Nef is a critical HIV-1 accessory factor shown to promote viral pathogenesis by altering host cell signaling pathways. Nef has been shown to bind several members of the Src family of protein-tyrosine kinases, and these interactions have been implicated in the pathogenesis of HIV/AIDS. The studies summarized below investigated this key interaction between virus and host cell proteins. We explored the direct effect of Nef interaction on Src family kinases (SFKs) using Saccharomyces cerevisiae, a well-defined system in which c-Src expression arrests yeast cell growth in a kinase-dependent manner. The seven SFKs found in HIV target cells wre expressed in yeast; each was found to be active alone, but repressed by co-expression of the negative regulatory kinase Csk. We then co-expressed each SFK with both Csk and HIV-1 Nef and found that Nef selectively activated Hck, Lyn, and c-Src among SFKs. We then used our yeast-based system to identify small molecule inhibitors of the active Nef:Hck complex using the auto-dowregulated Hck-YEEI molecule. Yeast expressing the Nef:Hck-YEEI complex were used to screen a library of small heterocyclic compounds based on their ability to rescue growth inhibition. Two compounds identified in this screen potently blocked Nef-dependent HIV replication, indicating Nef:SFK complexes as valid targets for anti-HIV drug therapy. Finally, we used the yeast assay to identify novel mechanisms of Nef:SFK interactions. We screened a panel of primary Nef alleles containing the known SH3-binding elements and discovered four alleles whose proteins demonstrated altered activation of SFKs. Sequence examination revealed the existence of amino acid changes in regions not previously suspected to be involved in SH3-mediated interaction. Particularly intriguing are residues in a large unstructured loop that projects from the Nef core. These findings suggest that critical residues outside of the known SH3-binding motifs may affect SFK binding and activation. Together, the results presented here advance the field of HIV research by furthering our understanding of the interaction between the HIV-1 Nef virulence factor and the Src kinase family, as well as validating this virus:host cell interaction as a rational target for anti-HIV drug discovery.

Deficiency of DNA repair has been shown to lead to cancer in multiple human disease states. We therefore hypothesized that deficiency of Nucleotide Excision Repair is involved in the etiology of breast cancer. This work was carried out using a novel in vitro culture system for both normal and cancerous breast tissue developed in the Latimer laboratory. Primary cultures of non-diseased and tumor tissue were utilized in functional studies, and cell lines generated from these primary cultures were used for expression analysis. Cell lines were generated from the Latimer primary culture system without the use of exogenously transforming agents. Using the Unscheduled DNA Synthesis Assay, we have shown that stage I and stage II breast tumors are deficient in functional NER capacity as compared to non-diseased controls. Analysis of the steady-state mRNA expression levels of 20 genes in the NER pathway using an RNase Protection kit showed that 7 genes (CSB, XPA, XPB, TFIIHp52, TFIIHp44, TFIIHp34, and Cdk7) also showed a loss of expression in early stage tumor cell lines. Microarray analyses corroborated these trends and protein expression was also lost in early stage tumor cell lines for the XPA and CSB genes. In contrast to the loss of DNA repair capacity manifested in early stage tumors, stage III tumors showed an increase in repair capacity. Gain of mRNA expression and protein expression was seen for our candidate genes in late stage tumors. Because loss of expression was reversible and occurred in multiple genes and cell lines, we hypothesized that the mechanism responsible was epigenetic regulation of gene expression. Two early stage cell lines that demonstrated loss of expression of these candidate genes were treated with the demethylating agent 5-aza-2deoxycytidine, resulting in the re-expression of 3 genes: TFIIHp52, TFIIHp34, and Cdk7. In contrast, two of the candidate genes that were not reactivated by azacytidine treatment (CSB and XPA) showed a 7-element transcription factor framework that may begin to explain their coordinate regulation. The functional loss of NER capacity in early stage tumors is consistent with a transcriptional regulatory mechanism including both methylation and other transcription factor based functions.

