There has been much interest in the use of therapeutics based on ribonucleic acid interference(RNAi) to inhibit synthesis of mutant proteins ever since Elbashir et al. (Elbashir, S. M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K. and Tuschl, T., 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature. 411, 494-498.) found that synthetic double stranded small interfering ribonucleic acids (siRNAs) can initiate this evolutionarily conserved process in mammalian cells. Since RNAi is able to target single genes and therefore mitigate the underlying molecular pathology of diseases, RNAi-based therapeutics will most likely benefit monogenic neurodegenerative diseases such as Huntington’s disease. It is however particularly difficult to deliver exogenous materials such as siRNAs into neurons in vivo as the blood-brain barrier (BBB) isolates the brain from the vascular system and prevents permeation of most materials. Neurons also do not take up exogenous materials readily. Therefore, effective delivery of siRNAs into the brain remains one of the biggest challenges impeding their use as a potential neurotherapeutic. Layered double hydroxide (LDH) nanoparticles are a class of anionic clay materials that have demonstrated great potential as a DNA (deoxyribonucleic acid) delivery system for a variety of mammalian cell lines due to their unique physiochemical properties. This thesis examined the feasibility of LDH as a siRNA delivery system for cultured neurons and demonstrated that the delivered siRNAs are able to effectively down-regulate synthesis of a target protein with minimal toxicity. Experiments were conducted using double stranded DNAs (dsDNAs) initially, and siRNAs were then used to verify these results. It was shown that nucleic acids(dsDNAs and siRNAs) could successfully intercalate into pristine LDHs to form nucleic acid-LDH complexes that had properties suitable for use as a delivery system in mammalian cells. These studies established that LDHs and nucleic acid-LDH complexes were biocompatible with neurons isolated from embryonic day 17.5 mouse cerebral cortex, suggesting that LDH can be used for nucleic acid delivery into cultured neurons. LDHs were also shown to successfully deliver nucleic acids into a non-neural mammalian cell line (NIH 3T3 cells). Finally, this thesis demonstrated for the first time that LDHs were able to deliver siRNAs into neurons, providing encouraging preliminary evidence that sequence specific gene silencing of the Mus Musculus Deleted in Colorectal Cancer (DCC) gene had occurred. However, down-regulation of the DCC protein did not occur consistently, suggesting that further optimisation is needed to improve the efficacy of siRNA-LDH complexes to inhibit expression of target protein in neurons. In future, LDHs should be further developed as an efficient siRNA delivery system for therapeutic gene silencing in the central nervous system using a neurodegenerative disease model such as the Huntington’s disease mouse model, which closely phenocopies the human disease. This model will allow the in vivo efficacy of these nanoparticles to be tested and subsequently improved in order to deliver siRNAs locally and systematically into the brain.

