Flavobacterium psychrophilum est le pathogène de la maladie de l’eau froide ou du syndromede l’alevin qui provoque des pertes économiques importantes dans les élevages de truite arc-en-cielen Aquitaine. Au cours de cette thèse, nous avons développé dans un premier temps un nouveau testqPCR TaqMan en duplex qui cible les gènes (i) adnr16S de F. psychrophilum et (ii) β-actine de latruite arc-en-ciel ; il permet de déterminer le taux d’infection de tissus contaminés au cours d’unemême analyse.Dans un second temps, nous avons cherché à objectiver trois approches préventives quipourraient être mises en oeuvre pour limiter l’infection à F. psychrophilum dans les élevages. Lacaractérisation de biocides classiquement utilisés par la filière aquacole a permis de proposer auxaquaculteurs des méthodes de désinfection réellement efficaces contre F. psychrophilum mais quin’affectent pas le développement des oeufs ni des futurs alevins. En parallèle, l’impact de la mise enplace de traitement probiotiques a été caractérisé à la surface des oeufs, à la surface du mucus et/oule long du système digestif, dès les tout premiers stades de développement des truites, sur le nombreet l’intensité des crises de flavobactériose dans les élevages. Deux traitements microbiens ont étéretenus : l’E112® (DOXAL-FRANCE, Pessac) et le Bactocell® (LALLEMAND, Toulouse). Enfin, troispréparations vaccinales ont été préparées (« Adjuvant », « Autovaccin » et « Membrane Externe ») ettestées pour leur capacité à protéger les poissons de la flavobactériose. / Flavobacterium psychrophilum is the causative agent of cold water disease or the rainbow troutfry syndrome which causes important economic losses in rainbow trout breedings. During this thesis,we firstly developed a new duplex Taqman qPCR test which targets the following genes (i) F.psychrophilum adnr16S and (ii) rainbow trout β-actine. This test enables to obtain the infection rate ofcontaminated tissues, during a same analysis.Then, we tried to objectivize three preventive approaches which could be implemented inbreedings to limit the infection by F. psychrophilum. The characterization of biocides used inaquaculture allowed proposing to the fish farmers effective methods of disinfection against F.psychrophilum. These methods do not affect the development of eggs or future fish larva. Moreover,the impact of the implementation of probiotics on the surface of eggs, on the surface of the mucusand/or the digestive system, from the very first steps of trout development has been characterized onthe number and the intensity of flavobacteriosis outbreaks in the breeding farms. Two probiotictreatments were selected: l’E112® (DOXAL-FRANCE, Pessac) and Bactocell® (LALLEMAND, Toulouse).Finally, three vaccine preparations (« Adjuvant », « Auto-vaccine » and « Outer Membrane ») wereprepared and tested for their capability to protect farmed fish against flavobacteriosis.

Acute ammonia toxicity in rainbow trout was studied. This was carried out by injecting fish with various concentrations of ammonia dissolved in isotonic saline. In order to approximate conditions of natural toxicity, where ammonia would enter the fish at the gills, without the additional problems of environmental factors, the ammonium solutions were injected via a cannula implanted in the dorsal aorta. To determine if a differential toxicity existed in fish in relation to high pH and low pH ammonium solutions, ammonium bicarbonate and ammonium chloride were chosen for injection. Hydrogen ion, and total ammonia, concentrations were measured in blood sampled from the dorsal aorta, both before and after injection. In order to determine if, during the course of the injection, normal excretion rates would remove all of the injected ammonia, total ammonia and hydrogen ion concentrations were measured in blood sampled from both the dorsal and ventral aortae, and rates of extraction of ammonia from blood at the gills were calculated. Ammonia is toxic to fish. There was no significant difference between the dose of NH₄Cl and NH₄HCO₃ which killed fish. Therefore, unlike mammals, fish exhibited no differential toxicity to the ammonium compounds tested. Injection of NH₄Cl decreased pHa and injection of NH₄HCO₃ increased pHa. Both compounds increased the total ammonia concentration in the blood. Although in water the fraction of ammonia which exerts the toxic effects is unionized ammonia, within the fish it is the ionized fraction which exerts the toxic effects. The same dose of ammonium killed fish, but NH₄HCO₃-injected fish which survived had a much higher concentration of unionized ammonia in the blood than NH₄Cl-injected fish which died. Ammonia extracted from the blood in control fish was about one-fifth the amount which killed fish. This, together with the measured increases in blood ammonia following injection, demonstrate that, although ammonia is a normal excretory product of rainbow trout, the trout cannot increase excretion rates sufficiently to rid themselves of an ammonia load. Symptoms observed in fish following injection of ammonium solutions led to the conclusion that ammonia acts at the neural level. / Science, Faculty of / Zoology, Department of / Graduate

