One of the most common hallmarks of cancer is dysregulation of cellular apoptotic processes. A comprehensive knowledge of the underlying mechanisms of the apoptotic machinery is vital for the identification of new drug targets and the development of innovative agents that stimulate the cell death process in cancer cells. Studies have shown that the lysosomal protease cathepsin D is important in the extrinsic apoptotic pathway stimulated by the death receptor ligands for TNFR1 and FAS, as well as by oxidative stress and the protein kinase C inhibitor staurosporine. To date, the role of cathepsin D in the chemotherapy-induced apoptotic pathway has not been characterized. This project examined the role of the lysosomal protease cathepsin D in chemotherapy-induced apoptosis of HeLa and U937 cells. The data demonstrated that following stimulation of U937 cells with the chemotherapy drug VP-16, cathepsin D was released into the cytosol approximately 4 hours after drug treatment. This release was selective for cathepsin D, as cathepsin B and the lysosomal markers LAMP and â-hexosaminidase were not released into the cytosol following VP-16 treatment. Inhibitors of caspases and cathepsin D had no effect on cathepsin D release, demonstrating that cathepsin D release occurred independently of caspase and cathepsin D activities. Downregulation of cathepsin D expression in U937 and Hela cells using siRNA was found to inhibit cell death resulting from a variety of stimuli, including death receptor ligands, oxidative stress, PKC inhibitors, and importantly, chemotherapy drugs. In addition, U937 and HeLa cells expressing cathepsin D siRNA exhibited delayed cytochrome c release and caspase-3 activation following VP-16 treatment. Moreover, isolated mitochondria from wild-type U937 cells released cytochrome c in response to cytosolic extracts that were treated with cathepsin D, suggesting that cathepsin D acts on a cytosolic factor to induce cytochrome c release. Inhibition of caspases had no impact on cytochrome c release provoked by cathepsin D-cleaved cytosolic extract, demonstrating that caspases are not mediators of cathepsin D-induced cytochrome c release. Taken together, these results demonstrate that cathepsin D is an important component of the apoptotic pathway and that it acts via an intermediary cytosolic factor to promote cytochrome c release and caspase activation during chemotherapy-induced apoptosis.

Zinc is a ubiquitous heavy metal that binds to proteins involved in critical cellular processes. Apart from its necessary role, excessive release of intracellular free zinc (Zn) is neurotoxic under stressed conditions characteristic of ischemic or epileptic neuronal injury. Our earlier results indicated that Zn-induced cell death is exacerbated in neurons compared to supporting neuroglia, suggesting that astrocytes have means to upregulate Zn buffering mechanisms, i.e. the Zn-binding protein metallothionein (MT). The first aim of this dissertation sought to address whether MT can effectively maintain Zn levels at a non-toxic minimum. From these studies, we have identified a dichotomous role for MT protective as a Zn buffering agent and detrimental as an oxidant-labile source for toxic Zn. Previous studies demonstrated the role of Zn as a mitochondrial toxin. Although it has been widely speculated that Zn is taken up by the mitochondrial calcium uniporter, the evidence is not entirely convincing. In the second specific aim, we addressed the specific hypothesis that mitochondrial Zn uptake occurs though the uniporter. Using a novel model involving isolated mitochondria pre-incubated with a Zn-selective fluorophore and attached to glass coverslips, we demonstrated for the first time direct visualization of mitochondrial Zn transport. The third specific aim addresses the importance of mitochondria as dynamic intracellular ATP factories, whose intracellular trafficking is critical for neuronal viability. We hypothesized that elevated Zn would attenuate mitochondrial trafficking. Our results revealed that Zn inhibited mitochondrial movement at pathophysiological levels. Intriguingly, acute activation of phosphotidyl inositol 3-kinase was implicated in both Zn-mediated movement inhibition and toxicity, providing a novel role for this traditionally pro-survival signaling pathway. In summary, this dissertation identifies intracellular targets for Zn-mediated neurotoxicity. We specifically emphasize the relevance of mitochondria as a Zn target under two circumstances which are critically dependent on the Zn concentrations established direct mitochondrial interactions that may involve Zn transport, and indirect mitochondrial interactions that affect intracellular mitochondrial trafficking.