Infections with high-risk human papillomaviruses (HPVs) have been implicated in the pathogenesis of cervical carcinoma and a subset of squamous cell carcinoma of the head and neck (SCCHN). In this study, we compared the cellular gene expression profiles of HPV16-positive and HPV-negative oropharyngeal carcinomas with those of the normal oral epithelium. Using a high-density oligonucleotide microarray containing 22,215 human transcripts, we showed that 397 and 162 genes were differentially expressed in HPV16-positive and HPV-negative SCCHN, respectively, compared to the normal oral epithelium. Our studies also identified 59 differentially expressed genes in HPV16-positive SCCHN as compared to both HPV-negative SCCHN and normal oropharyngeal tissues. Such up-regulated genes included those involved in nuclear structure and meiosis (SYCP2), DNA repair (RFC5), and transcription regulation (ZNF238). Genes involved in proteolysis (KLK8) and signal transduction (CRABP2) were found to be down-regulated in HPV-positive SCCHN. Our results reveal specific gene expression patterns in HPV16-positive oropharyngeal squamous carcinomas and suggest that HPV infection could play an important etiologic role in these tumors. In another study using the same high-density microarray platform, we have analyzed the cellular gene expression profiles of five HPV-16 and two HPV-18 positive cervical cell lines, one HPV-negative cervical carcinoma cell line, and normal cervical tissue. Our results showed that 877 and 536 genes were differentially expressed in the HPV-positive cell lines compared to the normal cervix tissue and the HPV-negative cervical carcinoma cell line C-33A, respectively. We also found that a total of 57 genes were differentially expressed in the HPV-positive cell lines as compared to both the normal cervix and C-33A. Differentially expressed genes including those involved in cell proliferation such as the L-type amino acid transporter 1 (LAT1, also known as SLC7A5) and gene expression regulation like nucleosome assembly protein 1-like 3 (NAP1L3) were found to be affected for the first time in cervical cell lines. In situ hybridization of LAT1 and NAP1L3 mRNA performed using tissue-arrays (containing ~50 different cervical tumor samples per slide) showed that these genes are also affected in their expression in tumor tissues. These results could lead to the identification of new cellular pathways affected by the presence of HPV in cervical cells. We have also carried out studies to determine whether the expression of human microRNAs (miRNAs; small non-coding RNAs that have the ability to regulate gene expression) are affected by the presence of HPV DNA. For this purpose, we analyzed the expression of miRNAs in HPV-16 positive cervical cell lines and tissues. Twenty-seven miRNAs were differentially expressed in cervical cell lines containing integrated HPV-16 DNA compared to the normal cervix, while, only 6 miRNAs were differentially expressed in a cell line containing episomal HPV-16 DNA compared to the normal cervix. Furthermore, 10 miRNAs were affected in their expression in cell lines containing integrated HPV-16 DNA compared to C-33A. Interestingly, microRNA-218 (miR-218) was specifically underexpressed in cell lines, cervical lesions and cancer tissues containing integrated HPV-16 DNA as compared to both the HPV-negative cell line C-33A and the normal cervix. Expression of the HPV-16 E6 oncogene in transfected cells reduced miR-218 expression, and conversely, RNA interference of E6/E7 oncogenes in an HPV-16 positive cell line increased miR-218 expression. We also showed that miR-218 expression parallels that of the tumor suppressor gene SLIT2 whose intron encodes miR-218. Furthermore, exogenous expression of miR-218 in HPV-16 positive cell lines decreased expression of the epithelial-specific gene LAMB3 which is involved in cell migration and tumorigenicity. These findings demonstrate specific regulation of cellular miRNAs in the presence of an HPV oncogene and may contribute to a better understanding of molecular mechanisms involved in cervical carcinogenesis.

Apoptosis, or programmed cell death, is a process of crucial importance for maintaining a homeostatic balance between cell proliferation and death. In the present study a new mathematical model is presented that draws attention to the possible occurrence of bistability in mitochondria-dependent apoptotic pathways, as well as a transition from bistable to monostable behavior -either apoptotic or cytoprotective, under well-defined conditions. Bistability is proposed to be conferred by positive feedback loops that enhance caspase-3 activation pathways through mitochondria and by kinetic cooperativity in the formation of an apoptosome complex. It essentially ensures that cells will not die in the presence of relatively small pro-apoptotic effects, but will undergo apoptosis when perturbing conditions or levels of pro-apoptotic agents exceed certain threshold values. The passage from bistable to monostable cytoprotective behavior i.e., resistance to apoptosis, may be induced by decreasing the levels of Bax, a pro-apoptotic enzyme, in agreement with experimental observations; while the opposite passage to a pro-apoptotic monostable state may be triggered by a change in the levels of mitochondrial permeability transition pore complexes (PTPCs). Further computations shed light on the origins of the experimentally observed dichotomous effects of nitric oxide (NO), demonstrating that the relative concentrations of anti- and pro-apoptotic reactive NO species, and the interplay of glutathione, dominate the cell fate at long times (of the order of hours). Transient apoptotic effects may be observed in the presence of high levels of intracellular non-heme iron, the duration of which may reach up to hours, despite the eventual convergence to an anti-apoptotic state. The computational results thus point to the importance of the precise timing of NO production and external stimulation in determining the eventual pro- or anti-apoptotic role of NO. The same mathematical model (network of interactions) applied with different model parameters to different cell types demonstrates that cells with high levels of intracellular non-heme iron are resistant to apoptosis while those subjected to high levels of superoxide undergo pathological death, consistent with experimental observations.

