Widespread bird species endemic to the lowlands of the Guineo-Congolian rainforest of Africa have traditionally been thought to lack substantial intraspecific structure. This view owes largely to their widespread distributions coupled with the absence of discrete geographic variation in plumage. In the following chapters I examine the phylogeographic patterns of three such species using a combination of molecular and morphometric data. The three species investigated are the Green Hylia (Hylia prasina), the Red-tailed Bristlebill (Bleda syndactylus), and the Yellow-whiskered Greenbul (Andropadus latirostris). Using the phylogeographic patterns from these data, I assess 1) models of rainforest species diversification, 2) current taxonomy, and 3) implications for conservation of lowland rainforests. In contrast to the lack of discrete plumage variation in these species, phylogeographic analyses reveal a high degree molecular and morphological divergence. Furthermore, general patterns of geographical structure of the mtDNA data are largely congruent among the three species. Each species has unique mtDNA haplotype groups in Liberia, Ghana, Cameroon-Gabon, Central African Republic, and the eastern Democratic Republic of Congo. Andropadus latirostris, which also occurs in montane forests, has unique haplotype groups in these montane regions. Taken together the patterns of geographic variation in molecular and morphological datasets from these three species suggest a history of allopatric divergence via genetic drift, consistent with predictions of refugial diversification. There is also some evidence for a potential role for divergent selection along a longitudinal temperature gradient in shaping the morphometric diversity in Andropadus latirostris. Despite their widespread distributions and the potential for high gene flow, these three species exhibit a remarkable level of geographic structure across the lowland rainforest. This contradicts the prevailing view of widespread lowland species as large panmictic populations. This significant geographic structure has important implications for species and habitat conservation in Africa, where lowland forest is typically viewed as a continuous block of homogeneous habitat. The implication is habitat loss in one area would not result in a substantial loss of biodiversity because many of the species are widely distributed. The geographic structure evident in these data suggests that that is not the case.

Spartina alterniflora, the dominant vegetation in many salt marshes, has the potential to remobilize metals (Cd, Cr, Cu, Ni, Pb, Zn) from sediment to the local salt marsh environment. My research determined the influence of sediment metal concentration, sediment type and salinity on the uptake, distribution and accumulation of metals by S. alterniflora tissues and excreted salts. Spartina alterniflora was cultivated at a constant salinity in natural and dredge sediments with three levels of metal additions (control, low, and high). The distribution and concentration of metals in S. alterniflora was influenced by both metal treatments and sediment type; however, the metals varied in their uptake and distribution. The proportional distribution of metals among tissues was influenced only by the level of metal contamination. Metal concentrations in different tissues increased with increased levels of metal contamination, but sediment type also influenced tissue concentrations. In contrast, the concentration of metals in excreted salts did not increase in response to sediment contamination, and some metals in excreted salts decreased significantly as metal contamination increased. In a concurrent experiment, S. alterniflora was cultivated in dredge sediment with a low level of metal addition at three salinities (0 psu, 15 psu, 30 psu). The distribution of Pb and Zn in tissues and excreted salts was influenced by salinity. Further analysis of the results from these experiments showed that the concentration factor (CF) and translocation factor (TF) for tissues and excreted salts varied among metals, and were significantly influenced by metal contamination. To a lesser degree, sediment type, the interaction between metal treatment and sediment type, and salinity also influenced the CFs and TFs. Additionally, the concentrations of metals in excreted salts were determined in two coastal marshes of Louisiana. The concentration of metals was much lower than the only other previously published values from New Jersey, and the cause for this discrepancy is unclear. The physiological and geochemical bases for these uptake patterns are not well understood. However, my observations illustrate the complexity of metal uptake by S. alterniflora and have potentially significant implications for the trophic transfer of metals within salt marsh food webs.