The signal transduction field has recently seen a surge of interest in cascade scaffolding proteins. One of these, the Kinase Suppressor of Ras (KSR), has received a great deal of attention as a scaffold for the Ras/ERK signaling pathway. KSR interacts with both MEK and ERK, and possibly binds to Raf-1 as well. Very little is known about the regulation of KSR; however, it has been determined that membrane association is essential for its function in signal augmentation. KSR shares a high degree of sequence homology to Raf-1, including an almost identical phosphatidic acid binding region (PABR). Previous work in the Romero lab has determined the direct interaction of Raf-1 with phosphatidic acid is critical for its membrane recruitment. The PABR is a 35 amino acid sequence consisting of a poly-basic motif (PBM) flanked by two hydrophobic regions. Neutralization of the two arginine residues in the PBM abrogates the binding of Raf-1 to phosphatidic acid (PA), and consequently disrupts its membrane association. This thesis examines lipid-binding properties of the PABR and their potential role in the traffic and function of KSR. Using peptides corresponding to the PABR and tryptophan fluorescence spectroscopy, the data presented in the first section demonstrate that PA induces a blue-shift in the tryptophan emission spectra of WT KSR PABR, and this shift is specific for PA. The second section explores the cellular consequence of KSR PABR mutation. A KSR protein lacking the arginine residues in the PBM expressed in HIRcB fibroblasts retains its membrane-binding ability, but inhibits MEK and ERK phosphorylation in a dominant negative fashion. The data presented here support the conclusion that, although an intact PABR may not be essential for the membrane localization of KSR, it is essential for proper coupling of the pathway.

Head and neck squamous cell carcinomas (HNSCC) are characterized by upregulation of the epidermal growth factor receptor (EGFR). We previously reported that a gastrin-releasing peptide/gastrin-releasing peptide receptor (GRP/GRPR) autocrine growth pathway is activated early in HNSCC carcinogenesis. GRP can induce rapid phosphorylation of EGFR as well as p42/44 MAPK activation, in part via extracellular release of transforming growth factor alpha(TGF-alpha) by matrix metalloproteinases (MMP). Src family kinases have been reported to be activated by G-protein-coupled receptors (GPCRs) followed by downstream EGFR and MAPK activation. To further elucidate the mechanism of activation of EGFR by GRP in HNSCC, we investigated the role of Src family kinases. Blockade of Src family kinases using three different Src-specific tyrosine kinase inhibitors (A-419259, PP2 or PD0180970) decreased GRP-induced EGFR phosphorylation as well as MAPK activation. GRP also failed to induce MAPK activation in dominant-negative c-Src transfected HNSCC cells. Invasion and growth assays demonstrated that c-Src was required for GRP-induced proliferation or invasion of HNSCC cells. In addition to TGF-alpha release, GRP induced amphiregulin, but not EGF, secretion into HNSCC cell culture medium, an effect that was blocked by the MMP inhibitor, Marimastat. TGF-alpha and amphiregulin secretion by GRP stimulation was also inhibited by blockade of Src family kinases. Further investigation showed that TNF-alpha converting enzyme (TACE) underwent Src-dependent phosphorylation and translocation to the plasma membrane in a complex with c-Src and the p85 subunit of PI-3 kinase, where it regulated amphiregulin release. In addition, we identified that PDK1 kinase, a downstream target of PI-3 kinase, directly phosphorylated TACE. Knockdown of PDK1 augmented the anti-tumor effects of the EGFR inhibitor erlotinib. These findings implicate PDK1 as a new target in HNSCC and suggest that therapeutic strategies that block PDK1 may improve the clinical response to EGFR inhibitors. Combined targeting of GRPR and EGFR pathway also showed enhanced anti-tumor efficacy by inhibiting cancer cell proliferation, invasion and promoting apoptosis. Overall, these findings show the promises and benefits of combination therapy when targeting EGFR and GRPR pathways in head and neck cancer.

