In the United States in 2009, 192,370 women are expected to be diagnosed with invasive breast cancer, and 62,280 with in situ disease. About 70% of these cases are estrogen receptor positive (ER+). There are two isoforms of the ER, α and β, that differ somewhat in structure and action. ERβ expression plays a protective role in breast cancer, and selective targeting of this isoform would have many beneficial effects. Tamoxifen has long been the standard of care for patients with ER+ breast cancer. A major problem with tamoxifen is the development of drug resistance. One of the mechanisms proposed for the development of tamoxifen resistance involves the loss of ERβ expression. The first objective of this study was to screen a library of biphenyl C-cyclopropylalkylamides for their ability to function as ERβ-selective ligands. Two compounds were identified with modest selectivity for ERβ and anti-proliferative effects in breast cancer cells where they inhibited expression of c-Myc. The nuclear matrix (NM), the structural scaffolding of the nucleus, plays a major role in many fundamental processes of the cell. Using the ER+ breast cancer cell line MCF-7 and an antiestrogen resistant derivative, along with subtype selective ER ligands, alterations in the abundance of specific proteins present in the NM were identified using a mass spectrometry (MS)-based relative quantitative methodology. Some of the most interesting proteins with altered abundance are NuMA, serpin H1, hnRNP R, and dynein heavy chain 5. These proteins may represent putative biomarkers to customize treatment. The alterations also provide a mechanistic understanding of tamoxifen resistance. The NM was also investigated by MS in the earliest stage of breast cancer, ductal carcinoma in situ (DCIS), utilizing novel cell lines derived from normal (breast reduction), DCIS, and non-diseased contralateral breast surgical specimens. Two of the interesting proteins found to be altered in DCIS were HSP90 and EEF1D. These studies may provide biomarkers to aid in the diagnosis and treatment of breast cancer. In addition by understanding the mechanism behind the development of breast cancer, prevention becomes a possibility.

Caveolin-3 is a muscle specific scaffolding protein with both structural and signaling roles. Lack of caveolin-3 expression has been implicated in limb-girdle muscular dystrophy, along with distal myopathy and rippling muscle disease. These diseases are characterized by progressive muscle weakness and muscle wasting. Nicotinic acetylcholine receptor (nAChR) clustering and localization are important for efficient nerve to muscle contractile signal transmission. It is hypothesized that muscle weakness could originate through disrupted nAChR clustering, disrupting the efficiency of signaling from the motorneuron to the muscle. While the molecular mechanisms involved in nAChR clustering remain to be fully defined, we hypothesize caveolin-3 is important for nAChR clustering and overall neuromuscular junction function. Caveolin-3 and the nAChR co-localize and associate evidenced by immunofluorescence and immunoprecipitation. These results were replicated in differentiated wildtype myotubes treated with the nAChR clustering agent, neural agrin. In differentiated caveolin-3 null myotubes, agrin treatment yields a 60% reduction in nAChR clusters as compared to agrin treated wildtype myotubes. Agrin induces nAChR clustering, through activation of muscle specific kinase (MuSK) and downstream through Rac-1 activation. In differentiated wildtype myotubes, Rac-1 activation peaks at 1 hour of agrin treatment, while in differentiated caveolin-3 null myotubes there is dramatically reduced Rac-1 activation upon agrin treatment. Immunoprecipitation of MuSK shows that caveolin-3 and MuSK association peaks at 1 hour of agrin treatment in wildtype cells. This corresponds to the peak of MuSK phosphorylation which also occurs at 1 hour. Agrin induced MuSK phosphorylation was decreased more significantly than the overall decrease in MuSK expression in the caveolin-3 null cells as compared to the wildtype results. These results indicate a role for caveolin-3 in efficient nAChR clustering. Electromyography studies in anesthetized mice indicated lengthened latencies of the muscle action potential in the caveolin-3 null mice as compared to wildtype mice. There were also decreased overall electromyography (EMG) amplitude and EMG area under the curve in caveolin-3 null mice. Comparison of contractile strength in wildtype and caveolin-3 null animals indicated tetanic contractions to be less stable in the caveolin-3 null animals, though there was late potentiation in actual contractile strength. Lack of caveolin-3 affects the neuromuscular junction formation and transmission without affecting overall contractile strength. This research opens a novel view, that correct neuromuscular junction formation and neuromuscular transmission is important in the development of muscular dystrophies.

