Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Ces travaux ont consisté à optimiser une voie de synthèse des spirotétrahydro-β-carbolines. Une stratégie en trois étapes à partir du 3-(2-nitrovinyl)indole a été élaborée avec la séquence linéaire suivante : addition de Michael, réduction de la fonction nitro en amine et cyclisation de Pictet-Spengler. L’addition de Michael a été développée sans protection préalable de l’indole sous activation ultrasons. La réduction de la fonction nitro en amine a été menée en présence d’hypophosphites. Au cours de cette étude, la désoxygénation des cétones aromatiques a été identifiée comme réaction secondaire de la réduction et a fait l’objet d’un développement approfondi. Une nouvelle voie d’accès à des isatines fonctionnalisées en position C-5 par substitution des sels de diazonium a été mise au point notamment par la réaction de Heck-Matsuda qui a fait l’objet d’une étude plus complète. La réaction finale de Pictet-Spengler entre tryptamines et isatines aboutit aux spiroindolones. Finalement, l’activité biologique des dérivés spirotétrahydro-β-carbolines synthétisés a été évaluée et l’un d’entre eux a montré une très bonne activité envers le Plasmodium falciparum / The aim of this work was to optimize a new way of synthesis of spirotetrahydro-β-carbolines. A three-steps strategy from the 3-(2-nitrovinyl)indole was elaborated with the following linear sequence : Michael addition, reduction of the nitro function into amine and Pictet-Spengler cyclisation. Michael addition was developed without protection of the indole under sonication activation. Reduction of the nitro function into amine was carried out with hypophosphites. During this study, desoxygenation of aromatic ketones has been identified as a side reaction of the reduction and has been the subject of specific development. A new synthetic pathway to functionalized isatins on C-5 position via substitution of diazonium salts was developed, and Heck-Matsuda reaction has been more particularly studied. The final Pictet-Spengler reaction between tryptamines and isatins led to tetrahydro-β-carbolines. Finally, biological activity of spirotetrahydro-β-carbolines was evaluated and one of these compounds showed good activity against Plasmodium falciparum

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

Pyranonaphthoquinone lactones have been recently found to be selective inhibitors of the serine-threonine kinase AKT/PKB. AKT/PKB plays a major role in tumorigenesis, hence these compounds have a great potential to act as anti-cancer agents. They act by a novel bioreductive alkylation mechanism of inhibition of AKT/PKB. In this work, 7-deoxykalafungin, a pyranonaphthoquinone lactone and its deconstruction analogs were synthesized. The structural features of the compounds necessary to inhibit AKT1 potently and selectively were determined. It was observed that compounds with a pyran ring were more potent in inhibiting AKT1. Conversely, flexible compounds were found to be weak inhibitors of AKT1. Also, presence of a lactone ring was found to be favorable in inhibiting AKT1. Of the compounds tested, 7-deoxykalafungin was the most potent inhibitor of AKT1 (IC50 = 0.28 µM against AKT1) and compound 4-61 was the most potent inhibitor of PKA (IC50 = 0.43 µM against PKA).

The tetrahydroisoquinoline (THIQ) core structure is explored as a steroidomimetic nucleus with attractive pharmaceutical properties. A library was synthesised employing Pomeranz-Fritsch, Pictet-Spengler, Bischler-Napieralski strategies yielding 77 final targets, substituted at every position, for biological evaluation. Complementary strategies overcame synthetic difficulties, sometimes yielding two products in a single cyclisation. Three compounds were initially tested against a panel of 19 nuclear receptors (NRs) and exhibited broad substitution-dependent activity. 2-(4-Chlorophenyl)-1-isopropyl-1,2,3,4-tetrahydroisoquinolin-6-ol fully inhibited every NR at 100 µM, confirming the THIQ as a lead for optimisation. Compounds were evaluated for cytotoxicity against 60 cell lines by the NCI (USA), exhibiting moderate to insignificant cytotoxicity. Three compounds showed ca. 30-90% of average growth inhibition and were selected for a five dose test. Off-target evaluation highlighted compounds with activity against glucagon-like peptide 1 secretion, calcitonin gene-related peptide receptor antagonism and with >100% inhibition against the metabotropic glutamate receptor 2. Estrogen receptor-related receptor α (ERRα), a constitutively active orphan NR, is a hormone-dependent cancer target and diethylstilboestrol (DES), a known inverse agonist, possesses similarities to THIQs. THIQs tested against ERRα revealed no general SAR rules, but showed a lower degree of efficacy in a commercial TR-FRET assay, with 1-benzyl-2-(4-chlorophenyl)-4-methyl-1,2,3,4-tetrahydroisoquinolin-6-ol showing 79% efficacy at 100 µM as an inverse agonist, being more active than DES (64% at 100 µM). Inhibition of steroidogenic enzymes like 17β-hydroxysteroid dehydrogenase type 1 (17β-HSD1) is an emerging approach for the treatment of HDBC, compared to other current clinical strategies. THIQs evaluated against 17β-HSD1 showed good activity in both whole cell and cell lysate assays, with the best inhibitor, 2-(4-chlorophenyl)-4-isopropyl-1,2,3,4-tetrahydroisoquinolin-6-ol, possessing an IC50 value of 336 nM. The value of THIQ as a drug-like steroidomimetic scaffold is thus established and this work reveals straightforward strategies to optimise potency and selectivity for a range of potential targets by structural and stereochemical iteration.

