The human p53 protein is a homotetrameric, sequence-specific transcription factor. Approximately 30 different sites on p53 can be post-translationally modified by phosphorylation, acetylation, methylation, ubiquitinylation, neddylation or sumoylation. Post-translational modifications (PTMs) of p53 can affect its stability, its activity as a transcription factor and its interactions with specific proteins. Our research has focused on identifying and exploring the biological roles of these post-translational modifications on p53 to better understand how they modulate p53 function. Lysine methylation has recently been observed to occur on non-histone proteins, such as p53, suggesting broad cellular roles for the enzymes generating and removing methyl moieties. We are utilizing a combination of biochemical, molecular and proteomic strategies to identify and characterize novel methylated species of p53, as well as to identify the methyltransferase enzymes (HMTs) that generate these modified forms of p53. In work published this year, we have reported a novel lysine methylation event occurring on p53 and identified the HMT SET8/PR-Set7 as a new regulator of p53. Specifically, we found that SET8 monomethylates p53 on Lys382. This methylation event robustly suppresses p53-mediated transcription activation of highly responsive target genes, but has little influence on weak targets. The increase of acetylation at p53 Lys382 and phosphorylation at Ser20 that normally accompany DNA damage are impeded by SET8, suggesting functional crosstalk between this methylated form and other discrete p53 post-translational modifications. Finally, SET8 attenuates the pro-apoptotic and checkpoint activation functions of p53. Based on our data, we propose a model in which SET8-mediated p53 methylation tips the balance of p53 function away from cell elimination towards cell survival. Mouse models containing missense mutations at sites of post-translational modification of p53, developed as part of a collaboration, continue to be a valuable resource for investigating the complex effects of single or multiple PTMs of p53 in a physiological setting. We have analyzed knock-in mutants of two N-terminal phosphorylation sites, Ser18 and Ser23, which result in broadly reduced p53 activity. We used expression arrays to assess changes in mRNA levels in thymocytes following exposure to IR in mice with wild-type p53, p53S18A, p53S18AS23A, or p53-/-. Most p53 target genes that were induced in response to IR by wild-type p53 showed a reduced induction in the p53S18S and p53S18A/S23A samples. Although several pro-apoptotic genes were only moderately affected by the mutations, including Noxa, Puma and Bax, the induction of Lrdd, a key effector of p53-dependent apoptosis in thymocytes, was severely compromised by the p53S18AS23A mutation. As a class, genes repressed in response to IR by wild-type p53 exhibited a wider range in the effect of the mutations. For about half of the repressed genes, such as Chek1 and Cdc25a, the p53S18A/S23A mutation abolished the repression. Other genes showed a reduced repression in the mutant samples. A few p53 target genes, including Myc, showed nearly equal repression in wild-type and the p53S18A/S23A samples. These results illustrate the crucial involvement of PTMs of p53 in modulating the function of p53 as a regulator of transcription. During a screen for p53-induced genes following IR, we identified a wild type p53 induced phosphatase, Wip1. Wip1 is similar in sequence to the type 2C protein phosphatases, which are associated with stress response, sexual differentiation and cell cycle control in a range of organisms. The gene encoding human Wip1, PPM1D, is amplified in a significant portion of primary human breast tumors, most of which harbor wild type p53. Thus, Wip1 may act as an oncogene by functionally inactivating p53. The first target of Wip1 phosphatase activity to be identified was p38 mitogen-activated protein kinase (MAPK), which it specifically dephosphorylates and inactivates in the nucleus. Experiments with cultured cells indicated that either the absence of p38 MAPK or its inactivation by Wip1 overexpression facilitates tumor formation by oncogenic Ras. Recently, we have used the ApcMin/+ mouse model that has been useful in colon cancer studies to investigate the role of Wip1 in initiation and progression of intestinal cancer, as we previously found deletion of Wip1 to delay the onset of lymphomas. Using real-time PCR analysis of Wip1 mRNA in polyps and normal intestinal epithelium, we found that the level of Wip1 mRNA in polyps was increased five-fold. We found that Wip1-/-/ApcMin/+ mice survived longer than wt/ApcMin/+ mice and developed significantly fewer and smaller polyps compared to wt/ApcMin/+ polyps. Moreover, Wip1-/-/APCMin/+ polyps were primarily localized in the small intestine. These results show that deficiency of Wip1 dramatically suppresses polyp formation in the presence of ApcMin mutation, supporting the role of Wip1 as a critical factor in the regulation of ApcMin-driven polyposis. In addition, the data provides evidence that modulation of apoptosis of stem cells and suppression of tumorigenesis in theAPCmin model of intestinal cancer can be achieved through a regulation of the Wip1/p53-dependent signaling pathway. We previously reported a substrate motif for Wip1 that was defined using variants of a p38-based (180pT 182pY) diphosphorylated peptide. However, as the sequences surrounding the targeted residue and p38 and in ATM, a second Wip1 substrate, appear to be unrelated, we continued studies to more fully define the substrate recognition motifs for Wip1. Using a recombinant human Wip1 catalytic domain, we measured the kinetic parameters for variants of an ATM-based (1981pS) phosphopeptide. We found that Wip1 dephosphorylates phosphoserine and phosphothreonine in the p(S/T)Q motif, which is an essential requirement for substrate recognition. In addition, acidic, hydrophobic, or aromatic amino acids surrounding the p(S/T)Q sequence have a positive influence, while basic amino acids have a negative influence on substrate dephosphorylation. Our kinetics results allow discrimination between true substrates and non-substrates of Wip1, and we identified several new putative substrates that include HDM2, SMC1A, ATR and Wip1 itself. Site-directed mutagenesis analysis, combined with a three-dimensional molecular model of Wip1 with a bound substrate peptide that we developed, suggested that the important residues for ATM(1981pS) substrate recognition are similar, but not identical, to those for the p38(180pT 182pY) substrate. Results from this study will be useful for predicting new physiological substrates that may be regulated by Wip1. Recently, we identified a cyclic phosphopeptide inhibitor for the catalytic site of Wip1 based on its substrate specificity that showed > 12-fold selectivity toward rWip1 over PP2Cα, which has a catalytic domain that is 50% similar to that of Wip1. However, these molecules are not viable drug candidates due to their size and metabolic instability. Thus, using a pyrrole based scaffold, we have developed a series of small molecules that mimic the three-dimensional arrangement of the polar and hydrophobic functional groups of the best cyclic-peptide inhibitor. Iterative optimization cycles of design, sy [summary truncated at 7800 characters]
Showing the most recent 10 out of 31 publications
