It is widely accepted that activating mutations of genes encoding the Notch family of transmembrane receptors, specifically Notch1, are associated with oncogenic transformation. Previous data from our lab has shown that an active form of Notch1 (Nic) provides a protective effect against apoptosis in D011.10 T cells, and that this effect may be attributed to Nic binding the pro-apoptotic protein Nur77. Nur77 is an immediate early gene that is upregulated during negative selection of thymocytes and activation-induced apoptosis in D011.10 T cells. Nur77 upregulation is tightly regulated and requires MEF2D, NFAT, and the co-activator, p300, to effectively respond to apoptotic stimuli. In this report we show that Nic has the ability to interfere with the induction of transcription of Nur77, and that this interference is directly related to the inability of p300 to bind the Nur77 promoter in the presence of Nic. We also show that blocking Notch activation through gamma secretase-inhibitors or siRNA directed against Notch1 in T cell acute lymphoblastic leukemia (T-ALL) cell lines restores Nur77 upregulation in response stimuli. These observations support a model in which during thymocytes negative selection activating mutations of Notch1 inhibit the upregulation of a crucial proapoptotic molecule. Studies to determine the mechanism by which Nur77 induces apoptosis have indentified a unique translocation of Nur77 from the nucleus to the cytosol. It has been determined that once in the cytosol Nur77 interacts with members of the Bcl-2 family of proteins at the mitochondrial membrane. This interaction induces a conformational change of Bcl-2 so that is becomes pro-apoptotic instead of protective. Of similar interest is the role that Nur77 itself plays during the induction of activation-induced apoptosis which may be independent of Bcl-2 conformational change. In an effort to describe possible functions of Nur77, DO11.10 cells that have Nur77 under a tet-inducible promoter were observed for changes IP3R. Initial results from our lab suggest that Nur77 alone has the ability to induce cell death in DO11.10 cells using this tet-inducible system. Interestingly we have been able to identify distinct changes in IP3R isoforms during stimulation induced apoptosis and Nur77-dependent apoptosis. Current experiments are focused on a mechanism beyond the known function of the Nur77/Bcl-2 interaction; that Nur77 may also be acting as a physical barrier between the known anti-apoptotic interaction of IP3R and Bcl-2, leading to sustained calcium flux.

Breast cancer tops the list of new cancer cases and is predicted to be the second leading cause of cancer deaths in women in 2012. The primary objective of the present study was to provide insights into the molecular mechanisms underlying the aggressive growth and metastasis of triple-negative and basal-like breast cancers. To study increased growth and invasive behavior in triple-negative and basal-like breast cancers we utilize both an interesting and relevant cell culture model and examination of human tissue. S-phase kinase-associated protein 2 (SKP2) plays an important role in cell cycle regulation by targeting p27 for degradation. The cyclin-dependent kinase (CDK) inhibitor p27 regulates G1/S transition by binding cyclin/CDK complexes and abrogating its activity. By targeting p27 for degradation, SKP2 frees the complexes needed to progress into the S phase of the cell cycle. Evaluation of SKP2 expression in TMX2-28 revealed significantly higher levels than in other breast cancer cell lines. Despite the high levels of SKP2 expression, p27 protein was not reduced. However, levels of the Serine 10 phosphorylated form of p27 (pSer10p27), which has been associated with increased proliferation rates, was found to be increased. Furthermore, suppression of SKP2 completely eliminated the pSer10p27 and slowed cycle progression confirming the role of SKP2 in the aggressive growth of TMX2-28 cells. Assessment of mRNA from 30 frozen human breast cancers demonstrated that SKP2 is more highly expressed in ER&agr;-negative and basal-like breast cancers. Immunohistochemical analysis of 188 breast cancers and 50 benign reduction mammoplasty tissues confirmed that SKP2 is more highly expressed in ER&agr;-negative breast cancers and for the first time demonstrated