Sall1 is a zinc finger containing putative transcription factor that is robustly expressed during mammalian embryogenesis. In humans, the developmental disorder Townes Brocks Syndrome is associated with mutations in the SALL1 gene. Sall1-deficient animals die at birth due to kidney deficits; however, its function in the nervous system has not been characterized. During embryonic development Sall1 is expressed by cortical progenitor cells. In late embryonic and postnatal stages Sall1 is expressed by oligodendrocytes and glia, as well as regions of adult neurogenesis. In the absence of Sall1 forebrain derived structures, the cerebral cortex and olfactory bulbs, were specifically decreased in size, while no gross alterations in midbrain development was observed. This study investigated the cellular mechanisms of Sall1 function in the developing cortex. Alterations in progenitor cell number and the rate of neuronal differentiation were observed during cortical development in Sall1-mutant animals compared to controls. In the absence of Sall1 more cells are committed to early-born cortical structures, at the expense of the progenitor population. In addition, from mid-neurogenesis fewer cells are committed to later-born structures. Together, these findings suggest that early in development Sall1 promotes a progenitor state, and from mid-neurogenesis Sall1 promotes a neural fate. Furthermore, I propose that Sall1 regulates the transition from an early cortical progenitor cell to an intermediate progenitor cell. These findings suggest that Sall1 regulates cortical neurogenesis and progenitor cell maturation in the developing cortex. Sall1 is also expressed by peripheral and central components of the developing olfactory system. Alterations in neurogenesis and mitral cell production were observed in Sall1-deficient olfactory bulbs. In addition, the olfactory nerve failed to extend past the ventral-medial region of the olfactory bulb in Sall1-mutant animals. Intrinsic patterns of neurogenesis were observed during olfactory development in control animals and in Sall1-mutant animals, these patterns of neurogenesis were disrupted. These findings suggest a role for Sall1 in regulating neuronal differentiation and maturation in the developing olfactory system. Together, these findings suggest a conserved role for Sall1 in regulating neurogenesis and cellular maturation in the developing cortex and olfactory bulb.

SIRT1 is a NAD+ dependent deacetylase that targets many histone and non histone proteins, thereby regulating a broad range of physiological processes such as metabolism, reproduction, development, and cell survival. In this study, we have identified novel binding partners for SIRT1 and documented three different cellular processes that are affected by these novel interactions. Using a yeast two-hybrid screen, we have identified several potential binding partners for SIRT1. Transducin like Enhancer of split 1 (TLE1) and eIF2-alpha (eIF2) are two such proteins identified in the screen whose interaction with SIRT1 was further confirmed by co-immunoprecipitation. TLE1 is co-repressor for several transcriptional factors including NF-κB. We demonstrate that SIRT1 and TLE1 repress NF-κB activity and that the catalytic activity of SIRT1 may not be critical for this. Using knock-out cell lines, we further demonstrate that both SIRT1 and TLE1 are required for the down-regulation of NF-κB activity. Our results suggest that the interaction between SIRT1 and TLE1 is important for mediating repression of NF-κB activity, potentially through a deacetyalse independent mechanism. SIRT1 protects cells from genetoxic and oxidative stress, whereas phosphorylation of eIF2 is critical for translation attenuation and preferential expression of stress related genes under stress conditions. We demonstrate that SIRT1 depleted cells show higher levels of phosphorylated eIF2 and delayed expression of the stress response protein, CHOP. Furthermore, SIRT1 deficient cells show higher sensitivity to stress treatments and a delayed recovery of protein systhesis. SIRT1 associates with eIF2 regardless of stress condition, SIRT1s catalytic activity or the phosphorylation state of eIF2. These observations suggest a novel aspect of SIRT1 mediated regulation of cellular stress response. Both SIRT1 and the target of rapamycin (TOR) are involved in age related diseases and lifespan. We demonstrate for the first time that these two pathways are interconnected. We show that SIRT1 null mouse embryonic fibroblasts (MEFs) have larger cell morphology and upregulated mTOR signaling. Furthermore, SIRT1 activator reduces, whereas inhibitor activates the mTOR pathway. Rapamycin is effective in inhibiting mTOR activity in both SIRT1 positive and deficient cells. Finally, we show that SIRT1 physically associates with TSC2 in HeLa cells. These observations demonstrate that SIRT1 negatively regulates mTOR pathway upstream of mTOR complex-1 (TORC1), potentially, by regulating the TSC1/2 complex.

