Cytoskeletal motor proteins utilize the energy from ATP turnover to perform work along their filament tracks. Eg5 is a microtubule-based motor of the Kinesin-5/BimC family that is essential for bipolar spindle formation during eukaryotic cell division. I have performed a detailed kinetic analysis to define the minimal Eg5 ATPase mechanism. Eg5 association with the microtubule, ADP product release, and ATP binding are rapid steps in the mechanism, while ATP hydrolysis, phosphate product release, and detachment from the microtubule occur relatively slowly. Phosphate release coupled to MtEg5 dissociation represents the rate-limiting step. Therefore, Eg5 remains tightly associated with the microtubule during most of the ATPase cycle, thus adapting the motor for its function within the mitotic spindle. Monastrol is a small molecule that specifically inhibits Eg5. I performed experiments to define the mechanistic basis for Eg5 inhibition by monastrol. The ATPase activity of Eg5 is reduced with weakened binding to microtubules. Monastrol also stabilizes a subpopulation of non-productive complexes that slowly hydrolyze ATP. After ATP hydrolysis, phosphate product is rapidly released coupled to detachment from the microtubule. Therefore, monastrol inhibits Eg5 force generation to yield a motor that cannot function properly in the mitotic spindle. I have purified Eg5 in the nucleotide-free state to perform a detailed kinetic analysis in the absence of microtubules. ATP hydrolysis and phosphate product release are rapid steps in the mechanism, and the observed rate of these steps is limited by the relatively slow isomerization of the Eg5ATP collision complex. A conformational change coupled to ADP release is the rate-limiting step in the pathway. These studies provide insight into how the microtubule guides the structural transitions needed to form the ATP hydrolysis-competent state and for rapid ADP release. In addition, monastrol appears to bind weakly to the Eg5ATP collision complex, but after tight ATP binding, the affinity for monastrol increases, thus inhibiting the conformational change required for ADP product release. Taken together, we hypothesize that loop L5 of Eg5 undergoes an open to closed structural transition that correlates with the rearrangements of the switch-1 and switch-2 regions at the active site during the ATPase cycle.

Transcription elongation is an important regulatory step in the RNA polymerase II transcription cycle. Previous work from our laboratory and others strongly suggests that the Paf1 complex interacts with and regulates the activity of RNA polymerase II during transcription elongation. Affinity purification of Paf1 revealed that it exists in a complex in vivo with Ctr9, Rtf1, Cdc73 and Leo1 called the Paf1 complex. To further investigate the function of the Paf1 complex in vivo, we used a genetic approach to study the component Rtf1 and microarray analyses to identify Ctr9-regulated genes. Null mutations in RTF1 confer two phenotypes associated with defective transcription elongation, sensitivity to base analogs and the Spt- phenotype. To identify novel mutations in RTF1, we performed a genetic screen for rtf1 mutations that confer conditional mutant phenotypes. We identified three new rtf1 missense mutations. To identify proteins that functionally interact with the Paf1 complex, we performed a screen for multicopy suppressors of the rtf1 mutations. This screen identified NAB3 as a suppressor of rtf1. Nab3 interacts with the RNA polymerase II-associated hnRNP Nrd1. Together, Nrd1 and Nab3 are required for efficient 3'-end formation of certain nonpolyadenylated RNA polymerase II transcripts, including snoRNAs. Using assays to detect transcriptional readthrough of snoRNA genes, we have shown that deletion of certain Paf1 complex members causes strong defects in snoRNA 3'-end formation. By chromatin immunoprecipitation (ChIP) analysis, we have found that the Paf1 complex and Nab3 associate with snoRNA genes in vivo. In the second approach, expression analyses revealed a role for the Paf1 complex in directly regulating the expression of certain genes that are not transcribed during growth in rich media. ChIP experiments indicated that the Paf1 complex associates along the length of ARG1 and with the promoter region of SER3. Together, our results reinforce the idea that the Paf1 complex is involved in the production of mRNAs and provide evidence that the Paf1 complex also participates in snoRNA transcription and 3'-end formation during RNA polymerase II transcription.

