The Amazon rainforest encompasses over one billion acres of South America and sustains remarkable biodiversity. Despite the large body of research stemming from this region, little is known about the effects of geomorphology and primary productivity on the fauna of Amazonia, and on reptiles and amphibians in particular. In my dissertation, I examine differences in the abundance, biomass and species richness of secondary consumers in the leaf litter herpetofauna communities on young and ancient soils. Herein, I develop methods to utilize existing data sets and museum collections in new studies involving community biomass. I found that, although the process of preservation does change the size of specimens, for most species the differences are less than 4%. Furthermore, these changes can be quantified and taken into account when applying measurements made on preserved specimens to studies of living individuals. I also derive equations which can be used to calculate mass from length in anurans. Snout-vent length (SVL) has been measured on thousands of herps, some released in the field and many housed in research collections. Estimated using SVL measurements, mass of individuals can now be used to determine community biomass of populations that were initially sampled for other purposes. In the second half of this dissertation, I assess the effects of geomorphology on leaf litter herpetofauna in Neotropical lowland rainforests. At sites with similar latitude, elevation and climate, litter herpetofauna abundance, biomass and species richness are twice as high on younger soils. Using methodology developed in the first chapters, this comparison is expanded to include forests differing in latitude and climate. In this study, the trend of increased density, biomass and diversity on younger soils holds, lending further support to the hypothesis that geomorphology and primary productivity drive leaf litter herpetofauna community dynamics. These studies taken together provide the basis for a change in Amazonian management strategies. The prevailing notion that a few large reserves within the Amazon will be able to sustain Amazonian biodiversity is unfounded. Reserves and conservation policy must be designed around the local geologic history and forest dynamics of forest regions within the Basin.

Carbon monoxide-oxidizing bacteria are among the first colonists of recently formed volcanic deposits despite environmental conditions that challenge their survival, such as oscillating water regimes and lack of endogenous sources of organic carbon and nitrogen. Activity assessments and molecular surveys of the coxL gene (large subunit carbon monoxide dehydrogenase) across a vegetation gradient on a 1959 volcanic deposit on Kilauea Volcano (Hawaii) indicated that CO-oxidizing communities continue to expand, diversify and remain competitive during ecosystem development. Distinct compositional shifts occurred across the gradient with Firmicutes-like coxL sequences dominating clone libraries from unvegetated sites (Bare) and Proteobacteria coxL sequences dominating libraries from vegetated sites (Canopy). Water regimes at the Bare and Canopy sites were distinct, with the former experiencing sometimes extreme diurnal oscillations in water potential (i.e. near 0 MPa to -60 MPa) and the Canopy remaining near 0 MPa. However, CO-oxidizing communities at these sites did not exhibit differential adaptations to water stress, which may not be an important factor controlling CO oxidizer distributions in these sites. Of the major taxonomic groups represented in the coxL clone libraries, increased total richness across the transect was most highly correlated with the number of OTUs represented by ?-Proteobacteria sequences, many of which were closely related to Burkholderia coxL. A qPCR approach developed to target Burkholderia coxL indicated that absolute gene copy numbers increased from undetectable quantities in the Bare site to an average of 8.6 x 108 copies gdw soil-1 in the Canopy site. In addition, a 16S rRNA gene phylogenetic analysis of Burkholderia isolates obtained from these sites and their close relatives indicated that CO-oxidation may be a common trait among root-associated Burkholderia species. Their strong association with plants, abundance and ability to consume atmospheric levels of CO, may explain their contribution to CO-oxidizer community expansion during ecosystem development and indicates that they may be an important constituent of active CO-oxidizing communities in situ.

In the genus Thomomys, external phenotypic differences account for much of the variation among currently named species. Allozyme and mitochondrial studies of the Mexican species, T. umbrinus, have shown strong discordance between genetically defined clades and current species taxonomy. In this study, I analyzed mitochondrial and nuclear DNA sequences, chomosomes, and morphological characters, and used ecological niche modeling to investigate the phylogenetic status of a phenotypically divergent subspecies of T. umbrinus, T. u. atrovarius. Although quantitative morphometrics provided little discrimination among groups within T. umbrinus, the molecular data indicated that there are three allopatric clades. The Pacific coastal clade (T. u. atrovarius) is diagnosable based on multiple characters, including nuclear genotype, habitat preference, and qualitative morphological characters. Accordingly, I resurrect Allens (1898) species Thomomys atrovarius to represent members of this clade, which occurs in the Pacific coastal region of western Mexico.

