LIS1 and NudE(L) are partner proteins in a conserved pathway regulating the function of dynein and microtubules. Members of the LIS1 pathway play a critical role in fundamental cellular processes, such as differentiation, proliferation, and migration. Therefore, determining the precise nature of their actions will lead to significant insight into the complex process of embryogenesis. In this document, I present data revealing that cytLEK1, a large protein containing a spectrin repeat and multiple leucine zippers, is a novel component of this pathway through its direct interaction with NudE, as determined by a yeast two-hybrid screen. This is the first time that the cytLEK1 protein has been linked to a molecular pathway. I identified the exact binding domains in each molecule, and co-immunoprecipitation and colocalization studies confirmed the specificity of the interaction between cytLEK1 and NudE. Confocal deconvolution analysis revealed that cytLEK1 exhibits colocalization with endogenous NudE and with the known NudE binding partners, LIS1 and dynein. By localizing the NudE-binding domain of cytLEK1 to a small domain within the molecule, I was able to disrupt cytLEK1 function using a dominant negative approach, in addition to LEK1 knockdown, and thus examine the role of the cytLEK1-NudE interaction in cells. Consistent with a defect in the LIS1 pathway, disruption of cytLEK1 function resulted in alteration of cellular morphology and microtubule organization. Additionally, cells exhibited a severe inability to repolymerize their microtubule networks after nocodazole challenge. I also present data here examining, for the first time, the expression and localization of cytLEK1 in various cells and tissues. These experiments provide further evidence suggestive of the action of cytLEK1 as a regulator of the LIS1 pathway. Taken together, my studies reveal that cytLEK1 is essential for cellular functions regulated by the LIS1 pathway and may play a critical role during murine development.

Hypoxia inducible vascular endothelial growth factor (VEGF) plays a major role in initiation and regulation of retinal neovascularization, which is the leading cause of severe vision loss and irreversible blindness in developed countries. Src family kinases (SFKs) are involved in a broad spectrum of cellular events. However, their roles in VEGF-mediated pathological retinal angiogenesis are completely unknown. My investigation showed that in vitro SFKs were essential for hypoxia-induced VEGF expression in retinal glial Müller cells, a major source of VEGF secretion during the pathogenesis of retinopathy, and for VEGF signaling in retinal microvascular endothelial cells (RMECs). However, neither process required phosphorylation of the SFK activation loop Tyr416. In addition, in RMECs, coexpressed SFK members Src, Fyn and Yes each displayed distinct properties in the regulation of VEGF-mediated cell events. All three kinases were required for VEGF mitogenic signaling. VEGF-induced cell migration was significantly increased in Fyn-deficient cells and decreased in Yes-deficient cells. Interference of Fyn, but not Src or Yes, impaired VEGF-induced tube formation in RMECs. In vivo, in a rat model of oxygen-induced retinopathy (OIR), I found that a significant increase of SFK Tyr416 phosphorylation was specifically associated with pathological retinal angiogenesis, but not with physiological intraretinal vascularizaion. Müller cells were the source of the elevated phospho-SFK Tyr416 signal. VEGF expression was also highly increased in these OIR retinas. Intravitreous injection of a selective SFK inhibitor (PP2) significantly reduced retinopathy. These findings indicate that aberrant SFK signaling may be an important factor in the pathogenesis of retinal neovascularization. Increased SFK activity or individual SFK member(s) are potential targets for therapeutic intervention in VEGF-mediated retinopathy.