that triple-negative breast cancers are more likely to overexpress SKP2 than are non-triple-negative, but still ER&agr;-negative, tumors. In contrast to some previous reports, we did not observe an inverse relationship between SKP2 and p27 expression. Only 11% of tumors expressed high SKP2 and low p27, while 32% of tumors had high SKP2 and high p27. Although no significant relationship between SKP2 and p27 expression was observed in human breast cancers, a significant positive relationship was discovered between SKP2 and pSer10p27. Furthermore, high levels of SKP2 and pSer10p27 were observed significantly more often in ER&agr;-negative and triple negative breast tumors than in ER&agr;-positive breast cancers. Based on these results and those of the cell culture experiments showing complete elimination of pSer10p27 after suppression of SKP2 it appears that levels of pSer10p27 may be a better indicator of SKP2-dependent p27 degradation than are levels of p27. Therefore, that inhibiting SKP2 in triple-negative breast cancers expressing high levels of both SKP2 and pSer10p27 regardless of p27 levels may be a valid therapeutic approach. We determined that TMX2-28 lack MMP-1 mRNA, and MMP-2/MMP-9 protein expression; each of which is important in protease-dependent invasion. Furthermore, TMX2-28 cells have low expression of other genes key to protease-dependent invasion including Slug, Zeb 1, Zeb 2, Vimentin, Fibronectin and N-cadherin. RhoA is a member of the Rho superfamily of GTPases that acts as a molecular switch to control signal transduction and is critical to the amoeboid invasion mechanism. TMX2-28 cells have high expression of protease-independent invasion genes such as RhoA, ROCK 1, ROCK 2, and E-cadherin. Finally, treating TMX2-28 cells with a RhoA pathway inhibitor or an shRNA targeting RhoA significantly reduces their invasiveness. These data suggest that TMX2-28 cells use a RhoA-dependent, proteolytic-independent invasion mechanism. Collectively, the data presented here demonstrate the roles of SKP2 and RhoA in triple-negative and basal-like breast cancers, making both genes, as well as their pathways, desirable therapeutic targets. (Abstract shortened by UMI.)

Glioblastoma Multiforme (GBM) is the most common brain cancer and one of the most fatal forms of cancer overall. The average survival time is 10-14 months, and less than 10% of patients survive more than 5 years after diagnosis. It is characterized by extreme vasculature, chemo/radioresistance, and invasiveness into the normal brain. The current standard of care, which includes surgical removal of tumor, radiation, and the chemotherapeutic agent temozolomide, initially stunt tumor growth. Nevertheless, the tumor invariably rebounds and the patient succumbs to the disease. Therefore, there is an urgent need to develop new therapies for this devastating disease. YKL-40 is one of the most over-expressed proteins by GBM cells, and is elevated in the serum of patients with GBM. YKL-40 is implicated in a host of inflammatory diseases and has been shown to play a major role in the maturation of some cells of the immune system, especially macrophages. Thus, it has been suggested that YKL-40 may act as a prognostic biomarker for cancer and inflammatory disease. However, little is known about the role of YKL-40 in relation to cancer development and progression, and more work needs to be done to validate it as a biomarker or as a therapeutic target. It was the goal of the work described herein to uncover some of the key molecular mechanisms of GBM development and progression in the hopes of offering new therapeutic targets. Using a wide variety of in vitro and in vivo techniques, the role of a secreted glycoprotein YKL-40 in GBM was probed. It was demonstrated that YKL-40 enhanced angiogenesis, radioresistance, and progression of GBM cells. Moreover, inhibition of YKL-40 in mouse models markedly arrested tumor growth and vascularization, lending support to the idea of YKL-40 as a therapeutic target. Finally, YKL-40 drove GBM cells into a mesenchymal phenotype, where the tumor cells act as mural-like cells, supporting tumor vasculature networks. Hopefully, the results from these studies will offer the rationale to develop drugs against YKL-40 and potentially extend the lives of patients with this terrible disease.