Metalofármacos baseados em rutênio têm se mostrado promissores com relação à atividade anticancerígena frente a diversos tipos de tumores. Nosso grupo de pesquisa dedica-se ao estudo de compostos contendo o centro dimetálico de valência mista Ru2(II,III) coordenado a ligantes derivados de Faines (Fármacos anti-inflamatórios não esteroides), tendo demostrando o potencial desses complexos frente a glioma. O entendimento do modo de ação destes complexos requer o estudo de suas interações com biomoléculas presentes no meio biológico. Neste cenário, o presente trabalho teve por objetivo investigar a interação de três complexos de dirutênio-Faines, ou RuFaines, [Ru2(ibp)4Cl], [Ru2(ceto)4Cl] e [Ru2(npx)4(H2O)2]PF6 (ibp = ibuprofenato, ceto = cetoprofenato e npx = naproxenato), e também do precursor [Ru2(O2CCH3)4Cl], RuAc, com as principais proteínas presentes no soro humano, transferrina e albumina. Os complexos foram sintetizados e caracterizados conforme metodologias desenvolvidas no grupo. A interação destes complexos com a transferrina, em suas formas apo e holo, e com a albumina foi avaliada por técnicas como espectroscopia eletrônica, dicroísmo circular, fluorescência, e realizaram-se estudos de ultrafiltração com análise do aduto formado por ICP-OES e espectrometria de massas. Além disso, fez-se um estudo de captação celular dos complexos RuFaines por células de glioma humano da linhagem U-87. Os resultados demonstraram que os complexos de dirutênio-Faines interagem com ambas as proteínas séricas (transferrina (apo e holo) e albumina), de modo semelhante, mas que é distinto daquele observado para o complexo RuAc. A presença de íons Fe(III) nos sítios específicos da transferrina não afetou a interação dos complexos RuFaines, enquanto que um comportamento diferente foi observado para o RuAc. Verificou-se que todas as proteínas avaliadas (albumina, apo-transferrina e holo-transferrina) apresentam capacidades similares de retenção dos complexos (~ 70% da quantidade de Ru adicionada inicialmente), independentemente da natureza do ligante carboxilato coordenado. Estudos de captação celular mostraram que a interação dos complexos RuFaines com a transferrina não contribuiu para modificar a capacidade de entrada desses complexos na célula, em comparação com os metalofármacos livres. Em alguns casos, inclusive, a formação de aduto com a apo-transferrina teve um efeito contrário, diminuindo a captação de rutênio. Dessa forma, concluiu-se que o ciclo da transferrina provavelmente não é a principal rota de entrada nas células para os complexos estudados. / Ruthenium metallodrugs have shown promising antitumor activity against to several tumor types. Our research group is dedicated to study compounds containing the mixed-valence Ru2(II,III) dimetallic center coordinated to NSAIDs (nonsteroidal anti-inflammatory drugs) derived ligands, and have demonstrated the potential of these complexes against glioma. The understanding of the mode of action of these complexes requires the study of their interactions with biomolecules present in biological environment. In this scenario, the present work aimed to investigate the interaction of three complexes of diruthenium-NSAIDs, or RuNSAIDs, [Ru2(ibp)4Cl], [Ru2(ceto)4Cl] and [Ru2(npx)4(H2O)2]PF6 (ibp = ibuprofenate, ceto = ketoprofenate, npx = naproxenate), and also of the precursor [Ru2(O2CCH3)4Cl], RuAc, with the major proteins present in the human serum, transferrin and albumin. The complexes were synthesized and characterized according to methods developed in our group. The interaction of these complexes with transferrin, in the two forms apo and holo, and with albumin was evaluated by techniques as electronic spectroscopy, circular dichroism, fluorescence, and ultrafiltration studies accompanied by analysis of adducts by ICP-OES and mass spectrometry. Moreover, cellular uptake studies of the RuNSAIDs complexes by U-87 human glioma cells line were performed. The results demonstrated that the diruthenium-NSAIDs complexes interact with both proteins (transferrin (apo and holo) and albumin), in a similar way, but that is distinct of that observed for the RuAc complex. The presence of Fe(III) ions in transferrin specific binding sites did not affect the interaction of the RuNSAID complexes with the protein, while a different behavior was shown by RuAc. All the proteins studied here (albumin, apo-transferrin and holo-transferrin) showed similar capabilities for retention of the complexes (~ 70 % of the initial amount of Ru added), independently of the nature of the coordinated carboxylate ligand. Cellular uptake studies showed that the interaction of the RuNSAIDs complexes with transferrin did not contribute to modify the internalization capacity of these complexes, in comparison with the free metallodrugs. In some cases, the adduct formation with apo-transferrin showed an opposite effect, leading to the decrease of ruthenium uptake. The findings led to the conclusion that transferrin cycle probably is not the main entry way to the cells for the studied complexes.

