Membrane transport proteins are known to influence the absorption, distribution, metabolism, excretion and toxicity (ADMET) of drugs. At the onset of this thesis work, only a few structure-activity models, in general describing P-glycoprotein (Pgp/ABCB1) interactions, were developed using small datasets with little structural diversity. In this thesis, drug-transport protein interactions were explored using large, diverse datasets representing the chemical space of orally administered registered drugs. Focus was set on the ATP-binding cassette (ABC) transport proteins expressed in the canalicular membrane of human hepatocytes. The inhibition of the ABC transport proteins multidrug-resistance associated protein 2 (MRP2/ABCC2) and bile salt export pump (BSEP/ABCB11) was experimentally investigated using membrane vesicles from cells overexpressing the investigated proteins and sandwich cultured human hepatocytes (SCHH). Several previously unknown inhibitors were identified for both of the proteins and predictive in silico models were developed. Furthermore, a clear association between BSEP inhibition and clinically reported drug induced liver injuries (DILI) was identified. For the first time, an in silico model that described combined inhibition of Pgp, MRP2 and breast cancer resistance protein (BCRP/ABCG2) was developed using a large, structurally diverse dataset. Lipophilic weak bases were more often found to be general ABC inhibitors in comparison to other drugs. In early drug discovery, in silico models can be used as predictive filters in the drug candidate selection process and membrane vesicles as a first experimental screening tool to investigate protein interactions. In summary, the present work has led to an increased understanding of molecular properties important in ABC inhibition as well as the potential influence of ABC proteins in adverse drug reactions. A number of previously unknown ABC inhibitors were identified and predictive computational models were developed.

Cancer is a leading cause of mortality worldwide, responsible for nearly one in six deaths. Thus, there is a need for a greater understanding of cancer for the development of novel therapeutics. This master thesis project aims to compare the proteome signatures between dying and surviving cancer cells treated with diverse anticancer drugs. The first aim is to investigate if drug targets behave similarly and have the same sign (up- or down-regulation) in dying versus surviving cells. The second aim is to validate that combining the dying cancer cell’s proteome with the surviving cell’s can help improve drug target rankings for anticancer treatments. The third aim is to identify proteins and pathways involved in life and death decisions by comparing dying and surviving states in response to the anticancer drugs in different cell lines. First, we demonstrate that drug target behaviour in dying versus surviving cells is almost identical for nine diverse anticancer compounds with a correlation of 0.93. To identify drug targets, orthogonal partial least squares-discriminant analysis (OPLS-DA) modelling was performed to contrast the proteome signature of one anticancer drug against all other drugs and rank the proteins based on the magnitude of the model’s predictive component. There were occasions when the dying cells gave better rankings than the surviving ones. In some cases, the best target rankings were obtained when combining the data from both surviving and dying cells. To identify proteins and pathways involved in life and death decisions, OPLS-DA modelling contrasting the two states was performed, and heatmaps and scatterplots of dying and surviving log2 fold changes were made. As a result, several pathways involved in cell survival and cell death were identified. In addition, at least six proteins consistently differentially regulated between the surviving and dying cells were identified. Such proteins can be considered as putative survival (resistance) or sensitivity biomarkers and serve as potential drug targets for the development of novel anticancer agents.

The electrical properties of proteins in solution are important for their structure and function. Computational biophysics studies of proteins need accurate parameters to ensure that numerical simulations match physical reality. Past work in this eld has compared the electrical properties of proteins obtained from dielectric spectroscopy to numerical simulations of proteins in water with adjustment of pKa values to try to capture the inevitable changes in electrical conformation that will occur in a complex structure such as a folded protein. However, fundamental veri cation of the charge parameters of the amino acid building blocks in common molecular dynamics software packages with electrical experiments needs to be performed to have increased con dence in the results from numerical simulations. The aim of this thesis is to start from a fundamental building block, the single amino acid alanine, and to compare numerical simulations of this amino acid in water using parameters from commonly used charge structures in CHARMM, GROMOS, and OPLS, with electrical parameters obtained from dielectric spectroscopy experiments in the GHz range. To this end, multiple molecular dynamics simulations were performed to accurately determine how these different charge structures yield different dielectric increments. Additionally, a commercial RF dielectric measurement probe was modi ed to perform measurements on solutions containing alanine at different concentrations. Using regression, the dielectric increment of alanine is readily determined and compared with the numerical simulations. The results indicate that the CHARMM and OPLS parameters seem to adequately capture the charge con guration of alanine in solution, while the GROMOS parameters produce a dielectric increment but do not seem to adequately capture the charge con guration of alanine in solution. These studies lay the foundation for future studies of additional amino acids in solution as well as a stepping stone for larger simulations of the electrical properties of fully solvated proteins in solution.