The root of the Tibetan plant Rhodiola crenulata is part of eastern traditional medicine. Studies have suggested that members of the Rhodiola genus display anticancer properties. In this study we examine the effect of R. crenulata in a cellular model of invasive breast cancer, this disease being the second cause of cancer death among women in the US. Deregulation of the Wnt/β-catenin pathway has been frequently observed in breast cancers and appears to have a key role in the transformation of benign cells to a malignant form. Although mutations of the Wnt growth factor are rarely observed in cancer, the Wnt signaling pathway is often up-regulated by either mutations that result in stabilization of β-catenin or by hypermethylation and subsequent loss of expression of Wnt signaling antagonists like secreted Frizzled-Related Protein 1 (SFRP1) (Hanahan and Weinberg 2000; Miyoshi, Rosner et al. 2002; Reya and Clevers 2005; Suzuki, Toyota et al. 2008) (Hanahan and Weinberg 2000; Miyoshi, Rosner et al. 2002; Reya and Clevers 2005; Suzuki, Toyota et al. 2008) (Hanahan and Weinberg 2000; Miyoshi, Rosner et al. 2002; Reya and Clevers 2005; Suzuki, Toyota et al. 2008). We used an engineered cell line in which SFRP1 expression has been knocked down. These cells were derived from 76NTert cell line, an immortalized human mammary epithelium cell line. The resulting 76NTert-siSFRP1 cells display a mesenchymal-like phenotype, invasive behavior and are more resistant to apoptosis triggered by anchorage independent conditions, or anoikis. Additionally we used a highly invasive estrogen receptor negative (ER-), progesterone receptor negative (PR-), Her2/neu negative (triple negative) breast cancer cells line MDA-MB-231. Treatment of MDA-MB-231 and 76NTert-siSFRP1 cells with an extract of R. crenulata inhibited migration and invasion of both cell types, as compared to untreated cells. Furthermore, R. crenulata sensitizes cells to anoikis but does not increase γ-irradiation induced cell death. We provide evidence that death induced by R. crenulata does not occur through the inhibition of an epithelial-to-mesenchymal transition (EMT). Taken together, our initial results suggest R. crenulata as a potential therapeutic agent for breast cancer patients with mutations in the Wnt/β-Catenin signaling pathway. Additionally, we present evidence that R. crenulata exerts its anti-metastasis effect by inhibiting cell migration and increasing cell attachment to a substrate. We demonstrate that this effect occurs by R. crenulata's modulation of the Rho GTPase effector ROCK1. We further show evidence that R. crenulata's effect on ROCK is not limited to cancer cells (in vitro), but also affects isolated and intact embryonic tissues. We discovered that treatment of embryos with R. crenulata can cause a spina bifida phenotype, suggesting (1) that compounds in R. crenulata may prove detrimental to a developing embryo, and (2) the active compounds in R. crenulata may prove useful in the study of developmental anomalies that lead to conditions such as spina bifida. More importantly, our results suggest that pregnant women should avoid taking R. crenulata-containing supplements during pregnancy. Compounds in R. crenulata may be contraindicative to the pregnancy and cause injury to a developing fetus. The information provided may help health providers offer better advice on natural supplements to expecting mothers. We also characterized and identified multiple R. crenulata compounds and report predicted protein targets for these compounds. Finally, we provide evidence that R. crenulata affects cancer cell metabolism and suggest potential protein targets of the chemical components of this extract. This study provides new information that will help dissect the mechanisms of action of the R. crenulata compounds and possible synergies amongst them.

To determine whether metazoan mitochondria retain the capacity to evolve on their own in response to selection pressure, I conducted an eight-month selection experiment on six populations of the monogonont rotifer Brachionus plicatilis. Rotifers were cultivated in brine shrimp hatchers under constant light, and were periodically exposed to the protein synthesis inhibitors chloramphenicol or cycloheximide. I uncoupled mitochondrial from recombinant chromosomal heredity in the test populations, either by destroying resting eggs to isolate mtDNA within all-clone lineages, or by restricting chemical exposure to males to halt mtDNA transmission to offspring. Rotifers' tolerance for toxins was tested after 10, 23, and 30 exposures; partial mitochondrial 16S rRNA gene sequences were also obtained from rotifer populations after 30 treatments. A novel genetic polymorphism that correlated with an increase in chloramphenicol resistance in the CAP-treated asexual B. plicatilis population was noted. Other toxin-treated rotifer populations' gene sequences exhibited no such increase in frequency for mitochondrial variants, even in cases where a population's resistance to toxins did improve. This is consistent with my hypothesis that asexual rotifers' mitochondria would adapt to the CAP antibiotic most easily, whereas the males-only treatment groups' chromosomes would adapt more easily to the cytoplasmic toxin cycloheximide.