Microsporidiosis is an emerging disease among immunocompromised individuals who often present with chronic diarrhea. The intracellular, eukaryotic parasites responsible for this pathology can often disseminate, causing multiorgan infections. Dissemination of these pathogens is believed to occur through vehicular spread by macrophages. The macrophage response to microsporidia is poorly understood. The information, described herein, is focused on defining the host-pathogen interaction and subsequent inflammatory response in human monocyte-derived-macrophages (MDM) against Encephalitozoon spp. of microsporidia. Initial studies were designed to better define the infection kinetics in MDM using various microscopic analysis and novel staining approaches. Spore adherence and uptake occurs within the first 6 hr and parsitophorous vacuole formation within 24 hr after infection. Replication was shown to peak at 72 hr as measured by bromodeoxyuridine incorporation and spore formation by 120 hr. Treating the MDM with interferon gamma and bacterial lipopolysaccharide reduced parasitic burden. The role of MDM in initiating monocyte recruitment after infection was evaluated using co-culture chemotaxis assays, limited gene and protein arrays, ELISA, and neutralizing antibody assays. These studies identified three major monocyte chemoattractants, CCL2, CCL3, and CCL4 that were upregulated, produced, and secreted in response to Encephalitozoon infections. Furthermore, these were necessary for monocytic infiltration. Finally, investigations into the receptors involved in initiating host recognition and regulating chemokine production were examined. Toll-like receptor (TLR) 2 was shown to be activated by Encephalitozoon spores. Using siRNA gene knock-downs in MDM, TLR2 was revealed to activate NF-êB within 1 hr after parasite exposure resulting in the production of not only CCL3 and CCL4 but also two pro-inflammatory cytokines, TNF-á and IL-8. These results indicate that microsporidia are recognized by TLR2 and induce the production of chemotactic and inflammatory mediators needed for the recruitment of monocytes/macrophages, which allow for parasitic proliferation.

Top-down (consumer) versus bottom-up (resource) control of food webs has long interested ecologists. Here, I take advantage of a full-factorial design of ecosystem-wide manipulations of nutrient additions (loading rates 10x above background) and the significant reduction (~60%) of a key predator, the killifish Fundulus heteroclitus, in the tidal creeks of the Plum Island Estuary, Massachusetts. Prior to manipulations, annelids numerically constituted 97% of the infaunal community and the largest scale (creeks) accounted for little spatial variability in annelid populations and diversities. Tidal creeks were similar based on diversity indices, abundance, and community patterns, suggesting the tidal creeks are appropriate replicates/experimental units for manipulations. Using data collected before (2003) and after (2004-2006) manipulations began, I observed little evidence of top-down or bottom-up control on infaunal densities, biomass, or community structure in four different habitats along an inundation gradient. Using exclusion cages to remove all predators (primarily killifish and the grass shrimp Palaemonetes pugio) within fish removal treatments (in non-nutrient creeks), I found top-down control of surface feeding polychaetes including Manayunkia aestuarina and Streblospio benedicti. Shrimp body size increased with killifish reduction but not shrimp density, suggesting that shrimp may alter their behavior and exert stronger top-down control on infauna when killifish are removed. No corresponding decrease in benthic microalgae (BMA) occurred when infauna abundance increased, suggesting a weak infauna-BMA interaction. For epifauna on the marsh platform, I found that hydrobiid snails increased in the creek bank Spartina alterniflora with fish removal and treatments interacted antagonistically on the amphipod, Uhlorchestia spartinophila. The interaction likely resulted from the parasite-induced movement of U. spartinophila to the creek wall habitat. This movement, in turn, made the amphipod more susceptible to predation by the semipalmated sandpiper, Calidris pusilla. Top-down and bottom-up control has been thought to operate independently on saltmarsh invertebrates. I demonstrate that food-web phenomena such as trophic omnivory, behavioral modification and indirect effects increase complexity and preclude simple predictions of trophic control on benthic invertebrates. If these trends are widespread, then long-term, large spatial-scale studies may be required to more completely understand the relative importance of top-down and bottom-up control on benthic invertebrates.

