Since being first reported in the late 1980ies, the Enterobacterium Edwardsiella ictaluri has rose in prevalence to become one of the two most damaging pathogens affecting the channel catfish industry. Despite this significance of the pathogen, understanding of the development of the disease, especially its route of entry into the host and the earlier stages of the infection, is still incomplete. A series of challenges were conducted using bioluminescent E. ictaluri either by infecting fish through immersion or topical application of the bacteria directly on the intact or abraded epithelium. This showed that abraded fish developed septicemia and died faster than non-abraded ones. Furthermore, results from a co-habitation challenge suggested that the bacterium induced septicemia through the skin instead of becoming water-borne. Finally, a histological technique was developed allowing the determination that the bacteria radiated from the initial skin infection site and penetrated deeper into the tissue as the challenge progressed. These results all suggest that site of abrasion on the skin can act as a route of entrance for the pathogen into the fish, a fact never previously reported. Transposon mutagenesis was also performed to construct a library of 1728 mutants. Screening of this library allowed us to identify 16 genes which inactivation lead to a decrease in the bacterium ability to colonize the epithelium or cause mortality. Sequencing of these genes allowed the identification of RstA/B, a regulator of invasion genes in Salmonella enterica Typhimurium, a putative ribonuclease, similar to a Shigella protein regulating the expression of adhesin and a protein that constitutes the second member of a newly discovered adhesin family. Finally, to investigate the development of the infection, fish were infected by bioluminescent E. ictaluri and sampled at various time points. At each time point, nine organs (gills, muscles, intestine, spleen, liver, stomach, heart, head kidney and trunk kidney) were sampled, and their bioluminescence was measured and half of these organs were homogenized, serial diluted, and plate counts determined. This allowed confirmation of a complex disease pathogenesis during ESC involving a period of intense reproduction in the spleen, anterior and posterior kidneys followed by a sharp increase in the levels of bacteria in the blood.

Acoustically, teleost swimbladders have been considered resonant underwater bubbles. Contrary results indicating that bladders are tuned less sharply than such a bubble have been explained by damping of surrounding fish tissue. Recent findings in toadfish and weakfish, however, suggest that the bladder is a highly damped structure and that the frequency of the fish sounds is deternlined as a forced response to sonic muscle movement rather than by resonance of the bladder. In this study I examined acoustics and material properties of swimbladders in Tilapia (an auditory generalist) and the channel catfish (an auditory specialist). The swimbladder was struck with a piezoelectric impact hammer. Amplitude and timing characteristics of bladder sound and displacement were compared for strikes of different amplitudes. Most of the first cycle of sound occurs during swimbladder compression, indicating that the bladder rapidly contracts and expands as force increases during the strike. Harder hits are shorter in duration generate a similar displacement duration with an increasing number of shorter cycles and a 12-15 dB increase in sound amplitude. The frequency spectrum is broad, and the dominant frequency is driven by the strike and not the natural frequency of the bladder. The displacement waveform varies between species catfish exhibit a greater structural stiffness and lower amplitude movement and higher sound amplitude for an equivalent hammer strike. Material properties (peak load, stress, strain, Young's modulus), fiber direction (catfish only), and structural stiffiless of bladders exhibit various patterns suggesting that the bladder walls are not uniform structures. Additionally thickness varies regionally. Notably Young's modulus in both species is similar despite large difference in stress and strain. The modulus of catfish bladder increased 1600-fold when dried. Finally the bladder of both species had a high water content averaging about 70%. These data suggest that viscous damping caused by water in the bladder wall is a major factor responsible for acoustic properties of the teleost swimbladder.

The commercial production of channel catfish (Ictalurus punctatus) is major industry in Mississippi. Infections of channel catfish with the digenetic trematode Bolbophorus damnificus have often been associated with heavy economic losses in the industry. To efficiently control transmission of this trematode, the avian hosts need to be identified. In the first study, two American white pelicans, two double-crested cormorants, two great blue herons, and two great egrets were fed channel catfish infected with B. damnificus metacercariae. The presence of Bolbophorus damnificus ova in pelican feces at three days post infection (dpi) indicated the pelicans had patent infections. Mature B. damnificus were recovered from the intestines of both pelicans at 21 dpi. No B. damnificus infections were observed in the other bird species. In a second study, 33 American white pelicans, 34 double-crested cormorants, 35 great blue herons, and 32 great egrets were collected in the Mississippi Delta. The prevalence of B. damnificus in the American white pelican was 93.9%, with an average of 158 B. damnificus found per bird (range 0-681). Bolbophorus damnificus was not found in any of the other bird species. The results of these two studies confirm that the AWPE is the only proven natural host for B. damnificus. In a third study, two previously undescribed cercariae were found infecting rams-horn snails in commercial catfish ponds. In challenge studies, channel catfish were exposed to both cercariae types. Only one type of cercariae (type I) was infective to channel catfish. The first evidence of type I metacercariae was seen histologically at 14 dpi and grossly at 21 dpi. Development continued until 120 dpi, when both gross examination and histology suggested that the metacercariae were mature. The type I metacercariae appeared to cause little host damage. Molecular analysis of the 18S rRNA gene region indicated that the type I cercariae and metacercariae may be a species of Clinostomum. The data generated in these three studies provides additional information that can be used in the development of efficacious management schemes to control digenetic trematodes infecting commercial catfish.