Mitochondria were isolated from heart and mosaic muscle of rainbow trout (Salmo qairdneri R.). State 3 respiratory rates were determined at 5 and 15°C using pyruvate, malate, lactate, glutamate or acetyl—carnitine as substrate. The final three substrates were used to generate pH profiles. Pyruvate was oxidized at high rates in all cases, indicating good potential for aerobic carbohydrate metabolism. At 15°C, malate was an equally good substrate for heart mitochondria, while all substrates were oxidized at similar rates to pyruvate in muscle mitochondria. Maximal oxidation rates of heart mitochondria were greater than or equal to those of muscle. State 3 Q₁₀ for oxidation of most substrates in heart was approximately 2, except for malate which had a Q₁₀ of 3. Mitochondrial oxidation tended to be more sensitive to decreased temperature in muscle than in heart, particularly with respect to acetyl—carnitine and glutamate oxidation which in muscle had Q₁₀ values of 4 and 7, respectively. Based on RCR values at 5 and 15°C, there was no indication that membrane permeability to H⁺ ions was altered by a 10°C change in temperature in mitochondria from either tissue. At pH above 7.6 respiratory rates decreased with increasing pH. State 3 respiratory rate increased in heart mitochondria as pH decreased, below 7.6 while in muscle mitochondria, no such pH dependence was observed. RCR values were above 4 in all experiments except at high pH. Muscle mitochondria were the more sensitive to extreme pH with respect to RCR. Heart mitochondria had higher oxidative rates than those of muscle and were less sensitive to decreased temperature, in keeping with the greater oxidative demands of that tissue relative to mosaic muscle. Muscle mitochondria which typically face larger fluctuations in extramitochondrial pH in vivo than do those of heart, were less sensitive to pH in vitro. It was concluded that substrate utilization patterns and response to changes in temperature and extramitochondrial pH in the two mitochondrial populations was different and reflected both the intracellular environment of the mitochondria and the different needs of each tissue for aerobic energy supply. / Science, Faculty of / Zoology, Department of / Graduate

The role of carbonic anhydrase in carbon dioxide excretion and acid-base regulation in the rainbow trout, Salmo gairdneri has been investigated. While a significant amount of carbonic anhydrase was found in the blood of the trout, calculations based on red cell hemolysates suggest that the probable circulating levels of carbonic anhydrase activity in blood may not be sufficient to account for the observed carbon dioxide excretion. An analysis of carbonic anhydrase activity in whole blood from the trout revealed that intact fish erythrocytes, unlike mammalian erythrocytes totally fail to facilitate the dehydration of extracellular bicarbonate. The possible mechanism of this phenomenon has been examined; however the salient point was that fish red blood cells do not appear capable and therefore by implication apparently not necessary for the excretion of carbon dioxide at the gills of trout. The observed excretion of carbon dioxide in the trout was found to be accounted for by the gills and their compliment of carbonic anhydrase. This finding was based on the following observations. (1) Depletion of circulating blood carbonic anhydrase levels during severe anemia was without effect on carbon dioxide excretion rates or blood acid-base status. (2) Introduction of the carbonic anhydrase inhibitor, diamox into anemic fish produced a severe acid-base disturbance associated with a fall in observed carbon dioxide excretion. (3) Isolated perfused gill preparations excrete carbon dioxide at rates comparable to those observed in vivo from free swimming fish. (4) Carbon dioxide excretion in isolated gill preparations is abolished by diamox. The excretion of carbon dioxide in fish occurs via the movement of plasma bicarbonate into the branchial epithelium, where it is subsequently dehydrated into molecular carbon dioxide and excreted. A model is proposed and supportive evidence presented to account for the coupling of ionic exchange occurring across the gill with carbon dioxide excretion. The proposed model distinguishes between control of plasma hydrogen ion activity and regulation of plasma total carbon dioxide concentration per se. The functional significance of this pattern of carbon dioxide excretion for aquatic animals is discussed along with the implications for air breathing fish. / Science, Faculty of / Zoology, Department of / Graduate

