Microsporidia are a group of obligate intracellular fungal parasites that infect a wide range of vertebrates and invertebrates, and are of economic and academic interest. Some areas of their economic impact are in aquaculture where they can infect salmon and other fish species. In agriculture they have been considered as control agents for insect pests, but more importantly as likely contributing to colony collapse disorder of bees. As an academic topic, microsporidia are fascinating because they are the smallest and simplest eukaryotic cells and require eukaryotic host cells in order to complete their life cycle. Therefore one research avenue that moves forward both economic and academic interests is to use cultures of animal cells to support the growth and development of the microsporidia life cycle, including the production of spores. Although the use of animal cell cultures for studying the microsporidia of insect and mammals has a fairly large literature, fish cell cultures have been employed less often but have had some successes as reviewed in this thesis. Very short-term primary cultures have been used to show how microsporidia spores can modulate the activities of phagocytes. The most successful microsporidia/fish cell culture system has been relatively long-term primary cultures of salmonid leukocytes for culturing Nucleospora salmonis. Surprisingly, this system can also support the development of Enterocytozoon bienusi, which is of mammalian origin. Some modest success has been achieved in growing Pseudoloma neurophilia on several different fish cell lines. The eel cell line, EP-1, appears to be the only published example of any fish cell line being permanently infected with microsporidia, in this case Heterosporis anguillarum. These cell culture approaches promise to be valuable for describing the growth and development of the microsporidia and for documenting the responses of fish cells to infection. In this thesis, cell lines from warm water fish, goldfish, fathead minnow and zebrafish, and a coldwater species, rainbow trout, were explored as potential cellular hosts of two microsporidia species that have never been grown or associated with fish before. One is Anncaliia algerae, which is an aquatic microsporidium that most commonly infects mosquitoes. This microsporidia is one of the easiest species to grow in mammalian cells, with the rabbit kidney cell line, RK 13, being the most documented culture system. The other is Nosema apis, which is a pathogen of bees and for which few cell culture systems exist. The ability of warm water fish cell lines to support the life cycle of A. algerae was investigated first. Spores were purified from RK-13 cultures and added to cell lines from three warm water species as well as to an insect cell line. The cell lines were GFSK-S1 and GFB3C- W1 from goldfish skin and brain respectively, ZEB2J from zebrafish embryos, FHMT-W1 from fathead minnow testis, and Sf9 from ovaries of a fall armyworm moth. All cultures were maintained at 27 °C. Infection was judged to have taken place by the appearance of sporonts and/or spores in cells and occurred in all cell lines. Spores were also isolated from ZEB2J cultures and used to successfully infect new cultures of ZEB2J, RK-13 and Sf9. These results suggest that cells of a wide range of vertebrates support A. algerae growth in vitro and fish cells can produce spores infectious to cells of mammals, fish and insects. As ZEB2J was the most characterized of the fish cell lines and supported good A. algerae growth, this cell line was used in further studies described below to compare the efficacy of antimicrosporidial drugs and to test whether fish cells could support N. apis growth, but first A. algerae growth at lower temperatures was explored with cell lines from a coldwater fish. Cultures of cell lines from rainbow trout gill, RTgill-W1, and brain, RTbrain-W1, at 9, 18 and 21°C were evaluated for their ability to support the development of A. algerae. For up to 8 days after the addition of spores, living and DAPI stained cultures were examined by phase-contrast microscopy, allowing the identification of the meront, sporont, and spore stages in cultures at 18 and 21 °C. Meronts and sporonts were both spindle-shaped, but relative to meronts, sporonts were darker under phase contrast and brighter after DAPI staining. Spores were egg-shaped, phase- bright and intensely DAPI stained. These stages could not be identified conclusively in cultures at 9 °C, but their appearance at 18 °C sets a new low temperature for the growth of this species. The growth of A. algerae at room temperature allowed living cultures to be observed conveniently and videoed with a proprietary instrument, the Riveal microscope (www.quorumtechnologies.com). With this microscope, the development of A. algerae life cycle stages at room temperature was confirmed plus for the first time meront division and intracellular germination were captured on video. Spore germination in the absence of host cells and in response to 3 percent hydrogen peroxide was also observed by Riveal microscopy and for first time an abnormal germination phenomenon was clearly documented: polar tubes were extruded but the spore bodies retained the nuclei. ZEB2J cultures that had been infected with Anncaliia algerae spores were used