Salt marshes are areas of high primary production that in turn support high secondary production. Macrophytes, phytoplankton and a variety of benthic algae all contribute to the high primary productivity. Each has the potential to contribute to the saltmarsh food web making it difficult to determine organic matter resources important to secondary production in salt marshes. Furthermore, our understanding of saltmarsh food webs is complicated by anthropogenic effects such as nutrient inputs and exploitation of key predators. I utilized a combination of natural abundance stable isotopes, isotope tracer additions, gut content analysis and manipulative experiments to determine the diets of saltmarsh consumers collected from tidal creeks within Plum Island Estuary, Massachusetts, USA (42¢ª44'N, 70¢ª52'W) and to address the effects of nutrient additions and species modifications on the saltmarsh food web. Results indicate detritus of the vascular marsh plants, Spartina alterniflora and S. patens, was of limited dietary importance to many consumers. Instead, microalgae, including epipelic and epiphytic diatoms and phytoplankton were dominant basal resources in the saltmarsh food web. However, Spartina dietary contributions increased in the high-marsh intertidal later in the growing season. Furthermore, the addition of nutrients altered the relative contribution of basal resources to the diet of infauna. Specifically, nutrient additions facilitated a change in the diet of an oligochaete, Cernosvitosviella immota, from macrophyte detritus to epiphytic algae and a switch within algal resources from tidal resources (e.g., phytoplankton) to local sediment associated algae (e.g., epipelic and epiphytic microalgae) for two polychaetes, Nereis diversicolor and Manayunkia aestuarina. This strongly suggests that anthropogenic nutrient inputs have the potential to increase the relative importance of algae to primary consumers and ultimately higher trophic levels. My research further suggests the diet of an important intermediate omnivore, Palaemonetes pugio, was altered with the reduction in abundance of a top omnivore, Fundulus heteroclitus. Some P. pugio became more carnivorous when F. heteroclitus where reduced indicating top-down control of infauna by P. pugio mediated through behavior. In the long-term, anthropogenic effects could fundamentally alter food web structure by changing saltmarsh species composition and linkages between primary producers and higher trophic levels.

Nitric oxide is a gaseous signaling molecule that is produced by subsets of each cell type in the vertebrate retina. Though there is evidence that nitric oxide (NO) can affect multiple cellular processes in the retina, much remains unknown, especially with respect to its function in the inner retina. We have used a simplified system of cultured amacrine cells (interneurons that signal in the inner retina) to explore the role of nitric oxide in amacrine cell signaling. We find that physiological concentrations (100s of nM low μM) of nitric oxide (NO) transiently invert the sign of voltage responses mediated by GABA or glycine receptors by shifting the equilibrium potential for chloride (ECl-) to more positive values. The direction of the shift in ECl- is consistent with a transient elevation of intracellular chloride. The physiological consequence of this shift is that NO can switch inhibitory synapses into excitatory synapses. Manipulations of extracellular chloride demonstrate that the shift in ECl- is not caused by the transport of chloride across the plasma membrane into the cytosol. Instead, NO mediates a release of chloride from an internal compartment. Analysis of cellular pH using the pH indicator dye, SNARF-1AM, reveals that NO also induces a transient acidification of the cytosol that displays a similar time course to the cytosolic chloride elevation. Using measurements of chloride reversal potential (ECl-) to monitor changes in intracellular chloride levels, we found that alkalinization of the cytosol with NH4Cl resulted in a negative shift in ECl-, consistent with a decrease in internal chloride. Acidification of the cytosol with amiloride induced a positive shift in ECl-, consistent with a low cytosolic pH-driven increase in internal chloride. Furthermore, NO-induced positive shifts in the ECl- were reduced in a basic cellular environment. Finally, when we strongly buffered cytosolic pH with 125 mM HEPES in the recording pipet, we found that the ability of NO to alter cytosolic chloride levels was reduced. These results indicate that NO-induced changes in cellular pH are both sufficient and necessary to alter chloride distribution across internal membranes in neurons. The discovery that this redistribution can change the sign of central synapses has potentially broad implications for our understanding of the role of this signaling molecule in the CNS.