The catfish industry suffers losses primarily due to enteric septicemia of catfish and columnaris disease caused by Edwardsiella ictaluri and Flavobacterium columnare, respectively. Understanding the host-pathogen interactions is vital for prevention and eradication of these diseases. Hence, the overall objective of this study was to analyze whole cell proteomes of these two bacteria, and to determine the changes in E. ictaluri protein expression against in vitro iron-restriction and host serum treatment. High-throughput proteomic analysis of these bacteria was conducted using two-dimensional liquid chromatography followed by electrospray ionization tandem mass spectrometry (2-D LC ESI MS/MS) and two-dimentional gel electrophoresis coupled with matrix-assisted laser desorption/ionization time-oflight mass spectrometry (2-DE MALDI TOF/TOF). Identified proteins were clustered into functional groups using clusters of orthologous groups, and subcellular locations as well as possible functional relationships were determined. A total of 788 unique E. ictaluri and 621 unique F. columnare proteins were identified, which represented 12 and 28 pathways, respectively. Vertebrate hosts tend to chelate free iron of their body and make the environment hostile for bacteria. Hence, reduced availability of iron may cause significant stress for pathogens and is considered a signal that leads to alteration in virulent gene expression. Similarly, E. ictaluri might use the catfish blood stream effectively for quick systemic invasion. Hence, exposure to catfish serum components might reveal the ability of E. ictaluri to protect against host defense mechanisms. Using two-dimensional difference gel electrophoresis, responses of E. ictaluri due to in vitro iron-restriction and host serum treatment were determined. A total of 50 and 19 proteins were identified to be differentially expressed due to in vitro iron-restriction and catfish serum treatment, respectively. Among the differentially expressed proteins, several putative virulent determinants, immunogenic proteins, chaperones, and housekeeping genes were noted. To initiate functional studies, four differentially expressed E. ictaluri genes (lamB, glyS, malE, and sdhA) were mutated by inrame deletion. Results from this study provided experimental evidence for many predicted proteins. In addition, identification of differentially expressed proteins provided targets for further functional analysis, which could help elucidate pathogenic mechanisms of E. ictaluri.

Rhizophagus grandis is one of the rare examples of totally specific predators: it only preys upon the bark beetle Dendroctonus micans in its native range (Europe and Asia). In the course of coevolution in such predator-prey systems, prey are selected for avoidance and resistance mechanisms against predators. Previous laboratory trials suggested that R. grandis could lay more eggs in presence of exotic prey species with an ecology similar to that of D. micans, but which could not evolve resistances against this predator, such as the North American parasitic bark beetles D. punctatus and D. valens. The specificity of the association R. grandis-D. micans is thought to depend on chemical signals (kairomones) used by the predator to find a gallery of its prey and to adjust its oviposition to the number of prey larvae available inside. However, the exact nature of the chemical signals regulating R. grandis’ oviposition is still unknown. While these signals seem specific in nature, stimulation of the predator’s oviposition by exotic prey species suggests that specificity may be constrained by geography. Unlike most bark beetles, which kill living trees or feed on dying trees, D. micans completes its development on healthy trees as a true parasite. This niche is highly defended by tree toxicants (terpenes among others) to which both D. micans and R. grandis are resistant. Insects possess their own detoxification processes, but they may also rely on symbiotic microorganisms (bacteria and fungi) to contend with the specific constraints of their niche. In comparison to other bark beetles, microbial communities of parasitic bark beetles are yet poorly known, as in the case of D. micans and D. punctatus. Apart from detoxification, insect symbionts may provide nutritional supplementation, protection against pathogens, or components of the chemical communication, which affect the hosts as well as partners of the tritrophic relationship tree-bark beetle-natural enemies. The system R. grandis-D. micans is a unique opportunity to study several aspects of this coevolution which are poorly understood to date. The objectives of this thesis were:- (1) the comparison of the oviposition performances of R. grandis on the specific prey D. micans, and on the exotic prey D. punctatus and D. valens, in order to emphasize possible prey resistance mechanisms; - (2) the identification of the chemical signals regulating R. grandis’ oviposition, and the evaluation of their role in the specificity of the association with D. micans, in presence of the exotic prey D. punctatus and D. valens; - (3) the characterization of the bacterial and fungal communities associated with non-aggressive populations of the bark beetles D. micans, D. punctatus and D. valens, and investigating how these microbial symbionts may affect bark beetle hosts in respect