Despite their widespread use, the precise molecular actions of anesthetics and alcohol are unknown. Although anesthetics have made surgical intervention palatable, anesthetics are not free of alarming side effects such as anesthetic awareness and post-operative cognitive deficits. While alcohol is consumed for positive effects such as anti-anxiety, euphoria, and relaxation, alcohol consumption also results in adverse behavioral effects such as sedation, motor incoordination, and cognitive impairment. Chronic alcohol consumption results in alcohol withdrawal syndrome, tolerance, and dependence. Understanding the mechanisms of action of anesthetics and alcohol are critical for preventing anesthetic side effects and for developing effective treatments for alcohol abuse and alcoholism. Many putative targets of anesthetics and alcohol have been identified in brain including sodium channels, potassium channels, glutamate receptors, glycine receptors, and especially GABA receptors. Phasic and tonic inhibitory currents mediated by GABA type A receptors are sensitive to modulation by anesthetics and alcohol. α4 subunit-containing GABA A receptors mediate tonic inhibitory currents, are highly sensitive to GABA, and are strongly implicated in the effects of volatile, intravenous, and neurosteroid anesthetics. Global α4 knockout mice showed reduced tonic current that was not potentiated by the volatile anesthetic isoflurane or the neurosteroid anesthetic, alphaxalone. Specific Aim 1 tested the hypothesis that volatile and intravenous anesthetic effects are mediated via a4-containing receptors by comparing behavioral responses to these drugs in wild type and α4 KO mice. Results obtained indicate that while α4-containing receptors are required for the amnestic effects of isoflurane, they are not required for mediating the effects of volatile and intravenous anesthetics on other behavioral endpoints. Interestingly, α4-containing receptors are required for low dose alphaxalone-induced locomotor stimulation, but not high dose effects. α4-containing receptors, when paired with the δ subunit, have been proposed to possess a common binding pocket for ethanol and pharmacologic antagonists of ethanol action. Previous studies in the Homanics lab indicated that the ethanol-reversing effects of Ro15-4513, an imidazobenzodiazepine ethanol-antagonist, were dramatically reduced in α4 KO mice both at the cellular and the behavioral level. Specific Aim 2 tested the hypothesis that a4-containing receptors are required for the ethanol antagonistic effects of RY023, a derivative of Ro15-4513. α4 KO mice showed differential sensitivity to the effects of RY023 in the presence and absence of ethanol on loss of righting reflex and locomotor behavior. We conclude that α4 containing receptors are involved in some intrinsic effects of RY023 but not in the ethanol-antagonistic effects of RY023. This study of α4-containing receptors has advanced our understanding of anesthetic action and eliminated the theory of a unitary target for ethanol antagonism.

Electrophilic lipids are emerging as critical mediators of anti-inflammatory signaling pathways, although many biologically relevant electrophiles may still remain unknown. Nitro derivatives (NO2-FA) and á,â-unsaturated carbonyl derivatives of unsaturated fatty acids are naturally occurring electrophilic products of redox reactions, and can modulate a variety of cellular signaling processes including the transcriptional activity of the peroxisome proliferator-activated receptor-ã (PPARã). PPARã binds diverse ligands to regulate the expression of genes involved in metabolism and inflammation. Activators of PPARã include anti-hyperglycemic drugs such as thiazolidinediones (TZDs) and intermediates of lipid metabolism and oxidation that bind PPARã with very low affinity. Recently TZDs have raised concern after being linked with increased risk of peripheral edema, weight gain, and adverse cardiovascular events. In contrast, NO2-FA act as partial agonists of PPARã at nM concentrations and covalently bind PPARã via Michael addition. NO2-FA show selective PPARã modulator characteristics by inducing coregulator protein interactions distinctively different from those induced by the TZD Rosiglitazone. In further exploring the electrophilic lipidome, a new subclass of electrophilic lipid has been revealed. Using a recently developed â-mercaptoethanol (BME) alkylation reaction, followed by HPLC-MS/MS-based screening, we report six novel electrophilic fatty acid derivatives (EFADs) specifically formed during macrophage activation (RAW264.7 and THP-1 cell lines and primary macrophages treated with IFNã and LPS). The major EFADs are á,â-unsaturated oxo-derivatives of omega-3 fatty acids as confirmed by cell culture and in vitro studies and by MS/MS structural analysis. The isomers of two major EFADs were identified as 13- and 17-keto derivatives of docosapentaenoic acid (DPA) and docosahexaenoic acid (DHA). Purified cyclooxygenase-2 (COX-2) product profiles and treatment of activated macrophages with COX-2 inhibitors confirmed EFAD synthesis to be catalyzed by inducible COX-2, followed by hydroxy-dehydrogenase activity. EFAD production was increased 2.5 fold in activated macrophages treated with acetylsalicylic acid (ASA; aspirin). Internal standard-based quantification showed that EFADs are highly abundant electrophiles in activated macrophages, reaching intracellular concentrations as high as 350 nM. Importantly, EFADs form reversibly reactive covalent adducts with both proteins and small molecule thiols in activated macrophages, supporting a potential for post-translational protein modification-mediated cell signaling. Furthermore, synthetic isomers of EFAD-1 and -2 (17-oxo-DHA and 17-oxo-DPA, respectively) act as partial agonists of PPARã, activate Nrf2 (nuclear factor-erythroid 2-related factor 2)-dependent gene expression, and inhibit pro-inflammatory cytokine production and iNOS expression in IFNã and LPS-activated RAW264.7 cells and in primary macrophages. In conclusion, it has been demonstrated that upon activation macrophages generate omega-3 derived electrophilic signaling molecules at biologically relevant concentrations that act as autocrine mediators.