Mitochondria are the primary generators of ATP and are important regulators of intracellular calcium homeostasis. These organelles are dynamically transported along lengthy neuronal processes, presumably for appropriate distribution to cellular regions of increased need such as synapses. The removal of damaged mitochondria that produce harmful reactive oxygen species and promote apoptosis is also thought to be mediated by mitochondrial transport to autophagosomes. Mitochondrial trafficking is therefore important for maintaining neuronal and mitochondrial health while cessation of movement may lead to neuronal and mitochondrial dysfunctions. The demands for mitochondria differ between developing and mature neurons, and separate mitochondrial recruitment signals have been identified in each case. In the first aim, we examined how mitochondrial dynamics are affected by the development of synaptic connections in cortical neurons. We revealed reduced mitochondrial movement and elongated morphology in mature neurons which probably serve to optimize mitochondrial contact with synaptic sites. Synapses require mitochondria to supply ATP and regulate local [Ca2+]i for neurotransmission. The second aim investigated mitochondrial trafficking patterns relative to synaptic sites on axons and dendrites. We demonstrated that synapses are targets for long-term mitochondrial localization and dynamic recruitment of moving mitochondria, and that trafficking patterns are influenced by changes in synaptic activity. We also found that mitochondrial movement in dendrites is more severely impaired by neurotoxic glutamate and zinc exposures than in axons. These findings suggest a mechanism for postsynaptic dysfunction and dendritic degeneration in excitotoxicity. The third aim examined impaired mitochondrial transport as an early pathogenic mechanism in Huntingtons disease. Recent studies indicate that aggregates composed of mutant huntingtin fragments hinder axonal transport by sequestering wildtype huntingtin, cytoskeletal components and molecular motors. Our studies in cortical neurons demonstrated reduced mitochondrial trafficking specifically to sites of aggregates and impeded passage of moving mitochondria by aggregates resulting in discrete regions of mitochondrial accumulation and immobilization. In summary, this dissertation provides new insight into our understanding of mitochondrial trafficking, morphology and distribution in cortical neurons that are developing, synaptically mature, acutely injured, and diseased. We conclude that mitochondrial movement is dynamic in healthy neurons and that injured neurons exhibit different manifestations of impaired movement.

SCREENING ASSAY FOR SELECTIVE ESTROGEN RECEPTOR MODULATORS Elena Sirbu, BS University of Pittsburgh, 2006 Estrogen influences the development and progression of breast cancer and of other types of cancer, such as ovarian and lung cancer. The best strategy for prevention and treatment of estrogen dependent cancers is to selectively block estrogen activity in the affected estrogen dependent tissues. The beneficial role of estrogen in the other tissues should be preserved. One of the most common methods to prevent the harmful effects of estrogen is to block the estrogen receptor signaling. The intense research in the breast cancer treatment and prevention field produced a number of estrogen related compounds. The existing screening assays to test the selectivity and potency of these compounds have major limitations. I propose here the development and validation of a rapid screening assay for selective estrogen receptor modulators. This assay is based on the use of an ERE (estrogen response elements) to drive expression of a fluorescent protein that can be visualized directly in living cells. I presented here the first step in developing the screening assay, the generation and evaluation of two fluorescent clones, ERE-GFP and ERE-DsRed. The clones were introduced in CV-1 cells, together with ER, using transient transfection in order to test whether they are under tight estrogenic control. The cells were further treated with know ER ligands. These results predict that the clones function as expected. A robust signal resulted in the presence of estradiol, while with a pure antiestrogen such as ICI 182,780 resulted in very little red/green fluorescence. The vehicle control (ethanol) also elicited very little response (fluorescence). Further, these clones can be stably integrated in CV-1 cells together with either ER alpha or ER beta in order to develop a high content screening assay for SERMs. The new SERMs identified using this assay can be used eventually in therapy of breast or lung cancers or as hormone replacement. In addition, compounds that differentiate ER¦Á and ER¦Â will be valuable tools to further dissect ER signaling pathways. It is important to know more about coactivator recruitment, gene expression profile or about the response with ER mutations. This will lead to a better understanding of estrogen related cancers and will help designing new therapeutic approaches.