Progression through the eukaryotic cell cycle involves the coordinated activation and inactivation of cyclin dependent kinases. One such family member, Cdk1p in a complex with Cyclin B, is activated at the G2/M transition in all eukaryotes and initiates the chain of events that ultimately lead to the phenotypic changes that occur during the progression through mitosis. The inactivation of Cdk1p-cyclin B at mitotic exit must be coordinated with respect to chromosome segregation to link nuclear and cytoplasmic divisions. This inactivation not only involves irreversible destruction of Cyclin B, but also a reversal of Cdk1p dependent phosphorylation events by the conserved Cdc14 family of protein phosphatases. We have identified the Cdc14 family member, Clp1p, in the fission yeast Schizosaccharomyces pombe, and studied how the disruption of Cdk1p phosphorylation events influences the coordinated inactivation of mitotic Cdk1p activity. Clp1p specifically disrupts the Cdk1p self-sustaining amplification loop involving Cdc25p activation and Wee1p inactivation at the exit from mitosis. This disruption is temporally regulated to occur only after Cdk1p activity wanes, involving a mechanism of direct inhibition of Clp1p phosphatase activity through Cdk1p dependent phosphorylation during early mitosis. Clp1p also participates in an auto-amplification loop during mitotic exit as it auto-catalytically reverses the inhibitory phosphorylation events to increase its activity, and prevents further inhibition of its activity by attenuating Cdk1p activity. Together, these findings point to a simple regulatory circuit that couples Cdk1p activation with its inactivation mediated through regulation of Clp1p phosphatase activity.

The study of cytokinesis in fission yeast S.pombe has revealed a signaling network, the septation initiation network (SIN) that serves to coordinate cytokinesis with mitotic exit in the fission yeast Schizosaccharomyces pombe. The timing of cytokinesis and septum formation hinges upon the activation of the GTPase Spg1. Given the central role of Spg1 GTPase in regulating this pathway we investigated the role of Cdc7 its effector, and Byr4 the GAP component that serves to inactivate it. In order to elucidate Cdc7 regulation I have identified various functional domains within Cdc7. I show that a region adjacent to the kinase domain is responsible for association with Spg1 but insufficient for SPB localization. In addition I find that Cdc7 self associates which is required for binding to Spg1. Our findings indicate that there are potentially multiple modes of regulating Cdc7 function and hence the SIN. I have investigated the regulation of the GAP component Byr4 by phosphorylation as a means of regulating Spg1activation and further characterized its interaction with Spg1 and Cdc16. Together these results point to modes of regulating the activation of Spg1 and its effector as a means of controlling septum formation and cytokinesis.

The work contained in this document provides the first description and characterization of Bves, a gene product identified by our laboratory, in epithelial cell types. Studies addressed a central hypothesis that Bves plays a role in epithelial cell-cell interaction and integrity, which is supported by an initial protein distribution analysis at cell membranes in both cells and tissues. Localization, interaction, and functional assays establish that Bves is an essential component of the tight junction complex and interacts with the protein complex containing ZO-1 in epithelial cells. Furthermore, in vivo experiments demonstrate that Bves is clearly important for epithelial morphogenesis during X. laevis development, and establish an excellent model system to perform future studies. This dissertation provides an important contribution to the body of literature on the Bves by defining the essential nature of Bves in epithelial integrity and by providing the groundwork for further examination of how Bves participates in epithelial cell interaction.