G-protein coupled receptors (GPCRs) are the largest family of transmembrane proteins, constituting 2% of the human genome. They mediate signaling from a diverse set of ligands, ranging from photons to large peptides. Their intracellular signaling cascades are complex and highly malleable depending on cellular context. Yet, GPCR signaling in vivo is highly specific. Cells maintain this tight control over GPCR signaling through the expression of adaptor proteins. These adaptors regulate GPCR function and activation on many levels - they localize receptors to specific subcellular domains, assemble functional signaling complexes, alter the specificity of G-proteins coupling to the receptor, or regulate receptor traffic to and from the plasma membrane. By balancing the expression of these adaptor proteins, cells control where, when and how long GPCRs signal. Na+/H+ Exchanger Regulatory Factor 1 (NHERF1), also known as Ezrin binding phosphoprotein 50kDa (EBP50), is the prototypical PDZ adaptor protein. Expression of NHERF1 clusters parathyroid hormone type 1 receptor (PTH1R) and frizzled (Fzd) along actin stress fibers in non-polarized cells at the apical actin cap in polarized cells. In addition to proper localization, interaction with NHERF1 has several signaling manifestations. For PTH1R, interaction with NHERF1 can cause either a G-protein switch or can scaffold a PTH1R-PKA-calcium channel signaling complex. For Fzd, interaction with NHERF1 blocks Wnt-induced -catenin activation. NHERF1 knockout mice exhibit PTH-resistant phosphate excretion and enhanced PTH-induced vitamin D synthesis, as well as increased mammary duct density secondary to heightened Wnt-Fzd signaling. NHERF1 is a prime example of how PDZ adaptor proteins regulate GPCR localization and diversify GPCR signaling in physiologically significant ways.

Background Lung cancer is currently the second most prevalent form of cancer in the United States and is the leading cause of cancer-related deaths. Currently, there are no effective therapies for those diagnosed in the later stages of lung cancer. The c-Met receptor is a potential therapeutic target for NSCLC along with its ligand, hepatocyte growth factor (HGF). Signaling interactions between c-Met and the mutant Epidermal Growth Factor Receptor (EGFR) have been studied extensively, but the biological importance of lateral signaling to c-Met in EGFR wild-type tumors is minimally understood. Principal Findings Our observations indicate that wild-type EGFR, the receptor most often found in NSCLC tumors, can initiate delayed c-Met activation in NSCLC cell lines. EGFR ligands induce accumulation of activated c-Met which begins at 8 h and continues for 48 h. This effect is accompanied by phosphorylation of critical c-Met tyrosine residues. Gene transcription is required for delayed c-Met activation; however, phosphorylation of c-Met by EGFR occurs without production of HGF or secretion of other factors, supporting an internal mechanism that is independent of c-Met ligand. Lateral signaling is blocked by two selective c-Met tyrosine kinase inhibitors (TKIs), PF2341066 and SU11274, or with gefitinib, an EGFR TKI, suggesting kinase activities of both receptors are required for this effect. The c-Src pathway is essential for EGFR to c-Met communication. This appears to be up- and downstream of delayed c-Met activation. Pre-treatment with pan-SFK inhibitors, PP2 and dasatinib, abolishes delayed c-Met phosphorylation. A c-Src dominant-negative construct reduces EGF-induced c-Met phosphorylation compared to control, further confirming a c-Src requirement. Additionally, delayed c-Src association with c-Met and prolonged c-Src activation are observed following EGF addition. Inhibition of c-Met with PF2341066 and siRNA decreases the EGF-induced phenotypes of invasion by ~86% and motility by ~81%, suggesting that delayed c-Met activation is utilized by EGFR to potentiate its full biological effects possibly through STAT3. Combined targeting of c-Met and EGFR pathways lead to increased NSCLC xenograft anti-tumor activity. Conclusions and Significance Collectively, these data provide an alternative working model of prolonged EGFR signaling, whereby c-Met activation in NSCLC cell lines initiated by wild-type, non-amplified EGFR with wild-type, non-amplified c-Met maximizes EGFR-induced cell motility and invasion. With the identification of this novel pathway, the studies presented here demonstrate that inhibition of both EGFR downstream signaling and EGFR lateral signaling through the EGFR-c-Src-c-Met axis might be effective in treatment of NSCLC. Taken together, these findings will aid in the future development of combination EGFR and c-Met TKI treatments in clinical trials for NSCLC tumors that are wild-type for both EGFR and c-Met.

Bladder cancer, one of the most frequently diagnosed cancers, is a significant source of morbidity and mortality throughout the world. According to the American Cancer Society (2005), approximately 63,210 new cases will be diagnosed in the United States and bladder cancer will account for nearly 13,180 deaths. The current standard for detection of bladder cancer relies on cystoscopy, an invasive procedure, and cytology. Cytology has a high specificity, but lacks sensitivity in detection of low-grade tumors, as well as requires a trained pathologist for review. Because current diagnostic tools are less than optimal and because bladder cancer has a high rate of recurrence and long term monitoring is a necessity, a better diagnostic tool is needed. There is now a great interest in researching urine markers for bladder cancer. Our lab previously identified six nuclear structural proteins (BLCA 1-6) that are specifically expressed in bladder cancer tissue. The nuclear matrix is the support scaffold of the cell nucleus. This structure has a variety of functions, many of which have implications in cancer progression. The purpose of this dissertation is to examine changes in nuclear structural proteins. The hypothesis we propose is that changes in structural elements of the nucleus are involved in the progression of bladder cancer and can be developed into markers of this disease. Specifically this study had three goals. 1) to determine if BLCA-1 could be developed into a biomarker of bladder cancer, 2) to clone the gene encoding BLCA-1, and 3) to examine functional aspects of BLCA-4. A urine-based immunoassay was developed that can detect BLCA-1 in patients with bladder cancer with a specificity of 87% and sensitivity of 80%. Furthermore, this protein can be detected in serum of individuals with bladder cancer and may associate with the stage of disease. We also demonstrated that BLCA-4 can confer a growth advantage to cells over-expressing this protein. Over-expression of BLCA-4 led to many gene expression changes. BLCA-4 may play a role in bladder cancer pathobiology by altering genes that enhance proliferation and invasion, maintain blood flow for tumor cell survival, or enhance angiogenesis. Finally, we have been successful in cloning part of the cDNA that encodes for BLCA-1 and it appears to have a close homology to a novel metastasis related gene. In summary, this project has demonstrated that bladder cancer specific nuclear matrix proteins can be developed into markers of the disease and may play a functional role in bladder cancer pathobiology.