The Formosan subterranean termite (FST), a pest species native to China and introduced to the U.S., is obligatorily dependent on its gut microbiota. Using high-throughput 16S rRNA gene sequencing, the effects of long-term alcohol storage and geographic location on the bacteria composition of the FST colonies were investigated. Initial studies found the use of amplified ribosomal DNA restriction analysis (ARDRA) to be unpractical due to its higher costs compared to the direct sequencing of 16S rRNA gene sequences. Using nine FST colonies consisting of fresh and alcohol-stored Lousiana FST colonies and alcohol-stored China FST colonies, 237 bacteria ribotypes were identified from 1876 clones based on a <97% sequence similarity criterion. Twenty-four of the ribotypes were artifact sequences and were excluded from subsequent analyses. Most of the remaining ribotypes were novel (70.89% of total ribotypes). Termite-specific bacteria dominated the bacteria composition in the FSTs (66.45% of total clones). Only 3.34% of the total clones were similar to environmental bacteria. Thirteen bacteria phyla were represented: Bacteroidetes (42.91% of total clones), Firmicutes (30.49%), Spirochaetes (11.30%), Actinobacteria (5.70%), Proteobacteria (2.24%), Tenericutes (1.55%), candidate division Termite Group 1 (1.01%), candidate division TM7 (0.64%), Verrucomicrobia (0.59%), Planctomycetes (0.48%), candidate division Synergistes (0.21%), candidate division ZB3 (0.05%) and Cyanobacteria (0.05%). The Bacteroidetes ribotype previously identified to be dominant in FST from Japan, was also among the dominant phyla in all the FST colonies of this study (38.71% of total clones). Differential DNA degradation occurred in the alcohol-stored FST samples, leading to higher proportions of the gram-positive bacteria such as Actinobacteria, Bacilli and Clostridia and lower proportions of the gram-negative bacteria such as Bacteroidetes and Spirochaetes compared to the fresh FST samples. Long-term alcohol storage of the FST led to the discovery of less abundant ribotypes when the predominant ribotype was reduced. Geographic region did not show detectable influence on the FST bacteria composition. This was likely due to the multiple introduction of FST from China into the U.S. Future studies using T-RFLP to sample the bacteria community from FST colonies randomized across each geographical region would be useful in confirming the observations from this study.

LINE-1 (Long INterspersed Element-1 or L1) and Alu elements are important sources of structural variation in primate genomes because they are highly active retrotransposons with copy numbers of ~520,000 and >1.2 million within the human genome, respectively. Although the bulk of these elements have resided in their respective host genomes for a long time, and have thus accumulated random mutations, overall these elements retain high levels of sequence identity among themselves. The presence of many nearly-identical retrotransposons located close to each other (e.g., Alu-Alu or L1-L1 pairs) disposes their host genomes to unequal homologous DNA recombination events that generate genomic deletions and inversions of varying sizes. Through computational comparisons of the human and chimpanzee genome sequences, and using rhesus macaque and orangutan genome sequences as outgroups, we have identified species-specific genomic variation. In the first analysis, we identified human and chimpanzee-specific L1s and examined their sequence evolution. We show that L1 retrotransposition activity is slightly higher in the human lineage, relative to the chimpanzee lineage, and that L1s have experienced different evolutionary fates in these two lineages, resulting from random variation or competition between L1 subfamily lineages. Next, we analyzed the magnitude of Alu recombination-mediated deletions (ARMDs) in the chimpanzee lineage subsequent to the human-chimpanzee divergence (~6 million years ago). We have identified 663 chimpanzee lineage-specific deletions (involving a total of ~771 kb of genomic sequence) attributable to this process. The RefSeq databases indicate that 13 exons in six genes are annotated as either demonstrably or putatively functional in the human genome, and 299 intronic regions have been deleted through ARMDs in the chimpanzee lineage. In the third analysis, we characterize chromosomal inversion events between the human and chimpanzee genomes caused by inverted L1-L1 or Alu-Alu pairs. We have identified 49 retrotransposon recombination-mediated inversion (RRMI) loci and, among them, three RRMI loci contain inverted exonic regions in known genes. Therefore, we suggest that L1 and Alu elements have contributed to the genomic and phenotypic diversity between humans and chimpanzees since the divergence of the two species.