Genetic, biochemical, and biophysical methods have been adopted to characterize the interactions between ligand and transmembrane receptor, between receptor and cytoplasmic proteins, and between cytoplasmic proteins in the bacterial chemosensory system in order to elucidate the signaling mechanism at the molecular level. Ligand binding to the serine receptor (Tsr) of Escherichia coli has been studied using Isothermal Titration Calorimetry (ITC) to determine the number of moles of ligand bound per mole of receptor (n), the binding constant $(K\sb{\rm a}),$ and the enthalpy of binding $(\Delta H)$ of serine to the receptor. The n-value for serine binding to octyl-$\beta$-D-glucopyranoside (OG)-solubilized Tsr (n = 0.5) was consistent with one molecule of serine binding to a receptor dimer. The values for $K\sb{\rm a}$ and $\Delta H$ were equivalent for the membrane and OG-solubilized samples and were found to be $\rm 3.6\times 10\sp4\ M\sp{-1}$ and $-18$ kcal/mol at 27$\sp\circ$C. Covalent modification of Tsr at the sites of methylation was found to have only a modest effect on serine binding. The interaction between Tsr and the methyltransferase (CheR) was also studied using ITC. Tsr bound to CheR with 1 to 1 ratio and a dissociation constant of 2 $\mu$M at 29$\sp\circ$C. Ligand-binding or modifications at the methylation sites of Tsr did not affect CheR binding. Truncation of Tar cytoplasmic-fragment at the C-terminus by 16 amino acids (aa) completely abolished CheR binding. This result led to the identification of the CheR binding site which is located at the C-terminal 5-aa segment remote from the methylation regions. The C-terminal truncated and crosslinkable Tsr mutant ($\Delta$34, D36C) was then constructed, which was found to be a poor substrate for CheR by itself. Coexpression of this binding site deleted Tsr with intact Tsr rescued its methylation activity even in the crosslinked form. The mechanism of interdimer methylation was established through this study. The autophosphorylating kinase CheA donates a phosphoryl group to either of two response regulator proteins, CheY or CheB. With ITC, it was demonstrated that CheA and a CheA fragment composed of aa residues 1 to 233 (CheA$\sb{1-233}$ bound to CheY with similar dissociation constants of 2.0 and 1.2 $\mu$M at 28$\sp\circ$C, respectively, indicating that the CheY binding site is wholly within the 1-233 aa locus. CheB bound to CheA$\sb{1-233}$ with a $K\sb{\rm d}$ of 3.2 $\mu$M, and also bound to intact CheA with the same affinity. CheY was found to compete with CheB for binding to CheA$\sb{1-233}.$ Phosphorylated CheY, in the presence of 6 mM Mg$\sp{2+},$ has a significantly lower affinity for CheA (20% of unphosphorylated form), but mutations at the active site of CheY have little effect on CheY-CheA interaction, a mutation remote from the active site (A103V) produced a 10-fold reduction in $K\sb{\rm a},$ indicating the separation of the binding site from the active site and a significant conformational change upon phosphorylation.