Many species rely on phenotypically plastic traits to defend themselves against predators and the induction of these phenotypes require reliable environmental cues. In aquatic systems, defensive phenotypes are induced by chemical cues emitted during predation events. Using larval amphibians as a model system, my dissertation focuses on how prey use the different types of chemical information available from predators (kairomones) and prey (alarm cues) and how prey integrate their defensive decisions in response to chemical cue variation over space and time. Predation cues contain information on the identity of the predator (kairomones) and the identity of the attacked prey (alarm cues). I have shown that different alarm cues (from different predator diets) induce different magnitudes of prey defense and discovered that the magnitude of the response depends on the evolutionary divergence time between the diet and the responding prey. Because chemical cues from consumed prey induce different suites of traits than cues from starved predators or damaged prey, I have also performed experiments to determine the role the predators themselves play in producing the cue (i.e. releasing a kairomone or digesting alarm cues). I found that digestion of the prey is essential to induce the complete suite of defensive traits. Because induced defenses have associated costs, prey should balance these costs and benefits by fine-tuning their responses to their environment over space and time. To do this, prey must be able to detect and respond to changes in risk when they move into new environments (spatially) or when predators come and go (temporally). I have found that tadpoles can detect small differences in risk, but that experiencing pulses of risk, when compared to a constant risk, largely does not alter their defensive decisions. Collectively, this work demonstrates the important role of environmental cues in understanding the ecology and evolution of inducible defenses.

Human diseases such as spina bifida are caused by a failure in cell morphogenesis and tissue fusion. Dorsal closure in the Drosophila embryo is a model for these tissue closure processes where proper Jun N-terminal Kinase (JNK) signaling is necessary. JNK activity is required in the leading edge cells of the epithelial layer to modulate the cytoskeleton and cell shape, allowing the epidermis to close on the dorsal side of the embryo. The mixed lineage kinase (MLK), Slipper (Slpr), is the JNKKK which is responsible for activation of the pathway during dorsal closure. The pathway components that regulate Slpr, as well as upstream activation signals, are not yet identified. We have examined the involvement of the Ste20-like kinase Misshapen (Msn) to act as the JNKKKK in the JNK pathway during dorsal closure through a direct interaction between Msn and Slpr. By observing phenotypes of recombinant and heterozygous mutants of slpr and msn, we have examined the genetic interactions. Also, by using a non-biased screen, we have investigated unknown regulators of the Slpr-mediated JNK pathway which have an effect on dorsal closure. These techniques have begun to identify regulatory interactions of molecules within the JNK pathway, and have narrowed down regions of chromosome two which may contain new modifiers further regulating JNK signaling, in order to provide a robust and highly regulated tissue closure event.

Research has repeatedly demonstrated that herbivores can, at some times and in some places, control the distribution and abundance of plants. Consequently, explaining variation in herbivore control over plant communities is a central goal in ecology and evolutionary biology. Two major challenges have prevented theoretical progress in this area of research. First, although there are numerous hypotheses that attempt to explain variation in herbivore control over plant communities, theoretical reviews have focused on a single hypothesis. Thus, it has been unclear where these herbivore control hypotheses diverge in their predictions and rationale. Second, herbivore control hypotheses base their explanations on highly correlated vegetation characteristics, namely net primary productivity (NPP), plant vigor, plant apparency, plant tissue nitrogen, plant defenses, plant tolerance, and host plant concentration. Consequently, interpretations of field experiments and meta-analyses have been equivocal. To address the first problem, I simultaneously reviewed herbivore control hypotheses and their predictions and rationale. I demonstrate that these hypotheses can be synthesized into four central hypotheses based on NPP, plant size, resource availability, and host stem density. This provides researchers with few vs. many herbivore control hypotheses. To address the second problem, I simultaneously tested these hypotheses by experimentally manipulating resource availability, total stem density, plant species composition, and herbivore abundance under field conditions. I then monitored the response of herbivore abundance, damage to plants, and the reduction in plant mass due to herbivory. The experiments demonstrated that herbivory caused the strongest reductions in mean stem mass where per stem resource availability was lowest, regardless of where herbivore abundance and damage was greatest. This result supports the plant tolerance based resource availability hypothesis, which assumes that the ability of plants to tolerate herbivory increases as resource availability increases. In addition, herbivore control over both simple plant communities (i.e., monocultures) and complex plant communities (i.e., polycultures) was due to herbivory on the dominant plant species, Solidago canadensis. Together, these results suggest that future herbivore control hypotheses should focus on the effect of per-capita resource availability on the ability of dominant plants to tolerate herbivory.