Les cyclopropanes sont rencontrés dans de nombreux produits naturels ou synthétiques bioactifs. Les travaux réalisés portent sur le développement de couplages catalysés par le palladium permettant d'accéder à des cyclopropanes diversement substitués. Des couplages de Suzuki-Miyaura impliquant les cis- et trans-2-benzyloxy-cyclopropyltrifluoroborates de potassium ont pu être mis au point après un important travail d'optimisation. L'accès à des aminocyclopropanes par des couplages de type Hartwig Buchwald impliquant des iodures cyclopropaniques s'est révélé beaucoup plus difficile à mettre en œuvre. Leur faisabilité a été démontrée avec un exemple, en version intramoléculaire, mais les résultats n'ont pas pu être généralisés. Les premiers exemples de couplages de Sonogashira entre des iodures cyclopropaniques diversement substitués et des alcynes terminaux ont été décrits. Les alcynylcyclopropanes correspondants sont obtenus avec d'excellents rendements et rétention de configuration. Les cis-2-alcynylcyclopropanecarboxamides préparés par cette méthode peuvent subir une cyclisation 5 exo-dig en milieu basique et conduire à des énamides incorporant un motif 3-azabicyclo[3.1.0]hexane. En milieu acide, ces composés engendrent des ions N-acyliminiums bicycliques pouvant être impliqués dans des réductions ioniques ou des cyclisations de Pictet-Spengler menant à une grande diversité de composés hétérocycliques azotés originaux de manière totalement diastéréosélective.

The war between human and malaria has never stopped, and the development and application of antimalarial drugs has not eradicated this terrible disease. To fight drug-resistant malaria, many leads have been studied over the years. (1S,3R)-MMV008138 and MMV1803522 are two compounds that have been studied in the Carlier Group. My research focused on the structural variation of each of these compounds, in the hope that greater potency could be realized. Chapter 2 describes my work on (1S,3R)-MMV008138, which inhibits the enzyme PfIspD in the methylerythritol phosphate (MEP) pathway. This compound shows good in vitro potency against the drug resistant Dd2 strain of Plasmodium falciparum. However, this lead showed no activity in mouse models. This lack of activity may be due to poor metabolic stability of the compound. However, a significant increase in in vitro potency could also improve in vivo activity. Towards that end, I focused on further variation of the D-ring and A-rings. With the regard to the D-ring, we made five analogs of MMV008138 that replaced the 2,4-dichlorophenyl ring with dihalogenated thiophen-3-yl and thiophen-2-yl rings. We also explored the effect of installing a cyano group on the A-ring of MMV008138. Unfortunately, none of these new compounds were potent growth inhibitors of Dd2 strain P. falciparum. We conclude that the lead goes into a well-defined pocket within the PfIspD enzyme that only accommodates 2,4-dihalogenated phenyl D-rings. This pocket also cannot accept any substitution larger than F on the A-ring. Interestingly, the crystal structure of 5-cyano-substituted MMV008138 was obtained ((±)-2-50c). This is the first compound out of more than 100 analogs of MMV008138 family to be amenable to crystallization. The solid state conformation of the (±)-2-50c revealed that the C3-carboxyl group was in a pseudoequatorial orientation, and the C1-aryl group was thus in a pseudoaxial orientation. 