Insulin resistance is central to the pathogenesis of non-alcoholic fatty liver disease (NAFLD) and type 2 diabetes mellitus (T2DM). Intrahepatic triglyceride (IHTG), the primary feature of NAFLD, strongly predicts insulin resistance in the liver and peripheral (skeletal muscle and adipose) tissues. Hepatokines (e.g. fibroblast growth factor 21 (FGF21), leukocyte cell-derived chemotaxin 2 (LECT2), follistatin, selenoprotein P, and fetuin-A) are liver-derived proteins with capacity to exert endocrine effects and may potentially modulate the link between IHTG and peripheral insulin sensitivity/glycaemic control. Exercise is integral to the management of NAFLD and T2DM, with evidence suggesting that high-intensity exercise may provide the greatest benefits. Chapter 4 of this thesis demonstrates that, in individuals without chronic metabolic disease, plasma concentrations of FGF21 and LECT2 are higher, and follistatin lower, in individuals with overweight or obesity compared with normal weight individuals. Furthermore, FGF21 and follistatin are transiently elevated for up to 6 h after acute aerobic exercise (60 min at 60% V̇O2 peak). The response of follistatin to acute moderate-intensity exercise is also present in individuals with impaired glucose regulation (Chapter 5), but the response of FGF21 is abolished. A single bout of low-volume high-intensity interval training has no effect on FGF21, follistatin or fetuin-A in individuals with dysglycaemia (Chapter 5). Chapter 6 demonstrates that six weeks of sprint interval training (SIT) is feasible for men with NAFLD and reduces IHTG despite no change in body weight. Peripheral insulin sensitivity tends to increase after SIT but hepatic insulin sensitivity and circulating hepatokines remain unchanged. Through meta-analyses, Chapter 7 confirms that exercise training reduces IHTG, even in the absence of weight loss. However, the magnitude of this effect is greater when weight loss occurs and benefits increase proportionally. Exercise training improves basal hepatic insulin sensitivity, but evidence in this area is currently limited (Chapter 7). Collectively, the studies in this thesis demonstrate that some hepatokines may be sensitive to acute and chronic changes in energy metabolism. However, further evidence is required before definitive statements can be made. Exercise training, including SIT, has the potential to reduce IHTG in men with NAFLD, even in the absence of weight loss. However, the greatest benefits on IHTG will likely be elicited when exercise training is performed in combination with dietary energy restriction to elicit sustained reduction in body weight.

Fibroblast Growth Factor 21 (FGF21) is a hormone that is produced from the liver which has pleiotropic effects. Physiologically, FGF21 increases energy expenditure, increases glucose uptake, enhances glucose tolerance, and increases peripheral insulin sensitivity. Pharmacologically, FGF21 reverses obesity and diabetes in animal models and significantly improves metabolic profiles in humans through unknown mechanisms. We hypothesized that the physiological actions of FGF21 may provide insights to explain FGF21’s beneficial pharmacological effects. The overall theme of this work was to identify the elusive mechanism by which FGF21 regulates energy homeostasis. In chapter 1, I review some adipokines and hepatokines that regulate energy homeostasis. In chapter 2, I provide background on fibroblast growth factors (FGFs), metabolic FGFs, and the tissue-specific effects of FGF21. In chapter 3, I will review the role of growth factors in thermoregulation. In chapter 4, we use tissue-specific loss of function models to investigate the trajectory of FGF21’s thermogenic effects during prolonged cold. In chapter 5, we specifically address the necessity and sufficiency of FGF21 signaling directly to adipose tissue, and the contribution of the adipokine adiponectin in mediating FGF21’s metabolic effects. In chapter 6, I summarize our results, reflect upon the ramifications of these results, and briefly address potential future experiments given our results on the physiological and pharmacological actions of FGF21 in adipose tissues.

L’hépatokine FGF21 (Fibroblast Growth factor 21) joue un rôle primordial dans le contrôle de l’homéostasie énergétique. Des études chez l’Homme et l’animal mettent en évidence ses effets bénéfiques dans la lutte contre l’hyperglycémie, la dyslipidémie et l’obésité. Connue pour être induite en réponse au jeûne par le récepteur nucléaire PPARα (Proliferator Activated Receptor α), des études récentes suggèrent l’implication du facteur de transcription ChREBP (Carbohydrate Responsive Element Binding Protein) dans la réponse nutritionnelle de FGF21. Dans ce contexte, les objectifs de thèse ont été : 1) d’obtenir une meilleure compréhension de la régulation de FGF21 dans le foie par le jeûne et le glucose via les acteurs moléculaires ChREBP et PPARα ; 2) de déterminer la relevance physiologique de l’axe ChREBP-PPARα-FGF21 en réponse au glucose. Nos résultats mettent en évidence que l’expression hépatique de ChREBP est nécessaire à l’induction de FGF21 en réponse au glucose in vitro et in vivo. De manière inattendue, lorsque l’expression de PPARα est spécifiquement invalidée dans le foie, la réponse au glucose de FGF21 est diminuée de manière significative car ChREBP ne peut se lier à son élément de réponse de type ChoRE, présent sur le promoteur de fgf21. La réponse synergique de ChREBP et de PPARα sur FGF21 a été également mise en évidence dans des cultures primaires d’hépatocytes humains. Chez les souris invalidées pour PPARα dans le foie, l’absence de FGF21 circulant entraine une augmentation de la préférence au sucrose. Notre étude révèle l’existence d’un dialogue fonctionnel unique entre ChREBP et PPARα pour la régulation de FGF21 en réponse au glucose / The hepatokine FGF21 (Fibroblast Growth factor 21) plays an important role in the control of energy homeostasis. Studies in humans and animals have established FGF21 as an important therapeutic target for its beneficial effects on hyperglycemia, dyslipidemia and obesity. Induced in response to fasting by the PPARα nuclear receptor (Proliferator Activated Receptor α), recent studies suggest the involvement of ChREBP (Carbohydrate Responsive Element Binding) in the nutritional response of FGF21. In this context, the thesis objectives were: 1) to obtain a better understanding of the regulation of FGF21 in the liver by fasting and glucose via the molecular actors ChREBP and PPARα; 2) to determine the physiological relevance of the ChREBP-PPARα-FGF21 axis in response to glucose. Our results demonstrate that hepatic expression of ChREBP is necessary for the induction of FGF21 in response to glucose in vitro and in vivo. Unexpectedly, when PPARα expression is specifically invalidated in the liver, the glucose response of FGF21 is significantly decreased as ChREBP cannot bind to its ChoRE response element present on the fgf21 promoter. The synergistic response of ChREBP and PPARα to FGF21 was also demonstrated in primary cultures of human hepatocytes. In mice deficient for PPARα in the liver, the absence of circulating FGF21 leads to an increase in their preference to sucrose. Our study reveals the existence of a unique functional dialogue between ChREBP and PPARα for the regulation of FGF21 in response to glucose