Chromosomal alterations can arise from numerous events, including errors during cell division or repair of damaged DNA. Of these errors, segregational defects such as anaphase bridges and multipolar spindles play a major role in chromosomal instability, leading to tumorigenesis. Bridges can theoretically be produced by several mechanisms including telomere-telomere fusion, persistence of chromatid cohesion into anaphase or repair of broken DNA ends. DNA damage can induce anaphase bridges following exposure to agents such as hydrogen peroxide or ionizing radiation (IR). Our hypothesis is that while the majority of double strand breaks (DSBs) are repaired, to restore the original chromosome structure, incorrect fusion events also occur leading to bridging and that bridge formation allows cells to bypass the apoptotic pathways that are activated in response to DNA damage. To test this, we set out to determine what pathways the cells use to heal the damage and form bridges. Our data suggest that neither of the two major pathways used by the cell for repair of double strand breaks, homologous recombination (HR) and non-homologous end joining (NHEJ), is required for bridge formation. In fact, the NHEJ pathway seems to play a role in the prevention of bridges. When NHEJ is compromised, the cell appears to use HR to repair the break, resulting in increased anaphase bridge formation. Moreover, intrinsic NHEJ activity of different cell lines appears to be correlated with induction of bridges from DNA damage. Our preliminary data also suggest that cell lines with high levels of bridging are capable of apoptosis, yet further experiments are required to see if blocking bridging can enhance cell death. Multipolar spindles are aberrant mitotic figures when a cell divides into two or more poles, which can lead to uneven segregation of the chromosomes. In our studies, we found that IR treatment can lead to an increase in multipolarity shortly after treatment and changes the distribution of spindle pole components. Initial observations on the splitting of centrosomal proteins following IR treatment are presented.

Tbx6 was shown to be a T-box transcription factor expressed in the primitive streak and presomitic mesoderm of the developing mouse embryo; null mutations in Tbx6 resulted in ectopic neural tube formation in place of posterior somites and embryonic lethality. However nothing was known about expression, transcriptional activity, or downstream targets of Tbx6 protein. Using antibodies generated against Tbx6, Tbx6 was confirmed to be a 58 kDa protein in vivo, and Tbx6 protein and mRNA have similar spatial and temporal expression patterns. T-box proteins were defined by the presence of the T-domain, the DNA binding region. Given the conservation of the T-domain, it was not surprising that many of the characterized T-box binding sites were also somewhat conserved. Therefore, I first demonstrated that Tbx6 recognized the T palindromic consensus sequence and half-site sequence in vitro and then determined that the Tbx6 consensus binding site was 5- AGGTGTBRNNN -3 using a PCR-based binding site selection assay. Using luciferase reporter assays in cell culture, Tbx6 was determined to be a weak transcriptional activator. Lastly, initial studies with T and Tbx6 suggested that these T-box proteins do not synergistically activate a luciferase reporter. In other work, the spontaneous mutation rib-vertebrae was determined to be a Tbx6 hypomorphic mutation due to an insertion of 185 bp into a Tbx6 enhancer. Finally, I also sought to understand how the Notch signaling pathway and Tbx6 interacted. The Notch transcription factor, RBP-J?, was demonstrated to control the maintenance of Tbx6 in the PSM. Furthermore, Tbx6 bound to several putative binding sites in vitro within a previously known mesodermal Dll1 enhancer. These results suggested that Notch signaling functions upstream and downstream of Tbx6. The work described in this thesis used a variety of in vivo and in vitro approaches, molecular and biochemical, to gain further understanding into how Tbx6, a member of a large family of transcription factors, functions in the context of a developing embryo in the presence of other family members to regulate the expression of target genes.