as an in vitro test system to evaluate the curative actions of albendazole, fumagillin, and three fluoroquinolones; ciprofloxacin, norfloxacin, and ofloxacin. For each drug at concentrations above 50 µg/ml, the viability of ZEB2J cell declined sharply so concentrations of 10 and 20 µg/ml were studied. At these concentrations the drugs had little effect on the morphology and germination A. algerae spores. Each of the fluoroquinolones failed to prevent A. algerae from infecting ZEB2J cells and from growing to the same extent as in untreated ZEB2J cultures. Adding albendazole or fumagillin to cultures did not prevent A. algerae from infecting ZEB2J cells but impeded the growth and accumulation of A. algerae life-cycle stages. However, albendazole treatments caused a significant fraction of the ZEB2J cells to have nuclear abnormalities. Fumagillin reduced the intensity of infections within a ZEB2J cell, although the number of infected cells in a culture was not reduced. Over 5 days of infection with A. algerae the accumulation of ZEB2J cells in cultures was reduced but fumagillin treatment restored the accumulation to control levels. These results suggest that fumagillin has some potential as a treatment for A. algerae infections. ZEB2J was exposed to Nosema apis spores from the western honey bee (Apis mellifera). Bees were collected from hives that had been naturally infected and confirmed polymerase chain reaction (PCR) to have N. apis. Frozen bees were crushed in water to yield a mixture of bee parts, pollen grains, yeast, and microsporidial spores. The mixture was filtered and then centrifuged through Percoll to produce a pellet of spores that was resuspended in L-15 with 10 percent fetal bovine serum (FBS). Aliquots of this were added to ZEB2J cultures. Cultures were observed periodically for up to 24 days with a combination of phase contrast microscopy and of fluorescence microscopy, usually after staining with 4’,6-diamidino-2-phenylindole (DAPI). Although earlier life cycle stages were not observed, structures that were concluded to be either sporonts, sporoblasts and/or spores were seen, but these were in less than 5 percent of the fish cells. These N. apis life cycle stages had grown in ZEB2J because some appeared to be inside the cells and often they were arranged around the nucleus of the host cell rather than being randomly distributed in cultures. Despite repeated rinsing over a three week period, all cultures were ultimately lost due to yeast from the original spore preparations over growing the fish cell cultures. The overarching observation of this thesis is that fish cells in culture have been shown for the first time to support the growth A. algerae, and possibly N. apis. This suggests that the cells of vertebrates might support the growth of a wide range of microsporidia species that normally are associated with insects. In turn this suggests restriction of a microsporidial species to a particular animal group is unlikely accomplished at the cellular level but through physiological systems expressed at the organismal level and disturbances in these systems might lead to infections in new groups of animal hosts. The overarching observation of this thesis has two general implications for future studies. Firstly, for studying the expression of antimicrosporidia mechanisms in fish cells, the ZEB2J/A. algerae co-culture system promises to be useful. Secondly, for microsporidia species that are difficult to grow in culture, cell lines from a wide range of vertebrate and invertebrate species should be explored and one possibility for N. apis is fish cells.

Microsporidia are a group of obligate intracellular fungal parasites that infect a wide range of vertebrates and invertebrates, and are of economic and academic interest. Some areas of their economic impact are in aquaculture where they can infect salmon and other fish species. In agriculture they have been considered as control agents for insect pests, but more importantly as likely contributing to colony collapse disorder of bees. As an academic topic, microsporidia are fascinating because they are the smallest and simplest eukaryotic cells and require eukaryotic host cells in order to complete their life cycle. Therefore one research avenue that moves forward both economic and academic interests is to use cultures of animal cells to support the growth and development of the microsporidia life cycle, including the production of spores. Although the use of animal cell cultures for studying the microsporidia of insect and mammals has a fairly large literature, fish cell cultures have been employed less often but have had some successes as reviewed in this thesis. Very short-term primary cultures have been used to show how microsporidia spores can modulate the activities of phagocytes. The most successful microsporidia/fish cell culture system has been relatively long-term primary cultures of salmonid leukocytes for culturing Nucleospora salmonis. Surprisingly, this system can also support the development of Enterocytozoon bienusi, which is of mammalian origin. Some modest success has been achieved in growing Pseudoloma neurophilia on several different fish cell lines. The eel cell line, EP-1, appears to be the only published example of any fish cell line being permanently infected with