<p> Knowledge into uptake rate and interactions of the metals via the gastrointestinal tract of freshwater fish is vital, in order to provide tools to protect and to sustain aquatic biota. An in vitro stomach and gut sac technique was utilized to investigate uptake rates of essential metals (copper, zinc and nickel) and non-essential metals (cadmium, lead and nickel) at luminal concentrations of 50fJ.M via the gastrointestinal tract Metals had no effect on the fluid transport rates via the gastrointestinal tract except for copper at the stomach. The stomach emerged as small but important site for metal absorption and interaction. Essential metals were absorbed at approximately the same rate as nonessential metals via the gastrointestinal tract. Copper, zinc, nickel, silver, and lead showed statistical correlation between rate of absorption and mucus binding via gastrointestinal tract, an important first finding for the development of a Biotic Ligand Model (BLM) for the gastrointestinal tract of trout. </p> <p> There was an antagonistic effect of calcium on cadmium uptake at the stomach but not at the intestine. Zinc and calcium exhibited synergistic interaction at the stomach but no interaction at the intestine. These results showed the possibility of the transporters DMTl to mediate copper and cadmium uptake via the intestine; hZip 2 to mediate copper and zinc uptake at the stomach; and Mzip 4 or ZTLl to mediate zinc uptake via the gastrointestinal tract of trout. These results can be used to develop a BLM for the gastrointestinal tract of fish. </p> / Thesis / Master of Science (MSc)

<p> Juvenile rainbow trout (Oncorhynchus myldss) were chronically exposed (90 days) in synthetically softened water ([Ca2+]=50, [Na+]=100 J.tequiv·L-1), to sublethal low pH (5.2) and a simulated global warming scenario (+20C), added to the natural summer thermal cycle ofinshore Lake Ontario. Two studies were conducted over the periods June September 1993 and 1994 in order to examine the effects ofthe sublethal stressors under conditions ofunlimited food ration (satiation feeding twice daily= 10% dry body weight·day-1), and a limited food ration (4% dry body weight·day-1) respectively. The addition of 2°C and sulphuric acid was designed to result in four treatment conditions: i) Control conditions; ambient water temperature and pH; ii) ambient water temperature and pH 5.2; iii) simulated global warming at ambient pH; iv) a combination of simulated global warming and pH 5.2. Year to year variation in temperature provided the trout in the satiation-fed study with an ambient ~peraturerange of 13-24°C, while those in the limited ration study experienced a range of 16.5-21°C. Consequently, the trout in the treatments with an additional 2°C experienced temperatures close to the upper incipient lethal level, particularly the trout in the satiation feeding study. Apparent specific dynamic action raised routine metabolic rates in all treatments to -75% and 55% M02(max) in the satiation and limited ration studies respectively, the difference of 20% indicating the influence of an unlimited feeding regime on metabolism. Trout in the satiation feeding study increased in wet body mass by 30-50 g, while trout in the limited ration study increased by only 3-4 g. Whole body proximate composition of the trout fed to satiation changed over time with large increases in lipid content, small increases in protein content, and compensating decreases in water content in all treatments. No such changes occurred in the limited ration trout, although whole body lipid and carbohydrate were highly variable. In both studies, the addition of20C resulted in decreased growth, with an accompanying depression in appetite in the satiation fed trout, especially at peak temperatures. Surprisingly, trout exposed to low pH alone exhibited improved growth in both treatments. Energy budgets indicated that the addition of 2°C reduced gross energy intake and increased fecal (and Ulllidentified) energy losses resulting in lowered conversion efficiencies, while in limited ration trout, energy expenditure was slightly higher. Trout exposed to low pH exhibited higher gross energy intake and gain, and more efficient energy conversion under unlimited food conditions while trout with limited rations expended the least metabolic energy and exhibited lower nitrogen energy losses. The surprising lack of ionoregulatory disturbance in these pH 5.2 exposed trout in both studies suggests tbat the availability of NaCl in the diet was compensating for branchial ion losses, and perhaps driving appetite in the satiation fed trout Where ration was limited, reductions in activity level may have contributed to energy conservation and consequently improved growth. A 22Na+ flux experiment conducted at the end of the limited ration exposure, in which the fish were exposed to a challenge concentration of H+ (pH 4.2), provided evidence for improved recovery of ionoregulatory balance in trout which had been chronically exposed to low pH. Overall, the combination ofincreased global temperatures and sublethal low pH results in increased physiological costs for juvenile rainbow trout, most noticeably when summer temperatures peak. Ration level is of integral importance when considering the degree of impact of such environmental conditions. </p> / Thesis / Master of Science (MSc)