Calmodulin (CaM) is a major intracellular calcium receptor. There is abundant calmodulin (CaM) in the oocytes and eggs of B. germanica during vitellogenesis and early embryogenesis. The accumulation of CaM in oocytes may be for immediate use in the oocytes and/or in preparation for later stages of their development. Previous investigation from this laboratory suggested that maternal follicle cells are the most likely source of this CaM. Tissue culture labeling with $\sp{35}$S methionine showed a 13-fold higher rate of synthesis of CaM in the follicle cells than in oocyte preparations (Zhang & Kunkel, 1994). The high rate of biosynthesis of CaM in the follicle cells, and the absence of extracellular CaM in transit in the hemolymph suggested that CaM is made in the follicle cells and transferred to the oocytes. In order to obtain more information about the site of CaM synthesis I isolated total RNA from different tissues that could potentially contribute to the high amounts of follicular CaM and measured the amounts of CaM transcripts during development. I show that isolated whole follicles accumulate more transcripts for CaM than the fat body. The steady state levels of CaM transcripts increases 150 fold during the 4 day developmental period under study. This is in addition to a 32 fold increase in total follicle RNA during the period. Steady state levels of CaM transcripts in whole follicles also show a pattern of increase disproportionate to the increase in volume of the whole follicle. In comparison steady state levels of actin transcripts increase 35 fold during the same developmental period. At 96 hr post feeding, in a given amount of total RNA, follicle cell total RNA contains 3 times more CaM transcripts than whole follicle total RNA, and 70 times more CaM transcripts than the fat body tissue. The oocyte total RNA collected from material expelled from the whole follicle contains less than 10% of the amount of CaM transcripts available in the follicle cells. The fat body tissue preparation shows little developmental increase in steady state levels of CaM transcripts despite a 4 fold increase in total RNA. In my investigation into the control of the accumulation of this transcript I found that deprivation of JH, by head ligation, not only causes atresia of the follicles, but also reduces CaM transcript accumulation. Reconstituting JH titer by injection allows a selected population of follicles to develop to full size and also reinstates steady state CaM transcript levels above that of unligated controls. The results of my study makes the CaM gene a potentially important target for the study of JH action in follicle cells during oogenesis.

Fluorescent ratiometric imaging and spectrofluorometric analysis was used to study two signal transduction mechanisms in the stamen hair cells of Tradescantia virginiana. The first study determined the metabolic pathway necessary for the inactivation of Inositol 1,4,5-trisphosphate mediated calcium release from intracellular stores in the living plant cell. Tradescantia stamen hair cells, preloaded with the calcium sensitive ratiometric dye fura-2-dextran, were injected with analogs of Inositol 1,4,5-trisphosphate and cytosolic calcium levels monitored by ratiometric imaging. The injected analogs were selected due to their insensitivity to various kinases and phosphatases for which Inositol 1,4,5-trisphosphate is a substrate. We determined that the 5-phosphatase is the preferred pathway for inactivation of Inositol 1,4,5-trisphosphate in the living plant cell. The second study investigated cyclic ADP-ribose mediated calcium release in the intact plant cell and determined the presence of the metabolic machinery necessary to synthesize cyclic ADP-ribose from its precursor NAD+. Cyclic ADP-ribose was observed to cause calcium release in the stamen hair cells of Tradescantia that were preloaded with the calcium sensitive dye fura-2-dextran. Evidence of cyclic ADP-ribose synthesis was determined using two experimental techniques. Homogenates of the sea urchin Lytechnicus piclus were used as bioassays to detect cyclic ADP-ribose in extracts of Tradescantia stamen hair cells that were incubated with [special characters omitted]NAD+. Cyclic ADP-ribose synthesis was detected from fluorimetric analysis of the homogenate as the calcium sensitive dye fluo-3 was present in the homogenate to detect cyclic ADP-ribose mediated calcium release from sea urchin egg microsomes. We also determined cyclic ADP-ribose synthesis by injection of fura-2-dextran loaded stamen hair cells with [special characters omitted]NAD+ and observing a delayed calcium increase in the cytosol. These results establish the metabolic fate of inositol 1,4,5-trisphosphate in plant cells and demonstrate the biochemical capability for plant cells to synthesize cyclic ADP-ribose to mediate calcium release in plants.

