ABCD2 (D2) has been proposed as a peroxisomal long-chain acyl-CoA transporter that is essential for very long chain fatty acid metabolism. In the livers of mice, D2 is highly induced by fenofibrate, a PPARα ligand that has been widely used as a lipid lowering agent in the treatment of hypertriglyceridemia. To determine if D2 is a modifier of fibrate responses, wild-type and D2 deficient mice were treated with fenofibrate for 14 days. The absence of D2 altered expression of gene clusters associated with lipid metabolism, including PPARα signaling. Using 3T3-L1 adipocytes, which express high levels of D2, we confirmed that knock-down of D2 modified genomic responses to fibrate treatment. We next evaluated the impact of D2 on effects of fibrates in a mouse model of dietinduced obesity. Fenofibrate treatment opposed the development of obesity, hypertriglyceridemia, and insulin resistance. However, these effects were unaffected by D2 genotype. We concluded that D2 can modulate genomic responses to fibrates, but that these effects are not sufficiently robust to alter the effects of fibrates on diet-induced obesity phenotypes. Although proposed as a peroxisomal transporter, the intracellular localization of D2, especially in adipose tissue, has not been validated with direct experimental evidence. Sequential centrifugation of mouse adipose homogenates generated a fraction enriched with D2, but lacked well-known peroxisome markers including catalase, PEX19, and ABCD3 (D3). Electron microscopic imaging of this fraction confirmed the presence of D2 protein on an organelle with evidence of a dense matrix and a diameter of ~200 nm, the typical structure and size of a microperoxisome. D2 and PEX19 antibodies recognized distinct structures in mouse adipose. Immunoisolation of the D2-containing compartment from adipose tissue confirmed the scarcity of PEX19. Proteomic profiling of the D2 compartment revealed the presence of proteins associated peroxisome, endoplasmic reticulum (ER), and mitochondria. We conclude that D2 is localized to a distinct subclass of peroxisomes that lack many peroxisome proteins and may physically associate with mitochondria and the ER.

Les objectifs de cette thèse étaient d'apporter une percée conceptuelle pour une compréhension en profondeur des mécanismes moléculaires des protéines intrinsèquement désordonnées (IDPs) et de leurs rôles dans la physiologie cellulaire de Chlamydomonas reinhardtii. La combinaison d’approches expérimentale et bioinformatique m’a permis d’identifier 682 protéines thermorésistantes chez C. reinhardtii. Parmi celles-ci, 299 protéines sont systématiquement prédites comme potentielles IDP par quatre algorithmes de prédiction de désordre. Nos résultats indiquent que le pourcentage désordonné moyen de ces protéines prédites comme étant des IDPs est d'environ 20%, et la plupart d'entre elles (~70%) sont adressées à d'autres compartiments que la mitochondrie et le chloroplaste. Leur composition en acides aminés est biaisée par rapport à d'autres IDPs de la base de données de protéines désordonnées (DisProt). Ces IDPs potentielles jouent des fonctions moléculaires diverses, et 54% d'entre elles sont des cibles de phosphorylation.Notre travail a également augmenté l’état des connaissances sur l'adénylate kinase 3 (ADK3), une enzyme contenant une région intrinsèquement désordonnée (IDR). Cette enzyme a été isolée par notre approche globale pour caractériser les IDPs de l’algue verte. L’extension C-terminale désordonnée (CTE) de cette enzyme lui confère de nouvelles fonctions comme par exemple, la formation d’un complexe bi-enzymatique avec la glycéraldéhyde-3-phosphate déshydrogénase (GAPDH), la régulation (négative) de l'activité GAPDH avec le NADPH comme cofacteur, et le rôle de chaperon pour la GAPDH en la protégeant de la dénaturation par traitement thermique et de l’agrégation. / The objectives of this work were to bring a conceptual breakthrough for an in-depth understanding of the molecular mechanisms of intrinsically disordered proteins (IDPs) and their roles in the cellular physiology of Chlamydomonas reinhardtii. Using experimental approaches, 682 heat-resistant proteins were identified as putative IDPs. Among them, 299 proteins were consistently predicted as IDPs by all four disordered predictors. The mean percentage of disordered residues content of these IDPs is about 20%, and most of them (~70%) are addressed to other compartments than mitochondrion and chloroplast. These newly identified IDPs from C. reinhardtii have a biased amino acid composition as regard to other IDPs from the Database of protein disorder (DisProt). Furthermore, they play diverse molecular functions, and 54% of them are the targets for phosphorylation. Our work also revealed more knowledge of the IDR-containing protein adenylate kinase 3 (ADK3) that was extracted by heat-treatment. Its disordered C-terminal extension (CTE) brought new functions to this protein. For instance, via its CTE, ADK3 can form a bi-enzyme complex with glyceraldehyde-3-phosphate dehydrogenase (GAPDH), down-regulates the NADPH-dependent GAPDH activity, and behaves as a chaperone for GAPDH against its aggregation and inactivation under heat-treatment.