1H NMR spectroscopic studies in CD3OD-D2O were carried out to determine the solution conformation. As expected from previous studies of ester derivatives of MMV008138, these studies indicated that in solution, 2-5 would adopt both the C3-carboxyl pseudoequatorial and pseudoaxial conformations. In Chapter 3, I describe the synthesis of analogs of the antimalarial drug candidate MMV1803522. This β-carboline-3-carboxamide shows good in vitro growth inhibition potency of Dd2 strain P. falciparum, operating by a still unknown mechanism. To investigate the pharmacophore of this lead, I first sought to determine whether the pyridine N (i.e. N2) of the β-carboline was important for in vitro potency. I prepared series of carbazole analogs of MMV1803522, which replace N2 with a CH. These compounds potently inhibited the growth of Dd2 strain P. falciparum. These results suggest that N2 of MMV1803522 is not involved in any energetically significant interactions with its target protein. To further identify the pharmacophore, we prepared truncated analogs lacking the A- and B- rings (biphenyl analogs), and tricyclic analogs that feature a reversed indole moiety. Unfortunately, the biphenyl analogs and reversed indole analogs show no growth inhibition at 10,000 nM the highest concentration tested. Lastly, I describe analogs of MMV1803522 in which the 3,4-dichlorophenyl ring of MMV1803522 was replaced with halogenated thiophene. This substitution was tolerated, but compounds were roughly half as potent as MMV1803522. / Doctor of Philosophy / Malaria, mainly caused by the infection of P. falciparum, is a serious worldwide disease. In 2020, there were 241 million cases of malaria infections and over 600,000 deaths from malaria. Combinations of commercially available antimalarial compounds, such as chloroquine, mefloquine and artemisinin, are commonly used as combination therapies to treat malaria. Since different antimalarial compounds have different mechanisms of action, this combination strategy can greatly slow down the spread of drug-resistant parasites. However, multiple drug-resistant strains of P. falciparum have been reported. Therefore, there is an urgent need for new antimalarial compounds with novel mechanisms of action. This dissertation involves my research on the investigation and optimization of two novel antimalarial compounds, MMV008138 and MMV1803522. MMV008138 is an inhibitor of the MEP pathway, which is an essential metabolic pathway and attractive target for antimalarial therapies, in malaria parasites. The parasites cannot survive, with the MEP pathway inhibited. Since the MEP pathway is not present in human, the MMV008138 molecule is unlikely to have toxicity to human. The MMV008138 molecule has been demonstrated to have great in vitro performance of inhibiting the MEP pathway in several studies, however, the in vivo performance in mouse models is yet to improve. This may be due to the poor metabolic stability of this compound. The compound decomposes in the mouse body before it takes effect. To enhance the metabolic stability and potency, I performed chemical modifications on the A- and D-rings of the MMV008138 compound. An X-ray crystal structure was obtained to help elucidate the conformer distribution of MMV008138. This crystal structure can be used to guide our understanding of the docking of this compound to the target enzyme in the future. MMV1803522 is another compound that shows great potency in vitro and in vivo. This compound is fully oxidized and contains four aromatic rings. However, the target enzyme and the mechanism of action of MMV1803522 is yet to be discovered, and the structure-activity relationship between the chemical structure and the biological activity of this molecule is still unknown. Therefore, I have developed synthetic methods to synthesize a series of compounds that are structurally similar to the MMV1803522 and found that potency of this molecule is not due to the nitrogen on the C-ring. Also, the number and size of the ring structures in the MMV1803522 may be crucial for this molecule to exhibit great potency in vitro and in vivo.