For animals that live in a reasonably variable environment the capacity for learning and memory allow them to adapt to the changes they experience. Ecological factors that vary between habitats can affect a range of learning behaviours. Less attention has been directed at how this variation may affect memory processes, or how different ecological variables might interact when shaping cognition and behaviour. Therefore one aim of this thesis was to investigate how different ecological variables shape memory abilities and to test whether those same variables affect other related behaviours such as learning. In order to test this, I selected natural populations of a temperate freshwater fish, the three-spined stickleback (Gasterosteus aculeatus) from pond and river habitats that were proposed to differ in predation pressure, and assayed their learning, memory and other behavioural traits. Pond and river populations differed in their memory and orientation behaviour. An interaction between pond/river habitat and predation pressure affected learning rate, and a similar interaction affected temperament behaviours. Two further studies were conducted to address how captive rearing environments and typical handling procedures affect behaviour in different species. Rearing environment affected memory, but not learning or temperament behaviours in three-spined sticklebacks. Handling caused stress responses in three-spined sticklebacks, Panamanian bishops (Brachyraphis episcopi) and Rainbow trout (Oncorhynchus mykiss), but handling with a water filled scoop compared to a traditional dip-net decreased these responses in three-spined sticklebacks and Panamanian bishops, and also affected behaviour in Panamanian bishops. The results presented in this thesis suggest that ecological variables play a substantial role in shaping learning, memory and other behavioural traits in fish, and highlight the utility of behavioural assays in answering welfare-based questions.

Memoria para optar al título profesional de Ingeniero Agrónomo / Dentro de la acuicultura chilena, una de las especies más relevantes es la trucha arcoíris (Oncorhynchus mykiss), ocupando actualmente el cuarto lugar en producción de de especies comerciales en Chile con un volumen de 94.717 toneladas, registrando un retorno de 606 millones de dólares el año 2015. Los estudios relacionados a la genética de este animal se centran en la búsqueda de genes asociados a rasgos de interés productivo y su aplicación para el mejoramiento de su cultivo. En la actualidad, un área especialmente importante es el desarrollo de dietas más amigables con el medio ambiente, que se enfocan en el remplazo de insumos como proteínas y aceites de origen animal por otros de origen vegetal.