Rhodopsin is the best characterized member of the large, pharmaceutically important, family of G-protein-coupled receptors (GPCRs), and serves as a prototype for understanding GPCR activation. In this thesis, we aim at understanding the activation mechanism of rhodopsin. To this aim, we first performed an in-depth analysis of the conformational motions of rhodopsin predicted by two elastic network models, Gaussian Network Model (GNM) and Anisotropic Network Model (ANM). We compared these motions with the extensive amount of experimental data, and developed a model for rhodopsin activation. We tested the model with Meta II fluorescence decay rates measured to characterize the deactivation of rhodopsin mutants. We find that our results correctly predict 93% of the experimentally observed effects in 119 rhodopsin mutants for which the decay rates and misfolding data were measured, including a systematic analysis of Cys->Ser replacements. Next, in order to incorporate atomic details and the effects of membrane and water molecules into our model, we developed a new protocol named ANM-restrained molecular dynamics (MD). In this protocol, we used multiple ANM modes as restraints to guide MD simulations. By using this protocol, we were able to sample biologically relevant, large scale motions of the protein that are otherwise not accessible to the conventional timescales MD simulations. Furthermore, we explored the evolution of the multiple ANM global modes with realistic deformations favored by a detailed atomic force field in the presence of the explicit environment. Remarkably, with this method, we identify a highly hinge site, which does not change with several rounds of applying normal modes as restraints. This hinge site includes residues that are directly affected by the isomerization of retinal, as well as those stabilizing the resulting all-trans conformation of the chromophore. The CP ends of the helices H3, H4, H5, and H6 and the connecting loops are found to enjoy an enhanced mobility facilitated by this hinge site. Several new interactions are observed to contribute to the mechanism of signal propagation from the retinal binding pocket to the G-protein binding sites in the CP domain.

Lim1 has been reported to regulate the head, nerve system and kidney development in mouse, zebrafish and Xenopus. In the present study, we demonstrate that Xlim1 plays a pivotal role in the pronephric specification and development in Xenopus. In Xenopus embryos at 4-cell and 8-cell stage, loss of function experiments using DEED depletion showed that down-regulation of Xlim1 expression severely impairs the formation of pronephros, assessed by pax8 expression. Overexpressing Xlim1 by injecting constitutively active Lim1 constructs in Xenopus embryos induce remarking enlargement of pronephros at various stages during development, assessed by both in situ hybridization and immunohistochemistry staining. Furthermore we studied the working window of Xlim1 on pronephric specification in mesoderm and observed the peak time point is at stage 10.5 while the whole working window is between stage 9 and 12.5. To understand the mechanism of Xlim1s regulation in kidney development, then we performed the microarray analysis on the induced pronephric RNA sample extracted from animal caps, targeting couple of genes that either are the members involved in the lim1 pathway or in the networking that induce the pronephric specification. In summary, our data suggest that Lim1 acts as a potent factor that determines the early specification of pronephros. Its expression levels at the early stage critically influences the specification of the pronephric tubules. More knowledge about XLim1s pathway in kidney development will be studied via examining the gene targets screened from microarray analysis.

