BIOCHEMISTRY IDENTIFICATION AND REGULATION OF P53 TARGET GENES IN PRIMARY HUMAN EPIDERMAL KERATINOCYTES KRISTY L. SCHAVOLT Dissertation under the direction of Professor Jennifer A. Pietenpol The p53 tumor suppressor gene is mutated in half of all human cancers. p53 mediates the cellular response to a variety of cell stress signals through transcriptional regulation of target genes that play roles in cell cycle arrest, DNA repair, and apoptosis. Over one hundred target genes are directly regulated by p53, though mechanisms regarding p53 regulation of gene transcription are not fully understood. To gain further insight to the processes of p53 target gene regulation, the dissertation research described herein resulted in the identification of novel targets of p53, as well as analysis of factors that dictate the ability of p53 to selectively regulate target genes in response to genotoxic stress. The role of p53 family members p63 and p73 in p53-mediated signaling pathways was investigated as well. A primary human keratinocyte model system was utilized to avoid genetic and epigenetic alterations found in established cell lines. To identify novel target genes, fragments of p53-bound DNA isolated by chromatin immunoprecipitation (ChIP) were functionally tested for p53 regulation using a yeast screen. To complement the ChIP/yeast screen, microarray analysis was performed to identify changes in gene expression upon ectopic expression of p53 and deltaNp63alpha. In these studies, Ras-related associated with diabetes (RRAD), modulator of apoptosis 1 (MOAP-1), and zinc finger protein 90 (ZFP90) were identified as potential p53 transcriptional targets. Finally, timing of transcription factor binding to target gene regulatory regions was examined using ChIP. At target gene consensus binding sites, p53 occupancy increased and concomitantly, deltaNp63alpha occupancy decreased. Correspondingly, expression of a panel of known p53 target genes was increased by p53 and decreased by deltaNp63alpha in microarray studies. Additionally, neither the presence of constitutively bound p53 nor deltaNp63alpha dictates the constitutive binding of RNA polymerase II (pol II) to target gene proximal promoters. Rather, the location of the consensus binding site (promoter or intron) dictates the constitutive binding of pol II. This research allows for a more complete understanding of p53 transcriptional regulation of target genes at the levels of promoter occupancy and gene expression, as well as a potential role of deltaNp63alpha in p53-regulated signaling. Understanding mechanisms of p53 target gene regulation is essential for comprehending how alterations in p53 signaling result in increased tumorigenesis. Approved___Jennifer Pietenpol________ Date_____11.30.06______

The Receptor for Advanced Glycation End Products (RAGE) contains a single transmembrane helix, a small cytosolic domain, and an extracellular region (sRAGE) composed of three Ig-like domains (V, C1, C2). RAGE has been implicated in complications arising from diabetes and chronic inflammation and is widely thought to be a therapeutic target. In this dissertation, production of purified sRAGE and the five single and tandem domain constructs enabled biophysical and structural characterization. The results, including an x-ray crystal structure of VC1, show that the V and C1 domains form an integrated structural unit. In contrast, C2 is attached to VC1 by a flexible linker and is fully independent. Studies of the interaction with a known RAGE ligand, Ca2+-S100B, revealed a major contribution from the V domain but clearly defined contributions to binding from the C1 domain. The implications of these results are discussed with respect to models for sRAGE quaternary structure and RAGE signaling.

The ErbB family of receptor tyrosine kinases regulates cell growth, differentiation, and tumorigenesis. This dissertation contains two independent studies of ErbB-1 and ErbB-4. In one study, the intracellular juxtamembrane region of ErbB-1 was investigated for its role in kinase activation. By creating deletions and mutations in the juxtamembrane region of ErbB-1, the data defined an essential requirement for the juxtamembrane region in the allosteric mechanism of tyrosine kinase activation. The second study focused on proteolytic processing of ErbB-4. Activation of ErbB-4 results in ectodomain and transmembrane domain cleavage events, but the parameters that must be fulfilled to initiate ectodomain proteolysis remain poorly defined. The localization of ErbB-4 and its sheddase to membrane microdomains was evaluated as a regulatory step in sequential proteolytic processing.

