to facilitate RAF-MEK-ERK signaling, KSR1 plays a key role in regulating subcellular localization of RAS effector kinases. We have detected PSCS in several kinds of human tumor cells and in KRAS-induced mouse lung tumors, suggesting that these localized complexes are a ubiquitous feature of oncogenic signaling. We found that PSCs are also transiently induced by serum growth factors in normal cells with delayed kinetics (4-6 hr after GF stimulation). We propose that mutant RAS constitutively activates this late phase of GF signaling such that effector kinases become localized to a perinuclear compartment and access key substrates that drive cell proliferation and tumorigenesis. Our current studies are aimed at elucidating the molecular basis for RAS-induced formation of PSCs, focusing on the role of signaling adaptor proteins associated with specific classes of perinuclear endosomes. In the future, PSC components may prove to be effective targets for cancer therapy, as well as providing novel biomarkers to identify tumor cells in tissue samples. In addition, PSCs could be used to monitor tumor responses to anti-cancer drugs that inhibit the RAS-ERK pathway.
Showing the most recent 10 out of 19 publications