Les plantes médicinales constituent une source inexhaustible de composés (des produits naturels - PN) utilisé en médecine pour la prévention et le traitement de diverses maladies. L'introduction de nouvelles technologies et méthodes dans le domaine de la chimie des produits naturels a permis le développement de méthodes ‘high throughput’ pour la détermination de la composition chimique des extraits de plantes, l'évaluation de leurs propriétés et l'exploration de leur potentiel en tant que candidats médicaments. Dernièrement, la métabolomique, une approche intégrée incorporant les avantages des technologies d'analyse moderne et la puissance de la bioinformatique s’est révélé un outil efficace dans la biologie des systèmes. En particulier, l'application de la métabolomique pour la découverte de nouveaux composés bioactifs constitue un domaine émergent dans la chimie des produits naturels. Dans ce contexte, le genre Acronychia de la famille des Rutaceae a été choisi sur la base de son usage en médecine traditionnelle pour ses propriétés antimicrobienne, antipyrétique, antispasmodique et anti-inflammatoire. Nombre de méthodes chromatographiques modernes, spectrométriques et spectroscopiques sont utilisées pour l'exploration de leur contenu en métabolites suivant trois axes principaux constituant les trois chapitres de cette thèse. En bref, le premier chapitre décrit l’étude phytochimique d’Acronychia pedunculata, l’identification des métabolites secondaires contenus dans cette espèce et l'évaluation de leurs propriétés biologiques. Le deuxième chapitre vise au développement de méthodes analytiques pour l'identification des dimères d’acétophénones (marqueurs chimiotaxonomiques du genre) et aux stratégies utilisées pour la déréplication de ces différents extraits et la caractérisation chimique des composés par UHPLC-HRMSn. Le troisième chapitre se concentre sur l'application de méthodologies métabolomique (RMN et LC-MS) pour l'analyse comparative (entre les différentes espèces, origines, organes), pour des études chimiotaxonomiques (entre les espèces) et pour la corrélation des composés contenus avec une activité pharmacologique. / Medicinal plants constitute an unfailing source of compounds (natural products – NPs) utilised in medicine for the prevention and treatment of various deceases. The introduction of new technologies and methods in the field of natural products chemistry enabled the development of high throughput methodologies for the chemical composition determination of plant extracts, evaluation of their properties and the exploration of their potentials as drug candidates. Lately, metabolomics, an integrated approach incorporating the advantages of modern analytical technologies and the power of bioinformatics has been proven an efficient tool in systems biology. In particular, the application of metabolomics for the discovery of new bioactive compounds constitutes an emerging field in natural products chemistry. In this context, Acronychia genus of Rutaceae family was selected based on its well-known traditional use as antimicrobial, antipyretic, antispasmodic and anti-inflammatory therapeutic agent. Modern chromatographic, spectrometric and spectroscopic methods were utilised for the exploration of their metabolite content following three basic axes constituting the three chapters of this thesis. Briefly, the first chapter describes the phytochemical investigation of Acronychia pedunculata, the identification of secondary metabolites contained in this species and evaluation of their biological properties. The second chapter refers to the development of analytical methods for the identification of acetophenones (chemotaxonomic markers of the genus) and to the dereplication strategies for the chemical characterisation of extracts by UHPLC-HRMSn. The third chapter focuses on the application of metabolomic methodologies (LC-MS & NMR) for comparative analysis (between different species, origins, organs), chemotaxonomic studies (between species) and compound-activity correlations.