Malaria continues to threaten human beings, causing a staggering number of more than 400,000 deaths each year. Although effective treatment and prevention methods are available, rapidly emerging resistance towards existing drugs is of great concern, and the need for novel antimalarial compounds are still urgent. The Malaria Box lead molecules MMV008138 and MMV665831 are promising in this regard, due to their apparently novel antimalarial mechanisms of action. The target of MMV008138 is the PfIspD enzyme in the MEP pathway, which is absent in humans. This difference makes the PfIspD a great target. However, while MMV008138 shows potency against Plasmodium falciparum-infected human erythrocytes in vitro, no efficacy was observed in a humanized mouse model or a P. berghei infected mouse in vivo. In order to block potential metabolic spots and to probe for steric demand, a series of analogous featuring C1-deuteration, methyl substitution on B- and C-ring, and an ethylene bridge were prepared. The effect of various substitution on the tetrahydro-β-carboline conformation and D-ring orientation was studied. In the course of preparing the C1-Me analog of MMV008138 featuring 2',4'- dichloro substitution, unexpected ring-expanded azepane products were isolated. Later it was found that the desired product could be isolated when the imine formed was treated with acid at lower temperature. Other intermediates possessing a 2ʹ- substituent were also isolated under the low temperature acid treatment protocol, which upon heating in acid gave the ring-expanded azepane we initially isolated. A mechanism was proposed to account for the formation of the azepane as well as other intermediates. The driving force of the expansion reaction was explored, and the hypothesis that the steric interaction between the 2ʹ-substituent and the C1-Me was supported via DFT calculation and conformational analysis. MMV665831 is another potent hit from the Malaria Box, and it appears to inhibit the hemoglobin endocytosis process of P. falciparum. The structure–activity relationship of MMV665831 was studied with analogues featuring modifications on C2-benzamide, C3-ester, C-7 phenol, as well as the phenolic Mannich base moiety. Modifications at phenolic Mannich base moiety leads to the discovery of an analogue that is twice as potent toward cultured P. falciparum compared to MMV665831. We were worried the phenolic Mannich base moiety might act as the precursor of toxic quinone methide intermediates, and designed two analogs to block this potential toxicophore. Although the modification resulted in reduced potency, this result proved that the potency of MMV665831 does not stem from the formation of quinone methides. Unfortunately, MMV665831 did not reduce parasitemia in P. berghei- infected mice. Fast hepatocyte metabolism was observed for MMV665831, and the loss of in vivo efficacy was discussed in comparison with other phenolic Mannich bases with similar hepatocyte stability. / Doctor of Philosophy / In the fight against malaria, one concerning issue is the rapidly emerging resistance towards existing drugs. The continuous development of antimalarials with novel mechanism of action is greatly needed. To accelerate the development of novel antimalarials, an open access ensemble of 400 compounds that are toxic to the malaria parasite known as the Malaria Box, has been made available. My work involves the optimization of two compounds from this ensemble, MMV008138 and MMV665831. MMV008138 kills the malaria parasite by inhibiting an enzyme named PfIspD, which is absent in human. In the parasite an enzyme called PfIspD is responsible for the biosynthesis of IPP and DMAPP, two chemical building blocks that are essential for all cells. It is unlikely that MMV008138 will interrupt with the biosynthesis of IPP and DMAPP in human, since we use another enzyme to synthesize them. Although MMV008138 shows great in vitro potency, but did not protect a live mouse from malaria infection. The lack of in vivo efficacy could stem from the rapid metabolism of MMV008138, and analogs aimed to prevent metabolism were designed and prepared. While preparing analogs featuring 2ʹ-substitution, the desired product was not found, but other unexpected by-products were isolated. The conditions that leads to both the desired products and the by-products were found, and the mechanistics detail of this unexpected reaction were studied. During the blood-stage, which causes malaria symptoms in human, the Plasmodium falciparum parasite invades and feeds on human red blood cells (erythrocytes). The parasite destroys human hemoglobin through a multistep process that begins by transporting the hemoglobin from the red blood cell into itself, a process called endocytosis. MMV665831 appears to interfere with this endocytosis process of P. falciparum, thus starving the parasite of its food. Analogs of MMV665831 were prepared to probe for the effect on potency, and one compound that is twice as potent in cultured parasites was found. The structure of MMV665831 contains a potentially unstable moiety, which might lead to toxicity in humans. Two analogs with the problematic moiety removed were designed and prepared, and one still shows antimalarial activity, showing that the reactivity of the potentially unstable moiety is not the reason for the antimalarial activity of MMV665831. However, MMV665831did not protect P. berghei-infected mice (murine malaria) in vivo, and the reason for the loss of efficacy was discussed.