Iron-sulfur (Fe-S) clusters are ubiquitous prosthetic groups that function in diverse fundamental life processes. However, the biogenesis of iron-sulfur clusters in vivo is not a spontaneous process. Previous studies indicated that Fe-S clusters maybe synthesized by three major systems: the Nif, the ISC, and the SUF systems. Among these three systems, IscU, NifU are the scaffold proteins for the Fe-S clusters assembly. IscA and its paralog, SufA are proposed as an alternative iron-sulfur cluster assembly scaffold proteins. The cysteine desulfurases: IscS, NifS and SufS catalyze desulfurization of the L-cysteine and provide sulfide for Fe-S clusters assembly. However, the iron donor for the Fe-S clusters assembly remains poorly understood. In this research, we reported that IscA is a strong iron binding protein. Under physiological conditions, if only iron is available, iron will bind to IscA. The addition of L-cysteine to this iron-bound IscA mobilizes the iron center in IscA and transfer iron to IscU for the Fe-S cluster assembly. However, if both iron and sulfide are available, Fe-S clusters are preferred to be assembled in IscU. Under oxidative stress conditions, IscA fails to bind ferrous iron due to the oxidation of its iron binding thiolate groups. CyaY, an E. coli homology of Frataxin is able to bind iron under oxidative stress conditions and effectively alleviate the production of the deleterious hydroxyl free radicals. Nevertheless, unlike IscA, CyaY cannot function as an efficient iron donor for the Fe-S clusters assemlby due to its weak iron binding property. We also investigated the repair mechanism for the NO-modified aconitase B [4Fe-4S] clusters. We found that E. coli [4Fe-4S] aconitase B is readily converted to the protein-bound DNICs by NO in vitro and in vivo. L-cysteine and oxygen are required for decomposition of the protein-bound DNICs. We further demonstrated that a complete repair of the NO-modified aconitase B requires two sequential steps: decomposition of the protein-bound DNICs requires both L-cysteine and oxygen, and the reassembly of Fe-S clusters, which requires Fe-S clusters assembly machinery.

Phytoplankton and benthic microalgae (BMA) may both contribute to the sediment organic matter pool available to benthic consumers. Recent studies have highlighted the presence and ecological roles of benthic microalgae (BMA) on the continental shelf but studies from the north-central Gulf of Mexico are limited. Here we use photosynthetic pigment analysis and microscopic examination of sediment microalgae to investigate how the biomass, composition, and degradation state of sediment-associated microalgae varies along the Louisiana inner shelf across a continuum of water column and sediment conditions. The contribution of microalgae to higher trophic levels was also examined. Three sandy shoals and surrounding muddy sediments with depths ranging from 4 to 20 m were studied. Light levels (< 1 to 30% of surface PAR) were sufficient for benthic photosynthesis although likely limiting at most locations. Sediment chlorophyll a concentrations ranged from 8 to 77 mg m-2 with no significant differences among locations. Pigment data suggested that sediment microalgae were primarily diatoms at all locations. Epipelic pennate diatoms, (considered indicative of BMA) made up a significantly greater proportion of sediment diatoms at sandy (50% to 98%) compared to muddy stations (16 to 56%), and sediment total pheopigment concentrations on the sandy stations (<20 mg m-2) were significantly lower than concentrations at nearby muddy stations (>40 mg m-2) suggesting that BMA predominate in shallow sandy sediments and that phytodetritus predominates at muddy stations. Our results also suggest that the relative proportion of phytodetritus in the benthos was highest where phytoplankton biomass in the overlying water was greatest, independent of sediment type. Stable isotopes indicated that, I found BMA was the primary basal resource for infauna where BMA was consistently predominant in the sediment algae. In sediments containing a mix of BMA and phytoplankton, isotope data indicated both resources were important to infauna. Our study demonstrates that, when available, BMA is utilized by shelf infauna that, in turn, support higher trophic levels.