Pollen tube growth, the process that transports the sperm cell to the ovule, is fundamental for plant sexual reproduction. The actin cytoskeleton is essential to the process of pollen tube elongation. To further understand how actin is regulated in the pollen tube, I characterized two actin regulatory proteins: profilin and plant-villin (ABP-135) from Lilium longiflorum . Profilin, a small actin monomer binding protein, is abundant in pollen. By a combination of rapid fixation, immunological staining and live cell analysis, I show that profilin is a soluble protein evenly distributed in the cytosol. After estimating its intracellular concentration, I elevated its concentration by microinjection to study the effect on cytoplasmic streaming and cell growth. Increasing profilin concentration by 25% resulted in half-maximal growth inhibition, while a 60% increase was necessary to half-maximally inhibit cytoplasmic streaming. My results suggest that actin polymerization is essential for tube growth, that profilin is an actin-monomer sequestering protein, capable of regulating actin's polymerization state, and that the participation of actin on growth is separable from that on cytoplasmic streaming. I characterized a second actin-binding protein, originally identified as an F-actin bundling factor, which co-localizes with actin cables in the pollen tubes. It is composed of several isoforms, and is widely distributed in different plant organs. By cloning its cDNA, I identified this protein as the plant homologue of villin; these are calcium dependent severing, capping, nucleating and bundling proteins. The actin cytoskeleton may be regulated via this plant villin, which could respond directly to the calcium gradient present at the tip of the pollen tube.

This dissertation is in two parts. The first part investigates the role of lipid signalling pathways in an intracellular phenomenon i.e. the formation of the nuclear envelope. The second part is a study of lipid signalling pathways in extracellular phenomenon, i.e. HeLa cell adhesion and spreading on a gelatin substrate. The disassembly and formation of the nuclear envelope are crucial steps in the progression of mitosis. Nuclear envelope dynamics involve many steps and since vesicular transport, binding and fusion of vesicles are essential for the formation of the nuclear envelope it was our interest to explore whether there were any parallel pathways, such as found in the secretary pathways, that were also used in the formation of the nuclear envelope. There is little information on the proteins of membrane vesicles that reconstitute the nuclear envelope and practically none about their lipid composition. It was therefore important to determine their lipid structure in order to proceed with investigating whether, during the formation of the nuclear envelope, similar signalling pathways to those in the vesicle trafficking operate. Cytoplasmic membrane vesicle fractions (MVs) from sea urchin eggs which, contribute to the nuclear envelope were studied. The phospholipid composition of the membrane vesicles, MV1, MV2 and MV3 was determined using a novel approach for direct quantification of phospholipids from two-dimensional 31 P-1H nuclear magnetic resonance spectroscopy with isotropic mixing. MV2 and MV3 have similar compositions typical of the ER and the Golgi membranes. However, MV1 is mainly composed of phosphatidylinositol with phosphatidylcholine being the minor phospholipid present in MV1. Furthermore, we determined that phosphatidylinositol specific phospholipase C (PI-PLC) promoted nuclear envelope formation. However, in the absence of MV1, PI-PLC, did not induce nuclear envelope formation. Inhibition of membrane vesicle fusion in a dose dependent manner, by wortmannin (a specific inhibitor of the PI-3kinase pathway) suggested that the PI-3 kinase branch of the phosphatidylinositol pathway may be involved in the formation of the nuclear envelope. These experiments indicate that the inositol phospholipid pathways, as in constitutive membrane trafficking, are also involved in the formation of the nuclear envelope. Furthermore, it is the first time that a biological membrane, MV1, with such an unusual composition has been reported and that it has been demonstrated the phosphatidylinositol hydrolysis is involved in the formation of the nuclear envelope. In the second part the goal was to determine which lipoxygenase metabolites were involved in facilitating HeLa cell adhesion and spreading to a gelatin substrate. Reverse phase HPLC methods demonstrated that HeLa cell homogenates converted arachidonic acid into 12-, 15- and 5-hydroperoxyeicosatetraenoic (HETEs), indicating that 12, 15 and 5 lipoxygensases are active in HeLa cells. In order to investigate which lipoxygenase pathway are required to induce cell spreading the lipoxygenase pathway was inhibited by 25uM nordihydroguaiaretic acid (NDGA), a specific inhibitor of the lipoxygenases and MK866, a specific inhibitor of leukotriene biosynthesis. The effect of the different enantiomers of 12-BETE and 15-HETE on the reversal of NDGA inhibition was determined. The leukotrienes that overcome the inhibition of cell spreading by NDGA and MK866 were also determined. It can be concluded that 12,15 and 5 lipoxygenase enzymes are present and active in HeLa cells; and 12-HETE, 15-HETE, leukotrienes LTB4, LTD4 and LTC4 play an active role in HeLa cell spreading on a gelatin substrate.