La lavendamycine est une molécule d'origine naturelle isolée en 1981 et formée par Streptomyces lavendulae qui est une bactérie. Cette molécule pentacyclique possède des propriétés antivirale, antimicrobienne et anticancéreuse. En effet, elle permet d'inhiber l'activité enzymatique des topoisomérases de type I (enzymes nécessaires à la réplication de l'ADN lors de la division cellulaire). Cependant, son intérêt est assez limité du fait de sa toxicité élevée pour l'organisme humain. Des travaux antérieurs réalisés au laboratoire ont permis de synthétiser des analogues de la lavendamycine. Ces analogues diversement substitués se sont révélés moins toxiques mais de faible activité biologique. Ces composés possèdent une certaine mobilité conformationnelle (rotation autour de liaison entre les cycles). Il serait alors intéressant de conférer une rigidité structurale à ces molécules par la création d'un nouveau cycle afin d'obtenir des structures analogues à celle de la camptothécine (substance inhibitrice la plus connue des topoisomérases de type I).<br /><br />Pour synthétiser ces analogues de la lavendamycine à géométrie contrainte, nous avons d'abord préparé un premier analogue pentacyclique possédant un groupement hydroxyméthyle en position 3. L'étape-clef de cette synthèse est une réaction de Pictet-Spengler entre une quinoléine fonctionnalisée et l'ester méthylique du tryptophane. Dans un premier temps, nous avons tenté de créer le nouveau cycle via une cyclisation intramoléculaire en milieu basique au niveau d'un nucléofuge (introduit à partir du groupe hydroxyle) et de l'azote pyrrolique du cycle D. Mais la préparation des composés à groupements nucléofuges a posé de gros problèmes et cette approche n'a pas abouti. En revanche dans un deuxième temps, une cyclisation intramoléculaire réalisée à partir d'un aldéhyde (obtenu à partir du groupement hydroxyméthyle) et de l'azote pyrrolique du cycle D a permis de former la structure hexacyclique souhaitée. Les deux premiers analogues de la lavendamycine à géométrie contrainte ont ainsi été obtenus en huit et neuf étapes avec des rendements globaux de 9,9% et 5,1%.<br /><br />Nous avons poursuivi notre étude par une fonctionnalisation des analogues à mobilité conformationnelle réduite. En effet, nous avons d'abord préparé des analogues possédant un groupement amino en position 7 sur le cycle A. Cette fonctionnalité a été introduite très tôt dans la synthèse et elle a nécessité une protection sous forme de ditosylamide. Nous avons également préparé des analogues qui tout en conservant ce groupement amino en position 7 avaient soit un groupement méthoxy en position 6 sur le cycle A, soit un groupement éthyle sur la position 4 du cycle B. Malgré des difficultés, ces trois séries d'analogues ont été préparées en neuf ou dix étapes avec des rendements globaux compris entre 1,2% et 4,8%.<br /><br />Ces analogues de la lavendamycine à géométrie contrainte, ou non, ont été soumis à des tests biologiques par le NCI (National Cancer Institute). Ils ont montré généralement une certaine activité biologique mais qui n'était pas suffisamment importante pour être exploitée en tant qu'agents anticancéreux.

Ces travaux ont consisté à optimiser une voie de synthèse des spirotétrahydro-β-carbolines. Une stratégie en trois étapes à partir du 3-(2-nitrovinyl)indole a été élaborée avec la séquence linéaire suivante : addition de Michael, réduction de la fonction nitro en amine et cyclisation de Pictet-Spengler. L'addition de Michael a été développée sans protection préalable de l'indole sous activation ultrasons. La réduction de la fonction nitro en amine a été menée en présence d'hypophosphites. Au cours de cette étude, la désoxygénation des cétones aromatiques a été identifiée comme réaction secondaire de la réduction et a fait l'objet d'un développement approfondi. Une nouvelle voie d'accès à des isatines fonctionnalisées en position C-5 par substitution des sels de diazonium a été mise au point notamment par la réaction de Heck-Matsuda qui a fait l'objet d'une étude plus complète. La réaction finale de Pictet-Spengler entre tryptamines et isatines aboutit aux spiroindolones. Finalement, l'activité biologique des dérivés spirotétrahydro-β-carbolines synthétisés a été évaluée et l'un d'entre eux a montré une très bonne activité envers le Plasmodium falciparum