Metalofármacos baseados em rutênio têm se mostrado promissores com relação à atividade anticancerígena frente a diversos tipos de tumores. Nosso grupo de pesquisa dedica-se ao estudo de compostos contendo o centro dimetálico de valência mista Ru2(II,III) coordenado a ligantes derivados de Faines (Fármacos anti-inflamatórios não esteroides), tendo demostrando o potencial desses complexos frente a glioma. O entendimento do modo de ação destes complexos requer o estudo de suas interações com biomoléculas presentes no meio biológico. Neste cenário, o presente trabalho teve por objetivo investigar a interação de três complexos de dirutênio-Faines, ou RuFaines, [Ru2(ibp)4Cl], [Ru2(ceto)4Cl] e [Ru2(npx)4(H2O)2]PF6 (ibp = ibuprofenato, ceto = cetoprofenato e npx = naproxenato), e também do precursor [Ru2(O2CCH3)4Cl], RuAc, com as principais proteínas presentes no soro humano, transferrina e albumina. Os complexos foram sintetizados e caracterizados conforme metodologias desenvolvidas no grupo. A interação destes complexos com a transferrina, em suas formas apo e holo, e com a albumina foi avaliada por técnicas como espectroscopia eletrônica, dicroísmo circular, fluorescência, e realizaram-se estudos de ultrafiltração com análise do aduto formado por ICP-OES e espectrometria de massas. Além disso, fez-se um estudo de captação celular dos complexos RuFaines por células de glioma humano da linhagem U-87. Os resultados demonstraram que os complexos de dirutênio-Faines interagem com ambas as proteínas séricas (transferrina (apo e holo) e albumina), de modo semelhante, mas que é distinto daquele observado para o complexo RuAc. A presença de íons Fe(III) nos sítios específicos da transferrina não afetou a interação dos complexos RuFaines, enquanto que um comportamento diferente foi observado para o RuAc. Verificou-se que todas as proteínas avaliadas (albumina, apo-transferrina e holo-transferrina) apresentam capacidades similares de retenção dos complexos (~ 70% da quantidade de Ru adicionada inicialmente), independentemente da natureza do ligante carboxilato coordenado. Estudos de captação celular mostraram que a interação dos complexos RuFaines com a transferrina não contribuiu para modificar a capacidade de entrada desses complexos na célula, em comparação com os metalofármacos livres. Em alguns casos, inclusive, a formação de aduto com a apo-transferrina teve um efeito contrário, diminuindo a captação de rutênio. Dessa forma, concluiu-se que o ciclo da transferrina provavelmente não é a principal rota de entrada nas células para os complexos estudados. / Ruthenium metallodrugs have shown promising antitumor activity against to several tumor types. Our research group is dedicated to study compounds containing the mixed-valence Ru2(II,III) dimetallic center coordinated to NSAIDs (nonsteroidal anti-inflammatory drugs) derived ligands, and have demonstrated the potential of these complexes against glioma. The understanding of the mode of action of these complexes requires the study of their interactions with biomolecules present in biological environment. In this scenario, the present work aimed to investigate the interaction of three complexes of diruthenium-NSAIDs, or RuNSAIDs, [Ru2(ibp)4Cl], [Ru2(ceto)4Cl] and [Ru2(npx)4(H2O)2]PF6 (ibp = ibuprofenate, ceto = ketoprofenate, npx = naproxenate), and also of the precursor [Ru2(O2CCH3)4Cl], RuAc, with the major proteins present in the human serum, transferrin and albumin. The complexes were synthesized and characterized according to methods developed in our group. The interaction of these complexes with transferrin, in the two forms apo and holo, and with albumin was evaluated by techniques as electronic spectroscopy, circular dichroism, fluorescence, and ultrafiltration studies accompanied by analysis of adducts by ICP-OES and mass spectrometry. Moreover, cellular uptake studies of the RuNSAIDs complexes by U-87 human glioma cells line were performed. The results demonstrated that the diruthenium-NSAIDs complexes interact with both proteins (transferrin (apo and holo) and albumin), in a similar way, but that is distinct of that observed for the RuAc complex. The presence of Fe(III) ions in transferrin specific binding sites did not affect the interaction of the RuNSAID complexes with the protein, while a different behavior was shown by RuAc. All the proteins studied here (albumin, apo-transferrin and holo-transferrin) showed similar capabilities for retention of the complexes (~ 70 % of the initial amount of Ru added), independently of the nature of the coordinated carboxylate ligand. Cellular uptake studies showed that the interaction of the RuNSAIDs complexes with transferrin did not contribute to modify the internalization capacity of these complexes, in comparison with the free metallodrugs. In some cases, the adduct formation with apo-transferrin showed an opposite effect, leading to the decrease of ruthenium uptake. The findings led to the conclusion that transferrin cycle probably is not the main entry way to the cells for the studied complexes.

Gemini surfactants are versatile gene delivery agents because of their ability to bind and compact DNA and their low cellular toxicity. The aim of my dissertation work was to develop non-invasive mucosal formulations of novel amino acid-substituted gemini surfactants with the general chemical formula C12H25(CH3)2N+-(CH2)3-N(AA)-(CH2)3-N+(CH3)2-C12H25 (AA= glycine, lysine, glycyl-lysine, lysyl-lysine). These compounds were formulated with a model plasmid DNA, encoding for interferon-γ and green fluorescent protein, in the presence of helper lipid, 1,2 dioleyl-sn-glycero-phosphatidyl-ethanolamine. Formulations were assessed in Sf 1 Ep epithelial cells. Among the novel compounds, plasmid/gemini/lipid (P/G/L) nanoparticles formulated using glycine- and glycyl-lysine substituted gemini surfactants achieved significantly higher gene expression than the parent unsubstituted compound. The key physicochemical properties, e.g. size, surface charge, DNA binding, and toxicity of P/G/L complexes were correlated with transfection efficiency. The presence of amino-acid substitution did not interfere with DNA compaction and contributed to an overall low toxicity of all P/G/L complexes, comparable to the parent gemini surfactant. A cellular uptake mechanistic study revealed that both clathrin- and caveolae-mediated uptake were major uptake routes for P/G/L nanoparticles. However, amino acid substitution in the gemini surfactant imparted high buffering capacity, pH-dependent increase in particle size, and balanced DNA binding properties. These properties may enhance endosomal escape of P/12-7NGK-12/L resulting in higher gene expression. Finally, the P/G/L complexes were incorporated into an in-situ gelling dispersion containing a thermosensitive polymer, poloxamer 407, and a permeation enhancer, diethylene glycol monoethyl ether (DEGEE). A 16% w/v poloxamer concentration produced a dispersion that gelled at body temperature and exhibited sufficient yield value to prevent formulation leakage from the vaginal cavity. The formulations were prepared with a model plasmid, encoding for red fluorescent protein, and administered topically to rabbit vagina. In agreement with our in vitro results, confocal microscopy revealed that glycyl-lysine substituted gemini surfactant exhibited higher gene expression compared to the parent unsubstituted gemini surfactant. This provided proof-of-concept for use of amino acid-substituted gemini surfactant in non-invasive mucosal (vaginal) gene delivery systems with potential therapeutic applications. These formulations will be developed with therapeutically relevant genes to assess their potential as genetic vaccines. In addition, new gemini surfactants will be developed by grafting other amino acids via glycine linkage to retain conformation flexibility and enhance endosomal escape of DNA complexes for higher transfection efficiency.