of their ecology as well as the tritrophic relationship tree-bark beetle-natural enemies. Our results show that R. grandis laid equivalent numbers of eggs with the native prey D. micans and with the exotic prey D. punctatus and D. valens, which could illustrate that R. grandis’ specificity is constrained by geography but which does not emphasize possible prey resistance mechanisms against the long-standing predator. We identified robust candidates to the stimulation and inhibition of R. grandis’ oviposition which should be confirmed in bioassays. Finally, we report that the bacterial and fungal communities of non-aggressive populations of D. micans, D. punctatus and D. valens are mainly composed of widespread environmental Enterobacteria and yeasts, and we discuss the various ways they may influence bark beetle hosts in respect of their life histories including the attacked trees and their natural enemies. Overall, this thesis illustrates the need to encompass all levels of complexity, from prey symbionts and semiochemicals to predators, to study systems like R. grandis-D. micans. / Rhizophagus grandis est un des rares exemples de prédateurs totalement spécifiques :présent en Europe et en Asie, il se nourrit exclusivement du scolyte Dendroctonus micans. Dans de telles associations, la coévolution prédateur-proie mène à la sélection de mécanismes d’évitement ou de résistance chez la proie, contre le prédateur. De précédentes observations en laboratoire ont suggéré que R. grandis pouvait pondre davantage d’œufs en présence de proies exotiques d’écologie comparable à celle de D. micans, mais qui n’auraient pu développer de résistance à l’encontre de ce prédateur, à l’instar des scolytes parasitiques nord-américains D. punctatus et D. valens. Il semble que la spécificité de l’association R. grandis-D. micans repose sur des signaux chimiques (kairomones) qui permettent au prédateur de repérer les galeries de sa proie et d’y ajuster le nombre d’œufs déposé à la quantité de larves de proies disponible. Néanmoins, on ignore encore la nature exacte des signaux régulant l’oviposition de R. grandis. Le fait que l’oviposition soit stimulée par des proies exotiques indique que la spécificité de cette association pourrait être limitée par des barrières géographiques. Contrairement à la plupart des scolytes qui tue des arbres ou s’attaque à des arbres mourants, D. micans se développe intégralement sur des arbres en bonne santé, en véritable parasite. Cette niche est fortement défendue par les composés toxiques de l’arbre (en particulier les terpènes), auxquels à la fois D. micans et R. grandis sont résistants. Certains insectes possèdent leurs propres processus de détoxification, mais ils peuvent également bénéficier de l’aide de microorganismes symbiotiques (bactéries et champignons). En comparaison avec les autres scolytes, les communautés microbiennes associées aux scolytes parasitiques est très peu documentée, comme dans le cas de D. micans et D. punctatus. En dehors de la détoxification, les symbiotes d’insectes peuvent contribuer à leur nutrition, les protéger contre des pathogènes, ou intervenir dans la communication chimique, ce qui affecte leurs hôtes comme les autres acteurs de la relation tritrophique arbre-scolyte-ennemi naturel. L’association R. grandis-D. micans est une opportunité unique d’étudier des aspects méconnus de la coévolution. Les objectifs de cette thèse étaient de :- (1) comparer le nombre d’œufs pondu par R. grandis sur sa proie spécifique, D. micans, et sur les proies exotiques D. punctatus et D. valens, afin de mettre en évidence de possibles résistances ;- (2) identifier les signaux chimiques qui régulent l’oviposition de R. grandis, et évaluer leur rôle dans la spécificité de l’association avec D. micans, en présence des proies exotiques D. punctatus et D. valens ;- (3) caractériser les communautés bactérienne et fongique associées aux populations parasitiques des scolytes D. micans, D. punctatus et D. valens, et investiguer comment ces microorganismes symbiotiques peuvent influencer leurs hôtes, selon leurs contraintes écologiques, ainsi que leurs ennemis naturels et arbres-hôtes. Nos résultats révèlent une oviposition équivalente de R. grandis en présence de la proie native D. micans et des proies exotiques D. punctatus et D. valens, ce qui illustre que la spécificité de cette association pourrait être limitée par des barrières géographiques mais ne met pas en évidence de possibles résistances à l’encontre du prédateur de longue date. Nous avons identifié des candidats robustes à la stimulation et à l’inhibition de la ponte de R. grandis, et leur rôle devrait être confirmé par des bioessais. Enfin, nous rapportons que la communauté microbienne associée aux populations parasitiques des scolytes D. micans, D. punctatus et D. valens est principalement constituée d’Entérobactéries et de levures répandues, et nous discutons des différentes façons dont ces symbiotes peuvent affecter leurs hôtes et autres acteurs de la relation tritrophique arbre-scolyte-ennemi naturel, selon leurs écologies respectives. Dans son ensemble, cette thèse souligne l’importance de considérer tous les niveaux de complexité biologique, des microorganismes associés aux proies jusqu’aux prédateurs, afin d’étudier des systèmes comme R. grandis-D. micans. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished