Membrane proteins are found in all kingdoms of life and are essential for cellular interactions with the environment. Although a large research effort have been put into this group many membrane proteins remains uncharacterized, both in terms of function and evolutionary history. We have estimated the component of α-helical membrane proteins within the human proteome; the membrane proteome. We found that the human membrane proteome make up 27% of all protein, which we could classify the majority of into 234 families and further into three major functional groups: receptors, transporters or enzymes. We extended this analysis by determining the membrane proteome of 24 organisms that covers all major groups of eukaryotes. This comprehensive membrane protein catalog of over 100,000 proteins was utilized to determine the evolutionary history of all membrane protein families throughout eukaryotes.  We also investigated the evolutionary history across eukaryotes of the antiviral Interferon induced transmembrane proteins (IFITM) and the G protein-coupled receptor (GPCR) superfamily in detail.  We identified ten novel human homologs to the IFITM proteins, which together with the known IFITMs forms a family that we call the Dispanins. Using phylogenetic analysis we show that the Dispanins first emerged in eukaryotes in a common ancestor of choanoflagellates and animals, and that the family later expanded in vertebrates into four subfamilies. The GPCR superfamily was mined across eukaryotic species and we present evidence for a common origin for four of the five main human GPCR families; Rhodopsin, Frizzled, Adhesion and Secretin in the cAMP receptor family that was found in non-metazoans and invertebrates, but has been lost in vertebrates. Here we present the first accurate estimation of the human proteome together with comprehensive functional and evolutionary classification and extend it to organisms that represents all major eukaryotic groups. Moreover, we identify a novel protein family, the Dispanins, which has an evolutionary history that has been formed by horizontal gene transfer from bacteria followed by expansions in the animal lineage. We also study the evolution of the GPCR superfamily throughout eukaryotic evolution and provide a comprehensive model of the evolution and relationship of these receptors.

Les glioblastomes sont des tumeurs au pronostic défavorable. L’échec des thérapies est lié à la présence de cellules souches cancéreuses (CSCs), résistantes aux traitements ; la caractérisation de ces cellules et l’identification de biomarqueurs sont donc primordiales. Le calcium contrôle de nombreux processus cellulaires ; parmi les éléments majeurs de la signalisation calcique, la Calmoduline (CaM) est impliquée dans différentes pathologies, dont des cancers, et est un puissant régulateur de l’état physiologique d’une cellule. CaM interagit avec de nombreuses protéines impliquées dans la régulation de l’homéostasie calcique de la cellule. Nous avons cherché à identifier et caractériser des antagonistes de CaM, inhibant différentiellement ces interactions. L’utilisation de ces antagonistes en tant que perturbateurs de l’homéostasie calcique a permis de mettre en évidence un marqueur caractérisé des CSCs de glioblastomes. D’autre part, l’étude comparée du protéome membranaire de CSCs issues de glioblastomes a permis de mettre en évidence la surexpression de clusters de différenciation et protéines impliquées dans la signalisation calcique. Ces protéines sont de potentiels marqueurs moléculaires des CSCs de glioblastome. / Glioblastomas are malignant tumor of poor prognosis. Therapeutic failure might be supported by cancer stem cells (CSCs); characterization of these cells and biomarkers identification are of most importance. Calcium controls numerous cellular process; beyond major elements of calcium signaling, Calmodulin (CaM) is involved in different pathologies and tumors, and is a powerful regulator of cell physiological state. CaM interacts with a plethora of proteins involved in cell calcium homeostasis regulation. We aimed at identifying and characterizing CaM antagonists, capable of differentially inhibiting these interactions. The use of these antagonists in calcium homeostasis disturbance led to the identification of a characterized marker of glioblastomas CSCs. In another approach, the comparative study of glioblastomas CSCs membrane proteome uncovered the overexpression of differentiation clusters and proteins involved in calcium signaling. These proteins are potential molecular biomarkers for glioblastomes CSCs.