The utilization of high-throughput -omics strategies, such as proteomics, in the analysis of breast cancer will function to define central molecular characteristics across a disease that is associated with a high degree of molecular heterogeneity. Data reported herein details the investigation of key subjects in breast cancer biology focused on the characterization of endogenous and experimentally-induced disease biology characteristics utilizing the application of LC-MS based proteomic analyses of both in vitro models of breast cancer as well as primary clinical samples. Results include a combined global and functional proteomic strategy to identify governing functional roles for mutually, differentially abundant proteins observed across three divergent cell line models of breast cancer. Further, evidence is presented which provides insights into the regulatory activity of the breast cancer-associated microRNA (miR-145) in several cell line models of breast cancer in which expression of this microRNA has been restored. Lastly, robust analyses are detailed focused on the identification of differential protein characteristics indicative of disease stage as well as of recurrent disease in breast cancer derived from proteomic analysis of formalin-fixed, paraffin embedded (FFPE) clinical samples. These studies contribute to the field of proteomics in the form of 1) providing robust experimental workflows directed towards investigation of functional themes and associated functional targets in large protein data sets 2) detailing strategies for navigating the application of proteomic analysis to microRNA target discovery and 3) further development and utilization of methodologies towards the proteomic analysis of clinical, FFPE tissue samples. Furthermore, these studies benefit the breast cancer community on several fronts including 1) the elucidation of provocative protein candidates which warrant further investigation for their role in regulating disease mechanisms underlying v breast cancer biology and 2) through the discovery of diagnostic markers indicative of discrete subtypes and stages of disease progression in breast cancer. The results reported herein detail disease-specific protein abundance characteristics associated with neoplastic progression in breast cancer that will benefit further expansion of the basic biological understanding of this disease and describes novel proteins for further evaluation as biomarker candidates for the diagnosis of breast cancer.

The Phosphatase of Regenerating Liver (PRL) family, consisting of PRL-1, PRL-2, and PRL-3, is a group of prenylated phosphatases that are candidate cancer biomarkers and therapeutic targets. Individual PRLs are over-expressed in a variety of cancer cell lines and tissues, and elevated PRL expression has been associated with tumorigenesis and metastasis. Although several studies have documented that altered expression of PRL-1 or PRL-3 can influence cell proliferation, migration and invasion, there is an absence of knowledge about the biological functions of PRL-2. Thus, the current study was designed to evaluate the role of PRL-2 in cell migration and invasion in human cancer cells. I found that four lung cancer cells, including A549, over-expressed PRL-2 when compared with normal lung cells. PRL-2 suppression by siRNA or shRNA markedly inhibited cell migration and invasion. PRL-2 suppression by siRNA decreased p130Cas and vinculin expression, increased phosphorylation of Ezrin on tyrosine 146, and decreased ERK phosphorylation upon serum stimulation. There were no significant changes in total p53, Akt and c-Src expression levels or their phosphorylation status, suggesting PRL-2 suppression could inhibit tumor cell migration and invasion through a Src-independent p130Cas signaling pathway. Ectopic expression of wild type PRL-2, a catalytic inactive C101S mutant, and a C-terminal CAAX deletion revealed a requirement for both the PRL-2 catalytic functionality and prenylation site. Expression of wild type but not the mutant forms of PRL-2 caused ERK phosphorylation and nuclear translocation, and promoted tumor cell migration and invasion. These results support a model in which PRL-2 promotes cell migration and invasion through an ERK-dependent signaling pathway. In addition, thienopyridone, a previously reported PRL inhibitor, showed antiproliferative activity in a concentration-dependent manner, and decreased cell migration and invasion. In summary, these studies demonstrate for the first time that PRL-2 regulates cell migration and invasion in non-small cell lung cancer, and I propose that PRL-2 stimulates cell migration and invasion through an ERK signaling pathway.

The majority of current pharmacological treatments for cancer target rapidly dividing cells, a characteristic of most cancer cells. Unfortunately, these treatments also affect cells that normally divide at a rapid rate, such as cells of the digestive tract, hair follicles, and bone marrow, which limits the efficacy of chemotherapy due to toxic side effects. Reducing the drug dose to evade these side effects, however, often impairs efficacy and encourages drug resistance. Therefore, new unbiased approaches are required to identify new drug combinations with existing effective cancer chemotherapeutics. I therefore exploited data from a short interfering RNA (siRNA) high throughput screen targeting 5,520 unique druggable genes, which comprises gene products that are theoretically good targets for drug development. I used the siRNA screening methodology to identify novel combination chemotherapies for the treatment of glioblastoma multiforme (GBM), the most common and aggressive form of human primary brain tumors. My hypothesis is that unrecognized chemosensitivity nodes exist for the microtubule destabilizing agent vinblastine. GBM cells were treated with a sub-lethal concentration of vinblastine and identified gene products that sensitized cells to vinblastine. Using a series of statistical methods, followed by target identification assays, I found gene products that sensitized GBM cells to vinblastine, implicating siRNA screening technology as an efficient, unbiased method for identifying potentially novel anticancer treatments.