En ce début de XXIème siècle l’essor des nanotechnologies est indéniable et ne semble pas près de ralentir. En effet, les secteurs industriels de pointe misent sur les innovations remarquables permises par les nouvelles propriétés de la matière à l’échelle nanométrique. Cependant, les efforts scientifiques et budgétaires, mis en place pour comprendre les potentiels risques sanitaires des nano-objets pour l’homme et l’environnement, s’avèrent peu fructueux tant la diversité des nano-objets est étendue. De plus, il n’existe pas de consensus règlementaire pour la nanotoxicologie, ce qui entraine un manque d’homogénéité, et parfois de cohérence, entre les données issues de la recherche. Dans ce contexte, la présente étude porte sur l’amélioration de la compréhension de l’activité biologique des nanoparticules. Des nanoparticules fluorescentes de silice, synthétisées à façon, ont permis d’étudier séparément l’impact de deux de leurs caractéristiques physico-chimiques (taille et fonctionnalisation de surface). L’altération membranaire, la réaction pro-inflammatoire et la génération de stress oxydant induites par la mise en contact des nanoparticules avec une lignée de macrophages murins ont été étudiées. En complément, des méthodes permettant de distinguer les nanoparticules internalisées de celles adsorbées à la membrane cellulaire ainsi que d’observer en cinétique la phagocytose, ont été développées. La nature et l’intensité des effets biologiques observés ont montré que les nanoparticules n’étaient pas inertes et que leur impact dépendait bien de leur taille et de leur fonctionnalisation de surface ainsi que de la dose étudiée. / In the 21st century nanotechnologies are growing fast and are continuously evolving. Indeed, high-tech industries rely on remarkable innovations enabled by the new properties of matter at the nanoscale. In addition, the nanomedicine field seems hopeful to cure human health problems such as heart or liver diseases, brain damage or cancers. However, despite expansive scientific efforts the potential risks of nano-objects on human health and on the environment are still poorly understood due to the wide variety of nano-objects. In addition, no consensus about nanotoxicology exists yet, sometimes leading to inhomogeneous and inconsistent findings. In this context, the present study aimed at a better understanding of the nanoparticles biological effects. Fluorescent silica nanoparticles with well-controlled size and surface functionalization were synthesized to study separately the impact of these two physico-chemical characteristics on cell response. Murine macrophages (from the RAW 264.7 cell line commonly used as a reference for nanotoxicology studies) were exposed to the different nanoparticles, and membrane alteration, induction of pro-inflammatory effect and generation of oxidative stress were investigated. In addition, in order to distinguish uptaken nanoparticles from those adsorbed at the cell membrane and to observe phagocytosis over time, methods were developed. Results demonstrated that the studied nanoparticles were not inert. Moreover the biological effects were found to depend on nanoparticles size, surface functionalization and dose.

L'alpha-1 antitrypsine (A1AT) est un inhibiteur de sérine protéase sécrété principalement par le foie et libéré dans la circulation où sa concentration physiologique est de 1,5 à 3,5 g/L. La principale fonction de l'A1AT est d'inhiber l'activité de l'élastase des neutrophiles (NE) afin de maintenir l'équilibre protéase/anti-protéase dans les poumons. Son déficit (A1ATD) touche plus de 3,4 millions d'individus dans le monde chez qui l'élastase des neutrophiles décompose l'élastine, provoquant ainsi une diminution de l'élasticité du poumon ainsi qu'une dégradation de son tissu conjonctif. En conséquence, l'A1ATD entraîne des troubles respiratoires tels que l'emphysème ou la maladie pulmonaire obstructive chronique et ceux qui en sont atteints nécessitent des injections fréquentes d'A1AT purifiée à partir du sang d'un donneur. Cependant, l'A1AT plasmatique est hétérogène dans son état de glycosylation et sa qualité varie d'un lot à l'autre. De plus, il y a un risque, même très faible, de transmission d'agents pathogènes avec