microsporidia, in this case Heterosporis anguillarum. These cell culture approaches promise to be valuable for describing the growth and development of the microsporidia and for documenting the responses of fish cells to infection. In this thesis, cell lines from warm water fish, goldfish, fathead minnow and zebrafish, and a coldwater species, rainbow trout, were explored as potential cellular hosts of two microsporidia species that have never been grown or associated with fish before. One is Anncaliia algerae, which is an aquatic microsporidium that most commonly infects mosquitoes. This microsporidia is one of the easiest species to grow in mammalian cells, with the rabbit kidney cell line, RK 13, being the most documented culture system. The other is Nosema apis, which is a pathogen of bees and for which few cell culture systems exist. The ability of warm water fish cell lines to support the life cycle of A. algerae was investigated first. Spores were purified from RK-13 cultures and added to cell lines from three warm water species as well as to an insect cell line. The cell lines were GFSK-S1 and GFB3C- W1 from goldfish skin and brain respectively, ZEB2J from zebrafish embryos, FHMT-W1 from fathead minnow testis, and Sf9 from ovaries of a fall armyworm moth. All cultures were maintained at 27 °C. Infection was judged to have taken place by the appearance of sporonts and/or spores in cells and occurred in all cell lines. Spores were also isolated from ZEB2J cultures and used to successfully infect new cultures of ZEB2J, RK-13 and Sf9. These results suggest that cells of a wide range of vertebrates support A. algerae growth in vitro and fish cells can produce spores infectious to cells of mammals, fish and insects. As ZEB2J was the most characterized of the fish cell lines and supported good A. algerae growth, this cell line was used in further studies described below to compare the efficacy of antimicrosporidial drugs and to test whether fish cells could support N. apis growth, but first A. algerae growth at lower temperatures was explored with cell lines from a coldwater fish. Cultures of cell lines from rainbow trout gill, RTgill-W1, and brain, RTbrain-W1, at 9, 18 and 21°C were evaluated for their ability to support the development of A. algerae. For up to 8 days after the addition of spores, living and DAPI stained cultures were examined by phase-contrast microscopy, allowing the identification of the meront, sporont, and spore stages in cultures at 18 and 21 °C. Meronts and sporonts were both spindle-shaped, but relative to meronts, sporonts were darker under phase contrast and brighter after DAPI staining. Spores were egg-shaped, phase- bright and intensely DAPI stained. These stages could not be identified conclusively in cultures at 9 °C, but their appearance at 18 °C sets a new low temperature for the growth of this species. The growth of A. algerae at room temperature allowed living cultures to be observed conveniently and videoed with a proprietary instrument, the Riveal microscope (www.quorumtechnologies.com). With this microscope, the development of A. algerae life cycle stages at room temperature was confirmed plus for the first time meront division and intracellular germination were captured on video. Spore germination in the absence of host cells and in response to 3 percent hydrogen peroxide was also observed by Riveal microscopy and for first time an abnormal germination phenomenon was clearly documented: polar tubes were extruded but the spore bodies retained the nuclei. ZEB2J cultures that had been infected with Anncaliia algerae spores were used as an in vitro test system to evaluate the curative actions of albendazole, fumagillin, and three fluoroquinolones; ciprofloxacin, norfloxacin, and ofloxacin. For each drug at concentrations above 50 µg/ml, the viability of ZEB2J cell declined sharply so concentrations of 10 and 20 µg/ml were studied. At these concentrations the drugs had little effect on the morphology and germination A. algerae spores. Each of the fluoroquinolones failed to prevent A. algerae from infecting ZEB2J cells and from growing to the same extent as in untreated ZEB2J cultures. Adding albendazole or fumagillin to cultures did not prevent A. algerae from infecting ZEB2J cells but impeded the growth and accumulation of A. algerae life-cycle stages. However, albendazole treatments caused a significant fraction of the ZEB2J cells to have nuclear abnormalities. Fumagillin reduced the intensity of infections within a ZEB2J cell, although the number of infected cells in a culture was not reduced. Over 5 days of infection with A. algerae the accumulation of ZEB2J cells in cultures was reduced but fumagillin treatment restored the accumulation to control levels. These results suggest that fumagillin has some potential as a treatment for A. algerae infections. ZEB2J was exposed to Nosema apis spores from the western honey bee (Apis mellifera). Bees were collected from hives that had been naturally infected and confirmed polymerase chain reaction (PCR) to have N. apis. Frozen bees were crushed in water to yield a mixture of bee parts, pollen grains, yeast, and microsporidial spores. The mixture was filtered and then centrifuged through Percoll to produce a pellet of spores that was resuspended in L-15 with 10 percent fetal