Multiple exocyclic DNA adducts arise from reactions of lipid and DNA peroxidation products with DNA bases. These endogenous lesions are mutagenic; therefore, monitoring adduct levels may provide a means of assessing genomic exposure to oxidative damage in human populations. The metabolic processing of endogenously formed exocyclic DNA adducts has now been investigated for the purpose of developing non-invasive markers of oxidative damage. The pyrimidopurinone deoxynucleoside adduct, M1dG, was found to undergo enzymatic oxidation to produce 6-oxo-M1dG. The corresponding base adduct, M1G, was subject to sequential oxidation producing first 6-oxo-M1G and then 2,6-dioxo-M1G. The enzymes xanthine oxidoreductase (XOR) and aldehyde oxidase (AO) catalyzed the oxidation of M1dG and M1G. Additionally, the unsubstitued etheno base adduct, 1,N2-etheno-Gua, and the substituted etheno base adduct, heptanone-1,N2-etheno-Gua, were oxidized to produce 2-oxo-etheno-Gua and 2-oxo-hepatanone-etheno-Gua, respectively; XOR catalyzed these oxidations. The deoxynucleoside adducts 1,N2-etheno-dG and 1,N6-etheno-dA were substrates for phosphorolysis to base adducts (1,N2-etheno-Gua and 1,N6-etheno-Ade) by purine nucleoside phosphorolase (PNP). A body of evidence now exists that demonstrates exocyclic DNA adducts are subject to enzymatic oxidation and phosphorolysis reactions (sometimes in tandem). Exocyclic base adducts, with planar and unsaturated exocyclic rings, are the best substrates for enzymatic oxidation. Base adducts containing substitutions such as deoxyribose on the imidazole ring of M1dG and a heptanone chain on the etheno ring of heptanone-1,N2-etheno-Gua do not prohibit oxidative metabolism. The enzymes of purine catabolism (XOR, AO, and PNP) are also involved in the metabolic processing of exocyclic adducts. Based on in vivo metabolism studies at near-physiological concentrations of exogenously administered M1dG, the metabolic processing of exocyclic adducts is likely to occur at their physiological rate of production in animals and humans. In total, these results demonstrate that exocyclic adducts are subject to metabolic processing (oxidation and phosphorolysis). Furthermore, metabolites of endogenously produced DNA adducts are likely to be produced in vivo, and may provide a new class of biomarkers to assess exposure to endogenous sources DNA damage.

Histone deacetylase 3 (HDAC3) is the enzymatic component of transcriptional repression complexes recruited by the nuclear hormone receptors. Inactivation of HDAC3 in cancer cell lines triggered apoptosis, and removal of Hdac3 in the germ-line of mice caused embryonic lethality. HDAC3 binding sites within the hematopoietic transcription factor RUNX1 were required for RUNX1-mediated gene repression. Therefore, Hdac3 was conditionally deleted in the postnatal mouse hematopoietic system (Mx:Hdac3). The Mx:Hdac3 mice had increased numbers of double positive (c-Kit+/Sca+) hematopoietic stem cells, yet these progenitor cells not able to survive or proliferate in a colony forming assay. Mx:Hdac3 mice also developed reversible liver hypertrophy, so Hdac3 was constitutively deleted in the liver (Alb:Hdac3). The Alb:Hdac3 mice also developed hepatomegaly, which was the result of hepatocyte hypertrophy. These morphological changes coincided with significant imbalances between carbohydrate and lipid metabolism. Liver-specific deletion of Hdac3 triggered changes in gene expression consistent with inactivation of repression mediated by nuclear hormone receptors. Loss of Hdac3 also increased the levels of Pparã2, and treatment of these mice with a PPARã antagonist partially reversed the lipid accumulation in the liver. In addition, gene expression analysis identified mammalian target of rapamycin (mTOR) signaling as being activated after deletion of Hdac3, and inhibition by rapamycin affected the accumulation of neutral lipids in the Hdac3-null livers. As the Alb:Hdac3 mice aged, accumulation of endogenous DNA damage and disrupted metabolism most likely contributed to the development of non-alcoholic steatohepatitis (NASH)-like symptoms, and ultimately, hepatocellular carcinoma. Thus, Hdac3 plays a critical role in mediating normal cellular homeostasis in both the adult hematopoietic system and liver, as demonstrated by the Mx:Hdac3 and Alb:Hdac3 conditional deletion models.

The replication of DNA occurs through a complex series of steps involving the coordinated action of many proteins. DNA polymerase alpha/primase (pol-prim) is a critical DNA replication factor that synthesizes short RNA-DNA primers on the leading and lagging strands of DNA being actively replicated. Pol-prim contains a DNA polymerase subunit (p180), an accessory subunit (p68), and two DNA primase subunits (p58 and p48). This dissertation describes the biochemical and structural characterization of two folded, globular pol-prim domains: p58C and p68N. Sub-cloning, bacterial expression, purification, and analysis of p58C revealed an essential iron-sulfur cluster, the first such cofactor identified in a DNA replication protein. Sub-cloning, bacterial expression, and NMR analysis enabled the determination of the solution structure of p68N. The binding of p68N to the SV40 large T antigen helicase domain was measured by isothermal titration calorimetry and the structure was used to model and test the molecular basis for this interaction. These results enabled an updated model to be generated for the action of pol-prim in SV40 DNA replication.

The cyclooxygenase enzymes (COX-1 and COX-2) catalyze the conversion arachidonic acid (AA) to prostaglandin H2 (PGH2), which is the precursor to biologically active prostanoids. The primary mechanism of action of non-steroidal anti-inflammatory drugs (NSAIDs) is the inhibition of prostaglandin biosynthesis by binding within the active site of the COX enzymes. Naproxen, a non-selective NSAID, has been marketed as analgesic and anti-inflammatory agent for over thirty years. In the work presented herein, the structure and dynamics of naproxen binding to COX is elucidated. We determined a 1.7 Å crystal structure of naproxen complexed to COX-2, which indicates that naproxen is bound similarly to other arylpropionic acid inhibitors with the carboxylate moiety making critical interactions at the base of the active site. Interestingly, we identified a novel interaction at the top of the active site between Trp-387 and the p-methoxy moiety of naproxen. Each of the major functional groups of naproxen is required for inhibitory activity suggesting that the development of more potent and/or COX-2 preferring naproxen analogs may be difficult. However, we have synthesized derivatives of naproxen with single atom modifications at the 6-position resulting in COX-2-preferring inhibitors. The crystal structure of one of these analogs, p-methylthio naproxen, bound to COX-2 suggests that the analogs occupy relatively the same conformation as naproxen within the active site. COX-2 has the ability to metabolize alternative fatty acid substrates in addition to AA including the endocannabinoids, 2-arachidonoylglycerol (2-AG) and anandamide (AEA). Previous studies have shown that ibuprofen and mefenamic acid are weak, competitive inhibitors of COX-2 mediated AA metabolism, but potent, non-competitive inhibitors of 2-AG oxygenation; this phenomenon was dubbed substrate-selective inhibition. In the present work, we demonstrate that a series of reversible inhibitors are significantly more potent inhibitors of 2-AG oxygenation compared to AA whereas a series of tight-binding inhibitors block the oxygenation of both substrates by COX-2 with comparable potency. Furthermore, (R)-arylpropionates, which were previously thought to lack COX inhibitory activity, are potent inhibitors of COX-2-mediated 2-AG oxygenation. A highly substrate-selective inhibitor may represent a novel analgesic agent that lacks the deleterious side effects associated with the use of traditional NSAIDs.

This dissertation describes the crystal structures of two distinct metal-binding proteins: Escherichia coli Iron-sulfur cluster protein A and the carboxyltransferase subunit of the acetyl-coA carboxylase enzymes from Staphylococcus aureus and Escherichia coli. Iron-sulfur cluster protein A (IscA) belongs to an ancient family of proteins responsible for iron-sulfur cluster assembly in essential metabolic pathways preserved throughout evolution. The crystal structure of Escherichia coli IscA reveals a novel fold in which mixed beta-sheets form a compact alpha-beta sandwich domain. In contrast to the highly mobile secondary structural elements within the bacterial Fe-S scaffold protein IscU, a protein which is thought to have a similar function, the great majority of the amino acids which are conserved in IscA homologues are located in elements which constitute a well-ordered fold. However, the 10-residue C-terminal tail segment which contains two invariant cysteines critical for the Fe-S binding function of IscA is not ordered. In addition, the crystal packing reveals a helical assembly which is constructed from two possible tetrameric oligomers of IscA. The rates of severe, multi-drug resistant bacterial infections, including those caused by pathogens previously confined to the hospital setting, have increased dramatically in both hospital and community populations. Acetyl-coA carboxylase is a central metabolic enzyme that catalyzes the committed step in fatty acid biosynthesis: biotin-dependent conversion of acetyl-coA to malonyl-coA. This work presents the structures of the bacterial carboxyltransferase subunits from two prevalent nosocomial pathogens, Staphylococcus aureus and Escherichia coli. Both structures reveal a small, independent zinc-binding domain that appears to shield the active site during the catalytic process. The zinc domain of bacterial carboxyltransferase, which lacks a complement in the primary sequence or structure of the eukaryotic homologue, is a feature that yields promise for the structure-based design and development of new, selective antimicrobial classes.