l'administration d'A1AT purifiée par plasma. Par conséquent, il existe un besoin pour une version recombinante. La glycoprotéine mature possède trois sites de N-glycosylation comprenant principalement des structures de type complexe bi-antennaires afucosylées et α-2,6-di-sialylées, A2G2S2 (6,6). Bien que la glycosylation ne soit pas essentielle à l'activité inhibitrice de l'A1AT, il a été démontré qu'elle a un impact significatif sur sa demi-vie in vivo. Notamment, l'acide sialique, un monosaccharide terminal chargé négativement présent sur les N-glycanes, aide à prolonger la demi-vie de l'A1AT dans le sérum en empêchant l'interaction entre l'avant-dernier galactose (Gal) du N-glycane et les récepteurs hépatiques des asialoglycoprotéines (ASGPRs), composés de deux sous-unités appelées lectines hépatiques (HL) 1 et 2, qui se lient aux glycoprotéines asialylées contenant un Gal terminal et conduisent à leur dégradation. Par conséquent, il est important de produire A1AT dans un système d'expression qui peut effectuer les modifications post-traductionnelles (PTM) appropriées à des fins thérapeutiques. Jusqu'à présent, la production d'A1AT recombinante (rA1AT) a été tentée dans différents systèmes d'expression cellulaire avec un succès limité. Malgré la disponibilité de diverses lignées cellulaires, les cellules ovariennes de hamster chinois (CHO) ont été largement utilisées pour la production de glycoprotéines thérapeutiques car ces cellules sont compatibles avec des stratégies de glyco-ingénierie pour produire des glycoprotéines recombinantes composées de glycanes de type humain. Cependant, ces cellules synthétisent des N-glycanes de type complexe comprenant de la fucosylation centrale et de l'acide sialique lié en α-2,3. Par conséquent, dans ce projet, l'objectif était de développer une version recombinante d'A1AT avec un profil de glycosylation humaine exprimée en cellules CHO modifiées et qui se prête à des utilisations thérapeutiques. À cette fin, dans notre étude, nous avons d'abord empêché l'α-2,3 sialylation ainsi que la fucosylation centrale en éliminant les gènes responsables via la technologie CRISPR/Cas9, suivie de la surexpression de l'α-2,6‐sialyltransférase humaine à l'aide d'un système d'expression inductible au cumate. Nous avons ensuite montré la supériorité du promoteur inductible CR5 pour l’expression de A1AT par rapport à cinq promoteurs constitutifs forts couramment utilisés dans l'industrie. En utilisant le promoteur CR5, nous avons généré des populations de CHO stables modifiées par glyco-ingénierie produisant plus de 2,1 g/L pour la forme native et 2,8 g/L pour la version mutée d'A1AT avec des N-glycanes analogues au produit clinique dérivé du plasma, la Prolastin-C. L'effet bénéfique de la supplémentation en N‐acétylmannosamine du milieu de culture cellulaire sur la glycosylation de l'A1AT a également été démontré. Enfin, nous avons montré que l'activité anti‐élastase des rA1ATs est comparable à celle de la Prolastin-C, et que la substitution des résidus méthionines critiques par des valines rendait A1AT significativement plus résistante à l'oxydation. Nous avons ensuite étudié l'impact de la glycosylation d'A1AT sur son interaction avec les orthologues d'ASGPR. Pour cela, nous avons initialement utilisé un test d'internalisation cellulaire basé sur la lignée cellulaire hépatique humaine HepG2 connue pour exprimer les ASGPRs à sa surface et avons examiné leur interaction avec les rA1ATs possédant divers profils de glycosylation. Comme le test d'internalisation basé sur les cellules HepG2 a démontré un faible rapport signal sur bruit (SNR) ainsi qu'un niveau élevé de signal de fond d'internalisation, nous avons cherché à développer un nouveau test basé sur des cellules CHO surexprimant des orthologues ASGPR recombinants. Alors que la sous-unité HL-1 humaine seule était suffisante pour lier et internaliser l'A1AT asialylée, les sous-unités HL-1 et HL-2 étaient nécessaires pour former des récepteurs fonctionnels et ayant une forte affinité pour les ASGPR de rat et de souris. Afin d'améliorer le SNR de notre test cellulaire d'internalisation, le tri cellulaire activé par fluorescence (FACS) a été utilisé pour enrichir les populations de cellules CHO pour celles exprimant des niveaux