bovine serum (FBS). Aliquots of this were added to ZEB2J cultures. Cultures were observed periodically for up to 24 days with a combination of phase contrast microscopy and of fluorescence microscopy, usually after staining with 4’,6-diamidino-2-phenylindole (DAPI). Although earlier life cycle stages were not observed, structures that were concluded to be either sporonts, sporoblasts and/or spores were seen, but these were in less than 5 percent of the fish cells. These N. apis life cycle stages had grown in ZEB2J because some appeared to be inside the cells and often they were arranged around the nucleus of the host cell rather than being randomly distributed in cultures. Despite repeated rinsing over a three week period, all cultures were ultimately lost due to yeast from the original spore preparations over growing the fish cell cultures. The overarching observation of this thesis is that fish cells in culture have been shown for the first time to support the growth A. algerae, and possibly N. apis. This suggests that the cells of vertebrates might support the growth of a wide range of microsporidia species that normally are associated with insects. In turn this suggests restriction of a microsporidial species to a particular animal group is unlikely accomplished at the cellular level but through physiological systems expressed at the organismal level and disturbances in these systems might lead to infections in new groups of animal hosts. The overarching observation of this thesis has two general implications for future studies. Firstly, for studying the expression of antimicrosporidia mechanisms in fish cells, the ZEB2J/A. algerae co-culture system promises to be useful. Secondly, for microsporidia species that are difficult to grow in culture, cell lines from a wide range of vertebrate and invertebrate species should be explored and one possibility for N. apis is fish cells.

Lors de la mise en place d’une interaction hôte-parasite, les principales barrières à franchir sont les mêmes quel que soit l’hôte considéré. Il faut que le parasite rencontre l’hôte puis qu’il soit capable d’échapper à ses systèmes de défenses. Pour cela, au-cours de la coévolution, les parasites ont ainsi développé des stratégies moléculaires leur permettant de pirater les réseaux de l’hôte, menant à l’établissement d’un dialogue moléculaire. Les microsporidies, qui sont des parasites intracellulaires obligatoires, ont, du fait de leur forte dépendance vis-à-vis de leur hôte, probablement développé des stratégies très poussées de piratage. L’objectif de cette thèse a été d’initier le décryptage du dialogue moléculaire qui s’établit entre une microsporidie et son hôte à deux niveaux d’intégration. Au niveau cellulaire, l’étude de la réponse protéique de cellules fibroblastiques humaines à l’infection par Anncaliia algerae a permis de suggérer l’existence d’une stratégie originale de leurre de l’hôte grâce à l’expression d’un élément transposable. Au niveau tissulaire, l’étude de la réponse protéique d’intestin d’abeilles infectées par Nosema ceranae a révélé une perturbation de l’homéostasie du tissu intestinal pouvant être à l’origine d’un impact négatif de l’infection sur le taux de renouvellement de l’épithélium. Un suivi du taux de multiplication des cellules souches intestinales lors d’une cinétique d’infection nous a permis de conforter cette hypothèse. Le suivi de l’expression de gènes impliqués dans les voies de signalisation contrôlant ce taux de renouvellement a confirmé une perturbation de l’homéostasie intestinale de l’abeille. Cependant, les mécanismes par lesquels les microsporidies arrivent à se développer chez leurs hôtes ne sont pas connus et méritent d’être explorés. / Within the host-parasite interaction, the parasite need to cross the same barriers whatever the host considered. First, the parasite has to meet its host and to escape its defense systems. For this purpose, the parasites have developed, during coevolution, molecular strategies allowing them to hijack the host networks, leading to the set-up of a real molecular crosstalk. Microsporidia, which are obligate intracellular parasites, have probably developed very sophisticated strategies to hijack their host cell functions as they are strongly dependent to their hosts. The objective of this thesis was to pave the way to the deciphering of the molecular dialogue that takes place during the interaction between a microsporidia and its host, at two different integration levels. At the cellular level, the study of the proteome response of human fibroblast cells to the infection by Anncaliia algerae allowed us to suggest the existence of a lure strategy used by A. algerae to bypass the host response. At the tissue level, the study of the midgut proteome response of honeybees infected by Nosema ceranae revealed a disturbance of the intestinal homeostasis. These results lead us to the hypothesis of a negative impact of the infection on the midgut epithelium renewal rates. This assumption was confirmed by a monitoring of the multiplication rate of intestinal stem cells during a kinetics of infection and of the expression of genes implicated in the signaling pathways controlling this renewal. However, the underlying mechanisms allowing microsporidia to develop in hosts are not known and deserve to be explored.