Despite the theoretical link between the ecology and the population genetics of species, little empirical evidence is available that corroborates the association. Here, I examined genetic variation in 40 co-distributed species of lowland Neotropical rainforest birds that have populations isolated on either side of the Andes, Amazon River, and Madeira River. I found widely varying levels of genetic divergence among these taxa between the same biogeographic barriers. My investigation of the extent to which ecological traits predicted the level of cross-barrier divergence revealed a significant relationship between the forest stratum at which a species forages and the level of within-population and cross-barrier genetic differentiation. Canopy species had statistically lower divergence values across the Andes and two riverine barriers than did understory birds. I hypothesize that the association reflects an effect of dispersal propensity on the geographic structuring of genetic variation, and, consequently, on the ancestral and extant effective population sizes of each species. This is the first large-scale avian comparative study to document a significant association between ecological traits of a species and its level of genetic differentiation. I examined further the contrasting genetic patterns revealed previously by comparing the range-wide mitochondrial (mtDNA) phylogeography of two canopy and two understory species of lowland Neotropical rainforest birds. All species exhibited divergence between cross-Andean populations. Unlike canopy species, understory birds were structured at smaller spatial scales, particularly across riverine barriers of the Amazon basin. Surprisingly, estimates of isolation-by-distance, a proxy for dispersal propensity, are similar within areas of endemism for all taxa suggesting levels of gene flow are comparable through contiguous habitat in canopy and understory species. Lastly, I examined the multilocus phylogeography of three previously studied species with contrasting mtDNA patterns to investigate the role of historical demography in cross-Andean divergence. Demographic estimates using an isolation-with-migration model suggest among-taxa variance in cross-Andean divergences reflects a history of staggered isolation versus a simultaneous isolating event. Nuclear sequence data reveal asymmetrical gene flow in two species marked by relatively shallow cross-Andean divergence, further evidence of differential effectiveness of the Andes as a barrier to gene flow among co-distributed taxa.

Pseudomonas aeruginosa encodes a putative topoisomerase with sequence similarity to the type IB topoisomerase enzyme from vaccinia virus. Residues in the active site are conserved, notably Tyr292 which would be predicted to form the transient covalent bond to DNA. The gene encoding the P. aeruginosa topoisomerase I (PAT) was cloned and expressed in E. coli. The enzyme relaxes supercoiled DNA, while a mutant of P. aeruginosa containing a Tyr292 to Phe substitution at the active site was found to be catalytically inert. This is consistent with the role of Tyr in forming the covalent intermediate. Like vaccinia topoisomerase (VT), PAT relaxes DNA in the absence of ATP. Unlike VT, PAT does not relax supercoiled DNA without MgCl2 (or MnCl2) present. In addition, high concentration of NaCl is not able to substitute for MgCl2 as seen for VT. A truncated derivative of the topoisomerase lacking residues 1-98 relaxes DNA in the absence of the N-terminal domain, with both full length and truncated enzyme exhibiting equivalent requirements for divalent cations. Data shows that P. aeruginosa encodes a functional topoisomerase with significant similarity to the type IB enzyme encoded by poxviruses. Fluoroquinolones are antibacterial agents in clinical use with activity against DNA gyrase and DNA topoisomerase IV. P. aeruginosa is an opportunistic pathogen causing life-threatening diseases. Sparfloxacin, enrofloxacin and norfloxacin fluoroquinolones, are able to inhibit the PAT at high concentrations but other drugs belonging to this family are unable to do so. VT is ¡Ö32 KDa and one-third the size of the human topoisomerase ¡Ö 100 KDa. It shares sequence and biochemical similarities with the human topoisomerase. The VT binds duplex DNA with stringent specificity for transesterification at 5`-(C/T) CCTT site, where the 3` phosphate of the incised strand is linked to the Tyr274 of the enzyme to form a covalent cleavage complex. The fluoroquinolone enrofloxacin inhibits relaxation of supercoiled DNA by VT in a Mg2+-dependent fashion. Further results indicate that the mechanism by which enrofloxacin inhibits VT is by preventing formation of the covalent complex which suggest that fluoroquinolones may be structurally optimized to target type IB topoisomerases.