Obesity is characterized by an overabundance of fat cells, or adipocytes, and is currently a global epidemic. Adipocytes store energy, respond to insulin, and participate in diverse endocrine signaling pathways by secreting adipokines. Obesity leads to the dysregulation of adipocyte function, which could lead to numerous metabolic diseases, such as Type II diabetes. Understanding the mechanisms that regulate adipocyte development and involvement in endocrine signaling will lead to a greater understanding of the importance of fat metabolism in full-body health. Many lines of evidence demonstrate the importance of various transcription factors in regulating adipogenic processes, and one of which is the Signal Transducer and Activator of Transcription (STAT) proteins. Work by a variety of laboratories has demonstrated that STAT proteins, particularly STAT5A, are activated and induced during adipogenesis and play an important role in adipose tissue development. Novel studies highlighted in this dissertation demonstrate that pyruvate dehydrogenase kinase (PDK)-4, a known regulator of glycolysis, is highly induced in adipocytes by growth hormone (GH) or prolactin (PRL) in a STAT5 dependent manner. Under these conditions, the induction of PDK4 is accompanied by insulin resistance. Because adipocytes also express other STATs, we observed the effects of other STAT activators on adipocytes both in vitro and in vivo. Adipocytes are responsive to gp130 cytokines, and these cytokines have diverse functions in adipocytes, as well as other tissues. All gp130 cytokines share glycoprotein 130 as a common transducer protein in their functional receptor complex and typically activate STAT3. Several gp130 cytokines differentially affect adipogenesis and insulin-stimulated glucose uptake, as well as crosstalk with other members of the gp130 family to alter one anothers signaling. Novel studies highlighted in these studies demonstrate that adipocytes both in vitro and in vivo are responsive to neuropoietin (NP), which can inhibit adipogenesis and negatively regulate insulin signaling. Importantly, NP does not activate the LIF receptor, although previously reported to do so, and does not crosstalk with other gp130 cytokines. An in-depth analysis of STATs, as well as their activators, will enable us to understand how their role in adipocytes could contribute to the pathogenesis of metabolic disease.

Genomic imprinting is a process that leads to the silencing of one allele of a gene in a parent of origin specific manner. Genes that are involved in this process are often regulated in clusters, one of which is the Peg3 (Paternally expressed gene 3) imprinted domain. We investigated this region for both CpG islands and long antisense transcripts, two common features of imprinted gene clusters. First, we performed a systematic survey of DNA methylation status of the CpG islands in this region of the mouse, cow, and human genomes. We identified two previously unreported differentially methylated regions (DMR): one in the promoter region of mouse Zim3 and another in the promoter region of human USP29. The PEG3-CpG island is the only DMR that is conserved among these three species. PEG3 has been implicated in several types of cancer, so we examined the methylation status of several CpG islands in this region using human tumor derived DNA. The CpG islands near PEG3 and USP29 both showed hypermethylation in DNA derived from breast and ovarian tumors. Second, we identified an antisense transcript to ZIM2 (zinc finger imprinted gene 2) called ZIM2as in the human, chimpanzee, and orangutan. In non-primate mammals, the 5 side of the Peg3 imprinted domain is bounded by a cluster of olfactory receptor (OLFR) genes which may curtail the spread of imprinting. We report the presence of two previously unreported DMRs near the ZIM2as promoter region. The CpG island distal to ZIM2as was methylated allele-specifically in the human testis, while the CpG island proximal to the ZIM2as promoter showed a mosaic methylation pattern in the chimpanzee. Two CpG islands near the promoter region of ZIM2as showed different methylation patterns in these three species. Overall, this work provides a firm foundation for future studies of the Peg3 imprinted domain. It represents the first systematic study of DNA methylation in the Peg3 imprinted region. It also describes an antisense transcript that has formed in the great ape PEG3 imprinted domain which may control the extension of this imprinted domain.