Dynamic spatiotemporal expression of the nodal gene and its orthologs is involved in the dose-dependent induction and patterning of mesendoderm during early vertebrate embryogenesis. In loss of function studies, a strong knockdown of Xenopus antivin/lefty (Xatv/Xlefty) function was achieved by coinjecting translation- and splicing-blocking morpholino oligonucleotides that target both the XatvA and XatvB alloalleles. A greater expansion of the Organizer and mesendoderm tissues in my studies than noted in previous Xlefty/Xatv knockdown experiments in Xenopus, with long-term maintenance of expanded axial tissues strongly suggests that the regulation of Xenopus nodal-related (Xnr) signaling by Xlefty/Xatv is essential for proper cell layer specification during early embryogenesis, and tissue patterning at late stages. Inhibitors specific to Xnr signaling were used to provide evidence that Xnr-mediated induction was inherently long-range in the Xlefty/Xatv-deficient embryo, essentially being capable of spreading over the entire animal hemisphere. While the expansion of Xnr1 and Xnr2 expression is limited to the marginal zone by Xlefty/Xatv deficiency, inhibition of Xbra function using Xbra-EnR in Xlefty/Xatv-deficient embryos caused a much larger increase in the level and spatial extent of Xnr expression. However, Xnr2 expression was constrained to the superficial cell layer in any experimental conditions, suggesting a fundamental tissue-specific competence to express Xnrs. These studies reveal a two-level suppressive mechanism for restricting the strength, range and duration of Xnr signaling via both Xlefty/Xatv-mediated extracellular inhibition and Xbra-mediated indirect transcriptional repression. Therefore, the tight regulation of Xnr signaling and expression by multiple influences is essential for precisely refining cell layer specification and patterning during gastrulation. Conserved Nodal/Lefty/Pitx2 cassette in left LPM has been implicated in left-right (L-R) axis patterning during post-gastrulation and later asymmetric morphogenesis. Detailed genetic pathways for these processes, however, are still elusive because of lack of focused studies on L-R development. Under this rationale, I tried to identify new genes involved in L-R pathways using several screening methods such as PCR-based subtractive cDNA screening and microarrays. Though any candidates were not isolated in these studies, results suggest that an unbiased and rapid screening with Affymetrix GeneChipsË seems to be an appropriate method for isolating new molecules involved in L-R patterning because of its reliable detection of controls, such as Xnr1 and Xlefty/Xatv.

Proper function of the brain requires that neurons adopt different morphologies and connections. In the nematode <i>C. elegans</i>, VA and VB motor neurons arise from a common precursor cell but adopt different morphologies and accept input from separate sets of command interneurons. In <i>unc-4</i> mutants, VA motor neurons are miswired with VB-type inputs. We have proposed that miswiring results when VB genes are ectopically expressed in the VAs in <i>unc-4</i> mutants. Previous work revealed that UNC-4 functions with the UNC-37/Groucho co-repressor protein to repress the VB-specific genes <i>acr-5, del-1, glr-4</i>. However, our genetic data rule out roles for these VB genes in synaptic choice. To identify the missing <i>unc-4</i> target genes, a microarray-based strategy for profiling VA motor neurons was adopted. A comparison of VA-specific transcripts isolated by mRNA-tagging from wildtype and <i>unc-37</i> mutant animals revealed ~250 upregulated transcripts in <i>unc-37</i> animals. One of these genes, <i>ceh-12</i>, is the <i>C. elegans</i> homolog of HB9, a homeodomain transcription factor with conserved roles in motor neuron fate in flies and vertebrates (Arber et al 1999, Broihier and Skeath 2002). In <i>C. elegans, ceh-12</i>::GFP is exclusively expressed in VB motor neurons in wildtype animals. In <i>unc-4</i> and <i>unc-37</i> mutants, <i>ceh-12</i>::GFP is also expressed in VA motor neurons as suggested by the microarray data. Thus, CEH-12 is a strong candidate for an UNC-4 target gene that regulates synaptic choice. To test this idea, the <i>unc-4</i> promoter was used to drive CEH-12 expression in wildype VA motor neurons. These animals exhibit an Unc-4 like backward movement defect, as expected for a model in which ectopic CEH-12 is sufficient to impose VB type inputs. In addition, we also showed that <i>ceh-12</i> deletion mutants are partial suppressors of Unc-4 movement, thereby confirming that CEH-12 is also required for the Unc-4 miswiring defect. We conclude the VB-specific gene, <i>ceh-12</i>, is normally repressed in VA motor neurons to prevent the imposition of VB-type inputs. The incomplete suppression of <i>unc-4</i>, however, suggests that UNC-4 also controls other downstream target genes that function in parallel pathways to regulate synaptic choice.