élevés d'orthologues ASGPR. Enfin, en utilisant des structures de glycanes remodelés par voie enzymatique de Prolastin-C, nous n'avons observé aucune internalisation lorsque les glycanes sont terminés avec α-2,6-Neu5Ac ni α-2,8-Neu5Ac-α-2,6-Neu5Ac par l’ASGPR de l'humain, du rat et de la souris. D'autre part, l'absorption de Prolastin-C portant des glycanes bi-antennaires avec une branche terminée par de l'acide sialique α-2,3 et l'autre par du galactose terminal, par l'ASGPR de souris a été statistiquement plus élevée que celle de l'humain et du rat. En somme, l'A1AT recombinante résistante à l'oxydation décrite dans ce projet pourrait représenter un meilleur médicament biothérapeutique tout en offrant une alternative sûre et plus stable pour la thérapie d'augmentation. Nous avons également contribué à une meilleure compréhension de l'impact de la sialylation de l'A1AT sur son internalisation cellulaire par les orthologues ASGPR. / Alpha-1 antitrypsin (A1AT) is a serine protease inhibitor secreted primarily by the liver, and released in the circulation where its physiological concentration is 1.5-3.5 g/L. The main physiological function of A1AT is to inhibit the activity of neutrophil elastase (NE) to maintain the protease/anti-protease balance in the lung. The A1AT deficiency (A1ATD) is affecting more than 3.4 million individuals worldwide where neutrophil elastase breaks down elastin, thereby causing a decrease in the elasticity of the lung as well as a degradation of its connective tissue. As a result, A1ATD leads to respiratory disorders such as emphysema or chronic obstructive pulmonary disease. Treatment of this health condition requires frequent injections of A1AT purified from donor blood. However, plasma A1AT is heterogeneous in its glycosylation state and its quality varies from batch to batch. Moreover, there is a risk, however very low, of pathogen transmittance with plasma-purified A1AT administration. Therefore, there is a need for recombinant version. The mature glycoprotein has three N-glycosylation sites possessing mostly afucosylated, α-2,6-di-sialylated bi-antennary complex-type structures, A2G2S2 (6,6). Though glycosylation is not essential for A1AT's inhibitory activity, it has been shown to have a significant impact on its in vivo half-life. Notably, sialic acid, a terminal negatively charged monosaccharide present on N-glycans, helps to prolong the half-life of A1AT in serum by preventing the interaction between the penultimate galactose (Gal) of the N-glycan and the hepatic asialoglycoprotein receptors (ASGPRs), composed of two subunits termed hepatic lectin (HL) 1 and 2, which bind to asialylated glycoproteins containing terminal Gal and lead to their degradation. To this extend, it is important to produce A1AT in an expression system that can carry out the appropriate post-translational modifications (PTMs) for therapeutic purposes. Thus far, the production of recombinant A1AT (rA1AT) has been attempted in different cell expression systems with limited success. Despite the availability of various cell lines, Chinese hamster ovary (CHO) cells have been widely used to produce therapeutic glycoproteins as these cells can tolerate glycoengineering strategies to produce recombinant glycoproteins with human-like glycans. However, these cells synthesize complex-type N-glycans with core-fucosylation along with α-2,3-linked sialic acid. Therefore, in this research project, the aim was to develop a recombinant version of A1AT with human glycosylation pattern expressed in genetically engineered CHO cells that would be amenable to therapeutic uses. To this end, in our study, we first prevented α-2,3 sialylation as well as core-fucosylation by eliminating the corresponding genes via CRISPR/Cas9 technology, followed by overexpressed human α-2,6‐sialyltransferase using a cumate‐inducible CHO expression system. We then showed superiority of the CR5 inducible promoter compared to five strong constitutive promoters commonly used in the industry. Using the CR5 promoter, we generated glycoengineered stable CHO pools producing over 2.1 g/L of the wild-type and 2.8 g/L of the mutein forms of A1AT, with N‐glycans analogous to the plasma‐derived clinical product, Prolastin-C. The effect of