Our genetic inheritance begins with one genome copy from each parent. The presence of one or more errors in either the maternal or the paternal genome can lead to genetic disease or disruption of the embryonic program and potential loss of the offspring. Numerous chemical and physical toxicants are known to produce germ line mutagenesis based on their ability to produce DNA sequence mutations. Exposures to these cytotoxic and mutagenic agents pose a risk for human offspring. However, very little is known regarding the sensitivity of the epigenetic patterns involved in reproductive functions to adverse effects of chemical agents. The present study was undertaken to address the potential for ethylene dichloride (EDC) to disrupt the epigenetic programming of the paternal genome in mammalian (mouse) sperm. Unfortunately, it is very difficult to study developing germ cells such as spermatogonia and their precursor spermatogonial stem cells within the in vivo mouse testis model. Thus, an in vitro mouse spermatogonial stem cell model was developed that enabled the triggering of a spermatogenesis differentiation pathway in these stem cell cultures. Since the epigenetic imprint patterns are reported to be established in the spermatogonial stem cell of prior to meiosis, this in vitro model enabled the treatment and study of the effects of EDC on DNA, 5-methlcytosine and histone modifications in the paternal gamete. Paternally imprinted genes, such as H19, Gtl2, and Rasgrf1, showed changes in histone methylation modifications in ethylene dichloride treated spermatogonia. These data demonstrate that ethylene dichloride can disrupt the genomic imprint in developing sperm, and thus perturb the embryonic programming of potential offspring in the mouse model. This work suggests that EDC may have the potential to cause genetic diseases in offspring from exposed males.

Ergosterol is the major sterol found in the membranes of Chlamydomonas reinhardtii. While past studies have identified some ergosterol mutants in C. reinhardtii, very little is known about sterol biosynthesis pathways in this species. With the elucidation of the Chlamydomonas genome, bioinformatics analysis has allowed us to determine potential genes involved in ergosterol biosynthesis. With this knowledge, a working model of the pathway was designed for future analysis. Several of the ergosterol biosynthetic genes were analyzed in respect to their role and involvement in flagellar regeneration. These genes were upregulated during the regrowth of the flagella. Also Chlamydomonas strains lacking flagella were analyzed by Q-RT PCR to determine what role ergosterol biosynthetic genes played in the absence of their flagella. Finally, one of the genes with homology to the yeast sterol C-5 desaturase, ERG3, was chosen for further analysis. To test whether ERG3 of C. reinhardtii had a similar function, yeast Saccharomyces cerevisiae ERG3 knockout strains were created to assess whether a plasmid expressing the Chlamydomonas ERG3 could complement the deletion. These erg3 null mutants were transformed with a vector expressing ERG3 cDNA from C. reinhardtii driven by the yeast ADH1 promoter, and this expression was able to restore ergosterol biosynthesis and reverse phenotypes associated with lack of ERG3 function. Complementation of these erg3 null phenotypes strongly suggests that ERG3 in C. reinhardtii functions as a sterol C-5 desaturase. Results from this dissertation provides the groundwork for future experimentation in the field of sterol lipid research in Chlamydomonas reinhardtii.

In this dissertation I explored variation in vocalizations of the avian genus Synallaxis (Passeriformes: Furnariidae) and the use of vocal characters to determine phylogenetic relationships at the species level. Vocalizations of suboscine passerines have recently gained favor as a proxy for genetic material in systematic studies; however, individual and geographic variation are both widespread in suboscines, complicating the definition and analysis of vocal characters. This is particularly true for species that have simple, phylogenetically conserved vocalizations. Moreover, the use of vocal characters in suboscine systematics depends on the confident assessment of homologous vocalizations, as well as a thorough understanding of their role in the behavioral processes fundamental to reproductive isolation and speciation. My research was based upon a collection of approximately 1200 Synallaxis recordings, representing every species and most subspecies in the genus. I identified presumably homologous vocalizations (songs) and analyzed variation within and among individuals, populations, and species. Song playback induced statistically significant changes in character measurements within individuals, although individual diagnosability was generally high even in classifications based on few measurements. Geographic variation was pronounced in a widespread, polytypic Andean species (S. azarae) but nonexistent in a related monotypic lowland form (S. frontalis). In comparisons both within and among species, a high degree of vocal convergence within a limited multivariate space and between geographic and elevational extremes obscured potentially significant patterns of variation in other characters, and underscore the need for large sample sizes and complete geographic sampling in suboscine vocal studies. Little phylogenetic information was recovered when vocal characters were mapped onto a molecular phylogeny of Synallaxis. Only one clade showed a strong pattern of phylogenetic conservatism, and there was no relationship overall between the number of vocal differences and genetic distance. The low incidence of autapomorphies and low consistency index for syntactical vocal characters suggest that vocalizations are constrained by morphology and prone to convergence in distantly related species. My results demonstrate substantial variation in suboscine vocalizations and suggest that vocal characters may not be reliable indicators of phylogenetic relationships at broad taxonomic scales, particularly among behaviorally and ecologically diverse taxa.