The work presented in this dissertation examines the DNA cleavage activity of human topoisomerase II. Type II topoisomerases are required for removing DNA knots and tangles and are important chemotherapy drug targets. The ability to cleave DNA is critical to the cellular and pharmacological functions of human type II topoisomerases. To study the mechanism of DNA cleavage, we developed a system that isolates topoisomerase II-mediated DNA scission from ligation using substrates containing a 3-bridging phosphorothiolate at the scissile bond. The characteristics of topoisomerase IIá-mediated cleavage of phosphorothiolate oligonucleotides were identical to those seen with wild-type substrates, except that no ligation was observed. The unidirectional accumulation of cleavage complexes enabled us to examine the effects of topoisomerase II poisons on the rate of DNA cleavage and will be valuable for future studies on DNA scission and the topoisomerase II-DNA cleavage complex. Topoisomerase II modulates DNA topology by generating double-stranded breaks in DNA. However, it is unclear how the enzyme coordinates its protomer halves. Results demonstrate that a nick at one scissile bond dramatically increases the rate of cleavage by human topoisomerase IIá at the opposite-strand scissile bond. We propose that this enhanced second strand activity coordinates the two topoisomerase II active sites, allowing the enzyme to create double-stranded breaks. We also found that nicks poison topoisomerase II and generate new cleavage sites. Finally, the role of metal ions in the DNA cleavage reactions of human topoisomerase IIá and IIâ were characterized. Results demonstrate that both enzymes employ a two-metal-ion mechanism for DNA cleavage. Also, an interaction between one divalent metal ion and the 3-bridging atom of the scissile phosphate greatly enhances enzyme-mediated DNA cleavage, most likely by stabilizing the leaving 3-oxygen. Further, there is an important interaction between a second metal ion and a non-bridging atom of the scissile phosphate that stimulates DNA cleavage mediated by topoisomerase IIâ. Although this non-bridging interaction frequently is postulated in models for enzyme-mediated DNA cleavage, our evidence with topoisomerase IIâ is the first biochemical demonstration that a metal ion contacts this position during scission.

Non-enzymatic modification by glucose of proteins induces protein damage through the formation of advanced glycation end products (AGEs). AGEs have been implicated in the pathogenesis of a number of diseases including diabetes. Pyridoxamine (PM) is an in vitro inhibitor of the formation of AGEs, which acts by blocking the conversion of the Amadori intermediate to AGEs. In both animal experiments and human clinical trials, PM has shown promising results in delaying the onset of diabetic renal disease. The goal of this study was to use model proteins to gain insight into how glucose modifications alter protein functionality and how PM prevents the loss of protein function. Biochemical analyses of ribonuclease, lysozyme, bovine serum albumin, and ubiquitin provided insight into the damaging effects of glucose modifications on protein function, which included the formation of the prominent AGE, carboxymethyllysine (CML), and the degradation and cross-linking of proteins. PM prevented CML formation, protected protein integrity, and preserved enzyme function. Mass spectrometry analyses show that in lysozyme residues K-96 and K-97 are more susceptible to CML formation, and PM protects against CML formation on these residues. Notably, K-96 and K-97 are in close proximity to the active site, suggesting that CML, but not Amadori, formation near the active site is detrimental to lysozyme activity. Also, PM was found to protect against the oxidation of tryptophan residues, W-62 and W-63. We suggest that the same mode of PM protection may also occur in vivo and may be responsible for PM efficacy.