Snf1/AMPK family members mediate the nutrient stress response in eukaryotes. In yeast, Snf1 is required for survival during glucose deprivation. There are three different Snf1 isoforms; each contains the catalytic alpha subunit, one of three regulatory beta subunits, and the regulatory gamma subunit. In glucose limitation, Snf1 is stimulated by activation loop phosphorylation by three partially redundant Snf1-activating kinases (SAKs), Sak1, Tos3, and Elm1. In glucose abundance, the PP1 phosphatase Glc7 dephosphorylates Snf1. We investigated the relationships between the three SAKs and three Snf1 isoforms, the roles of the SAK nonkinase domains, and the mechanism(s) controlling glucose-regulated Snf1 phosphorylation. Since Snf1 assembles into three complexes, the possibility existed that each SAK is dedicated to phosphorylating a single Snf1 isoform. To test this, we generated strains lacking different combinations of SAKs and beta subunits. While our results indicate that the SAKs are not dedicated to specific Snf1 isoforms, each SAK exhibited distinct abilities to activate Snf1, depending on the beta subunit and stress imposed on cells. Domain deletion analysis was employed to characterize the SAKs nonkinase domains. Deletion of Sak1 and Tos3 (but not Elm1) C-termini diminished the ability to activate Snf1. Deletion of the Elm1 C-terminal domain abrogated Elm1s participation in morphogenetic signaling. Thus, the SAK C-terminal domains contribute to pathway specificity. Additional deletion mutants revealed an N-terminal Sak1 motif essential for Snf1 signaling. These data demonstrate that nonkinase N- and C-terminal domains are critical for SAK pathway specification. Snf1 phosphorylation status is regulated by integration of the reaction rates of phosphorylation by the SAKs and dephosphorylation by Glc7. We sought to determine which reaction(s) is glucose-regulated. SAKs immunoprecipitated from extracts of cells grown in high- and low-glucose conditions exhibited similar activity toward Snf1, suggesting the SAKs exhibit constitutive activity. Snf1 dephosphorylation, however, is glucose-regulated. When de novo phosphorylation was inhibited, Snf1 phosphorylation was stable in low glucose but rapidly lost upon glucose addition. Glc7 catalytic activity is not glucose-regulated since another Glc7 substrate was dephosphorylated in both glucose conditions. Therefore, Snf1 phosphorylation status is controlled by Snf1s availability to serve as a Glc7 substrate.

MSUD is a serious liver-based metabolic disorder caused by a deficiency in the branched-chain alpha-ketoacid dehydrogenase (BCKDH) complex. Resulting branched-chain amino acid (BCAA) accretion in the body mainly affects the brain, which in most cases results in permanent neurological dysfunction or death without life-long attentive care. Recently it was shown liver transplantation alone restored BCKDH to a level sufficient to correct MSUD. To test novel therapies, a mouse model of intermediate MSUD (iMSUD) was created (Homanics et al., 2006), which mimicked human iMSUD. Therefore, this dissertation focused on the investigation of liver-directed therapeutic approaches to correct MSUD. In the first aim, iMSUD mice were further characterized and determined to closely mirror the human disease phenotype. iMSUD mice suffered from developmental delay, seizures, and altered brain amino acid and neurotransmitter concentration. iMSUD brains also displayed histological abnormalities while liver morphology was normal. In the second aim, adeno-associated viral (AAV) vectors were used to deliver E2 to the liver. However, no significant improvement was determined in AAV-treated iMSUD mice compared to controls. The most likely reasons this study was unsuccessful were low treatment dose, a weak albumin promoter, and possible competition and interaction between AAV-derived and iMSUD transgene-derived E2. The third aim focused on hepatocyte transplantation (HTx). iMSUD-HTx mice had a 75% reduction in BCAA/alanine levels compared to iMSUD controls. BCKDH activity was increased, and Real Time qPCR detected donor-derived E2 in the liver. Dopamine and serotonin, along with several related metabolites, were corrected to control levels. Body weight at weaning and survival were also significantly improved in iMSUD-HTx mice. The fourth aim focused on differentiated embryonic stem cell (ESC) transplantation. Differentiated ESCs expressed liver-specific markers after 3 days in culture and BCKDH activity was significantly increased over undifferentiated ESC populations. Liver-like ESC engraftment was verified up to 1 month following transplantation into wildtype mouse liver. In summary, iMSUD mice were determined to be a superior model to test novel liver-directed therapies. Our findings of partial metabolic correction of iMSUD in a mouse model by HTx were very encouraging. Therefore, liver-directed therapeutic intervention for human MSUD should be investigated further.