Collectively, my results highlight the utility of the P matrix as a tool for studying the integration of complex traits. The extreme stability of P in T. commodus suggests that it is likely to act as a constraint on the evolution of call structure in this species. This insight, together with the link between call structure and wing morphology, illustrates the value of treating evolution as a multivariate process.

The creation of biodiversity involves the evolution of new species. Recent trends in the study of speciation have increased the emphasis on the role of ecology in adaptation and the evolution of reproductive isolation. This includes examining the relative contributions of different types of selection, the role of gene flow and the genomic changes that occur during ecological speciation. The search for speciation genes continues, however our growing knowledge of how the genome translates into phenotypes means we should now consider a broader molecular basis of speciation, which includes genetic, transcriptomic and potentially epigenetic variation that contribute to phenotypic variation. This thesis addresses the molecular basis of speciation by using three different complementary methods to examine the early stages of ecological speciation and the evolution of premating reproductive isolation between two incipient species of the cactophilic fly, Drosophila mojavensis. First, the genetic basis was examined through the sequencing of two candidate genes underlying reproductive isolation (Chapter 2). Second, the historical biogeography of population divergence was uncovered using multiple sequenced loci (Chapter 3). Lastly, gene expression across the whole transcriptome associated with phenotypic plasticity and mating success was assessed (Chapter 4). Further, the role of epigenetic imprinting in the population divergence of a freshwater fish, Girardinichthys multiradiatus, was examined through sequencing of a well known gene involved in sexual conflict (Chapter 5). These studies find that uncovering the genetic variation underlying speciation is difficult, especially when there is extensive phenotypic plasticity. Further, gene expression plasticity may play an important role in the evolution of premating isolation, and this includes a role for epigenetic mechanisms of gene expression. Additionally, it is important to assess the demographic scenario of population divergence to put into context the ecological and functional data on divergent groups. Through these studies this thesis examines the genetic, expression and epigenetic variation associated with on-going population divergence, and emphasises the need to consider the potential role of the full range of gene expression changes during ecological speciation.

Thesis (Ph. D.) -- University of Texas Southwestern Medical Center at Dallas, 2004. / Vita. Bibliography: 291-299.

Tight barriers are crucial for animals. Insect respiratory cells establish barriers through their extracellular matrices. These chitinous-matrices must be soft and flexible to provide ventilation, but also tight enough to allow oxygen flow and protection against dehydration, infections, and environmental stresses. However, genes that control soft, flexible chitin-matrices are poorly known. We investigated the genes of the chitinolytic glycosylhydrolase-family 18 in the tracheal system of Drosophila melanogaster. Our findings show that five chitinases and three chitinase-like genes organize the tracheal chitin-cuticles. Most of the chitinases degrade chitin from airway lumina to enable oxygen delivery. They further improve chitin-cuticles to enhance tube stability and integrity against stresses. Unexpectedly, some chitinases also support chitin assembly to expand the tube lumen properly. Moreover, Chitinase2 plays a decisive role in the chitin-cuticle formation that establishes taenidial folds to support tube stability. Chitinase2 is apically enriched on the surface of tracheal cells, where it controls the chitin-matrix architecture independently of other known cuticular proteins or chitinases. We suppose that the principle mechanisms of chitin-cuticle assembly and degradation require a set of critical glycosylhydrolases for flexible and not-flexible cuticles. The same glycosylhydrolases support thick laminar cuticle formation and are evolutionarily conserved among arthropods.

A Thesis Submitted to the School of Graduate Studies in the Partial Fulfillment of the Requirements for the Degree Master of Science / Caenorhabditis elegans and one of its close relatives, Caenorhabditis briggsae, are animal models that are commonly used for comparative studies to understand the evolution of developmental mechanisms and gene function. Although the two species appear nearly identical morphologically, comparative genomic analyses have revealed interesting differences between the genomes. Whether such differ- ences contribute to changes in developmental mechanisms and signalling pathways is an active area of research. One of the most well studied phenotypes associated with C. elegans signalling pathways are those that affect the specification of vulval tissue. Within the system of vuval development, mutants that exhibit the Mul- tivulva (Muv) phenotype are important as they show inappropriate divisions of vulva cells, which model tumour formation. Comparing gene function in different species genetic backgrounds can lead to an understanding of how genetic differ- ences contribute to different responses in cancer development. Genetic screens, conducted in our laboratory, yielded several genes whose loss of function results in a Muv phenotype and identified a novel regulator of C. briggsae vulval devel- opment, Cbr-ivp-3. Using the nematode C. briggssae as experimental system, we have characterized the tumour suppressor gene, Cbr-ivp-3, which impacts cell sig- nalling and cell division. I have carried out molecular genetic analyses of ivp-3 in both C. briggsae and C. elegans and have begun to characterize the functional role of Cbr-ivp-3. The findings in this thesis suggest that Cbr-ivp-3 is functioning to negatively regulate EGF/Cbr-lin-3. / Thesis / Master of Science (MSc) / The nematodes Caenorhabditis elegans and Caenorhabditis briggsae, are commonly used for comparative studies to understand the evolution of developmental mechanisms and gene function. Although both species appear morphologically similar, comparative genomic analyses reveal differences between genomes. Comparing gene function in different genetic backgrounds can lead to an understanding of how genetic differences contribute to different responses in cancer development. Genetic screens have yielded several genes whose loss of function results in a Multivulva phenotype, showing inappropriate division of vulva cells, modeling tumor formation. We have carried out molecular genetic analyses of ivp-3, a novel regulator of C. briggsae vulval development, in both species and have found that Cbr-ivp-3 is regulating vulva development by negatively regulating EGF/Cbr-lin-3.