Prostaglandin G/H synthases (PGHS), commonly referred to as cyclooxygenases (COX-1 and COX-2), catalyze a key step in the synthesis of biologically active prostaglandins (PGs), the conversion of arachidonic acid (AA) into prostaglandin H₂ (PGH₂). PGs have important functions in a variety of physiologic and pathologic settings, including inflammation, cardiovascular homeostasis, reproduction, and carcinogenesis. However an evaluation of prostaglandin function in early mammalian development has been difficult due to the maternal contribution of prostaglandins from the uterus. The emergence of zebrafish as a model system has begun to provide some insights into the roles of this signaling cascade during vertebrate development. In zebrafish, COX-1 derived prostaglandins are required for two distinct stages of development, namely during gastrulation and segmentation. During gastrulation, PGE₂ signaling promotes cell motility, without altering the cell shape or directional migration of gastrulating mesodermal cells via the G-protein coupled prostaglandin E₂ receptor (EP4). During segmentation, COX-1 signaling is also required for posterior mesoderm development, including the formation of vascular tube structures, angiogenesis of intersomitic vessels, and pronephros morphogenesis. We propose that deciphering the role for prostaglandin signaling in zebrafish development could delineate mechanistic details underlying various disease processes that result from perturbation of this pathway and uncover novel potential therapeutic targets.

The lining of the gastrointestinal tract, a single layer of epithelial cells, forms a dynamic barrier that protects the organism from harmful substances present in the lumen. Maintenance of a healthy gastrointestinal tract requires a delicate balance of proliferation, migration, and apoptosis. In response to damage of the intestinal epithelium, cells immediately surrounding the wounded area rapidly migrate into the denuded area to close the wound and restore the integrity of the epithelium in a process called restitution. A number of growth factors found in the gastrointestinal tract accelerate restitution in in vitro models of intestinal injury. Several of these factors are currently being tested in clinical trials as therapeutic treatments for inflammatory bowel diseases, of which the most promising so far is epidermal growth factor (EGF). However, the molecular mechanisms by which these factors activate intestinal epithelial cell migration are poorly defined. Using an in vitro model of intestinal epithelial cell wound closure developed in our laboratory, we have previously shown that EGF treatment increases intestinal epithelial cell migration to accelerate wound closure. In this dissertation work I have explored the molecular mechanism by which EGF stimulates intestinal epithelial cell migration. The experimental data presented here demonstrate that the small GTPase Rac is required for EGF stimulated cell migration. Treatment of intestinal epithelial cells with EGF rapidly activates Rac and enhances lamellipodia formation at the wound margin. EGF receptor (EGFR) kinase activity is required for increased cell migration and Rac activation in response to EGF. Activation of Src family kinases and phosphoinositide 3-kinase by EGF are required for Rac activation, demonstrating a novel signaling pathway downstream of EGFR that contributes to intestinal epithelial cell migration. These data contribute to our knowledge of the complex molecular mechanism that control intestinal epithelial repair with implications for understanding the role of growth factor regulation of this small GTPase in intestinal cellular migration and differentiation programs in vivo.

Constitutively activating mutations of members of the RAS family of small G proteins provide an important oncogenic contribution to a significant percentage of human cancers. Despite nearly 25 years of investigation, important questions about the function of these proteins and their relationships to one another remain unresolved. We have undertaken a series of studies to address some of these questions and herein report novel findings regarding the relationship of these family members to one another and their respective contributions to the progression of human disease. Briefly, we examine the uniquely potent oncogenic contribution of K-RAS to cell lines derived from human tumors and identify some previously unappreciated findings regarding the mechanisms by which this protein promotes tumorigenesis. Interestingly, we present data to support that K-RAS requires the activity of another family member, NRAS to realize its full oncogenic potential in certain contexts, and that represents the first biochemical evidence of a relationship between RAS family members. We additionally challenge earlier notions about effector pathways long thought to be essential for RAS-driven transformation, as well as disprove an accepted and widespread assay for measuring the activity of RAS proteins.