N‐acetylmannosamine supplementation to the cell culture media on the A1AT glycosylation was also demonstrated. Finally, we showed that the anti‐elastase activity of rA1ATs is comparable to that of Prolastin-C, and that substitution of critical methionine residues with valines rendered A1AT significantly more resistant to oxidation. We then studied the impact of A1AT glycosylation on its interaction with ASGPR orthologs. For this, we initially used a cell-based internalization assay based on the human HepG2 hepatic cell line known to express ASGPRs at its surface and examined their interaction with rA1ATs possessing various glycosylation profiles. As HepG2 cell-based internalization assay demonstrated poor signal-to-noise ratio (SNR) as well as high level of background internalization signal, we then aimed at developing a new assay based on CHO cells overexpressing recombinant ASGPRs orthologs. While human HL-1 subunit alone was sufficient to bind and internalize asialylated A1AT, both HL-1 and HL-2 subunits were required to form functional and high affinity receptors for the rat and mouse ASGPRs. To enhance SNR of our cell-based uptake assay, fluorescence-activated cell sorting (FACS) was used to enrich the CHO pools for cells expressing high levels of ASGPR orthologs. Finally, using enzymatically remodelled glycan structures of Prolastin-C, we observed no uptake when glycans are terminated with α-2,6-Neu5Ac nor α-2,8-Neu5Ac-α-2,6-Neu5Ac by human, rat, and mouse ASGPR orthologs. On the other hand, the uptake of Prolastin-C bearing bi-antennary glycans with one branch terminated with α-2,3 sialic acid and the other with terminal galactose, by mouse ASGPR was observed to be statistically higher than that by human and rat ASGPR orthologs. Collectively, the oxidation-resistant recombinant A1AT described in this project could represent a viable biobetter drug while offering a safe and more stable alternative for augmentation therapy. We also contributed a better understanding of the impact of A1AT sialylation on its cellular uptake by ASGPR orthologs.

La thèse présentée décrit la préparation et l’étude d’auto-assemblages élaborés à partir de conjugués amphiphiles Tat-b-poly(triméthylène carbonate) (Tat-b-PTMC) doués de propriétés d’internalisation cellulaire conférées par le segment peptidique Tat. L’objectif principal de ces travaux était d’établir et de comprendre les liens entre la structure macromoléculaire, les caractéristiques colloïdales et l’activité biologique de ces systèmes. Les efforts de précision moléculaire et de caractérisation fournispour la synthèse des chimères Tat-b-PTMC a permis de corréler finement leurs structures chimiquesaux paramètres physico-chimiques des nanoparticules obtenues. Grâce à une approche expérimentale transverse combinant des études de biologie cellulaire et de biophysique, le mécanisme d’interaction in vitro de ces nanoparticules avec les cellules HeLa a pu être en partie élucidé. Enfin, un camouflage électrostatique pH-sensible a été mis au point pour tenter de moduler leur activité et d’augmenter leur sélectivité vis-à-vis de l’environnement tumoral. / This thesis work deals with preparation and study of cell-penetrating self-assemblies from amphiphilicpolymer-b-peptide Tat-b-poly(trimethylene carbonate) conjugates. Tat-b-PTMC chimeras withtunable hydrophilic fractions were synthesized, thoroughly characterized and self-assembled inaqueous buffer into size-tunable, highly monodisperse core-shell nanoparticles, presenting a full Tatcorona. Their physico-chemical profiles were assessed by complementary imaging (AFM, TEM) andscattering techniques (multiangle DLS, SANS) and correlated with their molecular architectures.Their transduction ability in vitro on HeLa cells and interaction mode with phospholipid membraneswere studied with a view to correlate their physico-chemical profiles with their biological properties.This interdisciplinary approach partially shed light on the interactions at play in the cellular uptakeprocess. With the ultimate goal of improving pharmacological characteristics, we finally endeavoredto develop an ON/OFF PEGylation strategy to harness the cell penetrating power of thosebiomacromolecular self-assembled systems.

No description available.