In spontaneously hypertensive rats (SHR) the hypersensitivity of the renal vasculature to angiotensin II (Ang II), compared to Wistar-Kyoto rats (WKY), appears to be the determining factor in the development and progression of hypertension. Recent evidence indicates that the ERK cascade and NAD(P)H oxidase generation of superoxide are involved in smooth muscle contraction, and Phospholipase D (PLD) generation of phosphatidic acid is involved in activation of ERK and NAD(P)H. Importantly, Ang II-mediated PLD activity is greater in aortic smooth muscle from SHR compared with WKY; however, this signaling pathway has not been examined in the kidney vasculature. The purpose of these studies were to define Ang II-mediated signal transduction mechanism(s) involved in PLD regulation in WKY and SHR preglomerular smooth muscle cells (PGSMCs). The goals of this study were to determine: 1) whether Ang II-mediated PLD activity is greater in SHR; 2) the Ang II signaling pathway(s) responsible for regulating PLD activity, and whether they are altered in SHR; and 3) whether PLD-mediated generation of phosphatidic acid is involved in Ang II-induced activation of the ERK cascade. The data indicates that the mechanisms leading to activation of PLD are similar in WKY and SHR and PLD is required for Ang II activation of ERK; however, Ang II more potently activates PLD in SHR. Further analysis indicates that the AT2 receptor inhibits AT1 receptor/RhoA-dependent activation of PLD through a nitric oxide/cGMP-dependent phosphorylation of RhoA at serine 188, which promotes RhoGDI inhibition of RhoA. These experiments expose two key differences between WKY and SHR PGSMCs: 1) SHR have an increased AT1/AT2 receptor ratio; and 2) SHR are less sensitive to nitric oxide and cGMP. Therefore, the hypersensitivity of the SHR to Ang II may be due to an imbalance in Ang II receptors and/or impaired AT2 receptor-mediated signaling within the kidney vasculature.

Huntington's Disease (HD) belongs to the CAG repeat family of neurodegenerative diseases and is characterized by the presence of an expanded polyglutamine (polyQ) repeat in the huntingtin (htt) gene product. PolyQ-expanded htt accumulates within large aggregates in various subcellular compartments, but are more often localized within the nucleus. The sequestration of proteins essential to cell viability may be one mechanism that accounts for toxicity generated by polyQ-expanded proteins. Nuclear inclusions containing polyQ-expanded htt recruit the transcriptional cofactor, CREB-binding protein (CBP). PolyQ toxicity appears to involve alterations of gene transcription and reduced neuronal cell viability. In the HT22 hippocampal cell line, we found that toxicity within individual cells induced by polyQ-expanded htt was associated with the localization of the mutant htt within either nuclear or perinuclear aggregates. However, in addition to CBP recruitment, we found that CBP ubiquitylation and degradation can be selectively enhanced by polyQ-expanded htt. Thus, selected substrates may be directed to the ubiquitin/proteasome-dependent protein degradation pathway (UPP) in response to polyQ-expanded htt within the nucleus. While both the polyQ domain and the histone acetyltransferase domain (HAT) of CBP have been found to interact with polyQ-expanded htt, deletion of either domain does not affect its enhanced degradation in the presence of polyQ-expanded htt in HT22 cells. Thus, enhanced degradation of CBP in cells expressing polyQ-expanded htt may not involve a direct interaction between CBP and htt. It seems likely specific enzymes in the UPP may be activated by htt and selectively target proteins such as CBP for degradation. Since molecular chaperones are found in the aggregates containing polyQ-expanded proteins, misfolding of polyQ-expanded proteins may play a key role in polyglutamine disease pathogenesis. In a number of some studies, HDJ-2, a member of DnaJ family molecular chaperones, was found to reduce aggregation and toxicity induced by polyQ-expanded proteins. In contrast, we show that HDJ-2 is unable to rescue aggregate formation of polyQ-expanded htt in transfected HEK293 fibroblast cells, nor is it recruited into these aggregates in vivo in a HD transgenic mouse model. Thus, molecular chaperone effects on polyQ-expanded protein induced toxicity could be cell-type specific or influenced by the developmental state of the culturable cells. These factors must be considered in any attempts to use chaperones as potential therapeutic targets in polyglutamine diseases.