Memoria para optar al título Profesional de Ingeniero Agrónomo. Mención: Producción Animal / En el presente estudio se estimaron heredabilidades, correlaciones genéticas y fenotípicas para un stock de truchas arcoíris que muestran el rasgo doble ciclo reproductivo anual. Se evaluaron las variables de presencia de uno o dos desoves en el transcurso de un año (DCRA), el tiempo que pasa entre dos desoves en una misma hembra (longitud del ciclo reproductivo, LCR), los pesos de las hembras al primer desove, al marcaje, a los 380 y 625 días, las edades de las hembras en su primer y segundo desove. Todo esto con el objetivo de estudiar y comparar la conducta productiva y reproductiva de truchas arcoíris doblecicladoras y monocicladoras. Se trabajó con una población de trucha arcoíris (n = 1570), mantenida en piscícola Huililco Ltda. En la estimación de heredabilidades se utilizó un modelo mixto, el que fue resuelto utilizando el programa DFREML, mientras que en la estimación de las correlaciones genéticas se utilizó el programa MTDFREML. Además, se estimaron las correlaciones fenotípicas junto con la respuesta correlacionada a la selección para las variables de peso al seleccionar solo para LCR. Las heredabilidades obtenidas para presencia de DCRA y LCR fueron medias: 0,19 ± 0,10 y 0,28 ± 0,11, respectivamente; y superiores a lo que se espera para caracteres reproductivos. Esto es positivo, pues permite obtener respuesta a la selección para estos rasgos. Las heredabilidades para las distintas variables de peso y edad, fluctuaron dentro de lo esperado, entre 0,15 y 0,26, excepto para peso al marcaje que dio 0,93 ± 0,12, que puede estar sobreestimada por efectos del ambiente común en la fase de alevinaje. Las correlaciones genéticas y respuestas correlacionadas entre LCR y los pesos fueron positivas, indicando que a mayor LCR se observa un mayor peso. En general las truchas monocicladoras presentaron mayores pesos y edades al primer y segundo desove que las doblecicladoras. Sin embargo, no se observaron diferencias en el peso a los 625 días de edad, indicando que el fenómeno DCRA al parecer no disminuiría los pesos al momento de la cosecha. / In the present study heritabilities, genetic and phenotypic correlations for a rainbow trout stock that show the characteristic double annual reproductive cycle were estimated. Variables evaluated were presence of one or two spawns in the course of one year (DCRA), days between two spawn in the same female (length of the reproductive cycle, LRC), weights of females at first spawning, to tagging, to the 380 and 625 days, and ages of the females in their first and second spawning. The main objective was to study and compare the productive and reproductive behavior of the doublecycler and monocycler rainbow trout. We worked with a rainbow trout population (n = 1570), kept at Huililco Ltd. farm. In the heritability and genetic correlation estimations a mixed model using a derivative free algorithm of restricted maximum likelihood (DFREML) was used. In addition, phenotypic correlations were estimated with correlated response to selection for weight variables selecting only for LCR. Heritabilities obtained for the presence of DCRA and LRC were medium: 0.19 ± 0.10 and 0.28 ± 0.11, respectively, and higher than expected for reproductive characters. These estimations of heritabilities allow higher response to the selection for both traits. Heritabilities for the different age and weight variables fluctuate within the expected range between 0.15 and 0.26, except for weight marking that gave 0.93 ± 0.12, which could be due to common environmental effects in alevín stage. Genetic correlations between LRC and correlated responses and weights were positive, indicating that the higher LRC shows a greater weight. The trout monocycler generally had higher weights and ages at first and second spawning that doublecycler. However, no differences in weight at 625 days of age were observed, indicating that the phenomenon does not seem to diminish DCRA weights at the time of harvest.

There is an increasing scientific acceptance that fish may feel some sort of fear, pain and distress, which in turn feeds a growing concern for their welfare. Humans impact the wellbeing of a large number of fish in various ways, one of them being through research. Welfare legislation in the UK demand welfare considerations for all animals used in scientific procedures. Furthermore, welfare and enrichment needs for fish are included in the Appendix A of the European Convention for the Protection of Vertebrate Animals used for Experimental and Scientific Purposes. As fish are extensively used in research, changing their housing and husbandry to improve welfare is of importance, since fish kept in laboratories are most likely subjected to impoverished environments. Although enrichment programs have been shown to improve health and welfare in various animal species, little is known of their potential for application to juvenile rainbow trout. How best to improve barren experimental tanks for female juvenile rainbow trout used in regulatory research was the broad aim of this PhD. In this thesis, three enrichment strategies for rainbow trout have been examined, using physiological and behavioural welfare indicators. The first study assessed the effects of semitransparent shelters on trout welfare, and a clear message became evident; that shelters of this design should not be considered enrichment for rainbow trout as they had several significant negative impacts, indicating chronic stress in fish from shelter tanks relative to fish in a barren environment. The second study investigated impacts of reduced visual access to conspecifics in the same tank. Habitats with low visual contact between individuals have been suggested to reduce aggression for a range of species, and I have shown that visual barriers appeared to be beneficial to trout as well. The final experiment evaluated effects of high and low water currents on the wellbeing of rainbow trout, and results indicated increased fish welfare when water currents were supplied.