Caenorhabditis elegans and its congener, C. briggsae are excellent animal models for the comparative study of developmental mechanisms and gene function. Gupta lab is using the vulval tissue in these nematodes as a system to investigate conservation and divergence in signal transduction pathways. Genetic screens conducted earlier in our laboratory recovered several mutants that cause multivulva (Muv) phenotype. The Muv genes act as tumor suppressors and negatively regulate the proliferation of vulval precursors. Genetic and molecular work on these genes has revealed that C. briggsae vulva developmental utilizes novel genes representing a new phenotypic class termed ‘Inappropriate Vulva cell Proliferation (IVP)’ (Sharanya et al., 2015). This indicates that the signaling mechanism in C. briggsae specifies vulval cell fates differently from C. elegans. Interestingly, it has been found that Cbr-ivp mutants show higher levels of Cbr-lin-3 (EGF) transcript, indicating that these genes act genetically upstream of Cbr-lin-3, similar to SynMuv family members in C. elegans. Moreover, RNAi knockdown of the Cbr-lin-3 transcript resulted in the suppression of the multivulva phenotype in mutant animals. Similar suppression was also observed when a MAP kinase inhibitor was used in the previous study. In addition, the role of two other novel negative regulators of cell proliferation, Cbr-lin(bh1) and Cbr-lin(bh3) was also investigated. Preliminary findings on these regulators suggested that both Cbr-lin(bh1) and Cbr-lin(bh3) exhibiting a heritable Muv phenotype and are found to be located on Chromosome I and III respectively. Identification of novel genes and further characterization will help us understand the molecular function of genes and their involvement in the regulation of vulval cell differentiation. The findings of my research work will provide a background for future studies to understand the role of novel genes in reproductive system development. Overall, these results provide evidence that although the morphology of vulva is similar in the two nematode species, underlying mechanisms of development appear to have diverged. / Thesis / Master of Science (MSc)

Neisseria meningitidis is the bacterium responsible for meningococcal meningitis and septicaemia in humans. Meningococcal disease is primarily a disease of young children, characterized by rapid deterioration from first symptoms to death, with an 11% fatality rate and a global distribution. Patterns of genetic diversity in meningococcal populations provide an account of their evolutionary history and structure, which can be inferred by population genetics modelling. Understanding these phenomena can inform control and prevention strategies, and provides interesting case studies in evolution. The aim of this thesis is to develop population genetics techniques for inferring the evolutionary history of meningococci. I begin by reviewing the field, and justifying the use of coalescent methods in modelling microparasite populations. Inference on carriage populations of meningococci under the standard neutral model and the neutral microepidemic model is performed using a modification to approximate Bayesian computation. AMOVA and Mantel tests are used to quantify the differentiation between carriage and disease populations, and the extent to which geography and host age structure carriage populations. The results are used to propose revised coalescent models for meningococcal evolution. The role of natural selection in shaping meningococcal diversity is investigated using a novel method that utilises an approximation to the coalescent and reversible-jump Markov chain Monte Carlo to detect sites under selection in the presence of recombination. Having performed a simulation study to assess the statistical properties of the method, I apply it to the porB antigen locus and seven housekeeping loci in N. meningitidis. There is strong evidence for selection imposed by the host immune system in the antigen locus, but not the housekeeping loci which are functionally constrained. Finally I discuss the future direction of population genetic approaches to understanding infectious disease.

This thesis focuses on the role and historical development of strategies of experimental domestication in scientific knowledge production, with a particular focus on the function of the laboratory strains known as 'wild types' in the model organism systems of classical genetics, where they play the role of standing in for the 'natural' instance of the species so that variation may be measured. As part of establishing how lab wild types came to assume this role, I have situated them within a much longer historical trajectory that tracks how changes in the manner that European intellectual traditions conceptualised the domestic-wild divide were linked to the development of new forms of scientific domestication and knowledge production. These new developments required that existing domesticating practices be intensified, expanded and analogised in order to better control, capture and comprehend 'wild' nature. My first two chapters introduce the domestic-wild divide by discussing both contemporary and ancient interpretations of it. In my third and fourth chapter, I explore the roots of the knowledge regime of European scientific domestication. I highlight Francis Bacon's campaign to use knowledge of domesticating practices to restore human dominion, before showing how Linnaeus later re-conceptualised the natural economy as an autonomous order and original order, with domestication reinterpreted as an artful transformation of nature requiring human maintenance to prevent reversion to its wild 'natural state'. I identify this idea of the wild as original and the domestic as derivative and artificially maintained as the basis of the original wild type concept. In my fifth chapter, I discuss Darwin's attempt to unite the domestic and wild under common laws of variation and selection, including his argument that reversion was simply a product of a return to ancestral conditions of existence. I observe that Darwin's theory of variation was problematic for the effort to bring wild nature under controlled conditions for study, so in my sixth and seventh chapters discuss how this difficulty was resolved, first by experimental naturalists both before and after Darwin who utilised vivaria and microscopes to bring pieces of nature indoors, and then by Weismann and Galton's sequestration of heredity, which helped persuade scientists that domestication was not in itself a cause of germinal variation. In my eighth and ninth chapter, I detail how sequestration led the early Mendelians de Vries and Bateson to assume that wild types could be brought into the lab from nature and purified into true-breeding strains. I discuss their differing atomist and interactionist perspectives on wild type, with de Vries favouring 'elementary species' as units of nature, whereas Bateson held wild types and mutants to represent normal and abnormal forms of the species respectively. In my last chapter, I cover the replacement of Bateson's interactionist genetics by the reductionist genetics of the Morgan group and argue that this led to a disintegration of wild types into their component genes. I conclude with a discussion of what wild type strains in classical genetics were meant to be representative of, and end by establishing that whilst these strains may not wholly be representative of their species, they are nonetheless useful tools for scientific knowledge production.