Deregulation of normal cell cycle control is essential for malignant transformation. The Cdc25A dual-specificity phosphatase promotes cell cycle progression by dephosphorylating and activating the cyclin-dependent kinases. Cdc25A has oncogenic and anti-apoptotic activity and is overexpressed in many human tumors. The mechanisms by which Cdc25A is overexpressed in human cancer are unknown. Cdc25A protein levels are downregulated by cell cycle checkpoints in response to genotoxic stress; cell cycle checkpoints are frequently compromised in tumor cells. In addition, under normal physiologic conditions, the half-life of Cdc25A protein is short. Alterations to physiologic Cdc25A regulatory mechanisms could be sufficient to result in oncogenic overexpression of this cell cycle regulatory protein. While Cdc25A downregulation in response to genotoxic stress occurs through defined signal transduction pathways, regulation of Cdc25A protein levels in non-stressed cells is poorly understood. The purpose of this thesis was to examine the physiological regulation of Cdc25A protein levels in human tumor cells. The goals of our studies were: 1) to investigate regulatory mechanisms of Cdc25A protein levels in non-stressed human tumor cells; 2) to understand how Cdk2 kinase activity regulates Cdc25A protein levels; and 3) to explore the mechanism by which Cdk2 kinase activity regulates Cdc25A protein turnover. The results of our studies revealed that Cdc25A protein half-life in non-stressed interphase cells is regulated, in part, by Cdk2 kinase activity, and that Cdk2 does not regulate Cdc25A turnover by affecting several known signal transduction pathways that control Cdc25A protein stability. Recent reports on the role of ubiquitin ligases in physiologic Cdc25A turnover have identified several phosphorylation sites that are necessary for efficient Cdc25A recruitment to ubiquitin ligases. The kinase(s) responsible for phosphorylating these serine residues remain to be identified, although Cdk2 could be one prime candidate. While initial reports of the interactions between Cdc25A and Cdk2 focused on an auto-amplification feedback loop that results in increased catalytic activity of both proteins, it now appears that Cdk2 also regulates Cdc25A stability and plays an important role in regulating Cdc25A protein levels during interphase progression.

The overall hypothesis of this thesis project was that bleomycin hydrolase (BLMH) has biologically specific and unique functions in the central nervous system (CNS). BLMH is a multifaceted papain superfamily cysteine protease that has importance in drug metabolism. The physiological functions of this protease are unknown as are factors regulating its expression. Immunohistochemical examination of B6.129Blmhtm1Geh/J null and control animals showed no gross abnormalities; however, marked global astrogliosis was observed in the null aged animals. To define the role of BLMH in the brain, the behavioral phenotype of hybrid [129S6-Blmhtm1Geh/J X B6.129 Blmhtm1Geh/J]F1 null and littermate controls was characterized using multiple behavioral paradigms. Deletion of Blmh was found to result in deficits among young animals in water maze probe trials. Retention of target platform location during the probe trials requires both learning and memory as well as sensory and locomoter skills. No overt sensory or motor deficits were noted in Blmh null F1 hybrids. The profile of BLMH expression and its regulation in the CNS was studied next. Inducible transcription of BLMH was evaluated in the context of a putative role in the processing of MHC I epitopes. BLMH was found to be differentially regulated in microglia and astrocytes. In microglia, Blmh protein was significantly induced by gamma interferon or tumor necrosis factor a, whereas in astrocytes, no change in protein expression was observed. Treatment of microglial derived cell lines with both gamma interferon and tumor necrosis factor a revealed synergistic effects between the cytokines. BLMH protein induction was accompanied by increased Blmh mRNA. These results suggest that cell specific regulation of BLMH is an important control mechanism for this protease. These data also provide further evidence for a targeted immune related biological function for BLMH. It is concluded that BLMH potentially has multiple unique and biologically important functions within the brain.