The actin cytoskeleton is essential for a vast array of cellular processes and behaviors including migration, cell division, cell adhesion, intracellular trafficking, and maintenance of cell shape. Regulation of cytoskeletal dynamics is achieved through the actions of a diverse group actin-binding proteins. The actin-binding protein Apxl, is a member of the Shroom protein family, which also includes Apx and KIAA1202. Shroom, the most well-characterized member of this family, binds and bundles actin stress fibers and is required for apical constriction of the neuroepithelium during neural tube closure in mice and Xenopus embryos. Apxl was named for its similarity to Apical Protein Xenopus (Apx), a regulator of an amilioride-sensitive sodium channel. All Shrm family members possess at least two of three conserved domains; a N-terminal PDZ domain, a centrally located ASD1 (APX/Shroom Domain) and a C-terminal ASD2 domain. Because of its similarity to Shroom, mouse Apxl was sequenced and cloned in order to begin initial characterization of the protein. Western blot analysis has shown that mAxpl is expressed in the majority of adult tissues. Immunofluorescence analysis of frozen sections has demonstrated that Apxl is specifically expressed in multiple populations of polarized cells, such as the neuroepithelium, vascular endothelium, and the epithelium of renal tubules. The subcellular localization of Apxl was investigated and Apxl was found to reside at the plasma membrane of non-adherent cells and in the apical compartment of polarized cells, possibly through interactions with cortical actin or members of the apical junctional complex. Analysis of Apxl deletion proteins has revealed that the ASD1 domain is crucial for proper localization,while the requirement for the PDZ domain varies in different cell lines. Cytochalasin D treatment of Rat1 fibroblasts has indicated that disruption of the actin cytoskeleton perturbs Apxl localization. Additionally, Apxl directly binds actin through its ASD1 domain in F-actin cosedimentation experiments. Apxl is expressed in multiple polarized cell types where it binds cortical actin and localizes to the apical junctional complex. Although the biological function of Apxl is unknown, its expression pattern, subcellular localization, and similarity to Shroom suggest that Apxl may play a role in regulation of cellular architecture throughout development.

My dissertation work was focused to characterize three members of the Kinesin superfamily in vitro. Kinesins are required in the cell for the correct localization and directed transport of proteins, DNA, RNA and cellular organelles. The three molecular motors of the kinesin superfamily studied here are Kar3Cik1, Kar3Vik1 and Nod. These three kinesins are unconventional in that they do not motor to the microtubule plus end transporting cargo over long distances, like Kinesin-1. Kar3Cik1 and Kar3Vik1 are Saccharomyces cerevisiae Kinesin-14s, exhibiting minus end directionality characteristic of Kinesin-14s. Kar3Cik1 is essential for meiosis and for karyogamy, or mating in yeast. Kar3Cik1 also has non-essential roles in mitosis. Kar3Vik1 plays an important role at the spindle pole body during yeast mitosis. My work has shown that the two heterodimers interact with the microtubule in very different ways. Cik1 targets Kar3 to microtubule plus ends and enhances the Kar3-instrinsic depolymerizing ability. In contrast Vik1 binds the microtubule in addition to the Kar3 motor domain and depresses the Kar3 depolymerizing ability. Both of these functions correlate with the in vivo phenotypes and suggest different mechanisms of action for the two heterodimers. Nod is a Drosophila melanogaster orphan kinesin proposed to provide a polar ejection force to stabilize chromosomes at the metaphase plate in meiosis. Our work characterized Nod as a kinesin that regulates microtubule dynamics by binding to the microtubule plus end and promoting microtubule polymerization. These results provide a mechanistic explanation for the polar ejection force observed in vivo. My dissertation work has provided knowledge about the many ways in which different kinesins can interface with and regulate microtubule dynamics.

The effects of environmental variation on the phenotypes expressed by organisms have gradually gained interest in biology. This interest has been sparked by the realization that environment-dependent phenotypic expression (i.e. phenotype plasticity) can improve an organisms fitness when different environments are encountered. The challenge for researchers is to determine the importance of phenotypic plasticity to the various fields of biology. A major goal in community ecology is to understand predator-prey interactions within natural communities. Recently, ecologists have focused their attention towards the inducible defenses of prey with the realization that prey are not simply passive participants but instead express a variety of inducible defenses. This dissertation explores the ecology and evolution of inducible defenses using freshwater snails and their predators as a model system. My specific objectives were to identify the adaptive significance of induced responses against predators, to explore the importance of development for understanding inducible defenses, and to address the phenotypic and fitness consequences of spatial and temporal variation in predation risk on prey species. Snails were extremely flexible in their ability to respond to different predator environments. It was evident in each of my experiments that snails were able to alter a unique suite of traits with different predators and integrate their phenotypic responses to environments that contained multiple predator species. Moreover, phenotypic trade-offs resulting from internal resource competition among traits appear to be the underlying mechanism for the expression of inducible defenses. I also found induced defenses can come at the cost of reduced growth, delayed reproduction, or reduced fecundity. However, these costs were dependent on the identity of the predator. By incorporating a developmental perspective, I was able to document that snails have wide developmental windows for formation of defenses but narrow windows for the reversal of defenses. Moreover, the lag time associated with the formation of some defenses can limit the benefits of the defense. Lastly, responses to predators early in development can constrain future responses to different predators and have dramatic impacts on fitness.